UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The complex socio-spatial architecture of Rhizopogon spp. mycorrhizal networks in xeric and mesic old-growth… Beiler, Kevin Jon 2011

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

Item Metadata

Download

Media
24-ubc_2012_spring_beiler_kevin.pdf [ 17.92MB ]
Metadata
JSON: 24-1.0072536.json
JSON-LD: 24-1.0072536-ld.json
RDF/XML (Pretty): 24-1.0072536-rdf.xml
RDF/JSON: 24-1.0072536-rdf.json
Turtle: 24-1.0072536-turtle.txt
N-Triples: 24-1.0072536-rdf-ntriples.txt
Original Record: 24-1.0072536-source.json
Full Text
24-1.0072536-fulltext.txt
Citation
24-1.0072536.ris

Full Text

THE COMPLEX SOCIO-SPATIAL ARCHITECTURE OF RHIZOPOGON SPP.  MYCORRHIZAL NETWORKS IN XERIC AND MESIC OLD-GROWTH INTERIOR DOUGLAS-FIR FOREST PLOTS    by  Kevin Jon Beiler  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  (Vancouver)     December 2011  ? Kevin Jon Beiler, 2011  iiAbstract  Mycorrhizal networks (MNs) can influence tree establishment and resource competition but little is known regarding their underlying architecture in situ.  This study examined the socio-spatial architecture of MNs between Rhizopogon spp. genets and interior Douglas-fir (Pseudotsuga menziesii var. glauca (Beissn.) Franco) trees in an old-growth forest.  MN features were contrasted between plots with xeric versus mesic soil moisture regimes as a proxy for changes in site water stress anticipated with climate change.  My objectives were to: (1) describe the fine-scale spatial patterns and autecological traits of R. vesiculosus and R. vinicolor mycelia systems and compare these between xeric and mesic plots; (2) describe the spatial patterns and architecture of Rhizopogon spp. MNs at the forest stand scale; (3) contrast MN architectures between phytocentric and mycocentric perspectives and between xeric and mesic plots, and identify critical determinants of MN architectures.  Rhizopogon vesiculosus mycelia occurred deeper, were more spatially prolific, and colonized more tree roots than R. vinicolor mycelia.  Both species were associated with moist microsites within plots, and had more prolific mycelia in mesic compared to xeric plots.  The occurrence of R. vesiculosus shifted in the presence of R. vinicolor towards deeper soil horizons, suggesting competition and foraging strategy are important for niche partitioning between these species.  At the forest stand scale, Rhizopogon spp. genets spanned tens of metres and colonized up to 19 trees, but R. vesiculosus genets were larger and linked more trees than R. vinicolor genets.  Multiple tree cohorts were linked, with saplings and mature trees sharing the same fungal genets.  Across all plots, the physical size of individual trees or fungal genets was positively related to their MN connectivity.  This together with size asymmetries among different genets and trees resulted in the self-organization of complex, hierarchical scale-free MN architectures.  The MNs appear robust to random perturbations but susceptible to the loss of large trees or fungal genets.  No MN structural differences were found between phytocentric and mycocentric models or between xeric versus mesic plots.  The pervasive mycelia and extensive MNs formed by R. vesiculosus and R. vinicolor could influence interior Douglas-fir stand dynamics and resistance to water stress.      iii Preface  In chapter 2, I developed the project under the advisement of committee members, conducted field and laboratory work and data analysis, and prepared a manuscript for publication. Daniel Durall and Suzanne Simard significantly contributed towards project development, interpretation of results and manuscript revisions. Valerie LeMay assisted with data analysis and manuscript revisions.  A version of chapter 3 has been published; Beiler KJ, Durall DM, Simard SW, Maxwell SA, Kretzer AM. 2010. Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts. New Phytol 185(2): 543-553.  I developed the project under the advisement of committee members, conducted the field and laboratory work with technical staff assistance, analyzed the data, and wrote the first draft of the manuscript. Daniel Durall and Suzanne Simard made a substantial contribution to project development, the interpretation of results and manuscript revisions. Sheri Maxwell assisted with the optimization of molecular methods and sample processing. Annette Kretzer provided critical advice during project development and reviewed the manuscript.  In chapter 4, I developed the project under the advisement of committee members, conducted field and laboratory work with technical staff assistance, analyzed the data, and prepared a manuscript for publication. Daniel Durall and Suzanne Simard made a substantial contribution towards project development and manuscript revisions. Valerie LeMay assisted with data analysis and manuscript revisions.     ivTable of contents  Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iii Table of contents .................................................................................................................... iv List of tables.......................................................................................................................... viii List of figures .......................................................................................................................... ix List of acronyms used .......................................................................................................... xiii Glossary ................................................................................................................................ xiv Acknowledgements .............................................................................................................. xvi Dedication ........................................................................................................................... xviii Chapter 1: Introduction ......................................................................................................... 1 1.1 Networks in ecology ............................................................................................................. 1 1.1.1 The problem of ecosystem complexity ............................................................................. 1 1.1.2 Ecological thresholds ........................................................................................................ 1 1.1.3 Network modeling and multi-scale analysis ..................................................................... 3 1.2 Interior Douglas-fir forests .................................................................................................... 4 1.2.1 Douglas-fir natural history ................................................................................................ 4 1.2.2 Interior Douglas-fir forests of south-central British Columbia, Canada ........................... 5 1.2.3 Interior Douglas-fir forest management ........................................................................... 7 1.3 Mycorrhizal symbiosis .......................................................................................................... 8 1.3.1 Plant-fungal mycorrhizal associations .............................................................................. 8 1.3.2 Mycorrhizal network ecology ........................................................................................... 9 1.3.3 Mycorrhizal network architecture ................................................................................... 11 1.3.4 Ectomyorhizal fungal population and community ecology ............................................ 12 1.3.5 Rhizopogon ..................................................................................................................... 13 1.4 Project objectives ................................................................................................................ 15 1.4.1 Overview ........................................................................................................................ 15 1.4.2 Chapter 2 objectives ....................................................................................................... 15 1.4.3 Chapter 3 objectives ....................................................................................................... 16  v1.4.4 Chapter 4 objectives ....................................................................................................... 17 Chapter 2: Vertical niche partitioning occurs between sister species of Rhizopogon fungi on mesic and xeric sites in an interior Douglas-fir forest .................................................. 18 2.1 Introduction ......................................................................................................................... 18 2.2 Materials and methods ........................................................................................................ 21 2.2.1 Plots and sampling .......................................................................................................... 21 2.2.2 Molecular analysis .......................................................................................................... 23 2.2.3 Data analysis and statistics ............................................................................................. 25 2.3 Results ................................................................................................................................. 26 2.3.1 Spatial patterns of Rhizopogon spp. genets .................................................................... 26 2.3.2 Rhizopogon spp. habitat associations ............................................................................. 27 2.3.3 Rhizopogon spp. mycelial growth form .......................................................................... 28 2.3.4 Mycelia continuity, root colonization patterns and networking ..................................... 28 2.4 Discussion ........................................................................................................................... 29 2.4.1 Spatial patterns of Rhizopogon spp. genets .................................................................... 29 2.4.2 Rhizopogon spp. habitat associations ............................................................................. 30 2.4.3 Rhizopogon spp. ?exploration type? ................................................................................ 31 2.4.4 Influence of soil moisture regime ................................................................................... 33 2.4.5 Conclusions .................................................................................................................... 34 Chapter 3: Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts................................................................................................................ 41 3.1 Introduction ......................................................................................................................... 41 3.2 Materials and methods ........................................................................................................ 44 3.2.1 Plots and sampling .......................................................................................................... 44 3.2.2 Molecular analysis .......................................................................................................... 45 3.2.3 Data analysis and statistics ............................................................................................. 46 3.3 Results ................................................................................................................................. 47 3.3.1 Characteristics of trees and fungi involved in the network ............................................. 47 3.3.2 Mycorrhizal network architecture ................................................................................... 48 3.4 Discussion ........................................................................................................................... 49 3.4.1 Mycorrhizal networks link multiple tree cohorts ............................................................ 49 3.4.2 Fungal genet parameters ................................................................................................. 49 3.4.3 Network parameters ........................................................................................................ 51  vi3.4.4 Implications for network functioning and forest management ....................................... 52 Chapter 4: The self-organization of ectomycorrhizal networks in xeric and mesic old-growth interior Douglas-fir forest ....................................................................................... 62 4.1 Introduction ......................................................................................................................... 62 4.2 Materials and methods ........................................................................................................ 66 4.2.1 Plots and sampling .......................................................................................................... 66 4.2.2 Molecular analysis .......................................................................................................... 68 4.2.3 Statistics and network modeling ..................................................................................... 69 4.3 Results ................................................................................................................................. 72 4.3.1 Sampling results ............................................................................................................. 72 4.3.2 Architecture of phytocentric MNs and sub-network components .................................. 73 4.3.3 Architecture of mycocentric MNs .................................................................................. 74 4.3.4 Factors contributing to mycorrhizal network architecture .............................................. 74 4.3.5 Network properties in xeric versus mesic soil moisture regimes ................................... 75 4.4 Discussion ........................................................................................................................... 76 4.4.1 Self-organized, complex architecture of Rhizopogon spp. MNs .................................... 76 4.4.2 Architecture of Rhizopogon spp. MNs and species-specific components ...................... 77 4.4.3 Contrasts between phytocentric and mycocentric models .............................................. 78 4.4.4 Determinants of mycorrhizal network architecture ........................................................ 79 4.4.5 Network architecture in xeric and mesic soil moisture regimes ..................................... 82 4.4.6 Summary ......................................................................................................................... 83 Chapter 5: Conclusion ........................................................................................................ 101 5.1 Contributions to current knowledge .................................................................................. 101 5.1.1 Summary ....................................................................................................................... 101 5.1.2 Chapter 2 synopsis: research conclusions and implications ......................................... 101 5.1.3 Chapter 3 synopsis: research conclusions and implications ......................................... 103 5.1.4 Chapter 4 synopsis: research conclusions and implications ......................................... 105 5.2 Project strengths and limitations ....................................................................................... 109 5.2.1 Primary strengths of project .......................................................................................... 109 5.2.2 Project scope and limitations ........................................................................................ 