{"http:\/\/dx.doi.org\/10.14288\/1.0099481":{"http:\/\/vivoweb.org\/ontology\/core#departmentOrSchool":[{"value":"Forestry, Faculty of","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/dataProvider":[{"value":"DSpace","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeCampus":[{"value":"UBCV","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/creator":[{"value":"Ie, Bryan","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/issued":[{"value":"2009-07-06T21:21:04Z","type":"literal","lang":"en"},{"value":"2000","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#relatedDegree":[{"value":"Master of Science - MSc","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeGrantor":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/description":[{"value":"Despite the keen interest in using native grass species for restoration purposes,\r\nlittle is known about their ecology and genetics. By examining the population genetic\r\nstructure, gene flow, and mating system of Elymus glaucus, recommendations for the\r\nsuccessful growth and transfer of seed can be made.\r\nIsozymes and morphological traits were used to study 40 Elymus glaucus\r\npopulations located within British Columbia. F[sub st] values, based upon 21 isozyme loci,\r\nwere high (0.65) suggesting that species' diversity is predominantly distributed between\r\nrather than within populations. Q[sub st] analysis, an index analogous to F[sub st], was used to\r\ndescribe population differentiation of the morphological traits measured. Continuous\r\ntraits displayed an average Q[sub st][sup c] of 0.80 while the discontinuous trait mean Q[sub st][sup d] was 0.44.\r\nIt seems that the distribution of diversity follows the same trend set by isozyme\r\ndistributions in that morphological diversity of this native grass species is partitioned\r\nbetween rather than within populations. F[sub is] and F[sub it] estimates showed a deficiency of\r\nheterozygote individuals. This may be due to inbreeding, a colonization effect, or a\r\nrecent evolutionary bottleneck. Mating system analysis of three Vancouver Island\r\npopulations indicates that outcrossing does occur within Elymus glaucus. The\r\ndistribution of Elymus glaucus populations seems characteristic of species undergoing\r\nmetapopulation dynamics. This observation is supported by its high F[sub st], low geographic\r\nstructuring (isolation by distance), and the patchiness of its environment.\r\n\r\nQuestions addressed by this thesis pertain to the degree of adaptation and\r\nplasticity this grass possesses. In nature, is there any indication of genetic or\r\nmorphological structuring? Will a high degree of genetic diversity be enough for this\r\ngrass to successfully evolve and adapt to different environmental conditions? Or does\r\nphenotypic plasticity hold the key to its survival in non-local habitats? The answers to\r\nthese questions will help uncover the evolutionary life history of this native grass species\r\nand facilitate the development of successful strategies for reclaiming disturbed habitats\r\nusing native grasses.","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/aggregatedCHO":[{"value":"https:\/\/circle.library.ubc.ca\/rest\/handle\/2429\/10267?expand=metadata","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/extent":[{"value":"4002015 bytes","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/elements\/1.1\/format":[{"value":"application\/pdf","type":"literal","lang":"en"}],"http:\/\/www.w3.org\/2009\/08\/skos-reference\/skos.html#note":[{"value":"R E C L A I M I N G DISTURBED HABITATS USING N A T I V E GRASSES: THE GENETIC STORY OF Elymus glaucus (BLUE WILDRYE) by B R Y A N IE B.Sc , The University of British Columbia, 1997 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF M A S T E R OF SCIENCE in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Forest Science; Genetics Program) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A February 2000 \u00a9 Bryan Alexander Ie, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of fO&J&^C S\"<-H^HC&-The University of British Columbia Vancouver, Canada Date K M ^ C L H - 2GOQ DE-6 (2\/88) Abstract Despite the keen interest in using native grass species for restoration purposes, little is known about their ecology and genetics. By examining the population genetic structure, gene flow, and mating system of Elymus glaucus, recommendations for the successful growth and transfer of seed can be made. Isozymes and morphological traits were used to study 40 Elymus glaucus populations located within British Columbia. F s t values, based upon 21 isozyme loci, were high (0.65) suggesting that species' diversity is predominantly distributed between rather than within populations. Q s t analysis, an index analogous to F s t , was used to describe population differentiation of the morphological traits measured. Continuous c d traits displayed an average Q st of 0.80 while the discontinuous trait mean Q s t was 0.44. It seems that the distribution of diversity follows the same trend set by isozyme distributions in that morphological diversity of this native grass species is partitioned between rather than within populations. F j s and F, t estimates showed a deficiency of heterozygote individuals. This may be due to inbreeding, a colonization effect, or a recent evolutionary bottleneck. Mating system analysis of three Vancouver Island populations indicates that outcrossing does occur within Elymus glaucus. The distribution of Elymus glaucus populations seems characteristic of species undergoing metapopulation dynamics. This observation is supported by its high F s t , low geographic structuring (isolation by distance), and the patchiness of its environment. Questions addressed by this thesis pertain to the degree of adaptation and plasticity this grass possesses. In nature, is there any indication of genetic or morphological structuring? Wil l a high degree of genetic diversity be enough for this grass to successfully evolve and adapt to different environmental conditions? Or does phenotypic plasticity hold the key to its survival in non-local habitats? The answers to these questions will help uncover the evolutionary life history of this native grass species and facilitate the development of successful strategies for reclaiming disturbed habitats using native grasses. iv Abstract if Table of Contents iv List of Tables vi List of Figures ix Acknowledgements xi Chapter I: Introduction 1.1 Restoration and its Importance 1 1.2 Elymus glaucus 7 1.3 Objective --9 Chapter II: Materials and Methods 2.1 Seed Collections 12 2.2 Electrophoresis 15 2.3 Genetic Diversity 16 2.4 Gene Flow 17 2.5 Genetic Distance 17 2.6 Common Garden 18 2.7 Morphological Measurements 19 2.8 Data Analysis 24 Chapter III: Results 3.1 Germination 30 3.2 Gel Scoring 30 3.3 Genetic Variation 32 3.4 Genetic Structure 34 3.5 Gene Flow 37 3.6 Genetic Distance 39 3.7 Continuous Traits 41 3.8 Discontinuous Traits (Q s t ) 43 3.9 Correlation Analysis 43 Chapter IV: Discussion 4.1 Species Genetic Diversity 45 4.2 Genetic Diversity within Populations 48 4.3 Genetic Structure 50 4.4 Hierarchical Analysis of F s t 52 4.5 Gene Flow 55 4.6 Mating System 56 4.7 Random Genetic Drift 59 4.8 Metapopulation Dynamics 60 4.9 Continuous vs Discontinous Traits 63 4.10 Selective Neutrality of Isozymes 64 4 . 1 1 Q s t \u00b0 v s Fst 6 5 4 . 1 2 Q s t d v s F s t 68 Chapter V: Conclusion 5.1 Genetic Story 70 5.2 Restoration Implications 70 5.3 Recommendations for Further Research 73 Bibliography 75 Appendix r 89 List of Tables Table Ha. Type of data measured (isozyme, morphological, or both), population name, Biogeoclimatic Ecosystem Classification (Meidinger and Pojar 1991), forest district, and elevation of 29 B.C. populations of Elymus glaucus. (13) Table lib. Population name, Biogeoclimatic Ecosystem Classification, Forest District, and Elevation in meters. (21) Table He. Description of continuous traits measured. (22) Table lid. Description of discontinuous traits characterized by the assumption of dominance and recessiveness. (26) Table Ilia. Measures of Genetic Diversity in 26 Populations of Elymus glaucus; mean sample size per locus, mean number of alleles per locus, percentage of polymorphic loci*, and observed and expected heterozygosity. (33) Table IHb. Isozyme diversity for Elymus glaucus at 21 loci at the species, within population, and among population levels. Parameters used were percentage of polymorphic loci (PLP), mean number of alleles per locus (Al), expected heterozygosity within populations(Hs), total heterozygosity (Ht), and F-statistics. (35) Table IIIc. Hierarchical F s t for populations nested within Factors and Factors nested within the Total. (36) Table Hid. Population Outcrossing Rates. Population name, no. of families assayed, number of loci, multilocus t-estimate (tm), singlelocus t-estimate (ts), and difference between multi and singlelocus t estimates (Ritland 1996). (38) Table Hie. Morphological differentiation among populations for continuous (Q s t ) and discontinuous (Qst.d) traits. (42) Table IVa. Several studies contributed to the data set for grasses (166) and other plants (666). Species and population level measures of genetic diversity. Percentage of polymorphic loci (Ps & Pp), Average number of alleles per locus (As & Ap), Total genetic diversity (H t), Intrapopulational Diversity (H s), Proportion of intrapopulational Diversity (G s t ) . (Godt and Hamrick 1998). (46) Table A l . Populations grouped by elevation in meters. (92) Table A2. Populations grouped by average yearly snowpack. (93) Table A3. Populations grouped by average yearly temperature. (93) Table A4. Populations grouped by average yearly precipitation. (94) Table A5. Populations grouped by the number of frost free days. (94) Table A6. Populations grouped by soil type. (94) Table A7. Enzymes, number of loci resolved, buffer system, and stain components. (96) Allele Frequencies (97) List of Figures Figure 1. General locations of Elymus glaucus populations sampled within British Columbia. (14) Figure 2. Picture of Elymus glaucus illustrating the different traits measured and characterized. (23) Figure 3. Cluster analysis using unweighted pair group method. (40) Figure A l . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Malate Dehydrogenase. (89) Figure A2. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Shikimate Dehydrogenase. (89) Figure A3. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Diaphorase. (90) Figure A4. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Isocitric Dehydrogenase. (90) Figure A 5 . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Phosophoglucoisomerase. (90) Figure A 6 . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Malic Enzyme. (91) Figure Al. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Phosphoglucomutase. (91) Figure A 8 . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Glutamate Dehydrogenase. (91) Figure A 9 . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Alcohol Dehydrogenase. (92) xi Acknowledgements I'd like to thank all the grass for cooperating so well during this study. Besides the grass, I would like to thank Dr. Kermit Ritland for supporting me through my research and offering constructive and timely advice throughout this project. His knowledge of genetics and math always seemed to astound me as I walked back from his office trying to figure out exactly what just happened during our meetings. Without Kermit's help, this project would definitely not have been successful. Dr. Carol Ritland also contributed greatly. Those little conversations we had helped instill ideas within my head which later became integrated into this thesis. In addition, Kermit and Carol's knowledge and experience with isozymes was essential for surviving the Saltmines. I'd like to thank my Mom and Dad for giving me the opportunity to pursue my university education. Their sacrifices have not been overlooked and I appreciate and will remember every single one. Even though they have little knowledge of what I do at school everyday, they have contributed more than they will ever know. Phil and Carla Burton and all the gang at Symbios Research and Restoration were instrumental in first collecting the seed for the study but then creating and maintaining the common garden in Smithers, British Columbia. I would like to thank all those individuals up north for making me feel welcome during my stay at the research station in Smithers. I would like to express my deepest gratitude to Adi Nokiani (A.K.A. Iranian Warrior) for unknowingly but diligently helping me measure all those morphological XII traits at the common garden in Smithers; Philip Hoy and David Chu for entertaining and keeping me company on trips to the Island. Lisa Marie O'Connell, the undisputed queen of isozymes, always there for me when I needed help finding things I had misplaced in the lab, performing the \"hi-tech stuff, and answering questions that I should have already known. Thanks to Dilara Ally for shaping my beliefs on what \"good science\" is all about. Thanks to Charles Chin-Lin Chen, the bumbos worker, the Saltminer, and the inspiring fountain of knowledge for shaping my mandarin vocabulary, helping me with computer programs, discussing details on my data that I should have done, and for the words, \"NO TIME! SO SERIOUS! IT TRUE!\" T-man was instrumental in my travels on the Island and in the lab for his love of Riverdance helped me get through some of the most trying times in the Saltmines. Shirley Pang's ability to order chemicals on very little notice cannot be overlooked - thank you Shirley for putting up with my lapses of disorganization. I'd like to thank Carol Goodwille and Dawn Marshall, two post-docs that I could always depend on when the other grad students just didn't know. Thanks also to Dr. Fred Ganders and Dr. Sally Aitken for their many insightful comments on this thesis. The second to last thank you goes out to Mr. Vincent Lai for allowing me to be close to U.B.C. and teaching me how to be a good cleaner. Lastly, I would like to thank my friends, both new and old, because they are the ones that helped spur me to finish line so that we could all have one big party in Japan. Thanks everybody. 1 Chapter I: Introduction 1.1 Restoration and its Importance Over the last few years, ecological restoration has been a major item on the agenda for many biologists, land managers, and naturalists (Dobson et al. 1997; Havens 1998; Lavendel 1999; Millar and Libby 1989; Montalvo et al. 1997; Muller et al. 1998; Noble and Dirzo 1997; Yaffee 1999). Land disturbances such as logging, mining, and road construction have opened up an important niche that must not go unfilled (Veverka 1998). These industrial activities and countless other disturbances have the potential to reduce and eliminate native vegetation used by a diverse array of organisms as forage and habitat material. Plant species native to the environment help deliver nutrients and stability to the soil in order to promote the growth of a natural healthy ecosystem while preventing erosion and sediment seepage into delicate streams essential to the surrounding wildlife. More importantly, the conversion of natural habitat into industrial and agricultural landscapes poses a great threat to the natural biodiversity of the surrounding plant community (Franklin 1993; Millar and Libby 1989). Research must be done in this field in order to provide adequate restoration of these priceless ecosystems as well as to recreate and re-establish a self sustainable plant community. In the past, disturbances by anthropogenic activities such as the ones listed above have been ameliorated using plant species shown to be robust in many different circumstances (Eliason and Allen 1997; Richardson 1998). These species are generally 2 chosen for convenience, either through availability or economic cost. Unfortunately, these attributes are usually possessed by species that are non-native to the restoration site. Introductions like these have been termed \"exotics\" (Millar and Libby 1989). The introduction of \"exotic\" species has often resulted in adverse effects on the \"natural\" ecosystem to which they were introduced. Evidence supporting this claim can be observed by the domination of landscapes by weedy exotic plants competing with existing native species. The literature abounds with cases describing how non-native noxious species have invaded ecosystems. For example, disturbance of coastal sage scrub in southern California has led to extensive displacement by exotic grasses (Eliason and Allen 1997). There have been widespread invasions by newly introduced plant species to the Tiwi islands in northern Australia (Fensham and Cowie 1998), some having the potential of degrading large areas of habitat essential for wildlife. Native mussels along the coast of southern California have declined due to invasive species from the north (Geller 1999). These are just a few examples in the literature that describe the potential hazards of introducing exotic species into the environment. Another hazard associated with the introduction of exotic species is the contamination of existing gene pools. This can potentially result in the disruption of adaptive gene complexes (Huxel 1999; Slatkin 1987) and the inadvertent creation of new species by the hybridization of exotics with