Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The evolutionary origin of "black" kokanee (Oncorhynchus nerka) and their genetic and phenotypic diversity… Moreira, Amanda Leigh 2014

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

Item Metadata

Download

Media
24-ubc_2014_november_moreira_amanda.pdf [ 3.4MB ]
Metadata
JSON: 24-1.0165554.json
JSON-LD: 24-1.0165554-ld.json
RDF/XML (Pretty): 24-1.0165554-rdf.xml
RDF/JSON: 24-1.0165554-rdf.json
Turtle: 24-1.0165554-turtle.txt
N-Triples: 24-1.0165554-rdf-ntriples.txt
Original Record: 24-1.0165554-source.json
Full Text
24-1.0165554-fulltext.txt
Citation
24-1.0165554.ris

Full Text

The evolutionary origin of “black” kokanee (Oncorhynchus nerka) and their genetic and phenotypic diversity in the Anderson and Seton lakes system by  Amanda Leigh Moreira  B.Sc., The University of British Columbia, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  MASTER OF SCIENCE in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Zoology)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  October 2014  © Amanda Leigh Moreira, 2014  ii Abstract  In order to conserve intraspecific biodiversity, it is crucial to identify genetic and phenotypic variation among populations and to understand the evolutionary processes that drive the evolution and persistence of such variation. Oncorhynchus nerka exists as two basic life history forms: the anadromous form, sockeye salmon and the non-anadromous form, kokanee. Unique populations of “black” kokanee are found in Lake Saiko, Japan, and in Anderson Lake and Seton Lake in the southwestern interior of British Columbia. They are distinct from other populations of O. nerka in that “black” kokanee spawn between 20 to 70 meters below the surface of each lake and spawn in the winter or early spring, whereas O. nerka typically spawns in the autumn in streams or shallow lake beaches. Using mitochondrial DNA and nine microsatellite loci, I investigated the evolutionary origin of eastern and western North Pacific populations of “black” kokanee. Both data sets support the hypothesis that “black” kokanee in Anderson Lake, Seton Lake and Lake Saiko have a polyphyletic origin resulting from repeated episodes of parallel evolution. Further, using variation at nine microsatellite loci, I demonstrated that “black” kokanee in the Anderson and Seton lakes system are genetically distinct from sympatric populations of sockeye salmon in Gates and Portage creeks, and that Anderson and Seton lake “black” kokanee are modestly distinct from one another. Anderson and Seton lake “black” kokanee differ dramatically from one another in standard length at maturity, but no differences were found between the two populations in size-adjusted maximum body depth or in gill raker numbers. My results provide an example of parallel evolution in north temperate freshwater fishes as well as evidence of yet another facet of distinctive biodiversity within O. nerka that needs to be recognized and accounted for in conservation planning for the taxon.   iii Preface  I was responsible for data collection, analysis, and interpretation. This project was conducted under supervision of Eric Taylor who contributed to data collection, analysis, and interpretation. Tissue samples were collected with the help of Gene Tisdale (Tisdale Environmental Consulting Inc.), William Alexander and William Terry (residents of Seton Portage). Tissue samples were also provided by Matthew Casselman (UBC Forest and Conservation Sciences), Steve Latham (Pacific Salmon Commission), David Levy (Levy Research Services Ltd), Shannan May-McNally (UBC Zoology), and Tetsuji Nakabo (Kyoto University). Additional DNA samples were from the Beaty Biodiversity Museum’s fish DNA archive provided by Eric Taylor (UBC Zoology). Eric Taylor developed the genetic protocols that I used in the project.    iv Table of Contents  Abstract .......................................................................................................................................... ii!Preface ........................................................................................................................................... iii!Table of Contents .......................................................................................................................... iv!List of Tables ............................................................................................................................... viii!List of Figures ............................................................................................................................... ix!Acknowledgements ........................................................................................................................ x!Dedication ..................................................................................................................................... xii!Chapter 1: Introduction ................................................................................................................ 1!1.1! Biodiversity and conservation ............................................................................................ 1!1.2! Diversity of fishes ............................................................................................................... 2 1.3! Oncorhynchus nerka ........................................................................................................... 3!1.4! “Black” kokanee: Gwenish ................................................................................................. 6!1.5! “Black” kokanee: Kunimasu ............................................................................................ 10!1.6! Research questions ........................................................................................................... 12 1.7! Hypotheses and predictions .............................................................................................. 13!Chapter 2: The evolutionary origin of “black” kokanee (Oncorhynchus nerka) ................... 15!2.1! Introduction ...................................................................................................................... 15!2.2! Methods ............................................................................................................................ 18!2.2.1! Lakes and samples ..................................................................................................... 18 2.2.2! Mitochondrial DNA ................................................................................................... 19 2.2.3! Mitochondrial DNA data analysis ............................................................................. 20  v 2.2.4! Microsatellite DNA ................................................................................................... 21 2.2.5! Microsatellite DNA data analysis .............................................................................. 21 2.3! Results .............................................................................................................................. 23 2.3.1! Mitochondrial DNA ................................................................................................... 23 2.3.2! Microsatellite DNA ................................................................................................... 27 2.4! Discussion ......................................................................................................................... 33 2.4.1! Polyphyletic origin of “black” kokanee .................................................................... 33 2.4.2! Phenotypic divergence ............................................................................................... 34 2.4.3! Low genetic diversity ................................................................................................ 37 2.4.4! Phylogeographic structure ......................................................................................... 38 2.4.5! Conclusions ............................................................................................................... 39 Chapter 3: Genetic and morphological differentiation of “black” kokanee (Oncorhynchus nerka) in the Anderson and Seton lakes system ........................................................................ 41!3.1! Introduction ...................................................................................................................... 41!3.1.1! Biocomplexity ........................................................................................................... 41 3.1.2! Diversity within Oncorhynchus nerka: sockeye salmon and kokanee ...................... 41!3.1.3! “Black” kokanee ........................................................................................................ 43!3.2! Methods ............................................................................................................................ 46!3.2.1! Microsatellite DNA ................................................................................................... 46 3.2.2! Microsatellite DNA data analysis .............................................................................. 48!3.2.3! Morphological analysis ............................................................................................. 50!3.2.4! Morphological data analysis ...................................................................................... 51 3.3! Results .............................................................................................................................. 52  vi 3.3.1! Within population microsatellite DNA variation ...................................................... 52 3.3.1.1! Hardy-Weinberg Equilibrium and linkage disequilibrium ................................ 52!3.3.1.2! Temporal variation ............................................................................................ 53!3.3.1.3! Allelic richness and genetic diversity ................................................................ 53!3.3.1.4! Unique alleles in “black” kokanee .................................................................... 57 3.3.2! Among population microsatellite DNA variation ..................................................... 58 3.3.2.1! FST comparisons ................................................................................................. 58!3.3.2.2! Bayesian analysis of population structure ......................................................... 60!3.3.2.3! DAPC analysis ................................................................................................... 65!3.3.2.4! Hierarchical analysis of genetic structure .......................................................... 68 3.3.3! Morphological differentiation ................................................................................... 68 3.3.3.1! Within population variation ............................................................................... 68 3.3.3.2! Between population variation ............................................................................ 69 3.4! Discussion ......................................................................................................................... 72 3.4.1! Within population genetic variation .......................................................................... 72 3.4.2! Among population genetic variation ......................................................................... 73 3.4.3! Anderson and Seton lakes “black” kokanee .............................................................. 79 3.4.4! Morphological variation ............................................................................................ 80 3.4.5! Comparisons with Japanese “black” kokanee (kunimasu) ........................................ 83 3.4.6! Conclusions ............................................................................................................... 85 Chapter 4: General discussion ................................................................................................... 87!4.1! Summary of results ........................................................................................................ 87 4.2! Conservation implications ............................................................................................. 88! vii 4.3! Future studies ................................................................................................................. 90!4.4! Conclusion ..................................................................................................................... 91 References .................................................................................................................................... 92!Appendix .................................................................................................................................... 107!A.1! ND1 maximum likelihood tree .................................................................................. 107!A.2! ND2 maximum likelihood tree .................................................................................. 108 A.3! Population genetic summary statistics ....................................................................... 109!   viii List of Tables  Table 2.1 Summary of mitochondrial DNA (mtDNA) haplotypes............................................... 25 Table 2.2 Summary of populations used for microsatellite anaylsis ............................................ 29 Table 2.3 STRUCTURE results .................................................................................................... 32 Table 2.4 Frequency of unique alleles .......................................................................................... 32 Table 3.1 Summary of populations used for microsatellite analysis ............................................ 48 Table 3.2 Summary of population genetic statistics ..................................................................... 55 Table 3.3 Frequency of unique alleles .......................................................................................... 57 Table 3.4 Pairwise FST comparisons ............................................................................................. 59 Table 3.5 STRUCTURE results .................................................................................................... 61 Table 3.6 Microsatellite variation partitioned heirarchically  ....................................................... 68 Table 3.7 Summary of morphological variation ........................................................................... 69 Table A.3 Summary of chapter 2 population genetic statistics .................................................. 109            ix  List of Figures  Figure 1.1 Map of Oncorhynchus nerka's geographic distribution................................................. 4 Figure 1.2 Picture of mature sockeye salmon and kokanee ............................................................ 6 Figure 1.3 Map of the Anderson and Seton lakes system ............................................................... 8 Figure  1.4  Picture  of  mature  Seton  Lake  “black”  kokanee ............................................................. 9 Figure 1.5 Picture of mature Anderson Lake  “black”  kokanee .................................................... 10 Figure  1.6  Picture  of  mature  Lake  Saiko  “black”  kokanee ........................................................... 12 Figure 2.1 Maximum likelihood tree ............................................................................................ 26 Figure 2.2 Neighbor-joining tree .................................................................................................. 30 Figure 2.3 STRUCTURE results .................................................................................................. 31 Figure 3.1 Map of British Columbia ............................................................................................. 47 Figure 3.2 STRUCTURE results using the entire dataset ............................................................ 62 Figure 3.3 STRUCTURE results using the Anderson/Seton lakes system populations (K = 3) .. 63 Figure 3.4 STRUCTURE results using the Anderson/Seton lakes system populations (K = 4) .. 64 Figure 3.5 DAPC results using entire dataset ............................................................................... 66 Figure 3.6 DAPC results using the Anderson/Seton lakes system populations ............................ 67 Figure 3.7 Female standard length histogram ............................................................................... 71 Figure 3.8 Male standard length histogram .................................................................................. 72 Figure A.1 ND1 maximum likelihood tree ................................................................................. 107 Figure A.2 ND2 maximum likelihood tree ................................................................................. 108    x  Acknowledgements  Firstly, I would like to thank my supervisor, Eric (Rick) Taylor, for taking me as a student/getting me out of being a barista for life and keeping me as a student even when I thought I could not do it. Thank you for taking a chance on me and giving me this amazing opportunity. I would like to thank David Levy for introducing Rick to this project and answering all my questions. You were an invaluable resource. Also, thank you to my committee members, Jordan Rosenfeld and Colin Brauner for your contributions and suggestions. Jessica Mackenzie, I cannot thank you enough for the support and experiences that you have given me. You are an amazing person to aspire to be.  Thank you to my fellow lab mates: Monica Yau, Shannan May-McNally, and Jenn Ruskey for your assistance in the lab. Thank you for all the laughs! I wish you all the best to wherever you go from the Taylor Lab. I would also like to thank Rebecca Piercey for all the help she gave me and for listening to me complain about everything. Thank you for driving in a blizzard with no winter tires to Lillooet.  Many people and organizations helped me collect samples and I am grateful to all of them: Matthew Casselman (UBC Forest and Conservation Sciences), Gene Tisdale and colleagues (Tisdale Environmental Consulting), Steve Latham (Pacific Salmon Commission). I honestly believe that William Alexander and his brother William Terry are the nicest people I have ever met. They helped collect Anderson Lake Gwenish and introduced me to the cultural significance of Gwenish and salmon in general.  My office mates have created an amazing support system. Thank you, Kaylee Byers, for being  hilarious  and  listening  to  all  my  “problems.”  You  are  literally  the  funniest  person  I  know.    xi Thank you, Kim Gilbert, for sending me cat pictures and introducing me to other grad students. Thank you, Seth Rudman, for all the stories of the ridiculous things that have happened to you. I could not have asked for three better people to share a tiny office with. Thank you to honorary office mate, Alison Porter, for always distracting me with the long/off topic conversations we have.  I do not think I would have stayed sane without: Nick Adamson, Karen Cheung, Bob Foerster, Michael Hulland, Pearly Li, Jeremy Orsted and Danielle Scott. Thank you all for keeping me laughing. Also, I am grateful to the staff at Bean Around the World for providing me with free coffee and being a refuge from grad school. And thank you to the customers for all your words of encouragement. I am so eternally grateful for Brittney Soobotin and Krista Soobotin. You two are truly amazing. There are no words to describe how supportive you have been.  I do not think I could have asked for a more supportive and helpful family. Thank you Mom and Dad for everything you have done to support me and for taking us every year to see spawning kokanee. Thank you to my siblings: Carmen, Bryce, and Darrin. This was a group thesis and I do not think I would have finished it without the help that each of your provided. I would also like to thank the most important members of the Moreira family: Sampson, Merry, Pippin and the late Salem.    xii Dedication  For Carmen, Bryce, and Darrin. Without you I would have no grammar, no computer and no Gwenish. Thank you for the unconditional support.   1 Chapter 1: Introduction  My  thesis  explores  the  evolutionary  origin  of  “black”  kokanee  and  evaluates  the  genetic  and  phenotypic variation that they represent within the Anderson and Seton lakes system. This introductory chapter reviews the importance of conserving biodiversity specifically focusing on fishes, reviews the diversity found within Oncorhynchus nerka, and introduces unique populations of O. nerka. This chapter ends with my specific research questions, hypotheses and predictions.  1.1 Biodiversity and conservation A major goal of conservation biology is to preserve phenotypic and genetic diversity, as well as the evolutionary processes and adaptive landscapes that produce such variability, and this goal has increasingly been applied to the conservation of intraspecific diversity (Meffe and Carroll 1994; Moritz 2002; Allendorf and Luikhart 2007). Conservation of intraspecific diversity contributes to the persistence of species over a range of habitats and through changing environments (Luck et al. 2003; Duffy 2009; Bolnick et al. 2011). Intraspecific diversity, however, is often overlooked by the general public and by managers because they are focussed on ecosystem and species conservation (Hughes et al. 1997).  Defining intraspecific units of conservation can be conceptually and practically difficult (Moritz 1994; Crandall et al. 2000; Wood and Gross 2008), but there have been various approaches proposed (e.g., Allendorf et al. 1997; Taylor et al. 2011, 2013). Widely distributed species that possess a wealth of diversity are especially problematic because first one needs to gain a better understanding of their current ecological roles and evolutionary diversity before  2 focusing on the potential biological consequences of their loss (e.g., Taylor et al. 2013). This is especially concerning if such populations are unknowingly declining. Therefore, to conserve intraspecific diversity effectively, we must first document it and secondly understand the processes that have produced and maintained that diversity in a changing environment (Moritz 2002; Pressey et al. 2007). With that said, fishes are a model group to study all levels of biodiversity and conservation.  1.2 Diversity of fishes Fishes make up more than half of all extant vertebrate species and exhibit considerable morphological, ecological, and behavioural diversity at all taxonomic levels (Volff 2005; Helfman et al. 2009). At the intraspecific level, extensive and taxonomically unrecognized diversity occurs in temperate freshwater species (reviewed by Behnke 1972; Taylor 1999). Evidence of such diversity can be seen, for example, in the threespine stickleback (Gasterosteus aculeatus, McPhail 1984; Schluter 1996), European whitefish (Coregonus lavaretus, Sendek 2004), pygmy whitefish (Prosopium coulterii, Gowell et al. 2012), lake whitefish (Coregonus clupeaformis, Bernatchez and Dodson 1990), rainbow/steelhead trout (Oncorhynchus mykiss, Keeley et al. 2005, 2007), brook char (Salvelinus fontinalis, Dynes et al. 1999), brown trout (Salmo trutta, Ferguson 1989) and sockeye salmon/kokanee (Oncorhynchus nerka, Taylor 1999). The majority of intraspecific diversity is maintained by some degree of spatial or temporal reproductive isolation that is further reinforced by selection for additional ecological differentiation (e.g. growth rates, age of maturity, habitat and/or food preference) (Behnke 1972). Differences in trophic niches that promote differentiation in feeding ecology (threespine stickleback, lake whitefish, brown trout, references above) are common features promoting  3 intraspecific divergence (Schulter 1996; Taylor 1999). Salmonids (salmon, trout, char and their relatives), however, exhibit extensive variability in reproductive biology that is expressed as intraspecific diversity in size  and  time  of  maturation  (e.g.,  seasonal  “races”  such  as  spring  and  fall Chinook salmon (Oncorhynchus tschawytsha,  “winter”  and  “summer”  steelhead  trout,  reviewed in Groot and Margolis 1991; Quinn 2005). Another striking example of divergence in reproductive ecology is the presence of alternative life-history types of O. nerka: sockeye salmon and kokanee, as described below.   1.3 Oncorhynchus nerka Intraspecific diversity, particularly in reproductive ecology and life history, is exceptionally notable in Oncorhynchus nerka, which includes two main forms: sockeye salmon, the anadromous (or sea-going) form, and kokanee, the nonanadromous (or freshwater-resident) form (Nelson 1968; Burgner 1991). Both forms are indigenous to tributaries in the North Pacific, ranging from southern Kamchatka and Japan in the western Pacific to the Columbia River in the eastern Pacific (Burgner 1991, Fig. 1.1).  