Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ecological genetics and effects of inbreeding and white pine blister rust on genetic structure of whitebark… Bower, Andrew David 2006

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

Item Metadata

Download

Media
831-ubc_2006-199428.pdf [ 7.99MB ]
Metadata
JSON: 831-1.0075061.json
JSON-LD: 831-1.0075061-ld.json
RDF/XML (Pretty): 831-1.0075061-rdf.xml
RDF/JSON: 831-1.0075061-rdf.json
Turtle: 831-1.0075061-turtle.txt
N-Triples: 831-1.0075061-rdf-ntriples.txt
Original Record: 831-1.0075061-source.json
Full Text
831-1.0075061-fulltext.txt
Citation
831-1.0075061.ris

Full Text

E C O L O G I C A L G E N E T I C S A N D E F F E C T S O F INBREEDING A N D W H I T E PINE B L I S T E R R U S T O N G E N E T I C S T R U C T U R E O F W H I T E B A R K PINE (Pinus albicaulis Engelm.) by ANDREW DAVID BOWER B. Sc. University of California at Berkeley, 1994 M . Sc. Oregon State University, 1998 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY In THE FACULTY OF GRADUATE STUDIES (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA August 2006 © Andrew David Bower, 2006 , A B S T R A C T In this thesis I investigate the ecological genetics of whitebark pine (Pinus albicaulis Engelm.), a five-needle pine found near treeline in western North America. Throughout its range whitebark pine is experiencing declines due to the introduced disease white pine blister rust and successional replacement due to fire suppression. Conservation and restoration efforts are needed to reverse this decline; however, little is known about the genetics of whitebark pine as related to quantitative traits. A common garden experiment was used to assess rangewide genetic variation and geographic differentiation of quantitative traits. Significant variation among provenances was detected for all traits except spring cold injury. Geographic differentiation was weak to moderate for growth and biomass traits, but was strong for fall cold injury and date of needle flush. Both of these variables were significantly associated with mean temperature of the coldest month at of source location. These results were used to determine guidelines for seed movement in whitebark pine management and restoration. Isozyme analysis of seed tissues was used to determine the mating system and presence and strength of inbreeding depression in quantitative traits. Regional estimates of outcrossing rate (/,„) ranged from 0.73 to 0.93, with a mean of 0.86. Family mean t„, values were skewed towards outcrossing; however, some individuals experienced substantial inbreeding. Biomass in the southern B.C. region was the only trait and geographic origin with significant inbreeding depression, and is predicted to be reduced by 19.6% in this region. Isozyme analysis of bud tissue was used to determine genetic diversity (H 0 and H e) and fixation index (Fis) for three age cohorts, from 14 sites. Significant evidence of inbreeding (Fis>0) was found in all age cohorts. When sites were stratified by level of blister rust infection, differences in Fjs and H 0 among cohorts were only significant when level of infection was low. A significant negative association with level of rust infection was found for H 0 only in the mature cohort. This ii suggests that when differential selection due to rust is weak, more heterozygous individuals are favoured. However, more homozygous individuals appear to have higher fitness when rust infection levels are high. 111 T A B L E O F C O N T E N T S Abstract ii Table of contents iv List of tables vii List of figures ix Acknowledgements x Co-authorship statement xii Chapter 1: Thesis introduction and literature review 1 1.1 Introduction 1 1.2 Literature Review 2 1.2.1 Ecology of whitebark pine 2 1.2.1.1 Biology and seed dispersal 2 1.2.1.2. Population genetics 5 1.2.1.3 Ecological threats 6 1.2.2 Ecological influences on plant population structure 7 1.2.3 Quantitative trait variation in forest trees 8 1.2.4 Inbreeding....... '. 10 1.2.4.1 Genetic basis of inbreeding 10 1.2.4.2 Purging 11 1.2.4.3 Plant mating systems 12 1.2.5 Heterozygosity-fitness correlation 13 1.3 Thesis introduction 14 1.4 References 17 Chapter 2: Genetic diversity and geographic differentiation in quantitative traits, and seed transfer guidelines for whitebark pine (Pinus albicaulis Engelm.) 23 2.1 Introduction 23 2.2 Materials and Methods 27 2.2.1 Sample materials 27 2.2.2 Data analysis 28 2.3 Results 31 2.3.1 Soil temperature effects 31 2.3.2 Genetic variation and differentiation 32 2.4 Discussion 35 2.4.1 Effects of soil temperature 35 iv 2.4.2 Genetic variation 35 2.4.3 Environmental effects on quantitative traits 36 2.4.4 Comparison of genetic differentiation 37 2.4.5 Seed transfer guidelines 38 2.5 Acknowledgments 41 2.6 References 49 Chapter 3: Mating system and inbreeding depression in whitebark pine {Pinus albicaulis Engelm.)..52 3.1 Introduction 52 3.2 Materials and methods 55 3.2.1 Sample materials 55 3.2.2 Isozyme analysis 56 3.2.3 Data analysis 57 3.3 Results 59 3.3.1 Inbreeding coefficients 59 3.3.2 Outcrossing rates 59 3.3.3 Inbreeding depression 60 3.4 Discussion 61 3.4.1 Outcrossing rate and inbreeding coefficients 61 3.4.2 Inbreeding depression 63 3.5 Acknowledgements 66 3.6 References 71 Chapter 4: Changes in genetic structure of whitebark pine (Pinus albicaulis Engelm.) associated with inbreeding and white pine blister rust infection '. 76 4.1 Introduction 76 4.2 Materials and methods 78 4.2.1 Sample materials • 78 4.2.2 Isozyme analysis 79 4.2.3 Data analysis 79 4.3 Results 80 4.4 Discussion 81 4.5 Acknowledgements 86 4.6 References 92 Chapter 5: Geographical and seasonal variation in cold hardiness of whitebark pine (Pinus albicaulis Engelm.) 96 5.1 Introduction 96 5.2 Materials and methods 98 5.2.1 Source materials 98 5.2.2 Freeze testing and phonological observations 99 5.2.3 Statistical analysis 100 5.3 Results 102 5.3.1 Seasonal variation 102 5.3.2 Differentiation among regions in fall and spring 103 5.3.3 Genetic control and correlations 103 5.3.4 Environmental effects on cold hardiness 104 5.4 Discussion 104 5.5 Acknowledgements 108 5.6 References • 115 Chapter 6: Thesis conclusions 117 6.1 Introduction 117 6.2 Major Findings 118 6.2.1 Genetic diversity and local adaptation of quantitative traits 118 6.2.2 Mating system and inbreeding depression 120 6.2.3 Relationship of genetic structure with inbreeding and white pine blister rust infection 121 6.3 Future Research 122 6.4 References 126 vi LIST O F T A B L E S Table 1.1. Reported relationships between individual heterozygosity and adaptive traits for several conifer species 16 Table 2.1. Whitebark pine provenances sampled, number of seedlings tested, geographic coordinates, and climatic information 42 Table 2.2. Soil temperature treatment least-squares-means, standard deviations, and significance level for eight quantitative traits 43 Table 2.3. Significance level of provenance effect in ANOVA, percent of variation accounted for by provenance and family, and genetic differentiation (QST) for nine quantitative traits in ambient (A) soil temperature treatment and seven traits in cold (C) treatment 43 Table 2.4. Reported values of genetic differentiation for whitebark pine and other stone pine (Pinus subsection Cembrae) species : 44 Table 2.5. Correlations among provenance means for nine quantitative variables in ambient soil temperature treatment. Correlations when |r|>0.5 significant at <x=0.05 after Bonferroni adjustment for number of correlations tested (n=36) 44 Table 2.6. Correlations among provenances for seven climatic variables 45 Table 2.7. Correlations among provenance means for quantitative and climatic variables 45 Table 2.8. Canonical correlation analysis of the relationship between provenance mean quantitative traits and climatic variables 45 Table 2.9. Correlations between quantitative canonical variables and the quantitative variables, and between climate canonical variables and both climate and quantitative variables 46 Table 3.1. Number of provenances (Prov.), families (Fam.), total seedlings genotyped, and mean number of seedlings genotyped by family using isozyme analysis and number of families grown in two temperature treatments(amb. and cold) in a common garden (CG) experiment for three sampling regions 67 Table 3.2. Population mean estimates of multi- and single-locus outcrossing rates (tm and ts), mean parental inbreeding coefficient (Fp) and multilocus correlation of paternity (rp) with standard errors in parentheses 67 Table 3.3. Regression line slope (inbreeding load) for family mean of log-transformed seedling traits on Fe in two soil temperature treatments for 18 families from southern B.C 67 Table 3.4. Outcrossing rates of four stone pine species (Pinus subsection Cembrae) and mean outcrossing rate of subsection cembrae and a sample of genus Pinus 68 Table 4.1. Study sites, geographic locations, and white pine blister rust infection 87 Table 4.2. Fixation indices (Fis) at three life stages for several conifers 87 Table 5.1. Geographic regions of whitebark pine tested for cold injury 109 vn Table 5.2. Dates and test temperatures o f artificial freeze tests 109 Table 5.3. Sources of variation, p-values for F-statistics in A N O V A and % of total variance for fall and spring cold injury 110 Table 5.4. Phenotypic correlations among regional means (below diagonal) and family means (above diagonal) for cold injury in different seasons, with number of observations (n) in parentheses.. 110 Table 5.5. Freezing tolerance o f species o f Pinus in the subgenus Strobus (adapted from Oohata and Sakai 1982) I l l v i i i LIST O F F I G U R E S Figure 2.1. Distribution of whitebark pine and locations of provenances tested in common garden experiment. Dashed lines separate the Southern, Rocky Mountain and Northern regions 47 Figure 2.2. Regression of first quantitative canonical score (QS1) on standardized mean temperature of the coldest month (MTCM) for 41 whitebark pine provenances in three geographic regions. Axes scales are standard deviations and bracket indicates value of LSD 0.20 48 Figure 2.3. Scatterplot of first two quantitative canonical scores (QC1 and QC2) based on eight quantitative traits for 41 provenances of whitebark pine. Axis scales are standardized values. Symbols refer to geographic groups shown in table 1 48 Figure 3.1. Regional mean estimates of observed parental and offspring inbreeding coefficient (Fp and F0). Error bars are + 2 standard errors 69 Figure 3.2. Distribution of multilocus outcrossing rate by region 69 Figure 3.3. Family mean ln(biomass) vs. expected equilibrium inbreeding coefficient (Fe) for three regions with regression line for southern B.C. region 70 Figure 4.1. Map of the range of whitebark pine and locations and level of blister rust infect of 14 sample locations 88 Figure 4.2. Mean fixation index (Fjs) by cohort for low, moderate, and high rust infection level sites (error bars are 95% confidence intervals 89 Figure 4.3. Mean observed heterozygosity (H0) by cohort for low, moderate, and high rust infection level sites (error bars are 95% confidence intervals) 90 Figure 4.4. Scatterplot of (a-c) fixation index (F;s) and (d-f) observed heterozygosity (HG) of whitebark pine versus infection percent for 14 sites 91 Figure 5.1. Range of whitebark pine with designated regions and parental source locations of trees tested for cold hardiness 112 Figure 5.2. Seasonal change in mean LT50 of whitebark pine. Estimates of winter cold hardiness are truncated at the minimum testing temperature (-70°C.) 113 Figure 5.3. Regional LSmean cold injury vs. latitude (a-1 through 4) at four test dates and vs. mean temperature of the coldest month (MTCM) for fall and spring (b-1,2) indicating clinal variation in fall cold hardiness associated with temperature. Error bars are + 2 standard errors 114 i x A C K N O W L E D G E M E N T S I would like to express my thanks to everyone who has helped me during my time at UBC. My advisor, Sally Aitken, who I also view as a colleague and friend, has provided me with the opportunity to do this project, as well as providing financial support, answers, advice, guidance, and review and critique of my work and writing along the way. I also thank my committee members, Mike Whitlock, Jeannette Whitton, and Alvin Yanchuk, for their advice, suggestions, and critiques of my work. I would also like to thank my thesis examiners, Yousry El-Kassaby, Sean Graham, and Diana Tomback for their feedback which improved this thesis. The staff of the Faculty of Forestry and the Department of Forest Sciences were invaluable in making sure that the administrative details of being a grad student were looked after. I am especially grateful to all of the people and organizations that provided the seed and other material used in this study: regions 1, 5 and 6 of the U.S. Forest Service, specifically Randy Menke, Stewart Cook, Donna Stubbs, Paul Berrang, Chris Jensen, Tom DeSpain, Vicky Erickson, Dave Doede, Nancy Lankford, Paul Stover, Jay Kitzmiller, Donna Dekker-Robertson, Aram Eramian, Richard Sniezko, and Jerry Berdeen; the B.C. Ministry of Forests, Surrey Tree Seed Centre, especially Dave Kolotelo who provided seed, materials, resources, and advice on stratification and germination; B.C. Parks E.C. Manning and Tweedsmuir Provincial Parks; Bob Brett of Snowline Ecological Consulting, Whistler, B.C., and Don Pigott of Yellow Point Propagation, Ladysmith, B.C. This project could not have happened without the hard work of many people in the raised beds, the lab, and the field, including: Joanne Tuytel, Jodie Rrakowski, Christine Chourmouzis, Pia Smets, Makiko Mimura, Dane Szohner, Milena Semproni, Washy Gapare, Megan Harrison, Karolyn Keir, and Seane Trehearne. Tongli Wang and Andreas Hamann provided advice on data analysis, and Tongli also provided all of the climate date for the Canadian provenances. Climate data from the U.S. provenances were provided by Gerald Rehfeldt of the U.S. Forest Service. Stefan Zeglen of the B.C. Ministry of Forests provided maps, advice, and data on blister rust surveys in B.C., and Yousry x El-Kassaby and Jodie Krakowski provided the data set for the mating system analysis from the Manning and Baldy populations. Funding for this project was provided by the British Columbia Forestry Investment Account through the Forest Genetics Council of B.C. to the Centre for Forest Gene Conservation at UBC and by the Mary and David Macaree Fellowship, the TimberWest Forest Limited Fellowship in Forest Sciences, and the Donald S. McPhee Fellowship. Lastly, but certainly not least, I would like to thank my wife, Dorothy Watson for her support through the last 4-plus years. She helped me in the lab and the field, provided moral support, editing, reviewing, photography, and never seemed to get sick of hearing me talk about trees. I also want to thank my son, Russell, who brings such joy and light to my life and stirs the love in my heart. xi C O - A U T H O R S H I P S T A T E M E N T This thesis has been written in manuscript form, with chapter 5 already published and chapters 2 through 4 to be submitted for publication. For al l components o f this project I took the lead in design o f the research program, experiment establishment, data collection and analysis, and manuscript preparation. M y supervisor, Sally Aitken, helped me to plan the project, provided funding, and reviewed and edited all manuscript chapters, and is a co-author on the research chapters (chapters 2-5). x i i C H A P T E R 1: T H E S I S I N T R O D U C T I O N A N D L I T E R A T U R E R E V I E W 1.1 I N T R O D U C T I O N Whitebark pine (Pinus albicaulis Englem.) is a high-elevation, five-needle pine, found at and near treeline on mountain peaks from central British Columbia south to California, and east to Alberta, Wyoming, and Montana. Throughout its range it is experiencing a decline in health, reproduction, and regeneration. There are several factors contributing to whitebark pine's decline, including successional replacement due to decades of fire suppression and attack by mountain pine beetle (Dendroctonus ponderosae Hopkins). However, the factor that has had the most significant impact on whitebark pine is the introduced disease white pine blister rust (caused by the fungus Cronartium ribicola J.C. Fisch. ex Rabh.). Since its introduction in Vancouver, B.C. in 1910, white pine blister rust has spread throughout the range of whitebark pine, causing widespread mortality and reduced seed production. Concern has been growing regarding the fate of whitebark pine. It has been declared a Species of Concern in the State of Washington, and is being considered for listing under the Canadian Species At Risk Act. It is a keystone species in high-elevation ecosystems, providing a major food source for wildlife. It also aids slope stability and tempers melting of the snowpack in the spring, and has high aesthetic value for high-alpine and wilderness recreationists. Conservation and restoration efforts are needed to halt or reverse the losses caused by blister rust and fire suppression. However, due to its low economic importance, little is known about the genetics whitebark pine, other than its mating system and genetic diversity of neutral molecular markers. Genetic variation and geographic differentiation in quantitative traits and the impacts of inbreeding and blister rust infection on genetic diversity are all unknown. In this thesis, I use a seedling common garden experiment and isozyme analysis of seed and bud tissues to fill some of the gaps in scientific knowledge of the genetics of whitebark pine. I investigate genetic diversity and geographic differentiation of quantitative traits, mating system and 1 inbreeding depression, and the impacts of inbreeding and blister rust infection on genetic diversity. I also use my results to develop guidelines for seed transfer for use in conservation or restoration efforts. The results from this work will be valuable for planning conservation and restoration activities to help ensure that whitebark pine is maintained as an integral part of these high-elevation ecosystems. 1.2 L I T E R A T U R E R E V I E W 1.2.1 Ecology of Whitebark Pine 1.2.1.1 Biology and seed dispersal Whitebark pine (Pinus albicaulis Engelm.) is a high elevation tree species that is distributed from 37° to 55° N . and from 130° to 110° W. (Critchfield and Little 1966). It is restricted to the upper subalpine forest with typical habitats of ridge crests and steep southwestern-facing slopes that experience high winds and shallow snow. Typical habitats are at or near timberline where it grows with Engelmann spruce (Picea engelmannii Parry), subalpine fir (Abies lasiocarpa (Hook.) Nutt.), and mountain hemlock (Tsuga mertensiana (Bong.) Carr.) (Arno and Hoff 1989). Whitebark pine is commonly a pioneer species on disturbed sites, especially recent burns (Tomback 1982). However, whitebark pine is regarded as a persistent serai species, as climax communities are most prevalent rangewide and successional communities appear more geographically restricted (Campbell and Antos, 2003). It is the only North American member of the stone pines (Pinus subsection Cembrae) (Price et al. 1998; but see Gernandt 2005), a group of five species of pines that share the trait of having indehiscent cones (cones that do not open as they ripen). This characteristic is proposed to be an adaptation to seed dispersal by two species of nutcrackers (Nucifragia, family Corvidae) that are mutualists with the stone pines (Lanner 1982, Tomback and Linhart 1990, Lanner 1996). The large, nutritious seeds are an important food source for these birds as well as for squirrels (Tamiasciurus spp.) and grizzly bears (Ursus arctos ssp. horribilis Ord) (Mattson and Reinhart 1994, Mattson et al. 2001). 2 In North America, the Clark's nutcracker (Nucifraga columbiana Wilson) is the primary agent of dispersal of whitebark pine seeds (Hutchins and Lanner 1982, Lanner 1982, Tomback 1982, Linhart and Tomback 1985, Tomback and Linhart 1990, Lanner 1996, Tomback 2001). They harvest ripe seeds from the unopened cones, and bury thousands of seeds in the ground for later consumption (Hutchins and Lanner 1982, Lanner 1982, Tomback 2001). Seed are cached on steep, often windswept, south-facing slopes as well as open, disturbed terrain, particularly recent burns at any distance from a few meters to several hundred meters at subalpine elevations used communally by local nutcracker populations or even greater distances in lower elevation communal storage areas (Hutchins and Lanner 1982, Tomback 1978). Typically one to 15 seeds are stored per cache, with an average of three to five seeds, buried one to three cm deep (Hutchins and Lanner 1982, Tomback 1982). The seed dispersal patterns of nutcrackers influence several aspects of whitebark pine's population structure. These include growth form, stand structure, genetic relationships among individuals within populations and the genetic diversity among populations. They also account for the elevational and geographical occurrence of whitebark pine (Tomback and Schuster 1994, Tomback 2001). Nutcrackers can carry seeds over 20 km (Vander Wall and Balda 1977) and may return to recover and recache previous year's seeds, some of which could have already germinated (Hutchins and Lanner 1982). These seed dispersal behaviors appear to randomize the distribution of related clumps of trees, resulting in relatedness within clumps, but little family structure among clumps. In addition, this results in little genetic structure among populations, as revealed by low levels of population differentiation for genetic markers (Furnier et al. 1987, Bruederle et al. 2001). Whitebark pine has three main growth forms: single stem, multiple-stem clumps, and the shrubby krummholz form (Arno and Hoff 1989). These clumps may be a single individual with multiple forks resulting from damage or environmental influences, or may be multiple distinct individuals, resulting from the simultaneous germination of several seeds (Linhart and Tomback 1985, Tomback and Schuster 1994). It has been shown that in many cases, the clumpy growth habit 3 of whitebark pine results from the caching o f seeds by the nutcracker (Hutchins and Lanner 1982, Tomback and Schuster 1994). Because of the seed foraging behavior o f the nutcracker, it is l ikely that multiple seeds are collected from a single cone and cached together (Hutchins and Lanner 1982, Tomback 1982). A significant proportion of these tree clumps contain multiple individuals which are related as half- or full-sibs (Schuster and Mit ton 1991). The proximity of related individuals means that these clumped stems may be l ikely to experience non-random mating, resulting in a higher proportion of inbred offspring than solitary trees (Krakowski et al. 2003). The origin o f whitebark pine is unknown and it has been suggested that ancestral forms o f Clark 's nutcracker and whitebark pine crossed the Bering Straight land bridge to arrive in North America (Tomback and Linhart 1990). A phylogenetic analysis using chloroplast D N A ( c p D N A ) markers appears to support this theory (Krutovskii et al. 1995). Phylogenetically, whitebark pine is closely related to the four other stone pine species (Krutovskii et al. 1995, Belokon et al. 1998, Liston et al. 1999). It is most closely related to Swiss stone pine (Pinus cembra) and Siberian stone pine (P. sibirica), and a time o f divergence estimated from the genetic distance between these species is from 0.6-1.3 mil l ion years ago. This corresponds wel l to the approximate time of the opening o f the Bering Straight 1.8-3.5 mi l l ion years ago (Krutovskii et al. 1995).. Fossil evidence shows that whitebark pine has been present in the Yellowstone basin for over 100,000 years, and that it survived the last glaciation in glacial refugia throughout much of the Northern U S A Rocky Mountains (Baker 1990). Its current occurrences in the southern part o f its range in the Great Basin and Oregon Cascades are thought to be remnants of a much larger subalpine woodland that existed during glaciation. These glacial refugia were putative sources for the subsequent expansion north o f two distinct c p D N A haplotypes (Richardson et al. 2002). 4 1.2.1.2 Population genetics Genetic variation of whitebark pine has been assessed using molecular markers at scales ranging from a single watershed to most o f its range, but quantitative traits that might reveal local or regional adaptation have not been studied. Several studies have analyzed genetic diversity in neutral molecular markers o f whitebark pine (Yandell 1992, Jorgensen and Hamrick 1997, Bruederle et al. 1998, Stuart-Smith 1998, Richardson et al. 2002, Krakowski et al. 2003). Expected heterozygosities (H e ) reported in these studies were within the range o f other stone pines (mean H e = 0.189, range 0.092-0.257) (Politov et al. 1992, Jorgensen and Hamrick 1997); however, most o f the reported values fall somewhat below the mean for other pines with wind-dispersed seed in section Strobus (0.219) (Bruederle et al. 2001). The amount o f genetic variation due to population differentiation (F S T) reported in these studies ranged from 0.025 to 0.088. A n average o f 5.3% of total genetic variation was among populations, while the vast majority of variation resides within populations. This number is slightly higher than the mean for the four other Cembrae species (F S T = 0.046) (Goncharenko et al. 1993a, 1993b, Krutovskii et al. 1995, Tani et al. 1996, Potenko and Ve l ikov 1998, 2001, Belokon et al. 2005, but see Szmidt 1982 in El-Kassaby 1991) but lower compared to the mean for nine species o f pine with wind-dispersed seed in section Strobus ( F S j = 0.086) (Bruederle et al. 1998). Overall, this is typical of pines, which usually have more than 90% of their genetic diversity within populations (Ledig 1998). This indicates that most o f the variation in whitebark pine is within and among individuals within populations, and that populations of bird-dispersed pines are not strongly differentiated for putatively selectively neutral molecular markers. 5 1.2.13 Ecological threats Whitebark pine has become an ecological symbol of the effects of altered fire regimes and the introduction of exotics to western North America (Tomback et al. 2001). It has become seriously threatened in many areas throughout its range due to successional replacement resulting from fire suppression, mountain pine beetle (Dendroctonus ponderosae), and the impact of white pine blister rust, an introduced disease caused by the fungus Cronartium ribicola (Campbell and Antos 2000). White pine blister rust affects five-needle pines in the section Strobus, and whitebark pine is more susceptible than most North American 5-needle pines to this disease (Bedwell and Childs 1943, Bingham 1972, Hoff et al. 1980). Evidence of blister rust resistance has been found in whitebark pine (Hoff et al. 1980), but to date, the most powerful resistance mechanism described in white pines, major gene resistance (MGR), has not been detected in this species. MGR is a single dominant gene conferring immunity to white pine blister rust, and has been detected in other white pine (Strobus) species including western white pine (P. monticola Dougl. ex D. Don), sugar pine (P. lambertiana Dougl.), limber pine (P.flexilis James) and southwestern white pine (P. strobiformis Engelm.) (Kinloch and Littlefield 1977, Kinloch et al. 1999, Kinloch and Dupper 2002). Blister rust is present throughout the range of whitebark pine in British Columbia (Campbell and Antos 2000, Zeglen 2002), and although levels of infection tend to be highest in the south and west of the province, areas of high or low infection can be found throughout the range of occurrence of the disease, depending on local environmental conditions. Scientists are in unanimous agreement that restoration and conservation efforts are needed to stop or reverse the effects of blister rust and fire exclusion (McCool and Freimund 2001). Silvicultural techniques such as prescribed burning or clearing openings can be used to encourage seed caching by nutcrackers and thus promote natural regeneration, but in stands with an inadequate natural seed source or those that need to be regenerated quickly, planting seedlings is the suggested restoration practice (Hoff et al. 2001). 6 1.2.2 Ecological Influences on Plant Population Structure Plant populations have both a spatial and temporal genetic structure that may be manifested among geographically diverse populations, within local groups, or even in the progeny of individuals. Ecological factors that influence reproduction and dispersal are likely to be particularly important in determining the type of genetic structure maintained by a species, both within and among populations (Loveless and Hamrick 1984). The genetic structure of whitebark pine is influenced by the potential for non-random mating due to both its growth habit, and the dispersal of its seed by birds. Natural selection generates population genetic structure along with the selectively neutral forces of migration, drift, mutation and recombination (reviewed by Charlesworth et al. 2003). A meta-analysis by Gerber and Griffen (2003) of the roles of inheritance and selection on functional traits in non-woody plants found that direct selection is the primary force in functional trait evolution and population ' differentiation, although indirect selection through correlated traits accounts for a substantial portion of total selection, and often appears to reinforce direct selection. Selection on vegetative performance (plant size and growth) was stronger than on phenology, physiology, and morphology, but physiological traits had the highest heritabilities. The magnitude of selection depended on the fitness measure assessed (vegetative fitness, fertility, or cumulative fitness) and in general was negatively correlated with heritability (Geber and Griffen 2003). Several reviews by Hamrick and others have examined a number of life history traits and ecological factors and how they are related to genetic structure (Loveless and Hamrick 1984, Hamrick et al. 1991, Hamrick et al. 1992). The factors assessed included taxonomic status, regional distribution, breeding system, life cycle, reproductive morphology, mode of reproduction, pollination mechanism, seed dispersal system, phenology, life cycle, timing of reproduction, successional stage and geographic range. These reviews collectively showed that genetic diversity, as estimated by expected heterozygosity, and population differentiation (G S T) are inversely related. Low values of G S T and high diversity both within species (Hes) and populations (Hep) were found in long-lived, late successional, woody, wide spread, outcrossing, temperate and tropical species with wind-dispersed seed. The reverse was true for annual, early 7 successional, endemic, selling or mixed mating, animal dispersed, boreal-temperate species. However, there is a lot of variation among species within these life history types, and these factors only account for approximately 25% of the genetic variation among species (He) (Loveless and Hamrick 1984, Hamrick et al. 1991, Hamrick et al. 1992). Whitebark pine presents an interesting intermediate between these groups. It is monoecious and predominantly outcrossing, wind pollinated and long lived, which are factors usually associated with low G S T and high diversity, but is also subject to selfing or biparental inbreeding, is commonly a pioneer species on disturbed sites, and its seeds are animal dispersed, which are characteristics usually associated with higher G S T and lower diversity (Lanner 1982, Arno and Hoff 1989, Tomback and Linhart 1990, Tomback et al. 1993, Tomback 2001, Krakowski et al. 2003). 1.2.3 Quantitative Trait Variation in Forest Trees Patterns of variation across distinct geographic units are widely recognized and well documented in plants. In their review, Linhart and Grant (1996) reference a number of the early works that documented this. The term genecology was coined to describe this concept by Turreson (1922), and over time, the adaptation of plants to their environment has received attention in four main areas; 1) adaptation to microhabitats, 2) selection for life history traits, 3) selection acting at various stages of the life cycle (demographic genetics), and 4) physiological adaptations to specific environments (reviewed by Hamrick 1982). Quantitative genetic variation has been described for a number of temperate and boreal forest tree species at the scales ranging from a single island or watershed, to several U.S. states or Canadian provinces (Morgenstern 1996). These studies, as well as countless others, were based on seedlings from different provenances that covered the breadth of the geographic unit of their study grown in a common garden to assess differences in quantitative traits; Traits studied have included growth, phenology, cold hardiness, net photosynthesis, transpiration rate, stomatal conductance, vulnerability to cavitation and water use efficiency, and quantitative characters such as cotyledon number, germination percent, days to germination, seed 8 weight, height, diameter, and root and shoot biomass. These traits were compared to geographic variables (i.e., latitude, longitude, elevation, slope, aspect, and distance from the coast or Cascade crest) and modeled climatic variables (e.g., seasonal and annual temperature, precipitation, aridity, and frost free period) describing parent tree environments to determine the amount of genetic variation among populations potentially reflecting local adaptation. Environmental factors that are geographically variable within the range of a species can result in regional differentiation in gene frequencies. These factors are most often found in gradients, and natural selection along these gradients can lead to clinal variation (Morgenstern 1996). Species have an ecological tolerance and within that a preference for a particular zone along the environmental gradient, with maximum abundance for a given population found at some intermediate optimum (Endler 1977). In species that show significant geographic differentiation along environmental gradients, elevation is often associated with genetic differentiation, along with latitude, and distance from a major geographic feature, such as the Pacific coast or the Cascade crest (Campbell 1979, Rehfeldt et al. 1984, Campbell 1986, Rehfeldt 1989, 1991, 1993, 1994a, 1994b, Mitton 1995, Benowicz et al. 2001, Bennuah et al. 2004, Wei et al. 2004, St. Clair et al. 2005). Clinal variation is common in forest trees (Morgenstern 1996). Growth and phenological traits are usually the most closely correlated with climatic factors (Matyas 1996). In the past, genecological studies have related quantitative traits to geographic variables, which were used to represent the complex three-dimensional environmental gradients associated with clines in these traits. Sophisticated climate models are now available which can be used to estimate a suite of climatic variables describing local environments (Rehfeldt 1995, Rehfeldt et al. 1999b, Hamann and Wang 2005, Wang et al. 2006). Recently, genecological studies of forest trees have moved towards directly using climatic rather than geographic location data allowing variation in quantitative trait's to be studied relative to the environmental gradients driving local adaptation (Rehfeldt 1995, van Niejenhuis and Parker 1996, Rehfeldt et al. 1999a, Rehfeldt et al. 1999b, Benowicz et al. 2000, Hamann et al. 2000, Chuine et al. 2001, Guy and Holowachuk 2001, 9 Leinonen and Hanninen 2002, O'Neill et al. 2002, Beaulieu et al. 2004, O'Neill and Aitken 2004, Wei et al. 2004, St. Clair et al. 2005, Wang et al. in press). Most studies of temperate and boreal species show mean annual temperature to be one of the most important factors influencing local adaptation. However, other climatic variables (e.g. frost free period, number of frost free days, extreme hot or cold temperature, difference between summer and winter temperature (continentality), annual and summer precipitation and aridity) have been shown to be related to a number of quantitative traits in various species of forest trees, including growth, biomass, phenology, survival, number of resin canals, and stable carbon isotope ratio (8 I3C). 