110 5.2.2.1 The general problem of complexity: mycorrhizal networks in curtus ................. 110 5.2.2.2 Sampling limitations: spatial extent and grain ..................................................... 111 5.2.2.3 Temporal limitations ............................................................................................ 112  vii5.3 Applications and inferences .............................................................................................. 113 5.4 Future research directions ................................................................................................. 117 5.4.1 Cryptic species coexistence .......................................................................................... 117 5.4.2 Mycorrhizal network structure and phenology ............................................................. 117 5.4.3 Mycorrhizal networking within individual tuberculate mycorrhizas ............................ 118 5.4.4 Kin selection via mycorrhizal networks ....................................................................... 119 Bibliography ........................................................................................................................ 120 Appendices ........................................................................................................................... 143 Appendix A Chapter 2 supplemental material ............................................................................... 143 A.1 Rhizopogon morphology ............................................................................................... 143 A.2 Plot characteristics and sampling .................................................................................. 144 A.3 Mycelia continuity: auxiliary results ............................................................................ 147 Appendix B Chapter 3 supplemental material ............................................................................... 148 B.1 Incidence matrix showing the number of times each Rhizopogon vesiculosus and/or R. vinicolor genet was encountered on the roots of each interior Douglas-fir tree. ....................... 148 B.2 Ancillary illustrations ................................................................................................... 149 Appendix C Chapter 4 supplemental material ............................................................................... 152 C.1 Incidence matrices showing associations between Rhizopogon spp. genets and interior Douglas-fir trees in 100 m2 plots with xeric or mesic soil moisture regimes. ........................... 152 C.2 Estimation of the plot edge effect ................................................................................. 154     viii List of tables  Table 2.1    Characteristics of study plots including the depth of soil layers, occurrence frequency of ground cover substrates and plants, and frequency of Rhizopogon spp. mycorrhizae, hyphae, and rhizomorph samples among 20 cm3 soil blocks ........................... 35 Table 2.2    Multi-dimensional contingency tests revealed the distribution of mycelia density classes (spatial pervasiveness) differed significantly between Rhizopogon spp. and between mesic versus xeric soil moisture regimes for each species.  R. vesiculosus genets formed dense ?mycelial mats? more frequently than R. vinicolor, and both species had more spatially pervasive mycelia in mesic plots versus xeric plots (? = 0.05). ............................................. 36 Table 3.1     Properties of microsatellite DNA loci used to discriminate among individual Pseudotsuga menziesii var. glauca trees and Rhizopogon spp. fungal genets.  Values corresponding to R. vinicolor are given in parentheses following those corresponding to R. vesiculosus; NA = not applicable. ........................................................................................... 54 Table 3.2    Size and networking characteristics of tree nodes by cohort class in a 30 x 30 m plot of interior Douglas-fir forest; mean values are reported ? one SD. ................................ 55 Table 3.3    Attributes of Pseudotsuga menziesii var. glauca trees linked through Rhizopogon spp. genets in a 30 x 30 m plot and characteristics of the resulting mycorrhizal network. .... 56 Table 4.1    Physical characteristics of the six 10 x 10 m study plots. ................................... 85 Table 4.2    The percent cover and frequency of forest floor substrates and understory vascular plant and bryophyte species among 1 m2 quadrat grid frames (n = 36 per plot) sampled in 5 m intervals within 30 x 30 m  plots of Pseudotsuga menziesii var. glauca forest with either mesic or xeric soil moisture regimes. ................................................................... 86 Table 4.3    Attributes of mycorrhizal network formed between mixed-aged interior Douglas-fir trees linked through Rhizopogon vesiculosus and R. vinicolor fungal genets. .................. 87 Table 4.4   The frequency distribution of interior Douglas-fir tree cohorts linked through either Rhizopogon vesiculosus or R. vinicolor genets in mycorrhizal networks, including all trees with roots found in Rhizopogon spp. mycorrhizas inside 10 x 10 m plots. .................... 89 Table 4.5    Attributes of mycorrhizal networks from the mycocentric perspective, with Rhizopogon vesiculosus and R. vinicolor genets as nodes linked through shared host trees in 10 x 10 m plots of xeric and mesic mixed-aged interior Douglas-fir forest. .......................... 90  ixList of figures  Figure 2.1    Map showing the locations of the six study plots in an interior Douglas-fir forest, with three plots (1-3) on upper slope positions with xeric soil moisture regimes and three plots (4-6) on lower slope positions with mesic soil moisture regimes (see main text).  Contour lines are labeled in US customary units (feet above sea level) and are based on 1:50,000 scale digital data (Garmin MapSource, Olathe, Kansas USA). ............................... 37 Figure 2.2    The median depths of occurrence (cm) of hyphae, rhizomorphs and tuberculate mycorrhizas averaged across Rhizopogon spp. genets from six independent plots differed significantly between species (i.) and between xeric and mesic soil moisture regimes for each species (ii.-iii.) (different letters indicate |P < t| < 0.05).  Boxplots show the median (middle line), mean (dashed line), and range (upper and lower bounding lines). ................................ 38 Figure 2.3    The spatial frequency, continuity, and pervasiveness of R. vesiculosus (in blue colour tones) and R. vinicolor mycelia (in red colour tones), excavated from six independent 2 x 0.2 x 0.2 m transect plots with either xeric or mesic soil moisture regimes in a Pseudotsuga menziesii var. glauca forest.  Blocks show the median density class of Rhizopogon spp. mycelia among the F and H forest floor layers (top layer of plots) and A and B mineral horizons (bottom layer of plots), with white indicating absence (no evidence of Rhizopogon spp.), gray indicating the presence of unidentified or desiccated tuberculate mycorrhizas, and the intensity of blue or red colour saturation increasing according to the density classes: scarce (? 1 tubercle and patchy, sparse hyphae/ rhizomorphs), diffuse (? 10 tubercles and uniformly sparse hyphae/ rhizomorphs), patchily dense (? 10 tubercles or patchily dense hyphal mats), or dense (? 10 tubercles or uniformly dense hyphal mats). Two genets each of R. vesiculosus and R. vinicolor were encountered in plot 6 (arrows point to sampling units where secondary genets occurred). ................................................................. 39 Figure 2.4    Frequency distributions of Rhizopogon vesiculosus and R. vinicolor mycelia density classes (pervasiveness) among organic and mineral horizons of the forest floor and in xeric and mesic soil moisture regimes.  Rhizopogon vesiculosus mycelia were more pervasive (P < 0.05) than R. vinicolor mycelia, and both species were more pervasive in mesic compared to xeric soil moisture regimes. ............................................................................... 40  xFigure 3.1    The top-down spatially implicit topology of Rhizopogon spp. genets and interior Douglas-fir trees in a 30 x 30 m plot. The plot (square outline) contains 67 trees of mixed age (green shapes, sized relative to each tree?s diameter). Black dots mark Rhizopogon ectomycorrhiza sample locations (n = 401), 338 of which were associated with a specific tree and fungal genet based on microsatellite DNA analysis. Samples representative of each fungal genet are outlined in differing colours. Rhizopogon vesiculosus genets (n = 14) are shaded with a blue background, and R. vinicolor genets (n = 13) with pink. Lines illustrate the linkages between tree roots encountered in Rhizopogon ectomycorrhizas and corresponding source trees aboveground (?root spans?) and are coloured according to tree genotype. An arrow points to the most highly connected tree, linked to 47 other trees through eight R. vesiculosus genets and three R. vinicolor genets inside the plot. Some trees, mycorrhiza samples, and/or genets may be obscured by overlapping features. ..................... 57 Figure 3.2    Spatially implicit network model showing interior Douglas-fir trees linked via shared colonization by Rhizopogon spp. genets. Circles represent tree nodes, sized according to tree diameter and coloured with four shades of yellow or green that increase in darkness with increasing age class. Lines represent the Euclidean distances between trees that are linked.  Line width increases with the number of links between tree pairs (i.e., repeated links through multiple fungal genets). An arrow points to the most highly connected tree, linked to 47 other trees through eight R. vesiculosus genets and three R. vinicolor genets inside the plot. Some tree nodes and their links may be obscured by overlapping features. .................. 58 Figure 3.3    The degree distribution of tree nodes linked through a mycorrhizal network, showing the number of trees linked to x number of other trees with roots colonized by Rhizopogon spp. in a 30 x 30 m plot of uneven-aged interior Douglas-fir forest.  Two or more trees were considered linked if they were colonized by the same fungal genet.  Node degrees were higher on average among older trees (cohorts 3-4) than younger trees (cohorts 1-2), leading to a skewed degree distribution characteristic of a scale-free network model. .......... 59 Figure 3.4    Mycorrhizal network topology showing the loss of connectivity between interior Douglas-fir trees linked through Rhizopogon spp. genets with the hypothetical removal of (a) the most highly linked hub tree (k = 47), (b) the most linking fungal genet (colonized 19 trees), (c) 20% of networking trees selected randomly, or (d) trees from the  xioldest cohort (also 20% of networking trees). Note the rotation by 90? from empirical model orientation. .............................................................................................................................. 60 Figure 3.5    Histograms showing the degree distributions of interior Douglas-fir tree nodes linked through Rhizopogon spp. genets following the hypothetical removal of (a) the most highly linked hub tree (k = 47), (b) the most linking fungal genet (linked 19 trees), (c) 20% of networking trees selected randomly, or (d) trees from the oldest cohort (also 20% of networking trees). ................................................................................................................... 61 Figure 4.1    Photographs showing mixed-aged stand structures of xeric (a-c) and mesic (d-f) study plots. .............................................................................................................................. 92 Figure 4.2    Socio-spatial network models depicting the architecture of mycorrhizal networks, with interior Douglas-fir tree nodes linked through Rhizopogon spp. genets in 100 m2 plots with xeric (a-c) or mesic (d-f) soil moisture regimes. Nodes (circles) are sized relative to tree diameter and darken from yellow to forest green with increasing age class. Models are scaled relative to plots where Rhizopogon spp. mycorrhizas were sampled (squares with dashed lines). .................................................................................................... 93 Figure 4.3    Scatterplot showing the monitonically negative relationship between the node degree of trees (number of neighboring trees each tree node was liked to) and the network clustering coefficients of those trees, in mycorrhizal networks formed between mixed-aged interior Douglas-fir trees and Rhizopogon spp. genets in six plots with xeric (filled shapes; Spearman?s rho = -0.469, P < 0.01) or mesic (open shapes; Spearman?s rho = -0.763, P < 0.01) soil moisture regimes. .................................................................................................... 94 Figure 4.4    Rhizopogon vesiculosus (i) and R. vinicolor (ii) sub-network components of mycorrhizal networks formed among Pseudotsuga menziesii var. glauca trees in six 10 x 10 m plots with either xeric (a-c) or mesic (d-f) soil moisture regimes. The degree of connectivity and density of links between trees differed significantly between R. vesiculosus and R. vinicolor linked network components (P < 0.05, Wilcoxon signed-rank tests, see main text) with R. vesiculosus forming more links between more trees than R. vinicolor. ............. 96 Figure 4.5    Mycorrhizal networks depicting Rhizopogon vesiculosus and R. vinicolor genets as nodes (circles filled with blue or pink, respectively) linked through shared interior Douglas-fir host trees in plots with xeric (a-c) or mesic (d-f) soil moisture regimes. Nodes are sized relative to the proportion of Rhizopogon spp. mycorrhiza samples in 10 x 10 m plots  xiirepresenting each genet; lines represent tree links between nodes, with line width corresponding to the number of host trees shared between genets. The lower right plot (f) is shown at 75% scale relative to others. All but one genet were linked through shared hosts, and this connectivity was not influenced by fungal species or soil moisture regime. ............ 97 Figure 4.6    In mycorrhizal networks modeled with Rhizopogon spp. genets as nodes linked through shared interior Douglas-fir trees, the frequency distribution of genet node degrees (the number connections each node has to other nodes) was positively skewed, with a few genets of each species having higher than average node degrees. .......................................... 98 Figure 4.7    The number of trees colonized by R. vesiculosus genets (filled circles) and R. vinicolor genets (open circles) in mycorrhizal networks increased in relation to the number of times genets were sampled within 10 x 10 m plots, which provides a proxy for the link strength of fungal genets sampled within similar spatial extents (see main text). .................. 99 Figure 4.8    Generalized linear mixed model results showing the predicted values for response variables (a) tree node degree, (b) normalized node degree, and (c) node clustering coefficient in relation to increasing tree DBH in cm, for trees linked in mycorrhizal networks through Rhizopogon vesiculosus (?VES?) or R. vinicolor (?VIN?) in plots with xeric (?DRY?) or mesic (?MOIS?) soil moisture regimes (see main text). ................................................... 