It is generally accepted that contemporary kokanee populations evolved from sockeye salmon multiple times independently in different locations throughout its range in the last 15,000 years (Ricker 1940; Nelson 1968; Foote et al. 1989). This is supported by the observations that: (i) the geographic distribution of kokanee is almost wholly contained within that of sockeye salmon, (ii) sockeye salmon can produce kokanee when introduced to new systems (and vice versa), and (iii) kokanee are polyphyletic as inferred from phylogeographic analysis (Ricker 1940, 1959; Nelson 1968; Taylor et al. 1996).   4  Figure 1.1. Map highlighting the broad natural distribution of Oncorhynchus nerka (sockeye salmon and kokanee) ranging from northern Japan to Washington State (map modified from Steinback and Fuller 2005).      Sockeye salmon and kokanee have considerably different life histories. Both sockeye salmon and kokanee adults spawn in streams tributary to lakes or on submerged lakeshore beaches and the young of both forms migrate to a nursery lake. Sockeye salmon then spend one to three years feeding in the lake before migrating to the ocean where they spend another one to three years growing and maturing (Burgner 1991; Wood 1995). From the ocean, they return to freshwater to spawn and die. In contrast, kokanee mature and spawn over two to four years entirely in freshwater (Ricker 1940).  Migratory life history differences between sockeye salmon and kokanee are accompanied by divergence in numerous traits (Nelson 1968). For instance, the two forms differ genetically in  5 growth and developmental rates (Wood and Foote 1990; Foote et al. 1994), salt-water adaptability (Foote et al. 1992), morphological and meristic traits (Wood and Foote 1996; Foote et al. 1999), and sustained swimming performance (Taylor and Foote 1991). Most of these traits may be a result of selection stemming from the differential physiological demands of anadromy and differences in trophic and predation ecology in the sea and in lakes (e.g., Taylor and Foote 1991). Owing to differences in marine versus freshwater productivity and the demands of migration, sockeye salmon are typically much larger at maturation than kokanee (approximately 40-70 cm versus 20-30 cm, respectively). Assortative mating by size between the forms promotes further divergence and reproductive isolation when the forms are sympatric (Foote and Larkin 1988; Taylor 1999). Sockeye salmon and kokanee do, however, typically exhibit similar colouration: both turn from a countershaded silver on the sides and black/grey on the back to bright red with an olive green head and blackish-red fins when they reach maturity (Craig and Foote 2001; Fig. 1.2). Molecular analyses based on allozyme allele frequency variation and minisatellite and mitochondrial DNA variation as well as in quantitative traits have shown that sockeye salmon and kokanee are genetically distinct in allopatry as well as when they are sympatric (Foote et al. 1989; Wood and Foote 1990; Taylor et al. 1996; Wood and Foote 1996). Specifically, sockeye salmon and kokanee form distinct gene pools in sympatry (spatially and temporally) in many British Columbia lakes (e.g., Takla, Shuswap and Babine lakes) with the divergent forms found in the same lake often more closely related to each other than they are to the comparable form in different lakes (Taylor et al. 1996; Wood and Foote 1996). Fully viable and fertile hybrids between sockeye salmon and kokanee can, however, be formed experimentally as well as in nature (Wood and Foote 1996). Despite such gene flow in nature, the forms remain genetically  6 distinct likely as a result of selection against hybrids associated with their very different life histories (Wood and Foote 1990; Wood and Foote 1996).    Figure 1.2. Mature sockeye salmon (top two fish) and kokanee (bottom four fish) sampled from a tributary of Takla Lake, British Columbia. In both the sockeye salmon and kokanee, females are at the top and males are at the bottom (C. Foote, Vancouver Island University).  The white scale marker is approximately 1 m in length.    1.4 “Black” kokanee: Gwenish  In addition to the fascinating divergence within O. nerka represented by sockeye salmon and kokanee, there is also considerable variation within both of these forms. One particularly remarkable and understudied example of such diversity is found within Anderson Lake and Seton Lake in the Bridge/Seton River system of the Fraser River drainage in the southwestern interior of British Columbia (Fig.  1.3).  A  unique  form  of  kokanee,  locally  called  “Gwenish,” is found in these lakes. Gwenish are black at maturity (Fig. 1.4, 1.5) and have spawning behaviour that is distinct within Oncorhynchus. Gwenish spawn late in the season, typically starting in  7 November in Seton Lake and in January in Anderson Lake (Levy 2012). In Anderson Lake, Gwenish are on average three to four years old at maturity, whereas those in Seton Lake are two to three years old at maturity (Morris and Caverly 2004). Additionally, Anderson Lake Gwenish are larger than Seton Lake Gwenish even when fish of the same age class are compared. In both lakes Gwenish spawn between 20 to 70 meters below the surface of the lake (typically greater than 50 meters) (Morris and Caverly 2004; Levy 2012). After spawning, Gwenish float to the lake’s  surface  due to their distended swim bladders (Morris et al. 2003; Levy 2012). Their distended  swim  bladders  give  them  a  deep  body  profile  that  is  distinct  from  “regular”  kokanee  populations (Levy 2012). It is hypothesized that Gwenish seek out deep water sites for spawning owing to the presence of upwelling of relatively warm, oxygenated water at these areas (Morris and Caverly 2004).  In the Anderson and Seton lakes system there are also spawning populations of sockeye salmon in Gates Creek (spawning in July to September) and Portage Creek (spawning in September to November) (Komori 1997). Sockeye salmon have also been observed (in smaller numbers) in the Seton River, which connects Seton Lake to the Fraser River, but they are believed to be occasional strays that enter the river during periods of adverse migratory conditions (Levy 2012). While sockeye salmon and Gwenish are broadly sympatric in both lakes, the sockeye salmon juveniles feed almost exclusively in Seton Lake where the two forms likely compete for trophic resources (Shortreed et al. 2001).   8  Figure 1.3. Anderson Lake (50.63°N, 122.38°W) and Seton Lake (50.70°N, 122.15°W) in the Bridge/Seton  system  of  BC’s  southwestern interior. Seton Lake is 24.3 km2 with an average depth of 85 meters. Anderson Lake is 28.3 km2 with an average depth of 140 meters.   9  Figure 1.4. Mature Gwenish (Oncorhynchus nerka) collected from Seton Lake (male above, female below) Lillooet, British Columbia in November 2013 during peak spawning season. Photo by: Amanda Moreira.   10  Figure 1.5. Mature Gwenish (Oncorhynchus nerka) collected from Anderson Lake (female above, male below) Lillooet, British Columbia in January 2014 during peak spawning season. Photo by: Amanda Moreira.   1.5 “Black” kokanee: Kunimasu Remarkably, despite the broad distribution of kokanee in watersheds across the North Pacific, the only other known occurrence of “black” kokanee is in Lake Saiko, Japan (near Tokyo) (Fig. 1.6).  “Black” kokanee,  or  “kunimasu”, in Lake Saiko, were transplanted from their native Lake Tazawa, in northern Japan, after hydroelectrical development in 1935 increased Lake  Tazawa’s  acidity (Nakabo et al. 2011). It was believed that the introduced kunimasu had gone extinct in Lake Saiko until their rediscovery in 2010 (Nakabo et al. 2011). Eyed eggs were also transplanted into several other lakes, but there is no evidence that these transplants were successful.  In  Japan,  native  “regular”  kokanee  from  Lake  Akan,  on  the  island  of  Hokkaido,  were   11 introduced into both Lake Saiko and Lake Tazawa. Consistent with most North American kokanee, these populations spawn in streams in the fall (Nakabo et al. 2011). In addition to being black at maturity, like Gwenish, kunimasu spawn at average depths of 30-40 meters in lakes from March to April (Nakabo et al. 2011). In Lake Tazawa, kunimasu were reported to spawn at depths of 15 to 300 meters in February, and they were also observed spawning year round. It is hypothesized that kunimasu are adapted to the cool water temperatures at these depths (between 3.9 to 5 °C; Nakabo et al. 2011). There have been no reports of floating or distended swim bladders in post-spawning kunimasu, a trait that is observed in Gwenish. In Japan, “black” kokanee are recognized as a separate species, Oncorhynchus kawamurae, based primarily on differences from  sympatric  “regular”  kokanee in gill rakers (36-45 versus 31-35  in  “regular”  kokanee),  pyloric  caeca (46-81 versus 55-82), anal fin rays (14-16 versus 13-15), and pectoral fin rays (15-17 versus 15-18) (Nakabo et al. 2014). Genetic divergence from sympatric  “regular”  kokanee  at  five microsatellite loci was also reported (Nakabo et al. 2011; Muto et al. 2012). Hybridization between the two types of kokanee was found to be rare, but there is evidence of past introgression although it is not known if such introgression occurred before or after both were introduced into Lake Saiko (Muto et al. 2012).  The similarities between Gwenish and kunimasu, the diversity they represent, and their relatively unknown biology raise several intriguing questions relevant to the evolutionary origin and the potential genetic and morphological distinctiveness within and between these forms of Oncorhynchus nerka.    12  Figure 1.6. Male kunimasu (Oncorhynchus kawamurae) collected off Nishin-okoshi, Lake Saiko, Yamanashi, Japan at 40 meters depth (photo courtesy of T. Nakabo).    1.6 Research questions My thesis encompasses two areas of investigation to test hypothesis about the distinctiveness (patterns) and evolutionary origin (process) of “black” kokanee and their relevance to the conservation of this form of O. nerka. Concordance among a number of traits, including genotypic and phenotypic variation, can provide strong evidence for distinctiveness within a species (e.g.,  Rising  and  Avise  1993;;  O’Donnell  et  al.  2004;;  Taylor  et  al.  2011). Therefore, I used mitochondrial DNA, nuclear DNA and morphological data sets to assess the distinctiveness and origin of  “black”  kokanee. Specifically, my research questions were:  1. What is the degree of phylogeographic structure and evolutionary origin of “black” kokanee?  13 2. Are “black” kokanee in the Anderson and Seton lakes system genetically distinct from each other and from sympatric sockeye salmon within the Oncorhynchus nerka complex, and  are  these  two  populations  of  “black”  kokanee  morphologically  divergent  from  each  other?  1.7 Hypothesis and predictions For question 1, I tested the hypothesis that “black” kokanee evolved postglacially at least twice, once in the eastern North Pacific and once in the western North Pacific. Multiple, independent divergences between forms within single taxa are common in freshwater fishes (Ryman and Stahl 1981; McPhail 1984; Hindar et al. 1991; Taylor 1999) including O. nerka (Foote el al. 1989; Taylor et al. 1996; Frazer and Russello 2013).  This hypothesis led me to predict that “black” kokanee would form at least two distinct phylogenetic groups: one native to the western Pacific and one native to the eastern Pacific. The “black”  kokanee  in Anderson Lake and Seton Lake most likely diverged from sockeye salmon populations spawning in Gates Creek or Portage Creek in the Anderson/Seton watershed. Within-lake divergences have been reported in other sockeye salmon/kokanee systems (e.g., Wood and Foote 1996; Taylor et al. 1996; Frazer and Russello 2013).  For question 2, I tested the hypothesis that “black” kokanee are distinct within O. nerka not only in colouration and reproductive ecology, but also genetically and morphologically. Across their geographic range, sockeye salmon demonstrate considerable genetic variation and population structure (e.g., Beacham et al. 2006). Kokanee also exhibit substantial genetic differentiation across their range (Wood and Foote 1996) even within lake systems (Taylor et al. 1997; Taylor 2013; Frazer and Russello 2013). Based on previous O. nerka population genetic  14 studies  and  the  phenotypic  differences  between  forms,  I  predicted  that  “black”  kokanee  would  be  genetically distinct from other populations of kokanee and from sympatric sockeye salmon populations. Additionally, due to differences in spawning timing and location as well as sizes at maturity between “black”  kokanee  in  Anderson  and  Seton  lakes, I predicted that there would be genetic heterogeneity within and between the lakes that will be observable as local population structure. Morphological variability (standard length, depth and gill raker counts) was also used as a measure of distinctiveness between the  “black”  kokanee  within  Anderson Lake and Seton Lake. Morphological variation may reflect local adaptation arising from divergent selection associated with differences in spawning environments, from phenotypic plasticity, or some combination thereof. Previous studies based on morphology in kokanee have shown that meristic traits such as gill rakers and vertebral counts are variable within and among lakes (e.g., Vernon 1957; Nelson 1968). The description and identification of genetic population structure, biodiversity, and taxonomic relationships in a constantly changing environment are necessary pieces of information to better understand “black” kokanee. Considering only sockeye salmon and kokanee there is already significant diversification within O. nerka, and “black” kokanee represent an additional, but understudied, aspect of such biodiversity.       15 Chapter 2: The evolutionary origin of “black” kokanee (Oncorhynchus nerka)  2.1 Introduction Understanding the patterns and origins of intraspecific diversity is important for conserving as many  genetic  “building  blocks”  of  individual  species  as  possible  (Waples 1995). For populations with distinct phenotypes, a key first step to better understand such diversity is to elucidate its evolutionary origin, especially if similar traits are found in more than one population. The multiple, independent divergence of similar phenotypic traits is widely recognized as a general phenomenon within species of north temperate freshwater fishes (Waples et al. 2004). For example: various life-history  types,  ecotypes,  and  “morphs”  have  arisen  independently  in  different geographic areas in brown trout (Salmo trutta, Hindar et al. 1991), rainbow smelt (Osmerus mordax, Taylor and Bentzen 1993), lake whitefish (Coregonus clupeaformis, Pigeon et al. 1997), threespine stickleback (Gasterosteus aculeatus, McPhail 1984; Schluter 1996), Chinook salmon (Oncorhynchus tshawytscha, Waples et al. 2004) and sockeye salmon/kokanee (Oncorhynchus nerka, Taylor et al. 1996). Parallel divergences within species are most likely the result of repeated opportunities for geographic isolation from Pleistocene glaciations, widespread dispersal opportunities, and filling of vacant niches in postglacial environments (Rice 1984; Taylor et al. 1996; Schluter 2000). The taxon Oncorhynchus nerka (sockeye salmon and kokanee) represents a fascinating system to study intraspecific diversity, in particular life history and reproductive system variation, and the evolutionary processes that have shaped this diversity in nature. It is generally accepted that extant populations of kokanee are the result of multiple, independent, post-glacial divergences from sockeye salmon throughout the range of O. nerka (Ricker 1940; Taylor et al.  16 1996). Evidence to support this scenario is supported by patterns of genetic similarity and allele sharing between the two forms (Foote el al. 1989; Taylor et al. 1996). The polyphyletic nature of kokanee is in agreement with the observation that the distribution of kokanee is generally contained within that of sockeye salmon (Ricker 1940). Further, kokanee and sockeye salmon populations have been found in lakes previously devoid of O. nerka after the introduction of either form suggesting that the forms can give rise to one another (Ricker 1940; Scott 1984; Kaeriyama et al. 1992). Island populations of kokanee (Japan, Vancouver Island and Alaskan islands) have likely resulted from divergence from sockeye salmon colonization after deglaciation (Taylor et al. 1996).  Evidence of parallel evolution of life history characteristics is not restricted to kokanee/sockeye salmon divergences within the O. nerka complex. For example, Wood et al. (2008)  hypothesized  that  “lake-type”  sockeye  salmon  (i.e.,  those  where juveniles reside and feed in lakes for at least one year before seaward migration) have evolved multiple, independent times from  geographically  proximate  ”sea/river-type”  sockeye  salmon  (i.e.,  juveniles  reside  in  spawning streams or migrate to sea as age 0 fish) an idea that the authors supported with mitochondrial  DNA  data.  In  addition,  “beach”  and  “stream”  spawning  ecotypes  of  kokanee  appear to have evolved from independent colonization of beach habitats by stream spawning kokanee in five British Columbia lakes (Duncan, Kootenay, Okanagan, Wood, and Tchesinkut lakes) in less than 10,000 years (Taylor et al. 2000; Frazer and Russello 2013). An understudied aspect of diversity within O. nerka is the existence of so-called  “black”  kokanee.  “Black”  kokanee are found in Lake Saiko, Japan and in Anderson Lake and Seton Lake in the southwestern interior of British Columbia, Canada. The populations on either side of the Pacific Ocean share similar unique spawning behaviour and colouration: deep lake spawning, black  17 colouration, and late spawning time from November to February in BC (Levy 2012) and March to April in Japan (Nakabo et al. 2011). These unusual features set them apart from sockeye salmon and kokanee which typically spawn in shallow water areas of streams or lake beaches, possess red body colouration and green heads at spawning, and spawn in the late summer to autumn (Burgner 1991).   In this chapter, I tested the hypothesis that “black” kokanee evolved at least twice: once in the eastern North Pacific and once in the western North Pacific, post-glacially. In doing so, I tested alternative hypotheses, single versus multiple origins, concerning the unknown origin of these unique “black” kokanee populations on either side of the Pacific Ocean. If the multiple origins hypothesis is supported then this suggests that selection has played a prominent role in the evolution  of  “black”  kokanee,  as it is unlikely that random genetic drift has caused the evolution of similar traits in two geographically distant locations especially given the evidence for parallel evolution in other traits in O. nerka and north temperate fishes more generally (Schluter 1996, 2000; Taylor 1999).  To test this hypothesis, I used mitochondrial DNA sequence variation and microsatellite DNA allele frequency variation. Mitochondrial DNA has a maternally, non-recombining mode of inheritance and typically exhibits extensive intraspecific polymorphism (Avise et al. 1987; Thomas et al. 1986). Mitochondrial DNA has been informative in understanding the evolutionary and phylogeographic relationships within species in a variety of taxa (Avise et al. 1987; Bernatchez and Wilson 1998; Avise 2000). In comparison, microsatellites are highly polymorphic, typically neutral, tandem repeat loci and are biparently inherited (Jarne and Lagoda 1996). These features of microsatellites and their high mutation rate have proven useful to understand the genetic relationships below the species level as well as to infer evolutionary  18 origins among populations (Lehmann et al. 1996; Abdul-Muneer 2014). The use of two independent data sets provides a powerful test to study the evolutionary processes and divergence patterns among “black” kokanee, and O. nerka in general.  The independent origin of “black” kokanee populations would reinforce the importance of parallel evolution as a central process of evolutionary change in the history of north temperate fishes while highlighting further diversity found within Oncorhynchus nerka.   2.2 Methods 2.2.1 Lakes and samples Seton Lake and Anderson Lake are part of the Fraser River drainage system, located west of the town of Lillooet in the southwestern interior of British Columbia (BC, see Fig. 1.3). Anderson Lake has a total surface area of 28.3 km2 and average depth of 140 m. It is fed by Gates Creek and empties into Seton Lake, which is also fed from water diverted from the Bridge River into Carpenter Reservoir. Seton Lake has a surface area of 24.3 km2 and an average depth of 85 m and it empties into the Fraser River at the town of Lillooet. There are two BC Hydro facilities in operation within the watershed: Seton River power dam and Shalath power generation facility. Rail tracks run along the length of both lakes on the north side (Levy 2012).  Samples of O. nerka were obtained from a variety of sources across the North Pacific (Tables 2.1,  2.2).  “Black”  kokanee  (post-spawners) samples were collected from the surface of Seton Lake using dipnets on November 20, 2013 with the help of Gene Tisdale (Tisdale Environmental Consulting Inc.). Adipose or pelvic fin samples were taken at the University of British Columbia (UBC) and stored in 95% ethanol for DNA analysis. Anderson Lake samples were collected in January 2013 by David Levy (Levy Research Services Ltd.) and I collected  19 samples on January 16, 2014. Samples were collected by surveying the beach for post spawning adults and adipose fin samples were taken on site at Anderson Lake and stored in 95% ethanol for DNA analysis.  Japanese “black” kokanee muscle tissue was graciously provided by Tetsuji Nakabo (Kyoto University). Portage Creek sockeye salmon fin tissue samples were provided by Matthew Casselman (UBC Forest and Conservation Sciences). Gates Creek sockeye salmon fin tissue samples were provided by Steve Latham (Pacific Salmon Commission, Vancouver). Hansen Creek, Lake Aleknagik, Alaska, sockeye salmon fin tissue samples were provided by Shannan May-McNally (UBC Zoology). Additional DNA samples from across O. nerka’s  range  were  obtained from the Beaty Biodiversity  Museum’s  fish  DNA  archive  (Eric  Taylor,  pers.  comm.)  for the analysis of phylogeographic structure (see Table 2.1).   2.2.2 Mitochondrial DNA Qiagen  spin  columns  were  used  to  extract  DNA  using  the  manufacturer’s  protocol.  The  final  elution was in 100 μL  of  buffer  “AE”  supplied  by  the  manufacturer.  I  used  the  polymerase  chain  reaction (PCR) to amplify an approximately 950 base pair (bp) long fragment of the mitochondrial DNA (mtDNA) NADH dehydrogenase-1 (ND1) subunit gene using the primers ND1F  (5’-ACCTCGATGTTGGATCAG-3’)  and  ND1R  (5’-TATTCGGCCAGGAAAAACAG-3’)  and  the  ND2  (approximately  900  bp  long  fragment)  subunit  gene  using  the  primers  ND2F  (5’-AGCACTACCAACGCCTGAC-3’)  and  ND2R  (5’-AACAAGGGCTGGGAGATTTT-3’)  designed by Eric Taylor from sequences in Gharrett et al. (2001) and Churikov et al. (2001). The PCR amplifications were carried out in a final volume of 50 μL using the following reagents (final concentrations): 5 μL 10x New England Biolabs (NEB) ThermoPol Buffer, 4 μL dNTP (5  20 mM), 1 μL of  each  primer  (10  μM),  0.3 μL NEB Taq, 37.2 μL distilled autoclaved water, and 1.5 μL of template DNA (between 50 – 100 ng/μL). The PCR conditions were as follows: initial denaturation 95 °C for 3 minutes, followed by 4 cycles of 95 °C denaturation (30 seconds), 55 °C annealing (30 seconds), 72 °C extension (90 seconds), 32 cycles of 92 °C denaturation (30 seconds), 54 °C annealing (30 seconds), 72 °C extension (90 seconds) and a final extension step at 72 °C for ten minutes. The PCR products were checked for quality and quantity on 1.5% agarose gel stained with SYBR Safe DNA gel stain and viewed under ultraviolet light. The PCR products were purified using Qiagen PCR purification columns and eluted in 20 μL distilled autoclaved water. Purified DNA samples were then sequenced in the reverse direction using ND1R and ND2R primers.   2.2.3 Mitochondrial DNA data analysis Sequences were aligned using the multiple sequence alignment ClustalW in BIOEDIT version 7.2.5 (Hall 1999). I combined the ND1 and ND2 sequences to increase the resolution in detecting distinct haplotypes. I used the software package MEGA version 6 (Tamura et al. 2013) for phylogenetic analysis. The best model for inferring phylogenetic relatedness was estimated using Bayesian Information Criterion (BIC). A maximum likelihood tree was constructed based on the Tamura-Nei sequence evolution model (including a Gamma distribution of among site variation in substitution rate – see Results) and bootstrapped with 2,000 replicates. Oncorhynchus tshawytscha (Chinook salmon) and Oncorhynchus kisutch (coho salmon) were used as outgroups (accession numbers AF392054.1 and EF126369.1, respectively).   