1.2.4 Inbreeding 1.2.4.1 Genetic basis of inbreeding The reduced fitness that typically accompanies an increase in homozygosity observed in progeny from matings among relatives is known as inbreeding depression. Empirical evidence of inbreeding depression is abundant (Charlesworth and Charlesworth 1987, Crnokrak and Roff 1999), and the genetic basis of inbreeding depression has been studied, reviewed and discussed extensively, with the two competing explanations referred to as the dominance (or partial dominance) and overdominance hypotheses (Charlesworth and Charlesworth 1987, Williams and Savolainen 1996, Byers and Waller 1999, Charlesworth and Charlesworth 1999, Carr and Dudash 2003). Inbreeding depression is generally thought to be the result of increased homozygosity of recessive or partially recessive deleterious alleles, which are masked or partially masked by dominant alleles in more heterozygous outbreds (the dominance hypothesis) (Bijlsma et al. 1999, Charlesworth and Charlesworth 1999, Carr and Dudash 2003). The overdominance hypothesis states that heterozygotes at a given locus have an advantage over homozygotes and the loss of heterozygosity in inbred progeny results in inbreeding depression. However, associative overdominance, which occurs when there is linkage disequilibrium between the neutral locus studied and two other loci that are under selection (Pamilo and Palsson 1998), can look like true overdominance. Little empirical evidence has 10 been found to support the overdominance hypothesis that heterozygosity per se is advantageous, and that heterozygotes have higher fitness relative to the homozygotes produced by inbreeding (Strauss and Libby 1987, Houle 1989, Charlesworth and Charlesworth-1999, Roff 2002). 1.2.4.2 Purging The dominance hypothesis predicts that inbreeding and selection should work together to reduce a population's mutational load (Charlesworth et al. 1990). Inbreeding increases the homozygosity of deleterious recessive alleles, increasing their phenotypic expression, making selection more efficient at removing or "purging" them from the population. Therefore, populations with a history of extensive inbreeding are expected to have lower mutational load, lower inbreeding depression, and higher mean fitness than ancestral populations (Carr and Dudash 2003). Under the overdominance hypothesis, when inbred lines are crossed, it is predicted that trait means will exceed those of outbred populations. In contrast, the dominance hypothesis predicts that trait means will only be restored to the level of outbred populations following crossing of inbred lines (Roff 2002). Evidence of purging, however, is inconclusive and can depend on how purging is evaluated (Crnokrak and Barrett 2002). Crnokrak and Barrett (2002) found substantial evidence of purging in a review including plant, insect, and mammal taxa, as did Husband and Schemske (1996) in natural plant populations. In contrast, Byers and Waller (1999) found varying evidence of purging in plants among families, populations and species. While some studies cite a lack of inbreeding depression as evidence of purging, fixation of deleterious alleles through drift can produce the same effect as selective removal of deleterious alleles. Without genetic variation in a trait, there can be no inbreeding depression (Carr and Dudash 2003). 11 1.2.4.3 Plant mating systems Inbreeding depression has likely played a significant role in the evolution of plant mating systems. Most models for the evolution of mating systems suggest only complete selling or complete outcrossing are evolutionarily stable states: outcrossing when inbreeding depression is strong (5=1-[fitness of selfed progeny/fitness of outcrossed progeny] > 0.5) and selfing when inbreeding depression is weak (5 < 0.5) (Lande and Schemske 1985, Charlesworth and Charlesworth 1987, Husband and Schemske 1996, Byers and Waller 1999). However, several models have predicted that intermediate selling rates may be stable depending on a number of parameters such as efficiency of reproduction, seed dispersal, and variation in selfing rate among generations, variation in inbreeding depression, and biparental inbreeding (Lloyd 1979, Holsinger 1986, Uyenoyama 1986, Holsinger 1991, Cheptou and Mathias 2001). Using a genetic model, Lande and Schemske (1985) predicted a bimodal distribution of plant mating systems with predominantly selfing and predominantly outcrossing as alternative stable states. Theory predicts that inbreeding depression will decrease with inbreeding as recessive deleterious alleles are expressed and purged through selection. Husband and Schemske (1996) found that empirical evidence supports the theory that selfing reduces inbreeding depression. In their review, they found that inbreeding depression in predominantly selfing species was significantly less than in predominantly outcrossing species. Selling is generally associated with annual plants, and outcrossing with perennials, but mean fitness and inbreeding depression are affected by longevity, and when mutation rates are high, fitness and inbreeding depression of perennials can approach those of annuals (Morgan 2001). Morgan (2001) suggests that accumulation of deleterious recessive mutations may partly explain the extraordinary levels of inbreeding depression found in forest trees, even at intermediate rates of self fertilization. Reviews of the distribution of plant mating systems have generally found evidence to support models that predict species to be either strongly outcrossing or selfing. In a review of 55 species of plants, Schemske and Lande (1985) present data that support this hypothesis. However, in using data from 345 species, Goodwillie et al. (2005) found a significantly different distribution with fewer predominantly selfing 12 taxa and no bimodality. In addition, they found a higher proportion of species with mixed mating systems (0.2 <t< 0.8) than Schemske and Lande (1985), although the difference was not significant. However, when pollination biology is considered, thedistribution of selfing rates in wind-pollinated species is indeed bimodal, but the distribution for animal-pollinated species shows that intermediate levels of outcrossing are common (Vogler and Kalisz 2001). The distributions of outcrossing rates differ significantly, and animal-pollinated taxa are almost twice as likely to experience mixed mating than wind pollinated taxa (Goodwillie et al. 2005). 1.2.5 Heterozygosity-fitness correlation The relationship, or lack thereof, between heterozygosity and fitness has been debated extensively and the literature on this topic is voluminous (see references in Britten 1996, David 1998). It has been shown in several conifers that inbred individuals are removed from cohorts through selection with age, and many populations of mature conifers exhibit an excess of heterozygotes (Mitton and Jeffers 1989). Mitton (1993) suggests that selection against inbred homozygotes would bring a population into Hardy-Weinberg equilibrium, but an excess of heterozygotes should only be the result of selection favoring heterozygotes in the outbred population. However, a fitness increase due to heterozygosity per se would imply overdominance where a heterozygote is fitter, on average, than either homozygote, although associative overdominance could also produce this result (Houle 1994). Most relationships between genomic heterozygosity and apparent superiority of heterozygotes may be just a measure of inbreeding depression due to recessive deleterious alleles in homozygotes resulting from inbreeding (Ledig et al. 1983, Savolainen and Hedrick 1995). Associations of fitness and heterozygosity are likely to be observed only when there are high levels of inbreeding, very small population sizes, or extreme population substructuring (Savolainen and Hedrick 1995). 13 While there is a large body of literature on the positive association between heterozygosity at putatively neutral loci and fitness components in conifers (see Table 1.1), much of this work has been questioned as to how accurately the loci used cumulatively represented genomic heterozygosity (Pamilo and Palsson 1998, Slate and Pemberton 2002, Balloux et al. 2004, Pemberton 2004, Slate et al. 2004). Associative overdominance has been cited as a more plausible explanation for the associations observed in empirical studies. Computer simulations of the power of associative overdominance in creating a correlation between individual heterozygosity and fitness in small populations showed that the association is dependent on both population size and the dominance coefficient. The association was stronger in small populations (n < 100) and when dominance was small (h = 0.01) so that deleterious mutations were masked in heterozygotes (Pamilo and Palsson 1998). A meta-analysis on results from both plant and animal studies reported a weak correlation (r = 0.133, p <0.10) between heterozygosity and growth rate for plants and animals combined, and r = 0.489 (p<0.01) for plants alone (Britten 1996). The high variability in correlations among taxonomic groups, and among species within taxa, indicates that this association many not be a general phenomenon, and strong evidence for a relationship between heterozygosity and fitness is clearly lacking. 1.3 THESIS I N T R O D U C T I O N Whitebark pine is experiencing drastic declines throughout its range due to the combined effects of infection and mortality due to white pine blister rust, successional replacement due to fire suppression, and attack by mountain pine beetle. Concern about potential local extirpation has lead to increased efforts of conservation and restoration. While several previous studies have investigated neutral genetic variation, information on genetic variation in quantitative traits, which is needed to guide these efforts, has been lacking. 14 In the following chapters, I investigate the ecological genetics, mating system, inbreeding depression, and the effects of inbreeding and white pine blister rust on genetic diversity of whitebark pine. In CHAPTER 2,1 use a common garden experiment to determine genetic diversity and geographic differentiation of quantitative traits, and the climatic factors influencing local adaptation, and use these data to determine seed transfer guidelines. In CHAPTER 3,1 use isozyme analysis of seed tissues to analyze the mating system and determine the presence and magnitude of inbreeding depression. In CHAPTER 4, fuse isozyme analysis of bud tissue to investigate how genetic diversity changes with cohort age, reflecting the effects of inbreeding, and with level of blister rust infection on different sites. In CHAPTER 5,1 employ artificial freeze testing to determine seasonal variation and geographic differentiation of cold hardiness of seedlings in the common garden. My major findings and future research questions are summarized and discussed in CHAPTER 6 15 Table 1.1. Reported relationships between individual heterozygosity and adaptive traits for several conifer species. Species Trait Association Reference Pinus rigida Diameter growth Positive Lediget al. 1983 Pinus taeda Survival Positive Bush and Smouse 1991 a Fecundity Positive tt ii Height growth Positive tt ii Diameter growth Positive tt Picea engelmannii Seedling growth Positive Mittonand Jeffers 1989 Pinus ponderosa Seedling growth Positive Mitton and Grant 1984 a Growth variability Positive Knowles and Grant 1981 Pinus contorta a Positive Knowles and Mitton 1980 Pinus attenuata a No correlation Strauss 1987 Pinus ponderosa Growth No correlation Linhart and Mitton 1985 Pinus sylvestris Growth No correlation Savolainen and Hedrick 1995 Reproduction No correlation Pinus radiata Growth Little correlation Strauss and Libby 1987 Pseudotsuga Growth rate No correlation* Bongarten et al. 1985 menziesii *with family mean heterozygosity 16 1.4 REFERENCES Arno, S. F., and R. J. Hoff. 1989. Silvics of whitebark pine (Pinus albicaulis), Gen. Tech. Rep. INT-GTR-253. U.S. Dept. of Agriculture, Forest Service, Intermountain Research Station, Ogden, UT. Baker, R. G. 1990. Late quaternary history of whitebark pine in the Rocky Mountains, in W. C. Schmidt and K. J. McDonald, editors. Proceedings - Symposium on whitebark pine ecosystems: ecology and management of a high-mountain resource; 1989 March 29-31; Bozeman MT; Gen. Tech. Rep INT-GTR-270. U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Bozeman, MT. Balloux, F., W. Amos, and T. Coulson. 2004. Does heterozygosity estimate inbreeding in real populations. Mol. Ecol. 13:3021-3031. Beaulieu, J., M . Perron, and J. Bousquet. 2004. Multivariate patterns of adaptive genetic variation and seed source transfer in Picea mariana. Can. J. For. Res. 34:531-545. Bedwell, J. L., and T. W. Childs. 1943. Susceptibility of whitebark pine to blister rust in the Pacific Northwest. J. F. 41:904-912. Belokon, M . M. , Y. S. Belokon, D. V. Politov, and Y. P. Altukhov. 2005. Allozyme polymorphism of Swiss stone pine Pinus cembra L. in mountain populations of the Alps and eastern Carpathians. Russ. J. of Genet. 41:1268-1280. Belokon, M . M. , D. V. Politov, Y. S. Belokon, K. V. Krutovskii, O. P. Malyuchenko, and Y. P. Altukhov. 1998. Genetic differentiation in white pines of section strobus: Isozyme analysis data. Doklady Akademii Nauk 358:699-702. Bennuah, S. Y., T. Wang, and S. N. Aitken. 2004. Genetic analysis of the Picea sitchensis x glauca introgression zone in British Columbia. Forest Ecol. Manag. 197:65-77'. Benowicz, A., R. D. Guy, and Y. A. El-Kassaby. 2000. Geographic pattern of genetic variation in photosynthetic capacity and growth in two hardwood species from British Columbia. Oecologia 123:168-174. Benowicz, A., S. J. L'Hirondelle, and Y. A. El-Kassaby. 2001. Patterns of genetic variation in mountain hemlock (Tsuga mertensiana (Bong.) Carr.) with respect to height growth and frost hardiness. Forest Ecol.Manag. 154:23-33. Bijlsma, R., J. Bundgaard, and W. F. Van Putten. 1999. Environmental dependence of inbreeding depression and purging in Drosophila melanogaster. J. Evolution Biol. 12:1125-1137. Bingham, R. T. 1972. Taxonomy, crossability, and relative blister rust resistance of 5-needled white pine. Pages 271-280 in R. T. Bingham, R. J. Hoff, and G. I. McDonald, editors. Biology of rust resistance in forest trees. Misc. Publ. 1221. U.S. Department of Agriculture, Washington, D.C. Bongarten, B. C , N . C. Wheeler, and K. S. Jech. 1985. Isozyme heterozygosity as a selection criterion for yield improvement in Douglas-fir. Pages 121-128 in Proceedings of the Canadian Tree Improvement Association. Britten, H. B. 1996. Meta-analyses of the association between multilocus heterozygosity and fitness. Evolution 50:2158-2164. Bruederle, L. P., D. P. Rogers, K. V. Krutovskii, and D. V. Politov. 2001. Population genetics and evolutionary implications. Pages 137-158 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. Bruederle, L. P., D. F. Tomback, K. K. Kelly, and R. C. Hardwick. 1998. Population genetic structure in a bird-dispersed pine, Pinus albicaulis (Pinaceae). Canadian Journal of Botany 76:83-90. Bush, R. M. , and P. E. Smouse. 1991. The impact of electrophoretic genotype on life-history traits in Pinus taeda. Evolution 45:481-498. Byers, D. L., and D. M . Waller. 1999. Do plant populations purge their genetic load? Effects of population size and mating history on inbreeding depression. Annu. Rev. Ecol. Syst. 30:479-513. Campbell, E. M. , and J. A. Antos. 2000. Distribution and severity of white pine blister rust and mountain pine beetle on whitebark pine in British Columbia. Can. J. For. Res. 30:1051-1059. 17 Campbell, E. M. , and J. A. Antos. 2003. Postfire succession in Pinus albicaulis - Abies lasiocarpa forests of southern British Columbia. Can. J. Bot. 81:383-397. Campbell, R. K. 1979. Genecology of Douglas-fir in a watershed in the Oregon cascades. Ecology 60:1036-1050. Campbell, R. K. 1986. Mapped genetic variation of Douglas-fir to guide seed transfer in southwest Oregon. Silvae Genet. 35:85-96. Carr, D. E., and M . R. Dudash. 2003. Recent approaches into the genetic basis of inbreeding depression in plants. Phil. Trans. R. Soc. Lond. B 358:1074-1084. Charlesworth, B., and D. Charlesworth. 1999. The genetic basis of inbreeding depression. Genet. Res. 74:329-340. Charlesworth, B., D. Charlesworth, and N . H. Barton. 2003. The effects of genetic and geographic structure on neutral variation. Annu. Rev. Ecol. Syst. 34:99-125. Charlesworth, B., D. Charlesworth, and M . T. Morgan. 1990. Genetic loads and estimates of mutation rates in highly inbred plant populations. Nature 347:380-382. Charlesworth, D., and B. Charlesworth. 1987. Inbreeding depression and its evolutionary consequences. Annu. Rev. Ecol. Syst. 18:237-268. Cheptou, P. O., and A. Mathias. 2001. Can varying inbreeding depression select for intermediary selfing rates? Am. Nat. 157:361-373. Chuine, I., S. N . Aitken, and C. C. Ying. 2001. Temperature thresholds of shoot elongation in provenances of Pinus contorta. Can. J. For. Res. 31:1444-1455. Critchfield, W. B., and E. L. Little, Jr. 1966. Geographic distribution of the pines of the world. USDA Forest Service, Washington, D.C. Miscellaneous publication 991. Crnokrak, P., and S. C. H. Barrett. 2002. Purging the genetic load: a review of the experimental evidence. Evolution 56:2347-2358. Crnokrak, P., and D. A. Roff. 1999. Inbreeding depression in the wild. Heredity 83:260-270. David, P. 1998. Heterozygosity-fitness correlations: new perspectives on old problems. Heredity 80:531-537. Endler, J. A. 1977. Geographic variation, speciation, and clines. Princeton University Press, Princeton, NJ. Furnier, G. R., P. Knowles, M . A. Clyde, and B. P. Dancik. 1987. Effects of avian seed dispersal on the genetic structure of whitebark pine populations. Evolution 41:607-612. Geber, M . A., and L. R. Griffen. 2003. Inheritance and natural selection on functional traits. Int. J. Plant Sci. 164:S21-S42. Gernandt, D. S., G. Geada Lopez, S. Ortiz Garcia, and A. Liston. 2005. Phylogeny and classification of Pinus. Taxon 54:29-42. Goncharenko, G. G., V. E. Padutov, and A. E. Silin. 1993a. Allozyme variation in natural populations of Eurasian pines. 2. Genetic variation, diversity, differentiation, and gene flow in Pinus sibirica DuTour in some lowland and mountain populations. Silvae Genet. 42:246-253. Goncharenko, G. G., V. E. Padutov, and A. E. Silin. 1993b. Allozyme variation in natural populations of Eurasian pines. 1. Population structure, genetic variation, and differentiation in Pinus pumila (Pall) Regel from Chukotsk and Sakhalin. Silvae Genet. 42:237-246. Goodwillie, C , S. Kalisz, and E. C. G. 2005. The evolutionary enigma of mixed mating systems in plants: occurrence, theoretical explanations, and empirical evidence. Annu. Rev. Ecol. Syst. 36:47-79. Guy, R. D., and D. L. Holowachuk. 2001. Population differences in stable carbon isotope ratio of Pinus contorta Dougl. ex Loud.: relationship to environment, climate of origin, and growth potential. Can. J. Bot. 79:274-283. Hamann, A., M . P. Koshy, G. Namkoong, and C. C. Ying. 2000. Genotype x environment interactions in Alnus rubra: developing seed zones and seed transfer guidelines with spatial statistics and GIS. Forest Ecol. Manag. 136:107-119. 18 Hamann, A., and T. Wang. 2005. Models of climatic normals for genecology and climate change studies in British Columbia. Ag. Forest Meteorol.:211-221. Hamrick, J. L. 1982. Plant population genetics and evolution. Am. J. Bot. 69:1685-1693. 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. Pages 283 in D. A. Falk and K. E. Holsinger, editors. Genetics and conservation of rare plants. Oxford University Press, New York. Jf*^  Hamrick, J. L., M . J. W.podt, and S. L. Sherman-Broyles. 1992. Factors influencing levels of genetic diversity in woody'plant species. Pages 420 in W. T. Adams, S. H. Strauss, D. L. Copes, and A. R. Griffin, editors. Population genetics of forest trees. Kluwer Academic Publishers, Dordrecht, The Netherlands. Hoff, R. J., R. T. Bingham, and G. I. McDonald. 1980. Relative blister rust resistance of white pines. Eur. J. Forest Pathol. 10:307-316. Hoff, R. J., D. E. Ferguson, G. I. McDonald, and R. E. Keane. 2001. Strategies for managing whitebark pine in the presence of white pine blister rust. Pages 346-366 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Holsinger, K. E. 1986. Dispersal and plant mating systems: the evolution of self-fertilization in subdivided populations. Evolution 40:405-413. Holsinger, K. E. 1991. Mass-action,models of plant mating systems: the evolutionary stability of mixed mating systems. Am. Nat. 138:606-622. Houle, D. 1989. Allozyme-associated heterosis in Drosophila melanogaster. Genetics 123:789-801. Houle, D. 1994. Adaptive distance and the genetic basis of heterosis. Evolution 48:1994. Husband, B. C , and D. W. Schemske. 1996. Evolution of the magnitude and timing of inbreeding depression in plants. Evolution 50:54-70. ' Hutchins, H. E., and R. M . Lanner. 1982. The central role of Clark's Nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55:192-201. Jorgensen, S. M. , and J. L. Hamrick. 1997. Biogeography and population genetics of whitebark pine, Pinus albicaulis. Can. J. For. Res. 27:1574-1585. Kinloch, B. B., and G. E. Dupper. 2002. Genetic specificity in the white pine-blister rust pathosystem. Phytopathology 92:278-280. Kinloch, B. B., and J. L. Littlefield. 1977. White pine blister rust: hypersensitive resistance in sugar pine. Can. J. Bot. 55:1148-1155. Kinloch, B. B., R. A. Sniezko, G. D. Barnes, and T. E. Greathouse. 1999. A major gene for resistance to white pine blister rust in western white pine from the western Cascade range. Phytopathology 89:861-867. Knowles, P., and M . C. Grant. 1981. Genetic patterns associated with growth variability in ponderosa pine. Am. J. Bot. 68:942-946. Knowles, P., and J. B. Mitton. 1980. Genetic heterozygosity and radial growth variability in Pinus contorta. Silvae Genet. 29:114-118. Krakowski, J., S. N . Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. Krutovskii, K. V., D. V. Politov, and Y. P. Altukhov. 1995. Isozyme study of population genetic structure, mating system and phylogenetic relationships of the five stone pine species (subsection cembrae, section strobi, subgenus strobus). Pages 279-304 in P. Baradat, W. T. Adams, and M.-S. G., editors. Population genetics and genetic conservation of forest trees. SPB Academic Publishing, Amsterdam, The Netherlands. Lande, R., and D. W. Schemske. 1985. The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39:24-40. Lanner, R. M . 1982. Adaptations of whitebark pine for seed dispersal by Clark's Nutcracker. Can. J. For. Res. 12:391-402. 19 Lanner, R. M . 1996. Made for each other: a symbiosis of birds and pines. Oxford University Press. Ledig, F. T. 1998. Genetic variation in Pinus. Pages 251-280 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Ledig, F. T., R. P. Guries, and B. Bonefeld. 1983. The relation of growth to heterozygosity in pitch pine. Evolution 37:1227-1238. Leinonen, I., and H. Hanninen. 2002. Adaptation of the timing of bud burst of Norway spruce to temperate and boreal climates. Silva Fenn. 36:695-701. Linhart, Y. B., and M . C. Grant. 1996. Evolutionary significance of local genetic differentiation in plants. Annu. Rev. Ecol. Syst. 27:237-277. Linhart, Y. B., and J. B. Mitton. 1985. Relationships among reproduction, growth rates, and protein heterozygosity in ponderosa pine. Am. J. Bot. 72:181-184. Linhart, Y. B., and D. F. Tomback. 1985. Seed dispersal by nutcrackers causes multi-trunk growth form in pines. Oecologia 67:107-110. Liston, A., W. A. Robinson, D. Pinero, and E. R. Alvarez-Buylla. 1999. Phylogenetics of Pinus (Pinaceae) based on nuclear ribosomal DNA internal transcribed spacer region sequences. Mol. Phylogenet. Evol. 11:95-109. Lloyd, D. G. 1979. Some reproductive factors affecting the selection of self-fertilization in plants. Am. Nat. 113:67-79. Loveless, M . D., and J. L. Hamrick. 1984. Ecological determinants of genetic structure in plant population. Annu. Rev. Ecol. Syst. 15:65-95. Mattson, D. J., K. C. Kendall, and D. P. Reinhart. 2001. Whitebark pine, grizzly bears, and red squirrels. Pages 121-136 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark Pine Communities: ecology and restoration. Island Press, Washington, D.C. Mattson, D. J., and D. P. Reinhart. 1994. Bear use of whitebark pine seeds in North America. Pages 321 in W. C. Schmidt and F.-K. Holtmeier, editors. International Workshop on Subalpine Stone Pines and Their Environment: The Status of Our Knowledge; 1992 September 5-11; St. Moritz, Switzerland; Gen. Tech. Rep. INT-GTR-309. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden, UT. Matyas, C. 1996. Climatic adaptation of trees: rediscovering provenance tests. Euphytica 92:45-54. McCool, S. F., and W. A. Freimund. 2001. Threatened landscapes and fragile experiences: conflict in whitebark pine restoration. Pages 263-284 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. Mitton, J. B. 1993. Theory and data pertinent to the relationship between heterozygosity and fitness. Pages 575 in N . Wilmsen Thornhill, editor. The natural history of inbreeding and outbreeding. University of Chicago Press, Chicago. Mitton, J. B. 1995. Genetics and physiological ecology of conifers. Pages 1-36 in W. K. Smith and T. M . Hinckley, editors. Ecophysiology of coniferous forests. Academic Press, San Diego. Mitton, J. B., and M . C. Grant. 1984. Associations among protein.heterozygosity, growth rate, and developmental homeostasis. Annu. Rev. Ecol. Syst. 15:479-499. Mitton, J. B., and R. M . Jeffers. 1989. The genetic consequences of mass selection for growth rate in Engelmann spruce. Silvae Genet. 38:6-12. Morgan, M . T. 2001. Consequences of life history for inbreeding depression and mating system evolution in plants. P. Roy. Soc. Lond. B Bio. 268:1817-1824. Morgenstern, K. E. 1996. Geographic variation in forest trees. UBC Press, Vancouver, B.C. O'Neill, G. A., and S. N . Aitken. 2004. Area-based breeding zones to minimize maladaptation. Can. J. For. Res. 34:695-704. O'Neill, G. A., S. N . Aitken, J. N . King, and R. I. Alfaro. 2002. Geographic variation in resin canal defenses in seedlings from the Sitka spruce x white spruce introgression zone. Can. J. For. Res. 32:390-400. Pamilo, P., and S. Palsson. 1998. Associative overdominance, heterozygosity and fitness. Heredity 81:381-389. 20 Pemberton, J. M . 2004. Measuring inbreeding depression in the wild: the old ways are the best. Trends Ecol. Evol. 19:613-615. Politov, D. V., K. V. Krutovskii, and K. V. Altukhov. 1992. Isozyme loci characteristics of gene banks of populations of cedar pines. Genetika 28:76-95. Potenko, V. V., and A. V. Velikov. 1998. Genetic diversity and differentiation of natural populations of Pinus koraiensis (Sieb. et Zucc.) in Russia. Silvae Genet. 47:202-208. Potenko, V. V., and A. V. Velikov. 2001. Allozyme variation and mating system of coastal populations of Pinus koraiensis Sieb. et Zucc in Russia. Silvae Genet. 50:117-122. Price, R. A., A. Liston, and S. H. Strauss. 1998. Phylogeny and systematics of Pinus. Pages 527 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Rehfeldt, G. E. 1989. Genetic variances and covariances in freezing tolerance of lodgepole pine during early winter acclimation. Silvae Genet. 38:133-137. Rehfeldt, G. E. 1991. A model of genetic variation for Pinus ponderosa in the Inland Northwest (U.S.A.): applications in gene resource management. Can. J. For. Res. 21:1491 -1500. Rehfeldt, G. E. 1993. Genetic variation in the ponderosae of the southwest. Am. J. Bot. 80:330-343. Rehfeldt, G. E. 1994a. Adaptation of Picea engelmannii populations to the heterogeneous environments of the Intermountain West. Can. J. Bot. 72:1197-1208. Rehfeldt, G. E. 1994b. Genetic structure of western red cedar populations in the Interior West. Can. J. For. Res. 24:670-680. Rehfeldt, G. E. 1995. Genetic variation, climate models and the ecological genetics of Larix occidentalis. Forest Ecol. Manag. 78:21-37. Rehfeldt, G. E., R. J. Hoff, and R. J. Steinhoff. 1984. Geographic patterns of genetic variation in Pinus monticola. Bot. Gaz. 145:229-239. Rehfeldt, G. E., N . M . Tchebakova, and L. K. Barnhardt. 1999a. Efficacy of climate transfer functions: Introduction of Eurasian populations of Larix into Alberta. Can. J. For. Res. 29:1660-1668. Rehfeldt, G. E., C. C. Ying, D. L. Spittlehouse, and D. A. Hamilton. 1999b. Genetic responses to climate in Pinus contorta: niche breadth, climate change, and reforestation. Ecol. Monogr. 69:375-407. Richardson, B. A., S. J. Brunsfeld, and N . B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Mol. Ecol. 11:215-227. Roff, D. A. 2002. Inbreeding depression: tests of the overdominance and partial dominance hypotheses. Evolution 56:768-775. Savolainen, O., and P. W. Hedrick. 1995. Heterozygosity and fitness - no association in Scots pine. Evolution 140:755-766. Schemske, D. W., and R. Lande. 1985. The evolution of self-fertilization and inbreeding depression in plants. II. Empirical observations. Evolution 39:41-52. Schuster, W. S. F., and J. B. Mitton. 1991. Relatedness within clusters of a bird-dispersed pine and the potential for kin interactions. Heredity 67:41-48. Slate, J., P. David, K. G. Dodds, B. A. Veevliet, B. C. Glass, T. E. Broad, and J. C. McEwan. 2004. Understanding the relationship between the inbreeding coefficient and multilocus heterozygosity: theoretical expectations and empirical data. Heredity 93:255-265. Slate, J., and J. M . Pemberton. 2002. Comparing molecular measures for detecting inbreeding depression. J. Evolution. Biol. 15:20-31. St. Clair, J. B., N . L. Mandel, and K. W. Vance-Borland. 2005. Genecology of Douglas fir in western Oregon and Washington. Ann. Bot.-London 96:1199-1214. Strauss, S. H. 1987. Heterozygosity and developmental stability under inbreeding and crossbreeeding in Pinus attenuata. Evolution 41:331 -339. 21 Strauss, S. H., and W. J. Libby. 1987. Allozyme heterosis in radiata pine is poorly explained by overdominance. Am. Nat. 130:879-890. Stuart-Smith, G. J. 1998. Conservation of whitebark pine in the Canadian Rockies: blister rust and population genetics. M . Sc. University of Alberta, Edmonton. Szmidt, A. E. 1982 in El-Kassaby, Y. A. 1991. Genetic variation within and among conifer populations: Review and evaluation of methods In Fineschi, S., M.E. Malvolti, F. Cannata, and H.H. Hattemer, editors. Biochemical markers in the population genetics of forest trees. SPB Academic Publishing, The Hague, The Netherlands. Tani, N. , N . Tomaru, M . Araki, and K. Ohba. 1996. Genetic diversity and differentiation in populations of Japanese stone pine (Pinus pumila) in Japan. Can. J. For. Res. 26:1454-1462. Tomback, D. F. 1978. Foraging strategies of Clark's nutcracker. Living Bird 16:123-161. Tomback, D. F. 1982. Dispersal of whitebark pine seeds by Clark's Nutcracker: a mutualism hypothesis. J. Anim. Ecol. 51:451-467. Tomback, D. F., S. F. Arno, and R. E. Keane. 2001. The compelling case for management intervention. Pages 3-25 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Tomback, D. F., and Y. B. Linhart. 1990. The evolution of bird-dispersed Pines. Evol. Ecol. 4:185-219. Tomback, D. F., and W. S. Schuster. 1994. Genetic population structure and growth form distribution in bird-dispersed pines. Pages 43-50 in W. C. Schmidt and F.-K. Holtmeier, editors. International Workshop on Subalpine Stone Pines and Their Environment: The Status of Our Knowledge; 1992 September 5-11; St. Moritz, Switzerland; Gen. Tech. Rep. INT-GTR-309. U.S. Dept. of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden, UT. Tomback, D. F., S. K. Sund, and L. A. Hoffmann. 1993. Postfire regeneration of Pinus albicaulis -height-age relationships, age structure, and microsite characteristics. Can. J. For. Res. 23:113-119. Tomback, F. 2001. Clark's Nutcracker: agents of regeneration. Pages 89-104 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Turreson, G. 1922. The genotypical response of the plant species to the habitat. Hereditas 3:211-350. Uyenoyama, M . K. 1986. Inbreeding and the cost of meiosis: the evolution of selling in populations practicing biparental inbreeding. Evolution 40:388-404. van Niejenhuis, A., and W. H. Parker. 1996. Adaptive variation in jack pine from north central Ontario determined by short-term common garden tests. Can. J. For. Res. 26:2006-2014. Vander Wall, S. B., and R. P. Balda. 1977. Coadaptations of the Clark's Nutcracker and the pinyon pine fore efficient seed harvest and dispersal. Ecol. Mon. 47:89-111. Vogler, D. W., and S. Kalisz. 2001. Sex among the flowers: the distribution of plant mating systems. Evolution 55:202-204. Wang, T., A. Hamann, D. L. Spittlehouse, and S. N . Aitken. 2006. Development of scale-free climate data for western Canada for use in resource management. Int. J. Climatol. 26:383-397. Wang, T., A. Hamann, A. Yanchuk, G. A. O'Neill, S. N . Aitkan. In press. Use of response functions in selecting lodgepole pine populations for future climates. Glob. Change Biol. Wei, R.-P., S. D. Han, N. K. Dhir, and F. C. Yeh. 2004. Population variation in growth and 15-year-old shoot elongation along geographic and climatic gradients in black spruce in Alberta. Can. J, For. Res. 34:1691-1702. Williams, C. G., and O. Savolainen. 1996. Inbreeding depression in conifers: Implications for breeding strategy. Forest Sci. 42:102-117. Yandell, U. G. 1992. An allozyme analysis of whitebark pine (Pinus albicaulis Engl.). M . Sc. University of Nevada, Reno. Zeglen, S. 2002. Whitebark pine and white pine blister rust in British Columbia, Canada. Can. J. For. Res. 32:1265-1274. 