natives (Ayres et al. 1999). The potential invasiveness of exotic plant species is another concern for biologists, land managers, and naturalists alike because competitors (herbivores, parasites, and pathogens) acting to 3 control these species, present in their endemic habitats, may not be present in their \"newly\" acquired habitats (Higgins et al. 1999). Ultimately, \"exotic\" introductions have the potential of disrupting the path of evolution that has been occurring for thousands of years. The above factors bring up many questions about the integrity of the proposed restoration site: Can we call an ecosystem that contains \"exotic\" components a \"natural\" ecosystem? Scientist such as Millar and Libby (1989) and Knapp and Rice (1994) have introduced genetic awareness regarding the introduction of exotic species. They question the validity of settling for short-term success rather than addressing the long-term consequences. If exotic species are satisfactory in achieving soil stability and general \"green up\", what long-term problems are we ready to settle for? How long can we continue to exploit our natural resources while replacing the void with species that were not present to begin with? How will the surrounding wildlife adapt to these new environments? The forestry industry in British Columbia has long recognized this problem and thus has been replanting tree species indigenous to the land. This method of restoration has been enhanced through the implementation of tree seed zones (Meidinger and Pojar 1991). Seed zones are developed through the use of a classification system which matches germplasm to geography, environmental history, and ecology of the land. These units are measured by a system known as the biogeoclimatic ecosystem classification 4 (BEC) (Meidinger and Pojar 1991). This type of classification is important because logged sites will now be restored with tree seedlings most compatible to the site, being from seeds of trees that had existed in a similar environment. The classification helps to match new seedlings to the restoration site similar to the ones their parents had previously flourished. New recruits are assumed to possess alleles, and\/or adaptive gene complexes suited for survival and healthy growth in their respective environments. Seed zones have been developed to match the adaptive potential of germplasm to the geographic region in which it is being planted. This has been an intense area of research for the forestry industry because as logging pressures increase, there is an increasing push to maintain the natural biodiversity of disturbed sites while simultaneously developing a healthy timber supply for the next generation as quickly and as efficiently as possible (Veverka 1998). It now seems rather logical that it is not only trees that need seed zones for successful restoration to occur, they need to be developed for the surrounding vegetation as well (shrubs, herbs, brushes, and grasses). These plants are essential for soil stability, erosion control, nutrient fixing, and in creating natural habitats and forage material for wildlife that depend on these plants for survival (Bratton and Meier 1998; Franklin 1993; Kinderscher and Tieszen 1998; Yaffee 1999). It has been determined through numerous population genetic studies that most plant species have genotypes possessing some form of population structuring. Genetic structure, described by Loveless and Hamrick (1984), refers to the non-random distribution of alleles and genotypes in space. It describes the degree of genetic differentiation both within and between populations and also provides information related to gene flow, genetic drift, and adaptation to microhabitats (Silander 1984; Slatkin and Barton 1989; Snaydon and Davies 1982; Wright 1978). These factors help to mold populations into sometimes very distinct groups within a common species and knowledge of genetic structure can aid one in defining the size or spatial extent of a randomly breeding population. Different groups separated by geographic distance posses various degrees of ecological interactions and adaptive potential suited to their landscape. For these reasons, knowledge and understanding of how alleles and genotypes are distributed within and among populations is fundamental, (Hamrick et al. 1991) especially before implementing restoration projects (as in this case) with native grass for British Columbia's diverse landscape. Recently, the keen interest in using grass species native to the environment for landscape restoration has resulted in a number of studies examining restoration, establishment, and genetics related to native grasses (Biedenbender and Roundy 1996; Bugg et al. 1997; Claassen and Marler 1998; Kinderscher and Tieszen 1998; Knapp and Rice 1994; Knapp and Rice 1996; Knapp and Rice 1998; Miao et al. 1997; Muller et al. 1998; Rosales et al. 1997; Walters et al. 1994; Williams and Davis 1996). Currently, most of the herbaceous plants used for re-vegetation consist primarily of domesticated grasses and legumes, mainly of European origin (Eliason and Allen 1997). The exotic plants used are inexpensive as well as fast growing which is why they are sown every year on thousands of hectares of roadsides, grazing lands, mine tailings, and clear cuts. However, as stated above, there is much concern and controversy as to what the impact 6 of these exotic species are capable of doing to the natural biodiversity in the long term (Higgins etal. 1999; Richardson 1998). The trend of utilizing native grasses for restoration is a positive step towards recreating the natural flora that once inhabited the landscape. However, this solution can not be pursued blindly. Because these programs are still in their infancy, one must always proceed with caution when developing and utilizing commercial sources of native grasses or any species for restoration purposes. This caution stems from the fact that the seeds used may contain only a limited amount of genetic variability or an unsuitable genetic composition needed for the project at hand (Havens 1998; Knapp and Rice 1994). Maladapted genotypes may be inappropriate or not optimal for the survival of certain germplasm once sown in a particular geographic region or habitat (Helenurm 1998). Seeds should therefore be collected and sown with consideration given to the genetic structure of the species. Doing so will allow one to \"customize\" the selection of germplasm for the particular restoration site of interest. The study of genetic variation therefore comes to the forefront because the success of future restoration projects may rely upon knowledge of the biology and population genetics of the candidate plant species. This information would greatly help restoration projects become more efficient and successful because they would aid in the development of seed zones for the transfer and use of various species over a wide geographic range. 7 1.2 Elymus glaucus In California, Oregon, and Washington, Elymus glaucus, commonly named blue wild rye, has been effectively used for restoration of disturbed habitats because it naturally resides in open clearings and disturbed sites. Blue wild rye has been characterized as a non-rhizomatous, allotetraploid species (2n=28) (Dewey 1968; Knapp and Rice 1996). It has been proposed through chromosomal and cytological studies that this highly variable grass most likely arose as a hybrid of Hordeum and Agropyron (Dewey 1968; Jensen 1993). As a member of the Poaceae family, it is a tufted perennial bunchgrass capable of growing between 50 cm and 150 cm tall. Foliage from this native West Coast grass is green to a purple-like glaucous colour (personal observation). They possess lax leaves of rough texture ranging between 5 cm and 20 cm long and 5 cm to 12 cm wide (Pojar and MacKinnon 1994). The defining characteristic of this grass species can be seen by its sometimes glaucus appearance and fairly distinct inflorescence which possesses a dense spike between 5cm to 20 cm long with awned or unawned lemmas (Pojar and MacKinnon 1994). Its importance in soil stability and erosion control has been exemplified by its vigorous, potentially deep, fibrous and well branched root system (Hoover era\/. 1948). There has been some research into the morphological variation and systematics of Elymus glaucus. Douglas et al. (1994) have recognized two subspecies: 1. Elymus glaucus Buckley ssp. virescens (Piper) A. Love 2. Elymus glaucus Buckley ssp. glaucus. Hitchcock (1971) gives one variety of the species as: Elymus glaucus var. jepsonii Davy while Taylor and MacBryde (1977) acknowledge three varieties: 1. Elymus glaucus Buckley var. breviaristatus Davy in Jepson 2. Elymus glaucus Buckley var. glaucus 3. Elymus glaucus Buckley war. jepsonii Davy in Jepson. Elymus glaucus has been known for its widespread distribution along the west coast of North America ranging from Southern Alaska to Southern California. In British Columbia, it is found in many diverse biogeoclimatic zones including the boreal white and black spruce (BWBS), coastal Douglas-fir (CDF), and coastal western hemlock (CWH) zones. This range has been determined for var. breviaristatus. Variety glaucus has been found in the mountain hemlock zone (MH), sub boreal spruce (SBS), interior douglas fir (IDF), sub boreal pine-spruce (SBPS), interior cedar hemlock (ICH), ponderosa pine bunchgrass (PPBG), CDF, and C W H zones. Variety jepsonii can also be found in the CDF and C W H zones (Taylor and MacBryde 1977). One reason for using this native grass for restoration and reclamation of disturbed habitats is because of its wide ecological range. Landscapes inhabited by E. glaucus include those ranging from dry to moist, open to shaded, and clay loam to sandy loam soils. This grass also resides in open coniferous and deciduous forests, rocky slopes, and 9 clearings from low to middle elevations (Hitchcock 1971). Elymus glaucus is common to meadows and riparian areas, copses and open forests, avalanche chutes, old burns, and cut-overs. Although it grows well throughout the west coast, it does have a tendency to decrease in frequency with increasing elevation (Klinka et al. 1989). Hassel noticed that E. glaucus grows in soils ranging from pH 5.5 to pH 7.5, shows moderate sensitivity to saline soils, and does not grow very well in shallow soils. Another important feature of its ecological distribution is that it rarely grows in pure stands. Stands are predominately mixed and populations seem to possess a patchy distribution with potentially large distances between populations (Knapp and Rice 1996 and personal observation). Because it is able to inhabit a variety of diverse environments alongside other plants, Elymus glaucus is an ideal species for re-vegetation and soil stabilization on cleared disturbed land. 1.3 Objective British Columbia is known for its geographic and ecological heterogeneity (Meidinger and Pojar 1991). In order for land managers to be successful in using native plant species for restoration, they must first investigate their natural biology and ecology. These encompass an organism's genetic structure, gene flow and mating system. By investigating these genetic parameters, one can infer patterns of selection, migration, and distribution, which can then be used to help implement the successful transfer and development of germplasm. 10 Genetic structure is the measure of genetic differentiation found among and within populations. This can be detected using neutral genetic markers such as isozymes. Isozymes were chosen for this study because they have shown to be relatively neutral, have already been optimized for this species (Knapp and Rice 1996) and are cost effective. Genetic differentiation found among natural populations can occur through two major mechanisms, random drift and selection, and these genetic markers will help to define these evolutionary factors (Haiti and Clark 1997). Because E. glaucus is an early successional species which occupies a wide range of habitats, it would be assumed that there is a high degree of selective divergence between populations. A test that measures the degree of selective divergence was introduced by Lande (1976). This test compares the observed additive genetic variance between populations with that expected from random drift. Lynch (1988), however, indicated that the calculation for expected variance requires the estimation of several unknown parameters such as mutation rate, time since divergence, and the effective population size. A n alternative approach when assessing population differentiation for morphological traits (Q s t) is to compare them to the distribution of neutral genetic markers such as isozymes (F s t); (Felsenstein 1986; Lande 1992; Rogers 1986; Spitze 1993; Wright 1951). Divergent selection for a morphological trait can be invoked i f Q st is greater than F s t . On the other hand, if morphological differentiation is measured to be 11 equal to or less than F s t , among population variance due to random genetic drift cannot be rejected and convergent and\/or stabilizing selection may be invoked as the cause for the reduction of differentiation of these traits. A study done by Knapp and Rice (1996) found a high level of F s t j low gene flow, and very high degree of homozygosity for E. glaucus populations in Washington, Oregon, and California. These results were coupled with low geographic correlation to genetic distance. My study will focus on whether the genetic structure of B.C. populations of E. glaucus are consistent with respect to Knapp and Rice's (1996) findings and will then attempt to interpret the evolutionary story of blue wild-rye. This will be aided by directly examining outcrossing rates of three Vancouver Island populations and comparing quantitative traits with isozymes. Results of this study will provide information towards seed transfer guidelines and the adaptive potential of this native grass species for the reclamation of disturbed habitats. 12 Chapter II: Materials and Methods 2.1 Seed Collections Seed from E. glaucus populations 1-18 were collected by Dr. Phil Burton and his associates at Symbios Research and Restoration. Seed collected by Symbios Research and Restoration were restricted to the zones between 52 degrees north and 60 degrees north latitude, and between the crests of the Coast Range and Rocky Mountains (Figure 1). I made collections in the Lower Mainland and on Vancouver Island (19-26, 34-36). Populations 19-21 were collected in December of 1997 while populations 22-26 were made in October of 1998. A description of the locations and general geography of these samples can be found in Table Ila and Figure 1. The objective when sampling E. glaucus populations was to obtain population representation from many different biogeoclimatic subzones within British Columbia. This would ensure a thorough investigation of the genetic structuring involved in shaping E. glaucus populations because many of British Columbia's temperature and precipitation regimes would have been represented in the collections. 13 Table Ha. Type of data measured (isozyme, morphological, or both), population name, Biogeoclimatic Ecosystem Classification (Meidinger and Pojar 1991), forest district, and elevation of 20 B.C. populations of Elymus glaucus. (both indicates morphological and isozyme studies). Data Type Population BEC subzone Forest District Elevation (m) Both 1 ICHmc3 Bulkley 802 Both 2 SBSmc Morice 922 Both 3 ICHvc Kalum 240 Both 4 SBSdk Morice 750 Both 5 ICHmc2 Kispiox 300 Both 6 SBSmc Lakes 653 Both 7 SBSdk Bulkley 550 Both 8 SBSmc Morice 1178 Both 9 SBSmw Prince George 900 Both 10 BWBSdkl Cassiar 924 Both 11 ICHmc3 Kalum 200 Both 12 ESSFmc Lakes 1224 Both 13 ICHvc Cassiar 732 Both 14 ICHwk Quesnel 960 Both 15 SBSmc Bulkley 840 Both 16 SBSdk Vanderhoof 854 Both 17 SBSmw Quesnel 809 Isozymes 18 ICHmc2 Kispiox 196 Isozymes 19 CDFmm Georgia Lowlands 75 Isozymes 20 CDFmm Georgia Lowlands 50 Isozymes 21 CWHvml Maritime 50 Isozymes 22 CDFmm Georgia Lowlands 75 Isozymes 23 CWHxml Georgia Lowlands 100 Isozymes 24 CWHxm2 Maritime 450 Isozymes 25 CWHvml Maritime 700 Isozymes 26 CWHvhl Maritime 0 Isozymes 34 CWHxm2 Maritime 180 Isozymes 35 CWHxm2 Maritime 200 Isozymes 36 CWHvhl Maritime 45 Figure 1. General locations of all Elymus glaucus populations sampled within British Columbia 15 Within each population, seeds were collected from individual culms over a 10 to 50 m collection radius. Seeds from individual culms representing populations 1 through 23, and 34 - 36 were pooled by their respective populations. From this pooled seed source, 30 randomly chosen seeds were used for electrophoresis. Populations 24, 25, and 26 were collected as progeny arrays in order to calculate outcrossing rates. 