21 2.2.4 Microsatellite DNA For a subset of populations (Table 2.2) nine microsatellite loci were assayed to assess the hypothesized independent origin of western and eastern Pacific “black”  kokanee  populations  using a data set independent from mtDNA. Nine microsatellite DNA loci were selected to assay based on clarity of resolution and degree of polymorphism: Omy77 (Morris et al. 1996), Ots100, 103, and 108 (Nelson and Beacham 1999), Oki10 and 29 (Smith et al. 1998), and One103, 108, and 110 (Olsen et al. 2000). The PCRs were carried out using fluorescently dye-labeled forward primers in 20 μL volumes  using  the  Qiagen  PCR  Multiplex  Kit  following  the  manufacturer’s  protocol. Three PCR multiplex groups and one single reaction were performed to increase scoring clarity: 1) Omy77, Ots103 and One103 (annealing temperature: 55 °C), 2) Ots100, Ots108 and Oki29 (57 °C), 3) One108, One110 and Oki10 (57 °C) and 4) Oki29 (58 °C). The PCR products were visualized and evaluated on a Beckman-Coulter CEQ 8000 automated genotyper.   2.2.5 Microsatellite DNA data analysis I used the program MICRO-CHECKER version 2.2.3 (van Oosterhout et al. 2004) to check for evidence of scoring errors and the presence of non-amplifying alleles. Tests for deviations from Hardy-Weinberg Equilibrium for each locus-population combination using an exact test with P values were estimated using a Markov chain method performed in GENEPOP version 4.2 (Raymond and Rousset 1995). Genotypic linkage disequilibrium between all pairs of loci within a population was also tested in GENEPOP using a Markov chain method. I used the microsatellite allele frequency data to construct a phylogenetic tree using the neighbor-joining method (NJ, Saitou and Nei 1987) and Cavalli-Sforza  and  Edwards’  (1967)   22 chord distance (DC) as a measure of genetic distance between populations performed in POPULATIONS version 1.2.31 (Langella 2000) and visualized in TREEVIEW version 1.6.6 (Page 1996). I used DC because it does not make assumptions based on mutation rates or models, only genetic drift (Cavalli-Sforza and Edwards 1967).  The NJ tree was bootstrapped with 5,000 replicates.  I also used the model-based Bayesian clustering program STRUCTURE version 2.3.4 (Pritchard et al. 2000) to visually represent and identify distinct genetic populations to demonstrate population independence on either side of the Pacific Ocean. The STRUCTURE analysis used correlated allele frequencies and employed the admixture model with a burn-in period of 100,000 iterations followed by an additional 200,000 iterations. This was replicated five times with the hypothesized number of genetic groups (K) ranging from one to seven. This range takes into account each population sampled and any potential additional substructure. Preliminary analyses using a K value larger than seven resulted in population structure with very low likelihoods that also were difficult to interpret biologically. The location prior option was used to assist clustering and the basic results did not differ from previous results obtained without using the location prior (Hubisz et al. 2009). Most Oncorhynchus species are thought to have survived the Pleistocene glaciations in one or both of a northern refugium in Beringia (the unglaciated portions of the Yukon River Valley and adjacent areas) and a southern refugium in unglaciated portions of the Columbia River and adjacent areas (Lindsey and McPhail 1986; McPhail and Lindsey 1986). Consequently, the STRUCTURE analysis was performed on a subset  of  the  microsatellite  data  using  Meadow  Creek  as  a  “southern”  refugium  outgroup  and  Hansen  Creek  as  a  “northern”  refugium  outgroup.  The  program  STRUCTURE  HARVESTER  (Earl and vonHoldt 2012) was used to evaluate the most likely number of populations (Pritchard  23 et  al.  2000)  and  incorporated  the  ΔK method of Evanno et al. (2005) to infer which model of population structure (i.e., value of K) is associated with the greatest second-order rate of change of the log probability of the data.   2.3 Results 2.3.1 Mitochondrial DNA After editing the unclear portions of each sequence, I analyzed an 856 base pair (bp) fragment of the ND1 subunit gene from 105 individuals (see Fig. A.1). The ND1 sequences resolved 12 haplotypes that were then sequenced at ND2 in order to increase the power to detect distinct haplotypes and resolve their inter-relationships. The ND2 subunit gene fragment (823 bp after editing) resolved 18 haplotypes in 82 individuals (see Fig. A.2). After combining the aligned fragments, 24 unique haplotypes were resolved in a 1,679 bp fragment from 82 individuals from across O. nerka’s  geographic  range.  The  24 combined ND1/ND2 sequences (Fig. 2.1, Table 2.1) differed from each other by an average of 0.25% sequence divergence. The mean sequence divergence between all O. nerka haplotypes and O. kisutch and O. tshawytscha outgroups was 12.1% and 10.3%, respectively. The best substitution model under BIC to build a maximum likelihood tree was the Tamura-Nei (1993) model including a Gamma distribution of substitution rates (TN93 + G). The Tamura-Nei (1993) model of substitution accounts for differences in substitution rates between nucleotides and the inequality of nucleotide frequencies (Tamura et al. 2013). Using a discrete Gamma distribution (+G) accounts for the substitution rate varying from site to site in a sequence (Tamura et al. 2013).  The combined ND1/ND2 tree depicts the phylogenetic relationships among haplotypes from sockeye salmon, kokanee and “black” kokanee (Fig. 2.1). Japanese “black” kokanee  24 grouped separately from all other forms of O. nerka from across its range (including Anderson and Seton “black” kokanee) at about 0.43% sequence divergence and with 58% bootstrap support (Fig. 2.1). There were no phylogenetic trees (separated ND1 or ND2, see Fig. A.1, A.2) that  grouped  “black”  kokanee  from  the  eastern and western Pacific all together. No haplotypes were shared amongst the “black” kokanee populations from the eastern and western Pacific and their haplotypes differed from each by an average of 0.34% sequence divergence (Table 2.1, Fig. 2.1). In contrast, Anderson Lake and Seton Lake “black” kokanee haplotypes clustered with haplotypes from other populations of O. nerka from across its range in North America and differed only slightly from other O. nerka haplotypes (0.21% sequence divergence, Fig. 2.1). For instance, the most common haplotypes in eastern Pacific O. nerka were also found in Anderson and Seton “black” kokanee e.g., (haplotypes 2 and 17) and these haplotypes were found in individuals ranging from Kamchatka to the Columbia River (Table 2.1, Fig. 2.1). Anderson Lake “black” kokanee sequences included two unique haplotypes (10 and 19, Table 2.1).             25 Table 2.1. Summary of 24 mitochondrial DNA (mtDNA) haplotypes present by location, watershed and sample size in 82 Oncorhynchus nerka (sockeye salmon, kokanee, “black” kokanee) from the combined ND1 and ND2 subunit sequences sequenced. Populations of “black” kokanee are represented with an “*”. BC = British Columbia, AK = Alaska, Y = Yukon.   Location  Watershed  Sample size  Haplotypes Present Klukshu Lake Alsek River, Y 7 14, 17 Katherine Lake Alsek River, Y 3 2 Tazimina River Bristol Bay, AK 3 7, 13 Woodey Island Bristol Bay, AK 5 2, 13, 17 Hansen Creek Bristol Bay, AK 10 2, 12, 13, 18, 23, 24 Okanagan River Columbia River, BC 1 17 Kootenay River Columbia River, BC 2 8, 17 Cowichan Lake Cowichan River, BC 3 11, 16 Anderson Lake* Fraser River, BC 5 2, 10, 17, 19 Seton Lake* Fraser River, BC 2 2 Sinmax Creek Fraser River, BC 1 2 Eagle River Fraser River, BC 2 2 Horsefly River Fraser River, BC 2 11, 17 Birkenhead Lake Fraser River, BC 2 2, 22 Adams River Fraser River, BC 1 11 Little Horsefly River Fraser River, BC 2 17, 21 Quesnel River Fraser River, BC 2 11, 20 Gates Creek Fraser River, BC 2 11 Portage Creek Fraser River, BC 2 2 Lake Saiko* Japan 10 1, 4, 9 Kronotsky Lake Kamchatka, Russia 10 5, 6, 13, 15, 17 Shale Creek Skeena River, BC 1 3 Dust Creek Skeena River, BC 1 2 Takla Lake Skeena River, BC 1 3 Babine Lake Skeena River, BC 2 11        26  Figure 2.1. Maximum likelihood tree using the TN93 + G model of evolution demonstrating the phylogenetic relationships among 24 haplotypes resolved from combing ND1 and ND2 mitochondrial DNA sequences of Oncorhynchus nerka (N=82) from the Columbia River (CR), Cowichan River (CoR), Alsek River (AR), Fraser River (FR), Skeena River (SK), Bristol Bay Alaska (BBA), Kamchatka (K) drainages. Bold haplotypes represent “black” kokanee from Japan (J) and Anderson Lake (AL) and Seton Lake (SL). Number of individuals with each haplotype is in parentheses. The percentage of trees in which the associated haplotypes clustered together is shown next to the branches (percentages less than 50% have been removed) from 2,000 bootstrap replicates. Dash lines represent a 33% reduction in branch length to enhance visualization among ingroup haplotypes.     27 2.3.2 Microsatellite DNA Most locus-population combinations were in Hardy-Weinberg Equilibrium after correcting for multiple comparisons using nine populations (20 out of 81 tests were not) (P < 0.018; Narum 2006). Of these 20 tests, and in all but two populations (Chedukuz Creek and Lake Saiko), seven deviations occurred at the Ots108 locus, showing evidence of heterozygote deficits that are most likely from null alleles detected by the MICRO-CHECKER analysis. Subsequent analyses were conducted with and without this locus. Other departures from Hardy-Weinberg Equilibrium were not concentrated to any specific loci or population. Interestingly, the Ots108 locus was difficult to amplify and genotype in all of the North American populations, but it consistently amplified in the Japanese samples. No genotypic linkage disequilibrium was detected between loci (five out of 324 tests showed significant linkage disequilibrium). Twenty-eight of the 324 tests provided “no  information”, an intractable situation where all rows or all columns marginal sums are 1 (Rousset 2013). Of these 28 tests, 26 were found in the Japanese populations and this is most likely due to a small sample (see Table A.3 and Chapter 3 for more details).  The NJ tree infers the genetic inter-relationships  between  the  “black”  kokanee  on  either  side of the Pacific Ocean along with O. nerka populations from Alaska, the Fraser River and the Columbia River (Fig. 2.2). This tree provides further support to my independent evolution hypothesis  as  Anderson  Lake  and  Seton  Lake  “black”  kokanee  clustered  separately  from  Lake  Saiko  “black”  kokanee  compared  to  sockeye  salmon  populations  within  the  same  watershed  as  well as other Fraser River sockeye salmon and kokanee populations. The same tree was obtained with and without the Ots108 locus.  To visually represent and give additional support to my independent origin hypothesis, I used the program STRUCTURE. Using five populations (Hansen Creek and Meadow Creek  28 populations were used as outgroups) and all nine loci, STRUCTURE returned four as the most likely number of populations (K = 4, Fig. 2.3). This was supported by the log-likelihood method (Table 2.3). The ΔK method supported K = 2 as the most likely number of populations grouping Meadow Creek, Hansen Creek, and Lake Saiko together in each independent replicate. The ΔK method supported K = 4 as the next most likely model. “Black”  kokanee  in  Anderson  Lake  and  Seton Lake grouped together while all other populations grouped separately from the Anderson/Seton cluster (Fig. 2.3). The same STRUCTURE results were produced without using the Ots108 locus with the K = 4 model being supported both by the log-likelihood method and the ΔK method (Table 2.3). There was no result produced by STRUCTURE that grouped the eastern  and  western  “black”  kokanee  populations  together.  Further,  the  STRUCTURE  analysis  was  run  using  the  admixture  model,  which  reflects  the  proportion  of  each  individual’s  genome from each hypothesized genetic group (K).  The  limited  amount  of  admixture  between  “black”  kokanee from the eastern and western Pacific demonstrates that they constitute distinct assemblages of alleles (Fig. 2.2, 2.3). For instance, unique alleles were found in both groups of kokanee: e.g., allele 310 at locus Oki10 in eastern Pacific  “black”  kokanee  and  allele  138  at  locus  Ots103 in western Pacific populations (Table 2.4). These results are consistent with my mtDNA analysis and support my hypothesis that “black”  kokanee  on  either  side  of  the  Pacific  Ocean  have had independent origins.       29 Table 2.2.  Summary of location, drainage, final sample size (after missing data were removed) and  ‘form’  of  Oncorhynchus nerka (kokanee,  sockeye  salmon,  “black”  kokanee) used for microsatellite  analysis.  “*”  represents  populations  used  in  the  STRUCTURE  analysis.  BC  =  British Columbia, AK = Alaska.             Location Drainage Sample size Form Chedakuz Creek Fraser River, BC 29 kokanee Davidson Creek Fraser River, BC 31 kokanee Meadow Creek* Kootenay River, BC 30 kokanee Anderson Lake* Fraser River, BC 80 “black” kokanee Seton Lake* Fraser River, BC 47 “black” kokanee Gates Creek Fraser River, BC 41 sockeye salmon Portage Creek Fraser River, BC 36 sockeye salmon Hansen Creek* Bristol Bay, AK 20 sockeye salmon Lake Saiko* Japan 12 “black” kokanee                  30 Figure 2.2. Neighbor-joining tree constructed using Cavalli-Sforza and Edwards (1967) chord distance (DC) among nine microsatellites loci from nine populations of Oncorhynchus nerka, sockeye salmon (S) and kokanee (K) from the Fraser River drainage (Gates Creek, Portage Creek, Anderson Lake, Seton Lake, Chedakuz Creek, Davidson Creek), Columbia River drainage (Meadow Creek), Bristol Bay Alaska (Hansen Creek) and Lake Saiko (Japan). “‹” represents “black”  kokanee.  Numbers represent percentage of 5,000 bootstrap replicates (percentages less than 50% have been removed).  31  Figure 2.3. The STRUCTURE bar plot of Oncorhynchus nerka (sockeye salmon and kokanee) sampled from five populations (x axis) assayed at nine microsatellite loci from: Meadow Creek (1), Anderson Lake (2), Seton Lake (3), Hansen Creek (4), and Lake Saiko (5). The most likely K value (genetic groups = colours) is four,  grouping  Anderson  and  Seton  lake  “black”  kokanee  into  one  (largely  “red”)  population.  Each vertical line represents an individual fish and the different colours represent the genetic contribution (Q, y axis) of the four genetic groups.    Q  32 Table 2.3. Results of five independent runs using nine microsatellite loci in STRUCTURE testing for the most likely number of genetic groups (K) using the posterior probability (mean LnP(K)) and the Evanno et al. (2005) method ('K).  NA  =  not  applicable.  “*”  represents  independent runs without using the Ots108 locus.         K Mean LnP(K) ' K Mean LnP(K)* 'K*1 -9601.52 NA -8644.64 NA 2 -9305.4 73.623252 -8400.6 33.079668 3 -9203.86 0.605418 -8259.4 2.35082 4 -9051.02 2.914966 -8149.4 73.069696 5 -9259.96 0.905084 -8212.7 0.612666 6 -9049.58 20.16409 -8240.42 0.74373 7 -9273.42 NA  -8326.94 NA     Table  2.4.  Overall  frequency  of  unique  alleles  per  locus  in  “black”  kokanee  found  in  both  Anderson Lake and Seton Lake (Gwenish, Oncorhynchus nerka), in Lake Saiko (kunimasu, Oncorhynchus kawamurae) and alleles found both in Gwenish and kunimasu, but not in the other O. nerka populations sampled. Frequencies calculated from the total number of alleles/locus/population.   Frequency of unique alleles   Gwenish Kunimasu Both Omy77 0.006 0.000 0.019 Ots103 0.011 0.000 0.083 Oki29 0.015 0.000 0.018 Ots108 0.003 0.000 0.009 Ots100 0.013 0.054 0.000 Oki10 0.026 0.003 0.017 One108 0.022 0.009 0.003 One110 0.010 0.000 0.000 One103 0.042 0.000 0.024           33 2.4 Discussion  2.4.1 Polyphyletic  origin  of  “black”  kokanee My results suggest a  polyphyletic  origin  of  “black”  kokanee  from at least two independent episodes of divergence on either side of the Pacific Ocean. This inference is supported by two independent data sets: the relatively high bootstrap support for  distinct  clusters  of  “black”  kokanee in the mitochondrial DNA (mtDNA) tree, the lack of clustering by phenotype in the microsatellite tree and the general lack of alleles shared between the eastern and western North Pacific  “black”  kokanee  populations.  “Black”  kokanee  in  Japan  and  in  the  Anderson/Seton  lakes  system do not share any mtDNA haplotypes and they did not form a monophyletic grouping. Further,  both  data  sets  showed  that  “black”  kokanee  from  Anderson  Lake  and  Seton  Lake  were  genetically more similar to each other and to other Fraser River sockeye salmon and kokanee populations than they are to Japanese  “black”  kokanee.  The STRUCTURE results and neighbor-joining tree also demonstrate the genetic similarity between Anderson Lake and Seton Lake “black”  kokanee  and  their  joint  similarity  to  other  Fraser  River  populations. This suggests that “black”  kokanee  from  the  Anderson/Seton  lake  system  are  more  similar  to  kokanee  and  sockeye  salmon  within  the  same  geographic  area  than  to  “black”  kokanee  from  Japan.  Additionally,  very  few unique alleles were shared amongst Japanese and  Anderson/Seton  “black”  kokanee  populations. A post-glacial  origin  for  “black”  kokanee  is  supported  by  the  lack  of  deep  phylogenetic  structure, low variation and large proportion of low frequency haplotypes (19 out of 24 haplotypes resolved). The overwhelming evidence of post-glacial polyphyletic origin of “regular” kokanee following multiple independent divergences between the forms of O. nerka further supports my findings (Foote et al. 1989; Taylor et al. 1996; Wood and Foote 1996; Taylor  34 et al. 1997; Frazer and Russello 2013). Furthermore, my findings are consistent with post-glacial parallel origins in other species complexes of north temperate freshwater fishes: Salmo trutta (Hindar et al. 1991), Osmerus mordax (Taylor and Bentzen 1993), Coregonus clupeaformis (Pigeon et al. 1997), Gasterosteus aculeatus (McPhail 1984; Schluter 1996) and Oncorhynchus tshawytscha (Waples et al. 2004) Alternatively,  “black”  kokanee  populations  could  have  had  a  single  origin  and  been much more widespread across O. nerka’s  geographic range, but, due to historical events, only two populations persisted. The lack of deep genetic distance, the existence of only two distantly located populations, and the lack of any known geographically intermediate populations despite the wide occurrence of kokanee, makes this alternative unlikely.   2.4.2 Phenotypic divergence “Black”  kokanee show parallel patterns of phenotypic (colour and reproductive ecology) divergence and my results provide evidence of genotypic divergence with both nuclear and mitochondrial DNA. Demonstrating parallel evolution of similar phenotypes by genetic divergence  can  implicate  the  role  of  natural  selection.  Inferences  can  be  made  that  “black”  kokanee populations on either side of the Pacific Ocean experience similar ecological conditions that have promoted parallel patterns of phenotypic and genetic divergence (Bell and Foster 1994; Schluter  and  Nagel  1995).  The  reproductive  habitats  and  behaviour  of  “black”  kokanee  may  result in unique selective regimes that promote parallel evolution of phenotypes as potential adaptations to their distinct reproductive ecology. Therefore, deep water and late spawning behaviour may be accompanied by selection for particular, but as yet unknown traits that have resulted in differentiation from  “regular”  kokanee  and  sockeye  salmon.  Evidence  of  unique   35 colouration and/or behaviour of adaptive significance can be seen in other salmonids: lake whitefish (Coregonus clupeaformis, Pigeon et al. 1997), beach/stream spawning kokanee (O. nerka, Taylor et al. 2000; Frazer and Russello 2013), Atlantic salmon (Salmo salar, Garcia de Leaniz et al. 2007), masu salmon (O. masou, Kano et al. 2010) and was reviewed in Fraser et al. (2011). In Lake Saiko, sympatric stream spawning kokanee are known to be reproductively isolated  from  “black”  kokanee  and  the  two  rarely  hybridize  (Nakabo  et  al.  2011;;  Muto  et  al.  2012). Comparatively, juvenile sockeye salmon reside and feed both in Anderson Lake and Seton Lake prior to migrating to the ocean, but there have been no observations of stream or beach spawning “black”  kokanee  in  these  lakes.  The  unique  reproductive  strategy  of  “black”  kokanee could be the result of competition for stream spawning sites and/or food in both Lake Saiko and the Anderson/Seton lakes system. For instance, the deep water spawning habitats of “black”  kokanee  may  be  associated  with  distinct  thermal  regimes  leading  to  different  times  at  first  feeding  for  sockeye  salmon  and  “black”  kokanee  in  the  same  lake  (Morris  and  Caverly  2004).  In addition, there may be more deep water spawning habitat in relation to stream spawning habitat that could facilitate deep water spawning. Morris and Caverly (2004) suggested that deep spawning would also reduce the risk of offshore alevin (larvae) and fry transport from spring winds that would normally occur in the much shallower stream or beach spawning sites. The deep water  spawning  sites  being  used  by  “black”  kokanee  may  be  associated  with  geothermal and/or groundwater upwelling that provides favorable temperature and oxygen content for spawning and incubation (Morris et al. 2003). Alternatively, it is hypothesized that “black”  kokanee  could  be  locally  adapted  to  spawn  at  cool  water  temperatures  (Nakabo  et  al.  2011). In Lake Saiko, the water temperature below 30 meters is estimated to be less than 5 °C  36 during spawning and temperatures at putative spawning sites in Anderson Lake and Seton Lake are estimated to be around 4 °C (Morris et al. 2003; Nakabo et al. 2011). Comparatively, “regular”  kokanee  and  sockeye  salmon  typically  spawn  in  approximately  9-13 °C water (Burgner 1991; Nakabo et al. 2011).  Decreased light intensity at depth may be causally related to the unusual colour of these “black”  kokanee  populations. Light clarity decreases with increasing depth (Moss 1998) and, as a result of spawning at such great depths; the characteristic nuptial red colouration of O. nerka could potentially be selected against because red-colour attenuates rapidly with depth (Reimchen 1989). For example, Reimchen (1989) found that the loss of red nuptial colouration in male threespine sticklebacks (Gasterosteus aculeatus) and the resulting black colouration in several localities in western North America is associated with low water clarity In addition, decreased visibility to predators would also result from colour loss (Moodie 1972). Given that carotenoids are  energetically  costly  (Martinkappi  et  al.  2009),  and  that  “black”  kokanee  flesh  does  not  appear  to contain an excess amount of carotenoids, any energy associated with carotenoid deposition could be allocated elsewhere. The eggs are a pale orange colour and appear to contain carotenoids  (personal  observation).  Further  research  is  needed  to  test  these  hypotheses  in  “black”  kokanee.   Recognizing  that  these  “black”  kokanee  populations  are  relatively  understudied  and  that  the  habitats  in  which  “black”  kokanee  occur  are  probably  not  unique  to  Anderson  Lake,  Seton  Lake  or  Lake  Saiko,  “black”  kokanee  could  potentially  be  in  other  north temperate lakes. Dunn Lake  north  of  Kamloops,  for  example,  may  also  produce  “black”  kokanee,  but  there  is  no  information regarding their spawning behaviour beyond that the kokanee are dark green to almost black in colour when spawning (Gene Tisdale, pers. comm.).   37 Even if natural selection favours the same phenotype in  populations  of  “black”  kokanee, mutations in different genes or different pathways could have also produced similar phenotypes (Conte et al. 2012; Lemay 2013). Alternatively, random genetic drift could have caused the divergence  of  “black”  kokanee,  although  this  is  unlikely  to  account  for  similar  phenotypes  repeatedly and independently, especially considering the evolutionary history of O. nerka across its geographic range.   2.4.3 Low genetic diversity The low levels of mtDNA sequence variation that I resolved are consistent with previous studies showing low biochemical and molecular variation within O. nerka in comparison to other species of Pacific salmon. For example, Allendorf and Utter (1979) examined allozyme allele frequency variation and found that O. nerka has the lowest levels of heterozygosity compared to other Pacific salmon and trout except coho salmon (O. kisutch). Taylor et al. (1997) amplified a highly variable portion of mtDNA control region in O. nerka and only resolved three haplotypes from across its geographic range. While Neilson et al. (1994) amplified the same portion in rainbow trout (O. mykiss), coho and Chinook salmon (O. tshawytscha) from 22 streams in northern California and resolved nine, five and six haplotypes, respectively.  