22 CHAPTER 2: Genetic Diversity and Geographic Differentiation in Quantitative Traits, and Seed Transfer Guidelines for Whitebark Pine {Pinus albicaulis Engelm.)1 Andrew D. Bower and Sally N. Aitken 2.1 I N T R O D U C T I O N Understanding the geographic structure of genetic variation and the environmental factors driving evolutionary processes is important for management of genetic resources. Movement of seeds from their collection site to other environments within their natural range may increase the risk of maladaptation (Campbell 1979). Reduced growth or mortality due to maladaptation could reduce the success of restoration projects and gene flow from planted trees into adjacent native trees could negatively affect populations adapted to local conditions (McKay et al. 2005). Variation in selectively neutral molecular markers primarily reflects the effects of genetic drift (population size and history), gene flow, and mutation, and does not accurately reflect adaptive evolutionary processes. Therefore, molecular marker results alone are not sufficient to describe differentiation among populations brought about by natural selection (Reed and Frankham 2001). Seed transfer should be guided by natural levels of genetic variation and local adaptation in quantitative traits specific to the species in question (Morgenstern 1996, Hufford and Mazer 2003, McKay et al. 2005). This information can be determined by growing seedlings from diverse geographic origins in a common environment and assessing differences among sources in these traits (e.g., growth, phenology, stress tolerance, and photosynthetic capacity). Mapped genetic variation and an understanding of genetic structure are used for managing breeding programs, for evaluating conservation of genetic resources, for predicting the possible effects of climate change, and for developing guidelines for seed transfer (St. Clair et al. 2005). While most genecological studies have been precursors for tree breeding programs, this information is often missing for species of low economic value. Given the potential for introduction of poorly adapted genotypes, information on 1 A version of this chapter will be submitted for publication. Bower, A. D. and Aitken, S. N . Genetic diversity and geographic differentiation in quantitative traits, and seed transfer guidelines for whitebark pine {Pinus albicaulis Engelm.). To be submitted to American Journal of Botany 23 local adaptation is needed to guide seed transfer from collection site to restoration site. There is a widespread need for restoration and often a limited supply of seed for whitebark pine (Pinus albicaulis Englem.) and geographic restrictions on seed transfer are needed for restoration and conservation of this threatened, ecologically important species. Whitebark pine is a high elevation, five-needle pine, and the only North American member of the stone pines (Pinus subsection Cembrae) (Arno and Hoff 1989, Price et al. 1998; but see Gernandt et al. 2005). Although of little commercial value, it has tremendous ecological value and is considered a keystone species. It provides a rich food source for a variety of animals; recolonizes recently burned areas; regulates snowmelt and reduces soil erosion; facilitates succession and fosters diversity in its ecosystem; and, it has high aesthetic value for alpine recreationists (Tomback et al. 2001). Whitebark pine is an important food source for the Clark's Nutcracker (Nucifragia columbiana Wilson), which is its primary seed dispersal agent and mutualist. The nutcrackers harvest seeds from the unopened cones and bury thousands of seed caches for later use (Tomback 1978, Hutchins and Lanner 1982, Lanner 1982, Tomback 1982). However, whitebark pine is in decline throughout most of its range from a synergism of natural and human-driven causes. Outbreaks of mountain pine beetle (Dendroctonus ponderosae Hopkins) and decades of fire suppression have led to mortality and successional replacement by shade tolerant species, but the greatest impact comes from the introduced disease white pine blister rust (caused by the fungus Cronartium ribicola J.C. Fisch. ex Rabh.). Since its introduction to western North America in 1910, blister rust has spread throughout nearly the entire range of whitebark pine, causing reduced cone production and widespread mortality (McDonald and Hoff 2001). The consequences of losing whitebark pine as a major component of the subalpine forest could include altered watershed hydrology and successional processes, homogenization of the subalpine landscape, impacts on grizzly bear (Ursus arctos horribilis Ord) nutrition, reduced nutcracker populations, and impaired aesthetic and recreational values (Tomback et al. 2001). Scientists are in agreement that whitebark pine ecosystems require immediate restoration to reduce 24 the effects of fire exclusion and blister rust (McCool and Freimund 2 0 0 1 ) . Silvicultural techniques can be used to encourage natural regeneration, but in stands with a damaged natural seed source, or those that need to be regenerated quickly, planting seedlings is the suggested restoration practice (Hoff et al. 2 0 0 1 ) , using blister rust-resistant seedlings when they are available. As a result of its low commercial value, whitebark pine has not been studied to the extent of many other conifers in North America. Genetic variation in whitebark pine has been assessed using molecular markers at scales ranging from a single watershed (e.g., Rogers et al. 1 9 9 9 ) to across most of its range using monoterpenes (Zavarin et al. 1 9 9 1 ) and molecular markers (both isozymes and organellar DNA) (Yandell 1 9 9 2 , Jorgensen and Hamrick 1 9 9 7 , Bruederle et al. 1 9 9 8 , Stuart-Smith 1 9 9 8 , Rogers et al. 1 9 9 9 , Richardson et al. 2 0 0 2 , Krakowski et al. 2 0 0 3 ) . However, levels of genetic variation and geographic differentiation in quantitative traits in whitebark pine are unknown. The relationship between quantitative trait variation and molecular measures of genetic diversity is a topic that has been reviewed and debated extensively (Frankham 1 9 9 9 , Whitlock 1 9 9 9 , Merila and Crnokrak 2 0 0 1 , Reed and Frankham 2 0 0 1 , Crnokrak and Merila 2 0 0 2 , Hendry 2 0 0 2 , McKay and Latta 2 0 0 2 ) . Population differentiation for quantitative traits can be estimated using Q S T , which is the analogue of F S T for genetic markers (Spitze 1 9 9 3 ) . Meta-analyses of plant and animal studies have shown that population differentiation in quantitative traits (life history and morphological traits combined) ( Q S T ) typically exceeds that estimated from neutral marker genes (allozymes and DNA markers combined) ( F S T ) (Merila and Crnokrak 2 0 0 1 , McKay and Latta 2 0 0 2 ) . This indicates a prominent role of natural selection in the differentiation of populations for quantitative traits (Lynch et al. 1 9 9 9 , Whitlock 1 9 9 9 ) . At a coarse scale, the ranges of many trees species are predicted to shift higher in latitude and elevation as a result of climate change (Davis and Shaw 2 0 0 1 ) . However, at a finer scale, projected vegetation responses include a combination of elevational and directional adjustments, as the location of suitable conditions for each taxon shift within a region (Barlein et al. 1 9 9 7 ) . Studies in the Greater Yellowstone Ecosystem predict whitebark pine to be highly affected under predicted future climate 25 scenarios, with the extent of its range dramatically decreased and increasingly fragmented due to the loss of current climatic conditions at high elevation that sustain this species (Romme and Turner 1991, Bartlein et al. 1997) There is concern that tree species may not have adequate genetic variation to adapt to these changes, or due to long generation times will not be able to adapt or disperse quickly enough to keep pace with the predicted changes (Hamrick 2004, Davis et al. 2005). Hamrick (2004), however, argues that many trees species are uniquely equipped to withstand directional environmental change when compared to other plant species. Their longevity and phenotypic plasticity should allow individuals and populations to survive a few decades of change, and high genetic mobility through pollen flow and long distance seed dispersal should facilitate colonization of newly available habitats. In addition, high levels of genetic diversity within populations for quantitative traits means that there may be phenotypes pre-adapted to new climatic conditions (Hamrick 2004), and prolific-seed production of trees exposes available genetic variation to selection which could increase the frequency of genotypes that tolerate these conditions (Davis et al. 2005). The goals of this study are to investigate levels of genetic diversity for quantitative traits and describe geographic patterns of genetic differentiation for these traits in whitebark pine throughout its range. These estimates are then compared with previously reported values of F S j in this species to determine the extent of differential selection and local adaptation of populations. We also compare quantitative traits for provenances to geographic and climatic data from the sources location to determine if clinal patterns are evident, and if so, which environmental variables have the greatest influence on this adaptation. We then use this information to develop seed transfer recommendations for restoration plantings. In addition, we discuss the potential response of whitebark pine to predicted climate change. 26 2.2 MATERIALS AND METHODS 2.2.1 Sample Materials Open pollinated seeds from 48 provenances of whitebark pine from across most of the range (Table 2.1 and Figure 2.1) were germinated in 2002 following stratification, using the protocol described by Burr et al. (2001). Germinants were sown into individual 10 in 3 Ray Leach "Cone-tainer" super cells (Stuewe and Sons, Inc., Corvallis, Oregon) for their first growing season. In November 2002, 10-month-old seedlings were transplanted into a raised nursery bed common garden in Vancouver, British Columbia (49° 13'N, 123° 6'W) and grown for two growing seasons. Seedlings were planted in an incomplete block Alpha design (Patterson and Williams 1976) with 12 replications, and ten four-tree by four-tree incomplete blocks within replications. Each replication contained 160 test trees, with families usually represented once per replication. If fewer than 12 seedlings were available for a family, another family from the same provenance was used as a substitute to fill all 12 replications. The AlphaPlus program (CSIRO 1996) was used to design the planting layout and assign seedlings randomly within replications. This program attempts to allocate each family to an incomplete block with each other family only once, when possible. Seedlings were planted at 9.5 x 10 cm spacing with one row of buffer trees for which data were not collected surrounding each raised nursery bed. Seedlings were kept well watered and were fertilized and weeded as needed to provide optimum growing conditions. As temperatures in Vancouver are much higher than those in the native environment, two temperature treatments were imposed: eight replications had ambient soil temperature (ambient treatment) and the remaining four replications (cold treatment) had cooled water pumped through hoses buried approximately 25 cm below the surface to which kept soil temperature consistently ~8°C cooler during the warmest part of the day. The timing of initiation of growth in the spring was observed for the 2003 and 2004 growing seasons, and at the end of the 2004 growing season survival, height growth, above-ground and below-ground oven-dry biomass were measured to determine total biomass and root:shoot ratio on all replications. Artificial freeze testing was performed on 5mm needle segments from all seedlings in the ambient 27 treatment in three replications in the fall of 2003 and four replications in the spring of 2004. The electrolyte leakage method was used to determine cold injury. Details of cold hardiness testing are given in Bower and Aitken (2006, Chapter 5). The final data set contained nine quantitative variables; data from all trees in both temperature treatments were available for third-year height increment, root dry biomass, shoot dry biomass, and total dry biomass, rootrshoot ratio, date of needle flush in 2003 and 2004. In addition, percent survival in each soil temperature treatment was tested for treatment effects. 2.2.2 Data Analysis SAS Version 8 (SAS Institute 1999) was used for all statistical analyses. Data were standardized to a mean of zero and standard deviation of one in order to eliminate scale differences among variables using PROC STANDARD. Preliminary analysis showed an increase in variability of residuals with an increase in predicted values, so a natural-log transformation was applied to height increment, root, shoot, and total biomass, and root:shoot ratio for all analyses to equalize variance. To test for differences between soil temperature treatments and genotype-by-environment interactions in the quantitative traits, PROC MIXED was used with the following model: yijklmn = /* + '/ + r{t\ + b(rt)iJk +Pl+ ptu + pr(t%, + f(p)Im + eiJklmn where Vyw„„, is the observed value for tree n in family m within provenance / in incomplete block k in rep j in soil temperature /, u is the overall mean, /, is the effect of temperature /, rit)y is the effect of rep j nested within temperature b(rt)ijk is the effect of incomplete block k nested within rep j within temperature i, pi is the effect of provenance I, ptu is the interaction of temperature i and provenance /, pr(t)jji is the interaction of provenance / and rep j within temperature i, and J[p)im is the effect of family m nested within provenance /, and ey!hm is the random residual error. Temperature, provenance, and provenance-by- temperature interaction were considered fixed, while all other effects were considered random. Provenance means were used to test for differences between treatments for 28 survival percentage using the above model with only the temperature and provenance effects, and their interaction. The cold treatment had lower replication, lack of cold injury testing, and the absence of key provenances at the northern and southern ends of the range compared to the ambient treatment. Thus, only the data from the ambient treatment were used in the analysis of provenance variation and the canonical correlation analysis. To test differences among provenances and to obtain estimates of variance components within each soil temperature, PROC MIXED was used with the REML variance component estimator and the following model: yijkin, =M + ri+ b{r)jj + pk + rpik + f(p)k! + eiJklm where terms for each effect are the same as listed above without the effect of soil temperature. Al l terms were considered random except for provenance, which was fixed. Genetic differentiation among provenances was estimated for all quantitative traits by calculation of Q s l : o - " l 2 2 where o~a is the among population (provenance) variance and o~w is the within population additive genetic variance. In this study the variance component for provenance ( o~2 ) was used as the among-population variance, and three times the variance component for family-within-provenance (3cr^( j) was used as the within-population variance. The within-population genetic variation was approximated as three times the family variance instead of four as is used for true half-sibs, because open-pollinated progeny of whitebark pine are more closely related than half-sibs due to moderate inbreeding and correlated paternity (Squillace 1974, Krakowski et al. 2003, Chapter 3). Values of Q s t were compared to published estimates for genetic markers (FSx or GST)-Pearson correlations among family least-squares-means for quantitative, climatic, and geographic variables were calculated using PROC CORR. A sequential Bonferroni adjustment was 29 used to ensure a = 0.05 over all comparisons within each group of variables (Rice 1989). Climatic variables used were mean annual temperature (MAT), mean temperature of the coldest month (MTCM), mean temperature of the warmest month (MTWM), mean annual precipitation (MAP), mean summer precipitation (MSP), annual heat:moisture index (AH:M) [(MAT + 10)/(MAP/1000)], summer heat:moisture index (SH:M) [(MWMT/(MSP/1000)], and frost free period (FFP). Climatic variables for provenances north of 48° were obtained from PRISM climatic data corrected for local elevation using the Climate BC model described by Wang et al. (2006). For provenances south of 48°, climatic data were obtained from a climate model using the thin plate splines of Hutchinson (2000) as illustrated for North America by McKinney et al. (2001). Geographic variables analysed were provenance latitude, longitude, elevation, and their squares to account for non-linear trends, and latitude times longitude (LTxLN) and latitude divided by longitude (LT/LN), which produce a grid from northwest-to-southeast and northeast to southwest, respectively. Clines in quantitative traits can be obscured when there are correlations among traits or if geographical structure is complex. In these cases, canonical correlation analysis is more efficient than regressing each trait on environmental variables separately (Westfall 1992). Several of the quantitative, climatic, and geographic variables were strongly intercorrelated (see Tables 2.5, 2.6, 2.7), so canonical correlation analysis was used to examine the relationship of quantitative traits with climatic and geographic variables, and to determine which variables are useful in explaining the gradients associated with clines in seedling traits. Again, only the data from the ambient treatment was used, with the climatic data and least-squares means for each provenance obtained from the analysis of variance for each quantitative trait. Canonical redundancy analysis was used to determine the proportion of variation in quantitative traits that can be accounted for by canonical correlations of the climatic and geographic data sets. Only quantitative variables with statistically significant (p<0.05) differences among provenances were included in this analysis. To determine whether variation in quantitative traits among provenances is clinal and to develop predictive equations for the construction of seed transfer guidelines, values of significant 30 canonical variables associated with the quantitative traits were regressed on the standardized key climatic variable with the highest loading for that canonical variable. The slope of this regression provides a rate of change in the canonical variable associated with the quantitative traits relative to the climatic variable. Rates of differentiation along climatic gradients were interpreted relative to the least significant difference among provenances at the 20% level (LSD 0.2). This reduces Type II error - accepting no differences among provenances when differences actually exist. Values of LSD for the canonical variables were obtained from a Duncan's Multiple Range test in PROC GLM using the model for testing variation among provenances described above. The rate of differentiation of the key quantitative traits associated with canonical vectors was determined as the change in the standardized climate variable associated with the LSD value of the canonical variable. The difference in the climate variable associated with significant genetic differentiation between provenances was calculated as the rate of differentiation multiplied by the standard deviation of the climate variable. Simple regressions of the climate variable on latitude, longitude, and elevation were used to determine the geographic distance associated with the rate of differentiation in the climate variable in order to make simple operational seed transfer recommendations. 2.3 R E S U L T S 2.3.1 Soil temperature effects Height increment and survival were significantly higher (15 and 26%, respectively) in the cold treatment than in the ambient treatment (Table 2.2). Means for biomass traits were also greater in the cold treatment, although the difference between treatments was not significant. The difference between treatments was greater for root mass than shoot mass, although neither was significant. The date of needle flush did not differ significantly between treatments and provenance-by-treatment interaction was not significant for any of the traits. Seedlings in the cold temperature treatments generally appeared healthier than those in the ambient treatment with darker green foliage at the end of the experiment. 31 2.3.2 Genetic variation and population differentiation In the ambient soil temperature treatment, significant differences were detected among provenances for all variables except root:shoot ratio and spring cold injury. Provenance accounted for only a small proportion of the variance for growth traits (height increment and biomass) as shown by the intraclass correlation coefficient (o2P/o2T) (Table 2.3). However, for date of needle flush and fall cold injury, provenance accounted for a substantial proportion of the variance. Similar patterns were observed in the cold soil temperature treatment; however, the differences among provenances were not significant for any of the biomass traits. Despite a lack of significant differences among provenances in the ANOVA, root:shoot ratio differed significantly among provenances in a Duncan's multiple range test. Genetic differentiation (Q S T ) among provenances was weak to moderate for the growth traits, with height increment having the largest value in both treatments (Table 2.3). Date of needle flush and fall cold injury showed strong differentiation among provenances regardless of treatment. A comparison of QST values with previously published values of F S T f o r whitebark pine (Table 2.4) shows that the quantitative traits with the weakest differentiation are similar to the highest estimates of differentiation in presumably neutral molecular markers from range-wide studies (Richardson et al. 2002, Jorgensen and Hamrick 1997), and the quantitative traits with the strongest differentiation have substantially higher QST estimates. Provenance means for quantitative traits show strong positive correlations (r>0.75) among height increment, root, shoot, and total biomass. High positive correlations were also observed between date of needle flush in different years, and between date of needle flush in 2004 and fall cold injury (Table 2.5). In the climate variables for the provenance origins, significant positive correlations (r>0.6) were detected among the temperature, among the precipitation, and among the aridity variables, as would be expected (Tables 6). There were also moderate (r>0.49) positive relationships of both mean annual temperature and mean annual precipitation with frost free period. 32 The only significant correlations observed for provenance means of quantitative traits with climatic variables were strong positive associations (r>0.6) between fall cold injury and mean temperature of the coldest month; and between date of needle flush in both years with mean temperature of the coldest month and summer aridity index (SH:M) (Table 2.7). Root:shoot ratio was positively correlated with latitude and negatively correlated with elevation, while the correlations of date of needle flush in both years had a strong negative correlation (|r|>0.7) with latitude and a moderate positive correlation (r>0.48) with elevation (Table 2.7). Only mean temperature of the coldest month and summer heatrmoisture index were negatively correlated with latitude, and only frost free period was negatively correlated with elevation (Table 2.7). In the canonical correlation analysis between provenance means for the quantitative traits and climatic variables for provenance origin, the first canonical correlation was significantly different from zero and explained 68% of the variance in the data (Table 2.8). The first pair of canonical variables demonstrates the effects of mean temperature of the coldest month as it had the highest loading with this variable (Table 2.9). The positive correlations of date of needle flush and fall cold injury with the first canonical variable (Figure 2.2) indicates that trees from provenances with higher mean temperature of the coldest month flush later in the spring and suffer higher cold injury in the fall. Scores for the first pair of canonical variables show a clear separation of the southern (Oregon and California) provenances from the Rocky Mountain and Canadian provenances (Figure 2.3). The second correlation was also significant and accounted for an additional 14% of the variation. The second pair of variables demonstrates the effect of the length of the growing season (FFP), with trees from provenances with longer growing seasons growing taller, and producing more biomass. Canonical redundancy analysis shows that the first two canonical correlations account for 24 and 17% (41% total) of the variation in quantitative traits, respectively. Using the geographic variables, the first four canonical correlations were significantly different from zero. The first pair of correlations demonstrated the effects of LT/LN and to a lesser degree both latitude and latitude2 (loadings=-0.92, 0.75, and 0.74, respectively). The second pair 33 demonstrated the effects of elevation and elevation2 (loadings=-0.75 and -0.73, respectively). The loadings for the third and fourth pair did not clearly demonstrate any single of group of variables (data not shown). The canonical redundancy analysis showed that the first four correlations with geographic variables accounted for 53% of the variance in quantitative traits. The first quantitative canonical correlation was regressed on mean temperature of the coldest month and the second on frost free period. Mean temperature of the coldest month explained a substantial proportion of the variance in the first canonical score (r2= 0.79, p<0.0001, Figure 2.3). It is clear from Figure 2.3 that the southern provenances have a large influence on the strength and slope of the regression line. When these provenances are excluded, the relationship is weaker (r2=0.31) but still significant (p=0.002). The relationship within the southern provenances alone was not significant (r2=0.06, p=0.43). The regression of the second canonical correlation on frost free period was also significant (p=0.001) but weak (^=0.23). The interval in mean temperature of the coldest month associated with a significant difference in the first quantitative canonical score (which largely reflects date of needle flush) was estimated as 1.16°C and the interval in frost free period associated with the second canonical score (which primarily represents height growth) was 16.2 days. The interval in mean temperature of the coldest month associated with a significant difference in the first canonical score (r2=0.49p<0.0001 for regression of mean temperature of the coldest month on latitude) translated to a geographic distance of 2.9° latitude or 709 meters in elevation. The interval in frost free period associated with a significant difference in the second canonical score translated to a difference of 1010 meters in elevation or 12.2° longitude (r2=0.23 p=0.001 for regression of frost free period on elevation). 3 4 2.4 DISCUSSION 2.4.1 Effects of soil temperature Whitebark pine is restricted to high-elevation, subalpine environments, where frosts can occur during any month of the year (Arno and Hoff 1989). The environment where the common garden experiment was grown (Vancouver, British.Columbia; elevation ~100m., MAT=10°C, MTWM=17.3°C, MTCM=3.2°C) was considerably warmer than its native habitat (Table 2.1). Although this warmer soil would enhance growth for most tree species, for whitebark pine, the ambient soil temperature was more stressful than the cold soil treatment, even with the soil kept moist. Despite the difference between the treatments (afternoon temperatures in this test at a depth of 15 cm varied from 19 to 25°C in the ambient treatment and 10 to 17°C in the cold treatment), the temperatures in the ambient treatment were still below that usually considered lethal for roots of most trees (Helgerson 1990). The darker color and superior health of the seedlings in the cold treatment indicated that higher soil temperature was a stress that appeared to be cumulative over the two growing seasons. 2.4.2 Genetic variation • - ' We observed significant differences among provenance means in most quantitative traits (Table 2.3), similar to many other widespread North American conifers (Morgenstern 1996). Differences among provenances accounted for a substantial proportion of the variance only for cold acclimation traits (date of needle flush and fall cold injury), indicating that provenances are under stronger differential selection for these traits than for growth. In the subalpine environments where whitebark pine grows, these traits most likely have a larger role than growth traits in determining local fitness and the ability to withstand abiotic stresses associated with local climate. 35 2.4.3 Environmental effects on quantitative traits Despite a higher percentage of the variance being accounted for by geographic than climatic variables, we have focused our discussion on relationships with climate. Genetic variation for quantitative traits in tree species with large distributions is usually the result of adaptation to local environments in combination with gene flow from nearby populations. Using geographic variables to explain spatial genetic patterns fails to account for local environmental variation and only predicts broad trends in genetic variation due to spatial autocorrelation in both data sets (Hamann and Wang 2005). Provenances differed significantly for nearly all traits. However, only traits related to adaptation to cold displayed clinal variation patterns that corresponded to climatic gradients (Table 2.7) indicating adaptation to local environment. In the common garden environment, provenances from higher latitudes and lower winter temperatures flushed earlier in the spring, suffered less cold injury in the fall, and allocated more biomass to shoots. The canonical correlation analysis supports the results of the simple correlations since the first pair of correlations showed a positive relationship between date of needle flush, and to some degree, fall cold injury, and mean temperature of the coldest month (Table 2.9 and Figure 2.3). Whitebark pine has a high level of cold hardiness compared to other conifers, but significant differences exist among geographic regions (Bower and Aitken 2006, Chapter 5). However, cold hardiness is a function of timing of initiation and loss rather than maximum level of hardiness attained, and requires a tradeoff between the length of the growing season and avoidance of early fall or late spring frosts (Howe et al. 2003). Despite a higher risk of spring cold injury, earlier flushing in the spring allows trees from higher latitudes to extend their growing season relative to trees from lower latitudes (Sagnard et al. 2002), a pattern common in temperate forest trees (Morgenstern 1996). The multivariate analysis revealed one clinal pattern that was not evident from the simple correlations: the second pair of canonical correlations showed a positive relationship of height increment and biomass traits with frost free period (Table 2.9). This relationship likely reflects local 36 adaptation of trees to length of growing season, since in common garden experiments, populations from colder locations usually stop growing earlier than plants from warmer locations (Howe et al. 2003). 2.4.4 Comparison of genetic differentiation Population differentiation (QST) in all quantitative traits except root:shoot ratio was greater than estimates of differentiation in neutral molecular markers ( F S T and G S T ) for whitebark pine (Table 2.3). Levels of population differentiation ( F S j ) previously reported in whitebark pine average 5.8%, which is slightly higher than most values reported for other stone pine species (Pinus subsection Cembrae) (Table 2.4). This indicates that most of the neutral variation in whitebark pine is among individuals within populations. Our results show that differentiation is much stronger for quantitative traits than neutral markers in whitebark pine. These results agree with the majority of studies that have compared quantitative and molecular variation and found Q S T > Fs-rin forest trees (Howe et al. 2003) as well as in other plants and animals (Merila and Crnokrak 2001, Reed and Frankham 2001, McKay and Latta 2002). This is evidence of differential natural selection among populations driving local adaptation in quantitative traits, and further indicates that cold adaptation traits have stronger selection pressures acting on them than the growth traits in whitebark pine. The relationship between date of needle flush and MTCM shown in the canonical correlation analysis supports this conclusion. Low temperatures, especially in winter appear to be the force driving local adaptation in date of needle flush. 