2.2 Electrophoresis Elymus glaucus seed was stored in either plastic film containers or put into small envelopes and stored at 5 \u00b0C. Seeds were germinated in petri plates lined with moistened Whatman #1 filter paper for two weeks. The enzymes examined in this study were those expressed in the tissue of 14 day old seedlings. Seedlings were ground separately in depressions drilled into a plexiglass tray using a flat bottom 1cm diameter stainless steel rod attached to a power drill. Tissue homogenization was performed using 2 drops of chilled Tris-HCL grinding buffer (Gottlieb 1981) over crushed ice. The extract was then covered with parafilm and frozen at -80 \u00b0C until assayed. Frozen extracts were allowed to thaw at room temperature and then absorbed into 2mm X 10mm wicks cut from Whatman #3 filter paper. Three populations were represented on each gel so that individual allele differences among populations could be discerned more accurately. Twelve percent starch gels, 2 running systems (tris citrate and sodium borate), and eight different enzyme stains were used in order to resolve the 21 loci observed (Table A7). Gels using the tris-citrate buffer system (Meizel and Markert 1967) were run at. 125 volts and 55 mA for 5 hours, while gels using the sodium borate buffer system (Poulik 1957) ran for 6.5 hours at 175 V and 60mA. Gels were run at 4 \u00b0C inside a refrigerator. Ice packs were placed on top of the gels once the wicks had been removed (45 minutes into the run) in order to aid in cooling during electrophoresis. Stain recipes were taken from Hillis et al. (1996) (Table A7). Although banding patterns seemed complicated for some of the loci, they were more easily interpreted after discussions with Kermit and Carol Ritland, Charles Chen, Lisa O'Connell, and Daniel Perry. Information regarding basic sub-unit structure also aided in scoring the various band patterns observed (Hillis, Moritz and Mable 1996). 2.3 Genetic Diversity Genetic diversity parameters were calculated using Biosys-2 (Swofford and Selander 1997). Diversity measures included percent of polymorphic loci (PLP), mean number of alleles per locus (A\/), and expected heterozygosity (He). Between and within populations measures of genetic diversity were partitioned using Biosys-2 in accordance to Wright (1978). Hierarchical analysis of F s t was calculated by nesting populations into different geographic and environmental variables. Populations were grouped by biogeoclimatic zone, elevation, mean yearly snow pack, mean number of frost free days in a year, predominant soil type, mean yearly precipitation, and mean yearly temperature. Grouping by forest district was also compared to see how geographic distance relates to 17 genetic distance. The population composition of these environmental and geographic groups can be found in the appendix (Tables A l to A6). 2.4 Gene Flow Gene flow was indirectly estimated using Slatkin and Barton's (1989) manipulation of Wright's F s t (Wright 1978). Populations 24, 25, and 26 were used to calculate outcrossing rates. Single and multilocus outcrossing rates were calculated using Mltr 1.1 (Ritland 1996) which uses progeny arrays to estimate both average single locus and multi locus population outcrossing rates. Outcrossing rates were calculated using 11 toi3 isozyme loci (Mdh, Idh, Skd, Dia, and Pgi). 2.5 Genetic Distance Allele frequencies were calculated using Biosys-2 (Swofford and Selander 1997) and input into Phylip 3.5c (Felsenstein 1993) in order to produce a U P G M A dendogram of genetic distances according to Nei's genetic distance (1978). This method of calculating genetic distance assumes that distances between populations arise from genetic drift. It is formulated for an infinite isoalleles model of mutation in which there is a rate of selectively neutral mutation and each mutation is assumed to result in a completely new allele. A l l loci examined are assumed to be under the same constraint of neutral mutation. Another assumption this model makes is that the initial genetic 18 variability in the population is at equilibrium between mutation and genetic drift, with the effective population size of each population remaining constant. A Mantel - Nei test was done using M A N T E L (Liedloff 1999). This test examines the correlation between genetic and geographic distance by creating a matrix based on pair-wise genetic and geographic distances. Geographic distances between populations were estimated by plotting a population's location according to its longitude and latitude. These points were then used to measure distances between populations. 2.6 Common Garden Common garden experiments are necessary when one wants to measure morphological variation due to genetic causes because variance attributed to environmental variation can be removed. Quantitative trait variation can be divided into two basic parts. One type of variation is due to environmental effects while the other is due to genetic effects. Environmental effects are negligible because the common garden provides a relatively constant environment for all accessions. This leaves us with predominately the genetic effects which is what we are interested in quantifying in this study. Dr. Phil Burton and his associates at Symbios Research and Restoration collected, planted, and maintained populations 1-17 and 27-33 in a common garden \/ seed orchard for the morphological analysis. Procedures in developing the common garden \/ seed 19 orchard are as follows. Elymus glaucus seeds were allowed to germinate in a green house for 6 to 10 weeks and then transplanted into the field during the spring and early summer of 1997. Seedlings were spaced 20 cm apart with an 80 cm space between each row in order to provide access for weeding and measurements according to the following design. Tim Mullin of Genesis Forest Science Canada Inc. prepared the orchard design. He prepared a generic design for 30 accessions consisting of 1200 plants using a computer program that calculates a permutated neighborhood design (Bell and Fletcher 1978). In order to maximize the diversity of accessions in the immediate neighborhood of each individual plant, this program assigns accession numbers to randomly spaced planting positions. This was done to maximize outcrossing between sampled populations. Each accession represents a different population. These planting positions consists of three alternating or triangularly spaced rows under the constraint that each plant is surrounded by an immediate neighborhood of 20 unique accessions and that the same accession did not occur as an immediate diagonal neighbor any more than 10 times. This randomized design also ensures that any environmental variation found within the common garden is distributed evenly to all accessions. 2.7 Morphological Measurements In July of 1998,1 went up to Smithers British Columbia in order to measure morphological traits of 20 individuals grown in the common garden from the populations listed in Table lib. The plants representing each population were chosen by randomly sampling 1 culm from each individual planted in different areas of the common garden. The traits measured on each plant are listed in Table lie and Table l id and illustrated in Figure 2. A l l 480 individuals were characterized and measured within a four day time span. 21 Table l i b . Population name, Biogeoclimatic Ecosystem Classification, Forest District, and Elevation in meters. Population BEC Subzone Forest District Elevation (m) 1 ICHmc3 Bulkley 802 2 SBSmc Morice 922 3 ICHvc Kalum 240 4 SBSdk Morice 750 5 ICHmc2 Kispiox 300 6 SBSmc Lakes 653 7 SBSdk Bulkley 550 8 SBSmc Morice 1178 9 SBSmw Prince George 900 10 BWBSdkl Cassiar 924 11 ICHmc3 Kalum 200 12 ESSFmc Lakes 1224 13 ICHvc Cassiar 732 14 ICHwk Quesnel 960 15 SBSmc Bulkley 840 16 SBSdk Vanderhoof 854 17 SBSmw Quesnel 809 27 ESSFmc Morice 1362 28 SBSdk Lakes 914 29 BWBSdk2 Cassiar 662 30 ICHmc2 Kispiox 447 31 ESSFwv Bulkley 1006 32 ESSFwk Quesnel 1190 33 SBSmk Quesnel 793 22 Table He. Description of continuous traits measured in plants grown in the common garden. Trait Measured Description Measurement Units Stem diameter Stem height Leaf length to width ratio Awn length Spike length(mm) Floret length(mm) below the first leaf millimeters soil to the base of the spike millimeters second leaf from base tip of floret to the end of the awn millimeters base of spike to the top of spike millimeters randomly chosen from the spike millimeters 23 Figure 2. Diagram of Elymus glaucus illustrating the different traits measured and characterized. 24 2.8 Data Analysis Quantitative traits were used in order to calculate Q s t , an index similar to that of F s t which partitions diversity both between and within populations (Podolosky and Holtsford 1995; Prout and Barker 1993; Spitze 1993; Yang et al. 1996). Q s t values for both continuous and discontinuous traits were estimated by doing an analysis of variance (ANOVA). A N O V A partitions the variance of these morphological traits within (Q 2 W) and between populations (Q 2 b). Wright's (1965) paper on population structure analysis shows how his F-statistics partitions total additive variance (Q2 t) into the among population component (Q2 D) and into the within population component (Q 2 W). Wright (1951) described that characters with a purely additive and neutrally evolving genetic basis had: 1) Q 2 w = ( l - Q s t ) Q 2 t Q 2 t represents the total genetic variance where all individuals are part of a panmictic population (Wright 1951). He also noted that: 2) Q 2 b = ( 2 Q s t ) Q 2 t Re-arranging equations 1 and 2 gives: 25 3) Qst = Q 2 b \/ ( O A + 2Q 2 W) (Lande 1992) The neutral expectation for Q s t is F s t for single locus genes in the same population (Felsenstein 1986; Lande 1992). Q s t was calculated for each trait and then averaged to give a final value for all the populations sampled. 26 Table Hd. Description of discontinuous traits characterized by the assumption of dominance and recessiveness. Trait Characterized Presence Absence Awn Serration Dominant Recessive Waxy coating on the stem Dominant Recessive Stem pubescence Dominant Recessive Waxy residue on the Spike Dominant Recessive Purple pigmentation on the first ligule Dominant Recessive Purple pigmentation in the spike Dominant Recessive Purple pigmentation in the leaf Dominant Recessive Purple pigmentation in the stem Dominant Recessive 27 Q s t can also be determined differently for traits that were quantified by the presence or absence of a phenotype (Discontinuous traits). This will be regarded as d* Q s t . Dominance of each discontinuous trait was inferred to be the presence of a certain phenotype (Table lid). The presence or absence of stem pubescence has been studied in Avena fatua by Jain and Marshall (1967) and they have shown that the presence of pubescence is dominant and under monogenic control. Presence or absence of colour spots (equivalent to purple ligule) have also been studied and the presence of colour was assumed to be dominant (Coffman 1961). Allele frequencies were calculated by inferring that the absence of a particular trait was caused by being homozygous recessive for that gene. Because E. glaucus has not been shown to be randomly mating (Knapp and Pace 1996), the assumption that this species does not violate Hardy Weinburg equilibrium can not be made. In order to incorporate non-random mating bias, the following calculations d* were made for Q st : The inbreeding coefficient (F;s) for E. glaucus is defined by: 4) F i s = ( H 0 - H O \/ H 0 Fj s measures the fractional reduction in heterozygosity of inbred populations relative to those that are randomly mating and which share the same allele frequencies. H 0 is defined as the expected proportion of heterozygotes and Hj describes the observed 28 frequency of heterozygosity. Because H 0 is equivalent to 2pq, the frequency of heterozygous genotypes in an inbred population can be re-written as: 5) Hi = 2 M ( l - F i s ) The frequency of A A homozygotes can also be written in terms of Fj s . If, for example, the proportion of A A genotypes is P and the allele frequency of A is p, therefore: 6) P + Hi \/ 2 = p If Hi = 2pq(l-\u00a5\\s), the equation can be written as: 7) ?=p-2pq(\\-F}S)\/2 From this equation, allele frequencies can be derived from the recessive genotypes and then used to calculate allele frequencies for the dominant allele (Hartl and Clark 1997). d* These allele frequencies were then used to calculate Q st by using Weir and Cockerham's (1984) estimation of Fst- This was done by calculating the variance of q 29 between populations per trait and then dividing this value by the product of the mean variance of p and q. This calculation was done for each trait and then averaged to give a d* final Q st for all the populations sampled. A correlation matrix was constructed using (JMP 1996) in order to compare the different traits with one another and to see whether there were any significant patterns shared between morphological characteristics and the environment. 30 Chapter III Results 3.1 Germination Seed germination was variable among the E. glaucus populations that were sampled. Fortunately a sufficient number of seedlings had germinated in order to be used for studying their genetic diversity. Collections made in September and October of 1997 germinated well with an approximate 80% success rate, while those collected later in the season (November and December of 1998) had a much lower germination success. Germination rates ranged approximately between 0% and 40%. It seems most likely that the majority of the filled seeds had already fallen by November and December. Collections made in late summer and early fall provide the greatest abundance of viable seeds. 3.2 Gel Scoring Initial scoring of isozyme banding patterns were complicated because of the allotetraploid nature of this species (2n = 28) (Dewey 1968; Jensen 1993). Evidence for allopolyploidy was observed by the appearance of several loci displaying homoallelic mobility and fixed heterozygosity. Fixed heterozygosity was observed for malate dehydrogenase, diaphorase, phosphoglucoisomerase, glutamate dehydrogenase, and alcohol dehydrogenase. This is the result of the interaction between two different monomorphic loci from separate genomes that have alleles for enzymes sharing similar 31 structure and function (Jaaska 1992). These alleles may possess slight differences in charge, size, and conformation indicated by the difference in mobility on the gel. Assuming that both alleles are able to interact and form a dimeric protein product, three bands would appear instead of two because of the protein-protein interaction. This phenotype would appear similar to a dimeric enzyme in its heterozygous form except that, in this case, all individuals (or a large proportion of individuals) would appear heterozygous in populations possessing fixed heterozygosity. Difficulty in scoring allotetraploids arises when a certain allele migrates to the same location as a different allele from a different locus. These bands are deemed to have homoallelic mobility. Homoallelic mobility is the result of proteins sharing a similar charge, size, and\/or conformation. This makes it difficult when attempting to discern the genotypes of certain individuals at these loci. This was apparent when staining for shikimate dehydrogenase and glutamate dehydrogenase (Figures A2 and A8). These loci were scored by comparing differences in banding intensities along with the presence or absence of fast or slow alleles. Darker bands in combination with absent or weaker banding indicates that alleles of different loci were overlapping (Illustrations of both fixed heterozygosity and homoallelic mobility can be found in the appendix Figures A l to A8). 32 3.3 Genetic Variation At the species level, E. glaucus displayed a moderately high degree of isozyme variation. Evidence for the high species diversity comes from having 78% of the loci screened being polymorphic (Ps). The mean number of alleles per locus was 2.0 alleles per locus (As) (Table Illb). Less diversity was observed within populations than between populations. The percent of polymorphic loci at the population level was only 22% (Pp) while the average number of alleles per locus for populations was 1.31 (Ap) (Table Illb). Table I l i a . Measures of Genetic Diversity in 26 Populations of Elymus glaucus; mean sample size per locus, mean number of alleles per locus, percentage of polymorphic loci*, and observed and expected heterozygosity. Mean no. Percentage Mean Heterozygosity Of alleles of loci Direct- HdyWbg Population Samplesize Per Locus Polymorphic* count expected Popl 32 1.3 (.1) 23.8 .049 (.029) .092 (.033) Pop2 32 1.4 (.1) 28.6 .055 (.055) .125 (.041) Pop3 16 1.3 (.1) 33.3 .006 (.006) .070 (.023) Pop4 32 1.4 (.1) 19.0 .010 (.004) .054 (.024) Pop5 32 1.3 (.1) 14.3 .004 (.003) .057 (.030) Pop6 32 1.4 (.1) 28.6 .021 (.011) .064 (.021) Pop7 32 1.3 (.1) 28.6 .013 (.009) .048 (.016) Pop8 32 1.2 (.1) 9.5 .013 (.008) .034 (.019) Pop9 32 1.2 (.1) 14.3 .013 (.011) .027 (.012) Pop 10 8 1.3 (.1) 33.3 .018 (.018) .128 (.044) Pop l l 32 1.3 (.1) 14.3 .033 (.023) .036 (.018) Pop 12 32 1.2 (.1) 4.8 .009 (.006) .012 (.006) Pop 13 32 1.4 (.1) 28.6 .012 (.008) .072 (.028) Pop 14 32 1.3 (.1) 9.5 .058 (.033) .057 (.030) Pop 15 32 1.2 (.1) 9.5 .027 (.017) .030 (.015) Pop 16 32 1.5 (.1) 38.1 .080 (.046) .176 (.050) Pop 17 32 1.4 (.1) 33.3 .045 (.037) .092 (.033) Pop 18 32 1.6 (.1) 52.4 .037 (.019) .193 (.044) Pop 19 32 1.2 (.1) 14.3 .001 (.001) .060 (.031) Pop20 32 1.2 (.1) 9.5 .000 (.000) .020 (.010) Pop21 32 1.2 (.1) 19.0 .006 (.006) .066 (.031) Pop22 32 1.2 (.1) 4.8 .010 (.006) .021 (.012) Pop23 32 1.2 (.1) 14.3 .009 (.005) .077 (.039) Pop34 32 1.4 (.1) 38.1 .030 (.016) .144 (.043) Pop35 32 1.4 (.1) 33.3 .027 (.015) .139 (.045) Pop36 32 1.2 (.1) 19.0 .012 (.010) .058 (.030) * A locus is considered polymorphic if the frequency of the most common allele does not exceed 0.95. Standard errors are in parentheses. 