100   xiii List of acronyms used  ANOVA Analysis of Variance BC British Columbia (Canada) CWD Coarse Woody Debris DF  Degrees of Freedom DNA  Deoxyribonucleic Acid EMF  Ectomycorrhizal Fungi IDF Interior Douglas-fir (Pseudotsuga menziesii var. glauca) biogeoclimatic ecological zone MANOVA Multivariate Analysis of Variance MN  Mycorrhizal Network, a.k.a. common mycorrhizal network PCR  Polymerase Chain Reaction SD Standard Deviation SMR  Soil Moisture Regime      xivGlossary  Glossary of terms as related to mycorrhizal networks  Terminology  Description Anastamosis the fusion of somatic hyphae in fungi, leading to reticulate mycelial networks Centrality  (re. betweenness centrality) the degree of a node, relative to the distribution of degrees in the network; describes the contribution of nodes to network connectivity with respect to their topological position in the network Clique a subset of three or more nodes that are maximally connected (each node is connected to the others) Cliquishness the likelihood that two associates of a node are themselves connected; a high clustering coefficient indicates greater cliquishness Cohort a group of individuals that are approximately the same age  Component- biological a distinct part or subgroup integrated within a biological system, such as an individual, species, or population within an ecosystem Component- network a maximal connected sub-network Cryptic species species that differ genetically but are hard to distinguish based on physical characteristics; often considered the same species without genetic analysis Degree (i.e., connectivity) the number of links a node has to other nodes    Degree distribution  the probability distribution of degrees among nodes in a complex network Genet  an individual fungal genotype or clone Hub  a node with a high degree of connectivity relative to the average degree distribution Link a path (connection) between two vertices (nodes); from the phytological network perspective an undirected path (edge) representing a fungal genet eij that colonizes the roots of two trees i and j or more (?hyperlink?) Mean path length the minimum number of link steps separating two nodes, averaged across all node pairs in the network Meta-network a complex network that can be partitioned into sub-network components Mycelium (pl., mycelia) the collective term for hyphae forming the vegetative tissues of a fungal genet    xvTerminology  Description Mycorrhizal network- mycocentric  an underground network where fungal genets are linked through a shared association with one or more host tree Mycorrhizal network- phytocentric  an underground network where the mycelia of mycorrhizal fungi link the roots of plants of the same or different species Node  points or vertices in a graph that are joined by a physical, functional, or physiological pathway Node clustering coefficient the density of links a node has relative to that of its neighbors; indicates the cliquishness among nodes in the network Network centralization  the variation among node degrees divided by the maximum variation possible in the network Network clustering coefficient the mean clustering coefficient among all nodes in the network; indicates the accessibility of links to nodes Network density  the number of linkages in a network, relative to the maximum possible number of linkages Network diameter  the longest of the shortest paths between any two nodes in the network Network motif a geodesic, pattern of connections, or sub-network that occurs significantly more than would be expected in a randomized network Normalized node degree the number of links a node has to other nodes relative to the total number of nodes in the network Path length (i.e., geodesic) the number of link steps separating two nodes Random network  a network model where nodes link randomly to other nodes, resulting in a probability degree distribution following a Gaussian curve; random networks are generally less cliquish and more easily traversed than regular networks Regular network  a network characterized by nodes with localized connectivity, high clustering coefficients and a probability degree distribution following a Gaussian curve; regular networks are generally cliquish and not easily traversed Scale-free network  a network in which some nodes (i.e., hubs) have a high centrality and degree of connectivity relative to the average among all nodes in the network, resulting in a skewed node degree distribution; they are both cliquish and easily traversed Small-world property  a network with a small topological diameter (eccentricity) relative to the total number of nodes in the network where non-neighboring nodes can still reach each other through a small number of link steps     xviAcknowledgements  This study would not have been possible without help from an extensive social network and a legacy of past researchers that provided the residual foundation of information to build on.  Hubs in this network include my supervisory committee and funding agencies, but there were many important nodes including technicians, forest managers, colleagues, friends and family. Regrettably, these include more persons than possible to credit; but the diversity of roles and relations linking them to the project helped to shape and strengthen the final product.    I especially thank my supervisors Daniel Durall and Suzanne Simard, whose earlier work on fungal-mediated carbon flow between plants captured my interest and imagination.  Their patience and support in and outside the workplace was unwavering, and the project would not have been possible without their advice and financial support.  Advisement provided by committee members Sally Aitken and Valerie LeMay also played a critical role in shaping the project and bringing it to completion.    I thank Thomas Bruns, Richard Hamelin, and William Mohn for serving on the dissertation examining committee and providing constructive feedback on an earlier version of this manuscript.  Karen Hodges, Melanie Jones, John Klironomos, Annette Kretzer, Michael Pidwirny, Jason Pither, and Michael Russello also provided important analytical or literary feedback during the project.  In particular, Melanie Jones was a repository of ectomycorrhizal knowledge and contributed largely to manuscript revisions; and Jason Pither helped guide me through the analysis and interpretation of challenging data sets.  Nusha Keyghobadi and Francis Martin provided quality feedback on the published version of chapter 3.  I am indebted to Mahsa Amirabbasi, Michaela Byrne, Anders Gon?alves da Silva and Philippe Henry for their advice regarding molecular methodologies.  I also thank Barb Lucente (Unit 2 Assistant, UBCO), Cindy Prescott (Professor & Associate Dean of Graduate Studies, UBC Faculty of Forestry), and the Faculty of Forestry Graduate Program staff for their assistance with program technicalities and student advocacy.       xviiThis research could not have been completed without the hard work of many field and lab assistants.  I especially thank Jenna Benson, Tanja Bergen, Jessie Brown and Fraser McIntosh for their help in the field.  These persons spent weeks away from home, braving inclement weather and mosquito swarms of epic proportions, to complete the unglamorous and tedious work of digging through soil in search of Rhizopogon spp. tubercles; and did so with enduring devotion.  Bill Clark, Sheri Anne Maxwell, and Mary Ann Olson helped with the optimization of lab protocols and molecular processing; and Jessica Baker, Jenna Benson, and Daniel Salloum each made substantial contributions towards sample processing.   I thank the BC Ministry of Forests for providing field sites, and especially thank Brent Olsen at the Kamloops Forest District and Andre Arsenault at the Southern Interior Forest Region for their assistance with site selection.  This project was funded by the Forest Investment Account- Forest Science Program from the BC Ministry of Forests, a University Graduate Fellowship from the University of British Columbia, and the National Science and Engineering Research Council of Canada.  Special thanks are owed to my parents Omar and Dolores who have supported me in numerous ways through my years of education, and my wife Maja who is my nearest and dearest network neighbor.              xviii Dedication                                                      To my family    1Chapter 1:   Introduction  1.1 Networks in ecology 1.1.1 The problem of ecosystem complexity In nature, all things are directly or indirectly connected.  The American naturalist John Muir is quoted as marveling ?when we try to pick anything out by itself, we find it hitched to everything else in the universe?.  Being no exception, forest ecosystems are complex, dynamic systems comprised of many interrelated biotic and abiotic parts and processes.  Their borders can appear well-defined, as seen in the forest-grassland interface and high elevation tree lines, but generally represent gradual shifts in species proportions over space or time.  Because this complexity defies easy description, forest ecosystems are best described by their emergent patterns and processes (e.g., spatial patterns, species diversity, productivity, nutrient cycling, etc.) or by the attributes of key biotic or abiotic components that are the principle determinants of the emergent properties (e.g., foundational, dominant, or keystone species, climate, edaphic factors, mutualistic networks, etc.).  Describing the emergent properties of a system, versus attributes of key system components, is analogous to holistic versus reductionist approaches to ecological studies.  Although often considered disparate methodological paradigms, these approaches can be used complimentarily to obtain a more complete understanding of an ecosystem?s universal patterns and processes in addition to underlying deterministic mechanisms (May, 1977; Levin, 1998; Bascompte, 2009b).  Knowledge of how these properties change across scales of space, time and complexity is necessary to preserve forest ecosystems and the services they provide.  1.1.2 Ecological thresholds  Peters et al. (2009) define ecological thresholds as ?the level or magnitude of an ecosystem process that results in a sudden or rapid change in ecosystem state?.  Thresholds may refer to an exogenous variable (e.g., geography, climate, edaphic factors, disturbances, etc.), endogenous processes (e.g., autogenic change, nutrient cycling, patch connectivity, interactions between organisms, etc.), or the emergent properties of the system itself (Scheffer et al., 2001; Peters et al., 2009).   2Ecosystem state changes, or regime shifts (see Sutherland, 1974), can result from the loss of one or a few principal biological components (dominant, foundational or keystone species, trophic guild, etc.) that play a central role in determining the structure and functioning of the system (Paine, 1966; Dayton, 1972; Jones et al., 1994; Ellison et al., 2005; Peters et al., 2009).  The mass-ratio hypothesis predicts that dominant species have a greater effect on ecological patterns and processes than subordinate or transient species (Grime, 1998).  There have been many examples throughout history where the loss of a single tree species disrupts ecosystem processes (species interactions, nutrient cycling, trophic dynamics, etc.) to the extent that the state of the ecosystem changes (Ellison et al., 2005).  For example, the American chestnut (Castanea dentata (Marsh.) Borkh.) was a major component of forests across eastern North America before it was functionally decimated by the introduction of the bark fungus Chryphonectria parasitica.  American chestnuts were large, fast-growing hardwood trees and prolific producers of nuts that were a major autumn food source for bear, deer, wild turkey, and other wildlife.  While their tannin-rich wood was recalcitrant to decay and provided important habitat for terrestrial and aquatic biota, their nutrient-rich leaves provided an important energy base for decomposer communities.  The functional decline of American chestnuts and subsequent replacement by other tree species profoundly altered these ecosystem patterns and processes (Smock & MacGregor, 1988; Ellison et al., 2005; Freinkel, 2007).  Thus, a better understanding of the autecological traits of dominant, foundational or keystone species and their interactions with the biotic and abiotic components of the system are needed to detect or predict potential ecosystem regime shifts (Suding et al., 2004; Miao et al., 2009; Peters et al., 2009; Lau et al., 2010).  In addition to key biotic or abiotic components, the stability of ecosystems may hinge on the diversity of components and the multiplicity of their interactions (MacArthur, 1955; Pimm, 1984; Levin, 1998; Peterson et al., 1998; McCann, 2000).  Attempts to link community diversity with ecosystem function and stability have focused on the role of transient species in providing increased efficiency of energy flow through path multiplicities (MacArthur, 1955; Ehrlich & Ehrlich, 1981; Naeem, 1998), functional replacements when dominant species are lost or environmental conditions change (Yachi & Loreau, 1999), and for providing communities with a diverse and complex adaptive toolkit in general (Tilman &  3Downing, 1994; Loreau, 2000).  Hypotheses proposed to account for these relationships include the ?diversity-stability hypothesis?, ?rivet hypothesis?, ?redundant species hypothesis?, ?idiosyncratic hypothesis?, ?insurance hypothesis?, and others (Goodman, 1975; Lawton & Brown, 1993; Lawton, 1994; Mikola & Setala, 1998; Yachi & Loreau, 1999).  Seemingly redundant species may interact at different scales of space or time, which creates cohesion in ecosystems through the nestedness and cross-scale connectedness of positive and negative feedbacks (MacArthur, 1955; Paine, 1966; Peterson et al., 1998; Okuyama & Holland, 2008).  The stress-gradient hypothesis infers that net effects of community interactions shift in importance from competition to facilitation with increasing environmental stress (Bertness & Callaway, 1994; Callaway, 1995; Brooker et al., 2008).  1.1.3 Network modeling and multi-scale analysis Modeling the properties of complex ecosystems in order to detect and predict system dynamics is one of the central problems in ecological studies (Holling, 1973; Levin, 1992; Ings et al., 2009).  Network analysis, based on the principles of graph theory, provides a tool for exploring the topology and dynamics of complex biological systems; where individuals, species, or species guilds are viewed as nodes linked through their ecological associations (Bray, 2003; Bascompte, 2009b; Lau et al., 2010; Parrott, 2010).  Moreover, it provides a framework to study systems from the perspective of emergent properties (top down) or component attributes (bottom up) concurrently and at multiple genetic, spatial and temporal scales (Proulx et al., 2005; Urban, 2005; Barab?si, 2009).    One of the most basic features of networks is their architecture, which refers to the orientation of nodes and the topology of associations between them.  The architecture of networks often follows one of three models: regular, random or scale-free; which depend on the distribution of links among nodes (degree distribution), the accessibility of links to nodes (clustering coefficients), the link steps separating nodes (path lengths), and the contribution of nodes to network connectivity with respect to their topological position or centrality (Albert et al., 2000; Bray, 2003; Newman, 2003; Parrott, 2010).  Each of these models may be spatially explicit, where the spatial orientation of nodes is specified; spatially implicit, where nodes are spatially referenced but links refer to aspatial relationships or processes; or  4purely aspatial, where no reference to space is made in the model (Dale & Fortin, 2010).  In ecological networks, architecture may comprise biotic or abiotic components, the relationships between components, and the spatiotemporal extent and topology of components and their relationships (Bascompte, 2010; Dale & Fortin, 2010).  Network architectures and their ecological implications are introduced in more detail in Chapters 3 and 4.        It should be emphasized that network analysis and biology share many terms in common but with different meanings, and it is important to distinguish between these terms according to the context.  For example, the previous paragraph refers to biotic and abiotic networking components (i.e., entities or elements), whereas a network component refers to a subset of interconnected nodes within a network (e.g., sub-networks of a meta-network).  Likewise, diameter could refer to the extent of spatial networks, or in graph theoretical terms the longest of the shortest path lengths (geodesics) required to traverse the network.  