37 2.4.5 Seed transfer guidelines While whitebark pine is distributed over a large range of latitudes, these results show that trees from a particular provenance are expected to be optimally adapted for only a portion of the environmental conditions experienced across the range of the species. Under the assumption that local sources are optimal and deviation of a provenance mean from the local source represents the degree of suboptimality, we have used the floating seed transfer model developed by Rehfeldt (1991, 1994) to determine seed transfer guidelines for restoration programs of whitebark pine. Seed transfer guidelines should be developed based on the traits showing the strongest local adaptation. Of the traits we assessed, date of needle flush is the trait that should be considered when moving seed to minimize the risk of maladaptation, as it shows the highest Q S T (0.43-0.63)(Table 2.3). Whitebark pine seed can be moved between areas differing by up to 1.2°C in mean temperature of the coldest month while maintaining a date of needle flush suitable for the local climate and with minimal risk of fall cold injury. Restoration ecologists, park managers, and foresters can more easily use seed transfer guidelines based on geographic distances than climatic differences and this difference in mean temperature of the coldest month corresponds to approximately 3° in latitude or 340 km. The difference in elevation required to distinguish genetically different populations (~700m) exceeds the elevational range of whitebark pine within 3° of latitude, so there should be no elevational restrictions on seed movement. The latitudinal patterns are similar to those for mountain hemlock (Tsuga mertensiana), a sympatric species in the western part of whitebark pine's range. It is suggested that transfer of T. mertensiana seed more than 3° north along the coast may result in significant frost damage to plantations (Benowicz et al. 2001). Differences in height growth and biomass in whitebark pine seedlings were related to the length of the growing season (frost free period); however, seed transfer distance based on these traits was too large to be of practical use in a conservation or restoration program. These guidelines are established based on a 20% risk of assuming no difference among provenances when a difference actually exists. Given the limited supply of whitebark pine seeds, it 38 may be necessary in some cases to exceed the distances established by these guidelines. However, exceeding these distances increases the chance of maladaptation and should be done only after weighing this risk against the need for immediate restoration (as opposed to waiting until a local seed source is available). If these guidelines are exceeded, then the distance of seed movement should be minimized to whatever extent possible. In western North America, MAT is predicted by global circulation models to rise by 3 to 5°C over the next century (Hulme et al. 1999). However, the role of climate change in the future of whitebark pine is uncertain. The predicted warming will likely cause phenotypes to shift northward to track conditions to which they are locally adapted (Davis and Shaw 2001) but climatic differences among regions should maintain clinal variation. It is likely that whitebark pine will shift both northward and higher in elevation, tracking the niche in which it can survive and be competitive. Seed transfer distances for several tree species in British Columbia are more restricted from the collection site south to the planting site than to the north, as sources from slightly south of the location of test sites grew better than local sources (British Columbia Ministry of Forests 1995). Restricting movement of seed to the south also provides a buffer against future climate warming. Therefore, movement of whitebark pine seed should be restricted to 3° to the north and 1° to the south in order to minimize maladaptation in current and future environments. Ideally, restoration activities should attempts to utilize seed from within the local provenance or the nearest provenance possible. Based on the relationship between MTCM and latitude, the predicted increase in temperature is equivalent to a shift northward of 7.5° to 10° latitude or approximately 840 to 1400 km. Surviving a shift of this magnitude will require adaptation of populations to a new climate, as this distance is beyond the boundary of genetically different populations of whitebark pine. Due to genetic variation, climate change will affect populations differently throughout the range of the species, and populations will vary in their rate of adaptation to new conditions (Davis et al. 2005). Climate models predict a dramatic decrease in the range of habitat suitable for whitebark pine with increases in temperature and C0 2 (Romme and Turner 1991, Bartlein et al. 1997). While increasing temperature may result in new 39 habitat available north of its current range, it is also likely to lead to an upward shift of the timberline and the range of whitebark pine, resulting in a smaller potential area for it to occupy. In addition, if climate change results in an increase in summer precipitation, then even the remaining subalpine environment could become unsuitable for whitebark pine (Romme and Turner 1991). Whitebark pine is long-lived and individuals may be able to persist in their present environments for decades or even centuries after climate becomes unsuitable for the survival of their offspring (Brubaker 1986). Whitebark pine communities may appear stable, however, after a disturbance, the mature plant community could be replaced by an entirely different suite of species (Romme and Turner 1991). The wide ecological breadth of genotypes and the propensity for high gene flow through bird mediated seed dispersal and windborne pollen flow will aid whitebark pine in adapting to new environments; however, the long delay to reproductive maturity will slow adaptation. The predicted increases in temperature will push whitebark pine beyond the geographic limits to which is locally adapted and will likely result in a dramatic reduction in suitable habitat, potentially decreasing genetic diversity. White pine blister rust has had a drastic impact on whitebark pine populations, and has a significant negative association with observed heterozygosity (Chapter 4). Depletion of variability is one of the genetic factors that can impede adaptation (Davis et al. 2005), and the introduction of pathogens, insect pests and invasive species could interact with climate change to drive some species to extinction (Hamrick 2004). Damage and mortality due to blister rust is a primary concern for whitebark pine; however, the potential affects of climate change should not be underestimated. Without restoration and conservation efforts using appropriate seed sources that will be adapted to new climatic conditions, the decline of whitebark pine is likely to be accelerated, with the potential of extirpation in some areas. The data and results presented here are crucial for restoration efforts that will be necessary to maintain whitebark pine as more than a minor component of the ecosystems in which it plays such a vital role. 40 2.5 ACKNOWLEDGEMENTS We thank the USDA Forest Service regions 1, 5, and 6, the British Columbia Ministry of Forests, E. C. Manning and Tweedsmuir Provincial Parks of British Columbia, and Bob Brett of Snowline Ecological Consulting, Whistler, B.C. for providing seed for this study. Many people provided assistance in this project, including Dave Kolotelo, Joanne Tuytel, Christine Chourmouzis, Dorothy Watson, Karolyn Keir, Megan Harrison, Dane Szohner, Pia Smets, Jodie Krakowski, Seane Treheame, and all of the members of the Aitken lab at UBC. Climate data were provided by Drs. Tongli Wang and Gerald Rehfeldt. Funding for this study came from the British Columbia Forestry Investment Account through the Forest Genetics Council of B.C. to the Centre for Forest Gene Conservation at UBC. Thank you to Drs. Alvin Yanchuk, Mike Whitlock, Jeannette Whitton, Yousry El-Kassaby, Sean Graham, and Diana Tomback for their helpful comments on earlier drafts of this manuscript. 41 Table 2.1. Whitebark pine provenances sampled, number of seedlings tested, geographic coordinates, and climatic information. Site Reg." Name # Trees Lat. Long. Elev. M A T M T W M M T C M FFP SH:M b No. Amb. Cold. <°N) (°W) (m) (°C) b (°C) b (°C) b (days)" 1 N Serb Creek 5 10 54.71 127.57 1385 0.7 11.4 -10.7 52 32.7 2 N Hunters Basin 13 29 54.53 127.18 1446 0.3 11.1 -11 48 34.8 3 N Morice Lake 2 - 54.04 127.48 1231 0.6 11.3 -11.1 31 44.9 4 N Kimsquit river 1 - 53.19 127.18 900 3 12.5 -7.2 83 32.9 5 N Heckman Pass 6 - 52.52 125.82 1526 0 9.9 -10.6 58 46.8 6 N Perkins Peak 3 1 51.83 125.05 1916 -1.8 7.9 -11.5 35 31.9 7 N Jesamond 42 19 51.27 121.87 1846 0.9 11.4 -9.1 45 45 8 N Lime Mtn. 19 13 51.10 121.67 1900 0.5 11.1 -9.4 39 52.9 9 N Darcy 46 9 50.53 122.58 1800 0.5 11.7 -9.8 46.1 45.3 10 N Blackcomb 57 40 50.10 122.90 1908 0.6 10 -7.5 43.7 22.4 11 N Thynne Mtn. 20 1 49.71 120.92 1785 1.9 12.7 -7.7 48.8 27.4 12 N Manning Park 93 40 49.10 120.67 2000 0.3 10.8 -8.6 43.8 48.1 13 N Baldy 14 - 49.17 119.25 2154 1.2 12.1 -8.6 39.6 37.5 14 R Copper Butte 11 9 48.70 118.46 2185 -0.5 10.4 -10.2 48.2 30.7 15 R Cotville 6 4 48.66 118.46 2154 -0.1 10.7 -10.1 49 32.5 16 R Snow Peak - 7 48.58 118.48 2185 0.5 11.2 -9.7 48.7 35.6 17 R Salmo Mtn. 13 11 48.97 117.10 2092 0 10.7 -9.3 61.2 21.2 18 R Hooknose Mtn. - 2 48.94 117.43 2215 0.5 11.5 -8.9 54 28.8 19 R Farnham Ridge 31 26 48.84 116.51 1846 1.5 12.4 -8.3 70.6 35.4 20 R North Baldy - 7 48.55 117.16 1877 2.6 13.9 -7 89 41.6 21 R Lunch Peak 37 14 48.38 116.19 1846 2.1 12.4 -6.6 84 23.9 22 R Our Lake 6 13 47.84 112.81 2277 0.2 11.4 -9.6 31.3 35.1 23 R Sheep Shed 22 19 47.52 112.80 2154 1 12.3 -8.9 33.4 43.5 24 R Granite Butte 19 8 46.87 112.47 2338 0.5 12 -9.1 32.8 47.1 25 R Blacklead Mtn. 23 41 46.64 114.86 2062 1.2 12.3 -8.1 32.1 26.2 26 R Gospel Peak 22 16 45.63 115.95 2154 1 11.9 -8.3 26.4 33.9 27 R Heavens Gate 10 3 45.38 116.51 2154 1.2 12.3 -8.3 40.6 49.2 28 R Mudd Ridge 23 15 45.90 113.45 2400 0.2 11.8 -9.5 17.5 29 29 R Quartz Hi l l 27 44 45.71 112.93 2646 -0.8 10.9 -10.2 15.8 37 30 R Little Bear 22 26 45.40 111.28 2154 2.1 14.4 -8.9 40.5 43.4 31 R Picket Pin 4 1 45.44 110.05 2892 -1.8 9.9 -11 20.6 24.9 32 R Hellroaring II 20 24 45.04 109:45 2892 -1.4 10.3 -10.4 21.7 36.5 33 R Sawtel Peak 26 18 44.54 111.41 2400 -0.1 12.9 -11.9 25.9 45.8 34 R Vinegar Hi l l 14 26 44.72 118.57 2338 0.5 11.4 -8.6 39.8 54.1 35 S Mt. Hood 20 22 45.39 121.66 1969 1.7 10.8 -4.9 48.4 15 36 S Newberry Crater 30 13 43.72 121.23 2100 2.9 12.9 -4.7 45.5 58.1 37 s Paulina Peak 18 11 43.69 121.25 2250 2 11.9 -5.3 42.3 59.6 38 s Batchelor Butte 16 1 43.26 122.68 2200 1.9 10.7 -3.9 44.1 40 39 s Tipsoo Peak - 13 43.22 122.04 2462 0.7 10 -5.4 34.4 35.3 40 s Moon Mtn. 6 4 43.20 122.65 2201 1.9 10.8 -3.9 43.9 45.2 41 s Pelican Butte 40 17 42.51 122.15 2462 1.1 10.4 -5 35.2 50.6 42 s Ball Mtn. 22 15 41.80 122.16 2363 2.2 11.4 -4.4 39.1 107.1 43 s Goosenest Summit 6 6 41.72 122.23 2506 1.5 10.6 -4.6 35.3 99.3 44 s Drakes Peak 26 16 42.30 120.15 2462 2.5 13.1 -5.3 48.1 80.3 45 s Crane Mountain 19 22 42.07 120.24 2538 2.2 12.7 -5.6 46.3 72.4 46 s Mt. Rose 11 - 39.30 119.90 2754 2.4 12.6 -4.8 61 88.4 47 s Stevens Peak 7 - 38.70 119.98 2923 1.6 11 -5.2 46.6 61.6 48 s Ebbetts Pass 2 - 38.50 119.80 2769 2.5 12 -4.4 46.9 72 a Region; N=northern, R=Rocky Mountain, S=southern b See text for explanation of variables 42 Table 2.2. Soil temperature treatment least-squares-means, standard deviations, and significance level for eight quantitative traits. Ambient Cold F-value (and p-value) of Variable LSMean SD LSMean SD Treatment Provenance TxP H T I N C 3 1.89 0.07 2.18 0.09 4.83 (0.05) 2.11 (O.01) 1.13(0.33) T D M " 1.08 0.08 1.14 0.11 0.62 (0.45) 1.75 (0.02) 0.81 (0.76) R M a 0.18 0.10 0.25 0.13 0.58 (0.46) 1.62 (0.04) 0.84 (0.71) SJVT 0.53 0.07 0.59 0.10 0.67 (0.43) 1.87 (0.01) 0.73 (0.85) R S R a -0.38 0.04 -0.38 0.05 0.37 (0.55) 1.24 (0.21) 0.71 (0.87) FL03 116.78 0.61 117.29 0.67 0.23 (0.64) 6.03 (O.001) 1.01 (0.47) FL04 99.47 1.23 96.63 1.56 4.57 (0.06) 7.78 (O.001) 0.96 (0.54) Survival 0.65 0.02 0.82 0.02 21.19(0.001) 1.38(0.28) 0.90 (0.76) a Natural log transformed HTINC=height increment, TDM=total dry mass, RM=root mass, SM=shoot mass, RSR=root shoot ratio, FL03=2003 date of needle flush, FL04=2004 date of needle flush Table 2.3. Significance level of provenance effect in A N O V A , proportion of variation accounted for by provenance and family, and genetic differentiation (QST) for nine quantitative traits in ambient (A) Variable Provenance <T2p/ 0 2 p/<T 2 T Q S T F-value p-value A - H T I N C a b 1.84 0.01 0.05 0.05 0.14 A - T D M a b 1.69 0.01 0.04 0.09 0.07 A - R M a b 1.65 0.02 0.03 0.07 0.08 A - S M a b 1.73 0.01 0.04 0.10 0.07 A - R S R a b 1.09 0.35 0.00 0.03 0.00 A - F L 0 3 3 6.51 <0.001 0.26 0.05 -0.47 A-FL04 3 5.49 <0.001 0.23 0.04 0.47 A - F C I * 2.59 O.001 0.19 0.06 0.36 A S C I * 1.11 0.33 0.05 0.07 0.12 C-HTINC" 1.57 0.04 0.07 0.07 0.13 C - T D M b 1.24 0.20 0.06 0.11 0.09 C - R M b 1.22 0.22 0.05 0.10 0.08 C - S M b 1.23 0.21 0.06 0.11 0.08 C - R S R b 1.09 0.36 0.01 0.08 0.01 C-FL03 3.29 <0.001 0.21 0.05 0.43 C-FL04 6.36 O.001 0.33 0.03 0.65 a A=ambient soil temperature treatment, C=cold soil temperature treatment, see above for explanation of variables b Natural log transformed * FCI=fall cold injury, SCI=spring cold injury 43 Table 2.4. Reported values of genetic differentiation for whitebark pine and other stone pine (Pinus subsection Cembrae) species. Populations Species Area F S T or G S T Reference 14 P. albicaulis BC, ID, MT, OR 0.075 Chapter 4 30 USA range-wide 0.034 Jorgensen and Hamrick 1997 14 a Great Basin 0.088 Yandell 1992 29 ii Canadian Rockies 0.062 Stuart-Smith 1998 17 ii British Columbia 0.061 Krakowski et al. 2003 18 ii Range-wide 0.046* Richardson et al. 2002 8 P. sibirica Russia -0.042 Goncharenko et al. 1993a 11 " Russia 0.025 Krutovskii et al. 1995 P. koraiensis Coastal Russia 0.016 Potenko and Velikov 2001 19 " Russia 0.015 Potenko and Velikov 1998 3 " Russian far east 0.040 Krutovskii et al. 1995 5 P. pumila Russia 0.043 Goncharenko et al. 1993b 3 Kamchatka perm. 0.021 Krutovskii et al. 1995 18 " Japan 0.170 Tanietal. 1996 5 P. cembra Alps & eastern 0.040 Belokon et al. 2005 Carpathians * O s t from cpDNA microsatellite data Table 2.5. Correlations among provenance means for nine quantitative variables in ambient soil temperature treatment. Correlations when |r|>0.5 significant at a=0.05 after Bonferroni adjustment for number of correlations tested (n=36). H T I N C 3 T D M a R M 1 S M a RSR a FL03 FL04 F C I T D M 0.79 R M 0.78 0.96 S M 0.76 0.98 0.91 RSR 0.00 -0.1 0.16 -0.27 FL03 -0.11 -0.2 -0.28 -0.15 -0.31 FL04 -0.24 -0.22 -0.33 -0.15 -0.41 0.86 F C I 0.12 0.06 0.01 0.08 -0.18 0.44 0.51 SCI 0.22 0.36 0.36 0.34 0.11 -0.47 -0.42 0.16 a Natural log transformed 44 M A T M T W M M T C M M A P M S P A H : M S H : M M T W M 0.66* M T C M 0.80* 0.23 M A P 0.29 -0.20 0.40 M S P -0.13 -0.09 -0.05 0.65* A H : M 0.08 0.24 0.01 -0.70* -0.57* S H : M 0.45* 0.22 0.53* -0.29 -0.63* 0.61* F F P 0.52* 0.2 0.28 0.49* 0.11 -0.29 -0.02 * significant at a=0.05 after Bonferroni adjustment for number of correlations tested (n=28) MAT=mean annual temp., MTWM=mean temp, warmest month, MTCM=mean temp, coldest month, MAP=mean annual precip., MSP=mean summer precip., AH:M=annual heat:moisture index, SH:M=summer heat:moisture index, FFP=frost free period Table 2.7. Correlations among provenance means for quantitative and climatic variables. H T I N C T D M a R M a S M a RSR a FL03 FL04 F C I SCI Lat. Long. Elev. M A T 0.34 0.16 0.1 0.2 -0.22 0.46 0.33 0.5 0.04 -0.38 -0.30 -0.11 M T W M 0.11 0.14 0.08 0.19 -0.25 0.17 0.02 0.07 0.27 -0.23 0.27 -0.02 M T C M 0.23 0.11 0.02 0.16 -0.33 0.66* 0.66* 0.63* -0.22 -0.70* -0.23 0.30 M A P 0.24 -0.01 0.00 0.00 0.02 -0.03 0.04 0.32 -0.08 0.06 -0.37 -0.29 M S P -0.04 -0.07 -0.07 -0.05 -0.02 -0.31 -0.15 -0.14 -0.04 0.13 0.32 -0.06 A H : M 0.02 0.14 0.12 0.15 -0.07 0.28 0.18 -0.15 -0.09 -0.15 0.09 0.18 S H : M -0.03 0.02 -0.05 0.05 -0.25 0.71* 0.60* 0.32 -0.28 -0.59* -0.18 0.38 F F P 0.2 -0.06 -0.08 -0.04 -0.1 -0.07 -0.12 0.33 0.17 0.21 -.045 -0.48* Lat. 0.17 0.05 0.18 -0.04 0.51* -0.71* -0.76* -0.46 0.36 Long. -0.40 -0.07 -0.16 0.00 -0.34 -0.08 0.06 -0.31 0.00 Elev. -0.40 -0.13 -0.25 -0.04 -0.48* 0.48* 0.60* 0.12 -0.36 * significant at a=0.05 after Bonferroni adjustment for number of correlations tested (n=72, 27, and 24) 3 Natural log transformed Table 2.8. Canonical correlation analysis of the relationship between provenance mean quantitative traits and climatic variables. Pair Coeff. SE Eigenvalue Prop. var. F ndf ddf P-value 1 0.94 0.02 7.75 0.68 2.94 64 151 <0.001 2 0.79 0.06 1.61 0.14 1.78 49 136 0.005 3 0.70 0.08 0.98 0.08 1.43 36 121 0.080 4 0.63 0.10 0.65 0.06 1.11 25 106 0.346 5 0.51 0.12 0.34 0.03 0.71 16 89 0.780 6 0.20 0.15 0.04 <0.01 0.24 9 73 0.988 7 0.17 0.15 0.03 <0.01 0.22 4 62 0.927 8 0.02 0.16 0.00 0.00 0.01 1 32 0.906 Coeff.=correlation coefficient, Prop. var.=proportion of variance accounted for, ndf=numerator degrees of freedom, ddf=denominator degrees of freedom 45 Table 2.9. Correlations between quantitative canonical variables and the quantitative variables, and between climate canonical variables and both climate and quantitative variables. Variable Quantl Quant2 Variable Climl Clim2 Variable Climl Clim2 HTINC a 0.08 • 0.84 MAT 0.75 0.45 HTINC 0.07 0.66 TDM" -0.05 0.60 MI WM 0.23 0.16 TDM -0.05 0.47 R M a -0.12 0.56 M T C M 0.95 0.26 RM -0.12 0.44 SM a 0.00 0.62 MAP 0.20 0.57 SM -0.00 0.48 FL03 0.91 -0.26 MSP -0.25 0.21 FL03 0.85 -0.21 FL04 0.90 -0.32 A H M 0.21 -0.27 FL04 0.85 -0.25 FCI 0.66 0.27 SH M 0.74 -0.33 FCI 0.62 0.21 SCI -0.30 0.40 FFP 0.14 0.62 SCI -0.28 0.32 Natural log transformed 46 Figure 2.1. Distribution of whitebark pine and locations of provenances tested in common garden experiment. Dashed lines separate the Southern, Rocky Mountain and Northern regions. 47 Figure 2.2. Regression of first quantitative canonical score (QS1) on standardized mean temperature of the coldest month (MTCM) for 41 whitebark pine provenances in three geographic regions. Axes scales are standard deviations and bracket indicates value of LSD 0.20. CO O a FT = 0.79 p < 0.001 • -2 M T C M • Southern • Rocky Mountain • Northern Figure 2.3. Scatterplot of first two quantitative canonical scores (QC1 and QC2) based on eight quantitative traits for 41 provenances of whitebark pine. Axis scales are standardized values. Symbols refer to geographic groups shown in table 2.1. • A • 3 2.5b 2 1.5 A 1 CM to o o* • S o u t h e r n • R o c k y M o u n t a i n • Northern • •1 . -fJ?5 0 A ° CL1.5 -2 -2.5 Q C S 1 48 2.6 REFERENCES Arno, S. F., and R. J. Hoff. 1989. Silvics of whitebark pine (Pinus albicaulis), Gen. Tech. Rep. INT-GTR-253. U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Ogden, UT. Bartlein, P. J., C. Whitlock, and S. L. Shafer. 1997. Future climate in the Yellowstone National Park region and its potential impact on vegetation. Conserv. Biol. 11:782-792. British Columbia Ministry of Forests. 1995. Forest practices code of British Columbia, seed and vegetative materials guidebook. British Columbia Ministry of Forests and British Columbia Ministry of Environment. Belokon, M . M. , Y. S. Belokon, D. V. Politov, and Y. P. Altukhov. 2005. Allozyme polymorphism of Swiss stone pine Pinus cembra L. in mountain populations of the Alps and eastern Carpathians. Russ. J. Genet. 41:1268-1280. Benowicz, A., S. J. L'Hirondelle, and Y. A. El-Kassaby. 2001. Patterns of genetic variation in mountain hemlock (Tsuga mertensiana (Bong.) Carr.) with respect to height growth and frost hardiness. Forest Ecol. Manag. 154:23-33. Bower, A. D., and S. N . Aitken. 2006. Geographic and seasonal variation in cold hardiness of whitebark pine (Pinus albicaulis Engelm.). Can. J. For. Res. 36:1842-1850. Brubaker, L. B. 1986. Responses of tree populations to climatic change. Vegetatio 67:119-130. Bruederle, L. P., D. F. Tomback, K. K. Kelly, and R. C. Hardwick. 1998. Population genetic structure in a bird-dispersed pine, Pinus albicaulis (Pinaceae). Can. J. Bot: 76:83-90. Burr, K. E., A. Eramian, and K. Eggleston. 2001. Growing whitebark pine seedlings for restoration. Pages 325-345 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. Campbell, R. K. 1979. Genecology of Douglas-fir in a watershed in the Oregon cascades. Ecology 60:1036-1050. Crnokrak, P., and J. Merila. 2002. Genetic population divergence: markers and traits. Trends Ecol. Evol. 17:501. CSIRO. 1996. Alpha+; Experimental designs for variety trials and many-treatment experiments. Canberra, Australia. Davis, M . B., and R. G. Shaw. 2001. Range shifts and adaptive responses to Quaternary climate change. Science 292:673-679. Davis, M . B., R. G. Shaw, and J. R. Etterson. 2005. Evolutionary responses to changing climate. Ecology 86:1704-1714. Frankham, R. 1999. Quantitative genetics in conservation biology. Genet. Res. 74:237-244. Gernandt, D. S., G. Geada Lopez, S. Ortiz Garcia, and A. Liston. 2005. Phylogeny and classification of Pinus. Taxon 54:29-42. Goncharenko, G. G., V. E. Padutov, and A. E. Silin. 1993a. Allozyme variation in natural populations of Eurasian pines. 2. Genetic variation, diversity, differentiation, and gene flow in Pinus sibirica DuTour in some lowland and mountain populations. Silvae Genet. 42:246-253. Goncharenko, G. G., V. E. Padutov, and A. E. Silin. 1993b. Allozyme variation in natural populations of Eurasian pines. 1. Population structure, genetic variation, and differentiation in Pinus pumila (Pall) Regel from Chukotsk and Sakhalin. Silvae Genet. 42:237-246. Hamann, A., and T. Wang. 2005. Models of climatic normals for genecology and climate change studies in British Columbia. Ag. Forest Meteorol.:211-221. Hamrick, J. L. 2004. Response of forest trees to global environmental changes. Forest Ecol. Manag. 197:323-335. Helgerson, O. T. 1990. Heat damage in tree seedlings and its prevention. New Forest. 3:333-358. Hendry, A. P. 2002. Q s^F s t? Trends Ecol. Evol. 17:502. Hoff, R. J., D. E. Ferguson, G. I. McDonald, and R. E. Keane. 2001. Strategies for managing whitebark pine in the presence of white pine blister rust. Pages 346-366 in D. F. Tomback, S. F. 49 Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Howe, G. T., S. N. Aitken, D. B. Neale, K. D. Jermstad, N . C. Wheeler, and T. H. H. Chen. 2003. From genotype to phenotype: unraveling the complexities of cold adaptation in forest trees. Can. J. Bot. 81:1247-1266. Hufford, K. M. , and S. J. Mazer. 2003. Plant ecotypes: genetic differentiation in the age of ecological restoration. Trends Ecol. Evol. 18:147-155. Hulme, M. , J. Mitchell, W. Ingram, J. Lowe, T. Johns, M . New, and D. Viner. 1999. Climate change scenarios for global impact studies. Global Environ^ Chang. 9:S3-S19. Hutchins, H. E., and R. M . Lanner. 1982. The central role of Clark's Nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55:192-201. Hutchinson, M . F. 2000. ANUSPLIN Version 4.1 user's guide. Australian National University, Centre for Resource and Environmental Studies, Canberra. Jorgensen, S. M. , and J. L. Hamrick. 1997. Biogeography and population genetics of whitebark pine, Pinus albicaulis. Can. J. For. Res. 27:1574-1585. Krakowski, J., S. N . Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. Krutovskii, K. V., D. V. Politov, and Y. P. Altukhov. 1995. Isozyme study of population genetic structure, mating system and phylogenetic relationships of the five stone pine species (subsection cembrae, section strobi, subgenus strobus). Pages 279-304 in P. Baradat, W. T. Adams, and M.-S. G., editors. Population genetics and genetic conservation of forest trees. SPB Academic Publishing, Amsterdam, The Netherlands. Lanner, R. M . 1982. Adaptations of whitebark pine for seed dispersal by Clark's Nutcracker. Can. J. For. Res. 12:391-402. Little, E. L., Jr. 1971. Atlas of United States trees, volume 1, conifers and important hardwoods. USDA For. Serv. Misc. Publ. 1146. Washington, D.C. Lynch, M. , M . Pfrender, K. Spitze, N. Lehman, J. Hicks, D. Allen, L. Latta, M . Ottene, F. Bogue, and J. Colboume. 1999. The quantitative and molecular genetic architecture of a subdivided species. Evolution 53:100-110. McCool, S. F., and W. A. Freimund. 2001. Threatened landscapes and fragile experiences: conflict in whitebark pine restoration. Pages 263-284 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. McDonald, G. I., and R. J. Hoff. 2001. Blister rust: and introduced plague, in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. McKay, J. K., C. E. Christian, S. Harrison, and K. J. Rice. 2005. "How local is local?" - A review of practical and conceptual issues in the genetics of restoration. Restor. Ecol. 13:432-440. McKay, J. K., and R. G. Latta. 2002. Adaptive population divergence: markers, QTL and traits. Trends Ecol. Evol. 17:285-291. McKinney, D. W., M . F. Hutchinson, J. L. Kesteven, and L. A. Venier. 2001. Canada's plant hardiness zones revisited using modern climate interpolation techniques. Can. J. Plant Sci. 81:129-143. Merila, J., and P. Crnokrak. 2001. Comparison of genetic differentiation at marker loci and quantitative traits. J. Evolution. Biol. 14:892-903. Morgenstern, K. E. 1996. Geographic variation in forest trees. UBC Press, Vancouver, B.C. Patterson, H. D., and E. R. Williams. 1976. A new class of resolvable incomplete block designs. Biometrika 63:83-92. Price, R. A., A. Liston, and S. H. Strauss. 1998. Phylogeny and classification of Pinus. Pages 49-91 in D. M. Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. 50 Potenko, V. V., and A. V. Velikov. 1998. Genetic diversity and differentiation of natural populations of Pinus koraiensis (Sieb. et Zucc.) in Russia. Silvae Genet. 47:202-208. Potenko, V. V., and A. V. Velikov. 2001. Allozyme variation and mating system of coastal populations of Pinus koraiensis Sieb. et Zucc in Russia. Silvae Genet. 50:117-122. Reed, D. H., and R. Frankham. 2001. How closely correlated are molecular and quantitative measures of genetic variation? A meta-analysis. Evolution 55:1095-1103. Rehfeldt, G. E. 1991. A model of genetic variation for Pinus ponderosa in the Inland Northwest (U.S.A.): applications in gene resource management. Can. J. For. Res. 21:1491-1500. Rehfeldt, G. E. 1994. Adaptation of Picea engelmannii populations to the heterogeneous environments of the Intermountain West. Can. J. Bot. 72:1197-1208. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Richardson, B. A., S. J. Brunsfeld, and N. B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Mol. Ecol. 11:215-227. Rogers, D. L., C. I. Millar, and R. D. Westfall. 1999. Fine-scale genetic structure of whitebark pine (Pinus albicaulis): Associations with watershed and growth form. Evolution 53:74-90. Romme, W. H., and M . G. Turner. 1991. Implications of global climate change for biogeographic patterns in the Greater Yellowstone ecosystem. Conserv. Biol. 5:373-386. Sagnard, F., C. Barberot, and B. Fady. 2002. Structure of Genetic diversity in Abies alba Mill , from southwestern Alps: multivariate analysis of adaptive and non- adaptive traits for conservation in France. Forest Ecol. Manag. 157:175-189. SAS Institute, I. 1999. The SAS system for windows, in. SAS Institute, Inc., Cary, North Carolina. Spitze, K. 1993. Population structure in Daphnia obtusa: quantitative genetic and allozymic variation. Genetics 135:367-374. Squillace, A. E. 1974. Average genetic correlations among offspring from open-pollinated forest trees. Silvae Genet. 23:149-156. St. Clair, J. B., N . L. Mandel, and K. W. Vance-Borland. 2005. Genecology of Douglas fir in western Oregon and Washington. Ann. Bot.-London 96:1199-1214. Stuart-Smith, G. J. 1998. Conservation of whitebark pine in the Canadian Rockies: blister rust and population genetics. M . Sc. University of Alberta, Edmonton. Tani, N., N . Tomaru, M . Araki, and K. Ohba. 1996. Genetic diversity and differentiation in populations of Japanese stone pine (Pinus pumila) in Japan. Can. J. For. Res. 26:1454-1462. Tomback, D. F. 1978. Foraging strategies of Clark's nutcracker. Living Bird 16:123-161. Tomback, D. F. 1982. Dispersal of whitebark pine seeds by Clark's Nutcracker: a mutualism hypothesis. J. Anim. Ecol. 51:451-467. Tomback, D. F., S. F. Arno, and R. E. Keane. 2001. The compelling case for management intervention. Pages 3-25 inD. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Wang, T., A. Hamann, D. L. Spittlehouse, and S. N. Aitken. 2006. Development of scale-free climate data for western Canada for use in resource management. Int. J. Climatol. 26:383-397. Westfall, R. D. 1992. Developing seed transfer zones. Pages 313-398 in L. Fins, S. T. Friedman, and J. V. Brotschol, editors. Handbook of quantitative forest genetics. Kluwer Academic Publishers, Dordrecht, The Netherlands. Whitlock, M . C. 1999. Neutral additive genetic variance in a metapopulation. Genet. Res. 74:215-221. Yandell, U. G. 1992. An allozyme analysis of whitebark pine (Pinus albicaulis Engl.). M . Sc. University of Nevada, Reno. Zavarin, E., Z. Rafii, L. G. Cool, and K. Snajberk. 1991. Geographic monoterpene variability of Pinus albicaulis. Biochem. Syst. Ecol. 19:147-156. 51 C H A P T E R 3: M A T I N G S Y S T E M A N D I N B R E E D I N G DEPRESSION IN W H I T E B A R K PINE (Pinus albicaulis Engelm.)2 Andrew D. Bower and Sally N. Aitken 3.1 I N T R O D U C T I O N The mating system of most Pinus species is classified as mixed, with outcrossing at a rate typically above 90% (Schemske and Lande 1985, Ledig 1998, O'Connell 2003). Spatial separation of male and female strobili and phenological offset of pollen maturation and female receptivity reduces self pollination (Sorensen 1994). Self-incompatibility has not been documented in pines (Sorenson 1982) and the mating system is flexible enough to allow selling and mating among related individuals (Ledig 1998). In addition, outcrossing rate varies both among populations and among individuals within a stand (reviewed by Mitton 1992) and among years (El-Kassaby et al. 1993). Variation in outcrossing rate has also been reported in Siberian stone pine (Pinus sibirica Du Tour) and Korean stone pine (P. koraiensis Sieb. Et Zucc.) (Krutovskii et al. 1995, Potenko and Velikov 2001), related species in Pinus subsection Cembrae (Price et al. 1998; but see Gernandt et al. 2005). Both monozygotic and polyzygotic polyembryony is common in pines, and may reduce inbreeding depression by allowing selection against inbred embryos without sacrificing seed yield (Sorenson 1982, Ledig 1998). Selection against inbred individuals within cohorts over time has been observed in numerous studies of pines that have compared genotypes of embryos or seed with older cohorts (Farris and Mitton 1984, Cheliak et al. 1985, Plessas and Strauss 1986, El-Kassaby et al. 1987, Furnier et al. 1987, Muona et al. 1987, Starova et al. 1990, Morgante et al. 1993). While seed and seedling cohorts often have a deficiency of heterozygotes due to some selfing or consanguineous mating, genotype frequencies in older cohorts usually approach Hardy-Weinberg equilibrium or heterozygote excess. 2 A version of this chapter has been submitted for publication. Bower, A . D. and Aitken, S. N . Mating system and inbreeding depression in whitebark pine {Pinus albicaulis Engelm.). Submitted to Tree Genetics and Genomes 52 The seeds of whitebark pine (Pinus albicaulis Engelm.) are wingless and primarily dispersed by the Clark's nutcracker (Nucifragia columbiana Wilson), birds in the family Corvidae. The nutcracker harvests seeds from the ripe cones and caches from 1 to 15 seeds per cache in the ground for later use as its primary food source (Hutchins and Lanner 1982, Tomback 1982). Seed dispersal by the nutcracker has been shown to result in a structure of clumps that often contain multiple, related genets (Linhart and Tomback 1985, Furnier et al. 1987). These groups of closely related individuals may increase opportunities for consanguineous mating (Tomback and Linhart 1990, Krakowski et al. 2003). Whitebark pine has been reported to have a mixed mating system, with an outcrossing rate averaging tm = 0.73 (Krakowski et al. 2003), which is considerably lower than most wind-pollinated pines. This low outcrossing rate is attributed to the spatial clustering of related individuals, and may reflect evolution to tolerate inbreeding as a result of the spatial structure produced by this species' seed biology and dispersal by the Clark's Nutcracker. This outcrossing rate, however, was estimated from just two populations separated by only approximately 100km in the northern portion of the species range, and may not reflect differences among populations and regions. Family mean outcrossing rates reported by Krakowski et al. (2003) had nearly a uniform distribution with individual tree outcrossing rates ranging from near zero to one, in contrast to the j-shaped distribution usually found in predominantly outcrossing conifers (El-Kassaby et al. 1987, El-Kassaby et al. 1993, Krutovskii et al. 1995, Rajora et al. 2002). While it is known that whitebark pine experiences some selfing and consanguineous mating, the magnitude of inbreeding depression is unknown. Inbreeding depression is studied by comparing relative fitness of inbred and outcrossed progeny. In plants this is usually accomplished using selfed and outcrossed progeny from controlled crosses involving the same seed parents where the associated inbreeding coefficients (F) are 0.5 and 0, respectively (Charlesworth and Charlesworth 1987, Williams and Savolainen 1996, Keller and Waller 2002). Inbreeding coefficients for other levels of relatedness can be determined if the pedigree is known. In the wild, where pedigrees are not known, inbreeding coefficients must be determined by genetic analysis (Hedrick and Kalinowski 2000), 53 where a heterozygote deficiency can be used as an indicator of inbreeding. Inbreeding in the Pinaceae results in reduced seed set and germination, as well as reductions in growth, survival, and reproduction (Sorensen and Miles 1982, Geburek 1986, Durel et al. 1996, Sorensen 1999, Wang et al. 1999, Woods et al. 2002, Johnsen et al. 2003, and reviewed by Schemske and Lande 1985, Charlesworth and Charlesworth 1987, Williams and Savolainen 1996). Species with long histories of small effective population sizes are expected to show lower inbreeding depression, whereas widespread, outcrossing species are expected to have substantial inbreeding depression (Lande and Schemske 1985). Historically large, predominantly outcrossing populations are expected to harbour a higher frequency of recessive lethal and sub-lethal alleles than smaller populations. Lethal and near-lethal alleles often result in early inbreeding depression in conifers (i.e., embryo abortion or mortality), while moderately deleterious alleles tend to affect quantitative characters such as growth and fecundity at later life stages (Sorensen 2001). Low levels of inbreeding such as sib-mating are predicted to be effective in purging deleterious genes of large effect, as homozygous recessive genotypes are exposed to selection (Fu et al. 1998). This suggests that whitebark pine may experience less inbreeding depression than other wind-pollinated conifers due to chronic, moderate inbreeding and often small effective population size. There are both theoretical grounds and empirical evidence to suggest that inbreeding can reduce the ability of a population to cope with disease (Frankham 2002, Altizer et al. 2003), and inbreeding depression in disease resistance has been reported in both animals and plants (Carr et al. 2003, Kumar 2004, Spielman et al. 2004, Stephenson et al. 2004, but see Matheson et al. 1995). Whitebark pine has been severely impacted throughout its range by the introduced disease white pine blister rust (caused by the fungus Cronartium ribicola J.C. Fisch. ex Rabh.) (Campbell and Antos 2000, McDonald and Hoff 2001, Zeglen 2002). The implications of inbreeding and the possible resulting inbreeding depression for fitness and survival of whitebark pine in the face of blister rust are unknown. The relationship, or lack thereof, between heterozygosity and fitness has been debated extensively and the literature on this topic is voluminous (Britten 1996, David 1998). Most positive 54 relationships between fitness and heterozygosity may be just a measure of inbreeding depression due to recessive deleterious alleles in homozygotes (Ledig et al. 1983, Savolainen and Hedrick 1995). While there is a large body of literature on the positive association between heterozygosity at putatively neutral loci and fitness components, it is not clear how accurately the loci used cumulatively represented genomic heterozygosity (Pamilo and Palsson 1998, Slate and Pemberton 2002). Rather than comparing heterozygosity and growth to study the effects of inbreeding, it may be more informative to compare parental outcrossing rate with the performance of a progeny array. For example, outcrossing rate of seed parents is positively correlated with growth of progeny in two species of eucalyptus (Eucalyptus grandis Hill ex Maid, and E. camaldulensis Dehnh.) and in lodgepole pine (Pinus contorta Dougl.) (Burgess et al. 1996; Sorensen 2001; Butcher and Williams 2002). The goals of this study are to 1) confirm the mating system and distribution of family level outcrossing rates of whitebark pine, and 2) to determine whether there is a relationship between inbreeding coefficients calculated from the parental outcrossing rate, and progeny performance in multiple quantitative traits. In addition, we have used the estimated inbreeding coefficient to quantify the level of inbreeding depression in these traits. 