34 3.4 Genetic Structure The high F s t value of 0.65 (Table Mb) indicates that the majority of genetic variation is distributed among rather that within populations. Approximately 65% of the total estimated variation observed is partitioned among populations while the remaining 35%) is accounted for within populations. Populations of E. glaucus were collected from a wide geographic range covering a diversity of terrain. Using Biosys-2 (Swofford and Selander 1997) a hierarchical analysis of F s t was performed to see how genetic variation is partitioned among populations nested within similar environments (Table IIIc). Hierarchical F s t results from Table IIIc are defined by the proportion of variation attributed to the different levels in the hierarchy. When looking at populations nested within environmental and geographic factors, it seems as though much of the variation is distributed between populations nested within factors. For example, i f I used B E C as the factor, when nesting individual populations within common BEC's, approximately 60% of the variation found within the species is distributed among populations within the factors while the remaining 40% can be accounted for between the different biogeoclimatic zones. Comparison 2 in Table IIIc depicts the genetic variation partitioned between and within factors. Looking at the ratios, there does not seem to be 35 much differentiation between factors however, much of the diversity detected within this species can be found within factors. Table IHb. Isozyme diversity for Elymus glaucus at 21 loci at the species, within population, and among population levels. Parameters used were percentage of polymorphic loci (PLP), mean number of alleles per locus (A\/), expected heterozygosity within populations(Hs), total heterozygosity (H t), and F-statistics. Levels P L P A \/ H t H s Fis Fit F s t Species 78% 2.0 0.065 0.023 0.68 .89 .65 Population 22% 1.3 n\/a n\/a n\/a n\/a n\/a Table IIIc. Hierarchical F s t for populations nested within Factors and Factors nested within the Total. Comparison 1 Comparison 2 Factor Populations nested within Factor Factor nested within Total BEC 0.59 0.12 Elevation 0.63 0.02 Snowpack 0.62 0.05 Frost free days 0.63 0.02 Soil type 0.64 0.01 Precipitation 0.62 0.04 Temperature 0.62 0.05 Forest District 0.60 0.10 37 3 . 5 Gene Flow The estimate of gene flow (Nw) between populations was low (Nm = 0.406). With low levels of gene flow (Nm < < 1), populations are geographically isolated from each other and are able to diverge genetically through drift, whereas at high levels (Nm > > 1), gene flow between populations will overcome the effects of drift and prevent local population differentiation (Slatkin 1994; Wright 1951). Outcrossing rates were calculated and the results of this analysis found that the mean multi-locus and single locus t-estimates were 0.55 and 0.46 respectively (Table Hid). These results indicate that E. glaucus samples collected from Vancouver Island have an outcrossing rate of approximately 50%. 38 Table H i d . Population Outcrossing Rates. Population name, no. of families assayed, number of loci, multilocus t-estimate (tm), singlelocus t-estimate (ts), and difference between multi and singlelocus t estimates (Ritland 1996). Population # families # loci tm ts tm-ts 24 7 13 .478 (.159) .408 (.167) .070 (.035) 25 9 13 .706 (.206) .563 (.158) .143 (.087) 26 11 11 .475(201) .409 (208) .066 (.007) Mean 9 12.3 .554 (.102) .460 (.090) .090 (.012) 39 3.6 Genetic Distance U P G M A cluster analysis using (Nei 1978) genetic distance was performed and the corresponding dehdogram is shown in Figure 3. From examining this tree, there does not seem to be any order in which populations are distributed, however, island populations (34, 35, and 36) seem to be grouped closely together. Generally, populations collected within the same forest district seem to be no more similar than when compared to populations from other districts. Further support for this conclusion can be drawn from the Mantel test which yielded a significant yet weak correlation (p = 0.01; r = 0.14) between geographic and genetic distance. 40 .60 .67 + + + ^ .73 t- + Similarity .80 .87 -+ +-.93 -+ +-1.00 ******** * * * *** POP 01 * POP 08 * * * * * * * *** pop 09 * * * ** POP 15 * ** POP 13 **** * ***** POP 17 * * * * * * * * ***** pop 19 * * * * ***** pop 02 * * * * * * * * * * * POP 03 * * * * ** POP 04 ************* POP 23 ********** * * * * * * * * * * ***************** pop 22 fc- * * ******* pop 10 * * * * * * * * * * * * * ******* pop 16 * * **** * pop 11 * * * * * * * * * * * * ***** * POP 12 * * * *** ********** POP 14 * ************** pop 18 * ************* pop 07 * ********* * * ************* pop 21 * * * * * * * **** pop 34 * * * ********************* pop 35 * * * * * * * * * * * * * * * * * * * * * * \" * * *********************************** pop 20 * * * ************************************* POP 05 * i r * * * * * * * * * * * * * * * pop 06 ***** pop 3 6 * * * * * * * * * * * * * * * * * * * * * * * B u l k l e y ICHmc3 Morice SBSmc2 P.G. SBSmw Bu l k l e y SBSmc2 Cas s i a r ICHvc Quesnel SBSmw Is l a n d CDFmm Morice SBSmc2 Kalum ICHvc Morice SBSdk I s l a n d CWHxml I s l a n d CDFmm Ca s s i a r BWBSdkl Vander SBSdk Kalum ICHmc3 lakes ESSFmc Quesnel ICHwk4 K i s p i o x ICHmc2 Bu l k l e y SBSdk Van CWHvml I s l a n d CWHxm2 Is l a n d CWHxm2 Is l a n d CWHvhl Is l a n d CDFmm Ki s p i o x ICHmc2 Lakes SBSmc2 + +--. 60 -+ +-. 67 -+ +-.73 -+ +-.80 -+ +-.87 .93 -+ + 1.00 Figure 3. Cluster analysis using unweighted pair group method. 41 3.7 Continuous traits Continuous trait Q st will be described as Q s t while Q st based on discontinuous traits will be symbolized as Q s t d . The alternative method of calculating Q s t d through the c use of allele frequencies will be symbolized as Q s t . Q S l values ranged from 0.61 for floret length to 0.91 for stem height (Table Hie). Q s t values for each continuous trait, with the exception of floret length, exceeded the F s t of 0.65 based on isozyme markers. Table Hie . Morphological differentiation among populations for continuous (Q s t ) and discontinuous (Q st d) traits. Standard deviations in parentheses. Continuous traits Qst c Discontinuous traits Q s t d Q s , d * Stem diameter 0.86 Awn serrations 0.39 0.31 Stem height 0.91 Waxy residue on spike 0.69 0.85 Leaf length: width 0.85 Waxy residue on stem 0.47 0.60 Awn length 0.73 Stem Pubescence 0.49 0.56 Spike length 0.86 Pigmentation on ligule 0.41 0.34 Floret length 0.61 Pigmentation on spike 0.37 0.24 Pigmentation on leaf 0.35 0.14 Pigmentation on stem 0.37 0.22 Average 0.80 (0.11) 0.44 (.11) 0.41 (.29) 43 3.8 Discontinuous Traits (Qst and Q st ) Mean Q st^ and Qst^* values were similar (0.44 and 0.41 respectively) and both c d much smaller than mean Q st (0.80). Q st estimates were generally lower than the neutral expectation (F s t). The only discrete trait that exceeded F s t was the one describing the presence or absence of wax on the spike. Q s t d ranged between 0.35 for leaf pigmentation to 0.69 for the presence or absence of wax on the spike and the mean was 0.44 (Table Hie). d* Q s t estimates were characterized by a large range in distribution. Waxy stem d* residue had a Q st of 0.85 while pigmentation analysis yielded estimates ranging d* between 0.14 and 0.24 (Table Hie). The only discontinuous trait with a Q s t higher than F s t was the presence or absence of a waxy spike. These results were very similar to those estimated by A N O V A (Q s t d ) . 3.9 Correlation Analyis A correlation matrix was ued to test whether morphological traits were inter-correlated or correlated with environmental variables like elevation, longitude, latitude, temperature, frost free days, snow pack, and precipitation. The only significant 44 correlation found between morphology and environment was a positive relationship between stem height and the average number of frost free days in a season (0.38). However, many significant correlations were found between morphological traits. Traits dealing with size variables were found to be a common factor when determining positive correlations. Stem height and stem diameter were positively correlated with one another (0.57) while at the same time, they were both negatively correlated with leaf length : width (-0.45 and -0.38 respectively). The correlation matrix also uncovered a positive correlation between spike length and stem diameter and spike length with stem height (0.52 and 0.62 respectively). Awn length was correlated (0.40) with stem height, spike length (0.46), and floret length (0.45). 45 Chapter IV: Discussion 4.1 Species Genetic Diversity Elymus glaucus possesses a high degree of genetic diversity when compared to other plants and grasses in the literature (Table IVa). This diversity was observed at the species level with 78% of the loci scored being polymorphic. The average percentage of polymorphic loci in other vascular plants and grasses were 51% and 60% of loci (Ps) respectively (Table IVa). The high amount of genetic diversity found within this native grass may be attributed to its widespread distribution. Other studies have shown that plants distributed over wide and more variable geographic ranges tend to contain more diversity across isozyme loci than species occupying more endemic and narrow ranges (Ge et al. 1999; Godt and Hamrick 1998; Hamrick and Godt 1989). 46 Table IVa. Several studies contributed to the data set for grasses (166) and other plants (666). Species and population level measures of genetic diversity. Percentage of polymorphic loci (Ps & Pp), Average number of alleles per locus (As & Ap), Total genetic diversity (Ht), intrapopulational Diversity (H s), Proportion of interpopulational Diversity (G s t ) . (Godt and Hamrick 1998). Group no. of no. of Pop. Loci Ps As Pp Ap H t H s Gst Other Grasses1 21 16 60.0 2.4 40.8 1.66 0.340 0.243 0.27 Other Plants2 12 17 50.8 1.9 34.1 1.52 0.295 0.225 0.22 E. glaucus BC 26 21 78.0 2.0 22.2 1.31 0.065 0.023 0.65* E. glaucus U S 4 20 20 80.0 3.3 31.4 1.4 0.085 0.038 0.55* 1&2 Hamrick and Godt (1998) 3 This study 4 Knapp and Rice (1996) * indicates F s t 47 The high genetic variability distributed in E. glaucus, as a species, could be a product of its widespread distribution in conjunction with its ability to adapt to a number of different climatic conditions. Elymus glaucus is an allotetraploid (Dewey 1968; Jenson 1993; Knapp and Rice 1996) and being an allotetraploid enhances species diversity because by definition, it is a species containing two different sets of genomes. This will undoubtedly increase overall species diversity and the biochemical diversity it creates may have some bearing on its ability to possess such a wide ecological amplitude. Fixed heterozygosity, common to all allopolyploids, offers much in the way of biochemical diversity (Allard et al. 1966). A l l individuals within the populations studied possessed fixed heterozygosity and this may have added to this species ability to flourish in a broad variety of niches. When comparing genetic diversity between northern (B.C.) and southern (U.S.) populations, U.S. populations of E. glaucus had a slightly larger percentage of polymorphic loci and mean number of alleles per locus at both the species and population levels (Table IVa). A possible explanation for this could be due to the northward recession of glacial ice sheets. Southern locations could have been a refugia for E. glaucus and as glacial ice sheets moved northwards, rare alleles present in southern populations would be lost in the more northerly advancing populations due to drift. Cwynar and MacDonald (1987) found this to be evident in their analysis of diversity for Lodgepole pine. They attributed the progressive northward decline in diversity as a consequence of small founding population sizes relating to the absence of rare alleles. 48 These small founding populations would then give rise to founding populations with a lower probability of possessing the rare allele. If we assume that there was a northward migration of E. glaucus due to the recession of the Cordilleran ice sheet 16 000 years ago (Pilou 1991) and that colonization of habitats was achieved by a limited set of individuals, these founding populations would have limited amounts of genetic diversity. The perpetual establishment of new founding populations from populations originally established by few individuals would then provide a reasonable explanation for the lower levels of diversity found when compared to U.S. populations for Pp and Ap for the populations studied in British Columbia. On the other hand, differences between Northern and Southern populations could have also simply been an artifact of assaying different isozymes, sampling techniques, and variations in locus resolution. Knapp and Rice's (1996) survey of genetic structure, gene flow, and mating system of Elymus glaucus was done using most of the same isozymes except they used Aconitase, Aspartate aminotransferase, and Phosphoglucomutase instead of Diaphorase and Isocitric dehydrogenase. 4.2 Genetic Diversity within Populations In contrast to the species level variation, E. glaucus possessed less genetic variation than other plant and grass species reported in the literature within populations (Table IVa). Godt and Hamrick's (1998) review found that the percentage of polymorphic loci at the population level (Pp) of other vascular plants and other grasses 49 was 41% and 34% respectively. The average number of alleles per locus found within populations (Ap) were 1.52 for vascular plants in general and 1.66 alleles for other grasses. E. glaucus populations in B.C. and the U.S. had lower values of Pp (22%) and Ap (1.31) than other vascular plants and grasses (Table IVa). These findings are expected because Elymus glaucus has generally been regarded as a self pollinating grass species. Population genetic theory of finite populations predicts a loss of variation as a result of inbreeding and random genetic drift (Falconer and Mackay 1996). The low levels of diversity found within populations could have also been the result of having populations founded by few individuals. When comparing these results to Elymus canadensis (Sanders et al. 1979) E. glaucus had much higher species diversity with respect to the percent of polymorphic loci, yet both shared the same low allelic diversity within populations. This may have been due to the contrasting distribution and range encompassed by both species. E. canadensis does not have as large of a range as E. glaucus, however, it is has been described like E. glaucus, as a self pollinating grass. Sanders (1979) proposed that a mating system composed of selfing, low levels of gene flow, and different flowering times between populations were the main contributors for such low levels of within population diversity. This explanation provided by Sanders (1979) is plausible for the results from this study except that I failed to assess flowering times for the different accessions of Elymus glaucus in the common garden. This information could have been useful in determining the cause for the low estimation of gene flow between populations within this native grass species. One could look at whether flowering times were 50 asynchronous between populations thereby causing pollen to be exclusive to the populations in which it came from. Inbreeding is definitely a plausible explanation for the loss of variation within populations. Supporting evidence for a predominately inbreeding mating system can be found in the 10 fold difference in expected heterozygosity between E. glaucus and other plants and grasses (Table IVa). Simple Mendelian genetics show that heterozygosity decreases with every generation of selfing (Griffiths et al. 1996). A study of inbreeding plant species by Allard et al. (1966) found that certain species contain substantial stores of genetic variability in combination with a well-integrated population structure. This is especially apparent in E. glaucus because of its relatively high genetic diversity as a species and population structuring indicated by the high F st value. I have discussed the possibility that inbreeding may have led to such low within population diversity however, I still have not explored the possibilities of random genetic drift, and founder's and bottleneck effects which can also reduce genetic diversity within populations. These will be addressed later with discussions focussed more towards the genetic structure, gene flow, and mating system of Elymus glaucus. 