In another study, Taylor et al. (1996) surveyed allele variation at two minisatellite loci from across O. nerka’s  range  and  found that two alleles at each of the two loci accounted for approximately 95% of the total variation. Churikov et al. (2001) sequenced seven regions of the mitochondrial genome (accounting for 97% of the genome) of three Pacific salmon species, pink (O. gorbuscha) sockeye, and chum (O. keta) salmon and two thirds of the haplotypes resolved in sockeye salmon were restricted to the ND1/ND2 segment. Also, mtDNA nucleotide diversity of sockeye salmon  38 was comparable to that of pink salmon and chum salmon, but haplotype diversity was substantially lower (Churikov et al. 2001).  Japanese kokanee and sockeye salmon populations have substantially lower levels of genetic variability at allozyme (Winans and Urawa 2000) and microsatellite loci (Young et al. 2004, Beachman et al. 2006) compared to North American and Russian populations. Additionally, most Japanese populations sequenced at the cytochrome b segment were monomorphic for a single haplotype (Yamamoto et al. 2011). Although my sample sizes were modest for Japanese kokanee, my results are consistent with low molecular variability in Japanese O. nerka. All 13 ND1 sequences were monomorphic (Fig. A.1). After combining the ND1 sequences with ND2 sequences only two additional haplotypes were resolved. The low genetic variation observed in Japanese O. nerka can probably be explained by the transplantation history (e.g., Tokui 1964; Masaoka et al. 1997) and that Japan is on the southwestern periphery of native O.  nerka’s range (Yamamoto et al. 2011).   2.4.4 Phylogeographic structure There are two major lineages of O. nerka that are hypothesized to have resulted from differentiation and divergence in two glacial refugia, Beringia in the north and Cascadia in the south, after the end of the last glacial period (Varnakskaya et al.1994; Taylor et al. 1996; Allendorf and Seeb 2000; Winans and Urawa 2000; Beacham et al. 2006). Within refugium divergences are often reflected in the current patterns of diversity across the range of a species (Waples et al. 2008). My findings provide some support for a minor phylogeographic break in O. nerka associated with the hypothesized divergence from at least two refugia during the most recent, Wisconsinan, glaciation: a northern, Beringian refuge, and a southern, Cascadian, refuge.  39 This division was observed in the mtDNA tree (at 87% bootstrap support for a small group consisting of Alaskan haplotypes) and the microsatellite DNA NJ tree (at 65% bootstrap support for a grouping linking Alaskan O. nerka and Japanese O. nerka). Similar phylogeographic distinctions resulting from past historical events are found in other species of Pacific salmon (e.g., O. mykiss: Taylor 1995, O. gorbuscha: Churikov and Gharrett 2002, O. tshawytscha: Waples et al. 2004), a variety of other fishes (e.g., Bernatchez and Dodson 1990; Bernatchez and Wilson 1998), and are consistent with our general understanding of phylogeography of the Pacific Basin (e.g., McPhail and Lindsey 1970; Jacobs et al. 2004). This phylogeographic distinction would become clearer with a larger sample size and inclusion of more populations from across O. nerka’s  geographic  range  but that is beyond the scope  of  this  chapter.  I  would  predict  that  Japanese  “black”  kokanee  would  group  with  a  more  northern  group  (such  as  suggested  by  the  microsatellite  tree)  and  Anderson/Seton  “black” kokanee would group with a southern group (Beacham et al. 2006). My results hint at multiple divergences within these phylogeographic lineages, which would also become clearer with a broader survey.  2.4.5 Conclusions It is important to understand how intraspecific diversity originates especially when natural selection is potentially driving the evolution of genetic variation. Similar ecological specializations observed in “black”  kokanee  in Lake Saiko and in the Anderson/Seton lakes system resulted from repeated evolutionary events rather than from a single divergence from the rest of O. nerka. In conclusion, my findings suggest that “black”  kokanee  have  evolved  at  least  twice across the North Pacific Ocean. This chapter has provided evidence from concordance of  40 two independent data sets of the role of parallel evolution in the origin of north temperate fish biodiversity particularly within O. nerka, and has revolved yet another nuance in the evolutionary history of Pacific salmon.             41 Chapter 3: Genetic and morphological differentiation of “black” kokanee (Oncorhynchus nerka) within the Anderson and Seton lakes system  3.1 Introduction  3.1.1 Biocomplexity The greatest challenge and perhaps the most important component in conservation biology is identifying and preserving intraspecific variation (Behnke 1972; Moritz 2002; Wood and Gross 2008). In particular, phenotypic and genetic variation among populations is a critical aspect of conservation (Allendorf and Luikhart 2007). Phenotypic specialization and genetic structure provide information regarding local adaptation and demographic independence of populations, which can influence recovery probability of declining populations (Taylor et al. 2000; Hoffmann and Willi 2009; Suk and Neff 2009). Intraspecific variation can buffer ecosystems from future changes in environmental conditions and lead to long-term sustainability of ecosystem services and functions (Bengtsson et al. 1997; Hilborn et al. 2003; Duffy 2009; Schindler et al. 2010). This  “biocomplexity”  is known to increase stability in populations and ecosystems. For example, Bristol Bay, Alaska, sockeye salmon (Oncorhynchus nerka) exist as numerous distinct and locally adapted populations, and conserving this “biocomplexity” has played an important role in maintaining stability in overall abundance of the population complex in the face of a fluctuating environment as well as providing the basis for a sustainable fishery (Hilborn et al. 2003).   3.1.2 Diversity within Oncorhynchus nerka: sockeye salmon and kokanee Oncorhynchus nerka provides an outstanding example of the broad range of diversity that can be found within a single species. First, O. nerka exists as two common life-history forms: the  42 sockeye salmon or anadromous (sea-going) form and the kokanee or non-anadromous (permanently freshwater-resident) form (Burgner 1991; Quinn 2005). Further, both within sockeye salmon and kokanee, extensive variation exists at multiple spatial and temporal scales, most of which is heritable. For example, variation in growth and maturation (Wood and Foote 1990, 1996), biochemical/molecular traits (Foote et al. 1989; Taylor et al. 1996; Wood and Foote 1996), colouration (Craig and Foote 2001; Craig et al. 2005), swimming ability (Taylor and Foote 1991) and morphology (Burgner 1991; Taylor and Foote 1991; Foote et al. 1999) have been well documented.  A most intriguing aspect of O. nerka is that of its variation in life history and reproductive behaviour that has been well documented (e.g., Wood 1995; Wood et al. 2008). Briefly, sockeye salmon exist as two main ecotypes: the sea/river type and the lake type. Sea/river type sockeye salmon spawn in tributaries with no access to lakes and juveniles feed and grow in side channels for several months to two years before migrating to the ocean where they grow further and initiate maturation before returning to freshwater streams and lake beaches to spawn (Wood 1995). The lake type of sockeye salmon typically spends one to two years in a lake before migrating to the ocean (Burgner 1991; Wood 1995). Lake type sockeye salmon can be further divided into beach-spawners and stream-spawners (Burgner 1991).  Considerable variation exists in kokanee some of which remains relatively understudied. Kokanee can also be divided into beach- and stream-spawning ecotypes (Burgner 1991; Taylor et al. 1997). Stream spawners are similar to sockeye salmon in both morphology and life history. They ascend tributary streams to spawn in gravel substrate and typically females defend spawning territories (Taylor et al. 1997; Shepherd 2000). In contrast, beach spawning kokanee spawn on submerged lakeshore beaches between 15 to 100 cm in depth and later than stream  43 spawners (Shepherd 2000). Beach spawners typically have less pronounced colouration and no mate or territory defense has been observed (Shepherd 2000; Frazer and Russello 2013).    3.1.3 “Black”  kokanee A relatively newly documented and understudied aspect of the variation within kokanee is the phenomenon known as the so-called  “black”  kokanee,  which  are  distinguished  from  “regular”  kokanee by the intense black nuptial colouration compared to the typical red-body/green head colouration. Curiously, to date “black”  kokanee  have  only  been  documented  in  a  single  lake  in  Japan and in two lakes in the Fraser River drainage of BC: Anderson Lake and Seton Lake (see chapter 2 and Fig. 1.3).  Along  with  their  distinct  colour,  some  of  the  spawning  behaviour  of  “black”  kokanee  in  Anderson Lake and Seton Lake appears to be unique not only within O. nerka but within Oncorhynchus generally. They spawn at depths of 20 to 70 meters (typically greater than 50 meters) on small gravel substrate with peak spawning November (Seton Lake) and in January (Anderson Lake) (Morris and Caverly 2004). Substrate size at putative spawning sites varies with depth and typically consists of small gravel but large gravel and finer substrates have also been observed (Morris and Caverly 2004). Large cobble and boulder substrates observed at putative spawning locations in Anderson Lake suggest that some “black”  kokanee  may be broadcast  spawning  (Morris  and  Caverly  2004).  In  addition,  “milling”  or  schooling  behaviour  has  been  observed  in  spawning  “black”  kokanee  Anderson  Lake,  but  remains  unknown  in  Seton  Lake. Deep water spawning is thought to cause swim bladder distension due to rising to the surface too quickly resulting in their unusual post-spawning floating (Morris and Caverly 2004; Stable  2004).  Interestingly,  some  “floaters” are still alive (personal observations).  44 Differences between the populations in both lakes are also observed. The majority of Anderson Lake spawners are three years of age, with a small number being four or five. In contrast, Seton Lake spawners are predominantly two years of age with few being three years (Morris and Caverly 2004). Differences in mean fork length between fish from the lakes correspond to the age differences with Anderson Lake males and female kokanee being larger than Seton Lake males and females (Morris and Caverly 2004). Black colouration in Anderson Lake fish is more striking in comparison to the duller Seton Lake spawners (personal observation, see Figs. 1.4, 1.5).  “Black”  kokanee  play  a  significant  ecological  role  in  maintaining  the  biodiversity  in  the  Anderson/Seton lakes system as  well  as  adding  to  the  productivity  and  complexity  of  the  lakes’  food webs. They provide food for piscivorous fishes such as bull trout (Salvelinus confluentus) and rainbow trout (O. mykiss) and for terrestrial predators such as bald eagle (Haliaeetus leucocephalus) and cougar (Puma concolor)  (Morris  and  Caverly  2004).  “Black”  kokanee  also  likely  play  a  role  in  nutrient  cycling  (Richey  et  al.  1975).  Further,  “black”  kokanee  are  important  socially  and  culturally  to  the  St’át’imc  First  Nation,  which  form  the majority of the population in the  surrounding  communities  in  the  Bridge/Seton  system.  “Black”  kokanee  are  used  as  a  crop  fertilizer,  a  winter  food  source,  and  a  way  to  transfer  traditional  St’át’imc  knowledge  (William  Alexander, pers. comm.).  In order to  better  understand  the  management  needs  of  “black”  kokanee  in  Anderson  Lake and Seton Lake, we need to understand if the two lakes house distinct populations of “black”  kokanee.  If  the  two  lakes’  populations  are  genetically  distinct,  for  instance,  then  this suggests that they might also be demographically distinct and require independent management in terms of habitat protection and harvest levels (see Palsbøll et al. 2007; Lowe and Allendorf  45 2010). Indeed, the differences in size and age at maturity between  Anderson  and  Seton  lakes’  “black”  kokanee  strongly  suggest  that  the  two  populations  are  not  the  same  demographically. In this chapter, I used microsatellite allele variation at nine loci to test the hypothesis that “black” kokanee are genetically distinct from each other and from broadly sympatric sockeye salmon within the Anderson/Seton lakes system. Such information is important to better understand the hierarchical nature of diversity within O. nerka both conceptually, but also in the context  of  ‘designatable  units’  (DUs)  and  Canada’s  Species at Risk Act. A DU is an intraspecific unit  of  diversity  that  is  both  “discrete”  (spatially  or  genetically)  from  other  such  units  and  where  such  discreteness  is  “significant”  to  the  evolutionary  legacy  of  the  species  (COSEWIC  2013;;  Taylor et al. 2013). This could be especially important because sockeye salmon are scheduled for assessment by the Committee on the Status of Endangered Wildlife in Canada in the near future (E. Taylor, pers. comm.). Specifically, I used microsatellite allele frequency variation to estimate the levels of divergence within and among populations. Genetic variation at neutral loci can provide a measure of population independence and are an ideal tool to investigate genetic diversity due to their high mutation rates and high levels of polymorphism (Suk and Neff 2009). This is the first study to quantify the level of genetic variation within the Anderson/Seton lakes system.  Furthermore, I conducted a morphological analysis (standard length, maximum body depth and gill raker counts) to test the extent of phenotypic variation within and between Anderson Lake and Seton Lake “black” kokanee. My morphological analysis provides an additional measure of diversity within a single lake system, variation that potentially reflects local adaptation from differences in spawning environments, or phenotypic plasticity (Nelson 1968; Taylor et al. 1997). Identifying population structure based on molecular and morphological  46 data will help to better understand the nature of O. nerka biocomplexity that exists in Anderson Lake and Seton Lake.   3.2 Methods  3.2.1 Microsatellite DNA Samples used in my microsatellite assay consisted of kokanee/sockeye salmon fin tissue from three lake systems: Tatelkuz Lake, Anderson/Seton lakes (both in the Fraser River drainage system) and Kootenay Lake (Columbia River drainage system) (Fig. 3.1). Tatelkuz Lake (upper Chedakuz Creek N = 30, Davidson Creek N = 31) and Kootenay Lake (Meadow Creek N = 30) kokanee samples were collected in 2012 and supplied by Eric Taylor. Tatelkuz Lake and Kootenay Lake  kokanee  were  used  as  “outgroups”  from  the  Fraser  River  and  Columbia  River,  respectively,  to  place  the  scale  of  molecular  variation  between  “black”  kokanee  and  sockeye  salmon from the Anderson/Seton lakes system within the context of variation between watersheds in the Fraser River and between the Fraser and Columbia rivers. Anderson Lake “black”  kokanee samples were collected in January 2013 by David Levy (Levy Research Services Ltd.) (N = 40) and I collected samples on January 16, 2014 (N = 44). Samples were collected by surveying the beach for post spawning adults and adipose fin samples were taken on site. I collected “black”  kokanee  (post-spawners) from the surface of Seton Lake on November 20, 2013 (N = 48) using dipnets and with the help of Gene Tisdale (Tisdale Environmental Consulting Inc.). Portage Creek sockeye salmon (N = 40) fin tissue was collected in 2013 and supplied by Matthew Casselman (UBC Forest and Conservation Sciences). Gates Creek sockeye salmon (N = 41) fin tissue were collected in 2013 and supplied by Steve Latham (Pacific Salmon Commission) (summarized in Table 3.1). Both Gates Creek and Portage Creek sockeye salmon  47 populations are native to the Anderson/Seton system and both have a history of receiving transplants from other sockeye populations, but only the transplantations into Portage Creek (from the lower Adams River also in the Fraser River drainage) were thought to be at least partially successful (Withler 1982) (see Fig. 1.3). Fin tissue was stored in 95% ethanol for later genetic analysis. The procedures and materials used for microsatellite amplification at the nine loci are outlined in Chapter 2.   Figure 3.1. Map of British Columbia showing the three lake systems: Tatlekuz Lake, Anderson/Seton lakes system in the Fraser River drainage and Kootenay Lake in the Columbia River drainage, used in the microsatellite analysis of sockeye salmon and kokanee (Oncorhynchus nerka).  48 Table 3.1. Summary of location, drainage, final sample size (after missing data was removed) and  ‘form’  of  Oncorhynchus nerka (kokanee,  sockeye  salmon,  “black”  kokanee)  used  for  microsatellite analysis.             Location Drainage Sample size Form Chedakuz Creek Fraser River, BC 29 kokanee Davidson Creek Fraser River, BC 31 kokanee Meadow Creek Kootenay River, BC 30 kokanee Anderson Lake 2013 Fraser River, BC 37 “black” kokanee Anderson Lake 2014 Fraser River, BC 43 “black” kokanee Seton Lake Fraser River, BC 47 “black” kokanee Gates Creek Fraser River, BC 41 sockeye salmon Portage Creek Fraser River, BC 36 sockeye salmon   3.2.2 Microsatellite DNA data analysis I used the program MICRO-CHECKER version 2.2.3 (van Oosterhout et al. 2004) to check for evidence of scoring errors and the presence of non-amplifying alleles (null alleles). Tests for deviations from Hardy-Weinberg Equilibrium for each locus-population combination using an exact test with P values were estimated using a Markov chain method performed in GENEPOP version 4.2 (Raymond and Rousset 1995). Genotypic linkage disequilibrium between all pairs of loci within a population was also performed in GENEPOP using a Markov chain method. To evaluate within population variation, I performed basic population genetic summary statistics (number of alleles per locus, allelic richness, observed heterozygosity and expected heterozygosity) in FSTAT version 2.9.3.2 (Goudet 1995) and GENEPOP. Among population variation (FST) was evaluated using pairwise population differentiation estimated as θ  (Weir  and  Cockerham 1984) in FSTAT. Means and 95% confidence intervals of FST were calculated by jackknifing over populations and significance levels determined in FSTAT were corrected for multiple simultaneous tests as outlined in Narum  49 (2006). Although in most instances, temporal variation is usually minor in comparison to spatial variation in salmonid fishes (e.g., Beacham et al. 2004; Taylor and Yau 2013), I tested for differentiation between temporal samples from Anderson Lake (2013, 2014) before proceeding with subsequent analyses.  Microsatellite variation was partitioned hierarchically by within population, among population  and  among  “black”  kokanee  and  sockeye  salmon  populations  within  the  Anderson/Seton lakes system by using an analysis of molecular variance (AMOVA) performed in GenAlEx version 6.5 (Peakall and Smouse 2012).   I used the model based Bayesian clustering method in STRUCTURE version 2.3.4 (Pritchard et al. 2000) to assess the hypothesized differences between populations of kokanee, sockeye  salmon  and  “black”  kokanee.  The  STRUCTURE  analysis  used  correlated  allele  frequencies and employed the admixture model with a burn-in period of 100,000 iterations followed by an additional 200,000 iterations. This was replicated ten times with K (hypothesized number of genetic populations characterized by a set of allele frequencies at each locus) ranging from one to eight. This range takes into account each population sampled and any potentially additional substructure. Preliminary analyses using a K value larger than eight resulted in population structure with very low likelihoods complicating biological interpretation. I used ten independent runs for each K to ensure results were consistent. The Anderson Lake and Seton Lake  “black”  kokanee populations did not give consistent results using five iterations so I used ten iterations to provide more consistent results. This is potentially caused by the weak differentiation between the two “black” kokanee populations (see below). The analysis was run with and without the location prior option that is used to assist clustering (Hubisz et al. 2009). The STRUCTURE analysis was performed on the entire microsatellite data set as well as with  50 the four populations from the Anderson and Seton lakes system with the same parameter set and K ranging from one to seven. The program STRUCTURE HARVESTER (Earl and vonHoldt 2012)  was  used  to  evaluate  the  most  likely  number  of  populations  and  incorporated  the  ΔK method of Evanno et al. (2005) to infer which model of population structure (i.e., value of K) is associated with the greatest second-order rate of change of the log probability of the data, and therefore most likely.   Discriminant analysis of principal components (DAPC) was used to visualize the variation among populations within the Anderson and Seton lakes system using kokanee from Meadow Creek, upper Chedakuz Creek and Davidson Creek as outgroups. The DAPC analysis is a multivariate clustering method that produces synthetic variables to maximize the among population variation and minimize the within population variation (Jombart et al. 2010). This method makes no assumptions about the underlying population genetic model, Hardy-Weinberg Equilibrium or linkage disequilibrium and it uses an analysis of variance model to quantify between group differentiation (Jombart et al. 2010). The DAPC analysis was conducted in R (R Development Core team 2012) using the ADEGENET package (Jombart et al. 2010).  3.2.3 Morphological analysis “Black”  kokanee  from  Anderson Lake (sampled in 2013 N = 30 and 2014 N = 19) and Seton Lake (sampled in 2013 N = 30) were used for morphological analysis. Carcasses from both lakes were kept on ice until arrival at UBC where they were frozen for subsequent morphological analysis. Samples were thawed overnight and then preserved in 10% formalin for a minimum of four days (Seton Lake samples) to a maximum of one week (Anderson Lake samples owing to their larger size). Swim bladders were deflated to assist in preservation. Samples were then  51 rinsed and soaked in water for 24 hours before being measured and then stored in 45% isopropyl alcohol. Basic morphological measurements included standard length and the distance from the insertion of the pectoral fin to the insertion of the dorsal fin (used as a measure of maximum body depth). Measurements were taken on the left side of the fish and the sex of each fish was recorded. The first gill arch on the left side was removed and stained in 1% potassium hydroxide with Alizarin red for two hours. Gill arches were rinsed and then soaked in distilled water for approximately 12 hours before gill rakers were counted using a dissecting microscope. Samples were then preserved in 45% isopropyl alcohol.    3.2.4 Morphological data analysis Body measurements typically scale positively with overall body size in fishes (e.g., Reist 1985) and thus assessment of differences in maximum body depth in “black” kokanee was required to account for the differences in overall body length between fish from Anderson Lake and Seton Lake. This was accomplished by using an allometric adjustment to standardize body depth to a common standard body length by first using an analysis of covariance (ANCOVA) to test for homogeneity between the slopes of depth versus length (using log10 values) for both lakes (Reist 1985). Then, I used the adjustment equation to standardized depth to standard length: Dadj  Do(L/Lo)b where Dadj is the adjusted body depth, DO is the observed body depth of the individual, L is the common standard length among all samples, LO is the observed standard length of the individual, and b is the allometric coefficient calculated using ANCOVA. Gill raker numbers were not adjusted for differences in body size because there was no significant correlation between  52 standard length and gill raker count within any of the three samples (Anderson Lake 2013, 2014, and Seton Lake 2013, r = 0.004 to 0.116, P > 0.08). Mature salmonids often exhibit significant sexual dimorphism in body shape (Quinn 2005); therefore, two-sample t-tests were performed to test for sexual dimorphism in standard length, size-adjusted maximum body depth and number of gill rakers for each lake. I compared the means of each measurement between Anderson Lake and Seton Lake as well as between sampling years using t-tests for independent samples.  