3.2 M A T E R I A L S A N D M E T H O D S 3.2.1 Sample Materials Seeds from 94 families from 7 populations in three geographic regions (Oregon, Montana, and southern British Columbia) (Table 3.1) were germinated in 2002 following the stratification protocol described by Burr et al. (2001) and sown individually into individual 10 in 3 Ray Leach "Cone-tainer" super cells (Steuwe and Sons, Corvallis, Oregon) for their first growing season. In November 2002, seedlings from 72 of these families (Table 3.1) were transplanted into a raised bed common garden in Vancouver, British Columbia (49° 13', 123° 6') and maintained for two growing 55 seasons. These families were part of a near range-wide collection of seed sown in a common garden experiment to assess genetic variation in quantitative traits (Chapter 2). Seedlings were planted in an incomplete block Alpha design (Patterson and Williams 1976) at 9.5 x 10 cm spacing in 12 total replications. The AlphaPlus program (CSJJR.O 1996) was used to design the planting layout and assign seedlings randomly within replications in two temperature treatments. Four replications had cooled water pumped through hoses buried approximately 25 cm below the surface to cool soil (cold treatment) and the remaining eight replications had ambient soil conditions (ambient treatment). Soil temperatures in the cold treatment were consistently ~8°C lower than in the ambient treatment during the warmest part of the day. Timing of initiation of growth in the spring was recorded for the 2003 growing season as the date that elongating needles were visibly separated. At the end of the 2004 growing season, survival, second year height growth increment, and total oven-dry biomass were measured all replications. Families represented by fewer than three seedlings were excluded from the analysis of inbreeding depression, leaving a total of 55 families. Forty-six families were included in the ambient treatment, 26 in the cold treatment, and 17 families in both treatments. The mean number of seedlings per family in the inbreeding depression analysis was 5.7 (SD = 1.2) for the ambient treatment, and 4.4 (SD = 1.8) for the cold treatment. 3.2.2 Isozyme analysis Maternal and offspring genotypes of individuals from five populations in two other geographic regions (Oregon and Montana) were determined from excess seeds. Seeds were dissected, and embryos (2N) and megagametophytes (N) stored separately at -80°C until used. Tissues were ground in a buffer slightly modified from Mitton et. al. (1977) and the supernatant absorbed onto 3 x 15 mm Whatman filter paper wicks. Samples were analyzed using electrophoresis of gels of 10.5% starch and 7.5% sucrose (w/v). Two buffer and electrode systems were used to analyze 9 loci following the methods of Conkle et al. (1982): 1) sodium borate (pH 8.0) (Conkle et al. system B) was used for AAT 1,3 (E.C. 2.6.1.1); and 2) morpholine citrate (pH 8.0) (Conkle et al. system D) was 56 used for IDH(E.C. 1.1.1.42), MDH3, 4 (B.C. 1.1.1.37), SKDH 1,2 (E.C. 1.1.1.25), and 6PGH 1,2 (E.C. 1.1.1.44). Both haploid megagametophyte and diploid embryo tissues were scored for all loci with the exception of AAT2, which resolved differently for the embryos and megagametophytes, presumably due to the expression of different loci. For this locus, only the diploid embryo data were used in the analysis. A sample of red pine {Pinus resinosa Ait.), an invariant homozygous species (Ledig et al. 2000), was run on each gel as a standard to aid in interpretation of loci. Both systems were run initially at 60 mV, with the wicks in place for 15 minutes, then removed. Gels with the sodium borate buffer were run at 200mV, and those with the morpholine citrate were run at 180mV for 4-5 h, until a dye marker migrated to within 2 cm of the end of the gel. 3.2.3 Data Analysis No consistent patterns of linkage disequilibrium have been reported for the loci used in this study in seeds or buds of whitebark pine (Furnier et al. 1986, Krakowski et al. 2003), so all loci were used in the genetic analysis. The latest version (V 3.1) of the MLTR program (Ritland 2002) was used to analyze the genotypic data. This program provides population-level estimates of both single and multilocus outcrossing rates (ts and tm), the multilocus correlation of paternity (rp) (the probability that a randomly chosen pair of progeny from the same mother shared the same father), and parental inbreeding coefficient based on inferred maternal genotype (Fp). It also estimates standard errors based on 1000 Newton-Raphson iterations and uses a method-of-moments to estimate individual tree outcrossing rates (Ritland 2002). Standard errors were used to calculate 95% confidence intervals around the estimated variables as ±2 standard errors of the means. A sample size of approximately 400 progeny is recommended for good population level estimates of outcrossing (standard error of 0.03-0.05) (Ritland 2002). Therefore, data from all populations within a region were combined in order to have samples sizes sufficient to obtain accurate estimations of outcrossing rate at the regional level. 5 7 Genotype data from two populations in southern British Columbia have already been analyzed using an older version of MLTR, and the details of electrophoresis and data analysis are given in Krakowski et al. (2003). The raw data for these two populations, provided by the authors, was reanalyzed using the updated version of MLTR which provides estimates of individual outcrossing rates using a procedure which is especially designed for small sample sizes which yields results that are less biased and have a smaller sampling error than previous versions of the program (Ritland 2002). The expected equilibrium inbreeding coefficient for each family was calculated following (Allardetal. 1968) as: where Fe is the estimated family mean equilibrium inbreeding coefficient of the offspring, and tm is the family multilocus outcrossing rate estimated using MLTR. This calculates the expected inbreeding coefficient assuming no inbreeding depression. The Fstat program (Goudet 1995) was used to calculate the fixation index of progeny (F0) as F = 1 - ( H o / H e ) (Wright 1951) (where Ff0 and H e are observed and expected heterozygosity, respectively) and 95% confidence intervals, for comparison with F p. The SAS system version 8 (SAS Institute 1999) was used for all statistical analysis. PROC REG was used to regress the family means of the log-transformed quantitative traits on Fe to test for and determine the magnitude of inbreeding depression for each of the three regions separately (Charlesworth and Charlesworth 1987, Keller and Waller 2002) as: where 5 is the proportional reduction in the trait mean of offspring due to inbreeding (inbreeding depression), and B is the slope of family mean log-transformed quantitative traits regressed on the mean inbreeding coefficient (Fe) for the families included in the common garden experiment. ( i - O [1] S = l-e [2] 58 3.3 R E S U L T S 3.3.1 Inbreeding coefficients Only the Oregon region had a mean parental inbreeding coefficient (Fp) significantly different from zero (Table 3.2). The difference was in the direction of an excess of heterozygotes (Fp<0) (Figure 3.1). In the progeny, inbreeding coefficient (F0) indicated a heterozygote deficiency in all three regions, but the difference from zero was significant only in the Oregon and southern B.C. regions (Figure 3.1). 3.3.2 Outcrossing rates At the regional level t„, was significantly different from unity (/„, = 1) for all three regions and mean single locus outcrossing rate (ts) was significantly different from unity for the Oregon and Southern B.C. regions (Table 3.2). The difference between /„, and ts was significantly different from zero only for the Southern B.C. region (Table 3.2), indicating some biparental inbreeding in these populations. In the Oregon and Montana populations inbreeding is primarily due to selfing. Across all regions, mean tm was 0.86. Multilocus estimates of the correlation of paternity (rp) were significantly different from zero only for southern B.C (Table 3.2). Values of rpm the southern B.C. region were positive, which indicates that in this part of the range there is more family structure within stands, as two randomly chosen individuals from the same maternal open-pollinated family had a probability of sharing a male parent of 0.162. In addition, there was a relatively low effective number of pollen donors (iVep = l/r p = 6.2, see Ritland 1989).' In Oregon and Montana, the probability that individuals within the same family share the same father is not different from that expected by chance (rp = 0). This indicates that in these regions there must be a large number of individuals contributing to a well-mixed pollen pool. A previous report of family mean tm rates in whitebark pine showed a nearly uniform distribution over the full range of possible values (Krakowski et al. 2003). In using the new version of MLTR, we found that the distribution of individual tm estimates, even using data from the same 59 two populations previously analyzed, was skewed towards higher levels of outcrossing compared to the previous study. Over all regions, rates for the majority of individuals (63%) were greater than or equal to 0.80. A considerable proportion of individuals (31%), had rates between 0.5 and 0.79, while 6% had rates below 0.49, and two individuals less than 0.10 (Figure 3.2). Of the six individuals with tm estimates below 0.50, five were from the Southern BC region, and all but two individuals from the Oregon region had rates over 0.70. Family mean outcrossing rates covered a large range from less than 0.1 to over 1.0. Possible reasons for t„, being greater then 1.0 are discussed by El-Kassaby et al. (1987) and include negative assortative mating caused by differences in phenology among and within trees, patchy allelic distribution due to non-random pollination patterns, and sampling error. While an outcrossing rate greater than 1.0 is biologically impossible, there is an error associated with each of these estimates and values greater than one indicate a greater likelihood that the progeny are fully outcrossed (Burgess et al. 1996). 3.3.3 Inbreeding depression Total biomass in the southern B.C. region was the only trait that exhibited measurable inbreeding depression. It was the only trait whose regression on expected inbreeding coefficient (Fe) was significantly different from zero (Table 3.3, Figure 3.3). In the cold temperature treatment, none of the traits were significantly related to Fe in any of the regions. Mean Fe over all southern B.C. families included in the common garden experiment was 0.252 (SD 0.064) and the mean predicted reduction in biomass due to inbreeding (8) from equation [2] was -0.196. Therefore, biomass is predicted to be reduced by 19.6% in this population due to inbreeding. 60 3.4 DISCUSSION 3.4.1 Outcrossing Rate and Inbreeding Coefficients Using the updated version of MLTR and including populations from other parts of the range of whitebark pine, we have confirmed the mixed mating system reported by Krakowski et al. (2003). Our estimated mean tm values were nearly identical to previously published values for the southern B.C. populations, but we have estimated a higher average outcrossing rate across all populations (mean t„, = 0.86 in the current study vs. 0.73 in Krakowski et al. 2003). Our regional tm estimates were similar to those reported for P. cembra L. but slightly lower than for P. sibirica and P. koraiensis, three of the four other stone pines (Pinus subsection Cembrae) and were slightly lower than the average value reported for 28 Pinus species (O'Connell 2003) (Table 3.4). Multi-genet tree clusters are rare in P. sibirica and P. koraiensis (Farjon 2005, K. Krutovskii and D. Politov pers. comm.), but are common in P. cembra (Tomback et al. 1993). This suggests that growth form may play a role in the level of outcrossing, with single trees having higher, outcrossing rates than tree clusters, due to relatedness among stems within clusters. Tree clusters are common in whitebark pine (Linhart and Tomback 1985, Furnier et al. 1987), however, no information was available on whether seed sampled in this study came from single trees or clusters. Due to changes in MLTR, large differences in family mean tm were noted between this study and values reported by Krakowski et al. (2003). It appears that family mean tm was often underestimated in the previous study, altering the distribution of tm among families. We also found considerable variation in family mean tm; however, the distribution was closer to a skewed normal (Figure 3.2) as opposed to the more uniform distribution reported earlier. In conifers, substantial differences have been reported in t,„ among populations (Krutovskii et al. 1995, O'Connell et al. 2001, Ledig et al. 2002, Potenko 2004), families (El-Kassaby et al. 1987, Ledig et al. 2000, Lewandowski and Burczyk 2000) and even crown positions (El-Kassaby et al. 1993). We found variation in /„, among geographic regions (Table 3.2) and individuals within a 61 population (Figure 3.2). Mean tm estimates for individual seed parents ranged from 0.015 to over 1.0, with 57.4% of the families greater than the overall mean of 0.846. Southern B.C. populations have a lower tm than populations from Oregon and Montana. In addition, in southern B.C. there is family substructuring, resulting in some biparental inbreeding (tm-ts > 0), and seeds from within the same open-pollinated family are more likely to share fathers than expected under random mating (rp = 0.162). The fixation index has been shown to increase with increasing latitude in whitebark pine (Krakowski et al. 2003), possibly due to patterns of postglacial recolonization. The area where the southern B.C. populations were sampled was glaciated at the last glacial maximum, while the Oregon and Montana populations were not, and there is evidence that the northern part of whitebark pine's range was recolonized from southern refugia (Richardson et al. 2002). Therefore, the lower t,„ in southern B.C. may reflect recolonization patterns and processes. Genetic diversity in these populations may have been reduced due to a bottleneck resulting from small population size of founder populations. Subsequently, individuals within these populations may have evolved to tolerate higher levels of inbreeding relative to individuals in non-glaciated populations. The parental generation either did not significantly differ from Hardy-Weinberg equilibrium (Montana and southern B.C.) or have an excess of heterozygotes (Oregon) (Figure 3.1). Comparisons of the parental (Fp) and offspring (F0) fixation indices show that F 0 was higher than F p in all regions. It is common in conifers for populations of mature trees to show a heterozygote excess while seeds have a heterozygote deficiency (Ledig et al. 2002). This phenomenon has been reported for a number of other pines, including the four other stone pine species (Krutovskii et al. 1995). The heterozygote deficiency in seeds is most likely due to inbreeding, and its disappearance by maturity suggests that there is selection against inbreds during the life cycle of the tree (Krutovskii et al. 1995). Our results reveal that this pattern is also evident in whitebark pine. 62 3.4.2 Inbreeding Depression The equilibrium inbreeding coefficient (Fe) is calculated under the assumption of no inbreeding depression. If inbreeding depression is present, then true inbreeding coefficient values (F) are lower than Fe. When true F (determined from assays of mature plants of inferred maternal genotypes) andF e calculated from estimated outcrossing rates were compared, a majority (88% of 64 plant species with mixed mating systems) were less inbred than expected (i.e., F< Fe) (Goodwillie et al. 2005). Therefore, the Fe values used to quantify inbreeding depression in this study may be overestimated and the calculated value of 8 may be viewed as an upper limit. Significant inbreeding depression was only detected in seedlings from southern B.C. growing in the ambient treatment. This region had the lowest outcrossing rate of those studied, as well as significant evidence of biparental inbreeding. Karkkainen et al. (1996) also found that selling may be more common in northern than in southern populations of Pinus sylvestris in Finland. However, they found lower levels of early inbreeding depression (% empty seeds) and fewer embryonic lethals in northern than in more southern populations. The lower level of early inbreeding depression was given as an explanation for the higher apparent self-fertilization rate, as inbred individuals had not been removed by selection. The lower /„, in whitebark pine in southern B.C. is compatible with the idea that reduced population sizes during post-glacial recolonization resulted in some purging of deleterious alleles affecting embryo survival. ^However, measurable inbreeding depression in biomass • i may indicate that deleterious alleles affecting'post-embryonic life stages may not have been so well purged and that the higher selfing rate in this region results in their expression. The level of inbreeding depression estimated for biomass (-20%) is within the range of values reported for other conifers at this level of inbreeding. Reported values of inbreeding depression in stem volume of other conifers at F=0.5 ranges from 29% at age four in Pinus radiata D. Don at (Wilcox 1983) to 66% and 44% at ages 11 and 14 for P. sylvestris L. (Lundkvist et al. 1987) to over 70% in Pseudotsuga menziesii Mirb. Franco and Abiesprocera Rehd., and nearly 80% at age 25 for P. ponderosa P. & C. Lawson (Sorensen 1999); and at F=0.25, reductions of around 20% were 63 predicted in P. menziesii at 90 years (Wang et al. 2004). Inbreeding depression in height growth has been reported for a number of tree species (see references in Sorensen and Miles 1982, Durel et al. 1996, Sorensen 1999) but no effect on height growth in seedlings was detected in this study. Whitebark pine is a slow growing relatively short-statured species, and a higher growth rate is not likely to lead to a selective advantage over other species. Survival depends on adaptation to withstand abiotic stresses rather than outcompeting other species, or it may be manifested at other times in life history (e.g., after reproductive maturity is reached). The lack of a significant relationship between height growth and Fe in this study may reflect the weaker role of height growth in overall fitness in this species. It may also be due to the fact that third year height growth increment was low, with a large range (mean = 8.8mm, range 0-81mm, data not shown) and high measurement error relative to the amount of growth. In addition, the maternal effect of seed size may be confounded with genetic differences in early height growth (Sorensen 2001). In contrast to the ambient treatment, no inbreeding depression was detected in any of the traits measured in the cold soil treatment. Whitebark pine is restricted to upper subalpine forests and ridge crests that experience cold, windy, snowy conditions, with cool, short summers where light frosts and snowfalls can occur in any month of the year (Weaver 2001). The environment where these seedlings were grown (Vancouver, B.C.) is warmer and is relatively benign for many tree species. However, for a species adapted to a cold, harsh environment, it appears that this warm environment was more stressful. Seedlings grown in the cold soil treatment had higher mean growth increment and survival than in the ambient treatment (Chapter 2), and appeared healthier with darker green foliage. Inbreeding depression has been shown to be environmentally dependent in some cases, and may be most strongly expressed in sub-optimal environments (El-Kassaby 1999, Hedrick and Kalinowski 2000, Armbruster and Reed 2005, Marr 2005), which in this study was the ambient treatment. Seed dispersal by the Clark's nutcrackers produces a clumpy distribution of related individuals, which results in a higher level of inbreeding in whitebark pine than other wind-pollinated 64 conifers (Linhart and Tomback 1985, Tomback and Schuster 1994, Krakowski et al. 2003). The failure to detect inbreeding depression for all traits in the Oregon and Montana regions and all but one trait in the southern B.C. region in a moderately stressful environment may indicate that whitebark pine has evolved to withstand the level of inbreeding that it experiences through the purging of deleterious alleles. However, there are a number of other possible reasons why inbreeding depression was not detected. Inbreeding depression may be weaker in whitebark pine for traits like height growth and biomass that are not directs measures of lifetime fitness or its components, such as seed germination. The selfing rate estimated for viable seeds or germinant seedlings is often lower than the primary rate of selfing at the time of fertilization, due to embryo mortality during seed development (Sorensen 1969, Husband and Schemske 1996, Williams and Savolainen 1996). Poiyembryony in whitebark pine was observed during germination in this study, as multiple embryos were noted in some seeds, and in rare instances two or more of the embryos germinated. The frequency of embryo competition and its effects on inbreeding depression in whitebark pine is unknown. Beyond the seed germination stage, selection against inbred individuals through mortality of germinants may have occurred before planting, in which case they would not be included in the estimate of survival. Individuals that died after planting would be accounted for in the estimates of 8 for survival, but not for other metric traits. If individual seedlings within a family differed in F, there was opportunity for selection against more inbred individuals prior to any measurement, which would result in 8 estimates biased downwards for these traits. The difference between F p and F 0 shows selection against inbred individuals, but we did not detect inbreeding depression in early survival. Therefore, if it occurs, inbreeding depression must either be early acting, possibly affecting seed development, germination, and emergence, or late acting, possibly affecting fecundity or tolerance to abiotic stress later in life. 65 3.5 A C K N O W L E D G E M E N T S We thank the USDA Forest Service regions 1, and 5, 6, and E. C. Manning Provincial Park of British Columbia, for providing seed for this study. Dr. Carol Ritland provided assistance in the isozyme lab, and Dr. Kermit Ritland provided assistance with MLTR. Many other people helped in various aspects of this project, including: Joanne Tuytel, Christine Chourmouzis, Dorothy Watson, Megan Harrison, Dane Szohner, Jodie Krakowski, and all of the members of the Aitken lab at UBC. Isozyme data for the southern B.C. region were provided by Dr. Yousry El-Kassaby. Funding for this study came from the British Columbia Forestry Investment Account through the Forest Genetics Council of B.C. to the Centre for Forest Gene Conservation at UBC. Thank you to Drs. Jeannette Whitton, Mike Whitlock, Alvin Yanchuk, Yousry El-Kassaby, Sean Graham, and Diana Tomback for their helpful comments on an earlier draft of this manuscript. 66 Table 3.1. Number of provenances (Prov.), families (Fam.), total seedlings genotyped, and mean number of seedlings genotyped by family using isozyme analysis and number of families grown in two temperature treatments (amb. and cold) in a common garden (CG) experiment for three sampling regions. Region # Prov. #Fam. Total Mean # #Fam. #Fam. # seedlings genotyped by C G amb. C G cold genotyped family (SD) Oregon 3 20 413 20.6(1.7) 16 8 Montana 2 19 308 16.2 (4.9) 12 14 Southern B. C. 2 55 1603 29.1 (3.3) 18 4 Table 3.2. Population mean estimates of multi- and single-locus outcrossing rates (t„, and ts), mean parental inbreeding coefficient (Fp) and multilocus correlation of paternity (rp) with standard errors in parentheses. Region tm ts rP(m> Oregon 0.90* (0.04) 0.87* (0.05) 0.04 (0.03) -0.16* (0.06) -0.06 (0.04) Montana 0.93* (0.03) 0.94 (0.04) -0.01 (0.03) 0.03 (0.14) -0.06 (0.04) Southern B.C. 0.73* (0.03) 0.68* (0.04) 0.06* (0.02) 0.03 (0.07) 0.16* (0.03) * Significant at a = 0.05 Table 3.3. Regression line slope (inbreeding load) for family mean of log-transformed seedling traits on Fe in two soil temperature treatments for 18 families from southern B.C. Trait Ambient Treatment Cold Treatment Slope p-value Slope p-value Ln(Height inc.) -0.06 0.86 -0.28 0.78 Ln(Biomass) -0.71 0.02* -0.25 0.82 Ln(Root:shoot ratio) 0.25 0.20 0.68 0.10 Ln(Survival) -0.28 0.25 -0.003 0.9937 * Significant at a = 0.05 67 Table 3.4. Outcrossing rates of four stone pine species (Pinus subsection cembrae) and mean outcrossing rate of subsection cembrae and a sample of genus Pinus. Species #pop t Range Reference P. albicaulis 0.729 0.722-0.736 Krakowski et al. 2003 P. albicaulis 7 0.863 0.732-0.978 This study P. cembra 1 0.808 Lewandowski and Burczyk 2000 P. cembra 1 0.686 Krutovskii etal. 1995 P. koraiensis 10 0.909 0.751-1.031 Potenko 2004 P. koraiensis 3" 0.974 0.920-1.034 Krutovskii et al. 1995 P. sibirica 9 0.894 0.817-0.980 Krutovskii et al. 1995 M E A N S SD Subsection Cembrae 4 spp. 0.861 0.083 grand mean of above Genus Pinus 28 spp. 0.878 0.124 O'Connell 2003 a populations included in this study 68 Figure 3.1. Regional mean estimates of observed parental and offspring inbreeding coefficient (Fp and Fp)- Error bars are ± 2 standard errors. 0.4 0.3 c 0 0.2 'o 0) o 0.1 o D) c 0 TJ 0) o .Q -0.1 _c -0.2 -0.3 • Fp O r e g o n M o n t a n a S o u t h e r n B . C . Figure 3.2. Distribution of multilocus outcrossing rate by region. 12 10 </> Q <D O m 6 Z 4 2 0 a S o u t h e r n B C • Montana O O r e g o n 0 *9 # # «P # # # A* # <£> # ^ o? c v N r > ^ ^ ^ ^ JV or \ . V ,<V ^ * ^ - ^ - ^ * ^ - ^ - ^ ' ^ - V V Mult i locus Outc ross ing Rate 69 Figure 3.3. Family mean ln(biomass) vs. expected equilibrium inbreeding coefficient (Fe) for three regions with regression line for southern B.C. region. w E o m c 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 O -0.2 0.2 0.4 O 0.6 R 2 = 0.28 p = 0.02 0.8 O Southern B C • Montana • Oregon 70 3.6 REFERENCES Allard, R. W., S. K. Jain, and P. L. Workman. 1968. The genetics of inbreeding populations. Adv. Genet. 14:55-131. Altizer, S., D. Harvell, and E. Friedle. 2003. Rapid evolutionary dynamics and disease threats to biodiversity. Trends Ecol. Evol. 18:589-596. Armbruster, P., and D. H. Reed. 2005. Inbreeding depression in benign and stressful environments. Heredity 95:235-242. Arno, S. F., and R. J. Hoff. 1989. Silvics of whitebark pine (Pinus albicaulis), Gen. Tech. Rep. INT-GTR-253. U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Ogden, UT. Britten, H. B. 1996. Meta-analyses of the association between multilocus heterozygosity and fitness. Evolution 50:2158-2164. Burgess, I. P., E. R. Williams, J. C. Bell, C. E. Harwood, and J. V. Owen. 1996. The effect of outcrossing rate on the growth of selected families of Eucalyptus grandis. Silvae Genet. 45:97-100. Burr, K. E., A. Eramian, and K. Eggleston. 2001. Growing whitebark pine seedlings for restoration. Pages 325-345 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. Butcher, P. A., and E. R. Williams. 2001. Variation in outcrossing rates and growth in Eucalyptus camaldulensis from the Petford Region, Queensland: evidence of outbreeding depression. Silvae Genet. 51:6-12. Campbell, E. M. , and J. A. Antos. 2000. Distribution and severity of white pine blister rust and mountain pine beetle on whitebark pine in British Columbia. Can. J. For. Res. 30:1051-1059. Carr, D. E., J. F. Murphy, and M . D. Eubanks. 2003. The susceptibility and response of inbred and outbred Mimulus guttatus to infection by Cucumber mosiac virus. Evol. Ecol. 17:85-103. Charlesworth, D., and B. Charlesworth. 1987. Inbreeding depression and its evolutionary consequences. Ann. Rev. Ecol. Syst. 18:237-268. Cheliak, W. M. , B. P. Dancik, K. Morgan, F. C. Yeh, and C. Strobeck. 1985. Temporal variation of the mating system in a natural population of jack pine. Genetics 109:569-584. Conkle, M . T., P. D. Hodgskiss, L. B. Nunnally, and S. D. Hunter. 1982. Starch gel electrophoreses of conifer seeds: a laboratory manual. Pacific Southwest Forest and Range Experiment Station, Forest Service, U.S. Department of Agriculture, Berkeley, CA Gen. Tech. Rep. PSW-64. CSIRO. 1996. AIpha+; Experimental designs for variety trials and many-treatment experiments. Canberra, Australia. David, P. 1998. Heterozygosity-fitness correlations: new perspectives on old problems. Heredity 80:531-537. Durel, C. E., P. Bertin, and A. Kremer. 1996. Relationship between inbreeding depression and inbreeding coefficient in maritime pine (Pinuspinaster). Theor. Appl. Genet. 92:347-356. El-Kassaby, Y. A. 1999. Phenotypic plasticity in western redcedar. Forest Genet. 6:235-240. El-Kassaby, Y. A., M . D. Meagher, and R. Davidson. 1993. Temporal variation in the outcrossing rate in a natural stand of western white pine. Silvae Genet. 42:131-135. El-Kassaby, Y. A., M . D. Meagher, J. Parkinson, and F. T. Portlock. 1987. Allozyme inheritance, heterozygosity and outcrossing rate among Pinus monticola near Ladysmith, British Columbia. Heredity 58:173-181. Farris, M . A., and J. B. Mitton. 1984. Population density, outcrossing rate, and heterozygote superiority in ponderosa pine. Evolution 38:1151-1154. Farjon, A. 2005. Pines: drawings and descriptions of the genus Pinus. Brill Academic Publishers, Leiden, The Netherlands. Frankham, R. 2002. Introduction to conservation genetics. Cambridge University Press, Cambridge, UK. 71 Fu, Y. B., G. Namkoong, and J. E. Carlson. 1998. Comparison of breeding strategies for purging inbreeding depression via simulation. Conserv. Biol. 12:856-864. Furnier, G. R., P. Knowles, M . A. Aleksiuk, and B. P. Dancik. 1986. Inheritance and linkage of allozymes in seed tissue of whitebark pine. Can. J. Genet. Cytol. 28:601-604. Furnier, G. R., P. Knowles, M . A. Clyde, and B. P. Dancik. 1987. Effects of avian seed dispersal on the genetic structure of whitebark pine populations. Evolution 41:607-612. Geburek, T. 1986. Some results of inbreeding depression in Serbian spruce (Picea omorika (Pane.) Purk.). Silvae Genet. 35:169-172. Gernandt, D. S., G. Geada Lopez, S. Ortiz Garcia, and A. Liston. 2005. Phylogeny and classification of Pinus. Taxon 54:29-42. Goodwillie, C , S. Kalisz, and E. C. G. 2005. The evolutionary enigma of mixed mating systems in plants: occurrence, theoretical explanations, and empirical evidence. Ann. Rev. Ecol. Syst. 36:47-79. Goudet, J. 1995. Fstat version 1.2: a computer program to calculate Fstatistics. J. Hered. 86:485-486. Hedrick, P. W., and S. T. Kalinowski. 2000. Inbreeding depression in conservation biology. Ann. Rev. Ecol. Syst. 31:139-162. Husband, B. C , and D. W. Schemske. 1996. Evolution of the magnitude and timing of inbreeding • depression in plants. Evolution 50:54-70. Hutchins, H. E., and R. M. Lanner. 1982. The central role of Clark's Nutcracker in the dispersal and establishment of whitebark pine. Oecologia 55:192-201. Johnsen, K. H., J. E. Major, and C. A. Maier. 2003. Selfing results in inbreeding depression of growth but not of gas exchange of surviving adult black spruce trees. Tree Physiol 23. Karkkainen, K., V. Koski, and O. Savolainen. 1996. Geographical variation in the inbreeding depression of Scots pine. Evolution 50:111-119. Keller, L. F., and D. M . Waller. 2002. Inbreeding effects in wild populations. Trends Ecol. Evol. 17:230-241. Krakowski, J., S. N. Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. Krutovskii, K. V., D. V. Politov, and Y. P. Altukhov. 1995. Isozyme study of population genetic structure, mating system and phylogenetic relationships of the five stone pine species (subsection cembrae, section strobi, subgenus strobus). Pages 279-304 in P. Baradat, W. T. Adams, and M.-S. G., editors. Population genetics and genetic conservation of forest trees. SPB Academic Publishing, Amsterdam, The Netherlands. Kumar, S. 2004. Effect of selfing on various economic traits in Pinus radiata and some implications for breeding strategy. Forest Sci. 50:571-578. Lande, R., and D. W. Schemske. 1985. The evolution of self-fertilization and inbreeding depression in plants. I. Genetic models. Evolution 39:24-40. Ledig, F. T. 1998. Genetic variation in Pinus. Pages 251-280 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Ledig, F. T., B. Bermejo-Velazquez, P. D. Hodgskiss, D. R. Johnson, C. Flores-Lopez, and V. Jacob-Cervantes. 2000. The mating system and genetic diversity in Martinez spruce, an extremely rare endemic of Mexico's Sierra Madre Oriental: an example of facultative selfing and survival in interglacial refugia. Can. J. For. Res. 30:1156-1164. Ledig, F. T., R. P. Guries, and B. Bonefeld. 1983. The relation of growth to heterozygosity in pitch pine. Evolution 37:1227-1238. Ledig, F. T., P. D. Hodgskiss, and V. Jacob-Cervantes. 2002. Genetic diversity, mating system, and conservation of a Mexican subalpine relict, Picea mexicana Martinez. Conserv. Genet. 3:113-122. Lewandowski, A., and J. Burczyk. 2000. Mating system and genetic diversity in natural populations of European larch (Larix decidua) and stone pine (Pinus cembrd) located at higher elevations. Silvae Genet. 49:158-161. 72 Linhart, Y. B., and D. F. Tomback. 1985. Seed dispersal by nutcrackers causes multi-trunk growth form in pines. Oecologia 67:107-110. Little, E. L. 1971. Atlas of United States trees, volume 1, conifers and important hardwoods. USDA For. Serv. Misc. Publ. 1146 Lundkvist, K., G. Eriksson, L. Norell, and I: Ekberg. 1987. Inbreeding depression in two field trials of young Pinus sylvestris (L.). Scand. J. For. Res. 2:281-290. Marr, A. B. 2005. Conservation genetics of small populations. University of British Columbia, Vancouver, B.C. Matheson, A. C , T. L. White, and G. R. Powell. 1995. Effects of inbreeding on growth, stem form and rust resistance in Pinus elliottii. Silvae Genet. 44:37-46. McDonald, G. I., and R. J. Hoff. 2001. Blister rust: and introduced plague. Pages 193-220 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Mitton, J. B. 1992. The dynamic mating systems of conifers. New Forest. 6:197-216. Mitton, J. B., Y. B. Linhart, S. W. Sturgeon, and J. P. Farrar. 1977. Protein polymorphisms detected in mature needle tissue of ponderosa pine. Am. J. Bot. 42:1154-1157. Morgante, M. , G. G. Vendramin, P. Rossi, and A. M . Olivieri. 1993. Selection against inbreds in early live-cycle phases in Pinus leucodermis Ant. Heredity 70:622-627'. Muona, O., R. Yazdani, and D. Rudin. 1987. Genetic change between life stages in Pinus sylvestris: allozyme variation in seeds and planted seedlings. Silvae Genet. 36:39-42. O'Connell, L. M . 2003. The evolution of inbreeding in western redcedar (Thuja plicata: Cupressaceae). University of British Columbia, Vancouver, B.C. O'Connell, L. M. , F. Viard, J. Russell, and K. Ritland. 2001. The mating system in natural populations of western redcedar (Thuja plicata). Can. J. Bot. 79:753-756. Pamilo, P., and S. Palsson. 1998. Associative overdominance, heterozygosity and fitness. Heredity 81:381-389. Patterson, H. D., and E. R. Williams. 1976. A new class of resolvable incomplete block designs. Biometrika 63:83-92. Plessas, M . E., and S. H. Strauss. 1986. Allozyme differentiation among populations, stands, and cohorts in Monterey pine. Can. J. For. Res. 16:1155-1164. Potenko, V. V. 2004. Genetic diversity and mating system of Korean pine in Russia. Pages 259 in R. A. Sniezko, S. Samman, S. E. Schlarbaum, and H. B. Kreiebel, editors. Breeding and genetic resources of five-needles pines: growth, adaptability and pest resistance; Proceedings of IUFRO Working Party 2.02.15. USDA Forest Service, Rocky Mountain Research Station, Ft. Collins, CO, Medford, OR. Potenko, V. V., and A. V. Velikov. 2001. Allozyme variation and mating system of coastal populations of Pinus koraiensis Sieb. et Zucc. in Russia. Silvae Genet. 50:117-122. Price, R. A., A. Liston, and S. H. Strauss. 1998. Phylogeny and systematics of Pinus. Pages 527 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Rajora, O. P., A. Mosseler, and J. E. Major. 