4.3 Genetic Structure The above results corroborate the F-statistic findings used to estimate population genetic structure (Wright 1978). F s t values from 21 isozyme loci (0.65) were high 51 suggesting that genetic diversity is distributed primarily among rather than within populations. High genetic structuring (Fst = 0.549) was also evident in U.S. populations distributed throughout Washington, Oregon, and California (Knapp and Rice 1996). The F-statistic values based on B.C. populations are also significantly higher than the average G st (F s t is equivalent to G s t ) of other grasses and vascular plants reviewed by Godt and Hamrick (1998); (Table IVa). The distribution of high genetic variability between populations as opposed to within populations predicts a mating system that has limited pollen and seed dispersal capabilities which usually results in low levels of gene flow (Sanders et al. 1979). High among population differentiation is characteristic of self-pollinating species with limited levels of gene flow. A high level of migration between populations usually results in a homogenizing effect with respect to genetic structure while species with restricted migration (low gene flow) promotes population divergence due to random genetic drift. Inbreeding within populations can also add to the genetic divergence of populations. This too ultimately decreases within population variation and increases the variation among populations (Golenberg and Nevo 1987; Knapp and Rice 1996; Sanders, Hamrick and Holden 1979). A literature review by Hamrick and Godt (1989) found that, on average, plants which produce seed predominately through self-pollination generally have 51% of their isozyme diversity partitioned between populations while the remaining 49% is found within populations. In their survey of plant population genetics, they also noted that 52 predominately outcrossing plant species showed lower average between population variability (14.8%) and higher within population variability (85.2%). Results observed for both U.S. and B.C. populations suggest that the genetic structuring in E. glaucus populations are the manifestations of a predominately self pollinating species with low levels of gene flow. 4.4 Hierarchical analysis of F s t Much of the population genetics related to assessing the genetic diversity for Elymus glaucus has been discussed. Sufficient evidence for population differentiation has been presented, however, there is a key question that has yet to be answered. This question relates to whether or not the observed genetic differences between populations are structured by environmental and\/or geographic factors. Restoration projects must be approached with caution when a general inventory of genetic structure uncovers high levels of population differentiation. There is a definite potential for significant environmental influences leading to selection for a highly organized population structure when such high between population differentiation exists. If this is the case, restoration projects approached blindly may have a high potential for failure. Genetic variability across the species range in concert with a high degree of population differentiation may suggest that E. glaucus has undergone some form of differential selection pressure resulting in local adaptation to British Columbia's vast environmental heterogeneity. Selection within several meters leading to genotypic 53 differences have been well documented in the past (Argyres and Schmitt 1990; Golenberg and Nevo 1987; Hsiao 1999; Nevo et al. 1997; Nevo et al. 1988; Nevo et al. 1994; Ransom et al. 1998; Ruiz et al. 1997; Volis and Mendlinger 1998). A study done by Golenberg and Nevo (1987) found multi-locus differentiation between 8 different loci for wild emmer wheat. Their transect of 4 populations seemed to be genetically differentiated with respect to climatic gradients. It is therefore important for restoration projects to assess whether the genetic structuring quantified by the high F s t is due to causes such as adaptation, selection, or drift. In total, 12 different forest districts were sampled and across these forest districts, seven different biogeoclimatic ecosystem zones were represented (not to mention numerous variants contained within some BEC zones). Samples were collected from a variety of habitats each having unique climates, average yearly temperatures, aridity, elevation, soil types, and surrounding vegetation. In order to test whether environmental gradients helped to structure populations, reciprocal transplant studies would be ideal (Claassen and Marler 1998; Greipsson et al. 1997; Helenurm 1998; Hester et al. 1998; Kinderscher and Tieszen 1998; Miao et al. 1997; Williams and Davis 1996). However, these studies are usually too time consuming and labour intensive. Another method that can help assess whether environment or ecology plays a role in population differentiation is to analyze the distribution of variation on a hierarchical level by nesting populations into groups that share similar environments or ecology (Holsinger and Mason-Gamer 1996). If environment does play a role in population differentiation, individuals from similar environments or geography would be more similar genetically than ones collected 54 from different environments (Knapp and Rice 1996; Knapp and Rice 1998; Nevo et al. 1997; Volis and Mendlinger 1998). This hierarchical analysis was done by nesting populations into groups designated by their biogeoclimatic ecosystem classification, elevation, mean yearly snow pack, mean number of frost free days, soil type, mean yearly precipitation, mean yearly temperature, and forest district (appendix Tables A l to A6). Based on the F s t results obtained from the hierarchical analysis, there seems to be little discernable evidence for a correlation between ecology (BEC, elevation, snow pack, frost free days, soil, elevation, precipitation, and temperature) and genetic relatedness. Populations sampled from ecologically similar environments were not necessarily genetically \"closer\" to ones that were collected from contrasting environments. Genetic variation was actually distributed primarily within these groups rather than between them (Table IIIc). Evidence supporting these conclusions is depicted by the high degree of population differentiation still apparent within ecological and geographical groupings. The average F s t between environmental factors was extremely low (mean 0.051, +\/- 0.04SD) indicating that there is little genetic variation partitioned between environmental groups while most of the variation can be found distributed within factors. One would expect the majority of genetic variability to be distributed between environmental groupings if climatic and geographic variables had in fact played a role in causing the high genetic structuring seen by the F s t (0.65). These results support the view that ecology and geography have little bearing on the distribution of selectively neutral genetic variability in Elymus glaucus and 55 that genetic drift may have played a major role in the random distribution and fixation of alleles. It is important to look at why genetic drift plays such an influential role in determining the genetic structure of this species. The factors discussed so far have all pointed towards the effects of random genetic drift and self pollination as being the driving force behind the divergence of populations. Founder effects could also explain the low within population diversity, deficiency of heterozygosity, and its patchy ecological distribution. Analysis of its gene flow and mating system may provide further evidence supporting these arguments. 4.5 Gene Flow High F st values can be the direct result of little gene flow between populations. Gene flow, as depicted by Slatkin and Barton's (1989) indirect method of estimated average gene flow was low (Nm = 0.406). This result is consistent with data presented thus far because the homogenizing effects of gene flow are not apparent. This estimate of gene flow relies on the variance in gene frequencies among different populations. It is based on a number of assumptions, one of which is that gene frequency variance depends on the number of migrants entering each population each generation. Whitlock and McCauley (1999) have indicated that this indirect method for the estimation of gene flow, due to its dependence on many assumptions, must be approached with caution. Although the Nm calculated from F s t relies on a number of assumptions, it still serves as a good 56 indication that gene flow between populations was very low. The high genetic variance distributed between populations for E. glaucus is most likely due to a combination of selfing and low gene flow between populations. If high levels of gene flow were present, both population divergence and the deficiency in heterozygosity measured within populations would not be present. Possible factors that may have reduced gene flow are environmental barriers which may have impeded seed and pollen dispersal (Maki and Morita 1998). Mountains, rivers, and other unsuitable habitats can potentially provide significant barriers that limit the dispersal of individuals from one population to the next. Along with these environmental obstacles, the patchy distribution of E. glaucus in scattered disturbed habitats provides an ideal opportunity for drift to act on neutral genetic markers because migration to and from other populations could be very difficult. 4.6 Mating System The main factors responsible for affecting gene flow are an organism's mating system and mode of seed dispersal. A plant's mating system can ultimately affect long term population processes such as the way genetic variation is partitioned and how novel genetic variation is spread through populations (Ritland 1983). Limited abilities for seed and pollen dispersal will likely cause low levels of gene flow and high deficiencies in heterozygosity. Inadequate seed dispersal mechanisms will increase the chance of inbreeding because mating will be limited to individuals residing in close proximity. The 57 opposite is true for outcrossing species where the proportion of heterozygosity is high relative to inbreeding species and the distribution of genetic variation is more homogeneous. Many mating systems have also been described by the mixed mating model where a proportion of zygotes are derived through self fertilization while the remaining are the product of random mating between other plants in the population (Ritland 1983). Knowledge of mating systems is important for restoration purposes because it enables one to formulate better decisions on how genetic sampling should be conducted in order to preserve natural biodiversity (Jain 1975). The literature (Knapp and Rice 1996; Pojar 1974a; Pojar 1974b) depicts E. glaucus as a predominantly self-pollinating species and thus far, results from the isozyme analysis have indirectly supported this claim. Elymus glaucus' inbreeding nature can be inferred from Wright's (1978) inbreeding coefficients derived from the isozyme analysis. There is an obvious deficiency in heterozygotes relative to Hardy-Weinburg expectations at both the population (F, s = 0.68) and species (F, t = 0.89) level. The F\\s value of 0.68 indicates that there is a 68% deficiency of heterozygosity at the population level relative to that expected under random mating. When combining all the populations into one large population, an 89% deficiency of heterozygotes exists. Mendelian genetics has shown that inbreeding (or partial inbreeding) decreases levels of heterozygosity. With limited seed dispersal, individuals located within close proximity will more than likely be full or half sibs and thus consanguineous matings will result in a deficiency of heterozygosity which ultimately decreases the amount of diversity found within 58 populations. Mating between close relatives can also occur i f populations are founded by few individuals which are closely related. Published reports and the data I have presented thus far have indirectly shown that E. glaucus is a self pollinating species. However, it seems rather odd that such a tall grass which relies on wind and animal dispersal of seed can maintain a predominately selfing and inbreeding mating system. Outcrossing rates were calculated and the results, based on single and multilocus outcrossing rates (Ritland 1996), estimated that the 3 populations examined outcrossed approximately 50% of the time. These results seem to contradict Wright's (1978) inbreeding coefficients relating to the observed deficiency of heterozygosity within E. glaucus populations. Results gained by single and multi locus t-estimates do not fully concur with the high degree of population differentiation (0.65) and the heterozygote deficiency indicated by Fj t and F; s . Certainly one would expect that a species which outcrosses approximately 50%> of the time must show some degree of gene flow resulting in a lower degree of heterozygote deficiency and ultimately leading to a more homogenous population structure. A n explanation for this apparent contradiction is a combination of founder effects, patchy ecology, and low levels of gene flow. Colonization of habitats would be achieved by a limited number of individuals. Because of the outcrossing rates measured, self pollination may not be the best way of explaining the low levels of heterozygosity found 59 within populations but instead, bi-parental inbreeding seems to be more likely in light of the multi and single locus outcrossing rates obtained from 3 Vancouver Island populations. Populations were found to be small and patchy (Knapp and Rice 1996; personal observation) and the patchiness of its population structure could be an obstacle for gene flow between populations. Bi-parental inbreeding would then result from founder effects because pollen involved in outcrossing would have a greater likelihood of fertilizing closely related individuals (relatives) than unrelated ones in these fairly isolated populations. This would explain the high deficiency of heterozygotes within populations. Further evidence characteristic of bi-parental inbreeding can be found in the differences measured between multi and single locus t-estimates. Multi locus outcrossing rates are consistently larger than single locus estimates. This is characteristic of bi-parental inbreeding because multilocus estimates tend to include inbreeding due just to selfing, and exclude biparental inbreeding. This is because true selling is manifested at all loci in an organism, while biparental inbreeding causes effective selfing at some gene loci and effective outcrossing at other loci; the multilocus estimator of selfing would classify such a progeny as completely outcrossed. 4 . 7 Random Genetic Drift Random genetic drift is one scenario that can explain the distribution of genetic variation that has been measured. Low measures of genetic diversity within populations compared to species diversity is likely due to the actions of drift leading to the random fixation of alleles. E.glaucus populations have predominately been found in small, patchy, and mixed distributions occupying outskirts of forest openings, roadsides, and meadows (Phil Burton, personal observation). Distances between suitable habitats may be quite large relative to its seed dispersal capabilities and thus, the potential for allele fixation increases with small, isolated populations (Falconer and Mackay 1996). Though outcrossing was measured to be moderate, due to founder effects and the fixation of alleles in relatively small isolated populations, it is most likely relegated to bi-parental inbreeding. Gene flow acting to homogenize populations does not seem to be great enough to overcome the action of random genetic drift because the distribution of genetic variability has shown that populations are not well correlated with geographic distance. 4.8 Metapopulation Dynamics Expanding on the hypothesis that random genetic drift was the cause for such high population differentiation is the drift related model of Metapopulation Dynamics. Some species, particularly those found in subdivided and fragmented habitats persist as metapopulations (Brookes et al. 1997; Dybdahl 1994; Neve et al. 1996; Ouborg 1993; Sinsch 1992; Thomas and Harrison 1992; Whitlock 1992). Metapopulation dynamics describe a collection of local populations which periodically undergo extinction and recolonization events. In terms of the distribution of genetic variance among local populations, extinction and recolonization can act to increase population differentiation through founder effects (Wright 1940) or it can work to reduce differentiation between 61 populations through homogenizing gene flow (Slatkin 1987). The dynamics of how a species is able to recolonize after extinction in a specific habitat ultimately determines its population genetics and evolutionary outcomes. The distributions of E. glaucus populations seem to be characteristic of species possessing metapopulation dynamics. The species' high F st, low geographic structuring, and patchy ecology support this observation. According to the metapopulation framework, differentiation between populations will increase if: 1) Colonists predominantly originate from a limited number of populations, even i f the number of colonists is large (Manicacci et al. 1992; Whitlock and McCauley 1990). 