Terms related to holistic network properties may not directly relate to attributes of individual nodes (e.g., network centralization and clustering coefficient versus the centrality and clustering coefficients of nodes); although a superficial distinction, I will refer to system properties and element attributes to distinguish between top-down and bottom-up network perspectives.  Throughout this manuscript I use network terminology as commonly applied in computer sciences (Newman, 2003), rather than graph theoretical lexis (e.g., nodes and hubs, rather than vertices; links rather than arcs or edges) and define these with respect to mycorrhizal networks in the Glossary (xi).  1.2 Interior Douglas-fir forests 1.2.1 Douglas-fir natural history Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) is native to western North America, where it occurs in many mountainous regions in pure stands or as a principal component in mixed species stands (Hermann & Lavender, 1990; Farrar, 1995; Griesbauer & Green, 2010).  The proportion of Douglas-fir relative to other tree species in mixed stands varies considerably based on interactions between climatic, geographic, and edaphic factors; the frequency, extent, and severity of disturbances; and the genetic composition, abundance, and  5spatial patterns of species present (Ryker, 1975; Ferguson & Carlson, 1991; Oliver & Larson, 1996; Shatford et al., 2007).  There are currently two Douglas-fir varieties recognized, including coastal Douglas-fir (P. menziesii (Mirb.) Franco var. menziesii) and interior Douglas-fir (P. menziesii var. glauca (Beissn.) Franco), which are strongly differentiated based on morphological, chemical, and genetic traits (Hermann, 1982; Gugger et al., 2010).  Chemical and cytological evidence suggests P. menziesii var. glauca may comprise distinct northern and southern subgroups partitioned by a gradual transition zone at approximately 44? latitude (Zavarin & Snajberk, 1973; von Rudloff, 1975; Li & Adams, 1989).  1.2.2 Interior Douglas-fir forests of south-central British Columbia, Canada Interior Douglas-fir trees are a foundational species, owing to their extensive distribution and their prominent role in shaping the structure, patterns and processes of the ecosystems they inhabit (Ellison et al., 2005; Simard, 2009).  Trees are thought to be the largest, longest-lived members of forest communities and their spatial patterns a defining feature of forest ecosystems.  Fogel and Hunt (1979) estimated these trees, together with their ectomycorrhizal fungal symbionts, to be responsible for up to 70 % of net primary productivity in interior Douglas-fir forests.  In life and after death, they fundamentally alter their surrounding environments (i.e., available light, nutrient regimes and above/belowground microclimates) and provide food and habitat for a multitude of macro- and microorganisms (Ellison et al., 2005).  In British Columbia (BC), a diverse assemblage of plant and wildlife species inhabit these forests (Hope et al., 1991).      6In the dry, cool climates of south-central BC, interior Douglas-fir forests occur in extensive pure stands that undergo gap-phase regeneration, resulting in self-perpetuating, multi-cohort stands (Hermann & Lavender, 1990; Vyse et al., 2006).  Pure interior Douglas-fir forests can occur in several alternate structural states (Simard, 2009); the composition and structure of these forests ultimately depend on the aforementioned interplay of allogenic and autogenic agents (e.g., biogeoclimatic conditions, disturbance regimes, biological legacies following disturbances, ecological interactions, etc.) (Franklin et al., 2002).  Nonetheless, some predictions can be made regarding potential climax forest structures based on site biogeoclimatic conditions and the autecological characteristics of tree species present (Minore, 1979).         Soil moisture regimes refer to the average amount of soil water annually available for evapotranspiration by vascular plants over multiple years (Klinka & Meidinger 1987).  Among climatic factors, precipitation and its indirect effects on soil water availability is the principal limiting factor that governs the species composition and structural characteristics of interior Douglas-fir forests (Lopushinsky, 1990; Case & Peterson, 2005; Littell et al., 2008; Griesbauer & Green, 2010).  For example, pure stands of interior Douglas-fir transition to bunchgrass and ponderosa pine (Pinus ponderosa Laws) ecosystems in arid climates.  Where precipitation is higher, interior Douglas-fir occurs as an early to mid-seral species along with numerous broadleaf and conifer species (Vyse et al., 1990; Meidinger & Pojar, 1991; Huggard et al., 2005).  It eventually becomes a subordinate species component or is replaced altogether by western redcedar (Thuja plicata Don), western hemlock (Tsuga heterophylla (Raf.) Sarg., spruce (Picea spp.), or true firs (Abies spp.) in cooler and wetter climates (Alexander, 1988).  Thus, pure stands of interior Douglas-fir tend to develop in semi-arid climates, where water availability is limiting.    Over the next century, the climate conditions of south-central BC are projected towards more extreme temperature fluctuations and an increased frequency and severity of droughts, wildfires, and outbreaks of insects and disease (Nitschke & Innes, 2008a; Nitschke & Innes, 2008b; Walker & Sydneysmith, 2008; Rodenhuis et al., 2009; Griesbauer et al., 2011).  While the distribution of interior Douglas-fir forests have undergone numerous cycles of  7expansion and contraction throughout their history, the rates of climatic changes expected over the next century would far exceed the rates at which these forest ecosystems are capable of migrating (Shafer et al., 2001; Hamann & Wang, 2006; Loarie et al., 2009; Gugger et al., 2010).  Moreover, increasing temperatures combined with past management practices could leave interior Douglas-fir trees more susceptible to water deficits, abiotic or biotic disturbances, or diebacks (Lopushinsky, 1990; Overpeck et al., 1990; Case & Peterson, 2005; Klenner et al., 2008; Littell et al., 2008), and further reduce the capacity of these forests to regenerate or disperse to form new stands (Perry et al., 1990; Shafer et al., 2001).  1.2.3 Interior Douglas-fir forest management  Following the turn of the 20th century, the predominant management paradigm for interior Douglas-fir forests had been to harvest the largest, straightest trees for their economic value (e.g., high-grading) and leave the remaining understory trees to grow into future stands (Isaac, 1956; Vyse et al., 1990; Curtis, 1998; O'Hara, 2002).  Although various forms of partial cutting were used for over a century in interior Douglas-fir forests, there remains little consensus as to the long-term efficacy of these treatments (Worthington & Staebler, 1961; Zeide, 2001; Maguire et al., 2006 and references therein).  The majority of studies report the initial release of younger cohorts from suppression (Krauch, 1956; Bailey & Tappeiner, 1998; Miller & Emmingham, 2001; Vyse et al., 2006; Renninger et al., 2007), though prolonged growth shock has also been reported (Staebler, 1956; Harrington & Reukema, 1983).  It is assumed that repeated high-grading could reduce the vigour and resilience of stands through dysgenic selection if not carefully implemented (Munger, 1950; Howe, 1995; Curtis, 1998).  Harvest methods currently used in the interior Douglas-fir forests of BC include variable retention or selection methods (Vyse et al., 2006), with the goal of maintaining the genetic integrity and structural complexity of stands to facilitate their return to conditions resembling the diversity of species, structures and ecological processes of naturally occurring late-successional forest ecosystems (Vyse et al., 1990; Franklin et al., 1997; Aubry et al., 1999; Halpern et al., 2005).     81.3 Mycorrhizal symbiosis 1.3.1 Plant-fungal mycorrhizal associations The mycorrhizal symbiosis is an ancient, predominantly mutualistic association that forms between fungi and the roots of approximately 80% of vascular plant species (Frank, 1885, as per Trappe 2005; Simon et al., 1993; Remy et al., 1994; Blackwell, 2000; Wang & Qiu, 2006; Smith & Read, 2008).  Mycorrhizal associations have independently evolved many times within plant and fungal lineages.  As such, broad arrays of morphological, physiological and ecological traits are represented, and relationships between the symbionts can range from mutualistic to parasitic depending on numerous endogenous and exogenous factors (Smith & Read, 2008).  In general, mycorrhizal fungi provide plants with increased access to water and nutrients in exchange for photosynthate.  Mycorrhizal fungi can form extensive hyphal networks in the soil capable of absorbing inorganic and/or organic nutrients otherwise inaccessible to their host plants.  Access to these hyphal networks has been shown to improve plant water use efficiency and resilience to water stress (Parke et al., 1983; Allen, 2007; Egerton-Warburton et al., 2007), which could be of paramount importance if climate change leads to an increased frequency and severity of droughts.  Plants also receive other benefits from mycorrhizal associations, such as protection from root pathogens and a buffer against extreme temperature, moisture, or pH levels and heavy metal concentrations (Trappe, 1977; Smith & Read, 2008).  In exchange, plants provide their fungal symbionts with excess photosynthates, which comprise the sole energy source for most mycorrhizal fungi.    Mycorrhizal types can be categorized based on the anatomical and morphological features of the plant-fugal interface (see Peterson & Massicotte, 2004; Peterson et al., 2004).  Many different types are found in association with the vascular and nonvascular plants within interior Douglas-fir forests, including ectomycorrhizas and arbuscular, ericoid, monotropoid, and orchid mycorrhizas.  Ectomycorrhizas are the predominant type associated with Douglas-fir tree roots (Trappe, 1977; C?zares & Smith, 1996; but see Smith et al., 1998) and these represent a taxonomically, morphologically and functionally diverse group of fungi that includes the Ascomycota, Basidiomycota, and Zygomycota phyla within the Kingdom Fungi (Allen et al., 2003).  The host specificity among species of plants and mycorrhizal fungi ranges from narrow to broad and can be context dependent.  For example, over 2,000 fungal  9species are estimated to associate with Douglas-fir trees, some of which colonize only Douglas-fir but many of which are less discriminating and associate with numerous plant species (Trappe, 1977), which can influence above and belowground community dynamics (Jones et al., 1997; Horton & Bruns, 1998; Bruns et al., 2002).  Moreover, a single fungal genet may associate with multiple plant species and form different types of mycorrhizae (different nutrient exchange interfaces) depending on the host, such that the physiological plasticity of the fungus provides a functional link between plants with highly divergent evolutionary histories (Read et al., 2000; Kottke et al., 2003; Peterson et al., 2004).     Ectomycorrhizal fungi (EMF) play a key role in forest ecosystems through their influences on plant growth, nutrient cycles, and above and belowground community dynamics (Trappe & Luoma, 1992; Francis & Read, 1994; Horton & van der Heijden, 2008).  The morphological and functional diversity of EMF species provides an adaptive toolkit for trees that serves to increase their niche breadth and resilience to stress in variable environments (Amaranthus, 1992; Read, 2002; Allen et al., 2003; Jones et al., 2010).  Fogel and Hunt (1979) estimated the combined standing crop of EMF mycelia and reproductive structures to comprise over 50 % of total soil C and N pools in a coastal Douglas-fir forest, and fungal tissues turned over 3-5 times faster than tree tissues and other forest floor substrates (see also Hunt & Fogel, 1983; H?gberg & H?gberg, 2002; Nilsson et al., 2005).  The sporocarps (e.g., mushrooms, truffles) produced by many EMF are important components of forest food webs, linking trophic guilds from microorganisms to megafauna (Maser & Maser, 1988; Claridge et al., 1999; Ashkannejhad & Horton, 2006).  1.3.2 Mycorrhizal network ecology Mycorrhizal networks (MNs), or common mycorrhizal networks, are formed when two or more plants of the same or different species are colonized by a continuous fungal mycelium.  Although their existence had been proposed since the late 1800?s (Trappe, 2005), a century passed before researchers began testing the potential ecological role of MNs through robust experimentation (Brownlee et al., 1983; Finlay & Read, 1986b; Finlay & Read, 1986a; Haystead et al., 1988).  By 1985, numerous studies had reported the strict associations of achlorophyllous plants with the mycelia of specific EMF species, and recognized that both  10partners depended upon photosynthate from neighboring conifer species (reviewed by Castellano, MA & Trappe, JM, 1985).  The recurrent parallel evolution of mixo- and mycoheterotrophy in kingdom Plantae (Leake et al., 2004), wherein understory plants obtain carbon from overstory plants via MN pathways, has now been well substantiated with isotope labeling studies and provides the most unequivocal evidence of the existence and functional relevance of MNs (Simard & Durall, 2004; Selosse et al., 2006).   Through their influences on the establishment and competitive dynamics of trees, MNs play a foundational role in shaping the spatial and genetic structure of forest communities (Allen et al., 2003; Simard, 2009; van der Heijden & Horton, 2009).  Radioactive and stable isotope analysis, along with the use of mesh barriers to isolate MN effects from other soil pathways, have repeatedly demonstrated the translocation of water, C, N, P, and other nutrients between the roots of the same or differing plant species via MNs (Simard et al., 2002; Leake et al., 2004; Smith & Read, 2008).  There is also potential for MNs to provide a route for plant hormones and other signalling molecules as recently demonstrated by Song et al. (2010), who found that MNs facilitated defense signalling between tomato plants (Lycopersicon esculentum Mill.) linked through the arbuscular mycorrhizal fungus Glumus mosseae.  Inter-plant nutrient transfers through fungal pathways have been demonstrated under natural conditions (Taber & Taber, 1984), including net C transfer following source-sink gradients from ectomycorrhizal Betula papyrifera to P. menziesii var. glauca trees (Simard et al., 1997a), between P. menziesii var. glauca seedlings (Teste et al., 2010), between arctic Betula nana shrubs (Deslippe & Simard, 2011), and between arbuscular mycorrhizal Acer saccharum trees and Erythronium americanum plants (Lerat et al., 2002).     Studies have demonstrated that linkage into a MN with residual trees or shrubs can improve the survival and productivity of newly establishing seedlings (Nara, 2006; McGuire, 2007; Bai et al., 2009) (but see Dehlin et al., 2008).  In interior Douglas-fir forests, MNs associated with residual trees in cutover areas appears to facilitate new regeneration by providing mycorrhizal inoculum and carbon, water or nutrient subsidies from the mature trees (Teste et al., 2009; Teste et al., 2010).  MNs may also intensify competition between trees (Kytoviita et al., 2003).  Like the mycorrhizal symbiosis in general, effects of MN formation on plant  11communities likely depend on numerous endogenous and exogenous factors (species combinations, resource availability) and vary over space and time (Allen et al., 2003; van der Heijden & Horton, 2009).  Although the ecological significance of inter-plant transfers remains a topic of debate (see e.g., Whitfield, 2007; van der Heijden & Horton, 2009), the existence of MNs in natural systems appears ubiquitous and incontrovertible.  