2002. Mating system and reproductive fitness traits of eastern white pine (Pinus strobus) in large, central versus small, isolated, marginal populations. Can. J. Bot. 80:1173-1184. Richardson, B. A., S. J. Brunsfeld, and N. B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Mol. Ecol. 11:215-227. Ritland, K. 1989. Correlated-matings in the partial selfer, Mimulus guttatus. Evolution 43:848-859. Ritland, K. 2002. Extensions of models for the estimation of mating systems using n independent loci. Heredity 88:221-228. SAS Institute, I. 1999. The SAS system for windows, in. SAS Institute, Inc., Cary, North Carolina. 73 Savolainen, O., and P. W. Hedrick. 1995. Heterozygosity and fitness - no association in Scots pine. Evolution 140:755-766. Schemske, D. W., and R. Lande. 1985. The evolution of self-fertilization and inbreeding depression in plants. II. Empirical observations. Evolution 39:41-52. Slate, J., and J. M . Pemberton. 2002. Comparing molecular measures for detecting inbreeding depression. J. Evol. Biol. 15:20-31. Sorensen, F. C. 1969. Embryonic genetic load in coastal Douglas-fir, Pseudotsuga menziesii var. menziesii. Am. Nat. 103:389-398. Sorensen, F. C. 1994. Frequency of seedlings from natural self-fertilization in Pacific Northwest ponderosa pine (Pinusponderosa Dougl. ex. Laws.). Silvae Genet. 43:100-107. Sorensen, F. C. 1999. Relationship between self-fertility, allocation of growth, and inbreeding depression in three coniferous species. Evolution 53:417-425. Sorensen, F. C. 2001. Effect of population outcrossing rate on inbreeding depression in Pinus contorta var. murrayana seedlings. Scand. J. For. Res. 16:391-403. Sorensen, F. C , and R. S. Miles. 1982. Inbreeding depression in height, height growth, and survival of Douglas-fir, ponderosa pine, and noble fir to 10 years of age. Forest Sci. 28:283-292. Sorenson, F. C. 1982. The roles of polyembryony and embryo viability in the genetic system of conifers. Evolution 36:725-733. Spielman, D., B. W. Brook, D. A. Briscoe, and R. Frankham. 2004. Does inbreeding and loss of genetic diversity decrease disease resistance? Conserv. Genet. 5:439-448. Starova, N . V., Y. A. Yanbaev, N. K. Yumadilov, E. N . Adler, V. A. Dukharev, and Z. K. Shigapov. 1990. Genetic variability in age groups of scotch pine. Genetika 26:314-319. Stephenson, A. G., B. Leyshon, S. E. Travers, C. N . Hayes, and J. A. Winsor. 2004. Interrelationships among inbreeding, herbivory, and disease on reproduction in a wild gourd. Ecology 85:3023-3034. Tomback, D. F. 1982. Dispersal of whitebark pine seeds by Clark's Nutcracker: a mutualism hypothesis. J. Anim. Ecol. 51:451-467. Tomback, D. F., and Y. B. Linhart. 1990. The evolution of bird-dispersed Pines. Evol. Ecol. 4:185-219. Tomback, D. F., F.-K. Holtmeier, H. Mattes, K. S. Carsey, and M . L. Powell. 1993. Tree clusters and growth form distribution in Pinus cembra, a bird-dispersed pine. Arctic Alpine Res. 24:374-381. Tomback, D. F., and W. S. Schuster. 1994. Genetic population structure and growth form distribution in bird-dispersed pines. Pages 43-50 in W. C. Schmidt and F.-K. Holtmeier, editors. International Workshop on Subalpine Stone Pines and Their Environment: The Status of Our Knowledge; 1992 September 5-11; St. Moritz, Switzerland; Gen. Tech. Rep. INT-GTR-309. U.S. Dept. of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden, UT. Wang, T., S. N . Aitken, J. H. Woods, K. Polsson, and S. Magnussen. 2004. Effects of inbreeding on coastal Douglas-fir growth and yield in operational plantations: a model-based approach. Theor. Appl. Genet. 108:1162-1171. Wang, T. L., R. Hagqvist, and P. M . A. Tigerstedt. 1999. Inbreeding depression in three generations of selfed families of silver birch (Betula pendula). Can. J. For. Res. 29:662-668. Weaver, T. 2001. Whitebark pine and its environment. Pages 41-73 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Wilcox, M . D. 1983. Inbreeding depression and genetic variances estimated from self- and cross-pollinated families of Pinus radiata. Silvae Genet. 32:89-96. Williams, C. G., and O. Savolainen. 1996. Inbreeding depression in conifers: Implications for breeding strategy. Forest Sci. 42:102-117. Woods, J. H., T. Wang, and S. N. Aitken. 2002. Effects of inbreeding on coastal Douglas-fir: nursery performance. Silvae Genet. 51:163-170. 74 Wright, S. 1951. Evolution and the genetics of populations. Vol. II. University of Chicago Press, Chicago, IL. Zeglen, S. 2002. Whitebark pine and white pine blister rust in British Columbia, Canada. Can. J. For. Res. 32:1265-1274. 75 C H A P T E R 4: C H A N G E S IN G E N E T I C S T R U C T U R E O F W H I T E B A R K PINE (Pinus albicaulis Engelm.) A S S O C I A T E D W I T H I N B R E E D I N G A N D W H I T E PINE B L I S T E R R U S T I N F E C T I O N 3 Andrew D. Bower and Sally N. Aitken 4.1 I N T R O D U C T I O N Exotic diseases can have a devastating impact on populations that have not coevolved with pathogens and have little or no natural resistance, potentially leading to. a loss of genetic variation. Diseases pose unusual challenges to conservation because of their potential to drive rapid changes in host abundance and genetic composition (Altizer et al. 2003). Species in many genera of forest trees have been severely impacted by introduced diseases (e.g. Pinus, Chamaecyparis, Quercus, Castanea, Ulmus, Fagus, Juglans, Cornus) leading to local extirpation, ecosystem deterioration, impacts on associated species, and economic losses (Zobel et al. 1985; Ledig 1991; Daughtrey & Hibben 1994; McDonald et al. 1998; Dane et al. 1999; Houston & Houston 2000; Mcllwrick et al. 2000). However, data on the effects of these diseases on genetic structure and diversity are sparse. It is well known that disease can have a significant impact on both the species and structural composition of forests; however, empirical demonstrations of the effect of diseases on host population numbers are difficult to conduct (Alexander et al. 1996). In addition, studies of the effects of diseases on genetic diversity are difficult because baseline, pre-epidemic genetic data are impossible to collect after a disease has swept through (McDonald et al. 1998). The array of genotypes of reproducing mature adults, the seeds they produce, and the seedlings that germinate and survive to form the next generation of reproducing plants are three distinct life stages in conifers (Shaw & Allard 1982). Conifers often show increasing levels of heterozygosity with age, as embryos typically have a deficiency of heterozygotes, yet populations of mature trees have an excess (Plessas & Strauss 1986; Politov et al. 1992; Ledig et al. 2000; Chapter 3). These studies have observed positive fixation indices [Fjs = 1 - (H 0 / He)] (where H 0 and H e are 3 A version of this chapter has been submitted for publication. Bower, A.D . and Aitken, S.N. Changes in genetic structure in whitebark pine (Pinus albicaulis Engelm.) associated with inbreeding and white pine blister rust infection. Submitted to Conservation Genetics. 76 observed and expected heterozygosities, respectively) (Wright 1951) at the seed stage as a consequence of inbreeding (low levels of selfing or consanguineous matings), but not in adults (Morgante et al. 1993). A positive F;s may be the result of differences in genetic structure among stands producing a Wahlund effect, but in most cases, inbreeding and subsequent selection against more homozygous inbred individuals appear to be of greater importance (Plessas & Strauss 1986). Whitebark pine (Pinus albicaulis Engelm.) experiences inbreeding in the form of selfing, and to a lesser extent, consanguineous mating (Krakowski et al. 2003; Chapter 3). In addition, the inbreeding coefficient (F) of seedlings is higher than that of mature trees, indicating selection against inbred individuals (Chapter 3). A previous study reported that F i s decreased and observed heterozygosity (H0) increased from west to east in whitebark pine in British Columbia (Krakowski et al. 2003). Stand-level infection by the introduced disease white pine blister rust (caused by the fungus Cronartium ribicola J. C. Fisch. ex Rabh.) in British Columbia also increases from west to east (Zeglen 2002), although infection varies widely among stands even within a mountain range (Campbell & Antos 2000). The trends in genetic diversity observed by Krakowski et al. (2003) could be the result of either postglacial or other migration patterns or selection by the disease against more homozygous, inbred individuals. However, in their study, trees of all ages were sampled, so any potential effects of inbreeding may be confounded with the level of disease. The goals of this study are to examine the effects of inbreeding on genetic diversity of whitebark pine by comparing F i s and H 0 among three age cohorts, and to determine if there is a relationship between level of blister rust infection on a site and these parameters. Based on the results of Krakowski et al. (2003), we predict that selection against inbred individuals will be stronger where the level of rust infection is high, thus producing a greater difference in F i s between seedlings and mature trees at these sites compared with sites where levels are low. 77 4.2 M A T E R I A L S A N D M E T H O D S 4.2.1 Sample Materials Thirty individuals from each of three age cohorts were sampled from 14 sites in British Columbia, Idaho, Montana, and Oregon in 2002 and 2003 (Table 4.1 and Figure 4.1). The seedling cohort was represented by three-year-old trees growing in a common garden experiment in Vancouver, British Columbia. The young and mature cohorts were sampled in the field from the same locations where seeds were previously collected for the common garden experiment. The young cohort comprised trees estimated to be less than 30 years old, while the mature cohort included larger, older trees of reproductive age. To assess any potential differences in genetic diversity between the groups of putatively resistant (clean) and susceptible (infected) trees on each site, the health status of each individual was recorded. Without genetic analysis, it is impossible to determine the number of distinct genotypes in multi-stemmed clumps (Tomback & Schuster 1994). Therefore, multi-stemmed clumps were conservatively regarded as single genotypes and only a single stem was sampled. If any stems in the clump were infected, that sample was considered to be from an infected individual. Buds were sampled from 60 individuals in the field (young and mature cohorts) and 30 individuals in the common garden experiment (seedling cohort) and immediately placed into separate plastic vials for transportation and storage at -80°C. A disease infection survey was performed on each site to estimate the level of white pine blister rust infection. The survey protocol generally followed Smith and Hoffman (2000). Strip transects were used to estimate infection levels by categorizing the first 50 trees greater than breast height (1.3 meters) on each site as 1) alive and clean: no evidence of blister rust; 2) infected and alive: alive with cankers, flagged or dead branches, or dead stems in multiple stemmed clumps with some living stems; 3) infected and dead: dead due to blister rust; or 4) dead unknown: dead but unable to determine cause of death. Four to six transects were assessed on most sites, except one site with very low whitebark pine density, where only one transect was assessed which evaluated most of the trees on this site. Infection percentage was calculated as the number of infected trees (both alive and dead) 78 out of the total number of trees assessed minus the number of dead from unknown causes. The mean infection percentage over all transects on a site was used to estimate the level of blister rust infection. Sites were classified by infection level as low (<50% infection, n = 5), moderate (50-75% infection, n = 5), or high (>75% infection, n = 4). These levels were chosen based on natural groupings within the infection level data that yielded approximately even numbers of sites in each group. 4.2.2 Isozyme Analysis Whole buds were ground in a buffer slightly modified from Mitton (1977) and the supernatant absorbed onto 3x15 mm Whatman filter paper wicks. Samples were analyzed using starch gel electrophoresis on gels of 10.5% starch and 7.5% sucrose (w/v). Two buffer and electrode systems were used to analyze 10 loci as follows: 1) Lithium borate/Tris-citrate (pH 8.3) (Selander et al. 1971) was used for PGI(E.C. 5.3.1.9), SKDH1, 2 (E.C. 1.1.1.25), PGM (E.C. 2.7.5.1), and TPI (E.C. 5.3.1.1); 2) Histadine-EDTA (Cheliak & Pitel 1984) was used for IDH(E.C. 1.1.1.42), MDH1, 2 (E.C. 1.1.1.37), GDH(E.C. 1.4.1.3) and ALD (E.C. 4.1,2.13). Gels were electrophoresed with wicks in place at half voltage for 30 minutes, then wicks were removed and the lithium borate gels run at 300 V and the histadine-EDTA gels run at 225 V. Electrophoresis continued until a marker dye reached to within 1 cm of the edge of the gel. Gels were sliced and stained using recipes as detailed in Cheliak and Pitel (1984), then fixed in a mixture of 1:5:5 glacial acetic acid:distilled watenmethanol. 4.2.3 Data Analysis The Fstat program (Goudet 1995) was used to determine expected (He) and observed heterozygosities (FJ0) and Wright's (1951) fixation index (Fis) for each cohort on each site. Fstat also was used to calculate Weir and Cockerham's (1984) F s t, F i s, and their p-values (using 1000 bootstrap randomizations) for each cohort, and F i s for sites grouped by infection level within each cohort. 79 SAS Version 8 (SAS Institute 1999) was used for all statistical analysis. To determine if cohorts differed for genetic parameters, PROC G L M was used for one-way analysis of variance (ANOVA). Duncan's multiple range test was used to determine significant differences among cohorts. The ANOVA was also conducted separately on groups of sites stratified by level of infection. Simple regression with PROC REG was used to determine if there was a significant association between level of blister rust infection and site latitude and longitude or genetic parameters. Paired t-tests were performed on data from the mature cohort to determine if clean and infected trees paired by site differed in H 0 and FjS. Sites varied in the number of clean and infected trees in this cohort (Table 4.1). To eliminate potential confounding due to the variation in sample size between clean and infected trees on each site, ¥,s was calculated as 1 - (H 0 for the group of clean or infected trees/He calculated for all trees on the site). The t-test was limited to sites that had a minimum of four individuals in each group. 4.3 R E S U L T S Across all sites, significant inbreeding (F,s > 0 at a = 0.05) was detected in all cohorts. When sites were stratified by level of rust infection, significant evidence of inbreeding was detected in all cohorts when rust infection level was moderate or high, but only in the seedling cohort when infection level was low (Figure 4.2). Differences in mean F i s values were marginally significant among cohorts (p = 0.056), and means of the seedling and young cohorts were significantly different (p<0.05). No significant difference was detected among cohorts for H 0 . When the level of rust infection was low, the mean PL of the seedling cohort was significantly lower than both the young and mature cohort in the multiple range test (p = 0.039). When the infection level was moderate or high, no significant differences were detected among cohorts for any genetic parameters (Figures 4.2 and 4.3). 80 No significant associations were found between stand level blister rust infection and latitude (p = 0.361) or longitude (p = 0.167). Regressions of the genetic parameters for each of the cohorts with blister rust infection levels showed a significant association only for H 0 in the mature cohort (p = 0.018). However, when a sequential Bonferroni adjustment is used to ensure an experiment-wise <x-level of 0.05 (Rice 1989), this relationship is not significant at a = 0.05/6 = 0.0083. Visual inspection of plots of F i s and H 0 versus infection percentage shows that the association between F i s and infection increases with cohort age while the association of H Q decreases (Figures 5a-f). While these results are suggestive of a relationship between genetic parameters and blister rust infection, they are not conclusive. Comparison of clean and infected trees within a site showed that for nine of the fourteen sites, H 0 was higher on average in the infected trees than in clean trees. However, paired T-tests for both F i s and H 0 showed that the differences between the group means were not statistically significant (p = 0.297 and 0.247 for F i s and H 0 , respectively). Genetic differentiation varied among cohorts (Fst = 0.075, 0.025, and 0.040 for the seedling, young, and mature cohorts, respectively), but all values are within the range previously reported for whitebark pine (Yandell 1992; Jorgensen & Hamrick 1997; Stuart-Smith 1998; Richardson et al. 2002; Krakowski et al. 2003). 4.4 DISCUSSION Populations of mature conifers often have an excess of heterozygosity while seed and seedlings usually have a deficiency (Ledig et al. 2000). Results from several other studies of conifers are summarized in Table 4.2. Except for whitebark pine and one Abies species, all of these species exhibit this pattern. The excess homozygosity in seeds is most likely due to inbreeding (Ledig et al. 2000), including self-pollination and consanguineous matings, and selection against inbreds is the most likely reason for the reduction in F i s with age (Krutovskii et al. 1995). For example, a study of Bosnian pine (Pinus leucodermis Ant.), which has reported selfing rates in the same range as 81 whitebark pine (18-28%), found that inbred individuals were not efficiently eliminated during seed maturation and germination, but were eliminated by age five (Morgante et al. 1993). We observed a significant decrease in F;s between the seedling and young cohorts when the level of rust infection was low, but no difference between cohorts when the level of rust was moderate or high. This suggests that when selection pressure due to rust is weak, more heterozygous individuals are favoured. However, blister rust appears to exert strong differential selection when levels of rust are moderate to high. Based on the geographic patterns in F;s, H 0 , and infection found by others (Zeglen 2002; Krakowski et al. 2003), we predicted that this differential selection would give additional advantage to more heterozygous individuals; however, our results suggest the opposite. When selection pressure due to blister rust is higher, heterozygosity levels either decrease non-significantly, or are nearly the same among cohorts. This means that inbreeding and blister rust are affecting the genetic diversity of whitebark pine in different ways. Inbreeding results in more homozygous individuals and a heterozygote deficiency in seeds and seedlings, but this effect usually decreases over time as trees age and density-dependent selection occurs. Meanwhile, more homozygous individuals appear to have higher fitness when challenged by white pine blister rust, maintaining the heterozygote deficiency typical of seedlings in both young and mature trees in the more infected populations studied. Previous studies of the effects of disease epidemics on genetic diversity have yielded varying results in both forest trees and agricultural crops. Burdon and Thompson (1995) showed a change in the distribution of resistance genotypes in wild flax (Linum marginale A. Cum. Ex Planch.) after infection by a rust epidemic caused by Melampsora lini (Ehrenb.). In an artificial population of Silene alba (Mill.) Krause, the proportion of resistant plants increased over a 3-year period after infection with the smut fungus Ustilago violaceae (Pers.) (Alexander & Antonovics 1995). In wild plant-pathogen populations, hosts should evolve towards increased resistance (Gilbert 2002). However, it is not always possible to determine if changes in genetic structure are correlated with increases in resistance (McDonald et al. 1998). The effect of an oak wilt epidemic on a live oak 82 (Quercus fusiformis Small) population in Texas is one of the few empirical studies of the effect of an exotic disease on the genetic structure of a forest tree (McDonald et al. 1998). Significant differences in both allele and genotypic frequencies were observed between pre- and post-epidemic populations, and an increase of heterozygotes was observed for two allozyme loci in the post-epidemic population. In addition, no multilocus associations were present in the pre-epidemic population; however there were some in the post-epidemic population. These results suggest dominant or multigenic control of resistance and the authors hypothesized that selection for increasing disease resistance was the dominant evolutionary force driving genetic change. In a study of American beech (Fagus grandifolia Ehrh.), populations of trees susceptible to beech bark disease had H 0 values 26% higher than resistant trees and showed an excess of heterozygotes, while resistant trees showed a deficiency of heterozygotes (Houston & Houston 2000). These data suggest that recessive genes may control resistance in these cases. Conversely, a heterozygote excess detected in populations of American chestnut (Castanea dentatd) that have persisted through the chestnut blight epidemic through sprouting suggests that selection has favoured more heterozygous individuals (Stilwell et al. 2003). Low levels of natural resistance to an introduced pathogen can result in high selection pressure, and evolution towards increased resistance could result in decreased levels of diversity due to a genetic bottleneck or selective sweep. In a comparison of populations of western white pine (P. monticola), Kim et al. (2003) found that a natural population with low rust pressure had higher polymorphism (P) and heterozygosity (He), and more unique alleles than a population with high rust pressure. Assuming that levels of genetic diversity in the two populations were similar prior to invasion by blister rust, this suggests that selection by rust resulted in reduced diversity. However, a comparison of trees from a phenotypically resistant seed orchard with non-selected trees from natural population showed no differences in P or H e , and no sign of a genetic bottleneck, despite substantial differences in blister rust resistance (Kim et al. 2003). Inferences regarding the level of resistance in whitebark pine populations sampled in this study are not possible; however, stronger differential selection on high rust sites may lead to higher levels of overall resistance in the remaining population, 83 albeit at the risk of dramatic reductions in population size and genetic diversity, and even local extirpations if resistance genes are recessive. The existence and nature of heterozygosity-fitness correlations has been debated extensively in the literature (reviewed by Britten 1996; David 1998) and cases of low heterozygosity have been associated with lower resistance to pathogens in both plants and animals (Spielman et al. 2004). Disease resistance mechanisms classified as "vertical", which are qualitative, operate to prevent establishment of the pathogen, and often equate to hypersensitivity; while "horizontal" mechanisms, which are quantitative, operate after initial infection to delay or reduce sporulation, and possibly extend the period of latent infection (van der Planck 1963; Burdon 2001). Both dominant and recessive resistance alleles have been identified in a variety of crop plants (Toyoda et al. 2002; Chu et al. 2004). To date, only one dominant vertical mechanism of resistance to blister rust (hypersensitivity) has been identified in the white pines (Kinloch & Littlefield 1977; Kinloch et al. 1999; Kinloch & Dupper 2002), while several horizontal mechanisms have been identified with both dominant and recessive modes of inheritance hypothesized for a number of pine-rust pathosystems (see references in Kinloch 1982). Disease resistance alleles are often dominant (Burdon 2001) although about 10% of resistance alleles in crops are recessive (Burdon & Thompson 1995). In a large, truly panmictic population, low frequency recessive alleles would rarely be expressed and be of little selective value (Burdon 2001), but forest trees typically have mixed mating systems with some selfing and bi-parental inbreeding (Ledig 1998). In wild populations, this mixed mating will enhance the expression of such genes. Although this will likely be associated with some inbreeding depression (Williams & Savolainen 1996), the cost could be small compared with the selective advantage of disease resistance (Burdon 2001). The overall rate of inbreeding in whitebark pine is approximately 14% (Chapter 3), which is the same as the mean oi Pinus Subsection Cembrae to which it belongs (Price et al. 1998; but see Gernandt et al. 2005), and only slightly higher than the mean of the genus Pinus (12%) (Chapter 3). Inbreeding rate of whitebark pine varies by population and can be as high as 27%; however, we observed no inbreeding depression in two of three regions 84 studied (Chapter 3). The relationship between H 0 and level of blister rust infection suggests, but does not confirm, that the higher fitness of more homozygous individuals when challenged by blister rust could reflect recessive alleles for resistance. Nearly 50% of the distribution of whitebark pine in the United States is within designated wilderness areas or national parks (R. Keane unpublished data), which requires that these areas be managed to preserve the wilderness character of the area (McCool & Freimund 2001). In addition, the rugged, relatively inaccessible habitats where whitebark pine grows makes active management and conservation efforts difficult, limiting potential silvicultural and restoration treatments. One of the few options to limit or reverse losses due to white pine blister rust is to increase genetic resistance (Hoff et al. 2001). Resistance to blister rust has been documented in whitebark pine (Hoff et al. 1980; Hoff et al. 2001); however, a program to test and breed for blister rust resistance likely would take a long time to develop to the point where resistant seed is being produced. Collection of seed from healthy individuals in heavily infected stands has been proposed for the initial screening of potentially resistant parent trees (Hoff et al. 2001; Mahalovich & Dickerson 2004). If genetic control of some genes for resistance is recessive, this can help guide a breeding program that seeks to increase rust resistance. Although controversial, inbreeding has been suggested as a breeding tool in conifers. Selfing is the most extreme form of inbreeding, but is only likely to be useful in species with relatively few lethal equivalents due to typically high inbreeding depression (Williams & Savolainen 1996). Whitebark pine appears to suffer lower levels of inbreeding depression in biomass production than most wind-pollinated conifers, and suffers no detectable inbreeding depression in seedling height growth and survival in a common garden (Chapter 3). If the genetic basis of some resistance mechanisms are determined to be recessive, this lack of inbreeding depression suggests that a system of inbreeding (either selfing or sib-mating) could be used to promote the expression of resistance alleles, allowing more rapid development of rust resistant seedlings than under outcrossing or random mating. 85 4.5 A C K N O W L E D G E M E N T S The authors thank the USDA Forest Service regions 1 , and 6, the British Columbia Ministry of Forests, B.C. Parks E. C. Manning and Tweedsmuir Provincial Parks, and Bob Brett of Snowline Ecological Consulting, Whistler, B.C. for providing seed for this study. Dorothy Watson, Dane Szohner, Jodie Krakowski, and Milena Semproni helped in the field and lab and Christine Chourmouzis created the map. Funding for this study came from the British Columbia Forestry Investment Account through the Forest Genetics Council of B.C. to the Centre for Forest Gene Conservation. Thank you to Drs. Michael Whitlock, Alvin Yanchuk, Jeannette Whitton, Yousry El-Kassaby, Sean Graham, and Diana Tomback for their helpful comments on an earlier draft of this manuscript. 86 Table 4.1. Study sites, geographic locations, and white pine blister rust infection. No. a Site Name State/ Lat. Long. Elev. Infection # trees in mature cohort Prov. (°N.) (°W.) (m) % Clean Infected 1 Heckman Pass BC 52.5 125.8 1525 57.0 2 28 2 Jesamond BC 51.3 121.8 1850 55.4 12 18 3 D'Arcy BC 50.5 122.6 1800 81.0 6 24 4 Thynne Mt. BC 49.7 120.9 1785 82.5 11 19 5 Blackcomb BC 50.1 122.9 1900 71.0 20 10 6 Manning Park BC 49.1 120.7 2000 40.0 13 17 7 Mt. Baldy BC 49.2 119.3 2150 44.3 11 19 8 Lunch Peak ID 48.4 116.2 1850 81.6 5 25 9 Granite Butte MT 46.9 112.5 2340 82.2 3 27 10 Hellroaring MT 45.0 109.4 3000 70.0 8 22 11 Sawtel Peak ID 44.5 111.4 2400 47.5 16 14 12 Gospel Hill ID 45.6 115.9 2100 45.5 8 22 13 Quartz Hill ID 45.7 112.9, 2650 62.7 12 18 14 Vinegar Hill OR 44.7 118.6' 2340 30.1 26 4 a Refers to number on Figure 4.1. Table 4.2. Fixation indices (Fjs) at three life stages for several conifers. Species F e m brvo Fseedling (age) Fmature (age) Reference Pinus albicaulis 0.203 (3) 0.141 This study 0.093 -0.034a Chapter 3 P. sylvestris 0.143 0.266(10) -0.123 (40) Starovaetal. 1990 0.12 0.006 (3) Muona et al. 1987 P. radiata 0.081 0.038 (3-6) -0.119(17-20) Plessas & Strauss 1986 P. leucodermis 0.126 -0.101 (5) -0.095 Morgante et al. 1993 0.142 -0.121 (5) -0.079 P. sibirica -0.06 -0.025 Politovetal. 1992 P. koraiensis -0.04 -0.046 Politovetal. 1992 P. cembra 0.012 -0.174 Krutovskii et al. 1995 P. pumila 0.213 -0.046 Politov et al. 2006 P. ponderosa 0.171 -0.022 Farris & Mitton 1984 P. monticola 0.011 -0.105 (10-34) El-Kassaby et al. 1987 P. banksiana 0.010 -0.064 Cheliaketal. 1985 Pseudotsuga menziesii 0.050 -0.027 (>25) Shaw& Allard 1982 Picea engelmanii 0.050 -0.065 (<80) Shea 1987 Abies lasiocarpa 0.074 -0.102 (<80) Shea 1987 A. bracteata 0.388 0.049 Ledig et al. 2006 Sequoiadendron 0.043 -0.021 Fins&Libby 1982 giganteum a Determined from inferred maternal genotype 87 Figure 4.1. Map of the range of whitebark pine and locations and level of blister rust infect of 14 sample locations. 88 Figure 4.2. Mean fixation index (Fjs) by cohort for low, moderate, and high rust infection level sites (error bars are 95% confidence intervals. 0.5 0.4 0.3 0.2 0.1 0 -0.1 -0.2 Low Moderate 0 Seedling • Young B Mature High Rust Infection Level C lass 89 Figure 4.3. Mean observed heterozygosity (H0) by cohort for low, moderate, and high rust infection level sites (error bars are 95% confidence intervals). 0.24 0.22 0.20 0.18 -\ 0.16 0.14 0.12 0.10 0 Seedling • Young B Mature Low Moderate Rust Infection Level C lass High 90 Figure 4.4. Scatterplot of (a-c) fixation index (FiS) and (d-f) observed heterozygosity (H0) of whitebark pine versus infection percent for 14 sites. 0.5 0.4 0.3 iC 0.2 0.1 0 -0.1 0.5 0.4 0.3 uc 0.2 0.1 0 -0.1 3 0.5 0.4 0.3 i£ 0.2 0.1 0 -0.1 a - Seedling 20 40 60 Infection % 80 b - Young c - Mature 100 20 40 60 8ft 100 Infection % 20 40 60 80 100 Infection % 0.3 0.25 £ 0.2 0.15 0.1 0.3 0.25 £ 0.2 0.15 0.1 20 40 60 80 100 Infection % e - Young f - Mature 20 40 60 80 100 Infection % R z = 0.39 p = 0.02 20 40 60 80 100 Infection % 91 4.6 REFERENCES Alexander, H. M. , and J. Antonovics. 1995. Spread of anther-smut disease (Ustilago violacea) and character correlations in a genetically variable experimental population of Silene alba. J. Ecol. 83:783-794. Alexander, H. M. , P. H. Thrall, J. Antonovics, A. M . Jarosz, and P. V. Oudemans. 1996. Population dynamics and genetics of plant disease: a case study of anther-smut disease. Ecology 77:990-996. Altizer, S., D. Harvell, and E. Friedle. 2003. Rapid evolutionary dynamics and disease threats to biodiversity. Trends Ecol. Evol. 18:589-596. Britten, H. B. 1996. Meta-analyses of the association between multilocus heterozygosity and fitness. Evolution 50:2158-2164. Burdon, J. J., and J. N . Thompson. 1995. Changed patterns of resistance in a population oiLinum marginale attacked by the rust pathogen Melapsora lini. J. Ecol. 83:199-206. Burdon, R. D. 2001. Genetic diversity and disease resistance: some considerations for research, breeding, and deployment. Can. J. For. Res. 31:596-606. Campbell, E. M. , and J. A. Antos. 2000. Distribution and severity of white pine blister rust and mountain pine beetle on whitebark pine in British Columbia. Can. J. For. Res. 30:1051-1059. Cheliak, W. M. , B. P. Dancik, K. Morgan, F. C. Yeh, and C. Strobeck. 1985. Temporal variation of the mating system in a natural population of jack pine. Genetics 109:569-584. Cheliak, W. M. , and J. A. Pitel. 1984. Techniques for starch gel electrophoresis of enzymes from forest tree species. Can. For. Serv. Petawawa Natl. For. Inst. Rep. No. Pi-X-42. Chu, Z., Y. Ouyang, J. Zhang, H. Yang, and S. Wang. 2004. Genome-wide analysis of defense-responsive genes in bacterial blight resistance of rice mediated by the recessive R gene xal3. Mol. Genet. Genomics 271:111-120. Dane, F., L. K. Hawkins, and H. W. Huang. 1999. Genetic variation and population structure of Castaneapumila var. ozarkensis. J. Am. Soc. Hortic. Sci. 124:666-670. Daughtrey, M . L., and C. R. Hibben. 1994. Dogwood anthracnose: a new disease threatens 2 native Cornus species. Annu. Rev. Phytopathol. 32:61-73. David, P. 1998. Heterozygosity-fitness correlations: new perspectives on old problems. Heredity 80:531-537. El-Kassaby, Y. A., M . D. Meagher, J. Parkinson, and F. T. Portlock. 1987. Allozyme inheritance, heterozygosity and outcrossing rate among Pinus monticola near Ladysmith, British Columbia. Heredity 58:173-181. Farris, M . A., and J. B. Mitton. 1984. Population density, outcrossing rate, and heterozygote superiority in ponderosa pine. Evolution 38:1151-1154. Fins, L., and W. J. Libby. 1982. Population variation in Sequoiadendron: seed and seedling studies, vegetative propagation, and isozyme variation. Silvae Genet. 31:102-110. Gernandt, D. S., G. Geada Lopez, S. Ortiz Garcia, and A. Liston. 2005. Phylogeny and classification of Pinus. Taxon 54:29-42. Gilbert, G. S. 2002. Evolutionary ecology of plant diseases in natural ecosystems. Annu. Rev. Phytopathol. 40:13-43. Goudet, J. 1995. Fstat version 1.2: a computer program to calculate Fstatistics. J. Hered. 86:485-486. Hoff, R. J., R. T. Bingham, and G. I. McDonald. 1980. Relative blister rust resistance of white pines. Eur. J. Forest Pathol. 10:307-316. Hoff, R. J., D. E. Ferguson, G. I. McDonald, and R. E. Keane. 2001. Strategies for managing whitebark pine in the presence of white pine blister rust. Pages 346-366 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D. C. 92 Houston, D. B., and D. R. Houston. 2000. Allozyme genetic diversity among Fagus grandifolia trees resistant or susceptible to beech bark disease in natural populations. Can. J. For. Res. 30:778-789. • Jorgensen, S. M. , and J. L. Hamrick. 1997. Biogeography and population genetics of whitebark pine, Pinus albicaulis. Can. J. For. Res. 27:1574-1585. Kim, M.-S., S. J. Brunsfeld, G. I. McDonald, and N. B. Klopfenstein. 2003. Effect of white pine blister rust (Cronartium ribicola) and rust-resistance breeding on genetic variation in western white pine (Pinus monticola). Theor. Appl. Genet. 