2) Dispersal is distance or direction biased or when the number of colonists is very small; and 3) When colonists originate from all the local populations in the metapopulation, i f the number of colonists does not exceed twice the level of gene flow among established populations (Whitlock and McCauley 1990). The most plausible scenario for the high differentiation found among E. glaucus populations is that colonization was achieved by migration of a few individuals (founder's effects) from a small number of populations because gene flow between 62 populations was estimated to be quite low. The weak correlation between genetic and geographic distance from the Mantel test may also imply that populations were founded almost randomly, very much like the classical metapopulation approach first described by Levins (1970). This also coincides with the metapopulation theory because populations with frequent colonization and extinction events are usually considered not in genetic equilibrium and therefore show little isolation by distance. The patchy ecological distribution is somewhat evident when one looks at the 0.21 difference between F j s and Fj t . The 20% discrepancy between F j s and Fj t is indicative of a Wahlund effect (Haiti and Clark 1997). The principle behind the Wahlund effect is that when one takes the mean of allele frequencies from several different populations, it will predict an increase in expected heterozygosity i f alleleic frequencies are patchy. These frequencies are used when calculating inbreeding coefficients and will therefore lower the inbreeding coefficient value for F\\s with respect to Fj t . The Wahlund effect provides further evidence in favour of the metapopulation interpretation because the patchy ecological distribution found for E. glaucus is complemented by the patchiness of its genetic distribution. Individual populations represent genetic subsets of the entire species diversity because the genetic make-up of a certain population contains merely a fraction (or pocket) of the entire species variation. This is consistent with the metapopulation model where populations have patchy asynchronous distributions (Hanski and Gilpin 1991; Hansson 1991; Husband and Barrett 1996; Manicacci et al. 1992; Thomas and Harrison 1992). 63 Elymus glaucus is commonly found in temporary habitats such as roadsides, forest openings, and impoverished landscapes which can lead to cycles of extinction and re-colonization; another important element of metapopulation theory. This grass is usually found in environments with a moderate to high supply of sunlight (personal observation). As forests and surrounding shrubs expand, there is greater competition for sunlight, water and nutrients thereby increasing the potential for local population extinction due to succession. Extinction of populations, however, may be followed by recolonization elsewhere because they have the ability to disperse to a more suitable location or available habitat patch via wind or animals. In a way, this ties into the genetic drift scenario because population turnover can act to accelerate genetic drift. Alleles are lost when populations disappear so that the genetic effective size of a metapopulation with turnover may only represent a small fraction of its census size (Gilpin 1991). Low within population diversity and a high degree of heterozygote deficiency demonstrated by both F j s and Fj t further support the scenario of repeated colonization and founder effects because alleles are lost when population turnover occurs. Repeated colonization and extinction processes may have also acted to increase population differentiation because populations were most likely founded by a small number of individuals. If these cycles occur frequently, there is little time for selection to occur and therefore we expect to see populations in genetic disequilibrium. Low geographic and environmental structuring, as shown by the Mantel test, hierarchical F st, and dendrogram based on genetic distance, is indicative of populations which are not in genetic equilibrium. 64 4.9 Continuous vs Discontinuous Traits There seems to be a large discrepancy in the partition of differentiation in morphological characters among populations between continuous (Q st ) and d c c discontinuous traits (Q st ). Q st based on 6 continuous traits (Qst = 0.80) were generally greater than the 8 discontinuous characters (Qs^ = 0.44); (Table Hie). An observable difference when partitioning variability between and within populations for continuous and discontinuous traits was that variability for continuous traits was partitioned predominantly between populations while within population variability was larger for discontinuous traits. Qst^ estimates for the presence or absence of pubescence and wax were higher than those calculated for the presence or absence of pigmentation. Pubescence and wax may be traits selected for in dry and warmer conditions. The high among population differentiation measured for these traits may be caused by the selective forces of dryer and warmer climates (Taiz and Zeiger 1991), however, this will be discussed in greater detail in the following sections. 4.10 Selective Neutrality of Isozymes The assumption that isozyme markers are selectively neutral is important for comparative studies of Q s t with F s t (Podolosky and Holtsford 1995; Prout and Barker 65 1993; Yang et al 1996). Slatkin (1987) indicated that it is rare to see that selection affects all loci similarly so that i f many loci are used, then it is safe to assume that the mean F s t for all loci is neutral eventhough a small number may be under weak selection. It is also a relatively safe assumption made by many studies that isozymes do have a fairly neutral evolutionary pattern. Twenty-one loci were used to calculate F st for E. glaucus and this was relatively large compared to other studies listed above. Assuming that selection affected each of these loci differently, it would be supporting evidence for neutrality. Spitze (1993) found that isozymes were not related to any fitness components in his study of Daphnia obtusa and therefore had ample evidence supporting the neutrality of the isozymes he studied. Correlation analysis between isozymes and fitness components for E. glaucus were not done because fitness components were not measured. However, evidence supporting the neutrality of the isozymes studied in E. glaucus includes the fact that there was little genetic structuring according to environmental and geographic groups. Another assumption that must be made when c d comparing Q st and Q st with F s t is that variance is limited to genetic effects only. Even though plants had been grown in a common garden and measurements were randomized, we cannot ignore dominance and maternally inherited environmental effects without proper progeny and generation studies. Dominance and maternal effects could have played a role in some of the quantitative traits measured and this would have given higher c d o estimates of Q s t and Q st because variability within populations (Q w ) would be 66 c d overestimated. Taking these factors into account would increase Q st and Q st estimates and therefore estimates presented by these results should be deemed minimum estimates. 4.11 Q s t C v s F s t Comparisons between Q st and F s t showed that continuous traits had a greater degree of between population variation than that of isozymes. Similar findings have been presented for many species, e.g., soya bean germplasm (Perry et al. 1991), Drosophila buzzatii (Prout and Barker 1993), Clarkia dudleyana (Podolosky and Holtsford 1995), and Pinus contorta (Yang et al. 1996). These authors attribute the differences in the distribution of isozyme and quantitative trait variation to differential selective forces. If we can assume that isozyme markers are neutral, diversifying selective forces must be present if Q s t estimates based on morphology are higher than F s t based on isozymes. Populations are more distinct from one another morphologically because continuous morphological traits seem to be responding more quickly to differential environmental selective forces and thus causing morphological divergence of populations. Q st for stem height, stem diameter, leaf length : width, and spike length were all much higher than the neutral expectation (F s t). Quantitative traits such as awn length and floret length had Q s t values similar to F s t . This would support that these traits were neutral with respect to selection. 67 Waldmann and Andersson (1998) compared both isozyme F st and quantitative trait Q st of two perennial plants. They compared Scabiosa columbaria, a locally common grassland plant with a widespread distribution, to Scabiosa canescens, a rare species with a narrow range of habitats. Comparisons between both widespread and rare species using isozyme based F s t and Q s t derived from continuous traits, showed that 5*. columbaria had a significantly larger mean Q s t than F s t while both Q s t and F s t were relatively equal among S. canescens. The conclusion made by Waldmann and Andersson (1998) was that diversifying selection in S. columbaria enabled it to adapt more readily to environmental variables than S. canescens. This would therefore explain the reason why S. columbaria had a much broader ecological range than S. canescens. These findings may also be applicable to E. glaucus because, like S. columbaria, it too possesses a wide distributional range. Diversifying selection may have enabled this native grass to adapt more readily to different environmental and geographic situations. Q s t based on body size of Drosophila buzzatii was also significantly larger than F st (Prout and Barker 1993). They also implied that selective differentiation was the cause for this difference. Yang et al. (1996) concluded that the selective differentiation of traits relating to size and stature could reflect the inherent adaptation of populations of P. contorta for rapid growth in order to escape suppression by neighbouring plants during establishment. Yang et al. (1996) found that Q st for wood specific gravity, stem diameter, stem height, and branch length were all significantly greater than F s t . For E. 68 glaucus, it seems as though traits reflecting size and stature have undergone some form of diversifying selection as well. This selection has shaped large-scale patterns of variation between E. glaucus populations. These examples which describe diversifying selection of quantitative traits as a precursor to adaptation provides good support for using E. glaucus for the reclamation of disturbed habitat because advances in morphological adaptation seem to be moving faster than the neutral expectation. 4.12 Q s t d v s F s t Average Q s t d estimates indicate that the distribution of variance for discrete traits was found more within populations than those for both continuous traits and isozyme markers. Assuming again that the distribution of isozyme variation is neutral, results supporting stabilizing selection for discrete traits seem to be apparent. This type of selection could be one towards an optimal mean. Traits relating to pigmentation are highly correlated with each other because the presence or absence of pigment (violet) is caused by the presence or absence of anthocyanins (Taiz and Zeiger 1991). Estimated Qs^ values for the presence or absence of wax and pubescence were similar to F st and higher than most of the other discrete traits analyzed. There may have been some diversyfying selection involved and this can be related to fitness and adaptation as seen in papers written by Yang et al. (1995), Waldman (1998), and Knapp (1998). Pubescence and wax is typical of many desert plants because hair enables plants 69 to create a thicker boundary layer and therefore restrict evapotranspiration (Personal Communication with Dr. Fred Ganders) while wax can function to reflect sunlight and reduce water loss. However, the correlation of these features to regions characteristic of more arid climates did not prove to be significant. A reciprocal transplant study would be appropriate i f one wanted to determine the adaptive significance of the different morphological traits measured. If we were to raise populations in contrasting environments, we would expect to see more pronounced differences among populations for some continuous and discrete traits. 70 Chapter V: Conclusion 5.1 Genetic Story Examination of Elymus glaucus populations using isozymes has found that the high amount of genetic diversity within this species is predominantly distributed between rather than within populations. Results have shown that plants collected within populations are similar to one another while populations remain quite distinct. The divergence of populations does not seem to be structured ecologically or geographically because populations found in similar environments are no more similar genetically than ones originating from contrasting environments. This leads us to the conclusion that random genetic drift is somewhat responsible for the observed genetic pattern. The patchy distribution and genetic disequilibrium of populations, in combination with low levels of gene flow and low frequency of heterozygotes supports substantial founder effects and thus leads me to favor the random genetic drift model of metapopulation dynamics. However, without adequate data relating to extinction and re-colonization of populations, the metapopulation approach for this species can only be inferred indirectly. 5.2 Restoration Implications Because the genetic variation seems to be randomly and widely dispersed, collections of germplasm for genetic conservation should be made from a variety of different habitats in order to sample the entire species' genetic diversity. However, it is 71 important not to ignore the large positive inbreeding coefficients because they indicate a large degree of heterozygote deficiency within natural populations. This suggests that before genetic diversity is maximized within the seed orchard, studies investigating the potential for outbreeding depression should first be conducted (Havens 1998). The danger for outbreeding depression can occur because adaptive gene complexes can be broken up when progeny are the result of outcrossed parentage. Genetic mixture of populations already adapted to different local conditions can result in the reduction of fitness caused by the breakdown of the gene complexes (Shields 1982; Templeton 1986). I did not detect any genetic adaptation or selection in this study but non-detection may have been due to using selectively neutral genetic markers such as isozymes. Random genetic drift seems to have played a big part in the distribution of E. glaucus variation If outbreeding depression is not a factor in E. glaucus, seed orchards designed to introduce variation within the seed lots may lead to offspring with an enhanced potential for adaptation. This would facilitate the introduction of E. glaucus into a variety of environmental niches and would therefore increase the probability for the successful reclamation of disturbed habitats. Preliminary transplants of outcrossed seeds produced by the diversifying effects of the seed orchard should be carefully monitored for outcrossing depression for several generations. If these initial projects are deemed successful, then there will be great future potential for the successful introduction of E. glaucus into disturbed sites. Presently, ex situ collections of Elymus glaucus are being made and germplasm is being developed in order to feed the demand for native grass restoration of disturbed habitats. Guidelines for seed distribution should take into account the genetic and morphological findings of this research. Collection and propagation of seed for habitat reclamation strategies using E. glaucus can be done in two different ways. One method is to use the diversity approach as mentioned above. This method utilizes the species' genetic variability throughout the sampling range in order to develop a seed source where the potential for genetic variation is maximized. Another approach for land reclamation is to focus on the maintenance of the local genetic integrity and evolutionary processes that seem to be occurring within this species. Reclamation of disturbed habitats while maintaining metapopulation dynamics requires the use of individuals from a limited set of populations. This will act to mimic bi-parental inbreeding found within natural populations because founding populations will be relatively homogenous with respect to genetic variation and diversity. Outcrossing between relatives will then occur with a higher probability (just like in natural populations). A compromise between both the diversity and metapopulation approaches to restoration can be accomplished by developing a seed source from several populations. The diversity infused into the seed source for recolonization would not be designed to encompass the entire diversity found within the species, but instead, designed to possess the diversity found within 3-4 different populations sampled from 1-2 forest districts. 73 This would still support a metapopulation approach to restoration because colonization would come from a limited set of populations. Moreover, the diversity approach would also be considered because a moderate degree of diversity within the seed source would be used in order to increase the chance for adaptation and decrease the probability for outcrossing depression. 