1.3.3 Mycorrhizal network architecture Information regarding the architecture, organizational mechanisms, and spatiotemporal patterns of MNs is needed to evaluate the ecological significance of their functional traits (e.g., effects of nutrient transfer between plants on tree regeneration, competition and mortality) (Amaranthus & Perry, 1994; Selosse et al., 2006).  The architecture of MNs includes the participating plants and fungi (biological components represented as links or nodes), the qualitative and quantitative traits of participants (genetics, morphology, physiology, and spatiotemporal patterns and plasticity), the relationships between participants (e.g., direct or indirect competition, facilitation, or trophic dynamics), and the emergent properties of their patterns and processes (e.g., extent and topology of components and their relationships over space and time).  Despite being a basic feature of networks, MN topology remains virtually unknown due to the cryptic, subterranean existence of associated fungi (Simard & Durall, 2004).  Such studies are greatly facilitated by contemporary molecular methods, because tree and fungal DNA can be extracted concurrently from ectomycorrhizal root tips collected in the field and PCR amplified using genotype-specific primers to distinguish between individual trees and fungal genets (Saari et al., 2005; Lian et al., 2006).    MNs can be modeled from the ?phytocentric? or ?mycocentric? perspective (with plants or fungi as nodes, respectively), in addition to more complex associative network models (Southworth et al., 2005; Selosse et al., 2006).  The inclusivity of MN models ranges from individual plant-fungal species pairs (Lian et al., 2006) to community-level scales of complexity (Southworth et al., 2005).  Illustrating plants as nodes linked through mycorrhizal fungi is unambiguous, and is relevant to forest managers interested in how MNs may influence forest dynamics such as regeneration, competition for resources, and resilience to  12stress (Smith & Read, 2008; Simard, 2009).  The complimentary perspective, with fungal genets as nodes linked through shared host trees, could provide insights into the carbon economics of EMF (e.g., sharing of and competition for plant derived carbon between fungal genets), tree host ?island effects? on fungal biodiversity (e.g., relationships between the size of a tree or patches of trees and EMF diversity), or the genetic specificity of tree and fungal symbionts (e.g., which tree genotypes or tree species sustain which EMF genets, species, or species guilds).  1.3.4 Ectomyorhizal fungal population and community ecology An adequate review of EMF population and community dynamics is beyond the scope of this dissertation, owing to the prodigious genetic, morphological, and functional diversity represented within the EMF guild (Cairney & Burke, 1996; Martin et al., 2001; Allen et al., 2003; Agerer, 2006).  Below, I highlight several fundamental EMF community traits relevant to the potential ecological significance of MNs as discussed in this dissertation.  A more detailed albeit incomplete review of factors contributing towards EMF community structure and dynamics is provided in chapter 2.  Allen et al. (2003) provide a comprehensive summary of EMF traits relevant to the interplay between EMF and host plant community dynamics; this subject has been the focus of numerous reviews (e.g., Francis & Read, 1994; Bever et al., 1997; Dahlberg, 2001; Johnson et al., 2005; Johnson et al., 2006; Simard, 2009; Klironomos et al., 2011).   EMF communities occur as a spatially complex, temporally dynamic mosaic of species assemblages (Horton & Bruns, 2001; Lilleskov et al., 2004) whose organization is influenced both by stochasticity (disturbance severity, dispersal limitation, etc.) and deterministic mechanisms (edaphic adaptation, host specificity, mycelia growth rate and morphology, competitive interactions, priority effects, etc.).  Species abundances are typically distributed as an inverse J-curve (Horton & Bruns, 2001; Erland & Taylor, 2002; Tedersoo et al., 2003); a common feature in ecosystems comprised of a few dominant and many subordinate or transient species (Grime, 1998).  The host-specificity of EMF species ranges from narrow to broad (Molina et al., 1992; Horton & Bruns, 1998; Hoeksema & Thompson, 2007; Ishida et al., 2007; Tedersoo et al., 2008a; Morris et al., 2009), and single-species, even-aged forests  13or even individual trees generally harbour diverse EMF assemblages (Allen & Allen, 1992; Bruns, 1995; Korkama et al., 2006; Peay et al., 2008).  Among site factors, soil moisture availability can play a major role in shaping the species composition of EMF communities (Erland & Taylor, 2002; Izzo et al., 2005; Querejeta et al., 2009).        While the number of studies investigating the coarse-scale spatial patterns of EMF communities is growing, there is less information regarding more precise species autecology and fine-scaled community structuring (Dickie et al., 2002b; Zhou & Hogetsu, 2002; Lilleskov et al., 2004; Genney et al., 2006; Koide et al., 2007; Kennedy, 2010; Pickles et al., 2010).  The lack of species-level autecological information among EMF creates an obstacle for understanding the architecture and plasticity of MNs, or their coarse-scale ecological roles (e.g., effects on forest stand structural complexity, productivity and resilience, and successional dynamics) (Simard & Durall, 2004; Selosse et al., 2006).  The extent and continuity of individual fungal genets over space and time is a particularly important life history trait for addressing the assumption that plant roots may be ?linked? through hyphal bridges in a MN (Amaranthus & Perry, 1994; Simard & Durall, 2004).  For this reason, I begin my investigation of MN architecture at fine spatiotemporal scales (chapter 2) to address the issue of whether roots of two or more trees colonized by the same fungal genet constitutes a continuous ?link? versus fragmented mycelia clones, and to better understand the autecology of the EMF species involved.  1.3.5 Rhizopogon Rhizopogon (Fries) vesiculosus and R. vinicolor (Boletales, Basidiomycota) were chosen as the focal ectomycorrhizal fungal species in this study for numerous reasons.  They are among the most frequent and abundant EMF species associated with coastal and interior Douglas-fir, and show strong specificity for these trees (Chu-Chou & Grace, 1981; Molina et al., 1999).  They commonly occur on establishing seedlings and mature trees alike (Chu-Chou & Grace, 1981; Jones et al., 1997; Cline et al., 2005; Twieg et al., 2007; Teste et al., 2010) and therefore have high potential for linking multiple tree cohorts in a forest (Roth & Berch, 1992; Molina et al., 1999).  Both species form large perennial genets spanning meters (i.e., R. vinicolor) to decameters (i.e., R. vesiculosus) in length (Kretzer et al., 2004; Kretzer et al.,  142005; see also Beiler et al., 2010).  They are capable of mobilizing recalcitrant nutrients from forest soils that would otherwise be inaccessible to plants (Li et al., 1992; Durall et al., 1994b; Smith & Read, 2008) and have been shown to increase seedling establishment, growth, and resistance towards root pathogens and drought (Bowen, 1968; Zak, 1971; Parke et al., 1983; Dosskey et al., 1990; Castellano, 1996).  The vessel-like rhizomorphs that characterize the mycelia of these species are particularly effective for vegetative dispersion (Simard et al., 1997b; Agerer, 2001) and in the uptake and redistribution of nutrients and water over ecologically relevant distances (Duddridge et al., 1980; Brownlee et al., 1983; Read & Boyd, 1986; Egerton-Warburton et al., 2007; Warren et al., 2008).  The unique tuberculate morphology of their mycorrhizas (a nodular form consisting of numerous tightly bundled fine-root tips encased by a hyphal sheath; see Trappe, 1965) contributes to the resilience of Douglas-fir trees by protecting roots from pathogens (Zak, 1971) and the effects of extreme temperatures and moisture deficits (Parke et al., 1983; Goodman & Trofymow, 1998).  Moreover, the distinct morphological characteristics of R. vesiculosus and R. vinicolor facilitate recognition in the field and the isolation of tree and fungal DNA in the lab.    Throughout the growing season, Rhizopogon vesiculosus and R. vinicolor are prolific producers of hypogeous sporocarps (Molina et al., 1999), which are truffle-like fruitbodies produced belowground (e.g., analogous to aboveground mushrooms, they are the vessels in which meiosis occurs and basidiospores develop).  These sporocarps are especially important in the diet of small mammals (e.g., squirrels), some of which are keystone species in food webs of North American forests (Maser & Maser, 1988).  In return, small mammals, some large mammals, and the predators of small mammals act as vectors for the long-distance dispersal of Rhizopogon basidiospores which pass unabated through their digestive tracts (Colgan III & Claridge, 2002; Ashkannejhad & Horton, 2006).  Rhizopogon basidiospores may remain viable in the soil for a decade or longer before germinating to form a haploid, monokaryotic mycelium (Bruns et al., 2009).  When two compatible monokaryotic mycelia meet, they fuse and form a dikaryotic mycelium (i.e., somatic hyphal cells contain two distinct but co-occurring haploid nuclei), which represents the dominant phase of the Rhizopogon life cycle.    15Taxonomy of the Rhizopogon  genus has been in flux since the seminal work of Smith and Zeller (1966), who placed R. vesiculosus and R. vinicolor into different sections (Fulviglebae and Villosuli, respectively) within Rhizopogon (a subgenus at the time).  The current placement of R. vesiculosus and R. vinicolor in subgenus Villosuli within the Rhizopogon genus is strongly supported by nuclear ribosomal DNA (nrDNA) ITS sequence data (Grubisha et al., 2002; Kretzer et al., 2003), and by their morphological and ecological similarities (Kretzer et al., 2004; Kretzer et al., 2005).  The frequent spatiotemporal co-occurrence of R. vesiculosus and R. vinicolor, despite their similar host preferences, phylogenies and mycorrhiza morphologies, provides a model system for investigating ecological and evolutionary mechanisms enabling species coexistence.    1.4 Project objectives 1.4.1 Overview The overarching objective of this study was to characterize the socio-spatial architecture of MNs linking mature overstory trees and understory regeneration in an old-growth interior Douglas-fir forest.  I used intensive DNA-based sampling across multiple spatial scales to: determine the fine-scaled spatial patterns and autecological traits of Rhizopogon vesiculosus and R. vinicolor genets in xeric and mesic soil moisture regimes (chapter 2), characterize the spatial patterns of Rhizopogon spp. genets and the architecture of MNs formed between different tree cohorts at the forest stand scale (chapter 3), contrast the emergent structural properties of MNs between phytocentric and mycocentric perspectives and between xeric and mesic soil moisture regimes, and determine the critical determinants of MN architectures (chapter 4).  Soil moisture regimes were contrasted to provide site series specific guidelines, test whether the importance of MN facilitation increases with site water stress, and as a proxy for changes in site water stress and regeneration potential with climate change.  1.4.2 Chapter 2 objectives In the first part of this study, I used high-resolution sampling at fine spatial scales (20 cm3 resolution continuously excavated across 1-2 m extents) to identify ecological factors contributing to niche differentiation between coexisting R. vesiculosus and R. vinicolor genets.  I contrasted the autecological traits of R. vesiculosus and R. vinicolor between plots  16with xeric versus mesic soil moisture regimes to assess the resilience of these species to soil water stress.  Specifically, the objectives of chapter 2 were to: (1) describe the 3-dimensional spatial patterns, microsite associations, and mycelia exploration type of individual R. vinicolor and R. vesiculosus genets; and (2) compare these traits between xeric versus mesic soil moisture regimes.    I hypothesized that R. vesiculosus and R. vinicolor coexist perennially in interior Douglas-fir forests via niche differentiation as determined by spatial segregation, differing microsite associations, or differing exploration strategies.  Based on previous studies suggesting R. vinicolor forms smaller genets than R. vesiculosus (Kretzer et al., 2004; Kretzer et al., 2005), I predicted that R. vinicolor would have a more concentrated mycelial growth form as a competitive strategy amongst resource-rich patches.  I predicted that both species would be more prevalent in xeric relative to mesic soils, due to their drought resistant root tip morphology and increased positive feedbacks between Rhizopogon spp. and trees in stressful environments as per the stress-gradient hypothesis (e.g., preferential allocation of tree photosynthate toward the most beneficial EMF) (i.e., Callaway et al., 2002; Espeland & Rice, 2007; Maestre et al., 2009).  1.4.3 Chapter 3 objectives In the second part of this study, a 30 x 30 m plot was intensively sampled to identify the coarse-scale spatial patterns and mycorrhizal networking architecture of Rhizopogon spp. genets and interior Douglas-fir tree roots in high sampling resolution.  The objectives of chapter 3 were to: (1) determine if R. vesiculosus and R. vinicolor genets colonize multiple tree cohorts; (2) describe the size and spatial patterns of the Rhizopogon spp. genets; (3) compare the networking qualities between R. vesiculosus and R. vinicolor genets; and (4) determine the architecture of a mycorrhizal network.  Based on the aforementioned size differences observed between R. vesiculosus and R. vinicolor genets (Kretzer et al., 2004; Kretzer et al., 2005), I predicted that R. vesiculosus would link more trees than R. vinicolor.  I also predicted that tree size differences between young and old cohorts would be reflected in differences in MN connectivity, resulting in a scale-free network architecture where large trees serve as hubs.  171.4.4 Chapter 4 objectives The primary objective of chapter 4 was to compare the complexity of MNs between sites that differed in soil moisture regimes to infer the relative importance of MNs in forest stand structure with increasing site water stress.  My specific objectives were fourfold: (1) to describe the universal emergent structural properties of MNs at the forest stand scale; (2) to contrast network architectures between phytocentric and mycocentric perspectives; (3) to identify the relative contributions of individual trees and Rhizopogon spp. genets in determining the architecture of the MNs; and (4) to contrast the architecture of MNs between forest plots with xeric versus mesic soil moisture regimes.  Because this study took place in a self-regenerating mixed-aged forest, I predicted that size asymmetries among trees and fungal genets would lead to complex, scale-free MN topologies.  As in chapter 3, I predicted that node connectivity and link strength would increase in relation to the size of trees and fungal genets, such that large mature trees and R. vesiculosus genets contribute more towards network connectivity than small trees and R. vinicolor genets.  I hypothesized that MNs from the mycocentric perspective would have more regular, cliquish network topologies than the phytocentric model, based on anecdotal evidence of EMF exhibiting less spatial overlap between conspecific genets compared to interior Douglas-fir tree roots at the spatial and taxonomic scales investigated.  Lastly, I expected greater MN connectivity in xeric stands relative to mesic stands in keeping with the stress-gradient hypothesis (e.g., Bertness & Callaway, 1994; Callaway, 1995).   18Chapter 2:   Vertical niche partitioning occurs between sister species of Rhizopogon fungi on mesic and xeric sites in an interior Douglas-fir forest  2.1 Introduction The organizational structure of species assemblages across space and time is a central theme in community ecology.  It is crucial for the assessment and conservation of vulnerable populations, and for predicting responses to autogenic and allogenic change.  