106:1004-1010. Kinloch, B. B. 1982. Mechanisms and inheritance of rust resistance in conifers. Pages 110-118. Resistance to Pests and Diseases in Forest Trees. Proceedings of the 3rd International Workshop on Genetics of Host-Parasite Interactions in Forestry, Sept. 1980. Wagenigen, The Netherlands. Kinloch, B. B., and G. E. Dupper. 2002. Genetic specificity in the white pine-blister rust pathosystem. Phytopathology 92:278-280. Kinloch, B. B., and J. L. Littlefield. 1977. White pine blister rust: hypersensitive resistance in sugar pine. Can. J. Bot. 55:1148-1155. Kinloch, B. B., R. A. Sniezko, G. D. Barnes, and T. E. Greathouse. 1999. A major gene for resistance to white pine blister rust in western white pine from the western Cascade range. Phytopathology 89:861-867. Krakowski, J., S. N . Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. Krutovskii, K. V., D. V. Politov, and Y. P. Altukhov. 1995. Isozyme study of population genetic structure, mating system and phylogenetic relationships of the five stone pine species (subsection cembrae, section strobi, subgenus strobus) in P. Baradat, W. T. Adams, and Muller-Starck, editors. Population genetics and genetic conservation of forest trees. SPB Academic Publishing, Amsterdam, The Netherlands. Ledig, F. T. 1991. Human impacts on genetic diversity in forest ecosystems. Oikos 63:87-108. Ledig, F. T. 1998. Genetic variation in Pinus. Page 527 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Ledig, F. T., B. Bermejo-Velazquez, P. D. Hodgskiss, D. R. Johnson, C. Flores-Lopez, and V. Jacob-Cervantes. 2000. The mating system and genetic diversity in Martinez spruce, an extremely rare endemic of Mexico's Sierra Madre Oriental: an example of facultative selfing and survival in interglacial refugia. Can. J. For. Res. 30:1156-1164. Ledig, F. T., P. D. Hodgskiss, and D. R. Johnson. 2006. Genetic diversity and seed production in Santa Lucia fir (Abies bracteatd), a relict of the Miocene broadleaved evergreen forest. Conserv. Genet. 7:383-398. Little, E. L., Jr. 1971. Atlas of United States trees, volume 1, conifers and important hardwoods. USDA For. Serv. Misc. Publ. 1146. Washington, D.C. Mahalovich, M . F., and G. A. Dickerson. 2004. Whitebark pine genetic restoration program for the intermountain west (United States). Page 181-187 in R. A. Sniezko, S. Samman, S. E. Schlarbaum, and H. B. Kreiebel, editors. Breeding and genetic resources of five-needles pines: growth, adaptability and pest resistance; 2001 July 23-27; Medford, OR, USA. IUFRO Working Party 2.02.15. Proceedings RMRS-P-32. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ft. Collins, CO. McCool, S. F., and W. A. Freimund. 2001. Threatened landscapes and fragile experiences: conflict in whitebark pine restoration. Pages 263-284 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D. C. McDonald, B. A., B. K. Bellamy, J. Zhan, and D. N . Appel. 1998. The effect of an oak wilt epidemic on the genetic structure of a Texas live oak population. Can. J. Bot. 76:1900-1907. 93 Mcllwrick, K., S. Wetzel, T. Beardmore, and K. Forbes. 2000. Ex situ conservation of American chestnut (Castanea dentata (Marsh.) Borkh.) and butternut (Juglans cinera L.), a review. For. Chron. 76:765-774. Mitton, J. B., Y. B. Linhart, S. W. Sturgeon, and J. P. Farrar. 1977. Protein polymorphisms detected in mature needle tissue of ponderosa pine. Am. J. Bot. 42:1154-1157. Morgante, M. , G. G. Vendramin, P. Rossi, and A. M . Olivieri. 1993. Selection against inbreds in early live-cycle phases in Pinus leucodermis Ant. Heredity 70:622-627'. Muona, O., R. Yazdani, and D. Rudin. 1987. Genetic change between life stages in Pinus sylvestris: allozyme variation in seeds and planted seedlings. Silvae Genet. 1987:1. Plessas, M . E., and S. H. Strauss. 1986. Allozyme differentiation among populations, stands, and cohorts in Monterey pine. Can. J. For. Res. 16:1155-1164. Politov, D. V., K. V. Krutovskii, and K. V. Altukhov. 1992. Isozyme loci characteristics of gene banks of populations of cedar pines. Genetika 28:76-95. Politov, D. V., M . M . Belokon, and Yu. S. Belokon. 2006. Dynamics of allozyme heterozygosity in the Far Eastern populations of Siberian dwarf pine, Pinus pumila (Pall.) Regel: comparison of embryos and maternal plants. Russ. J. Genet. 42:1-11. Price, R. A., A. Liston, and S. H. Strauss. 1998. Phylogeny and systematics of Pinus. Pages 527 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Richardson, B. A., J. Brunsfeld, and N. B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Mol. Ecol. 11:215-227. SAS Institute, I. 1999. The SAS system for windows. SAS Institute, Inc., Cary, North Carolina. Selander, R. K., M . H. Smith, S. Y. Yang, W. E. Johnson, and J. R. Gentry. 1971. Biochemical polymorphism and systematics in the genus Peromyscus I. Variation in the old-field mouse (Peromyscuspolionotus). Stud. Genet. VI. 7103:49-90. Shaw, D. V., and R. W. Allard. 1982. Isozyme heterozygosity in adult and open-pollinated embryo samples of Douglas-fir. Silva Fenn. 16:115-121. Shea, K. L. 1987. Effects of population structure and cone production on outcrossing rates in Engelmann spruce and subalpine fir. Evolution 41:124-136. Smith, J. P., and J. T. Hoffman. 2000. Status of white pine blister rust in the Intermountain West. West. N. Am. Naturalist 60:165-179.. Spielman, D., B. W. Brook, D. A. Briscoe, and R. Frankham. 2004. Does inbreeding and loss of genetic diversity decrease disease resistance? Conserv. Genet. 5:439-448. Starova, N . V., Y. A. Yanbaev, N. K. Yumadilov, E. N . Adler, V. A. Dukharev, and Z. K. Shigapov. 1990. Genetic variability in age groups of scotch pine. Genetika 26:314-319. Stilwell, K. L., H. M . Wilbur, C. R. Werth, and D. R. Taylor. 2003. Heterozygote advantage in the American Chestnut, Castanea dentata (Fagaceae). Am. J. Bot. 90:207-213. Stuart-Smith, G. J. 1998. Conservation of whitebark pine in the Canadian Rockies: blister rust and population genetics. Page 125. University of Alberta, Edmonton. Tomback, D. F., and W. S. Schuster. 1994. Genetic population structure and growth form distribution in bird-dispersed pines. Pages 43-50 in W. C. Schmidt and F.-K. Holtmeier, editors. International Workshop on Subalpine Stone Pines and Their Environment: The Status of Our Knowledge; 1992 September 5-11; St. Moritz, Switzerland; Gen. Tech. Rep. INT-GTR-309. U.S. Dept. of Agriculture, Forest Service, Rocky Mountain Research Station, Ogden, UT. Toyoda, K., N . C. Collins, A. Takahashi, and K. Shirasu. 2002. Resistance and susceptibility of plants to fungal pathogens. Transgenic Res. 11:567-582. van der Planck, J. E. 1963. Disease resistance in plants. Academic Press, New York. Weir, B. S., and C. C. Cockerham. 1984. Estimating F-statistics for the analysis of population structure. Evolution 38:1358-1370. 94 Williams, C. G., and O. Savolainen. 1996. Inbreeding depression in conifers: Implications for breeding strategy. Forest Sci. 42:102-117. Wright, S. 1951. Evolution and the genetics of populations. Vol. II. University of Chicago Press, Chicago, IL. Yandell, U. G. 1992. An allozyme analysis of whitebark pine (Pinus albicaulis Engl.). Page 73. University of Nevada, Reno. Zeglen, S. 2002. Whitebark pine and white pine blister rust in British Columbia, Canada. Can. J. For. Res. 32:1265-1274. Zobel, D. B., L. F. Roth, and G. M . Hawk. 1985. Ecology, pathology, and management of Port-Orford-cedar (Chamaecypafis lawsoniana). Gen. Tech. Rep. PNW-GTR-184. U.S. Department of Agriculture, Forest Service, Pacific Northwest Research Station, Portland, OR. 95 CHAPTER 5: Geographical and seasonal variation in Cold Hardiness of Whitebark Pine (Pinus albicaulis Engelm.)4 Andrew D. Bower and Sally N. Aitken 5.1 I N T R O D U C T I O N Studies of cold hardiness and cold injury in most conifers have concentrated on economically important lower-elevation temperate species that are in breeding programs (see reviews by Howe et al. 2003 and Aitken and Hannerz 2001). Many of these studies have evaluated genetic differences among provenances or families and calculated heritabilities and genetic correlations among cold adaptation traits for the purpose of predicting gains or correlated responses to breeding for growth. Whitebark pine (Pinus albicaulis Engelm.) is a high-elevation species that as a keystone species in subalpine zone and treeline forest communities and is of tremendous ecological value (Arno and Hoff 1989, Tomback et al. 2001, Tomback and Kendall 2001, Weaver 2001). It is threatened in many areas throughout its range due to extensive damage and mortality caused by white pine blister rust (Campbell and Antos 2000), an introduced disease caused by the fungus Cronartium ribicola. It has also been impacted by mountain pine beetle (Dendroctonus ponderosae) and successional replacement resulting from fire suppression. As a result, it has become an ecological symbol of the effects of altered fire regimes and the introduction of exotic diseases to western North America (Tomback et al. 2001). While no breeding programs currently exist for whitebark pine, concerns about population decline and local extirpation have increased restoration plantings and consideration for formal listing as a threatened species. Previous research has examined population differentiation using molecular markers, but no information regarding genetic variation in quantitative traits has been available. As a result, seed transfer guidelines have generally been conservative in order to minimize potential maladaptation of planted seedlings (Mahalovich and Dickerson 2004). Populations of forest trees are often well differentiated for cold adaptation traits (Howe et al. 2003); and, for a high 4 A version of this chapter has been published. Bower, A . D . and Aitken, S.N. 2006. Geographical and seasonal variation in cold hardiness of whitebark pine (Pinus albicaulis Engelm.). Can. J. For. Res. 36:1842-1850 96 elevation species, adaptation to cold is likely to be an important determinant of how far seed can be moved with minimal risk of maladaptation. Whitebark pine is distributed from 37° to 55° N , and from 130° to 110° W. (Critchfield and Little 1966). It is restricted to upper subalpine forests, with typical habitats of ridge crests and steep southwestern-facing slopes that experience high winds and shallow snow. The climate in whitebark pine habitats is cold, windy, snowy, and generally moist. In the cool, short summers of these habitats, frost and snowfall can occur in any month of the year, and in winter, strong winds and severe blizzards are not unusual (Arno and Hoff 1989). This means that whitebark pine must be adapted to extreme cold during its period of winter dormancy but must also be able to withstand freezing temperatures during active growth. Spring cold injury is strongly correlated with timing of bud flush in most tree species, whereas the relationship between fall cold injury and bud set is variable (Howe et al. 2003). Bud set and acclimation to cold are triggered by different environmental cues than bud flush and deacclimation, and fall and spring cold hardiness are for the most part genetically unrelated (Howe et al. 2003). The timing of bud set is influenced by photoperiod (short days), temperature, soil moisture, nutrition, and light quality, while bud flush is predominantly influenced by temperature following adequate chilling (Howe et al. 2003, Sakai and Larcher 1987). The first of three stages of cold acclimation in plants is also induced by short days, while cold temperatures induce the second stage. The third stage of hardiness is induced by very low temperatures, but even a brief thaw can result in the loss of hardiness to extreme cold associated with this stage (Sakai and Larcher 1987, Weiser 1970). During the winter, when trees are in the second or third stage of acclimation, most tree species are usually hardy to temperatures well below what they would experience in their native environment (Glerum 1973). However, most temperate and boreal species will suffer cold injury if the temperature drops more than a few degrees below freezing during the period of active growth (Sakai and Larcher 1987). For most temperate tree species studied, the risk of a summer freeze event 97 is low. However, due to its high elevation habitat, whitebark pine is at risk from freezing temperatures throughout the year. Cold hardiness testing in the summer, during the period of active growth is much less common than testing in the fall, winter, and spring when the risk of cold injury for most species is much higher. In most species there is little variation among groups in cold hardiness during the growing season and most studies have concentrated on cold hardiness in the fall just before elongation stops, and in the spring just after elongation resumes (Howe et al. 2003). These are the periods that pose the greatest risk of frost damage to temperate conifers. The goals of this study were to sample whitebark pine across all seasons to characterize the annual patterns of cold hardiness, then to intensively sample in fall and spring to determine geographic patterns and genetic control of cold hardiness traits during the critical periods of cold acclimation and deacclimation. The pattern of geographic differentiation can help determine the extent to which whitebark pine populations are adapted to their local environments, and the degree to which seed movement should be restricted in any conservation and restoration efforts. 5.2 M A T E R I A L S A N D M E T H O D S 5.2.1 Source materials Open-pollinated, 10-month-old seedlings from a nearly range-wide seed collection were planted in 2002 in raised nursery beds in a common garden experiment in Vancouver, British Columbia (49° 13', 123° 6'). Seedlings were planted in an Alpha design (Patterson and Williams 1976) with eight replications, and ten four-tree by four-tree incomplete blocks of 16 trees in each block within replication. The AlphaPlus program (CSIRO 1996) was used to design the planting layout. Seedlings were planted at 9.5 x 10 cm spacing with one row of buffer trees surrounding each raised nursery bed. One hundred and sixty families from ten geographic regions that covered the range of whitebark pine were selected for artificial freeze testing (Table 5.1 and Figure 5.1). Regions were delineated based on geographic grouping of source locations, and physiographic features when appropriate (e.g. west and east of the Continental Divide). Due to the varied availability of seed, 98 regions were represented in the common garden by varying number of families (Table 5.1) and as a result of mortality, a varying number of individuals per family were available for testing (1-3 for fall, and 1-4 for spring). Six of the ten regions were tested in summer and winter (Table 5.1) to determine seasonal patterns of cold hardiness. These regions were chosen to span the latitudinal and longitudinal range of the seed sources and determine if hardiness varied across this range. Seedlings from all regions were sampled for fall and spring testing. 5.2.2 Freeze testing and phenological observations Artificial freeze testing was used to determine cold injury once in each season using the electrolyte leakage method. The freeze testing protocol followed the method described by Hannerz et al. (1999) with an unfrozen control vial and multiple test temperatures. Needles were collected from seedlings, rinsed in distilled water, and then cut into 5 mm lengths. Five needle segments were placed in each test vial with 0.2 ml distilled water and a few grains of silver iodide for ice nucleation. Samples were placed in a programmable Tenney Environmental Chamber (model T20C-3), and held at 4°C to equilibrate. Control vials were removed from the chamber at this point. Test temperatures were selected to span a range of injury based on preliminary tests conducted one week prior to freeze testing. Two to five temperatures were selected, depending on the date and the range of variation observed in the preliminary trial (Table 5.2). Temperature was decreased at a rate of 4°C/h to the first test temperature and then held constant for one hour. Sample vials for that test temperature were removed and placed at 4°C to thaw. The temperature was again decreased at 4°C/h until the next test temperature was reached, and the process repeated. After thawing samples for approximately 2 h. in the refrigerator, an additional 3.3 mL of distilled water was added to each vial. Vials were refrigerated at 4°C for approximately 20 hours, then placed on a gravity shaker at room temperature for 1 h before measuring conductivity with a VWR portable conductivity meter (model 2052). The vials were then placed in a hot water bath at 90°C to heat kill samples, and refrigerated at 4°C for 24 hours. The heat-killed samples were shaken 99 at room temperature for 1 h before conducting the final conductivity measurement. To determine the genetic relationships between cold hardiness and growth phenology, the date of needle flush (when needle primordia visibly separated in fascicles) was assessed on seedlings in the spring both of 2003 and 2004 as a surrogate for date of growth initiation. 5.2.3 Statistical Analysis For each sample, the index of injury (J), as described by Flint et al. (1967) was calculated as follows: Where I, is the index of injury (percent) resulting from exposure to temperature t, R, = L/Lk, Ra = LJLd, R, is the relative conductance of the sample exposed to temperature t, R0 is the relative conductance of the unfrozen control, L, is the conductance of the leachate from the sample frozen at temperature t, Lk is the conductance of the leachate from the sample frozen at temperature t then heat killed, L0 is the conductance of the leachate from the unfrozen sample, and Ld is the conductance of the leachate from the heat killed, unfrozen control sample. The test temperature that resulted in the widest range of injury scores was used for statistical analysis on each date. For fall and spring, this was the colder of the two test temperatures, and for summer and winter it was an intermediate temperature (Table 5.2). The SAS system, Version 8 (SAS Institute 1999) was used for all statistical analysis. Individual values from all seedlings were used in all analyses. PROC REG was used for regression analysis to determine the temperature at which I, — 50 (LT50) (lethal temperature causing 50% injury) for each test date. To test for differences among geographic regions and obtain variance components, PROC MIXED was used with the REML variance component estimator and the following model: \WKRt-R0) ( 1 - * . ) y^jkh Im = fi + rt + b{r)tj + gk + rgik + f(g)kl + e. ijklm 100 where yiJkim is the observed value for tree m in family / within region k in incomplete block j in rep /, p is the overall mean, r, is the effect of rep i, b(r)ij is the effect of incomplete block j nested within rep i, gk is the effect of geographic region k, rgik is the interaction of rep i and geographic region k, andy(g)« is the effect of family / nested within region k, and eijlkm is the random residual error. All terms were considered random except for region, which was fixed. For winter and summer cold injury, PROC GLM was used in a reduced model including only region as an effect as sample size was inadequate to test other main effects for these samples. Preliminary analysis showed that due to the limited number of trees tested from the EAOR region (n = 5), spring cold injury values from this region exerted undue influence on the data analysis. Therefore, it was treated as an outlier and excluded from the analysis of spring cold injury. To assess the level of genetic control of fall and spring cold injury, individual tree heritabilities were estimated as 2 2 where cr f is the variance due to families and oe is the variance due to error. Open-pollinated progeny are more closely related than half-sibs due to moderate inbreeding and correlated paternity in this species (Krakowski et al. 2003). To account for this, the additive genetic variation was approximated as three times the family variance instead of a coefficient of relatedness of four as is used for true half-sibs (Squillace 1974). To evaluate the genetic relationship between fall and spring cold injury, and between cold injury in both seasons and needle flush, genetic correlations (rA) were estimated between traits measured on the same seedlings according to Falconer (1989): 101 where CovF is the estimated family covariance between traits x and y, and a2Fx and o2Fy are the estimated family variances of traits x and y respectively. Pearson correlations were used to examine the relationship among seasonal cold injury traits and date of needle flush using both regional and family means. A sequential Bonferoni adjustment was used to ensure a = 0.05 over all comparisons within each group of correlations (Rice 1989). To determine the influence of source location and determine whether variation in cold injury is clinal, regional LSMEANS were obtained from PROC MIXED, and PROC CORR was used to examine their relationship with latitude, longitude, mean annual temperature, mean temperature of the coldest month (MTCM) and frost free period. Climatic variables for provenances north of 48° were obtained from PRISM climatic data corrected for local elevation using the Climate BC model described by Wang et al. (2006). For provenances south of 48°, climatic data were obtained from a climate model using the thin plate splines of Hutchinson (2000) as illustrated for North America by McKinney et al. (2001). 5 .3 R E S U L T S 5.3.1 Seasonal Variation The average cold hardiness of whitebark pine needles varied greatly throughout the year from an LT50 of approximately -9°C during active growth in summer to a level in winter well below the minimum temperature achievable in our programmable ultra-low temperature freezer (-70°C) (Figure 5.2). The LT50 in fall and.spring changed rapidly over time, and thus depended on the date of sampling. Trees from different regions differed in LT50 in the summer and fall, with the interior regions (WSCD and ESCD) able to withstand the lowest temperature (LT50—10.5 and -11.7 in summer and LT50=-36.9 and -39.9 in fall for WSCD and ESCD, respectively) followed by the northern (NOBC) region (LT50=-11.0 and -37.8 for summer and fall, respectively). Trees from the southern regions were the least hardy in summer (LT50=-8.6 for SWOR) and fall (LT50=-26.3 for 102 CALI). In contrast, in the spring, the California trees were the most hardy (LT50=-36.4) and the northern regions were the least (LT50=-22.0). Phenotypic correlations among regional and family means for cold injury traits in different seasons and with date of needle flush showed that at the regional level, although correlations of summer cold injury with both fall and spring cold injury were high, none of the seasonal cold injury traits were significantly correlated when p-values were adjusted for the number of correlations (Table 5.4). The correlations of fall and spring cold injury with date of needle flush were equal in magnitude but opposite in direction; however, only the correlation between fall cold injury and needle flush was significant. At the family level, fall and spring cold injury were weakly but significantly correlated, and both summer and fall cold injury were correlated with date of needle flush. 5.3.2 Differentiation among regions in fall and spring Significant differences were detected among geographic region means for cold injury in both fall and spring (Table 5.3), as well as summer (p = 0.03). The regional means showed a significant trend with hardiness increasing with latitude of seed origin in the fall (Figure 5.3a2). In the spring, there was a decreasing trend in hardiness with latitude that approached significance (p = 0.06) (Figure 5.3a4). Mean winter cold injury was less than 45% for all regions when tested at -70°C, and no differences were detected among regional means (p = 0.34). 5.3.3 Genetic control and correlations Estimates of individual heritabilities for cold injury were moderate for both spring (h2 = 0.18) and fall (h2 = 0.28), and the genetic correlation between the two tests was weak (rA = 0.18). Spring cold injury (tested in 2004) was correlated to date of needle flush in 2004 {rA = 0.34) but fall cold injury (tested in 2003) was not related to date of needle flush in 2003 (rA = 0.02). 103 5.3.4 Environmental effects on cold hardiness Regional least squares mean fall cold injury was strongly related to mean temperature of the coldest month (r = 0.81, p = 0.005) (Figure 5.3b2). Regional mean spring cold injury was not significantly related to any of the geographic or climate variables. 5.4 DISCUSSION Populations of temperate forest trees are often well differentiated for cold adaptation traits, and this differentiation is strongly associated with geographic and climatic gradients (Howe et al. 2003). For example, Rehfeldt et al. (1984) found significant differences in cold hardiness among provenances in western white pine (Pinus monticola Dougl. ex D. Don), also found in Pinus section Strobus (Belokon et al. 1998). In contrast, Thomas and Lester (1992) found significant variation in cold hardiness in P. monticola both among families within provenances, and between the coastal and interior areas of British Columbia, but not among provenances within these areas. In mountain hemlock (Tsuga mertensiana (Bong.) Carr.), which is sympatric with whitebark pine in much of its range, Benowicz et al. (2001) also found significant differences in fall cold hardiness among provenances. Similarly, we found significant variation among regions both for fall and spring cold injury, suggesting adaptation of whitebark pine populations to local climate. When fully acclimated, even just prior to dehardening (the winter test was performed only 6 weeks before the spring test) whitebark pine is hardy to temperatures exceeding our testing capabilities (<-70°C). Oohata and Sakai (1982), in a test of 41 Pinus species, found that almost all species from colder regions were more winter hardy than species from warmer regions and the degree of winter cold hardiness corresponded well to winter cold in the natural distribution of the species (Table 5.5). They also found that three other stone pine species (Pinus section Strobus subsection Cembrae) and two related North American five-needle pine species were well adapted to extreme winter cold. It is clear that whitebark pine, like other pines that experience low winter temperatures, has evolved to withstand this stress. The very high hardiness also suggests that whitebark pine has 1 0 4 evolved to not only withstand cold winter temperatures, but to withstand temperatures well below what it would experience throughout its range, typical of most temperate and boreal tree species (Sakai and Larcher 1987). In the harsh environments where it is found, the ability to survive low temperatures is likely an important factor in whitebark pine's ability to persist at the upper limits of treeline, surviving abiotic stresses that other species cannot, and giving it a competitive advantage in these locations. Once chilling requirements are met and seedlings start accumulating heat sum, whitebark pine loses cold hardiness rapidly prior to the initiation of growth. Seedlings were hardy to temperatures below -70°C in early February, yet preliminary testing to determine temperatures for the spring freeze test five weeks later resulted in injury exceeding 60% at -40°C. A subsample of trees tested one week later at -10, -15, and -20°C yielded mean injury indices of 25, 40, and 66%, respectively. The LT50 rose over 55°C in a period of six weeks leading up to growth initiation. Despite a small sample size, freeze testing of actively growing needles in June showed that whitebark pine was substantially more cold hardy at this stage in our test than most conifers previously tested. Freeze testing during the period of shoot elongation on other species has shown that they are only hardy to a minimum of approximately -5°C, and little or no variation among genetic groups has been reported (Cannell and Sheppard 1982, Dormling 1982, Glerum 1973, Li et al. 2003, Nilsson and Walfridsson 1995, Rikala and Repo 1987, Stevenson et al. 1999). In this study, at -9°C in June, during needle elongation, mean injury was 44% and ranged from 25% to 70% depending on region. In comparison, freeze testing in late August and early September, when needles are in the first stage of cold acclimation (Weiser 1970), resulted in approximately 50% damage at -10°C in both Pinus contorta var. latifolia Engelm (Rehfeldt 1980) and P. monticola (Thomas and Lester 1992). Cold injury to trees usually exhibits higher heritability in the spring than in the fall, and genetic variation in midwinter hardiness is low or not significant in many species (Howe et al. 2003). Reported heritability values for other conifers range from 0.16-0.44 for fall cold injury traits and from 105 0.45-0.78 with one report of 0.12 for spring cold injury traits (Howe et al. 2003). Heritability estimates for fall cold injury were within the range reported for other conifers, indicating similar levels of genetic control for this trait; however, spring cold injury appears to be under weaker genetic control in whitebark pine than in most conifers. The genetic correlation between fall and spring cold injury was low, indicating that these two traits are only weakly related genetically. O'Neill et al. (2000) also found a low genetic correlation between fall and spring cold injury in two-year-old Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco), and quantitative trait loci (QTL) mapping of cold hardiness in Douglas-fir found little colocation of QTLs for fall and spring cold hardiness (Jermstad et al. 2001). This implies that selecting for one of these traits would most likely not have a strong effect on the other trait. Spring cold injury is genetically correlated with bud flush in whitebark pine, as in most species (Howe et al. 2003). For example, O'Neill et al. (2000) reported genetic correlations of -0.82 to -0.90 in Douglas-fir seedlings. The correlations between regional mean cold injury and climatic data suggest that fall cold injury is under stronger differential selection among populations than spring cold injury. The correlations of fall and spring cold injury .with MTCM suggest that regions that experience lower winter temperatures have been under selection to develop cold hardiness earlier in the fall (Figure 5.3bl-2). A similar relationship was found in mountain hemlock, where northern sources that experienced lower MTCM suffered less fall cold injury in artificial freeze tests (Benowicz et al. 2001). However, in the test environment, colder regions appear to be the first to deharden in response to warmer temperatures in the spring, resulting in higher damage in spring freeze testing. Moving seed from an area that experiences milder winter temperatures (i.e., a more southern location) to a planting site with colder temperatures may increase the risk of fall frost damage. However, this risk may decline over the next century with climate change. Most of the variation in adaptation to fall cold is among regions, with little variation among families; whereas, in the spring, loss of cold hardiness is family specific. Cold injury in the spring is 106 likely to be influenced by late winter temperature conditions in a given year, because growth initiation and deacclimation are dependent on heat sum accumulation once chilling requirements are met (Morgenstern 1996). Regional and family mean correlations between fall and spring cold injury were nearly the same in magnitude, but opposite in direction (Table 5.4). The significant positive correlation at the family level may indicate that cold hardiness, in general, is family specific. Families that generally experience lower levels of cold injury in the fall also experience lower levels of cold injury in the spring, regardless of their source. Whitebark pine is well adapted to cold, and exhibits significant regional variation for fall and spring cold hardiness. The moderate heritability of these traits means that there is potential for increasing cold hardiness through selection and may reflect the strength of past selection, especially for spring cold injury. It is clear that cold injury is not a major problem for whitebark pine in its natural environment, particularly in comparison to the biotic agents white pine blister rust and mountain pine beetle (Arno and Hoff 1989). Screening and breeding for blister rust resistance will be the most urgent and productive activity for restoration of whitebark pine forests and maintenance of their ecological functions. Although cold hardiness is not likely to be a trait that will be of great concern for the future of whitebark pine, our results show that whitebark pine is well adapted to cold throughout its range, but regions differ significantly in cold injury. Long distance seed movement, especially from south to north should be avoided or considered carefully in order to minimize the possibility of damage due to early fall or late spring freeze events. In order to direct seed movement, an assessment of transfer risk using multivariate analysis with a number of quantitative traits in addition to cold hardiness measured on these seedlings has been done and will be reported elsewhere (Chapter 2). 107 5.5 A C K N O W L E D G E M E N T S The authors thank the USDA Forest Service regions 1, 5, and 6, the British Columbia Ministry of Forests, B.C. Parks E. C. Manning and Tweedsmuir Provincial Parks, and Bob Brett of Snowline Ecological Consulting, Whistler, B.C. for providing seed and the help of the many people in the Aitken lab who helped collect and prepare needles for artificial freeze testing, especially Joanne Tuytel, Christine Chourmouzis, Pia Smets, Dorothy Watson, Jodie Krakowski, and Milena Semproni. Climate data was provided by Dr. Tongli Wang and Dr. Gerald Rehfeldt. Funding for this study came from the British Columbia Forestry Investment Account through the Forest Genetics Council of B.C. to the Centre for Forest Gene Conservation. Thank you to Dr. Jeannette Whitton and Dr. Alvin Yanchuk and two anonymous reviewers for their helpful comments on earlier drafts of this manuscript. 108 Table 5.1. Geographic regions of whitebark pine tested for cold injury. Region Code # Families tested # Trees tested Latitude (deg.)a Longitude (deg.)a Mean Elevation (range) (m) M T C M " Northern B.C. NOBC* 4 39 54.9 -127.2 1400 (1231-1446) -10.9 B.C. Central Coast BCCC 2 22 52.5 -125.8 1447 (900-1916) -10.8 Southwestern B.C. SWBC* 49 522 50.0 -121.7 1908 (1785-2000) -9.3 Eastern Washington -Northern Idaho EWNI 20 205 48.8 -118.5 2024 (1846-2462) -8.8 West Side of the U.S. Continental Divide WSCD* 21 234 46.9 -112.8 2175 (2062-2338) -7.2 East Side of the U.S. Continental Divide ESCD* 23 273 45.4 -111.4 2518 (2154-2892) -9.4 Northern Oregon NOOR 3 28 45.4 -121.7 1969 -4.7 Eastern Oregon EAOR 2 25 44.7 -118.6 2338 -8.6 Southwestern Oregon SWOR* 32 314 42.5 -121.3 2351 (2074-2538) -5.0 California CAL* 4 36 38.7 -119.9 2813 (2754-2923) -4.7 * Regions tested for seasonal variation of cold hardiness a Median latitude and longitude of provenances within region bMean temperature of the coldest month Table 5.2. Dates and test temperatures of artificial freeze tests. Season Test Date Test temperatures (°C) # trees # families # regions Fall Oct. 8, 2003 -30, -40 a 353 157 10 Winter Feb.4,2003 -38, -46, -54, -62a, -70 30 28 6 Spring Mar. 17, 2004 -12, -22 a 454 159 10 Summer June 2, 2004 -3, -6, -9a, -12 54 41 6 a Temperature used for data analysis 109 Table 5.3. Sources of variation, p-values for F-statistics in ANOVA and % of total variance for fall and spring cold injury. Fall Spring Source" p-value % variation p-value % variation Rep 0.543 0 0.214 2.5 Region 0.001 12.1 0.016 2.0 Rep*Region 0.826 0 0.993 0 Block(Rep) 0.023 8.7 0.127 4.9 Family(Region) 0.116 9.1 0.102 6.3 Error 70.1 84.3 a see text for abbreviations Table 5.4. Phenotypic correlations among regional means (below diagonal) and family means (above diagonal) for cold injury in different seasons, with number of observations (n) in parentheses. F A M I L Y M E A N S R E G I O N A L Date of Needle M E A N S Summer Fall Winter Spring Flush 2004 Summer 0.24 (37) 0.17(28) 0.05 (36) 0.37* (45) Fall 0.80 (6) -0.05 (23) 0.28* (155) 0.38* (156) Winter -0.04 (6) -0.04 (6) 0.02 (22) -0.08 (28) Spring -0.25 (6) -0.26 (10) 0.33 (6) -0.10(159) Date of Needle 0.66 (6) 0.70* (10) -0.32 (6) -0.70 (9)a Flush 2004 * significant at a=0.05 with sequential Bonferroni adjustment for number of comparisons (n=10) a EAOR region not included due to small sample size with region (n=5) exerting undue influence on correlation. 