5.2 Recommendations for Further Research Future research in this field should be done using reciprocal transplant studies. Because there were no significant geographic\/genetic patterns found within this species, plots representing all individuals from all collections should be established in each of the different biogeoclimatic zones within British Columbia. Because British Columbia spans a wide number of ecological zones, plots representing the extremes in temperature, moisture and elevation should be made. This will give us further insight into which populations are best suited for the different ecological conditions needed for land reclamation. These plots should not be limited to Elymus glaucus alone instead, it would be wise and more efficient to gather information on other native species as well so that a native seed mix can be designed for habitat restoration. Genetic and phenotypic plasticity may also be involved in the widespread distribution of this endemic grass because genetic patterns structured by environmental and geographic locations could not be found. As mentioned previously in this study, E. glaucus is an allotetraploid with several loci displaying fixed heterozygosity. The 74 biochemical diversity may translate into a sufficient degree of plasticity therefore facilitating the transfer of germplasm into a variety of different habitats. Further analysis using reciprocal transplants would be important in directly observing and quantifying adaptation, selection and plasticity. In addition to these reciprocal transplant studies, further analysis of E. glaucus' genetic structure and mating system on a finer scale (i.e. using microsatellites) would be useful in obtaining a more precise characterization of its within population variability. This would provide information related to seed sampling protocols within populations for future projects. Bibliography Allard, R. W., S. K. Jain and P. L. Workman, 1966. The Population Genetics of Inbreeding Species. Advances in Genetics 14. Argyres, A . Z., and J. Schmitt, 1990. Microgeographic Genetic Strucutre of Morphological and Life History Traits in a Natural Population of Impatiens Capensis. Evolution 45: 178-189. Ayres, D., D. Garcia-Rossi, H . Davis and D. Strong, 1999. Extent and Degree of Hybridization between Exotic (Spartina alterniflora) and Native (S. foliosa) cordgrass (Poaceae) in California, USA determined by Random Amplified Poymorphic D N A (RAPDs). Molecular Ecology 8: 1179-1186. Banner, A. , W. MacKenzie, S. Haeussler, S. Thomson, J. Pojar et al, 1993. A Field Guide to Site Identification and Interpretation for the Prince Rupert Forest Region. Ministry of Forests, Victoria B.C. Bell, G. D., and A . M . Fletcher, 1978. Computer Organized Orchard Layouts. Silvae Genetica 27: 223-225. Biedenbender, S. H. , and B. A . Roundy, 1996. Establishment of Native Semidesert Grasses into Existing Stands of Eragrostis lehmanniana in Southeastern Arizona. Restoration Ecology 4: 155-162. Bratton, S., and A . Meier, 1998. Restoration of Wildflowers and Salamanders in Southeastern Deciduous Forests. Restoration and Management Notes 16: 158-165. Brookes, M . , Y . Graneau, P. King, O. Rose, C. Thomas et al, 1997. Genetic Analysis of Founder Bottlenecks in the Rare British Butterfly Plebejus argus. Conservation Biology 11: 648-661. Bugg, R. L. , C. S. Brown and J. H . Anderson, 1997. Restoring Native Perennial Grasses to Rural Roadsides in the Sacramento Valley of California: Establishment and Evaluation. Restoration Ecology 5: 214-228. Claassen, V . P., and M . Marler, 1998. Annual and Perennial Grass Growth on Nitrogen-Depleted Decomposed Granite. Restoration Ecology 6: 175-180. Coffman, F. A. , 1961. Oat and Oat Improvement, pp. in American Society of Agronomy, edited by M . Series, Madison. DeLong, C , 1988. A Field Guide for Identification and Interpretation of Serai Aspen Ecosystems of the BWBScl Prince George Forest Region. Ministry of Forests, Victoria. Delong, C , A . MacKinnon and L. Jang, 1990. A Field Guide for Identification and Interpretation of Ecosystems of the Northeast Portion of the Prince George Forest Region. Ministry of Forests. DeLong, C , D. Tanner and M . J. Jull, 1993. A Field Guide for Site Identification and Interpretation for the Southwest Portion of the Prince George Forest Region, Victoria B.C. Delong, C , D. Tanner and M . J. Jull, 1994. A Field Guide for Site Identification and Interpretation for the Northern Rockies Portion of the Prince George Forest Region. Ministry of Forests, Victoria B.C. Dewey, D. R., 1968. Synthetic Hybrids of Agropyron dasystachyum x Elymus glaucus and Sitanion hystix. Botanical Gazette 129: 309-115. Dobson, A . P., A . D. Bradshaw and A . J. M . Baker, 1997. Hopes for the Future: Restoration Ecology and Conservation Biology, Science 277: 515-521. 77 Dybdahl, M . F., 1994. Extinction, Recolonization, and the Genetic Structure of Tidepool Copepod populations. Evolutionary Ecology 8: 113-124. Eliason, S. A. , and E. B. Allen, 1997. Exotic Grass Competition in Suppressing Native Shrubland Re-establishment. Restoration Ecology 5: 245-255. Falconer, D. S., and T. F. Mackay, 1996. Introduction to Quantitative Genetics. Longman, Essex. Felsenstein, J., 1986. Population Differences in Quantitative Characters and gene frequencies: A comment on papers by Lewontin and Rogers. American Naturalist 127:731-732. Felsenstein, J., 1993. Phylip, University of Washington. Fensham, R. J., and I. D. Cowie, 1998. Alien Plant Invasions on the Tiwi Islands. Extent, Implications, and Priorities for Control. Biological Conservation 83: 55-68. Franklin, J., 1993. Preserving Biodiversity: Species, Ecosystems, or Landscapes. Ecological Applications 3: 202-205. Ge, S., K.-Q. Wang, D.-Y. Hong, W.-H. Zhang and Y . - G . Zu, 1999. Comparisons of Genetic Diversity in the Endangered Adenophora lobophylla and its Widespread Congener, A. potaninii. Conservation Biology 13: 509-513. Geller, J. B., 1999. Decline of a Native Mussel Masked by Sibling Species Invasion. Conservation Biology 13: 661-664. Gilpin, M . , 1991. The Genetic Effective Size of a Metapopulation, Biological Journal of the Linnean Society 42: 165-175. 78 Godt, M . J., and J. L. Hamrick, 1998. Allozyme Diversity in the Grasses, pp. 11-29 in Population Biology of Grasses, edited by G. P. Cheplick. Cambridge University Press, New York N Y . Golenberg, E. M . , and E. Nevo, 1987. Multilocus Differentiation and Population Structure in a Selfer, Wild Emmer Wheat, Triticum dicoccoides, Heredity 58: 451-456. Gottlieb, L. D., 1981. Gene Number in Species of Asteraceae that have Different Chromosome numbers. Proceedings of the National Academy of Sciences U.S.A. 78: 3726-3729. Green, R. N . , and K. Klinka, 1994. A Field Guide to Site Identification and Interpretation for the Vancouver Forest Region. Ministry of Forests, Victoria B.C. Greipsson, S., H. Ahokas and S. Vahamiko, 1997. A Rapid Adaptation to Low Salinity of Inland-Colonizing Populations of the Littoral Grass Leymus arenarius. International Journal of Plant Science 158: 73-78. Griffiths, A . , J. Miller, D. Suzuki, R. Lewontin and W. Gelbart, 1996. An Introduction to Genetic Analysis. W.H. Freeman and Company, New York. Hamrick, J. L. , and M . J. Godt, 1989. Allozyme Diversity in Plant Species, pp. 43-63 in Plant Population Genetics, Breeding, and Genetic Resources, edited by A . H . D. Brown, M . T. Clegg, A. L. Kahler and B. S. Weir. Simauer Associates, Sunderland M A . Hamrick, J. L. , M . J. W. Godt, D. A . Murawski and M . D. Loveless, 1991. Correlations between Species Traits and Allozyme Diversity: Implications for Conservation Biology, pp. in Genetics and Conservation of Rare Plants, edited by D. A . Falk, and K . E. Holsinger. Oxford University Press, Oxford GB. Hanski, I., and M . Gilpin, 1991. Metapopulation Dynamics: Brief History and Conceptual Domain, Biological Journal of the Linnean Society 42: 3-16. Hansson, L. , 1991. Dispersal and Connectivity in Metapopulations. Biological Journal of the Linnean Society 42: 89-103. Hartl, D., and A . G. Clark, 1997. Principles of Population Genetics. Sinauer Associates, Sunderland, M A . Havens, K. , 1998. The Genetics of Plant Restoration: An Overview and a Surprise. Restoration and Management Notes 16: 68-72. Helenurm, K. , 1998. Outplanting and Differential Source Population Success in Lupinus guadalupensis. Conservaton Biology 12: 118-127. Hester, M . W., I. A . Mendelssohn and K . L. McKee, 1998. Intraspecific Variation in Salt Tolerance and Morphology in Panicum hemitomon and Spartina alterniflora (Poaceae). International Journal of Plant Science 159: 127-138. Higgins, S. I., D. M . Richardson, R. M . Cowling and T. H. Trinder-Smith, 1999. Predicting the Landscape-Scale Distribution of Alien Plants and Their Threat to Plant Diversity. Conservation Biology 13: 303-313. Hillis, D., C. Moritz and B. Mable, 1996. Molecular Systematics. Sinauer Associates Inc., Sunderland M A . Hitchcock, A. S., 1971. Manual of the Grasses of the United States. Dover Publications Inc., New York. 80 Holsinger, K. , and R. Mason-Gamer, 1996. Hierarchical Analysis of Nucleotide Diversity in Geographically Structured Populations. Genetics 142: 629-639. Hoover, M . , M . A . Hein, W. Dayton and C. O. Erlanson, 1948. The Main Grasses for Farm and Home. U.S. Department of Agriculture, Washington D.C. Hsiao, J. Y. , 1999. Genetic Diversity and Microgeographic Diversity of Yushan cane (Yushania niitakayamensis; Poaceae) in Taiwan. Molecular Ecology 8: 263-270. Husband, B., and S. Barrett, 1996. A Metapopulation Perspective in Plant Population Biology. Journal of Ecology 84: 461-469. Huxel, G., 1999. Rapid Displacement of Native Species by Invasion Species: Effects of Hybridization. Biological Conservation 89: 143-152. Jaaska, V. , 1992. Isoenzyme Variation in the Grass Genus Elymus (Poaceae). Hereditas 117: 11-22. Jain, S. K. , and D. R. Marshall, 1967. Population Studies in Predominantly Self -Pollinating species. X . Variation in Natural Populations of Avena fatua and A. barbata. The American Naturallist 101: 19-33. Jensen, K . B., 1993. Cytogenetics of Elymus magellanicus and its intra- and inter-generichybrids with Pseudoroegneria spicata, Hordeum violaceaum, E. trachycaulus, and E. glaucus. Genome 36: 72-76. JMP, 1996. JMP, SAS Institute. Kinderscher, K. , and L. L. Tieszen, 1998. Floristic and Soil Organic Matter Changes after Five and Thirty-Five Years of Native Tallgrass Prairie Restoration. Restoration Ecology 6: 181-196. 81 Klinka, K. , A . Ceska and A. M . Scagel, 1989. Indicator Plants of Coastal British Columbia. University of British Columbia Press, Vancouver B.C. Knapp, E. E., and K. J. Rice, 1994. Starting from Seed: Genetic Issues in Using Native Grasses for Restoration. Restoration and Management Notes 12: 40-45. Knapp, E. E., and K. J. Rice, 1996. Genetic Structure and Gene Flow in Elymus glaucus (blue wildrye): Implications for native Grassland Restoration. Restoration Ecology 4: 1-10. Knapp, E. E., and K. J. Rice, 1998. Comparison of Isozymes and Quantitative Traits for Evaluating Patterns of Genetic Variation in Purple Needlegrass (Nassella pulchra). Conservation Biology 12: 1031-1041. Lande, R., 1976. Natural Selection and Random Genetic Drift in Phenotypic Evolution. Evolution 30: 314-334. Lande, R., 1992. Neutral Theory of Quantitative Genetic Variance in an Island Model with Local extinction and colonization. Evolution 46: 381-389. Lavendel, B., 1999. Ecological Restoration in Academia. Ecological Restoration 17: 120-125. Levins, 1970. Extinction, Providence RI. Liedloff, A . , 1999. Mantel Nonparametric Test Caluclator. Loveless, M . D., and J. L. Hamrick, 1984. Ecological Determinants of Genetic Structure in Plant Populations. Annual Review of Ecological Systematics 15: 65-95. Lynch, M . , 1988. Design and Analysis of Experiments on Random Drift and Inbreeding Depression. Genetics 120: 791-807. 82 Maki, M . , and H. Morita, 1998. Genetic Diversity in Island and Mainland Populations of Aster Spathulifolius (Asteraceae). International Journal of Plant Science 159: 148-152. Manicacci, D., I. Olivieri, V . Perrot, A . Atlan, P. Gouyon et al, 1992. Landscape Ecology: Population Genetics at the Metapopulation Level. Landscape Ecology 6: 147-159. Meidinger, D., A . McLeod, A. MacKinnon, C. DeLong and G. Hope, 1988. A Field Guide for Identification and Interpretation of Ecosystems of the Rocky Mountain Trench, Prince George Forest Region. Meidinger, D., and J. Pojar, 1991. Ecosystems of British Columbia. Research Branch Ministry of Forests, Victoria. Meizel, S., and G. L. Markert, 1967. Malate Dehydrogenase Isozymes of the Marine Snail Ilynassa obsoleta. Archives of Biochemistry and Biophysics 122: 753-765. Miao, S. L. , R. E. Borer and F. H. Sklar, 1997. Sawgrass Seedling Responses to Transplanting and Nutrient Additions. Restoration Ecology 5: 162-168. Millar, C. I., and W. J. Libby, 1989. Disneyland or Native Ecosystem: Genetics and the Restorationist. Restoration and Management Notes 7: 18-24. Mitchell, W. R., and R. E. Green, 1981. Identification and Interpretation of Ecosystems of the Western Kamloops Forest Region. Ministry of Forests, Victoria B.C. Montalvo, A . M . , S. L. Williams, K. J. Rice, S. L. Buchmann, C. Cory et al, 1997. Restoration Biology: A Population Biology Perspective. Restoration Ecology 5: 277-290. 83 Muller, S., T. Dutoit, D. Alard and F. Grevilliot, 1998. Restoration and Rehabilitation of Species- Rich Grassland Ecosystems in France: A Review. Restoration Ecology 6: 94-101. Nei, M . , 1978. Estimation of Average Heterozygosity and Genetic Distance from a Small Number of Individuals. Genetics 83: 583-590. Neve, G., L. Mousson and M . Baguette, 1996. Adult Dispersal and Genetic Structure of Butterfly Populations in a Fragmented Landscape. Acta Ecologica 17: 621-626. Nevo, E., I. Apelbaum-Elkaher, J. Garty and A. Beiles, 1997. Natural Selection Causes Microscale Allozyme Diversity in Wild Barley and a Lichen at Evolution Canyon\", Mt. Carmel, Israel. Heredity 78: 373-382. Nevo, E., A . Beiles and D. Kaplan, 1988. Genetic Diversity and Environmental Associations of Wild Emmer Wheat in Turkey. Heredity 61: 31-45. Nevo, E., T. Krugman and A . Beiles, 1994. Edaphic Natural Selection of Allozyme Polymorphism in Aegilopsperegrina at a Galilee microsite in Israel. Heredity 72: 109-112. Noble, I., and R. Dirzo, 1997. Forests as Human-Dominated Ecosystems. Science 277: 522-525. Ouborg, N . , 1993. Isolation, Population Size and Extinction: The Classical and Metapopulation approaches applied to Vascular Plants along the Dutch Rhine-System. Oikos 66: 298-308. Perry, M . C , M . S. Mcintosh and A . K . Stoner, 1991. Geographic Patterns of Variation in the USDA Soybean Germplasm Collection: Allozyme Frequencies. Crop Science 31: 1356-1360. 84 Pilou, E. C , 1991. After the Ice Age. Return of life to Glaciated North America. University of Chicago Press, Chicago and London. Podolosky, R. H. , and T. P. Holtsford, 1995. Population Structure of Morphological Traits in Clarkia dudleyana I. Comparison of F s t Between Allozymes and Morphological Traits. Genetics 140: 733-744. Pojar, J., 1974a. Reproductive dynamics of Four Plant Communities of Southwestern British Columbia. Canadian Journal of Botany 52: 1819-1833. Pojar, J., 1974b. The Relation of the Reproductive Biology of Plants to the Structure and Function of Four Plant Communities, University of British Columbia. Pojar, J., and A. MacKinnon, 1994. Plants of Coastal British Columbia. Lone Pine Publishing, Vancouver. Poulik, M . D., 1957. Starch Gel Electrophoresis in a Discontinuous System of Buffers. Nature 180: 1477-1479. Powell, J. R., 1971. Genetic Polymorphisms in Varied Environments. Science 174: 1035 - 1036. Prout, T., and J. S. F. Barker, 1993. F Statistics in Drosophila buzzatii: Selection, Population Size and Inbreeding. Genetics 134: 369-375. Ransom, C. V. , J. J. Kells, L. M . Wax and M . S. Orfanedes, 1998. Morphological variation among Hemp Dogbane (Apocynum cannabinum) populations. Weed Science 46: 71-75. Richardson, D. M . , 1998. Forestry Trees as Invasive Aliens. Conservation Biology 12: 18-26. Ritland, K. , 1983. Estimation of Mating Systems, pp. in Isozymes in Plant Genetics and Breeding, Part A, edited by S. D. Tanksley, and T. J. Orton. Elsevier Science Publsihers B. V. , Amsterdam. Ritland, K. , 1996. Multilocus Mating System Program, MLTR. Rogers, A. , 1986. Populations Differences in Quantitative Traits as Opposed to Gene frequencies. American Naturalist 127: 729-730. Rosales, J . , G. Cuenca, N . Ramirez and Z. D. Andrade, 1997. Native Colonizing Species and Degraded Land Restoration in La Gran Sabana, Venezuela. Restoration Ecology 5: 147-155. Ruiz, M . , J. M . Carrillo and F. Varela, 1997. Relationship between some Geographical Parameters and Agro\/Morphological and Biochemical Characters in a Sample of Spanish landraces of Barley (Hordeum vulgare L.). Plant Genetic Resource Newsletter 112: 86-89. Sanders, T. B., J. L. Hamrick and L. R. Holden, 1979. Allozyme Variation in Elymus canadensis from the Tall Grass Prairie Region: Geographic Variation. The American Midland Naturalist 101: 1-12. Shields, W. M . , 1982. Philopatry, Inbreeding and the Evolution of Sex. State University of New York Press, New York. Silander, J. A . , 1984. The Genetic Basis of the Ecological Amplitude of Spartina patens. Botanical Gazette 145: 569-577. Sinsch, U . , 1992. Structure and Dynamic of a Natterjack Toad Metapopulation (Bufo calamitd). Oecologia 90: 489-499. 