Ectomycorrhizal fungi (EMF) play a key role in forest ecosystems through their influences on nutrient cycling and above and belowground community dynamics (Trappe & Luoma, 1992; Francis & Read, 1994).  As a community, EMF influence the nutrition and water use efficiency of plants, protect plant roots from pathogens and environmental stress, and link above and belowground communities through complex multitrophic interactions (Read, 2002; Bonfante & Anca, 2009).  However, little is known regarding the mechanisms driving the spatial structure and temporal dynamics of EMF communities due to their cryptic nature.    Representative EMF species have been shown to partition vertically in the soil (Taylor & Bruns, 1999; Dickie et al., 2002b; Rosling et al., 2003) and among forest floor microsites (Goodman & Trofymow, 1998; Tedersoo et al., 2008b).  Temporal shifts in EMF community structure have been documented over monthly or seasonal cycles (Koide et al., 2007; Courty et al., 2008; Pickles et al., 2010) and through decades of forest stand development (Twieg et al., 2007; Twieg et al., 2009).  EMF species also vary in their response to site water stress (Querejeta et al., 2009), the frequency and severity of disturbances (Stendell et al., 1999; Jones et al., 2003; Cavender-Bares et al., 2009), and environmental change (Fransson et al., 2001; Avis et al., 2003).  While these studies have advanced our understanding of EMF communities, information gaps persist regarding the life history traits of species, even among dominant community members, and limited taxonomic resolution may conceal fine-scaled structure within cryptic species clades (Lilleskov et al., 2004).    Niche theory provides a framework for describing the co-existence of organisms in a community, particularly those similar in form or ecology, through their partitioning of resources in various dimensions of ecological space (e.g., via different spatial or temporal  19patterns, trophic status, or natural histories).  This approach assumes that a habitat is heterogeneous, and physiological variation among competing species influences their success in any combination of biotic and abiotic habitat conditions (Tilman, 1982).  Competition for patchily distributed soil resources (e.g., host roots or concentrated nutrient patches) has been shown to influence EMF community dynamics and could lead to niche differentiation between species (Wu et al., 1999; Kennedy & Bruns, 2005; Kennedy, 2010).    The life history traits of EMF species ultimately determine their reproductive success in any combination of biotic and abiotic conditions in their environment.  These traits include host and habitat specificity, mycelia exploration type, persistence, modes of reproduction and dispersal, and the phenology and physiological plasticity of these traits throughout an individual?s life cycle.  The exploration type of EMF, as described by Agerer (2001), constitutes the spatial patterns and extent of mycelia growth through the soil, and is used to categorize EMF by their extramatrical foraging patterns.  Here, I expand on this classification scheme to include the spatial continuity and pervasiveness of mycelia and the spatial patterns of root colonization (e.g., root networking characteristics) exhibited by EMF species.  These attributes help clarify the carbon demands and acquisition strategies of EMF, which may not be evident from the anatomy and spatial organization of hyphae and rhizomorphs alone.  Fungal mycelia are typically distributed non-randomly in relation to the complexity and nutrient status of soils, which has been related to the physiological ability of fungi to exploit those environments (Read, 1992; Goodman & Trofymow, 1998; Cavender-Bares et al., 2009).  Thus, EMF exploration types may be associated with ecological functions (Allen & Allen, 1992; Agerer, 2001; Agerer, 2006).  For example, EMF species forming dense mycelial mats (radial, or ?closed? form) may exclude or displace neighbors from localized resource patches, while long-distance strand forming species (directed, or ?open? form) are more suited for evading competition or reaching irregularly distributed resources (Ogawa, 1985; Andrews, 1992).  Accordingly, mycelial connections between plant roots could range from local to long-distance and spatially continuous to discontinuous, and could either represent network links (a mycelium colonizing more than one plant) or loops (a mycelium colonizing multiple roots from the same plant).      20Rhizopogon vesiculosus and R. vinicolor sensu Kretzer et al. (2003) (Boletales, Basidiomycota) are closely related ?sister? species that are among the most frequent and abundant EMF colonizers of interior Douglas-fir (Jones et al., 1997; Goodman & Trofymow, 1998; Twieg et al., 2007).  Both species produce hypogeous sporocarps, form highly differentiated rhizomorphs and distinctive tuberculate mycorrhizas (?tubercles?; a nodular morphology consisting of numerous tightly bundled fine-root tips encased by a hyphal sheath), and share narrow host specificity for Douglas-fir (Molina & Trappe, 1994).  The fungal sheath that encases R. vesiculosus and R. vinicolor mycorrhizas protects the internal tree and fungal tissues from extreme temperatures and water deficits, and helps facilitate recovery when conditions become favorable (Parke et al., 1983; Goodman & Trofymow, 1998).  Moreover, the vessel-like rhizomorphs formed by fungi such as R. vesiculosus and R. vinicolor are capable of transporting water and dissolved nutrients over ecologically significant distances (Brownlee et al., 1983; Egerton-Warburton et al., 2007; Warren et al., 2008).  Studies have found that R. vesiculosus genets tend to grow larger than those of R. vinicolor (Kretzer et al., 2004; Kretzer et al., 2005), and that the larger genets of R. vesiculosus potentially link more trees through mycorrhizal networks than R. vinicolor genets (chapter 3 this dissertation).  Despite these insights, the mechanisms allowing R. vesiculosus and R. vinicolor to coexist in Douglas-fir forests remained unknown.   This study sought to identify ecological factors contributing to niche differentiation of R. vesiculosus and R. vinicolor in natural settings.  My first objective was to describe the 3-dimensional spatial patterns, microsite associations, and mycelia exploration type of individual R. vesiculosus and R. vinicolor genets in an interior Douglas-fir forest.  Herein, microsite features include the available ground cover substrates, vascular plants and bryophyte species across horizontal space, and differing soil layers, buried coarse woody debris (CWD) and rock surfaces in vertical space.  My second objective was to compare these traits between xeric versus mesic soil moisture regimes to assess the resilience of these species to soil water stress.  I hypothesized that R. vesiculosus and R. vinicolor coexist perennially in interior Douglas-fir forests via niche differentiation as determined by spatial segregation, differing microsite associations, or differing exploration strategies.  Given the smaller spatial extent of R. vinicolor genets compared to R. vesiculosus, I predicted that R.  21vinicolor maintains a competitive advantage within resource-rich patches (including host roots) via a more concentrated (pervasive) growth form.  I also predicted that both species would be more prevalent in dry soils relative to moist soils, due to their drought resistant root tip morphology and increased facilitative effects of mycorrhizas in stressful environments as per the stress-gradient hypothesis (e.g., preferential allocation of tree photosynthate toward the most beneficial symbionts) (Callaway et al., 2002; Espeland & Rice, 2007; Maestre et al., 2009).      2.2 Materials and methods 2.2.1 Plots and sampling This study took place in a mixed-aged, old-growth interior Douglas-fir (P. menziesii var. glauca (Beissn.) Franco) forest located near Kamloops, Canada (Table 2.1; Fig. 2.1).  The forest is in the Thompson Dry, Cool Interior Douglas-fir biogeoclimatic subzone (IDFdk) and receives approximately 44 cm of annual precipitation; the majority occurring as snow during winter months (Lloyd et al., 1990).  Mean maximum and minimum temperatures are 21.0?C and -4.2?C (mean annual = 3.4?C), with an average growing season of 166 days (April-October) (Environment Canada; Canadian climate norms, 1971-2000).  Mineral soils were Orthic Dystric Brunisols (Canadian System of Soil Classification 1998) with sandy loam texture and the humus form was a Hemimor (Green et al., 1993).  The F and H layers of the forest floor (partially to fully decomposed organic materials pervaded with fungal hyphae) were underlain by Ae (eluviated) and Bm (modified) mineral horizons, often with a sharp boundary between the H forest floor layer and A mineral horizon.    To examine soil moisture regime effects on species traits, I selected three plots with xeric site characteristics and three with mesic site characteristics within single-species stands of interior Douglas-fir forest.  Soil moisture regimes were distinguished based on mesoslope position (upper and lower slopes) and soil and vegetation characteristics as per Lloyd et al. (1990).  Xeric plots (?plots 1-3?) were located on upper slope positions and had an understory plant community dominated by pinegrass (Calamagrostis rubescens Buckl.), while mesic plots (?plots 4-6?) were located on lower slope positions with a high cover of red-stemmed feathermoss (Pleurozium schreberi Brid.) (Appendix A) (Lloyd et al., 1990).   22The six plots were located 150-600 m apart and contained a diversity of plants and forest floor substrates.  Mean forest floor depths (combined L, F, and H layers) differed between mesic versus xeric plots (t = 6.48, P < 0.01); and were generally deeper in mesic plots (6.20 ? 0.25 cm) compared to xeric plots (3.87 ? 0.13 cm).  Elevation ranged from 996-1,155 m above sea level.  All plots had gentle slopes (15-20%) and east/southeasterly aspects.    Preliminary sampling in a hierarchically-scaled grid (with 2 cm, 10 cm, 20 cm, 50 cm, and 1 m lag distances; data not shown) along with published work regarding the population structure of R. vesiculosus and R. vinicolor (Kretzer et al., 2005) indicated the appropriate spatial scale and corresponding sampling scheme for comparing species at coarse spatial scales using the fine-scaled attributes of individual genets.  In each plot, a total of 19 contiguous 20 cm3 ?soil blocks? were sampled from two perpendicular transects, spanning 2 m length x 20 cm width and crossing near the center, similar to the design of Zhou et al. (2001) (see Fig. 2.2, Appendix A).  One randomly selected plot (plot 5) was subsequently excavated to a contiguous 1 x 2 m extent (50 soil block total, including 14 shared with the initial plot) to provide auxiliary estimates of genet continuity, but these data were excluded from all species and plot comparisons (Appendix A).    Each 20 cm3 soil block was searched systematically from the top down for Rhizopogon-like hyphae, rhizomorphs, or tuberculate mycorrhizas.  For habitat association measures, soil substrata were pooled into four new soil layers: the F and H forest floor layers, the interface between the H forest floor layer and A mineral horizon, the A mineral horizon, and the B mineral horizon.  In addition to these four, I examined buried CWD and rock surfaces (rocks ? 15 cm circumradius) independent from the soil layers in which they occurred.  Substrata were pooled into three new soil layers for the classification of Rhizopogon spp. mycelia presence/absence and pervasiveness: the combined F and H forest floor layers, the A mineral horizon, and the B mineral horizon.  As a measure of pervasiveness, the occurrences of Rhizopogon spp. mycelia in each soil block and soil layer were classified into one of five ?density? classes, labeled as: absent, scarce, diffuse, patchily dense, and dense.  These classes were based on the number of Rhizopogon tubercles present and mycelia growth form (e.g., diffuse rhizomorphs versus dense mats of hyphae).  Specifically, mycelia density was  23classified as: absent where there was no visible Rhizopogon-like mycelia; scarce where there was ? 1 tubercle and sparse hyphae & rhizomorphs covering ? ? of the sample block; diffuse where there were ? 10 tubercles and sparse hyphae or rhizomorphs covering > ? of the sample block; patchily dense where there were ? 10 tubercles or with dense hyphal mats covering ? ? of the sample block; or dense where there were ? 10 tubercles or with dense hyphal mats covering > ? of the sample block.      Two or more samples of Rhizopogon spp. mycorrhizas, hyphae, or rhizomorphs were collected from each soil block and soil layer (when available), placed in 2 ml polypropylene tubes, and frozen at -20?C for molecular analysis.  The depth range of each sample was measured in cm from the surface of the L soil layer.  A total of 371 samples were collected, consisting of 58.2% tuberculate mycorrhizas and 24.3% hyphae or rhizomorphs.  Among mycorrhiza samples, 66% were identified as R. vesiculosus, 27.2% as R. vinicolor, and 6.8% could not be identified to species.  The Rhizopogon spp. were encountered as hyphae in equal proportions.  Remaining samples consisted of 11.9% bulk soil and 5.6% degraded tuberculate mycorrhizas, and served to confirm the absence of target species while also serving as negative controls against possible sample contamination from Rhizopogon spp. spores or isolated hyphal fragments.  Only samples verified to be R. vesiculosus or R. vinicolor by microsatellite DNA analysis were included in qualitative or quantitative descriptions.    2.2.2 Molecular analysis A random subsample of Rhizopogon-like rhizomorphs, mycelia and individual root tips within tuberculate ectomycorrhizas were subject to molecular processing.  Fungal DNA was extracted from samples using the Qiagen DNeasy? plant extraction kit and protocols (QIAGEN Inc., Mississauga, Canada).  Tree DNA was extracted concurrently with fungal DNA from tuberculate mycorrhizas.  Approximately 0.1 g of material was ground in 400 ?l of Qiagen?s AP1 lysis buffer mixed with 2 ?l anti-foam Y-30 emulsion (Sigma-Aldrich Canada Ltd., Oakville, ON) in a FastPrep 120 high-speed shaker (MP Biomedicals, Solon, OH) for 45 s at 6.5 m/s.  Following extraction, fungal DNA was diluted 10-fold in ultrapure deionized water.  Presence of DNA product was verified by running a subsample through agarose gels stained with SYBR-Safe (Invitrogen Canada, Inc., Burlington, ON),  24electrophoresed, and visualized under UV light (Gel Logic 440 Imaging System, Mandel Inc., Guelph, ON).  Genets of R. vesiculosus/R. vinicolor were distinguished by the loci Rv02, Rv15, Rv46, Rv1.34, Rve2.77 and Rve3.21 for both species, plus Rve2.10, Rve1.21 and Rve2.44 for R. vesiculosus, and Rv53 and Rv2.14 for R. vinicolor (Kretzer et al., 2004).  Interior Douglas-fir trees were distinguished using microsatellite markers developed by Slavov et al. (2004) targeting the loci PmOSU_1C3, PmOSU_1F9 and PmOSU_2D4.  Microsatellite DNA markers were selected based on their amplification success and replicability, polymorphism rates, and compatibility in multi-plex PCR amplification.  Polymerase chain reaction targeting the internal transcribed spacer (ITS) region between the 18S and 28S of rDNA using the ITS1F and ITS4 primer pair, followed by a digestion with the Alu I restriction enzyme and gel electrophoresis, was initially used to distinguish fungal species.  Once the accuracy of identifying these species with microsatellite loci alone was confirmed, the primer pairs Rve2.10 and Rve2.14 were used to distinguish between R. vesiculosus and R. vinicolor.  Microsatellite loci were amplified together in PCR multiplex groups of 2-4 primer sets each, with forward primers labeled at the 5? end with either 6FAM (Integrated DNA Technologies Inc., Coralville, USA), NED, VIC, or PET (ABI; Applied Biosystems, Foster City, USA) fluorescent dyes.  A ?PIG-tail? tack-on sequence was applied to the 5? end of reverse primers targeting interior Douglas-fir microsatellite DNA to reduce stutter banding as per Brownstein et al. (1996).  