110 Table 5.5. Freezing tolerance of species of Pinus in the subgenus Strobus (adapted from Oohata and Sakai 1982). Strobus Species Mean Warmth Mean Cold M C M T Freezing Subsection Index" Index" Tolerance (°C/month) (°C/month) (°C) Cembrae P. koraiensis 42.8 -78.4 -14.8 -70 P. pumila 20.0 -115.6 -18.3 <-70* P. cembra 23.7 -53.2 -7.3 <-70* P. albicaulis 6 17.4 -60.5 -8.0 <-70* Strobi P. strobus 62.2 -41.0 -6.7 <-80* P. monticola 46.6 -21.3 -1.4 <-80* P. ayacahuite 94.7 0.0 9.8 -15 P. peuce 50.6 -35.3 -5.7 -40 P. griffithii 71.0 -17.2 -2.2 -35 Cembroides P. cembroides 113.4 -1.2 8.0 -12 Gerardianae P. bungeana 81.0 -33.6 -7.2 -30 a Sum of estimated monthly mean temperatures above or below 5°C b This study * Uninjured at lowest temperature 111 Figure 5.1. Range of whitebark pine with designated regions and parental source locations of trees tested for cold hardiness. 112 Figure 5.2. Seasonal change in mean LT50 of whitebark pine. Estimates of winter cold hardiness are truncated at the minimum testing temperature (-70°C). 113 Figure 5.3. Regional LSmean cold injury vs. latitude (a-1 through 4) at four test dates and vs. mean temperature of the coldest month (MTCM) for fall and spring (b-1,2) indicating clinal variation in fall cold hardiness associated with temperature. Error bars are ± 2 standard errors. .—. 100 i £> 80 -p 60 o 40 X a> 20 "a c 0 a-1 Summer n.s. 0 35 40 45 50 55 Latitude (Deg. N.) 60 100 -i 0 s* £> 80 Inju 60 »»-o 40 X a> T3 20 c 0 a-2 Fall R = 0.485 p = 0.026 35 40 45 50 55 Latitude (Deg. N.) 60 100 -i 80 p 60 o 40 X CO T J 20 c 0 a-3 Winter n.s. 0 4> 35 40 45 50 55 Latitude (Deg. N.) 60 ~ 100 80 p X c 60 40 20 0 a-4 Spring R z = 0.4189 p = 0.061 35 40 45 50 55 Latitude (Deg. N.) 60 _ 100 ~ 80 •§. 60 *& 40 x a> 20 c 0 b-1 Spring b-2 Fall n.s. 0 -12 -10 -8 -6 MTCM (Deg. C.) 100 -i 80 p 60 **-O 40 X a> 20 •o c 0 Rz = 0.654 p = 0.005 -12 -10 - 8 - 6 -4 MTCM (Deg. C.) 114 5.6 R E F E R E N C E S Aitken, S. N., and M . Hannerz. 2001. Genecology and gene resource management strategies for conifer cold hardiness. Pages 23-53 in F. J. Bigras and S. J. Colombo, editors. Conifer Cold Hardiness. Kluwer Academic Publishers, The Netherlands. Arno, S. F., and R. J. Hoff. 1989. Silvics of whitebark pine (Pinus albicaulis). USDA For. Serv. Gen. Tech. Rep. INT-253. Belokon, M . M. , D. V. Politov, Y. S. Belokon, K. V. Krutovskii, O. P. Malyuchenko, and Y. P. Altukhov. 1998. Genetic differentiation in white pines of section strobus: Isozyme analysis data. Doklady Akademii Nauk 358:699-702. Benowicz, A., S. J. L'Hirondelle, and Y. A. El-Kassaby. 2001. Patterns of genetic variation in mountain hemlock (Tsuga mertensiana (Bong.) Carr.) with respect to height growth and frost hardiness. Forest Ecol. and Manag. 154:23-33. Burr, K. E., A. Eramian, and K. Eggleston. 2001. Growing whitebark pine seedlings for restoration. Pages 325-345 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press. Campbell, E. M. , and J. A. Antos. 2000. Distribution and severity of white pine blister rust and mountain pine beetle on whitebark pine in British Columbia. Can. J. For. Res. 30:1051-1059. Cannell, M. , and L. Sheppard. 1982. Seasonal changes in the frost hardiness of provenances of Picea sitchensis in Scotland. Forestry 55:137-153. Critchfield, W. B., and E. L. Little, Jr. 1966. Geographic distribution of the pines of the world. USDA For. Serv. Misc. Publ. 991. CSIRO. 1996. Alpha+; Experimental designs for variety trials and many-treatment experiments. Canberra, Australia. 31 pages. Dormling, I. 1982. Frost resistance during bud flushing and shoot elongation in Picea abies. Silva Fenn. 16:167-177. Falconer, D. S. 1989. Introduction to quantitative genetics; third edition, third edition. John Wiley & Sons, New York. Flint, H. L., B. R. Boyce, and D. J. Beattie. 1967. Index of injury - a useful expression of freezing injury to plant tissues as determined by the electrolytic method. Can. Jo. Plant Sci. 47:229-230. Glerum, C. 1973. Annual trends in frost hardiness and electrical impedance for seven coniferous species. Can. J. Plant Sci. 53:881-889. Hannerz, M. , S. N . Aitken, J. N . King, and S. Budge. 1999. Effects of genetic selection for growth on frost hardiness in western hemlock. Can. J. For. Res. 29:509-516. Howe, G. T., S. N. Aitken, D. B. Neale, K. D. Jermstad, N. C. Wheeler, and T. H. H. Chen. 2003. From genotype to phenotype: unraveling the complexities of cold adaptation in forest trees. Can. J. Bot. 81:1247-1266. Hutchinson, M . F. 2000. ANUSPLIN Version 4.1 user's guide. Australian National University, Centre for Resource and Environmental Studies, Canberra. Jermstad, K. D., D. L. Bassoni, N . C. Wheeler, T. S. Anekonda, S. N. Aitken, W. T. Adams, and D. Neale. 2001. Mapping of quantitative trait loci controlling adaptive traits in coastal Douglas-fir. II. Spring and fall cold-hardiness. Theor. Appl. Genet. 102:1152-1158. Krakowski, J., S. N. Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. Li , C , A. Vihera-Aarnio, T. Puhakainen, O. Junttila, P. Heino, and E. Tapio Palva. 2003. Ecotype-dependent control of growth, dormancy and freezing tolerance under seasonal changes in Betula pendula Roth. Trees 17:127-132. Little, E. L., Jr. 1971. Atlas of United States trees, volume 1, conifers and important hardwoods. USDA For. Serv. Misc. Publ. 1146. Washington, D.C. Mahalovich, M . F., and G. A. Dickerson. 2004. Whitebark pine genetic restoration program for the intermountain west (United States). Pages 259 in R. A. Sniezko, S. Samman, S. E. Schlarbaum, 115 and H. B. Kreiebel, editors. Breeding and genetic resources of five-needles pines: growth, adaptability and pest resistance; Proceedings of IUFRO Working Party 2.02.15. July 24-27, 2001. Medford, OR. USDA For. Serv. Rocky Mountain Research Station, Ft. Collins, CO. McKinney, D. W., M . F. Hutchinson, J. L. Kesteven, and L. A. Venier. 2001. Canada's plant hardiness zones revisited using modern climate interpolation techniques. Can. J. Plant Sci. 81:129-143. Morgenstern, E. K. 1996. Geographic variation in forest trees. UBC Press, Vancouver. Nilsson, J.-E., and E. A. Walfridsson. 1995. Phenological variation among plus-tree clones of Pinus sylvestris (L.) in northern Sweden. Silvae Genet. 44:20-28. O'Neill, G. A., S. N . Aitken, and W. T. Adams. 2000. Genetic selection for cold hardiness in coastal Douglas-fir seedlings and saplings. Can. J. For. Res. 30:1799-1807. Oohata, S., and A. Sakai. 1982. Freezing resistance and thermal indices with reference to distribution of the genus Pinus. Pages 694 in P. H. Li and A. Sakai, editors. Plant cold hardiness and freezing stress, vol. II. Academic Press, London New York. Patterson, H. D., and E. R. Williams. 1976. A new class of resolvable incomplete block designs. Biometrika 63:83-92. Rehfeldt, G. E. 1980. Cold acclimation in populations of Pinus contorta from the northern Rocky Mountains. Botanical Gazette 141:458-463. Rehfeldt, G. E., R. J. Hoff, and R. J. Steinhoff. 1984. Geographic patterns of genetic variation in Pinus monticola. Bot. Gaz. 145:229-239. Rice, W. R. 1989. Analyzing tables of statistical tests. Evolution 43:223-225. Rikala, R., and T. Repo. 1987. Frost resistance and frost damage in Pinus sylvestris seedlings during shoot elongation. Scand. J. For. Res. 2:433-440. Sakai, A., and W. Larcher. 1987. Frost survival of plants. Responses and adaptation to freezing stress. Ecological Studies Vol. 62. Springer-Verlag, Berlin. SAS Institute, I. 1999. The SAS system for windows. SAS Institute, Inc., Cary, North Carolina. Squillace, A. E. 1974. Average genetic correlations among offspring from open-pollinated forest trees. Silvae Genet. 23:149-156. Stevenson, J. F., B. J. Hawkins, and J. H. Woods. 1999. Spring and fall cold hardiness in wild and selected seed sources of coastal Douglas-fir. Silvae Genet. 48:29-34. Thomas, B. R., and D. T. Lester. 1992. An examination of regional, provenance, and family variation in cold hardiness of Pinus monticola. Can. J. For. Res. 22:1917-1921. Tomback, D. F., S. F. Arno, and R. E. Keane. 2001. The compelling case for management intervention. Pages 3-25 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Tomback, D. F., and K. C. Kendall. 2001. Biodiversity losses: the downward spiral. Pages 243-262 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D. C. Wang, T., A. Hamann, D. L. Spittlehouse, and S. N. Aitken. 2006. Development of scale-free climate data for western Canada for use in resource management. Int. J. Climatol. 26:383-397. Weaver, T. 2001. Whitebark pine and its environment. Pages 41-73 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Weiser, C. J. 1970. Cold resistance and injury in woody plants. Science 169:1269-1278. 116 CHAPTER 6: THESIS CONCLUSIONS 6.1 I N T R O D U C T I O N Whitebark pine (Pinus albicaulis Engelm.) has been severely impacted throughout its range by successional replacement as a result of fire exclusion and mortality due to mountain pine beetle (Dendroctonus ponderosae) and white pine blister rust (caused by the fungus Cronartium ribicola) (Campbell and Antos 2000, Tomback et al. 2001, Campbell and Antos 2003). It is an ecologically important species in high-elevation, subalpine ecosystems (Tomback et al. 2001), and represents an important component of our bioheritage as the only North American member of the stone pines (Pinus subsection Cembrae) (Price et al. 1998; but see Gernandt et al. 2005). Conservation is needed to protect its remaining genetic resources, and in many areas restoration will be necessary to stop or reverse the effects of blister rust and to maintain whitebark pine on the landscape (McCool and Freimund 2001). These conservation and restoration efforts should be guided by knowledge of the genetic diversity, population differentiation and degree of local adaptation. Genetic diversity of putatively selectively neutral molecular markers and analysis of the mating system of whitebark pine have previously been published (Furnier et al. 1987, Yandell 1992, Jorgensen and Hamrick 1997, Stuart-Smith 1998, Rogers et al. 1999, Bruederle et al. 2001, Richardson et al. 2002, Krakowski et al. 2003), but no information has been available regarding genetic diversity of quantitative traits and local adaptation. In addition, there have been no studies on the effects of blister rust on genetic diversity. Krakowski et al. (2003) reported that whitebark pine experiences higher rates of inbreeding than most wind-pollinated conifers due to the clumpy growth habit resulting from seed dispersal by the Clark's Nutcracker (Nucifragia Columbiana). Intermediate rates of outcrossing are less common in wind-pollinated plants due to the selective disadvantage usually experienced by inbred individuals (inbreeding depression) (Vogler and Kalisz 2001). It was not previously known whether whitebark pine experiences inbreeding depression or whether is has evolved to withstand these higher levels of inbreeding by purging deleterious alleles. 117 The goal of this thesis was to fill some of these gaps of scientific knowledge of the genetics of whitebark pine. I employed a common garden study to assess genetic variation and local adaptation in quantitative traits and to determine guidelines for seed movement in restoration efforts. I was also able to compare estimates of genetic differentiation in quantitative traits to previously published estimates based on molecular markers to determine if there is evidence of natural selection driving geographic differentiation. In addition, I used isozyme analysis of seed tissue to confirm the mating system, and by relating these results to those from the common garden experiment, was able to determine the presence and strength of inbreeding depression in quantitative traits. Isozyme analysis of bud tissue of three age cohorts from 14 sites enabled me to investigate the relationships of inbreeding and blister rust with genetic diversity. I also studied seasonal variation, genetic control and regional differentiation in fall and spring cold injury using artificial freeze tests. 6.2 M A J O R FINDINGS 6.2.1 Genetic diversity and local adaptation of quantitative traits Results from the common garden experiment revealed significant effects of soil temperature on height growth and survival. In contrast to my a priori expectations, height growth and survival were both higher in the colder treatment. For a species adapted to the cold climates of subalpine ecosystems, the relatively mild temperatures of coastal British Columbia were more stressful, even under well-watered conditions. Significant variation was detected among provenances for height increment, date of needle flush, and fall cold injury in both the ambient and cold treatments, as well as for biomass in the ambient treatment. Genetic differentiation among populations ( Q S T ) was weak to moderate for growth traits (height and biomass), but strong for cold adaptation related traits (date of needle flush and cold injury). Q S T values for all traits were higher than previously published estimates of population differentiation based on molecular markers ( F S T ) - Differential selection on populations driving local adaptation can be inferred from the higher values of Q S T relative to F S T -118 In addition to provenance variation, strong seasonal patterns were found in cold hardiness, and significant variation was detected among broad geographic regions in all seasons except winter. Cold hardiness of whitebark pine ranged from -9° in early summer to below -70° in winter. In summer and fall, trees from northern and interior regions experienced the lowest cold injury, and trees from California experienced the highest. In the spring, these geographic patterns of injury were reversed. Genetic control of fall and spring cold injury was moderate (h2 = 0.28 and 0.18, respectively), and their genetic correlation was weak (rA = 0.18). Spring cold injury was genetically correlated with date of needle flush (rA = 0.34), and fall cold injury was strongly correlated with mean temperature of the coldest month in the parental environment (r=0.81). Whitebark pine is well adapted to withstand the cold temperatures it may experience in its natural environment, and is hardier to cold than most other temperate forest trees. Canonical correlation analysis showed that mean temperature of the coldest month (MTCM) appears to be the primary climatic variable driving local adaptation in quantitative traits, specifically date of needle flush and fall cold injury. The positive relationship showed that areas that experience colder winter temperatures (more northerly provenances) tend to flush earlier in the spring and harden to cold earlier in the fall. Earlier flushing is likely an adaptation to extend the growing season, albeit at the risk of higher levels of cold injury in the spring. The analyses also showed a positive relationship between length of the growing season (FFP) and the growth traits height increment and biomass. Areas with longer FFP tended to grow taller and produce more biomass. The relationship between MTCM and date of needle flush was used to develop guidelines for seed transfer. Seed can be moved between areas differing in MTCM by 1.2°C, which corresponds to approximately 3° in latitude or 700 m in elevation. 119 6.2.2 Mating System and Inbreeding Depression Isozyme analysis of seed tissues from three geographic regions (Oregon, Montana, and southern British Columbia) was used to genotype parent trees and progeny arrays. Estimated mean inbreeding coefficients were higher in progeny than in parents in all three regions, suggesting selection against more homozygous, inbred individuals, a pattern common in conifers. Analysis of the mating system confirmed that whitebark pine experiences mixed mating. Regional multilocus estimates of outcrossing rate (tm) varied from 73 to 98% with a mean over all populations of 86%, which is slightly lower than most wind-pollinated conifers but higher than previous estimates for this species (Krakowski et al. 2003). Family mean outcrossing rates varied widely, but their distribution showed that whitebark pine is predominantly outcrossing. However, some individuals experience high levels of inbreeding, at least in the years that seed was sampled. Significant evidence of biparental inbreeding and substructure among paternal contribution (fp(m)) was found only in the southern British Columbia region. In addition, this region had the lowest mean outcrossing rate (73%). The area where the two populations in this region were sampled was glaciated during the last glacial maximum and recolonized from areas to the south. Geographic variation in mating system may be the result of recolonization patterns and processes, where in previously glaciated areas, genetic diversity was reduced due to a founder effect, and subsequent populations evolved to withstand higher levels of inbreeding in small founder populations relative to populations in non-glaciated areas. Southern British Columbia was also the only region in which significant inbreeding depression was detected. Of the quantitative traits measured in the common garden experiment, inbreeding depression was detected only for biomass in the ambient treatment. Using the mean equilibrium inbreeding coefficient [Fe = (l-t„)/(l+tm)] of the 20 families from southern British Columbia (Fe=0.25) growing in the common garden experiment, a decrease in biomass in this region of approximately 20% is predicted due to inbreeding. 120 6.2.3 Relationship of genetic structure with inbreeding and white pine blister rust infection. A previous study of genetic diversity of whitebark pine reported a significant increase in fixation index (Fis) and significant decrease in observed heterozygosity (H0) from south to north and from east to west in British Columbia and Alberta, Canada (Krakowski et al. 2003). These trends correspond to geographic patterns in the level of blister rust infection, which generally is highest in the southern and eastern part and lowest in the northern and western part of the range of whitebark pine in Canada (Campbell and Antos 2000, Zeglen 2002). It was suggested that trends in F i s and observed heterozygosity (H0) provided circumstantial evidence for selection by the pathogen against more homozygous genotypes or for increased tolerance in more heterozygous genotypes. This hypothesis was tested by estimating F i s and H 0 for three age cohorts (seedling, young, and mature trees) from 14 sites that varied in level of blister rust infection. I found significant evidence of inbreeding (F i s > 0) in all age cohorts. When sites were stratified by the level of blister rust infection, there was significant evidence of inbreeding in all cohorts when the infection level was moderate or high, but only in the seedling cohort when infection was low. In the seedling cohort, H 0 was significantly lower and F i s significantly higher than in the young and mature cohorts when infection level was low. Therefore, when selection pressure from blister rust is weak, it appears that more heterozygous individuals are favoured; however, when pressure is higher, more homozygous individuals are fitter. This shows that selection due to inbreeding and blister rust are affecting genetic diversity in different ways. Over time, selection against inbred individuals usually results in a decrease in heterozygote deficiency, however, more homozygous individuals appear to have higher fitness when challenged by rust. This maintains a heterozygote deficiency in mature trees in stands with high infection levels more typical of that in seed and young seedlings. H 0 and FjS showed no relationship with level of blister rust infection in the seedling and young cohort, but in the mature cohort, H 0 showed a decreasing trend and F i s an increasing trend when regressed on level of rust infection. This relationship shows that infection by 121 blister rust is affecting genetic structure of whitebark pine, and could indicate increased expression of recessive genes for resistance as a result of inbreeding. 6.3 F U T U R E R E S E A R C H The results of the research presented in this thesis have filled some of the knowledge gaps regarding the genetics of whitebark pine. However, there are still gaps, and additional questions raised by these results. The quantitative traits investigated in this research have provided evidence of differential selection driving local adaptation, and been useful in establishing seed transfer guidelines. However, an important suite of traits I did not assess deal with potential differences in seasonal growth patterns. In my study, it was not possible to determine the average daily growth rate, duration, and cessation of growth, which are traits that have been useful in assessing genetic differentiation in a number of other conifers such as Pinus contorta, P. ponderosa, P. monticola, Picea engelmannii, Larix occidentalis, Tsuga mertensiana, and the Abies procera/Abies magnifica complex (Rehfeldt et al. 1984, Rehfeldt 1988, Sorensen et al. 1990, Rehfeldt 1991, 1994, 1995, Benowicz et al. 2001). In addition, without data on growth cessation, it was not possible to examine its relationship between fall cold injury, traits which are often genetically related (Howe et al. 2003). Another suite of traits that may be useful in understanding the physiological mechanisms whitebark pine has used to adapt to local population are gas exchange parameters and the variables that can be calculated from them (e.g. maximum net instantaneous photosynthetic rate, transpiration rate, intercellular-to-ambient CO2 concentration ratio, mesophyll conductance, stomatal conductance, and photosynthetic water use efficiency) (Benowicz and El-Kassaby 1999, Benowicz et al. 2000). I have preliminary evidence that there are differences in photosynthetic rate from north to south, however, a more extensive sample is needed to establish these differences with certainty. Clinal patterns may be a function of test environments as well as population genotypes. Higher confidence can be placed in tests where differences in population trait means and genotype-by-environment interactions are both associated with environmental differences at population origins 122 (Campbell and Sorensen 1978). Replication of this type of common garden experiment in multiple locations throughout the range of whitebark pine would determine the existence and strength of genotype-by-environment interactions. Determination of these interactions can be used to develop norms of reaction for whitebark pine to climatic or other environmental variables, and may be helpful in refining seed transfer guidelines or determining proper seed sources for restoration plantings. In addition to the lack of multiple locations with different environments, this common garden experiment was also grown outside of the natural range of the species. Tests in nursery beds have been criticized because such environments may not be natural, and differences among populations in these artificial conditions may have little adaptive significant and relevance to field conditions (Campbell and Sorensen 1978). The coastal environment where this common garden experiment was established was substantially warmer and appeared to be more stressful for this species than its native high elevation, subalpine ecosystems. However, differences revealed among populations in any environment must have a genetic basis, and thus Q S T estimates from non-native environments can still be used to guide seed transfer and restoration decisions. Establishment of field tests at high elevations over a range of latitudes and longitudes would nonetheless provide additional insights into local adaptation of whitebark pine populations. Based on previously published estimates, it was thought that whitebark pine had a substantially lower outcrossing rate than most wind-pollinated conifers (73%) (Krakowski et al. 2003). While I have shown that over all regions, the mean outcrossing rate is closer to other pines (86%), there is a great deal of variation within the species both among populations and among seed parents within populations. I have suggested that the lower outcrossing rate in the southern British Columbia region is a result of post-glacial recolonization. Based on the current geographical distribution mtDNA haplotypes, Richardson (2002) hypothesized that Canadian populations were colonized from a refugia in central Idaho. Further investigations of outcrossing rates across a latitudinal gradient from areas that were both glaciated and ice-free during the last glacial maximum would allow this hypothesis to be tested. If outcrossing rates are lower in glaciated areas, this may 123 support the idea that a bottleneck during post-glacial colonization led to higher inbreeding in small founder populations relative to non-glaciated areas. In addition, comparison of outcrossing rates between single trees and stems within tree clumps will help to determine the potential effects of growth structure on inbreeding. No inbreeding depression was detected in most of the traits analysed in this study. However, lack of evidence does not necessarily mean that there is no inbreeding depression. The proportion of outcrossed individuals at the time of fertilization may be lower than the measured outcrossing rate if inbreeding depression has removed individuals from the measured population (Galloway et al. 2003). Seed that did not germinate and mortality after emergence prior to analyses may have led to higher measured outcrossing rates which would effect the estimation of inbreeding coefficient. This consequently would affect the relationship with measured trait means and the expression of inbreeding depression. Controlled pollinations can be used to generate crosses with different inbreeding coefficients from the same individuals by selfing, outcrossing, and mating with related individuals. Family mean performance in quantitative traits can then be related to inbreeding coefficient to calculate inbreeding depression. In addition, from these crosses, it will be possible to determine the primary selfing rate, and the relative self-fertility based on these crosses can also be used to estimate the genetic load (number of lethal equivalents per zygote) (Sorensen 1969). Information on the genetic load and inbreeding depression may be useful in inferring whether chronic inbreeding due to its clumpy growth structure has led to purging of deleterious alleles in whitebark pine, and how inbreeding might affect a breeding program for blister rust resistance. Identifying individuals resistant to white pine blister rust is likely to be crucial in the conservation and restoration of whitebark pine. The extensive experience gained from white pine blister rust resistance breeding programs for Pinus monticola and P. lambertiana in Idaho, Oregon, and California (McDonald et al. 2004) provides an established model on which to base a breeding program for whitebark pine. I found some evidence that more homozygous individuals may have higher fitness when challenged by blister rust, possibly due to the expression of recessive resistance 124 genes. This can be tested by using artificial inoculation of seedlings from open pollinated seed from phenotypically resistant individuals to test for heritable resistance and whether genetic control of the mechanisms is by single or multiple genes. Segregation ratios in the tested seedlings will help to determine inheritance of these and possibly other resistance mechanisms, and selected resistant parent trees can then be used in a breeding program for resistance. If recessive mechanisms are identified, information regarding the effects of inbreeding depression will be important in designing resistance breeding programs. For example, if whitebark pine experiences little or no inbreeding depression in most quantitative traits, then a system of inbreeding could be effective in increasing expression of these mechanisms in seedlings to be used in restoration plantings. Time to reproductive maturity in whitebark pine can be as long as 50 years (Arno and Hoff 1989), so growing resistant individuals from seed for use in a breeding program is not likely to be efficient. It may be possible to graft scion from identified resistant parent trees onto young rootstock of whitebark pine or possibly related species such as Pinus monticola or P. strobus. This would shorten the time needed to produce resistant seed for restoration. 125 6.4 R E F E R E N C E S Arno, S. F., and R. J. Hoff. 1989. Silvics of whitebark pine (Pinus albicaulis), Gen. Tech. Rep. INT-GTR-253. U.S. Department of Agriculture, Forest Service, Intermountain Research Station, Ogden, UT. Benowicz, A., and Y. A. El-Kassaby. 1999. Genetic variation in mountain hemlock (Tsuga mertensiana Bong.): quantitative and adaptive attributes. Forest Ecol. Manag. 123:205-215. Benowicz, A., R. D. Guy, and Y. A. El-Kassaby. 2000. Geographic pattern of genetic variation in photosynthetic capacity and growth in two hardwood species from British Columbia. Oecologia 123:168-174. Benowicz, A., S. J. L'Hirondelle, and Y. A. El-Kassaby. 2001. Patterns of genetic variation in mountain hemlock (Tsuga mertensiana (Bong.) Carr.) with respect to height growth and frost hardiness. Forest Ecol. Manag. 154:23-33. Bruederle, L. P., D. P. Rogers, K. V. Krutovskii, and D. V. Politov. 2001. Population genetics and evolutionary implications. Pages 137-158 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. Campbell, E. M. , and J. A. Antos. 2000. Distribution and severity of white pine blister rust and mountain pine beetle on whitebark pine in British Columbia. Can. J. For. Res. 30:1051-1059. Campbell, E. M. , and J. A. Antos. 2003. Postfire succession in Pinus albicaulis - Abies lasiocarpa forests of southern British Columbia. Can. J. Bot. 81:383-397. Campbell, R. K., and F. C. Sorensen. 1978. Effect of test environment on expression of clines and on delimitation of seed zones in Douglas-fir. Theor. Appl. Genet. 51:233-246. Furnier, G. R., P. Knowles, M . A. Clyde, and B. P. Dancik. 1987. Effects of avian seed dispersal on the genetic structure of whitebark pine populations. Evolution 41:607-612. Galloway, L. F., J. R. Etterson, and J. L. Hamrick. 2003. Outcrossing rate and inbreeding depression in the herbaceous autotetrapoid, Campanula americana. Heredity 90:308-315. Gernandt, D. S., G. Geada Lopez, S. Ortiz Garcia, and A. Liston. 2005. Phylogeny and classification of Pinus. Taxon 54:29-42. Howe, G. T., S. N . Aitken, D. B. Neale, K. D. Jermstad, N. C. Wheeler, and T. H. H. Chen. 2003. From genotype to phenotype: unraveling the complexities of cold adaptation in forest trees. Can. J. Bot. 81:1247-1266. Jorgensen, S. M„ and J. L. Hamrick. 1997. Biogeography and population genetics of whitebark pine, Pinus albicaulis. Can. J. For. Res. 27:1574-1585. Krakowski, J., S. N. Aitken, and Y. A. El-Kassaby. 2003. Inbreeding and conservation genetics in whitebark pine. Conserv. Genet. 4:581-593. McCool, S. F., and W. A. Freimund. 2001. Threatened landscapes and fragile experiences: conflict in whitebark pine restoration. Pages 263-284 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities: ecology and restoration. Island Press, Washington, D.C. McDonald, G. I., P. Zambino, and R. A. Sniezko. 2004. Breeding rust-resistant five-needle pines in the western United States: lessons from the past and a look to the future. Pages 28-50 in R. A. Sniezko, S. Samman, S. E. Schlarbaum, and H. B. Kreiebel, editors. Breeding and genetic resources of five-needles pines: growth, adaptability and pest resistance; Proceedings of IUFRO Working Party 2.02.15. RMRS-P-32. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station, Ft. Collins, CO. Price, R. A., A. Liston, and S. H. Strauss. 1998. Phylogeny and systematics of Pinus. Pages 527 in D. M . Richardson, editor. Ecology and biogeography of Pinus. Cambridge University Press, Cambridge. 126 Rehfeldt, G. E. 1988. Ecological genetics of Pinus contorta from the Rocky Mountains (USA): a synthesis. Silvae Genet. 37:131-135. Rehfeldt, G. E. 1991. A model of genetic variation for Pinus ponderosa in the Inland Northwest (U.S.A.): applications in gene resource management. Can. J. For. Res. 21:1491-1500. Rehfeldt, G. E. 1994. Adaptation oi Picea engelmannii populations to the heterogeneous environments of the Intermountain West. Can. J. Bot. 72:1197-1208. Rehfeldt, G. E. 1995. Genetic variation, climate models and the ecological genetics of Larix occidentalis. Forest Ecol. Manag. 78:21-37. Rehfeldt, G. E., R. J. Hoff, and R. J. Steinhoff. 1984. Geographic patterns of genetic variation in Pinus monticola. Bot. Gaz. 145:229-239. Richardson, B. A., S. J. Brunsfeld, and N. B. Klopfenstein. 2002. DNA from bird-dispersed seed and wind-disseminated pollen provides insights into postglacial colonization and population genetic structure of whitebark pine (Pinus albicaulis). Mol. Ecol. 11:215-227. Rogers, D. L., C. I. Millar, and R. D. Westfall. 1999. Fine-scale genetic structure of whitebark pine (Pinus albicaulis): Associations with watershed and growth form. Evolution 53:74-90. Sorensen, F. C. 1969. Embryonic genetic load in coastal Douglas-fir, Pseudotsuga menziesii var. menziesii. Am. Nat. 103:389-398. Sorensen, F. C , R. K. Campbell, and J. F. Franklin. 1990. Geographic variation in growth and phenology of seedlings of the Abies procera/A. magnifica complex. Forest Ecol. and Manag. 36:205-232. Stuart-Smith, G. J. 1998. Conservation of whitebark pine in the Canadian Rockies: blister rust and population genetics. M . Sc. University of Alberta, Edmonton. Tomback, D. F., S. F. Arno, and R. E. Keane. 2001. The compelling case for management intervention. Pages 3-25 in D. F. Tomback, S. F. Arno, and R. E. Keane, editors. Whitebark pine communities; ecology and restoration. Island Press, Washington, D.C. Vogler, D. W., and S. Kalisz. 2001. Sex among the flowers: the distribution of plant mating systems. Evolution 55:202-204. Yandell, U. G. 1992. An allozyme analysis of whitebark pine (Pinus albicaulis Engl.). M . Sc. University of Nevada, Reno. Zeglen, S. 2002. Whitebark pine and white pine blister rust in British Columbia, Canada. Can. J. For. Res. 32:1265-1274. 127 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075061/manifest

Comment

Related Items