3.3 Results 3.3.1 Within population microsatellite variation 3.3.1.1 Hardy-Weinberg Equilibrium and linkage disequilibrium Only one locus (Ots108) consistently showed evidence of one or more null alleles after analysis with MICRO-CHECKER. This locus also had lower than expected heterozygosity in seven out of the eight populations analyzed. Subsequent analyses were performed with and without Ots108 for comparison. Other loci showed minor evidence for null alleles, but no other locus showed consistent evidence across populations. There was no evidence of large allele dropout or scoring error due to stuttering.  Out of 72 tests for Hardy-Weinberg Equilibrium for each locus population combination, 17 tests departed from expectations of random mating showing statistically significant heterozygote deficiencies after correcting for multiple comparisons for eight populations (P < 0.018; Narum 2006). Of these 17 tests, seven were at Ots108 therefore the deficiencies of heterozygotes are most likely caused by non-amplifying alleles. The remaining ten tests were not concentrated at one specific locus or population. Tests for linkage disequilibrium between loci  53 resulted in statistically significant departures in 11 out of 288 tests but departures were not concentrated on specific locus pairs and were less frequent than expected by chance alone therefore each locus represents an independent measure of genetic variation.  3.3.1.2 Temporal variation Anderson Lake was sampled in 2013 and in 2014. Allele frequencies in temporal samples differed significantly at two of the nine loci assayed (Omy77 P < 0.005 and Oki10 P < 0.05). When comparing allele frequencies between Anderson Lake samples, most differences between years were caused by shifts in the presence or absence of less common alleles, or by an increase in the frequency of common alleles from one year to the next. Combining all nine loci, the temporal samples differed significantly in allele frequencies (FST = 0.0056, P < 0.05), but temporal variation accounted for only 0.68% of the total variation in allele frequencies. Consequently, I combined the temporal Anderson Lake samples for the subsequent analysis.   3.3.1.3 Allelic richness and genetic diversity The mean number of alleles per locus corrected for different sample sizes (allelic richness) ranged from 7.50 (Omy77) to 21.03 (One103) (Table 3.2). Averaged across all nine loci and populations, allelic richness was 13.10 per locus per population. Genetic diversity (expected heterozygosity) averaged across loci ranged from 0.58 (Chedakuz Creek) to 0.89 (Seton Lake), and ranged from 0.67 (Omy77) to 0.93 (Oki10) when averaged across populations. Genetic diversity averaged across all seven populations and loci was 0.81. Chedakuz Creek and Gates Creek consistently exhibited the lowest levels of within population diversity in my analysis (Table 3.2). Genetic diversity averaged across populations within the Anderson/Seton lakes  54 system (excluding the three outgroups) ranged from 0.74 (Oki29) to 0.94 (Oki10 and One103) with the overall average being 0.85.  “Black”  kokanee  in  both  lakes  show  similar  levels  of  within  population variation to each other as well as to Portage Creek sockeye salmon (Table 3.2).   55 Table 3.2.  Summary statistics at nine microsatellite loci in sockeye salmon (S), kokanee (K), and “black” kokanee (B) (Oncorhynchus nerka) from Chedakuz Creek K (CC), Davidson Creek K (DC), Meadow Creek K (MC), Anderson Lake B (2013 and 2014 combined) (AL), Seton Lake B (SL), Gates Creek S (GC), and Portage Creek S (PC).  Number of alleles/locus (A), expected heterozygosity (HE), observed heterozygosity (HO), allelic richness (AR) and sample size (N) for each loci and population. Significant departures from Hardy-Weinberg Equilibrium after correcting for multiple comparisons are represented by * (P < 0.018).   Population Locus  Omy77 Ots103 Oki29 Ots108 Ots100 Oki10 One108 One110 One103 Average CC           A 5 9 5 5 4 11 7 6 7 6.56 HE 0.465 0.829 0.651 0.388 0.333 0.842* 0.755 0.683* 0.729 0.58 HO 0.448 0.655 0.759 0.276 0.345 0.862 0.714 0.448 0.586 0.57 AR 4.897 8.93 5 4.965 3.966 10.93 7 5.999 6.931 6.51 N 29 29 29 29 29 29 28 29 29 28.89            DV           A 5 14 10 5 3 21 11 10 18 10.78 HE 0.781 0.905 0.841 0.719* 0.351* 0.951 0.832* 0.883 0.933 0.8 HO 0.742 0.71 0.839 0.161 0.226 0.935 0.774 0.742 0.839 0.66 AR 5 13.677 9.613 4.992 2.992 20.385 10.701 9.887 17.653 10.54 N 31 31 31 31 31 31 31 31 31 31            MC           A 10 18 11 14 17 25 16 16 22 16.56 HE 0.657 0.951* 0.702 0.862* 0.917 0.94 0.933 0.916* 0.956 0.87 HO 0.633 0.767 0.767 0.414 0.8 0.9 0.8 0.9 0.933 0.77 AR 9.793 17.853 10.6 13.86 16.716 24.056 15.79 15.53 21.516 16.19 N 30 30 30 29 30 30 30 30 30 29.89  56 Population Locus  Omy77 Ots103 Oki29 Ots108 Ots100 Oki10 One108 One110 One103 Average SL           A 9 19 16 16 16 32 14 15 36 19.22 HE 0.759 0.926 0.847 0.903* 0.867 0.963 0.825 0.9 0.971* 0.89 HO 0.674 0.907 0.872 0.333 0.787 0.935 0.778 0.851 0.915 0.78 AR 7.675 16.849 14.37 14.792 14.417 26.285 11.777 13.478 29.605 16.58 N 46 43 47 42 47 46 45 47 47 45.56            AL           A 12 24 23 17 20 38 16 17 54 24.56 HE 0.843 0.933 0.7 0.810* 0.897 0.966* 0.874 0.891 0.968 0.88 HO 0.759 0.872 0.725 0.416 0.835 0.825 0.825 0.887 0.975 0.79 AR 9.178 17.616 15.345 12.784 15.247 27.493 12.162 12.771 32.047 17.18 N 79 78 80 77 79 80 80 80 80 79.22            GC           A 7 13 12 8 12 16 6 9 23 11.78 HE 0.747 0.702 0.701* 0.732* 0.843 0.884 0.776 0.741 0.872 0.78 HO 0.8 0.683 0.585 0.067 0.725 0.829 0.675 0.61 0.775 0.64 AR 6.669 10.891 9.97 7.926 10.113 13.706 5.7 7.604 18.513 10.12 N 40 41 41 30 40 41 40 41 40 39.33            PC           A 10 13 17 12 12 25 14 18 22 15.89 HE 0.852 0.898 0.708 0.787* 0.781 0.951* 0.87 0.928 0.951 0.86 HO 0.806 0.889 0.714 0.241 0.694 0.833 0.778 0.972 1 0.77 AR 9.286 12.311 15.078 11.894 10.861 23.029 13.163 16.795 20.929 14.82 N 36 36 35 29 36 36 36 36 34 34.89  57 3.3.1.4 Unique alleles in “black” kokanee Most unique alleles were found at low frequency and most occurred in Anderson Lake and not in Seton Lake (Table 3.3). The total frequency of unique alleles found in Anderson Lake was 0.0086 and in Seton Lake  was  0.0050.  A  majority  (68.5%)  of  unique  alleles  found  in  the  “black”  kokanee populations were found at Oki10 (frequency = 0.032), Oki29 (0.039), and One103 (0.060). Interestingly, when comparing allele frequencies across all populations, alleles found in high  frequency  in  “black”  kokanee  were  typically  also  shared  with  Portage  Creek  sockeye  salmon rather than Gates Creek sockeye salmon.  Table  3.3.  Overall  frequency  of  unique  alleles  per  locus  in  “black”  kokanee  from  Anderson  Lake  (AL), only Seton Lake  (SL)  or  unique  to  both  populations  of  “black”  kokanee  (Both).  Frequencies calculated from the total number of alleles/locus/population.    Frequency of unique alleles    AL SL Both Omy77 0.004 0.002 0.005 Ots103 0.008 0.002 0.000 Oki29 0.011 0.011 0.017 Ots108 0.008 0.002 0.000 Ots100 0.002 0.000 0.000 Oki10 0.007 0.008 0.017 One108 0.007 0.008 0.002 One110 0.002 0.005 0.003 One103 0.037 0.012 0.011        58 3.3.2 Among population microsatellite variation: 3.3.2.1 FST comparisons  Comparing each locus separately, Ots108 (FST mean = 0.155, range = 0.018 – 0.428), Ots100 (FST = 0.132, range = 0.002 – 0.353) and Omy77 (FST = 0.106, range = 0.011 – 0.417) showed the highest mean levels of divergence when all seven populations were considered together.  The mean level of pairwise divergence (estimated using θ) among all seven populations and all nine loci was 0.080 (95% confidence intervals 0.051 – 0.112) (Table 3.4). Chedakuz Creek and Gate Creek consistently had the highest level of mean pairwise divergence from the other populations, 0.168 (range = 0.135 – 0.200) and 0.106 (range = 0.059 – 0.200), respectively. When only populations from the Anderson/Seton lakes system were included, the mean FST amongst populations from this system was 0.041 (95% confidence intervals 0.029 – 0.055). Gates Creek showed the greatest divergence from all other populations, FST = 0.071 (range = 0.059 – 0.083). All pairwise comparisons between populations, including the outgroup populations (P < 0.014) and without them (P < 0.020), were significant after adjusting for multiple  tests  (Narum  2006)  (Table  3.4).  Anderson  and  Seton  lake  “black”  kokanee  and  Portage  Creek sockeye salmon consistently showed the lowest mean level of divergence from each other (FST  = 0.019 and 0.035, respectively, all P < 0.020).   When I removed the Ots108 locus, the mean level of divergence among all populations was FST = 0.069 (95% confidence intervals 0.046 – 0.096) and all pairwise comparisons were significant after adjusting for multiple comparisons (P < 0.014). When only the Anderson and Seton lakes system populations were compared, the mean level of divergence across the eight loci was FST = 0.036 (95% confidence intervals 0.027 – 0.045) and all pairwise comparisons were significant after adjusting for multiple comparisons (P < 0.020). The program FreeNA  59 (Chapuis and Estoup 2007) was used to estimate FST taking into account null alleles. These results did not change the significance levels of my results using all nine loci or after removing Ots108.  Considering  the  low  level  of  differentiation  between  the  two  “black”  kokanee  populations, I used a G-test, which evaluates the distribution of alleles among populations to test for genetic differentiation and which is typically more powerful than FST-based analyses (Rousset 2013). Three of the nine loci were found to be significantly different between the two populations (Omy77, Oki29, and One108,  P  <  0.05).  Using  all  nine  loci,  the  “black”  kokanee  populations in Anderson and Seton lakes were significantly different from each other (P < 0.05). The results did not change after removing Ots108.   Table 3.4. Pairwise FST comparisons between Anderson/Seton lakes (Fraser River)  “black”  kokanee (*) and sockeye salmon, Tatelkuz Lake (Fraser River) kokanee and Kootenay Lake (Columbia River) kokanee (Oncorhynchus nerka) using nine microsatellite loci (including Ots108). Chedakuz Creek (CC), Davidson Creek (DV), Meadow Creek (MC), Anderson Lake (AL), Seton Lake (SL), Gates Creek (GC), and Portage Creek (PC). All comparisons are significant (P < 0.0137).                  DV MC AL* SL* GC PC CC 0.135 0.197 0.162 0.149 0.200 0.167 DV  0.100 0.061 0.067 0.123 0.054 MC   0.040 0.050 0.100 0.055 AL*    0.012 0.070 0.019 SL*     0.059 0.035 GC      0.083 PC                60 3.3.2.2 Bayesian analysis of population structure The STRUCTURE analysis using all seven populations and nine loci suggested that K = 6 was the most likely number of populations and it grouped  Anderson  Lake  and  Seton  Lake  “black”  kokanee into one population (supported by the log-likelihood method Table 3.5, Fig. 3.2). The 'K method supported K = 2 as the most likely number of populations and K = 6 as the second most likely. A K = 2 highlights the high level differentiation of Gates Creek sockeye salmon and Chedakuz Creek kokanee (Fig. 3.2). Portage Creek sockeye salmon (and to a lesser extent Gates Creek)  showed  signs  of  admixture  with  the  “black”  kokanee  populations,  but  were  suggested  to be separate genetic populations. This is consistent with my FST analysis and DAPC results (see below). The Chedakuz Creek and Davidson Creek kokanee populations (Tatlekuz Lake system) grouped separately but showed a small amount of admixture with one another. As a Columbia River outgroup, Meadow Creek kokanee also formed their own genetic population. Removing the Ots108 locus returned the same results.  When the Anderson and Seton lakes system populations were analyzed alone using all nine loci, a K = 3 was supported by the log-likelihood  method  (grouping  “black”  kokanee  together, Fig. 3.3). The 'K method supported a model of K = 2 (Table 3.5) grouping Portage Creek  sockeye  salmon  and  the  two  “black”  kokanee  populations  together.  This  is  consistent  with the low levels of differentiation between these populations, the clustering observed in the DAPC analysis (see below) and the Gates Creek population being quite distinct. The results were the same with and without Ots108. Neither the log-likelihood method nor the 'K method supported a model of K =  4  associated  with  separation  of  the  two  “black”  kokanee  populations  although  slight distinction is observed in their patterns of admixture (Fig. 3.4).   61 Table 3.5. Results of ten independent runs in STRUCTURE testing for the most likely number of genetic groups (K) using the posterior probability (mean LnP(K)) and the Evanno et al. method ('K).  NA  =  not  applicable.  “*”  represents  independent  runs  using  all  nine  loci  and  all  seven  populations  and  “Š”  represents  independent runs using all nine loci and four populations from the Anderson and Seton lakes system with K ranging from one to seven            K Mean LnP(K)* 'K * Mean LnP(K)Š 'KŠ 1 -14327.43 NA -9859.51 NA 2 -13638.24 31.410751 -9511.12 19.500255 3 -13230.04 5.99219 -9466.71 1.228284 4 -12950.46 16.084129 -9482.21 0.929703 5 -12853.32 0.581451 -9553.79 1.351255 6 -12627.09 18.011541 -9538.12 1.036939 7 -12819.49 1.05612 -9742.43 NA  8 -12788.13 NA  NA NA              62  Figure 3.2. The STRUCTURE bar plot for seven populations (x axis) of Oncorhynchus nerka (sockeye salmon and kokanee) sampled from Chedakuz Creek (1), Davidson Creek (2), Meadow Creek (3), Anderson Lake (4), Seton Lake (5), Gates Creek (6), and Portage Creek (7) and assayed at nine microsatellite DNA loci. The most likely model was K = 6 (supported by log-likelihood), which grouped “black”  kokanee  into  one  genetic  group  (light  blue).  Each vertical line represents an individual fish and the different colours represent the genetic contribution (Q, y axis) of the six genetic groups.         Q Q  63  Figure 3.3. The STRUCTURE bar plot for  “black” kokanee and sockeye salmon (Oncorhynchus nerka) sampled from the Anderson/Seton lakes watershed and assayed at nine microsatellite DNA loci.  “black”  kokanee  from Anderson Lake (1), Seton Lake (2), and sockeye salmon from Gates Creek (3), Portage Creek (4). The most likely model was K = 3 (supported by log-likelihood), which  grouped  “black”  kokanee  into  one  genetic  group (largely green). Relatively low admixture of allele frequencies with the other genetic groups (red and blue) demonstrates the  uniqueness  of  “black”  kokanee  within  the  Anderson/Seton  system,  Fraser  River  drainage. Each vertical line represents an individual fish and the different colours represent the genetic contribution (Q, y axis) of the three genetic groups.     Q  64  Figure 3.4. The STRUCTURE bar plot for  “black” kokanee and sockeye salmon (Oncorhynchus nerka) sampled from the Anderson/Seton lakes watershed and assayed at nine  microsatellite  DNA  loci.  “Black”  kokanee  from Anderson Lake (1), Seton Lake (2), and sockeye salmon from Gates Creek (3), Portage Creek (4). In this case, K = 4 was modeled which reveals some distinction between  the  two  “black”  kokanee  populations  (1,  2).   Q  65 3.3.2.3 DAPC analysis The DAPC analysis using all nine loci highlighted the differences and similarities among the seven populations retaining 95 principal components (PCs) representing 90% of the total variation in the data set across six discriminant functions (Fig. 3.5). The DAPC plots illustrated the distinctiveness of the Chedakuz Creek and Gates Creek populations, and, similarly to the STRUCTURE  results,  “black”  kokanee  clustered  closely  with  Portage  Creek  sockeye  salmon.  Using only Anderson/Seton lakes samples, 68 PCs representing 85% of the total variation across three discriminant functions were retained and this highlighted  the  distinctiveness  of  “black”  kokanee from other O. nerka (Fig. 3.6). The same results were observed after removing the Ots108 locus. Consistent with the FST and STRUCTURE analysis, the Portage Creek sockeye salmon  population  was  genetically  more  similar  to  “black”  kokanee  than  were  Gates  Creek  sockeye  salmon.  The  relatively  close  clustering  of  Anderson  Lake  and  Seton  Lake  “black”  kokanee together within both DAPC plots indicates that they are genetically similar to one another.            66   Figure 3.5. Discriminant analysis of principal components (DAPC) highlighting the genetic variability among populations and minimizing variation within populations among seven Oncorhynchus nerka populations assayed at nine microsatellite loci. Ninety percent of the variation was retained as by 95 principal components. Takelkuz Lake watershed (Fraser River): Chedakuz Creek kokanee (1), Davidson Creek kokanee (2), Kootenay Lake watershed (Columbia River): Meadow Creek kokanee, and Anderson/Seton watershed (Fraser River): (4) Anderson Lake “black” kokanee, (5) Seton Lake “black” kokanee, (6) Gates Creek sockeye salmon and (7) Portage Creek sockeye salmon.      1  2  3  4  5  6  7 1 - Chedakuz Creek2 - Davidson Creek3 - Meadow Creek4 - Anderson Creek5 - Seton Lake6 - Gates Creek7 - Portage Creek 67  1  2  3  4 1 - Anderson Lake2 - Seton Lake3 - Gates Lake4 - Portage Creek                Figure 3.6. Discriminant analysis of principal components (DAPC) highlighting the genetic distinctiveness of “black” kokanee populations from: (1) Anderson Lake and (2) Seton Lake from that of sockeye salmon from: (3) Gates Creek and (4) Portage Creek sockeye salmon (Fraser River drainage), (Oncorhynchus nerka) assayed at nine microsatellite loci. Eighty-five percent of the variation was retained as 68 principal components.  68 3.3.2.4 Hierarchical analysis of genetic structure When the microsatellite variation was partitioned hierarchically using AMOVA and using “form” as an a priori grouping  factor  (“black”  kokanee  or  sockeye  salmon)  in  the  Anderson/Seton lakes system, the majority of variation was found within populations (95.6%, P < 0.001) (Table 3.6). Among populations-within form variation accounted for 4.1% of the total variation  (P  <  0.001).  Between  “black”  kokanee  and  sockeye  salmon  variation  accounted  for  only 0.32% of the total variation, but was significant (P < 0.001). The magnitude or significance of the results did not change when I removed Ots108.  Table 3.6. Microsatellite variation using nine loci partitioned hierarchically using analysis of molecular variance in Anderson and Seton Lake “black” kokanee and sockeye salmon (Oncorhynchus nerka). Among  form  variation  represents  the  difference  between  “black”  kokanee and sockeye salmon.           Source of Variation % variation F-value P-value Within population 95.57 0.044 < 0.001 Among all populations 4.12 0.041 < 0.001 Between forms 0.32 0.003 < 0.001    3.3.3 Morphological differentiation: 3.3.3.1 Within population variation On average, males were larger (standard length and adjusted maximum body depth) than females in both lakes for (Table 3.7). Tests for differences between the sexes within each population (Seton Lake, Anderson Lake 2013, Anderson Lake 2014) for standard length and size-adjusted maximum body were all significant (P < 0.05). Mean gill raker counts did not differ between the sexes in either lake (both P > 0.05).   69 In Anderson Lake (both years combined), maximum body depth was positively correlated with standard length in females (r = 0.826, P < 0.0001) and males (r = 0.571, P = 0.001). In Seton Lake, maximum body depth was positively correlated with standard length in females (r = 0.654, P = 0.0005), but not in males (r = 0.256, P = 0.625).   Table 3.7. Summary of  morphological  variation  between  “black”  kokanee  (Oncorhynchus nerka) sampled from Anderson Lake in 2013 and 2014, and Seton Lake in 2013, separated by sex. All measurements are in mm (mean, SE). Sexually dimorphic traits include: standard length (SL) and size-adjusted maximum body depth (AD, standardized to a mean standard length of 242.2 mm and 191.3 mm for males and females, respectively). Gill rakers (GR) were not sexually dimorphic.  “*”  represents  significant  difference  between  sampling  years  (P  <  0.001).  “Š”  represents significant difference between the lakes (P < 0.001).  Trait Anderson Lake Anderson Lake Seton Lake 2013 2014 2013 Males    N 15 15 6 SL 255.3 (1.95)* 265.4 (1.62)* 152.2 (1.15)Š AD 87.9 (1.41) 85.5 (0.63) 86.5 (1.67)     Females    N 15 4 24 SL 243.8 (2.75)* 252.2 (1.00)* 148.4 (0.73)Š AD 62.4 (0.45) 62.7 (1.12) 62.5 (0.48)     GR 37 (0.32)* 39 (0.39)* 38 (0.51)    3.3.3.2 Between population variation Results from the analysis of covariance (ANCOVA) demonstrated that the slopes of maximum body depth versus length (using log10 values) were not significantly different for males (P = 0.748) or females (P = 0.387). Therefore, the maximum body depth of each sample was standardized to a common body length of 242.3 mm for males and 191.3 mm for females. The adjustment coefficient was calculated using the pooled slope of the log10 maximum body  70 depth versus log10 length regression line using all samples but separated for males (b = 1.200) and females (b = 1.084). Adjusting maximum body depth to a common standard length separately for males and females removed any positive correlation between depth and length in either lake (r < 0.100, P > 0.5) therefore effectively controlling for any effects of body size. Standard length was significantly different between sampling years for males (P = 0.0004) and females (P = 0.010) in Anderson Lake while size-adjusted maximum body depth was not significantly different between years for males (P = 0.129) or females (P = 0.839). Despite some differences between years within Anderson Lake, they were minor compared to differences between lakes so I combined the temporal samples to highlight the striking difference in size of maturity between the lakes. Standard length was significantly different between Anderson Lake (combined years) and Seton Lake both for males and females (P < 0.0001, Fig. 3.7, 3.8) while size-adjusted maximum body depth was not (P > 0.9, Table 3.7). Mean gill raker number was found to be significantly different (P < 0.006) between the temporal samples from Anderson Lake. There was, however, no significant difference in mean gill raker number between Anderson Lake and Seton Lake before, or after, combining the two sampling years in Anderson Lake (P > 0.05). The number of gill rakers ranged from 35-42 in Seton Lake (mean = 38), 35-42 in Anderson Lake 2013 (mean = 37) and 37-42 in Anderson Lake 2014 (mean = 39, Table 3.7).   71               Figure 3.7. Frequency distribution of standard lengths highlighting the differences in standard length (mm) between female kokanee (Oncorhynchus nerka) sampled from Anderson Lake (combined 2013 and 2014 data sets), mean = 245.45 mm (95% confidence intervals ± 4.86 mm) and Seton Lake, 148.40 mm (± 1.51 mm) (P < 0.0001).          Standard Length (mm)Frequency051015Seton LakeAnderson Lake125 140 155 170 185 200 215 230 245 260 275 72 Figure 3.8. Frequency distribution of standard length highlighting the statistically significant differences in standard length (mm) between male kokanee (Oncorhynchus nerka) sampled from Anderson Lake (combined 2013 and 2014 data sets), 260.36 mm (± 3.19 mm) and Seton Lake, 152.24 mm (± 2.95 mm) (P < 0.0001).     3.4 Discussion 3.4.1 Within population genetic variation Levels of within population variation in the Anderson and Seton lakes system were comparable, on average, to other population genetic studies of kokanee and sockeye salmon although comparisons should be used with some caution as different loci were used and each study is unique in terms of sample sizes and geographic scale of study. For example, Taylor et al. (2000) surveyed Okanagan Lake beach and stream spawning kokanee using eight microsatellite loci Standard Length (mm)Frequency01234567Seton LakeAnderson Lake135 150 165 180 195 210 225 240 255 270 285 73 (three of which were used in my analysis) and reported five to 23 alleles per locus and expected heterozygosities ranging from 0.49 to 0.91 (averaged across the six populations studied). In comparison to Anderson Lake and Seton Lake populations, the number of alleles per locus corrected for sample size ranged from 10 to 17 and expected heterozygosities ranged from 0.78 to 0.89. Comparing individual populations in the Okanagan Lake study of similar sample sizes, Anderson Lake and Seton Lake O. nerka had, on average, more alleles per locus. Furthermore, Young et al. (2004) studied 11 populations of O. nerka (sockeye salmon and kokanee) in the Lake Sammamish/Lake Washington basin using nine microsatellite loci; three of the same loci were used in my analysis. They reported relatively high levels of diversity with the mean number of alleles per locus ranging from 5 to 49 and heterozygosities averaged across populations ranging from 0.74 to 0.96 (Young et al. 2004). These comparisons suggest that within population variation between Anderson and Seton lakes O. nerka and that of spawning ecotypes of kokanee (Okanagan Lake system) and sympatric kokanee and sockeye salmon (Lake Sammamish/Lake Washington basin) are comparable across broad geographic areas.   