PCR amplifications were conducted in 10 ?L reaction volumes containing ultrapure H20, GeneAmp 10x PCR Buffer (ABI), 10 mg/ml BSA, 1 mM GeneAmp dNTP mixture (ABI), 2.5 mM MgCl2, 0.25-0.35 mM of each primer, 10 ng DNA template and 5 U/mL AmpliTaq Gold polymerase (ABI).  Temperature cycles began with a 95?C hot-start for 10 minutes; followed by 30 cycles of 45 s at 93?C, 30 s at 55?C, and 30 s at 72?C; and a final hold for 7 min at 72?C.  Microsatellite loci were analyzed using a DNA sequencer (3130XL genetic analyzer, ABI) and then genotyped using Gene Mapper software (V4.0, ABI).  Two or more samples were considered to represent an individual fungal genet or tree root system if alleles matched at all microsatellite loci analyzed.  The probabilities of identity of Rhizopogon spp. and interior Douglas-fir genotypes were estimated with the inclusion of additional samples collected in the vicinity of study plots and are reported in chapter 3.   252.2.3 Data analysis and statistics Statistical tests were performed using SAS 9.2.2 (Statistical Analysis Software, SAS Institute, Cary, USA) unless otherwise noted, with ? = 0.05.  Descriptive statistics are reported as a range with mean and standard deviation (SD).  The frequency of occurrence of Rhizopogon spp. mycelia within plots (presence/absence among 20 cm3 soil blocks) was compared between species and between mesic and xeric soil moisture regimes in a generalized linear binomial model with a logit link function using PROC GLIMMIX.  The median depths of DNA-identified mycelia (hyphae, rhizomorphs, or tubercles) were compared between species, and between soil moisture regimes for each species, in separate Wilcoxon Rank Sums tests using SPSS 16.0 Graduate Student Version for Windows (IBM Corporation, Somers, USA).  The minimum, maximum, and median depth of Rhizopogon spp. mycelia were first determined for each 20 cm3 soil block, and then averaged across genets separately for each of the six plots.    Associations between R. vesiculosus/R. vinicolor and available forest floor substrates (needle litter, CWD, rock surfaces, or bare ground), vascular plants and bryophyte species, and soil microsites (soil layers, buried CWD, and rock surfaces) were assessed using Manly?s alpha selectivity index as described in Krebs (1989).  I calculated the habitat associations of each Rhizopogon species, first assuming the presence of one species did not influence the presence of the other (i.e., random dispersal), then assuming competitive exclusion between species (i.e., sampling units occupied by one species were considered unavailable to the other).  The frequency of above or belowground microsite types were based on their presence or absence among 20 cm3 soil blocks in all plots.  Plants encountered in less than 5% of soil blocks were excluded from habitat association measures, as were lichens.    The frequency distributions of Rhizopogon spp. mycelia occurrence (presence/absence) and density class were compared between soil layers, soil blocks within transects, Rhizopogon species, and soil moisture regimes in multi-dimensional contingency tables using Chi-square tests.  A generalized linear binomial model with logit link function was obtained to predict the probability of R. vesiculosus or R. vinicolor mycelia occurrence (presence/absence) by soil blocks within transect plots, soil layers, and soil moisture regimes.  I then obtained a  26generalized linear multinomial logistic model with a cumulative-logit link function to predict the probability of each R. vesiculosus or R. vinicolor mycelia density class by soil blocks within transect plots, soil layers, and soil moisture regimes.  In both models, Euclidean distances between the centroids of each soil block and soil layer combination were incorporated to account for potential 3-dimensional spatial autocorrelation among samples using PROC GLIMMIX.  I first tested expanded versions of the models with soil blocks independent of transect plots and the inclusion of interactions between predictor variables (data not shown), but these results did not differ from those of the reduced models.    The mean number of trees colonized by genets and the mean number of trees linked by shared and spatially continuous mycelia were compared between Rhizopogon spp., and between soil moisture regimes for each species independently, in Wilcoxon Rank Sums tests using SPSS.  The spatial continuity of fungal genets was calculated using ?queen-case? adjacency (i.e., soil blocks with shared edges or corners) in ROOKCASE (Sawada, 1999) and is reported as the proportion of spatially joined soil blocks with a genet?s mycelium present out of the total number of joins possible within transect plots.    2.3 Results 2.3.1 Spatial patterns of Rhizopogon spp. genets Averaged by genet, the median depths at which R. vesiculosus and R. vinicolor mycelia were encountered differed significantly between species and between xeric and mesic soil moisture regimes for each species (each with P < 0.05 in separate Wilcoxon Rank Sums tests; Fig. 2.1).  Rhizopogon species and soil moisture regime were not significant factors in the generalized linear binomial model fit to predict the overall frequency of occurrence of R. vesiculosus and R. vinicolor mycelia among soil blocks in each plot (each factor with P > 0.10; Table 2.1).  Rhizopogon vesiculosus and R. vinicolor mycelia shared the same horizontal space in 8% of the 20 cm3 soil blocks sampled, but no overlap was observed between genets of the same species.  Both species were found concurrently in four of the six  transect plots, but multiple genets of the same species were only found in ?Plot 6?, where two R. vesiculosus and two R. vinicolor occurred but were segregated in horizontal space (Fig. 2.2).  272.3.2 Rhizopogon spp. habitat associations The dominant forest floor substrates and frequency among soil blocks included needle litter (50%), coarse woody debris (33%), rock ? 15 cm circumradius (13%), and bare ground (10%).  Among vegetation covering the forest floor, P. schreberi occurred in 55% of soil blocks and C. rubescens occurred in 51%, with Arcotostaphylos uva-ursi (L.) Spreng., Spiraea betulifolia Pall., Goodyera oblongifolia Raf., Rosa acicularis Lindl., and Viola renifolia A. Gray occurring in 5-10% of soil blocks.  I found no evidence of Rhizopgon spp. mycelia within the litter layer (L) of the forest floor.  The results of Manly?s habitat selection index further clarified the 3-dimensional spatial partitioning of R. vesiculosus and R. vinicolor among major ground cover components (needle litter, CWD, large rocks, bare ground, and species of vascular plants or bryophytes), and among belowground microsites (the forest floor, interface between the forest floor and A mineral horizon, A and B mineral horizons, CWD, and rock surfaces).  Habitat associations varied depending on whether the assumption of competitive exclusion between R. vesiculosus and R. vinicolor was made prior to the analysis.  Among ground cover, R. vesiculosus was associated with CWD, needle litter, rock, G. oblongifolia, P. schreberi, R. acicularis, and S. betulifolia when soil blocks having R. vinicolor were excluded from the analysis, but only rock, G. oblongifolia, P. schreberi, and R. acicularis with the inclusion of blocks having R. vinicolor.  Rhizopogon vinicolor was associated with CWD, needle litter, rock, C. rubescens, and S. betulifolia whether soil blocks with R. vesiculosus were included or excluded from the analysis.  Among belowground microsites, R. vesiculosus and R. vinicolor were both abundant at the interface between the H forest floor layer and A mineral horizon, while R. vinicolor was more abundant in forest floor layers (F + H) and buried CWD and R. vesiculosus was more abundant in the A and B mineral horizons and against rock surfaces.  However, R. vesiculosus was more strongly associated with buried CWD than the A and B mineral horizons when soil blocks containing R. vinicolor were excluded from the analysis.     282.3.3 Rhizopogon spp. mycelial growth form The spatial patterns and pervasiveness of R. vesiculosus and R. vinicolor mycelia within transect plots, including the median mycelia density class among forest floor layers and mineral horizons in each 20 cm3 soil block, are shown in Figure 2.2.  A greater proportion of R. vesiculosus occurrences were characterized as ?patchily dense? or ?dense? compared to R. vinicolor, and the occurrence of R. vinicolor was heavily biased toward forest floor layers, while R. vesiculosus was frequent in both the forest floor and mineral soils (Fig. 2.3).  Mycelia density class distributions differed significantly between Rhizopogon spp. and between soil moisture regimes for each species, whether analyzed separately by soil layer (the combined F and H forest floor layers, the A mineral horizon, and the B mineral horizon) or regardless of soil layer (Table 2.2).  Density class comparisons between species in the A and B mineral horizons, and between soil moisture regimes for R. vinicolor, should be interpreted with caution since R. vinicolor mycelia were largely absent from mineral horizons.  The results of the generalized linear mixed model showed the pervasiveness of mycelia was significantly influenced by species, soil layer, and soil moisture regime (species: F1,677 = 76.72, P < 0.01; soil layer: F2,677 = 31.82, P < 0.01; soil moisture regime: F1,677 = 5.23, P < 0.05).  2.3.4 Mycelia continuity, root colonization patterns and networking Of the fungal genets encountered two or more times as mycorrhizas (n = 10 out of 12 total genets encountered), all were found to colonize more than one tree (range = 2 - 7 trees, mean = 3.91 ? 2.07 trees).  The number of trees colonized by these genets differed significantly between Rhizopogon spp. (P = 0.024, W = 17, n1,2 = 5), with R. vesiculosus (range = 4 - 7 trees, mean = 5.60 ? 1.34 trees) generally colonizing more trees than R. vinicolor (range = 2 - 5 trees, mean = 2.80 ? 1.30 trees).  The spatial continuity of mycelia (the percent of ?queen? adjacencies out of the total possible in each transect plot) were similar between R. vesiculosus (range = 0 - 68 %, mean = 33 ? 27 %) and R. vinicolor (range = 0 - 55 %, mean = 18 ? 21 %).  However, R. vesiculosus mycelia occupied nearly twice the area (65 %) and had twice the frequency of queen adjacencies (42 %) as R. vinicolor (35 % and 17 % respectively) with the inclusion of auxiliary soil blocks sampled in the contiguous 1 x 2 m area of plot 5.  Similar to the tree colonization patterns, the number of spatially continuous  29?linkages? between roots differed significantly between species (P < 0.016, W = 16, n1,2 = 5), with R. vesiculosus genets (range = 5 - 38 links, mean = 18.20 ? 12.79 links) linking more trees than R. vinicolor genets (range = 1 - 6 links, mean = 2.60 ? 2.07 links).  Thirteen percent of fungal links between trees occurred within the same 20 cm3 soil blocks, and 33 % of links occurred within 40 cm distance or less.  The mean topological distance (units measured in 20 cm2 increments) between tree roots linked by spatially continuous mycelia was 1.5 units for R. vesiculosus and 1.6 units for R. vinicolor.  Additionally, 14 % of trees colonized by either Rhizopogon species were encountered more than once within spatially continuous mycelia (these represent ?loops? in terms of network linkages).  No significant differences were found in the mean numbers of trees colonized or linked by genets between mesic and xeric soil moisture regimes, nor were differences found in the spatial continuity of mycelia, for either R. vesiculosus or R. vinicolor.  2.4 Discussion 2.4.1 Spatial patterns of Rhizopogon spp. genets I found that genets of R. vesiculosus and R. vinicolor partitioned vertically where they overlapped in horizontal space, supporting my first hypothesis that niche differentiation between the two sister-species may occur through spatial segregation.  R. vesiculosus occurred at significantly greater depths than R. vinicolor by all measures of mycelial depth, including the minimum, median, or maximum depths (minimum and maximum depth comparisons not shown).  Moreover, although the two species frequently overlapped in horizontal space, different genets within each species did not overlap.  These results provide insight into processes for niche partitioning and species coexistence at fine scales, which heretofore has remained elusive for mycorrhizal fungi.  For example, the broader depth range of R. vesiculosus genets compared to R. vinicolor genets could reflect a more evasive competitive strategy or better access to water resources in deeper soils during dry periods.  Studies have demonstrated vertical stratification between coarse taxonomic groups of EMF (e.g., between different genera or species clades), suggesting that the vertical partitioning of space or resources may play a role in shaping EMF community structure and species coexistence (Dickie et al., 2002b; Rosling et al., 2003; Tedersoo et al., 2003; Baier et al., 2006; but see Zhou & Hogetsu, 2002).  302.4.2 Rhizopogon spp. habitat associations The associations between R. vesiculosus and particular forest floor substrates were influenced by the presence of R. vinicolor, but not vice versa.  A similar pattern was observed among microsites in the soil profile, where the prevalence of R. vesiculosus mycelia shifted from CWD toward rock surfaces and mineral soil horizons in the presence of R. vinicolor.  This could result from direct competition if R. vinicolor displaces R. vesiculosus from more productive microsites, or through indirect competitive advantages (resource exploitation or interactions with other organisms or the environment) if R. vinicolor is more equipped physiologically to exploit these areas or to defend them from common enemies (Culver, 1992).  For example, soil blocks in which R. vinicolor were encountered had a substantially deeper forest floor compared to those with only R. vesiculosus present (data not shown) which could reflect a direct or indirect competitive advantage afforded to R. vinicolor in these soil layers.  Among the types of ground cover present in transect plots, R. vesiculosus and R. vinicolor appeared indiscriminant among vascular plant and bryophyte species, and were commonly associated with needle litter, CWD, and rock surfaces.  The only plant negatively associated with both R. vesiculosus and R. vinicolor was Arctostaphylos uva-ursi, a shrub that associates with many of the same EMF species as interior Douglas-fir (Horton et al., 1999; Hagerman & Durall, 2004).  The most parsimonious explanation is that A. uva-ursi does not associate with R. vesiculosus or R. vinicolor, and EMF species it does associate with probably exclude the presence of Rhizopogon spp. in the A. uva-ursi rhizosphere due to competition.  The greatest overlap between R. vesiculosus and R. vinicolor microsite associations occurred along the interface between the H forest floor layer and the A mineral horizon.  This interface could be a productive microsite for EMF given the concentration of available host fine roots and eluviation of organic nutrients from the forest floor (Harvey et al., 1979; Schenk & Jackson, 2002; Rosling et al., 2003).  Mat-forming fungi such as Piloderma spp., Hysterangium spp., and Ramaria spp. have been found to be associated with this interface in P. menziesii var. menziesii forests (Entry et al., 1991; Read, 1992; Dunham et al., 2007).  Previous reports associated R. vesiculosus/R. vinicolor occurrence with coarse woody debris (Zak, 1971; Maser & Trappe, 1984; Kretzer et al., 2005), though no forest floor microsite  31associations were found when tested empirically by Goodman and Trofymow (1998).  In the present study, both R. vinicolor and R. vesiculosus (in the absence of R. vinicolor) were associated with CWD.  Studies have stressed the importance of CWD for moisture retention during periods of prolonged drought and as habitat for small mammals that are vectors for dispersing Rhizopogon spp. spores (Maser & Trappe, 1984; Amaranthus et al., 1989; Goodman & Trofymow, 1998).  