86 Slatkin, M . , 1987. Gene flow and Geographic Structure in Natural Populations. Science 236: 787-792. Slatkin, M . , 1994. Gene Flow and Population Structure. Princeton University Press, Princeton, NJ. Slatkin, M . , and N . H. Barton, 1989. A comparison of three Indirect Measures for Estimation of Average Levels of Gene Flow. Evolution 43: 1349-1368. Snaydon, R. W., and M . S. Davies, 1982. Rapid Divergence of Plant Populations in Response to Recent Changes in Soil Conditions. Evolution 36: 289-297. Spitze, K. , 1993. Population Structure in Daphnia obtusa: Quantitative Genetic and Allozymic Variation. Genetics 135: 367-374. Swofford, D., and R. Selander, 1997. Biosys-2. Taiz, L. , and E. Zeiger, 1991. Plant Physiology. The Benjamin\/Curnmings Publishing Company, Inc., Redwood City California. Taylor, R., and B. MacBryde, 1977. Vascular Plants of British Columbia: A descriptive Resource Inventory. The University of British Columbia Press, Vancouver, B.C. Templeton, A . R., 1986. Co-adaptation and Outbreeding Depression. Sinauer Associates, Massachusettes. Thomas, C. D., and S. Harrison, 1992. Spatial Dynamics of a Patchy Distributed Butterfly Species. Journal of Animal Ecology 61: 437-446. Veverka, 1998. Revegetating following Logging on Decomposed Granite. Restoration and Management Notes 16: 182-188. 87 Volis, S., and S. Mendlinger, 1998. Phenotypic variation and stress resistance in core and peripheral populations of Hordeum spontaneum. Biodiversity and Conservation 7: 799-813. Waldmann, P., and S. Andersson, 1998. Comparison of Quantitative Genetic Variation and Allozyme Diversity within and between Populations of Scabiosa canescens and S. columbaria. Heredity 81: 79-86. Walters, T. W., D. S. Decker-Walters and D. Gordon, 1994. Restoration Considerations for Wiregrass (Aristida stricta): Allozymic Diversity of Populations. Conservation Biology 8: 581-585. Whitlock, M . , and D. McCauley, 1990. Some Population Genetic Consequences of Colony Formation and Extinction: Genetic Correlations within Founding Groups. Evolution 44: 1717-1724. Whitlock, M . C , 1992. Non-equilibrium Population Structure in Forked Fungus Beetles: Extinction, Colonization, and the Genetic Variance among Populations. The American Naturalist 139: 952-970. Williams, S. L. , and C. A . Davis, 1996. Population Genetic Analyses of Transplanted Eelgrass (Zostera marina) Beds Reveal Reduced Genetic Diversity in Southern California. Restoration Ecology 4: 163-180. Wright, S., 1940. Breeding Structure of Populations in Relation to Speciation. American Naturalist 74: 232-248. Wright, S., 1951. The Genetic Structure of Populations. Annals of Eugenics 15: 323-354. 88 Wright, S., 1965. The Interpretation of Population Structure by F-statistics with Special regard to Mating Systems. Evolution 19: 395-420. Wright, S., 1978. Evolution and the Genetics of Populations. Chicago Press, Chicago. Yaffee, S., 1999. Three Faces of Ecosystem Management. Conservation Biology 13: 713-725. Yang, R . - C , F. C. Yeh and A . Yanchuk, 1996. A Comparison of Isozyme and Quantitative Genetic Variation in Pinus contorta ssp. latifolia by FSf. Genetics 142: 1045-1052. Yole, D., T. Lewis, A . Inselberg, J. Pojar and D. Holmes, 1989. A Field Guide for Identification and Interpretation of Engelmann Spruce-Subalpine Fir Zone in the Prince Rupert Forest Region British Columbia. Ministry of Forests, Victoria B.C. Appendix 11\u2014 1 \u2014 MDH1 (monomer) 22 \u2014 22\u2014 2 \u2014 1 \u2014 11\u2014 1 \u2014 MDH2 (monomer) 2 22 2 11 11 11 11 MDH3 (monomorphic) Figure A l . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Malate Dehydrogenase 11 \u2022 \u2022 1 1 \u2014 1 1 \u2014 12\u2014 22 SKD1 (monomer) homoallelic 11 1 2 \u2014 22 \u2014 2 2 \u2014 22 H SKD2 (monomer) mobility 11 null SKD3 (monomer) 22 \u2014 Figure A 2 . Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Shikimate Dehydrogenase 11\u2014 1 1 \u2014 null 11\u2014 1 1 ~ \" DIA1 (monomorphic) I \u2014 11\u2014 DIA2 (monomer) 22\u2014 2 \u2014 null 22 \u2014 II \u2014 11 \u2014 DIA3 (dimer) 22\u2014 2 2 \u2014 2 2 \u2014 22 \u2014 Figure A3. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Diaphorase 11 \u2014 1 \u2014 null IDH1 (monomer) 2 2 \u2014 2 \u2014 11\u2014 11\u2014 11\u2014 11 \u2014 IDH2 (monomorphic) Figure A4. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Isocitric Dehydrogenase 11 \u2014 11 \u2014 PGI1 (monomer) 22 \u2014 1 1 1 1 M fixed heterozygote 1 1 2 2 _ 11 PGI 2 (monomer) 1122 \u2122 fixed heterozygote 22 2 2 _ Figure A5. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Phosophoglucoisomerase 91 11 \u2014 11\u2014 11 \u2014 11\u2014 M E 1 (monomorphic) 11 1 null ME2(dimer) 1 2 \u2122 22~~ 2 \u2014 Figure A6. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Malic Enzyme 11 ~ ~ P G M 1 (monomorphic) 11 \u2014 P G M 2 (monomorphic) Figure A7. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Phosphoglucomutase J J \u2014 12\u2014 1 1 \u2122 1111\u2122 G D H 1 (monomer) \u2014 fixed heterozygote 2 2 \u2014 2222 \u2014 2 2 1 2 \u2014 2222 ~ G D H 2 (monomer) n u l l 11 \u2014 G D H 3 (monomer) Figure A8. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Glutamate Dehydrogenase 92 11 \u2014 A D H 1 \"\u2122\" fixed heterozygote 11 _ 12 1 1 M 1122 H 11 \u2014 A D H 2 (monomer) \"\"\" Fixed heteroygote 22 \u2014 2 2 2 2 H 22 H 12 A D H 3 (monomer) Figure A9. Banding patterns, locus designations, enzyme sub-unit structure, and genotype designations for Alcohol Dehydrogenase Table A l . Populations grouped by elevation in meters. Elevation groups Populations 0 to 250m 3,11 250m to 500m 5, 30 500m to 750m 4,6,7,13,29 750m to 1000m 1, 2, 9, 10, 14,15, 16, 17, 28, 33 1000m and above 8,12,27,31,32 93 Table A2. Populations grouped by average yearly snowpack. Snowpack groups Populations Ocm to 100cm 1,4,5,7, 11, 16, 28,30 100cm to 200cm 12,14,27,31 200cm to 300cm 2, 6, 8, 9, 10, 15, 17, 29,33 300cm and above 3, 13,32 Table A3. Populations grouped by average yearly temperature. Temperature groups Populations -3\u00b0C to 0\u00b0C 10, 12, 27, 29,31,32 0\u00b0Cto 1.5\u00b0C 2, 6, 8, 15 1.5\u00b0C to2.5\u00b0C 4, 7,16, 28, 33 2.5\u00b0C to 4\u00b0C 9, 14, 17 4\u00b0C and above 1,3,5, 11, 13,30 94 Table A4. Populations grouped by average yearly precipitation Precipitation groups Populations 250mm to 350mm 4, 7, 10, 16, 28, 29 350mm to 600mm 2, 6, 8, 9, 12, 17, 27 600mm to 800mm 30,33 800mm and above 1,3,5,11,13,14,31,32 Table A5. Populations grouped by the number of frost free days Frost Free day groups Populations 0 days - 80 days 10, 12,27, 29,31 81 days - 150 days 2, 6, 8, 15, 32 151 days - 180 days 1,3,5,9, 11, 13, 14, 17,30 180 days and above 4, 7, 16, 28, 33 Table A6. Populations grouped by soil type Soil type Populations Brunisolic 1,4, 5, 7, 9, 10, 11, 16, 17, 18, 19, 20, 28, 29, 30 Gray Luvisols 2, 6, 8, 12, 15 OHFP 3, 13,14 Podzols 21,22, 23,34,35,36 95 BEC for each population is described in Pojar (1991). Average yearly snow pack, temperature, number of frost free days, and precipitation were obtained through the use of numerous field guide books used for site identification for British Columbia (Banner et al. 1993; DeLong 1988; Delong et al. 1990; DeLong et al. 1993; Delong et al. 1994; Green and Klinka 1994; Meidinger et al. 1988; Meidinger and Pojar 1991; Mitchell and Green 1981; Yole et al. 1989). 96 Table A7. Enzymes, number of loci resolved, buffer system, and stain components. Enzyme # of loci scored Buffer Stain components Alcohol Dehydrogenase (ADH, E.C. 1.1.1.1) 3 Poulik 50ml0.1MTris-Hcl pH 8.5 5ml 95% ethanol 5mg N A D , 20mg M T T 5mg PMS Diaphorase (DIA) 3 Tris Citrate 50 ml O.lMTris-Hcl pH 8.0 12mg N A D H reduced lOmgMTT, 5mg PMS Glutamate Dehydrogenase (GDHE.C . 1.4.1.2) 2 Poulik 50ml O.lMTris-Hcl pH 8.5 400mg D-glutamate N A D , lOmgMTT 5mg PMS Isocitric Dehydrogenase (IDH, E.C. 1.1.1.42) 2 Tris Citrate 50ml 0.2M Tris-Hcl pH 8.0 lOOmg Isocitric Acid 5mg MgC12, 5mg N A D P 5mgPMS, lOmgMTT Malate Dehydrogenase ( M D H E . C . 1.1.1.37) 3 Tris Citrate 45ml 0.2M Tris-Hcl pH 8.0 45ml Malate Solution 5mg N A D , lOmg M T T 5mg PMS Phosphoglucoisomerase (PGI E.C. 5.4.2.2) 3 Tris Citrate 50ml 0.2M Tris-Hcl pH 8.0 20mg Fructose-6-Phosphate 2ml Glucose-6-Phosphate Dehydrogenase 5mg N A D P , 5mg MgC12 5mgPMS, lOmgMTT Phosphoglucomutase (PGM E.C. 5.4.2.2) 2 Poulik 50ml 0.2M Tris-Hcl pH 8.0 25mg D-Glucose-1-Phosphate, 5mg MgC12 5mg N A D P , lOmg M T T 5mg PMS Shikimate Dehydrogenase (SKD 1.1.1.25) 3 Trsi Citrate 50ml 0.1 Tris-Hcl pH 8.0 lOOmg Shikimate Acid 5mg N A P D , lOmg M T T 5mg PMS Allele Frequencies MDH1 Pop 1 Pop 2 Pop 3 Pop 4 Pop 5 Pop 6 Pop 7 Pop 8 Pop 9 1 1.000 1.000 1.000 1.000 0.984 0.984 1.000 1.000 1.000 2 0.000 0.000 0.000 0.000 0.016 0.016 0.000 0.000 0.000 N 32 32 16 32 32 32 32 32 32 MDH2 1 0.781 0.344 0.125 0.125 1.000 0.031 0.000 0.000 0.000 2 0.219 0.656 0.875 0.875 0.000 0.969 1.000 1.000 1.000 N 32 32 16 32 32 32 32 32 32 SKDH1 1 0.969 0.969 1.000 0.969 0.969 0.938 0.922 0.922 1.000 2 0.031 0.031 0.000 0.031 0.031 0.063 0.078 0.078 0.000 N 32 32 16 32 32 32 32 32 32 SKDH2 1 0.953 0.953 0.938 0.969 0.938 0.938 1.000 0.969 0.891 2 0.047 0.047 0.063 0.031 0.063 0.063 0.000 0.031 0.109 N 32 32 16 32 32 32 32 32 32 SKDH3 1 1.000 1.000 1.000 1.000 0.969 0.969 0.969 0.969 0.969 2 0.000 0.000 0.000 0.000 0.031 0.031 0.031 0.031 0.031 N 32 32 16 32 32 32 32 32 32 DIM 1 1.000 1.000 1.000 1.000 1.000 0.156 0.906 1.000 1.000 3 0.000 0.000 0.000 0.000 0.000 0.844 0.094 0.000 0.000 N 32 32 16 32 32 32 32 32 32 DIA2 1 0.781 0.750 1.000 0.984 1.000 1.000 1.000 1.000 0.938 2 0.219 0.250 0.000 0.016 0.000 0.000 0.000 0.000 0.063 N 32 32 16 32 32 32 32 32 32 PG11 1 0.250 0.313 0.938 0.656 1.000 0.969 0.906 0.250 0.000 2 0.750 0.688 0.063 0.344 0.000 0.031 0.094 0.750 1.000 N 32 32 16 32 32 32 32 32 32 PGI2 1 1.000 1.000 1.000 1.000 1.000 1.000 1.000 1.000 0.000 2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 N 32 32 16 32 32 32 32 32 32 IDH1 1 1.000 2 0.000 N 32 1.000 1.000 0.000 0.000 32 16 0.984 1.000 0.016 0.000 32 32 1.000 1.000 0.000 0.000 32 32 0.000 0.000 0.000 0.000 0 0 IDH2 1 2 N ME1 1 N ME2 1 2 N PGM1 1 N PGM2 1 N GDH1 1 2 3 N GDH2 1 2 N ADH1 1 2 N ADH2 1 2 3 N ADH3 1 3 N 1.000 0.219 0.125 0.000 0.781 0.875 32 32 16 1.000 1.000 0.000 32 32 0 0.813 0.875 0.000 0.188 0.125 0.000 32 32 0 1.000 1.000 1.000 32 32 16 1.000 1.000 1.000 32 32 16 1.000 1.000 0.875 0.000 0.000 0.125 0.000 0.000 0.000 32 32 16 1.000 1.000 1.000 0.000 0.000 0.000 32 32 16 1.000 1.000 0.813 0.000 0.000 0.188 32 32 16 1.000 1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 32 32 16 0.000 0.000 0.000 1.000 1.000 1.000 32 32 16 0.063 1.000 1.000 0.938 0.000 0.000 32 32 32 0.000 0.000 0.000 0 0 0 0.000 0.000 0.000 0.000 0.000 0.000 0 0 0 1.000 1.000 1.000 32 32 32 1.000 1.000 1.000 32 32 32 1.000 1.000 0.000 0.000 0.000 0.000 0.000 0.000 1.000 32 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 0.984 0.016 0.922 0.016 0.000 0.078 0.000 0.984 0.000 32 32 32 0.000 0.969 0.000 1.000 0.031 1.000 32 32 32 1.000 0.000 0.000 0.000 0.000 0.000 32 0 0 0.000 1.000 1.000 0 32 32 0.000 0.000 0.000 0.000 1.000 1.000 0 32 32 1.000 1.000 1.000 32 32 32 1.000 1.000 1.000 32 32 32 0.000 1.000 1.000 0.000 0.000 0.000 1.000 0.000 0.000 32 32 32 1.000 0.969 0.969 0.000 0.031 0.031 32 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 0.938 1.000 1.000 0.063 0.000 0.000 0.000 0.000 0.000 32 32 32 0.000 0.000 0.000 1.000 1.000 1.000 32 32 32 MDH1 1 2 3 N MDH2 1 2 N SKDH1 1 2 N SKDH2 1 2 N SKDH3 1 2 N DIM 1 N DIA2 1 2 N PGI1 1 2 N PGI2 1 2 N IDH1 1 N IDH2 1 2 N PoplO P o p l l Popl2 1.000 0.766 1.000 0.000 0.234 0.000 0.000 0.000 0.000 8 32 32 0.938 1.000 1.000 0.063 0.000 0.000 8 32 32 1.000 0.938 0.938 0.000 0.063 0.063 8 32 32 1.000 0.984 1.000 0.000 0.016 0.000 8 32 32 1.000 0.969 0.969 0.000 0.031 0.031 8 32 32 1.000 1.000 1.000 8 32 32 1.000 1.000 0.984 0.000 0.000 0.016 8 32 32 0.750 0.031 0.000 0.250 0.969 1.000 8 32 32 1.000 0.969 1.000 0.000 0.031 0.000 8 32 32 1.000 0.000 0.000 8 0 0 0.125 0.000 0.000 0.875 0.000 0.000 8 0 0 Popl3 Popl4 Popl5 1.000 0.969 1.000 0.000 0.031 0.000 0.000 0.000 0.000 32 32 32 0.000 0.984 0.031 1.000 0.016 0.969 32 32 32 0.953 1.000 0.922 0.047 0.000 0.078 32 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 0.969 0.969 0.969 0.031 0.031 0.031 32 32 32 1.000 1.000 1.000 32 32 32 0.938 0.656 1.000 0.063 0.344 0.000 32 32 32 0.063 1.000 0.000 0.938 0.000 1.000 32 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 0.000 1.000 1.000 0 32 32 0.000 1.000 1.000 0.000 0.000 0.000 0 32 32 Popl6 Popl7 Popl8 0.922 0.516 0.891 0.078 0.031 0.109 0.000 0.453 0.000 32 32 32 0.391 0.063 1.000 0.609 0.938 0.000 32 32 32 0.984 0.938 0.750 0.016 0.063 0.250 32 32 32 1.000 0.984 0.969 0.000 0.016 0.031 32 32 32 0.594 1.000 0.281 0.406 0.000 0.719 32 32 32 1.000 1.000 1.000 32 32 32 0.984 0.781 0.656 0.016 0.219 0.344 32 32 32 0.875 0.813 0.688 0.125 0.188 0.313 32. 32 32 1.000 1.000 1.000 0.000 0.000 0.000 32 32 32 1.000 1.000 1.000 32 32 32 0.469 1.000 0.313 0.531 0.000 0.688 32 32 32 ME1 1 N ME2 1 2 3 N PGM1 1 N PGM2 1 N GDH1 1 2 N GDH2 1 2 N ADH1 1 2 N ADH2 1 2 3 N ADH3 1 3 N 1.000 1.000 1.000 8 32 32 0.250 0.000 0.016 0.750 1.000 0.984 0.000 0.000 0.000 8 32 32 1.000 1.000 1.000 8 32 32 1.000 1.000 1.000 8 32 32 0.563 1.000 1.000 0.438 0.000 0.000 8 32 32 1.000 0.938 1.000 0.000 0.063 0.000 8 32 32 1.000 1.000 1.000 0.000 0.000 0.000 8 32 32 0.500 1.000 1.000 0.000 0.000 0.000 0.500 0.000 0.000 8 32 32 0.500 0.000 0.000 0.500 1.000 1.000 8 32 32 1.000 1.000 1.000 32 32 32 0.141 1.000 1.000 0.859 0.000 0.000 0.000 0.000 0.000 32 32 32 1.000 1.000 1.000 32 32 32 1.000 1.000 1.000 32 32 32 0.719 1.000 1.000 0.281 0.000 0.000 32 32 32 0.922 1.000 0.828 0.078 0.000 0.172 32 32 32 0.688 1.000 1.000 0.313 0.000 0.000 32 32 32 1.000 0.953 1.000 0.000 0.047 0.000 0.000 0.000 0.000 32 32 32 0.000 0.000 0.000 1.000 1.000 1.000 32 32 32 1.000 1.000 1.000 32 32 32 0.094 0.000 0.609 0.781 1.000 0.391 0.125 0.000 0.000 32 32 32 1.000 1.000 1.000 32 32 32 1.000 1.000 1.000 32 32 32 0.469 0.609 0.781 0.531 0.391 0.219 32 32 32 1.000 1.000 0.625 0.000 0.000 0.375 32 32 32 1.000 1.000 0.844 0.000 0.000 0.156 32 32 32 0.625 1.000 0.938 0.000 0.000 0.063 0.375 0.000 0.000 32 32 32 0.375 0.000 0.000 0.625 1.000 1.000 32 32 32 101 MDH1 MDH2 SKDH1 SKDH2 SKDH3 D I M DIA2 Popl9 Pop20 Pop21 Pop22 Pop23 Pop24 Pop25 Pop26 1 1.000 1.000 0.938 1.000 1.000 1.000 0.972 0.989 2 0.000 0.000 0.063 0.000 0.000 0.000 0.028 0.011 N 32 32 32 32 32 130 . 89 140 1 0.000 0.000 0.250 0.000 0.547 0.408 0.124 0.039 2 1.000 1.000 0.750 1.000 0.453 0.592 0.876 0.961 N 32 32 32 32 32 130 89 140 1 0.969 1.000 1.000 0.984 0.984 0.958 1.000 0.989 2 0.031 0.000 0.000 0.016 0.016 0.042 0.000 0.011 N 32 32 32 32 32 129 89 140 1 0.859 0.906 0.250 1.000 0.953 0.852 0.949 0.921 2 0.141 0.094 0.750 0.000 0.047 0.148 0.051 0.079 N 32 32 32 32 32 128 89 140 1 0.750 0.000 1.000 0.000 1.000 0.993 1.000 1.000 2 0.000 0.000 0.000 0.000 0.000 0.007 0.000 0.000 3 0.250 1.000 0.000 1.000 0.000 0.000 0.000 0.000 N 32 32 32 32 32 70 89 68 1 1.000 1.000 1.000 1.000 1.000 0.946 1.000 1.000 2 0.000 0.000 0.000 0.000 0.000 0.054 0.000 0.000 N 32 32 32 32 32 129 89 140 1 1.000 1.000 1.000 0.953 1.000 0.985 0.966 0.632 2 0.000 0.000 0.000 0.047 0.000 0.015 0.034 0.368 N 32 32 32 32 32 130 89 140 PGI1 PGI2 IDH1 IDH2 1 1.000 0.063 0.000 1.000 0.594 0.000 1.000 1.000 2 0.000 0.938 1.000 0.000 0.406 0.000 0.000 0.000 N 32 32 32 32 32 0 89 140 1 1.000 1.000 1.000 1.000 1.000 0.000 1.000 1.000 N 32 32 32 32 32 0 89 140 1 1.000 1.000 1.000 1.000 1.000 0.000 1.000 1.000 N 32 32 32 32 32 0 89 140 1 0.531 0.000 0.813 0.000 0.000 0.000 0.000 0.900 2 0.000 0.031 0.000 0.125 0.000 0.000 1.000 0.100 3 0.469 0.969 0.188 0.875 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 89 140 ME1 ME2 PGM1 PGM2 GDH1 GDH2 ADH1 ADH3 1 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 0.781 0.000 0.000 0.000 0.000 0.000 0.000 0.000 2 0.219 1.000 1.000 0.000 0.000 0.000 0.000 0.000 3 0.000 0.000 0.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 0.969 0.000 0.000 1.000 1.000 0.000 0.000 0.000 2 0.031 1.000 0.000 0.000 0.000 0.000 0.000 0.000 3 0.000 0.000 1.000 0.000 0.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 1 1.000 0.031 1.000 1.000 1.000 0.000 0.000 0.000 2 0.000 0.969 0.000 0.000 0.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 3 1.000 1.000 1.000 1.000 1.000 0.000 0.000 0.000 N 32 32 32 32 32 0 0 0 ","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/hasType":[{"value":"Thesis\/Dissertation","type":"literal","lang":"en"}],"http:\/\/vivoweb.org\/ontology\/core#dateIssued":[{"value":"2000-05","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/isShownAt":[{"value":"10.14288\/1.0099481","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/language":[{"value":"eng","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#degreeDiscipline":[{"value":"Forestry","type":"literal","lang":"en"}],"http:\/\/www.europeana.eu\/schemas\/edm\/provider":[{"value":"Vancouver : University of British Columbia Library","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/publisher":[{"value":"University of British Columbia","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/rights":[{"value":"For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https:\/\/open.library.ubc.ca\/terms_of_use.","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#scholarLevel":[{"value":"Graduate","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/title":[{"value":"Reclaiming disturbed habitats using native grasses : the genetic story of Elymus glaucus (blue wildrye)","type":"literal","lang":"en"}],"http:\/\/purl.org\/dc\/terms\/type":[{"value":"Text","type":"literal","lang":"en"}],"https:\/\/open.library.ubc.ca\/terms#identifierURI":[{"value":"http:\/\/hdl.handle.net\/2429\/10267","type":"literal","lang":"en"}]}}