3.4.2 Among population genetic variation  My  results  demonstrate  significant  microsatellite  divergence  between  “black”  kokanee  and  sockeye salmon from Anderson Lake and Seton Lake. Consistent differences were resolved using FST, Bayesian analysis of population structure and in discriminant analysis of principal components (DAPC). Comparing levels of population differentiation between studies should be interpreted with some caution because of the different markers used, the number of populations surveyed and the geographic scale of the study. Nevertheless, sympatric populations of kokanee and sockeye salmon are known to be genetically distinct even when they spawn at the same time  74 and in the same stream. For example, Foote et al. (1989) found allele frequency differences between sympatric kokanee and sockeye salmon in Takla, Babine and Shuswap lakes in British Columbia. Similar results were reported by Taylor et al. (1996) in the same three lakes as well as in other sympatric populations ranging from Kamchatka to the Columbia River using minisatellite loci. Wood and Foote (1996) studied variation at 15 allozyme loci and found that 18% of variation was due to differences between kokanee and sockeye salmon spawning at the same place and time in Takla Lake compared to 0.5% variation within forms between tributaries.  Young et al. (2004) found significant genetic divergence (0.071, measured as coancestry distances, i.e., d = -ln(1-θ); Weir 1996) using nine microsatellite loci between sympatric Issaquah Creek (Lake Sammamish basin, Washington State) kokanee and sockeye salmon. Differentiation may be the result of different spawning times between kokanee (early August) and sockeye salmon (late October) (Young et al. 2004). Interestingly, this level of differentiation is  comparable  to  the  level  of  differentiation  between  Anderson  and  Seton  lakes  “black”  kokanee  and Gates Creek sockeye salmon when measured as coancestry distances (0.072 and 0.061, respectively).  Levels of differentiation between sympatric forms of O. nerka may represent a widespread phenomenon within the species that have only recently become apparent with current advances in genetic analysis. Frazer and Russello (2013) analyzed differentiation between sympatric beach and stream spawning kokanee in five British Columbia lakes (Wood, Kootenay, Okanagan, Duncan, and Tcheskinut lakes) using 15 neutral loci and 15 expressed sequence tag (EST) linked loci that exhibit signatures of directional selection. Using only neutral loci, significant levels of differentiation were found in Okanagan, Kootenay, and Wood lakes (FST = 0.005, 0.008, and 0.032, respectively) (Frazer and Russello 2013). In contrast, divergence  75 between beach and stream spawning kokanee in two other lakes was found to be virtually non-existent (FST = -0.001, and 0.000 in Duncan and Tcheskinut lakes, respectively, Frazer and Russello 2013). Using lake-specific EST linked loci, levels of divergence were found to be much greater between beach and stream-spawning kokanee in each lake: Duncan FST = 0.017, Kootenay FST = 0.071, Okanagan FST = 0.081, Wood FST = 0.126, and Tcheskinut FST = 0.074 (Fraser and Russello 2013). Using loci that may be influenced by natural selection can be effective for management when neutral loci demonstrate modest divergence as a result of recent divergence or weak population structure (Russello et al. 2011). Levels of neutral genetic differentiation  between  sympatric  “black”  kokanee  and  sockeye  salmon  in  the  Anderson/Seton  lakes system are generally comparable (and higher than in some cases) to differences between sympatric beach and stream spawning kokanee (cf. Frazer and Russello 2013). In addition, Taylor  et  al.’s  (2000)  microsatellite  analysis  of  beach  and  stream  spawning  kokanee  in  Okanagan  Lake reported an FST = 0.018 as the level of population differentiation which is slightly lower than  I  observed  between  sympatric  sockeye  salmon  and  “black”  kokanee  in  the  Anderson/Seton  lakes system (FST = 0.023). Two of the three outgroups used in my analysis were chosen based on their relative distinctiveness within their respective watersheds; providing an appropriate comparison to evaluate  the  distinctiveness  of  “black”  kokanee.  Kokanee  from  upper  Chedakuz  Creek  were  found to be highly distinct within the Tatelkuz Lake system (Taylor 2013). Upper Chedakuz Creek kokanee had an average level of divergence of FST = 0.148 when compared to seven other kokanee populations within Tatelkuz Lake system using six of the same loci used in my study compared to an average level of differentiation among the other kokanee populations of from FST = 0.0517 to 0.0685 (Taylor 2013). Taylor (2013) suggested that the high divergence of the upper  76 Chedakuz  Creek’s  kokanee  was  related  to  greater  average  distance  from  the  other  populations in the watershed and because it was part of a different, but interconnected, lake system (Kuyakuz Lake). Upper Chedakuz Creek kokanee were also larger than the other populations of kokanee, which highlights their distinctiveness within the watershed (Taylor 2013).  Kokanee from Meadow Creek of the West Arm of Kootenay Lake (Columbia River system) are also known to be distinct, with an average level of divergence of FST = 0.018, when compared to four other Kootenay Lake populations, two of which were stocked with Meadow Creek  kokanee  (Anders  et  al.  2007).  Meadow  Creek  kokanee’s  relatively  high  level  of  divergence in the Kootenay Lake system may be related to its distant location at the head of the north arm of the lake (Anders et al. 2007). Both Chedakuz Creek and Meadow Creek kokanee are distinct within their respective watershed, some of which can be attributed, at least in part, to their relatively isolated locations. These outgroups provide context to the levels of differentiation found  in  “black” kokanee.  Within the Anderson/Seton lake system, the  average  level  of  divergence  of  “black” kokanee from Portage Creek and Gates Creek sockeye salmon was 0.022 and 0.062, respectively. Given the level of differentiation between Portage Creek and Gates Creek sockeye salmon  and  “black”  kokanee,  temporal  and  spatial  segregation  could  account  for  the  observed  divergence between these ecotypes of O. nerka. For instance, spawning timing and location are both known to cause genetic differentiation between sympatric populations of O. nerka (Taylor et al. 2000; Fillative et al. 2003; Lin et al. 2008; Muto et al. 2012; Frazer and Russello 2013) and other species within Oncorhynchus (McGregor et al. 1998; Hendry et al. 2002). In Anderson and Seton lakes, both sockeye salmon populations are stream spawners with different spawning times both  from  each  other  and  “black”  kokanee.  Portage  Creek  is  a  late  run  population  with  peak   77 spawning occurring in September to November (Beacham et al. 2004). Gates Creek is an early summer run population and peak spawning typically occurs in July to September (Beacham et al. 2004).  The  high  level  of  distinctiveness  of  “black”  kokanee  from  sympatric  sockeye  salmon  in  the Anderson/Seton lakes system is consistent with mixed stock fisheries analysis using 14 microsatellites and four single nucleotide polymorphism loci (Steve Latham, Pacific Salmon Commission, Vancouver,  pers.  comm.).  Here,  the  “black”  kokanee samples (which do not occur in the baseline samples from the Fraser River) were more similar to sockeye salmon from outside the Anderson/Seton lakes system (e.g., early Stuart River, Harrison River) than to sympatric sockeye salmon. The  “mosaic”  population  structure  of O. nerka, whereby neighboring lake populations are not always the most similar genetically to one another (Wood 1994, 1995; Beacham et al. 2004), could explain this pattern (Wood 1995; Withler et al. 2000).  Consistent with my analysis, several studies (e.g., Withler et al. 2000, Beacham et al. 2004, 2006) have found Gates Creek sockeye salmon to be quite distinct not only from other Fraser River populations, but also those throughout its range. Withler et al. (2000) reported that Gates Creek sockeye salmon had fewer alleles per locus than other Fraser River populations and a low proportion of rare alleles. Additionally, most pairwise FST comparisons between Gates Creek sockeye salmon populations and other Fraser River sockeye salmon populations were greater than 0.100 (Withler et al. 2000). Withler (1982) attributed the low diversity of Gates Creek sockeye salmon and their atypical allele frequencies to founder events and/or random genetic drift as a result of a small population size. Another possible explanation, although unlikely, could be the effect of previous transplantation of over 15 million eggs and juveniles from the Birkenhead River (lower Fraser River) to Gates Creek between 1915-1930, but there is  78 no evidence to support that this transplantation effort was successful or resulted in introgression (Withler 1982; Withler et al. 2000).   Portage Creek has a more successful history of transplantation. In 1950, 300,000 eyed eggs were transplanted from the lower Adams River (Thompson River drainage) to Portage Creek and 193,000 juveniles were transplanted into Anderson Lake (Aro 1979; Withler et al. 2000). Prior to the transplant, there was a small, native population in Portage Creek. While there is no evidence for substantial introgression, there is a high degree of genetic similarity between Portage Creek sockeye salmon and the lower Adams River reflecting some level of gene flow following these transplantations (Withler et al. 2000).  Interestingly and consistent with both the FST, DAPC and STRUCTURE analyses, “black”  kokanee  within  both  Anderson  and  Seton  lakes  tended  to  be  more  similar  to  the  Portage  Creek sockeye salmon rather than the sockeye salmon from Gates Creek (Table 3.4; Figs. 3.3, 3.6).  The  observed  level  of  admixture  between  the  “black”  kokanee  populations  and  Portage  Creek sockeye salmon may reflect some level of gene flow between the populations, but it is unknown whether this is reflective of pre-transplantation conditions or some effect of transplantation of sockeye from other systems (see above). Alternatively, the observation that Gates Creek sockeye salmon are so highly differentiated from most Fraser River populations could  result  in  “black”  kokanee  grouping  with  Portage  Creek  sockeye  salmon  more  as  a  consequence of the highly distinct nature of Gates Creek sockeye salmon than any real affinity with Portage Creek sockeye salmon per se.  Regardless,  “black”  kokanee  as  a  group  are  significantly distinct from sympatric sockeye salmon, as revealed by the hierarchical analysis of genetic structure.    79 3.4.3 Anderson  and  Seton  lakes  “black”  kokanee For  most  loci,  Anderson  Lake  “black”  kokanee  had  a  higher  number  of  unique  alleles  compared  to  Seton  Lake  “black”  kokanee  (Table  3.3). This is most likely due to the large sample size after combining temporal samples. Both populations did share a relatively high frequency of alleles that were not found in the other O. nerka populations sampled, especially at Oki10, Oki29, and One103 (Table 3.3).  The  level  of  differentiation  between  “black”  kokanee  populations  was  modest,  but  statistically significant using all nine loci and without Ots108. This result was consistent across all analyses including the DAPC and STRUCTURE analyses that most clearly demonstrated their genetic similarity to one another (Fig. 3.4, 3.6). Additionally, using genetic diversity as a test statistic (G-test)  showed  that  the  two  “black”  kokanee  populations  are  significantly  different  at three loci (Omy77, Oki29, and One108) and when all loci are considered together. These modest levels of differentiation are comparable (and higher in some instances) to the levels of differentiation found using neutral loci between beach and stream spawning kokanee (see above, Fraser and Russello 2013). These modest levels of differentiation may result from a recent, post-glacial  time  frame  for  divergence,  recent  and/or  historical  gene  flow  between  “black”  kokanee  populations or high effective population sizes. For instance, O. nerka colonized the Anderson and Seton lakes system about 11,000 years ago and adult population size were recently estimated at 5,000 in Seton Lake and 150,000 in Anderson Lake (Wood 1995; Morris and Caverly 2004).   The relatively weak, but significant differentiation that was  observed  between  the  “black”  kokanee populations in Anderson Lake and Seton Lake could be the product of different spawning locations and times. The peak spawning time in Anderson Lake is the beginning to the middle of January and in Seton Lake it is the beginning to the middle of November. Temporal  80 and spatial isolation are well-documented processes by which salmonid populations can diverge from one another (e.g., Hendry et al. 1995; Hendry and Day 2005)    3.4.4 Morphological variation Perhaps one of the more fascinating results was the substantial size difference between mature “black”  kokanee  in  the  two  lakes.  A  large  proportion  of  Anderson  Lake  “black”  kokanee  mature  at ages three and four years, which contrasts with that in Seton Lake.  In  Seton  Lake,  “black”  kokanee mature predominately at age two with a smaller proportion at age three. When the fish are compared at a common age class of three, Anderson Lake fish were still found to be larger (218.22 mm fork length in Anderson Lake compared to 198.11 mm fork length in Seton Lake) (Morris  and  Caverly  2004).  I  also  found  Anderson  Lake  “black”  kokanee  were  larger  than  Seton  Lake  “black”  kokanee  (254.61  mm  and  149.16  mm,  respectively),  but  I  did  not  compare  sizes  at  specific age classes.   A previous survey of Anderson  Lake  and  Seton  Lake  “black”  kokanee  (Morris  and  Caverly 2004) that was conducted in the pre-spawning season (early June) found that Seton Lake fish were larger than Anderson Lake fish (189.00 mm mean fork length and 143.25 mm, respectively) when samples of the same age class were compared (age three). By contrast, Morris and  Caverly  (2004)  later  surveyed  “black”  kokanee  after  spawning  and  reported  that  Anderson  Lake fish were larger than Seton Lake fish when the same age class was compared (see above). Potentially, this could suggest that the age-specific growth rates vary between the two lakes. Life history theory predicts that there are fitness related trade-offs between growth, reproduction and survival (Fisher 1930; Hirshfield and Tinkle 1975; Roff 1984). Furthermore, increased adult survival favours reduced reproductive effort and delayed maturation while increased juvenile  81 growth rate favours increased reproductive effort and early maturation (Hutchings 1993). Therefore, different growth rates between  juvenile  and  adult  “black”  kokanee  in  both  lakes  may  account for the observed age-specific size differences and age of maturity. Hutchings (1993) suggested that there might be age-specific variation in growth rate and survival among three populations of brook trout (Salvelinus frontinalis). Comparisons between Anderson and Seton lake “black”  kokanee  populations  in  terms  of  age  of  maturity  and  growth  rates  combined  with  information on age-specific survival and age-specific growth rates would help to elucidate the basis for the differences in age of maturity (Roff 1984; Hutchings 1993). Differences  in  standard  length  (and  associated  age  of  maturation)  of  adult  “black”  kokanee spawners could potentially reflect adaptations to different spawning environments (Blair et al. 1993; reviewed by Taylor 1991 and Fraser et al. 2011). Seton Lake is turbid from the inflow of glacial water from the Bridge River and Carpenter Reservoir, but Anderson Lake is very clear. Additionally, Anderson Lake has a higher plankton density than Seton Lake (2,622 mg dry mass/m2 and 422 mg dry mass/m2, respectively) (Shortreed et al. 2001). Interestingly, Geen and Andrews (1961) and Shortreed et al. (2001) both found that a variable, but substantial proportion of Gates Creek sockeye salmon fry migrate straight through Anderson Lake to reside and feed in Seton Lake. For example in 2000, densities of juvenile sockeye salmon and kokanee in Anderson Lake were estimated, using hydroacoustic and trawl surveys, at 1,057 fish per hectare and otolith strontium to calcium ratios (used to infer seawater entry) indicated that less than 5% of juvenile O. nerka sampled in Anderson Lake were sockeye salmon. In comparison, Seton Lake juvenile densities were estimated at 289 fish per hectare (Shortreed et al. 2001). The lower estimated population density of adult Seton  Lake  “black”  kokanee (only about 5,000 adults versus about 150,000 in Anderson Lake) could explain the lower juvenile densities and result in  82 lower competition for food in Seton Lake. Additionally, the Carpenter Reservoir and Bridge River diversions introduce sediment that may accumulate on the lake bottom, reducing the spawning  habitat  quality  for  “black”  kokanee  in Seton Lake (Morris et al. 2003).  The increased turbidity in Seton Lake probably decreases the visibility in the water column and thus the feeding efficiency of piscivorous fishes, particularly relative to that of planktivorous fishes like O. nerka (De Robertis et al. 2003). In Anderson Lake and Seton Lake, bull  trout  prey  on  “black”  kokanee  and  juvenile  sockeye  salmon;;  therefore,  the  increased  turbidity in Seton Lake may be beneficial for juvenile sockeye salmon because they will be less visible to predators, but their feeding rate will not be substantially affected (Morris and Caverly 2004; De Robertis et al. 2003). All together, the different biotic and abiotic conditions in Seton Lake could explain why a greater proportion of sockeye salmon juveniles feed in Seton Lake compared to Anderson Lake. Consequently, the younger age and smaller size at maturity in of “black”  kokanee  in  Seton  Lake  may  be  attributed  to  the  increased  competition  from  sockeye  salmon within the lake especially given the lower plankton biomass. Additional studies are needed to test these hypotheses.   Gill raker counts did not differ significantly between Anderson Lake and Seton Lake “black”  kokanee.  Gill  raker  counts  are  known  to  occasionally  vary  among kokanee populations within the same lake (e.g., Vernon 1957; Kurenkov 1977, but see Taylor et al. 1997). The lack of differences observed in my results suggests that the feeding ecology (size and composition of prey  items)  is  similar  between  the  populations  of  “black”  kokanee  in  the  two  lakes.    By  contrast,  Kurenkov  (1977)  reported  mean  gill  raker  counts  of  “few-rakered”  kokanee  (mean  gill  rakers  =  32)  and  “many-rakered”  kokanee  (mean  gill  rakers  =  43)  in  Lake  Kronotsky.  These  differences  in gill raker counts were associated with dramatically different diets of the two morphs of  83 kokanee; the “many-rakered”  kokanee fed on plankton while the “few-rakered”  kokanee fed on macrobenthos (Kurenkov 1977). The degree of colouration also varied between the two lakes. Anderson Lake fish exhibit a dramatic, uniform black colour at spawning and, unlike the Seton Lake fish, this colour did not fade after spawning. Seton Lake fish were lighter black to olive in colour, which faded to almost white on the ventral side after spawning. These differences in colour observed between the two lakes could result from water colour/transparency. Seton Lake is glacially turbid whereas Anderson  Lake  is  very  clear.  Glacial  turbidity  can  be  associated  with  “brighter”  coloured  fish  (e.g.  Northrup  et  al.  2010).  Further,  the  jaws  and  caudal  fins  of  Seton  Lake  “black”  kokanee  were  worn out or even missing, but Anderson Lake samples rarely had worn out caudal fins suggesting that the spawning substrate may be composed of large gravel or the spawners are removing sediment. The differences in physical damage from spawning could result from differences in spawning substrates and/or that broadcast spawning may occur in Anderson Lake, but not in Seton Lake (Morris and Caverly 2004).  3.4.5 Comparisons  with  the  Japanese  “black”  kokanee  (kunimasu) In Japan, stream-spawning kokanee population structure is characterized by low genetic differentiation with average levels of FST ranging from  -0.008 to 0.032 (Kogura et al. 2011). This low level of divergence is most likely due to intensive transplantation history and resultant mixing of diverse gene pools and perhaps to ancestral populations founded by a common gene pool because they are on the southern periphery of O. nerka’s  geographic  range  in  the  western  Pacific (Kogura et al. 2011). In Japan, kokanee populations are native to Lake Akan and Lake Shikotsu (on Hokkaido), but have been transplanted to more than 60 lakes with populations  84 being established in 22 lakes, including Lake Saiko (Yamamoto et al. 2011). Native spawning populations of sockeye salmon do not occur in Japan; the Abira River sockeye salmon population was derived from kokanee in Lake Shikotsu (Burgner 1991; Nakabo et al. 2011).  By  contrast,  striking  pairwise  differentiation  was  observed  between  sympatric  “black”  kokanee and stream spawning kokanee in Lake Saiko, Japan (FST =  0.134)  and  between  “black”  kokanee and stream spawning kokanee in Lake Akan (FST = 0.142) (Nakabo et al. 2011). In comparison, stream spawning kokanee in Lake Akan and Lake Saiko are dramatically less differentiated from each other (FST = -0.0005) (Nakabo et al. 2011). While there is a strong level of  divergence  between  “regular”  kokanee  and  “black”  kokanee  in  Japan,  especially  considering  the low genetic diversity in other Japanese populations, their differentiation is consistent with levels of differentiation among some populations of sympatric and allopatric North American kokanee or sockeye salmon populations (Varnakskaya et al.1994; Beacham et al. 2004; Nakabo et  al.  2011).  By  contrast,  my  results  suggest  that  “black”  kokanee  in  the  Anderson/Seton  lake  system have weaker differentiation from sympatric sockeye salmon populations than that reported  between  Japanese  “black”  kokanee  and  sympatric  “regular”  kokanee  (Nakabo  et  al.  2011). Although it is difficult to make direct comparisons owing to the different loci used in each of the studies, the apparently lower differentiation observed in the Anderson/Seton system may be due to differences in effective population sizes within each system resulting in lower genetic diversity or to the fact that some of the Japanese populations involve introduced populations. Additionally,  the  Anderson  Lake  and  Seton  Lake  “black”  kokanee  could  represent  a  more  recent  divergence  compared  to  Japanese  “black”  kokanee.    Interestingly, the phenotypes  of  North  American  and  Japanese  “black”  kokanee  tended  to  be quite  similar.  The  standard  length  of  Japanese  mature  “black”  kokanee  males  and  females   85 ranged from 178.0 mm to 268.7 mm and 183.2 mm to 235.6 mm, respectively (Nakabo et al. 2014). These lengths are more comparable to Anderson Lake (males: 241.7 mm to 278.8 mm, females: 226.3 mm to 261.8  mm)  than  to  Seton  Lake  “black”  kokanee  (males:  147.5  mm  to  155.3  mm,  females:  140.7  mm  to  155.9  mm).    Japanese  “black”  kokanee  gill  raker  counts  ranged  from 36 to 45 (mean 39, Nakabo et al. 2014) compared to 34 to 42 gill rakers for Anderson and Seton  lakes  “black”  kokanee.  At  maturity  in  both  sexes,  colouration  was  described  as  olive  green  to black when alive and becoming darker when dead (see Fig. 1.4 to 1.6; Nakabo et al. 2011, 2014). These phenotypic similarities provide further evidence to support the hypothesis that natural selection may be promoting parallel evolution  of  “black”  kokanee  phenotypes  (see  Chapter 2).  3.4.6 Conclusions In  summary,  my  findings  highlight  the  genetic  and  morphological  distinction  of  “black”  kokanee  in Anderson Lake and Seton Lake from sympatric sockeye salmon, from O. nerka in other areas and, to some extent, from each other. By contrast, the genetic differences observed between the Anderson  Lake  and  Seton  Lake  “black”  kokanee  populations  are  relatively  modest  compared  to  the striking differences in age and size at maturity.  These results have several management and conservation implications. First, my data supports the  potential  classification  of  “black”  kokanee  as  a  distinct  designatable  unit  (DU)  within O. nerka under  Canada’s  Species at Risk Act: (i) they are genetically distinct from sympatric sockeye salmon; (ii) their unusual reproductive ecology can reasonably be inferred to be the result of local adaptation; and (iii) Anderson Lake and Seton Lake are the only two known lakes  in  Canada  that  produce  “black”  kokanee  (see  Chapter  4).  Furthermore,  the  modest,  yet   86 significant, genetic differentiation between  Anderson  Lake  and  Seton  Lake  “black”  kokanee  in  addition to their differences in size and age at maturity and spawning time suggests that the two populations are also demographically independent and that they might qualify as separate DUs. Overall,  “black”  kokanee  exemplify  novel  diversity  found  within  a  taxon  that  is  already  renowned for spectacular levels of intraspecific diversity of evolutionary and conservation significance (e.g., Ricker 1972; Hendry 2001; Hilborn et al. 2003).   87 Chapter 4: General Discussion  4.1 Summary of results My thesis addresses some important questions regarding the unknown evolutionary origin of “black”  kokanee  and  the  genetic  and  phenotypic  diversity  they  represent  within  the  Anderson/Seton lakes system and within Oncorhynchus nerka. My thesis has resulted in three general findings:   1) “Black”  kokanee in Anderson Lake, Seton Lake and Lake Saiko have a polyphyletic origin caused by repeated episodes of parallel evolution in the eastern and western North Pacific Ocean. This is supported by the polyphyletic relationship of mitochondrial DNA haplotypes and the lack of microsatellite admixture between eastern and western North Pacific  “black”  kokanee  at  nine  loci. The independent origin of “black” kokanee populations reinforces the phenomenon of parallel evolution in the history of north temperate fishes (e.g., threespine stickleback, sockeye and kokanee, whitefish pairs; see reviews by Schluter 1996 and Taylor 1999). 2) “Black”  kokanee in Anderson Lake and Seton Lake are genetically distinct from broadly sympatric populations of sockeye salmon that spawn in Gates Creek and Portage Creek within the Anderson/Seton system, and also from other populations of Fraser River and Columbia River kokanee. The genetic differences within the Anderson/Seton system between  “black”  kokanee  and  sockeye  salmon  populations are most likely the result of spatial and temporal differences in spawning.   88 3) The striking morphological distinction represented in standard length at maturity between the two populations of “black” kokanee in the Anderson/Seton lakes system contrasts with their modest, but significant, genetic differentiation. Differences in sizes at maturity may be the result of different levels of competition, predation, turbidity and/or plankton levels between the lakes in addition to potential life history trade-offs. There were no differences observed in gill raker numbers, suggesting that  the  two  populations  of  “black”  kokanee have broadly similar feeding ecology. Further, the black colouration is also uniquely different between the two lakes perhaps as a result of local adaptation.  4.2 Conservation implications Anderson Lake and Seton Lake are the only known lakes in North America that produce “black”  kokanee  (also  known  locally  as  “Gwenish”).  Historically, Gwenish were extremely numerous; local residents recall seeing Anderson and Seton lake beaches with piles of post-spawning Gwenish (William Alexander, local Tsalalh resident, pers. comm.). This is, unfortunately, no longer the case. Local residents hypothesized that the reduction in abundance of Gwenish has been driven by the effects of increased sedimentation from the Seton Dam project, which is thought to be affecting their spawning habitat (William Alexander, pers. comm.). Effects of overfishing, hydroelectrical development, climate change, and other anthropogenic habitat changes may also have contributed to population declines in the Anderson/Seton lakes system (Levy 2012).  The  apparent  decline  of  Anderson/Seton  “black”  kokanee  combined  with  the  results  of  my thesis may have important implications for species at risk assessment in Canada. For instance, my data could be used to assess whether “black” kokanee in Anderson Lake and Seton  89 Lake qualify as distinct designatable units (DU) within O. nerka under the Species at Risk Act and as per The Committee for the Status of Endangered Wildlife in Canada’s (COSEWIC) guidelines for recognizing conservation units within taxa (COSEWIC 2013). Designatable units are  defined  as  “species, subspecies, variety, or geographically or genetically distinct population that may be assessed by COSEWIC, where such units are both discrete and evolutionarily significant”  (COSEWIC 2013). “Discrete” can be defined as occupying a distinct biogeographic region, being genetically distinct, or having a major range disjunction. In  the  context  of  “black”  kokanee,  my  results  have  provided  evidence  to  support  the  “discreteness”  of  these  unique  populations.  “Black”  kokanee  are  genetically  differentiated  from  other  populations  of  sockeye  salmon and kokanee across their range and they exhibit unique reproductive ecology and colour, all  of  which  can  be  used  to  help  define  them  as  “discrete.”   Defining  a  population  as  “evolutionarily  significant” includes evidence for some combination of: deep phylogeographic divisions, occupying a unique habitat such that local adaptations  are  likely  to  have  arisen,  being  the  only  surviving  assemblage  within  the  taxon’s  natural range and/or if the DU went extinct it would cause a gap in the species range (COSEWIC 2013). While I did not find deep phylogeographic lineages, the black colouration and unique deep-water spawning behaviour clearly qualifies as unique local adaptations. In addition, Anderson  Lake  and  Seton  Lake  are  the  only  known  lakes  that  produce  “black”  kokanee in Canada (and none have been reported from the United States).  The  conservation  status  of  “black”  kokanee  is  also  a  concern for local First Nations surrounding  the  Anderson  and  Seton  lakes  system.  “Black”  kokanee  were  identified  by  First  Nations as a species of special concern in the Bridge/Seton Water Use Plan process (Morris and Caverly 2004; David Levy, Larry Casper, Tsalalh Chief, pers. comm.). The  St’át’imc  First   90 Nation forms the majority of the population in the surrounding communities in the Bridge/Seton system. They value protecting the environment for future generations and living in harmony with all living things (Larry Casper, pers. comm.). Gwenish are used as a crop fertilizer, a winter food source, and as a way  to  transfer  traditional  St’át’imc knowledge. The cultural connection to fish and the services they provides increases the importance of studying such an understudied, but not undervalued, fish.   4.3 Future studies Understanding how intraspecific diversity originates and quantifying this diversity are important aspects to consider when conducting studies on biodiversity. My research is the first biological study on the evolutionary history and  genetic  variation  of  “black”  kokanee  in  the  Anderson  and  Seton lakes system and the first using the three known populations of “black” kokanee across the extensive range of O. nerka. Answering questions regarding the evolutionary history of, and quantifying genetic divergences among, these understudied populations of O. nerka have hopefully bridged some gaps regarding the generally unknown ecology and evolution of “black” kokanee.  Additionally, my thesis has presented new aspects of their biology to study. A better understanding of  the  specific  spawning  locations  of  “black”  kokanee  in  Japan  and  the  Anderson  and Seton lakes system would help elucidate how deep spawning originated and the associated costs and benefits of it. In Anderson Lake and Seton Lake the cumulative effects of hydroelectric dams and rail systems on putative spawning locations and rearing habitat need to be studied further. Also, investigation into what causes the unusual black colour of these kokanee populations could potentially help answer other questions regarding the unknown ecology of  91 these fish and the genetics of pigmentation. Assaying with outlier loci such as expressed sequence tag (EST) linked loci that exhibit signatures of directional selection (see Russello et al. 2011; Frazer and Russello 2013) would  also  provide  insight  into  local  adaptation  in  “black”  kokanee. Assaying with outlier loci would especially be beneficial considering the low levels of divergence  observed  with  neutral  loci  between  Seton  Lake  and  Anderson  Lake  “black”  kokanee. A more in depth  morphological  analysis  using  juvenile  and  adult  “black”  kokanee  would  help  understand the life history differences between the Anderson and Seton lake populations. Considering  that  there  are  more  questions  than  answers  regarding  “black”  kokanee,  the  future research possibilities are enticing.   4.4 Conclusion Studies of genetic and phenotypic variation are crucial in order to conserve the maximum evolutionary potential of a species (Fraser and Bernatchez 2001). Additionally, quantifying levels of genetic variation is beneficial because the stability of an ecosystem, and its ability to respond to a fluctuating or changing environment, appears to increase with increasing genetic variation (Hilborn  et  al.  2003;;  Pressey  et  al.  2007;;  Hughes  et  al.  2008).  “Black”  kokanee not only provide another example of parallel evolution found in north temperate freshwater fishes, but these unique populations of kokanee also increase the diversity found within a single species. The striking variation in black colouration, timing and depth of spawning, and morphological differences that exist among the  three  known  populations  of  “black”  kokanee  have  made  this  a  fascinating system to study.    92 References  Abdul–Muneer, P. M.  (2014). Applications of microsatellite markers in conservation genetics and fisheries management: recent advances in population structure analysis and conservation strategies. Genetics Research International, 2014:1–11.   Allendorf, F. W., and Luikart, G. (2007). Conservation and the genetics of populations. Blackwell Pub., Malden, MA.  Allendorf, F. W., and Seeb, L. W. (2000). Concordance of genetic divergence among sockeye salmon populations at allozyme, nuclear DNA and mtDNA markers. Evolution, 54(2): 640–651.  Allendorf, F. W., and Utter, F. M. (1979). Population genetics. In Fish Physiology (Hoar, W.S, Randall, D.J. and Brett, J.R., eds), vol. 8, pp. 407–454. Academic Press, New York.  Allendorf, F. W., Bayles, D., Bottom, D. L., Currens, K. C., Frissell, C. A., Hankin, D., Lichatowich, J. A., Nehlsen, W., Totter, P. C., and Williams, T. H. (1997). Prioritizing Pacific salmon stocks for conservation. Conservation Biology, 11(1): 140–152. Anders, P., Falcer, J., Powell, M., Andrusak, H., and Holderman, C. (2007). Initial microsatellite analysis of kokanee (Oncorhynchus nerka) population structure in the Kootenai/y River basin, Idaho, Montana, and British Columbia, report prepared by Cramer Fish Sciences, Moscow, ID, for Freshwater Fisheries Society of British Columbia and the Kootenai Tribe of Idaho.  Aro, K. V. (1979). Transfers of eggs and young Pacific salmon within British Columbia. Fisheries and Marine Service Technical Report Number 861. Avise, J. C. (2000). Phylogeography: The history and formation of species. Harvard University Press. Cambridge, MA.  Avise, J. C., Arnold, J., Ball, R. M., Bermingham, E., Lamb, T., Neigal, J. E., Reeb, C. A., and Saunders, N.C. (1987). Intraspecific phylogeography: The mitochondrial DNA bridge between population genetics and systematics. Annual Review of Ecology and Systemics, 18(1):489–522.   Beacham, T. D., Lapointe, M., Candy, J. R., McIntosh, B., MacConnachie C., Tabata, A., Kaukinen, K., Deng, L., Miller, K. M., and Withler, R. E. (2004). Stock identification of Fraser River sockeye salmon using microsatellites and major histocompatibility complex variation. Transactions of the American Fisheries Society, 133(5):1117–1137.    93 Beacham, T. D., McIntosh, B., MacConnachie, C., Miller, K. M., Withler, R. E., and Varnavskaya, N. (2006). Pacific rim population structure of sockeye salmon as determined from microsatellite analysis. Transactions of the American Fisheries Society, 135(1):174–187. Behnke, R. J. (1972). The systematics of salmonids fishes of recently glaciated lakes. Journal of the Fisheries Research Board of Canada, 29(6):639–671.  Bell, M. A., and Foster, S. A. (1994). Evolutionary biology of the threespine stickleback. Oxford University Press, Oxford, New York.  Bengtsson, J., Jones, H., and Setätä, H. (1997). The value of biodiversity. Trends in Ecology and Evolution, 12(9):334–336.  Bernatchez, L., and Dodson, J. J. (1990). Allopatric origin of sympatric populations of lake whitefish (Coregonus clupeaformis) as revealed by mitochondrial DNA restriction analysis. Evolution, 44(5):1263–1271.  Bernatchez, L., and Wilson, C. C. (1998). Comparative phylogeography of Nearctic and Palearctic fishes. Molecular Ecology, 7(4):431–452.   Blair, G. R., Rogers, D. E., and Quinn, T. P. (1993). Variation in life history characteristics and morphology of sockeye salmon in the Kvichak River system, Bristol Bay, Alaska. Transactions of the American Fisheries Society, 122(4):550–559.  Bolnick, D. I., Amarasekare, P., Araújo, M. S., Bürger, R., Levine, J. M., Novak, M., Rudolf, V. H. W., Schreiber, S. J., Urban, M. C., Vasseur, D. A. (2011). Why intraspecific trait variation matters in community ecology. Trends in Ecology and Evolution, 26(4):183–192.  Burgner, R. L. (1991). Life history of sockeye salmon. In: Groot, C. and Margolis, L. eds. Pacific Salmon Life Histories. University of British Columbia Press, Vancouver, pp. 3–117.  Cavalli-Sforza, L. L., and Edwards, A. W. F. (1967). Phylogenetic analysis: Models and estimation procedures. Evolution, 21(3):550–570.  Chapuis, M. P., and Estoup, A. (2007) Microsatellite null alleles and estimation of population differentiation. Molecular Biology and Evolution 24(3):621–631.  Churikov, D., and Gharrett, A. J. (2002). Comparative phylogeography of the two pink salmon broodlines: An analysis based on a mitochondrial DNA genealogy. Molecular Ecology, 11(6):1077–1101.  Churikov, D., Matsuoka, M., Luan, X., Gray, A. K., Brykov, V. A., and Gharrett, A. J. (2001). Assessment of concordance among genealogical reconstructions from various mtDNA segments in three species of pacific salmon (genus Oncorhynchus). Molecular Ecology, 10(9):2329–2339.   94 Conte, G. L., Arnegard, M. E., Peichel, C. L., and Schluter, D. (2012). The probability of genetic parallelism and convergence in natural populations. Proc. R. Soc. B., 279(1749):5039–5047. COSEWIC. (2013). Guidelines for recognizing designatable units. Appendix F5. Operations and Procedures Manual; Committee on the Status of Endangered Wildlife in Canada: Ottawa, ON, Canada.  Craig, J. K., and Foote, C. J. (2001). Countergradient variation and secondary sexual color: Phenotypic convergence promotes genetic divergence in carotenoid use between sympatric anadromous and nonanadromous morphs of sockeye salmon (Oncorhynchus nerka). Evolution, 55(2):380–391.  Craig, J. K., Foote, C. J., and Wood, C. C. (2005). Countergradient variation in carotenoid use between sympatric morphs of sockeye salmon (Oncorhynchus nerka) exposes nonanadromous hybrids in the wild by their mismatched spawning colour. Biological Journal of the Linnean Society, 84(2):287–305.   Crandall, K. A., Bininda–Emonds, O. R. P., Mace, G. M., and Wayne, R. K. (2000). Considering evolutionary processes in conservation biology. Trends in Ecology and Evolution, 15(7):290–295.  De Robertis, A., Ryer,C. H., Veloza, A., and Brodeur, R. D. (2003). Differential effects of turbidity on prey consumption of piscivorous and planktivorous fish. Canadian Journal of Fisheries and Aquatic Sciences, 60(12): 1517–1526. Duffy, J. E. (2009). Why biodiversity is important to the functioning of real-world ecosystems. Ecology Society of America, 7(8):437-444.  Dynes, J., Magnan, P., Bernatchez, L., and Rodríguez, M. A. (1999). Genetic and morphological variation between two forms of lacustrine brook charr. Journal of Fish Biology, 54(5):955–972.   Earl, D. A., and vonHoldt, B. M. (2012). STRUCTURE HARVESTER: a website and program to visualize STRUCTURE output and implement the Evanno method. Conservation Genetics Resources, 4(2):359–361.   Evanno, G., Regnaut, S, and Goudet, J. (2005). Detecting the number of clusters of individuals using the software STRUCTURE: a simulation study. Molecular Ecology, 14(8):2611–2620.  Excoffier L., Smouse, P., and Quattro, J. M. (1992). Analysis of molecular variance inferred from metric distances among DNA haplotypes: application to human mitochondrial DNA restriction data. Genetics, 131(2):479–491.  Excoffier, L., Estoup, A., and Cornuet, J. M., (2005). Bayesian analysis of an admixture model with mutations and arbitrarily linked markers. Genetics, 169(3):1727–1738.   95  Ferguson, A. (1989). Genetic differences among brown trout, Salmo trutta, stocks and their importance for conservation and management of the species. Freshwater Biology, 21(1):35–46.  Fillatre, E. K., Etherton, P., and Heath, D. D. (2003). Bimodal run distribution in a northern population of sockeye salmon (Oncorhynchus nerka): life history and genetic analysis on a temporal scale. Molecular Ecology, 12(7):1793–1805.  Fisher, R. A. (1930). The genetical theory of natural selection. Dover, New York, New York, USA. Foote, C. J. and Larkin, P. A. (1988). The role of male choice in the assortative mating of anadromous and non-anadromous sockeye salmon (Oncorhynchus nerka). Behaviour, 106(1/2), 43–62.  Foote, C. J., Clarke, W. C., Wood, C. C., and J. Blackburn. (1992). Circannual cycle of seawater adaptability in Oncorhynchus nerka: Genetic differences between sympatric sockeye salmon and kokanee. Canadian Journal of Aquatic Sciences, 46(1):99–109.   Foote, C. J., Mayer, I., Wood, C. C., Clarke, C. W., and Blackburn, J. (1994). On the developmental pathway to nonanadromy in sockeye salmon, Oncorhynchus nerka. Canadian Journal of Zoology, 72(3):397–405.  Foote, C. J., Moore, K., Stenburg, K., Craig, K. J., Wenburg, J. K. and Wood, C. C. (1999). Genetic differentiation in gill raker number and length in sympatric anadromous and nonanadromous morphs of sockeye salmon, Oncorhynchus nerka. Environmental Biology of Fishes, 54(3):161–173. Foote, C. J., Wood, C. C., and Withler, R. E. (1989). Biochemical genetic comparison of sockeye salmon and kokanee, the anadromous and nonanadromous forms of Oncorhynchus nerka. Canadian Journal of Fisheries and Aquatic Sciences, 46(1):149–158.  Fraser, D. J., and Bernatchez, L. (2001). Adaptive evolutionary conservation: towards a unified concept for defining conservation units. Molecular Ecology, 10(12):2741–2752.  Fraser, D. J., Weir, L. K., Bernatchez, L., Hansen, M. M., and Taylor, E. B. (2011). Extent and scale of local adaptation in salmonid fishes: Review and meta-analysis. Heredity, 106(3):404–420  Frazer, K. K., and Russello, M. A. (2013). Lack of parallel genetic patterns underlying the repeated ecological divergence of beach and stream–spawning kokanee salmon. Journal of Evolutionary Biology, 26(12):2606–2621.     96 Garcia de Leaniz, E. C., Fleming, I. A., Einum, S., Verspoor, E., Jordan, W. C., Consuegra, S., Aubin-Horth, N., Lajus, D., Letcher, B. H., Youngson, A. F.,Webb, J. H., Vøllestad, L. A., Villanueva, B., Ferguson, A. and Quinn, T. P. (2007). A critical review of adaptive genetic variation in Atlantic salmon: implications for conservation. Biological Reviews, 82(2):173–211.  Geen, G. H., and F. J., Andrew. (1961). Limnological Changes in Seton Lake Resulting from Hydroelectric Diversions. IPSFC progress report No. 8. New Westminster, BC. Gharret, A. J., Gray, A. K., and Brykov, V. (2001). Phylogeographic analysis of mitochondrial DNA variation in Alaskan coho salmon, Oncorhynchus kisutch. Fishery Bulletin, 99(4):528–544.  Goudet J. (1995). FSTAT version 2.9.3.1 Updated from Goudet, J. 1995. Journal of Heredity, 86(6):485–486.  Gowell, C. P., Quinn, T. P., and Taylor, E. B. (2012). Coexistence and origin of trophic ecotypes of pygmy whitefish, Prosopium coulterii, in a south–western Alaskan lake. Journal of Evolutionary Biology, 25(12):2432–2448.   Groot, C., and Margolis, L. (Eds.). (1991). Pacific salmon life histories. UBC press.  Hall, T. A. (1999). BioEdit: A user–friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acid Symposium Series, 41:95–98.   Hasegawa, M., Kishino H., and Yano T. (1985). Dating the human–ape split by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution, 22(2):160–174.  Helfman, G. S. (2009). The diversity of fishes: biology, evolution, and ecology. Wiley-Blackwell. Hoboken, NJ.  Hendry, A. P. (2001). Adaptive divergence and the evolution of reproductive isolation in the wild: an empirical demonstration using introduced sockeye salmon. In Microevolution Rate, Pattern, Process, p. 515–534. Springer Netherlands.  Hendry, A. P., and Day, T. (2005). Population structure attributable to reproductive time: isolation by time and adaptation by time. Molecular ecology, 14(4): 901-916.  Hendry, A. P., Leonetti, F. E., and Quinn, T. P. (1995). Spatial and temporal isolating mechanisms: the formation of discrete breeding aggregations of sockeye salmon (Oncorhynchus nerka). Canadian Journal of Zoology, 73(2): 339–352.  Hendry, M .A., Wenburg, J. K., Myers, K. W., and Hendry, A. P. (2002). Genetic and phenotypic variation through the migratory season provides evidence for multiple populations of wild steelhead in the Dean River, British Columbia. Transactions of the American Fisheries Society, 131(3):418–434.   97 Hilborn, R., Quinn, T. P., Schindler, D. E., and Rogers, D. E. (2003). Biocomplexity and fisheries sustainability. Proc. Natl. Acad. Sci., 100(11):6564–6568.  Hinder, K., Jonsson, B., Ryman, N., and Stahl, G. (1991). Genetic relationships among landlocked resident, and anadromous Brown Trout, Salmo trutta L. Heredity, 66(1):83–91.  Hirshfield, M. F., and Tinkle, D. W. (1975). Natural selection and the evolution of reproductive effort. Proceedings of the National Academy of Sciences of the United States of America, 72(6): 2227–2231.  Hubisz, M. J., Falush, D., Stephens, M., and Pritchard, J. K. (2009). Inferring weak population structure with the assistance of sample group information. Molecular Ecology Resources, 9(5):1322–1332.  Hughes, A. R., Inouye, B. D., Johnson, M. T. J., Underwood, N., and Vellend, M. (2008). Ecological consequences of genetic diversity. Ecology Letters, 11(6):609–623.   Hughes, J. B., Daily, G. C., and Ehrlich, P. R. (1997). Population diversity: its extent and extinction. Science, 278(5338), 689-692.  Hutchings, J. A. (1993). Adaptive life histories effected by age-specific survival and growth rate. Ecology, 74(3): 673–684.  Jacobs, D. K., Haney, T. A., & Louie, K. D. (2004). Genes, diversity, and geologic process on the Pacific Coast. Annual Review of Earth and Planetary Sciences, 32(1):601–652.  Jarne, P., and Lagoda, P. J. L. (1996). Microsatellites, from molecules to populations and back. Trends in Ecology and Evolution, 11(10):424–429.  Jombart, T., Devillard, S., and Balloux, F. (2010). Discriminant analysis of principal components: A new method for the analysis of genetically structured populations. BMC Genetics, 11(1):94–109.  Kaeriyama, M., Urawa, S., and Susuki, T. (1992). Anadromous sockeye salmon (Oncorhynchus nerka) derived from nonanadromous kokanee: Life history in Lake Toro. Scientific Reports of the Hokkaido Salmon Hatchery No. 46: 157–174.  Kano, Y., Kondou, T., and Shimizu, Y. (2010). Present status and conservation of the markless forms of stream-resident masu salmon Oncorhynchus masou (the so-called ‘iwame’) in Japanese mountain streams. Ichthyological Research, 57(1):78–84.  Keeley, E. R., Parkinson, E. A., and Taylor, E. B. (2005). Ecotypic differentiation of native rainbow trout (Oncorhynchus mykiss) populations from British Columbia. Canadian Journal of Fisheries and Aquatic Sciences, 62(7):1523–1539.   98 Keeley, E. R., Parkinson, E. A., and Taylor, E. B. (2007). The origins of ecotypic variation of rainbow trout: a test of environmental vs. genetically based differences in morphology. Journal of Evolutionary Biology, 20(2):725–736. Kogura, Y., Seeb, J. E., Azuma, N., Kudo, H., Abe, S., and Kaeriyama, M. (2011). The genetic population structure of lacustrine sockeye salmon, Oncorhynchus nerka, in Japan as the endangered species. Environmental Biology of Fishes, 92(4):539–550.  Komori, V. (1997). Salmon watershed planning profiles for the Bridge/Seton habitat management area. Report prepared for the Department of Fisheries and Oceans, Vancouver, BC.  Kurenkov, S. I. (1977). Two reproductively isolated groups of kokanee salmon, Oncorhynchus nerka kennerlyi, from Lake Kronotsky. Journal of Ichthyology, 17:526–534. Langella, O. (2000). POPULATIONS 1.2: population genetic software, individuals or population distance, phylogenetic trees. http://bioinformatics.org/~tryphon/populations/. [accessed 31 July 2014]. Lehmann, T., Hawley, W. A., and Collins, F. H. (1996). An evaluation of evolutionary constraints on microsatellite loci using null alleles. Genetics, 144(3):1155–1163.   Lemay, M., Donnelly, D., Russello, M. A. (2013). Transcriptome wide comparison of sequence variation informs SNP discovery in divergent ecotypes of kokanee salmon. BMC Genomics, 14(308):1–11. Levy, D. (2012). Implementation plan for the St’át’imc Government Services Fisheries Program: 2013 – 2017, report prepared by Levy Research Services Ltd., North Vancouver, BC, for St’át’imc Government Services, Lillooet, BC.  Lin, J., Quinn, T. P., Hilborn, R., and Hauser, L. (2008). Fine–scale differentiation between sockeye salmon ecotypes and the effect of phenotype on straying. Heredity, 101(4): 341–350. Lindsey, C.C., and McPhail, J.D. (1986). Zoogeography of fishes of the Yukon and Mackenzie basins. In The zoogeography of North American freshwater fishes. Edited by C.H. Hocutt and E.O. Wiley. John Wiley and Sons, New York. pp. 639–674.  Lowe, W.H., and Allendorf, F.W. (2010). What can genetics tell us about population connectivity?. Molecular Ecology, 19(15): 3038–3051.  Luck, G. W., Daily, G. C., Ehrlich, P. R. (2003). Population diversity and ecosystem services. Trends in Ecology and Evolution, 18(7):331–336.  Masaoka, T., Mutou, K., Oda, S., and Iwata, M. (1997). Origin and characteristics in the salmonid pedigree extending species at Nikko Branch, National Research Institute of Aquaculture. Annual Report of the National Research Institute of Aquaculture 14:1–26.  99 Matinkappi, J. B., Kekäläinen, J., Shatilova, Y., Parkkinen, J. (2009). Carotenoid Concentration of Arctic Charr (Salvelinus Alpinus L.) from Spectral Data. Advanced Concepts for Intelligent Vision Systems, Lecture Notes in Computer Science, 5807:562–570.  McGregor, A. J., Lane, S., Thomason, M. A., Zhivotovsky, L. A., Smoker, W. W., and Gharrett, A. J. (1998). Migration timing, a life history trait important in the genetic structure of pink salmon. N. Pac. Anadr. Fish Comm. BUll. No.1: 262–273.  McPhail, J. D. (1984). Ecology and evolution of sympatric sticklebacks (Gasterosteus): Morphological and genetic evidence for a species pair in Enos Lake, British Columbia. Canadian Journal of Zoology, 62(7):1402–1408.  McPhail, J. D., and Lindsey, C. C. (1970). Freshwater fishes of northwestern Canada and Alaska. Fisheries Research Board of Canada, Ottawa.  McPhail, J. D., and Lindsey, C. C. (1986). Zoogeography of the freshwater fishes of Cascadia (the Columbia system and rivers north to the Stikine). The zoogeography of North American freshwater fishes, 615–637.  Meffe, G. K., and Carroll, C. R. (1994). Principles of conservation biology. Sinaur Associates. Sunderland, MA.  Moodie, G. E. E. (1972). Predation, natural selection and adaptation in an unusual stickleback. Heredity, 28(2):155–167.  Moritz, C. (1994). Defining evolutionary significant units for conservation. Trends in Ecology and Evolutions, 9(1):373–375.  Moritz, C. (2002). Strategies to protect biological diversity and the evolutionary process that sustain it. Systemic Biology, 51(2):238–254.   Morris, A. R., and Caverly, A. (2004). 2003–2004 Seton and Anderson Lakes kokanee assessment, report to British Columbia Conservation Foundation and Ministry of Water, Land and Air Protection, Kamloops, BC.  Morris, A. R., Braumandl, E., Andrusak, H., and Caverly, A. (2003). 2002/2003 Seton and Anderson Lakes kokanee assessment: feasibility study and study design, report to British Columbia Conservation Foundation and Ministry of Water, Land and Air Protection, Kamloops, BC.   Morris, D. B., Richard, K. R., and Wright, J. M. (1996). Microsatellites from rainbow trout (Oncorhynchus mykiss) and their use for genetic study of salmonids. Canadian Journal of Fisheries and Aquatic Sciences, 53(1):120–126. Moss, B. (1998). Ecology of fresh waters: Man and medium, past to future. Oxford: Blackwell Science. Malden, MA.   100 Muto, N., Nakayama, K., and Nakabo, T. (2012). Distinct genetic isolation between “Kunimasu” (Oncorhynchus kawamurae) and “Himemasu” (O. nerka) in Lake Saiko, Yamanashi Prefecture, Japan, inferred from microsatellite analysis. Ichthyological Reserach. 60(2):188–194. Nakabo, T., Nakayama, K., Muto, N., and Miyazawa, M. (2011). Oncorhynchus kawamurae “Kunimasu,” a deepwater trout, discovered in Lake Saiko, 70 years after extinction in the original habitat, Lake Tazawa, Japan. Ichthyological Research, 58(2):180–183.   Nakabo, T., Tohkairin, A., Muto, N., Watanabe, Y., Miura, Y., Miura, H., Aoyagi, T., Kaji, N., Nakayama, K., and Kai, Y. (2014). Growth–related morphology of “Kunimasu” (Oncorhynchus kawamurae: family Salmonidae) from Lake Saiko, Yamanashi Prefecture, Japan. Ichthyological Reserach 61(2):115–130.  Narum S. P. (2006). Beyond Bonferroni: less conservative analyses for conservation genetics. Conservation Genetics, 7(5):783–787.  Nei, M., Tajima, F, and Tateno, Y. (1983). Accuracy of estimated phylogenetic trees from molecular data. II gene frequency data. Journal of Molecular Evolution, 19(2):153–170. Neilson, J. L., Gan, C., and Thomas, W., K. (1994). Differences in genetic diversity for mitochondrial DNA between hatchery and wild populations of Oncorhynchus. Canadian Journal of Fisheries and Aquatic Sciences, 51(S1):290–297. Nelson, J. S. (1968). Distribution and nomenclature of North American kokanee, Oncorhynchus nerka. Journal of the Fisheries Research Board of Canada, 25(2):409–414.   Nelson, R., and Beacham, T. (1999). Isolation and cross species amplification of microsatellite loci useful for study of Pacific salmon. Animal Genetics, 30(3):228–229. Northrup, S., Connor, M., and Taylor, E. B. (2010). Population structure of lake trout (Salvelinus namaycush) in a large glacial-fed lake inferred from microsatellite DNA and morphological analysis. Canadian Journal of Fisheries and Aquatic Sciences, 67(7):1171–1186. O’Donnell, K., Ward, T. J., Geiser, D. M., Kistler, H. C., and Aoki, T. (2004). Genealogical concordance between the mating type locus and seven other nuclear genes supports formal recognition of nine phylogenetically distinct species within the Fusarium graminearum clade. Fungal Genetics and Biology, 41(6):600–623.  Olsen, J.B., Wilson, S. L, Kretschmer, E. J., Jones, K. C., and Seeb, J. E. (2000). Characterization of 14 tetranucleotide microsatellite loci derived from sockeye salmon. Molecular Ecology, 9(12):2185–2187 Page, R. D. M. (1996). TREEVIEW: an application to display phylogenetic trees on personal computers. Computer Applications in the Biosciences, 12(4): 357–358.  101 Palsbøll, P. J., Berube, M., and Allendorf, F. W. (2007). Identification of management units using population genetic data. Trends in Ecology and Evolution, 22(1):11–16.  Peakall, R., and Smouse P. E. (2012). GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics 28, 2537–2539.  Pigeon, D., Chouinard, A., and Bernatchez, L. (1997). Multiple modes of speciation involved in the parallel evolution of sympatric morphotypes of lake whitefish (Coregonus clupeaformis, Salmonidae). Evolution, 51(1):196–205.  Pressey, R. L., Cabeza, M., Watts, M. E., Cowling, R. M., and Wilson, K. A. (2007). Conservation planning in a changing world. Trends in Ecology and Evolution, 22(11): 83–592.  Pritchard, J. K., Stephens, M., and Donnelly, P. (2000). Inference of population structure using multilocus genotype data. Genetics, 155(2):945–959.   Quinn, T. P. (2005). The Behavior and Ecology of Pacific Salmon and Trout. University of Washington Press, Seattle, Washington.  R Development Team. (2012). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna. Raymond M., and Rousset, R. R. (1995). GENEPOP (Version 1.2): Population genetics software for exact tests and ecumenicism. Journal of Heredity, 86:248–249.  Reimchen, T. E. (1989). Loss of nuptial color in threespine sticklebacks (Gasterosteus aculeatus). Evolution, 43(2):450–460. Reist, J. D. (1985). An empirical evaluation of several univariate methods that adjust for size variation in morphometric data. Canadian Journal of Zoology, 63(6):1429–1439.  Rice, W. R. (1984). Disruptive section on habitat preference and the evolution of reproductive isolation: a simulation study. Evolution, 38(6):1251–1260.  Richey, J. E., Perkins, M. A., and Goldman, C. R. (1975). Effects of kokanee salmon (Oncorhynchus nerka) decomposition on the ecology of a subalpine stream. Journal of the Fisheries Research Board of Canada, 32(6):817–820.   Ricker, W. E. (1959). Additional observations concerning residual sockeye and kokanee (Oncorhynchus nerka). Journal of the Fisheries Research Board of Canada, 16(6):897–902.   Ricker, W. E. (1972). Heredity and environmental factors affecting certain salmonid populations, In R.C. Simon and P.A. Larkin (eds.) The stock concept in Pacific salmon, p. 27–160.  H.W. MacMillan Lectures in Fisheries. Univ. British Columbia.Vancouver. BC.   102 Ricker, W. E. (1938). "Residual" and kokanee salmon in Cultus Lake. Journal of the Fisheries Research Board of Canada, 4a(3):192–218.   Ricker, W. E. (1940). On the origin of kokanee, a freshwater type of sockeye salmon. Proc. Roy. Soc. Canada Ser., 34(5):121–135.  Rising, J. D., and Avise, J. C. (1993). Application of genealogical concordance principles to the taxonomy and evolutionary history of the sharp–tailed Sparrow (Ammodramus caudacutus). The Auk, 110(4):844–856.  Roff, D. A. (1984). The evolution of life history parameters in teleosts. Canadian Journal of Fisheries and Aquatic Sciences, 41(6):989–1000.  Rousset, F. (2013). User manual for Genepop 4.2for Windows/Linux/Mac OS X Russello, M. A., Kirk, S. L., Frazer, K. K., Paul, J. A. (2011). Detection of outlier loci and their utility for fisheries. Evolutionary Applications, 5(1):39–52.  Ryman, N., and Stahl, G. (1981). Genetic perspectives of the identification and conservation of Scandinavian stocks of fish. Canadian Journal of Fisheries and Aquatic Sciences, 38(12):1562–1575.  Saitou, N., and Nei, M. (1987). The neighbor–joining method: a new method for reconstructing phylogenetic trees. Molecular Biology Evolution, 4(4):406–425. Schindler, D. E., Hilborn, R., Chasco, B., Boatright, C. P., Quinn, T. P., Rogers, L. A., and Webster, M. S. (2010). Population diversity and the portfolio effect in an exploited species. Nature, 465(7298):609–612.  Schluter, D. (1996). Ecological speciation in postglacial fishes. Proc. Roy. Soc. Lon., 351: 807–814.  Schluter, D. (2000). Ecological character displacement in adaptive radiation. The American Naturalist, 156(S4):S4–S16.  Schluter, D., and Nagel, L. M. (1995). Parallel Speciation by Natural Selection. The American Naturalist, 146(2):292–301.  Schwartz M. K., Luikart, G., and Waples, R. S. (2007). Genetic monitoring as a promising tool for conservation and management. Trends in Ecology and Evolution, 22(1):25–33.  Scott, D. (1984). Origin of the New Zealand sockeye salmon, Oncorhynchus nerka (Walbaum). Journal of the Royal Society of New Zealand, 14(3):245–249.   Sendek, D. (2004). The origin of sympatric forms of European whitefish (Coregonus lavaretus (L.)) in Lake Ladoga based on comparative genetic analysis of populations in north–west Russia. Annales Zoologici Fennici, 41:25–39.   103 Shephard, B. G. (2000). A case history: the kokanee stocks of Okanagan Lake. Proceedings of a conference on the biology and management of species and habitats at risk. L. M. Darling. Kamloops, B.C.: 609–616. Shortreed, K. S., Morton, K. F., Malange, K., and Hume, J. M. B. (2001). Factors limiting juvenile sockeye production and enhancement potential for selected B.C. nursery lakes, report for Fisheries and Oceans Canada, Cultus Lake, BC.  Smith, C. T., Koop, B. F., and Nelson, R. J. (1998). Isolation and characterization of coho salmon (Oncorhynchus kisutch) microsatellites and their use in other salmonids. Molecular Ecology, 7(11):1614–1616. Stables, T. B. (2004). Acoustic and video surveys of kokanee and their potential deepwater spawning habitat in Anderson Lake, BC. report to BC Conservation Foundation and BC Ministry of Land, Water, and Air Protection, Kamloops, BC. Steinback, C., and Fuller, A. (2005). Atlas of Pacific salmon: the first map–based status assessment of salmon in the North Pacific. Berkeley: University of California Press. Portland, OR. Suk, H. Y., and Neff, B. D. (2009). Microsatellite genetic differentiation among populations of the Trinidadian guppy. Heredity, 102(5):425–434.   Tallman, R. F., and Healey, M.C. (1994). Homing, straying, and gene flow among seasonally separated populations of chum salmon (Oncorhynchus keta). Canadian Journal of Fisheries and Aquatic Sciences, 51(3):577–588. Tamura, K and Nei, M. (1993) Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution, 10(3):512–526. Tamura, K., Stecher, G., Peterson, D., Filipski, A., and Kumar, S. (2013). MEGA6: Molecular Evolutionary Genetics Analysis Version 6.0. Molecular Biology and Evolution, 30(12): 2725–2729.  Taylor, E. B. (1991). A review of local adaptation in Salmonidae, with particular reference to Pacific and Atlantic salmon. Aquaculture, 98: 185-207.   Taylor, E. B. (1995). Genetic variation at minisatellite DNA loci among north pacific populations of steelhead and rainbow trout (Oncorhynchus mykiss). The Journal of Heredity, 86(5):354–363.  Taylor, E. B. (1999). Species pairs of north temperate freshwater fishes: Evolution, taxonomy, and conservation. Reviews in Fish Biology and Fisheries. 9(4):299–324.    104 Taylor, E. B. (2013). Microsatellite DNA analysis of populations of kokanee (Oncorhynchus nerka) in the Tatelkuz Lake/Chedakuz Creek Watershed, Interior British Columbia. report to AMEC, Vancouver, BC.   Taylor, E. B. and Foote, C. J. (1991). Critical swimming velocities of juvenile sockeye salmon and kokanee, the anadromous and non–anadromous forms of Oncorhynchus nerka. Journal of Fish Biology, 38(3):407–419.  Taylor, E. B., and Bentzen, P. (1993). Evidence for multiple origins and sympatric divergence of trophic ecotypes of smelt (Osmerus) in Northeastern North America. Evolution, 47(3):813–832.  Taylor, E. B., and Yau, M. (2013). Microsatellite DNA analysis of bull trout (Salvelinus confluentus), Arctic grayling (Thymallus arcticus), and mountain whitefish (Prosopium williamsoni) in the Peace River and tributaries near the proposed BC Hydro Site C hydroelectric development in northeastern British Columbia: 2006–2011. report to B.C. Hydro, Vancouver, BC.  Taylor, E. B., Darveau, C. A., and Schulte, P. M. (2013). Setting conservation priorities in a widespread species: Phylogeographic and physiological variation in the lake chub, Couesius plumbeus (Pisces: Cyprinidae). Diversity, 5(2):149–165.   Taylor, E. B., Foote, C. J., and Wood, C.C. (1996). Molecular genetic evidence for parallel life–history evolution within a Pacific salmon (sockeye salmon and kokanee, Oncorhynchus nerka). Evolution, 50(1):401–416.  Taylor, E. B., Harvey, S., Pollard, S., and Volpe, J. (1997). Postglacial genetic differentiation of reproductive ecotypes of kokanee Oncorhynchus nerka in Okanagan Lake, British Columbia. Molecular Ecology, 6(6):503–517.  Taylor, E. B., Kuiper, A., Troffe, P. M., Hoysaki, D. J., and Pollard, S. (2000). Variation in developmental biology and microsatellite DNA in reproductive ecotypes of kokanee, Oncorhynchus nerka: Implications for declining populations in a large British Columbia lake. Conservation Genetics, 1(3):231–249.  Taylor, E. B., Tamkee, P., Keeley, E. R., and Parkinson, E. A. (2011). Conservation prioritization in widespread species: The use of genetic and morphological data to assess population distinctiveness in rainbow trout (Oncorhynchus mykiss) from British Columbia, Canada. Evolutionary Applications, 4(1):100–115.   Thomas, W. K., Nielson, J. L., and Gan, C. (1994). Differences in genetic diversity for mitochondrial DNA between hatchery and wild populations of Oncorhynchus. Canadian Journal of Fisheries and Aquatic Sciences, 51(S1): 290–297. Thomas, W. K., Withler, R. E., and Beckenbach, A. T. (1986). Mitochondrial DNA analysis of Pacific salmonids evolution. Canadian Journal of Zoology, 64(5):1058–1064.   105 Tokui, T. (1964). Studies on the kokanee salmon (V). Transplantation of the kokanee salmon in Japan. Sci. Rep. Hokkaido Salmon Hatchery 18:73–90. Utter, F., and Ryman, N. (1993). Genetic markers and mixed stock fisheries. Fisheries, 18(8):11–21.  van Oosterhout, C., Hutchinson, W. F., Wills, D. M.P., and Shipley, P. (2004). MICRO–CHECKER: software for identifying and correcting genotyping errors in microsatellite data. Molecular Ecology Notes, 4:535–538.  Varnavskaya, N. V., Wood, C. C., and Everett, R. J. (1994). Genetic variation in sockeye salmon (Oncorhynchus nerka) populations of Asia and North America. Canadian Journal of Fisheries and Aquatic Sciences, 51(S1):132–146. Vernon, E. H. (1957). Morphometric comparison of three races of kokanee (Oncorhynchus nerka) within a large British Columbia lake. Journal of the Fisheries Research Board of Canada, 14(4):573–593.  Volff, J. N. (2005). Genome evolution and biodiversity in teleost fish. Heredity, 94(3):280–294.   Waples, R. S. (1995). Evolutionarily significant units and the conservation of biological diversity under the Endangered Species Act. In: Evolution and the Aquatic Ecosystem: Defining Unique Units in Population Conservation (eds Nielsen JL Powers GA), pp. 8–27. Symposium 17. American Fisheries Society, Bethesda, Maryland. Waples, R. S., Pess, G. R., and Beechie, T. (2008). Evolutionary history of Pacific salmon in dynamic environments. Evolutionary Applications, 1(2):189–206.  Waples, R. S., Teel, D. J., Myers, J. M., and Marshall, A. R. (2004). Life–history divergence in Chinook salmon: historical contingency and parallel evolution. Evolution, 58(2):386–403.  Weir, B. S. (1996). Genetic Data Analysis II. Sinauer Associates, Inc. Sunderland, MA. Weir, B. S. and Cockerham, C. C. (1984). Estimating F–statistics for the analysis of population structure. Evolution, 38(6):1358–1370.   Willi, Y., and Hoffmann, A. A. (2009). Demographic factors and genetic variation influence population persistence under environmental change. Journal of Evolutionary Biology, 22(1):124–133.  Winans, G. A., and Urawa, S. (2000). Allozyme variability of Oncorhynchus nerka in Japan. Ichthyological Research, 47(3):343–352. Withler, F. C. (1982). Transplanting Pacific Salmon: Canadian Technical Report of Fisheries and Aquatic Sciences No. 1079. Prepared for Department of Fisheries and Oceans, Nanaimo, BC.  106 Withler, R. E., Le, K. D, Nelson, R. J., Miller, K. M., and Beacham, T. D. (2000). Intact genetic structure and high levels of genetic diversity in bottlenecked sockeye salmon (Oncorhynchus nerka) populations of the Fraser River, British Columbia, Canada. Canadian Journal of Fisheries and Aquatic Sciences, 57(10):1985–1998. Wood, C. C, and Gross, M. R. (2008). Elemental conservation units: Communicating extinction risk without dictating targets for protection. Conservation Biology, 22(1):36–47.  Wood, C. C. (1995). Life history variation and population structure in sockeye salmon. American Fisheries Society Symposium, 17:195–216. Wood, C. C. and Foote, C. J. (1990). Genetic differences in the early development and growth of sympatric sockeye salmon and kokanee (Oncorhynchus nerka) and their hybrids. Canadian Journal of Fisheries and Aquatic Sciences, 47(11):2250–2260. Wood, C. C. and Foote, C. J. (1996).  Evidence for sympatric genetic divergence of anadromous and non–anadromous morphs of sockeye salmon (Oncorhynchus nerka). Evolution, 50(3):1265–1279.  Wood, C. C., Bickham, J. W, Nelson, R. J., Foote, C. J., and Patton, J. C. (2008). Recurrent evolution of life history ecotypes in sockeye salmon: implications for conservation and future evolution. Evolutionary Applications, 1(2):207–221.  Yamamoto, S., Kitamura, S., Sakano, H., and Morita, K. (2011). Genetic structure and diversity of Japanese kokanee Oncorhynchus nerka stocks as revealed by microsatellite and mitochondrial DNA markers. Journal of Fish Biology. 79(5):1340–1349. Young, S. F., Downen, M. R., and Shaklee, J. B. (2004). Microsatellite DNA data indicate distinct native populations of kokanee, Oncorhynchus nerka, persist in the Lake Sammamish Basin, Washington. Environmental Biology of Fishes, 69(1):63–79.   107 Appendix   Figure A.1. Maximum likelihood tree using the HKY model of evolution demonstrating the phylogenetic relationship among 12 haplotypes resolved from ND1 mitochondrial DNA sequences of Oncorhynchus nerka (N=105) from the Columbia River (CR), Cowichan River (CoR), Alsek River (AR), Fraser River (FR), Skeena River (SK), Bristol Bay Alaska (BBA), Kamchatka (K) drainages. Bold haplotypes represent “black” kokanee from Japan (J) and Anderson Lake (AL) and Seton Lake (SL). Number of individuals with each haplotype is in brackets. The percentage of trees in which the associated haplotypes clustered together is shown next to the branches (percentages less than 50% have been removed) from 1000 bootstraps. “Š”  represent  Japanese  “black”  kokanee.  “y”  represent  Anderson  Lake  and  Seton  Lake  “black”  kokanee.   108  Figure A.2. Maximum likelihood tree using the TN93+G model of evolution demonstrating the phylogenetic relationship among 18 haplotypes resolved from ND2 mitochondrial DNA sequences of Oncorhynchus nerka (N=82) from the Columbia River (CR), Cowichan River (CoR), Alsek River (AR), Fraser River (FR), Skeena River (SK), Bristol Bay Alaska (BBA), Kamchatka (K) drainages. Bold haplotypes represent “black” kokanee from Japan (J) and Anderson Lake (AL) and Seton Lake (SL). Number of individuals with each haplotype is in brackets. The percentage of trees in which the associated haplotypes clustered together is shown next to the branches (percentages less than 50% have been removed) from 1000 bootstraps. “Š”  represent  Japanese  “black”  kokanee.  “y”  represent  Anderson  Lake  and  Seton  Lake  “black”  kokanee.       109 Table A.3.  Summary statistics at nine microsatellite loci in sockeye salmon/kokanee (Oncorhynchus nerka) from Chedakuz Creek (CC), Davidson Creek (DV), Meadow Creek (MC), Seton Lake (SL), Anderson Lake (AL), Hansen Creek, Alaska (HCA), Gates Creek (GC), Portage Creek (PC), and Lake Saiko, Japan (J).  Number of alleles/locus (A), expected heterozygosity (HE), observed heterozygosity (HO), and sample size (N) for each loci and population. Significant departures from Hardy-Weinberg equilibrium after correcting for multiple comparisons are represented by * (P < 0.018).   Population Locus  Omy77 Ots103 Oki29 Ots108 Ots100 Oki10 One108 One110 One103 Average CC           A 5 9 5 5 4 11 7 6 7 6.56 He 0.465 0.829 0.651 0.388 0.333 0.842 0.755 0.683* 0.729 0.57 Ho 0.448 0.655 0.759 0.276 0.345 0.862 0.714 0.448 0.586 0.57 N 29 29 29 29 29 29 28 29 29 28.89            DV           A 5 14 10 5 3 21 11 10 18 10.78 He 0.781 0.905 0.841 0.719* 0.351* 0.951 0.832 0.883 0.933 0.88 Ho 0.742 0.710 0.839 0.161 0.226 0.935 0.774 0.742 0.839 0.66 N 31 31 31 31 31 31 31 31 31 31.00            MC           A 10 18 11 14 17 25 16 16 22 16.56 He 0.657 0.951* 0.702 0.862* 0.917 0.940 0.933 0.916 0.956 0.86 Ho 0.633 0.767 0.767 0.414 0.800 0.900 0.800 0.900 0.933 0.77 N 30 30 30 29 30 30 30 30 30 29.89                110 Population Locus  Omy77 Ots103 Oki29 Ots108 Ots100 Oki10 One108 One110 One103 Average SL           A 9 19 16 16 16 32 14 15 36 19.22 He 0.759 0.926 0.847 0.903* 0.867 0.963 0.825 0.900 0.971* 0.87 Ho 0.674 0.907 0.872 0.333 0.787 0.935 0.778 0.851 0.915 0.78 N 46 43 47 42 47 46 45 47 47 45.56            AL           A 12 24 23 17 20 38 16 17 54 24.56 He 0.843 0.933 0.700 0.81* 0.897 0.966 0.874 0.891 0.968 0.88 Ho 0.759 0.872 0.725 0.416 0.835 0.825 0.825 0.887 0.975 0.79 N 79 78 80 77 79 80 80 80 80 79.22            HCA           A 5 12 9 7 13 13 11 11 21 11.33 He 0.637 0.916 0.830 0.871 0.903 0.935 0.893 0.895 0.964 0.87 Ho 0.656 0.900 0.895 0.143 0.700 0.667 0.950 0.950 0.950 0.76 N 20 20 19 14 20 18 20 20 20 19.00            GC           A 7 13 12 8 12 16 6 9 23 11.78 He 0.747 0.702 0.701 0.732* 0.843 0.884 0.776 0.741 0.872 0.78 Ho 0.800 0.683 0.585 0.067 0.725 0.829 0.675 0.610 0.775 0.64 N 40 41 41 30 40 41 40 41 40 39.33      111 Population Locus  Omy77 Ots103 Oki29 Ots108 Ots100 Oki10 One108 One110 One103 Average PC           A 10 13 17 12 12 25 14 18 22 15.89 He 0.852 0.898 0.708 0.787* 0.781 0.951* 0.870 0.928 0.951 0.86 Ho 0.806 0.889 0.714 0.241 0.694 0.833 0.778 0.972 1.000 0.77 N 36 36 35 29 36 36 36 36 34 34.89            J           A 4 12 8 5 4 9 10 7 12 7.89 He 0.720 0.923 0.894 0.807 0.723 0.944 0.883 0.837 0.924 0.85 Ho 0.333 0.909 0.833 0.833 0.667 0.200 0.833 0.750 0.917 0.70 N 12 11 12 12 12 10 12 12 12 11.67   

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.24.1-0165554/manifest

Comment

Related Items