I also frequently encountered R. vesiculosus and R. vinicolor covering the lower surfaces of rocks, large lateral roots, and other impervious soil features; structures under which moisture is condensed.  Given the moisture deficits of these forests, fierce competition for such microsites would be expected among EMF species as well as saprotrophic fungi and other soil organisms.  2.4.3 Rhizopogon spp. ?exploration type? Extramatrical mycelia: Because R. vesiculosus and R. vinicolor genets tend to occur in similar frequencies across space despite the larger size of R. vesiculosus genets, I predicted that R. vinicolor could maintain a competitive advantage by concentrating its proliferation and colonization of tree roots within localized resource-rich patches.  However, dense mycelia mats were more frequently encountered among R. vesiculosus genets, and R. vesiculosus genets linked more trees across contiguous space than R. vinicolor.  This pattern suggests that patches of dense mycelia proliferation and root colonization support the extensive exploration strategy of R. vesiculosus genets, which in turn could help regulate internal water potentials or offset the greater carbon demands associated with this strategy (e.g., long-distance transport).  Alternately, these patterns could result from differences between R. vesiculosus and R. vinicolor growth rates (Wu et al., 1999) or differential investment in spatial proliferation versus enzymatic capabilities (Peay et al., 2008).  The gross morphology of R. vesiculosus and R. vinicolor genets were similar to the irregular mat-type foraging pattern described by Ogawa (1985), and those described for Suillus bovinus by Bending and Read (1995) in microcosms, characterized by scattered patches of dense proliferation linked through diffuse rhizomorphs.      32Researchers have categorized EMF into differing exploration types based on associations between mycelia morphology and ecological roles (Agerer & G?ttlein, 2003; Agerer, 2006; Hobbie & Agerer, 2010).  The mixed use of directed and radial mycelia growth by both Rhizopogon spp. likely reflects the selective proliferation among spatially discontinuous resources (roots or nutrients), and would be a particularly advantageous foraging strategy in these conditions (Andrews, 1992; Read, 1992).  In general, the ?long-distance? exploration type of these species could influence their competitive dynamics with other EMF species directly (i.e., as a mode for evading competition) or indirectly (i.e., by gaining priority access to resources) (Kennedy, 2010).  With the combined use of microcosms (Kennedy & Bruns, 2005) and field experimentation (Kennedy et al., 2007), Kennedy et al. reported the competitive dominance of Rhizopogon occidentalis over R. salebrosus when given first priority for colonizing Pinus muricata seedlings, but neither species was excluded with concurrent root colonization.  Thus, priority effects could explain why both Rhizopogon spp. in my study occurred in similar frequencies across space, despite evidence that R. vinicolor excludes R. vesiculosus from their mutual associated habitats.        Root colonization patterns:  Both R. vesiculosus and R. vinicolor were found to link the roots of different trees in close proximity and repeatedly across a spatially continuous mycelium.  Nearly 10 % of the 114 soil blocks sampled contained roots from multiple trees colonized by a single fungal genet, and the mean distance of continuous fungal ?links? between trees was less than 40 cm for both Rhizopogon species.  The proximity at which roots were linked could reduce the physiological costs of transporting carbon and other tree-derived materials throughout the fungal mycelium.  The number of trees linked through continuous mycelia were similar between Rhizopogon spp. genets in this study and Tricholoma matsutake genets in a Pinus densiflora forest as reported by Lian et al. (2006), despite considerable differences between the growth form of these species (T. matsutake mycelia grows radially, forming compound multi-genet ?shiros?).  I found that tree pairs were linked repeatedly through contiguous network ?loops?, suggesting that if links between roots should become disrupted, the opportunity exists for the fungal mycelium to re-bridge those gaps through anastamosis (Fricker et al., 2008; Rotheray et al., 2008; Boddy et al., 2010).  The presence of both ?links? and ?loops? between trees in my study suggests R. vesiculosus  33and R. vinicolor are nonselective in their colonization of interior Douglas-fir tree roots when encountered.  Estimates of mycelia continuity and root connectivity are reduced by plot edge and shape effects (determined from auxiliary sampling, Appendix A; see also chapters 3 & 4).  2.4.4 Influence of soil moisture regime Soil moisture conditions have been proposed as a potential niche axis along which EMF communities are structured in space (see references in Erland & Taylor 2002) and time (e.g., seasonal patterns, Izzo et al. 2005).  I found that R. vesiculosus and R. vinicolor mycelia were more pervasive in plots with mesic compared to xeric soil moisture regimes.  This could result from tree and root productivity being higher in the more resource-rich mesic sites, thus providing greater substrate for Rhizopogon colonization.  The pattern is also consistent with the prevalence of R. vesiculosus and R. vinicolor within microsites associated with moisture retention, suggesting these species could play an important role in interior Douglas-fir water dynamics.  However, the results were in contrast to my prediction that both species would be more prevalent in xeric conditions due to the competitive advantage of their drought-tolerant morphology coupled with an increased carbon investment in belowground sinks by moisture-stressed trees as per the stress-gradient hypothesis (Simard, 2009).  The overall pervasiveness of R. vesiculosus and R. vinicolor mycelia could help ameliorate any stand-level effects of soil moisture stress (i.e., through the redistribution of water by Rhizopogon mycelia or other multifarious interactions).  In a loblolly pine stand, Lorio et al. (1972) reported a positive relationship between soil moisture and the surface area of ectomycorrhizas, with an abundance of ?nodular? forms during periods of excess moisture.  Kennedy and Peay (2007) found that soil water stress influenced the relationships between three Rhizopogon species and Pinus muricata hosts, with suppressed root biomass and colonization rates observed in drier soils.  My finding that R. vesiculosus and R. vinicolor mycelia occurred significantly deeper in mesic soils compared to xeric soils was likely due to an interaction with forest floor depth, since mean forest floor depths were greater among mesic plots.  There were no significant differences in the microsite affiliations or root colonization patterns between plots with mesic versus xeric soil moisture regimes for either R. vesiculosus or R. vinicolor.   342.4.5 Conclusions I found consistent evidence of vertical segregation between R. vesiculosus and R. vinicolor.  Results of this study suggest differences between the competitive ability and foraging strategy of these species in relation to water acquisition are plausible mechanisms for this partitioning.  Moreover, I found that R. vesiculosus mycelia were more pervasive across 3-dimensional space, and colonized a greater number of trees within that space, than R. vinicolor mycelia.  These findings, together with previous research showing that R. vesiculosus has larger genets and links more trees than R. vinicolor at the forest stand scale, suggests these species have evolved distinct life history strategies that may not be evident from their gross anatomy, mycelia architecture, or taxonomic status.  Although both species had a higher incidence of dense mycelia proliferation in mesic plots compared to xeric soil moisture regimes, their microsite associations and soil exploration strategies were conserved across treatment replicates.  The results of this study suggest the characterization of EMF communities with greater spatial and taxonomic resolution using microsatellite methods may unearth cryptic structural complexity within these communities and help to better understand potential mechanisms leading to this complexity.  To my knowledge, this is the first study to demonstrate fine-scaled niche partitioning between cryptic EMF species.     35Table 2.1    Characteristics of study plots including the depth of soil layers, occurrence frequency of ground cover substrates and plants, and frequency of Rhizopogon spp. mycorrhizae, hyphae, and rhizomorph samples among 20 cm3 soil blocks  Xeric Soil Regime  Mesic Soil Regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6 Degrees latitude (plot center) 50.851512 50.85292 50.853952  50.852341 50.853988 50.855909 Degrees longitude (plot center) -120.5225 -120.522711 -120.522573  -120.518983 -120.519856 -120.518568         Number of excavated 20 cm3 soil blocks per plot 19 19 19  19 19 19 Mean depth (cm) of F, H forest floor layers 4.0 4.2 3.3  9.3 3.5 6.9 Mean bottom depth (cm) of Ae mineral horizon 8.0 8.0 6.9  12.0 7.7 10.3         Proportion of soil blocks with R. vesiculosus present 79.0 % 5.3 % 63.2 %  57.9 % 21.1 % 63.2 % Number R. vesiculosus samples collected 32 22 29  37 26 42 Proportion of soil blocks with R. vinicolor present 0.0 % 63.2 % 31.6 %  31.6 % 0.0 % 31.6 % Number R. vinicolor samples collected 0 23 17  33 0 31   Frequency of occurrence of major ground cover components:   Needle litter 0.0 % 100 % 84.2 %  37.3 % 31.6 % 61.7 % Coarse woody debris (CWD) 78.9 % 0.0 % 5.3 %  45.8 % 0.0 % 36.2 % Rock (? 15 cm circumradius) 0.0 % 0.0 % 0.0 %  40.7 % 0.0 % 0.0 % Pleurozium schreberi  89.5 % 0.0 % 10.5 %  62.7 % 100 % 55.3 % Calamagrostis rubescens  94.7 % 89.5 % 94.7 %  23.7 % 78.9 % 25.5 % Spiraea betulifolia  0.0 % 21.1 % 36.8 %  0.0 % 0.0 % 0.0 % Arcotostaphylos uva-ursi  0.0 % 0.0 % 63.2 %  0.0 % 36.8 % 0.0 % Goodyera oblongifolia  0.0 % 0.0 % 0.0 %  11.9 % 5.3 % 4.3 % Rosa acicularis  78.9 % 5.3 % 0.0 %  0.0 % 0.0 % 0.0 % Viola renifolia  78.9 % 0.0 % 0.0 %  0.0 % 21.1 % 0.0 %  36Table 2.2    Multi-dimensional contingency tests revealed the distribution of mycelia density classes (spatial pervasiveness) differed significantly between Rhizopogon spp. and between mesic versus xeric soil moisture regimes for each species.  R. vesiculosus genets formed dense ?mycelial mats? more frequently than R. vinicolor, and both species had more spatially pervasive mycelia in mesic plots versus xeric plots (? = 0.05). R. vesiculosus x R. vinicolor  DF ?2 |P > ?2| Mycelia pervasiveness in F, H forest floor layers  4 17.44 0.002 Mycelia pervasiveness in Ae mineral horizon  4 26.93 <0.001* Mycelia pervasiveness in Bm mineral horizon  4 65.90 <0.001* Median density class across all soil layers  4 13.62 0.009       R. vesiculosus in mesic vs. xeric soil     Mycelia pervasiveness in F, H forest floor layers  4 10.01 0.040 Mycelia pervasiveness in Ae mineral horizon  4 27.77 <0.001 Mycelia pervasiveness in Bm mineral horizon  4 22.81 <0.001 Median density class  across all soil layers  4 14.89 0.005       R. vinicolor in mesic vs. xeric soil     Mycelia pervasiveness in F, H forest floor layers  4 11.07 0.026 Mycelia pervasiveness in Ae mineral horizon  4 30.22 <0.001* Mycelia pervasiveness in Bm mineral horizon  4 5.62 0.230* Median density class across all soil layers  4 29.42 <0.001 *more than 15% of cells have expected frequencies <5 because R. vinicolor was largely absent from mineral soil horizons  37   Figure 2.1    Map showing the locations of the six study plots in an interior Douglas-fir forest, with three plots (1-3) on upper slope positions with xeric soil moisture regimes and three plots (4-6) on lower slope positions with mesic soil moisture regimes (see main text).  Contour lines are labeled in US customary units (feet above sea level) and are based on 1:50,000 scale digital data (Garmin MapSource, Olathe, Kansas USA).    38  Figure 2.2    The median depths of occurrence (cm) of hyphae, rhizomorphs and tuberculate mycorrhizas averaged across Rhizopogon spp. genets from six independent plots differed significantly between species (i.) and between xeric and mesic soil moisture regimes for each species (ii.-iii.) (different letters indicate |P < t| < 0.05).  Boxplots show the median (middle line), mean (dashed line), and range (upper and lower bounding lines).  39 Figure 2.3    The spatial frequency, continuity, and pervasiveness of R. vesiculosus (in blue colour tones) and R. vinicolor mycelia (in red colour tones), excavated from six independent 2 x 0.2 x 0.2 m transect plots with either xeric or mesic soil moisture regimes in a Pseudotsuga menziesii var. glauca forest.  Blocks show the median density class of Rhizopogon spp. mycelia among the F and H forest floor layers (top layer of plots) and A and B mineral horizons (bottom layer of plots), with white indicating absence (no evidence of Rhizopogon spp.), gray indicating the presence of unidentified or desiccated tuberculate mycorrhizas, and the intensity of blue or red colour saturation increasing according to the density classes: scarce (? 1 tubercle and patchy, sparse hyphae/ rhizomorphs), diffuse (? 10 tubercles and uniformly sparse hyphae/ rhizomorphs), patchily dense (? 10 tubercles or patchily dense hyphal mats), or dense (? 10 tubercles or uniformly dense hyphal mats). Two genets each of R. vesiculosus and R. vinicolor were encountered in plot 6 (arrows point to sampling units where secondary genets occurred).  40  Figure 2.4    Frequency distributions of Rhizopogon vesiculosus and R. vinicolor mycelia density classes (pervasiveness) among organic and mineral horizons of the forest floor and in xeric and mesic soil moisture regimes.  Rhizopogon vesiculosus mycelia were more pervasive (P < 0.05) than R. vinicolor mycelia, and both species were more pervasive in mesic compared to xeric soil moisture regimes.  41Chapter 3:   Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts1  3.1 Introduction Mycorrhizal networks (MNs), or the mycorrhizal fungal mycelia that connect two or more plants, are increasingly recognized as mediators of interactions among trees through their effects on tree survival, growth and competitive ability (Simard & Durall, 2004; Selosse & Duplessis, 2006; Whitfield, 2007).  Mycorrhizal networks provide a source of mycorrhizal fungal inoculum for establishing seedlings (Finlay & Read, 1986b; Nara, 2006), and a potential conduit for interplant transfer of water, carbon and nutrients (Simard et al., 1997a; He et al., 2004; Smith & Read, 2008; Warren et al., 2008).  Three major challenges of MN research are determining the architecture, function and ecological significance of MNs; considerable debate exists on all levels (Whitfield, 2007).  Historically, interest in MNs has focused on their formation and function in controlled artificial systems (Wu et al., 2001) and in natural ecosystems (Simard et al., 1997a; Lerat et al., 2002).  However, little remains known regarding the architecture of MNs in natural settings.  Architecture includes the physical components (e.g., nodes and links) of the network and their genetic complexity, the relationships between the components, and the spatial extent and topology of the components and their relationships.  Describing these attributes is a prerequisite to understanding how MNs function (e.g., in fungal colonization of plants, mycelial growth dynamics, or nutrient uptake/exchange between plants) and how they affect plant populations, communities, and forest dynamics (e.g., in tree regeneration, competition and mortality) (Nara & Hogetsu, 2004; Simard & Durall, 2004; Selosse et al., 2006).   Network theory provides a useful framework for describing the structure, function and ecology of MNs (Bray, 2003; Southworth et al., 2005; Selosse et al., 2006).  The architecture of networks is described using the node degree (i.e., number of links a node forms with other nodes), degree distribution (i.e., the distribution of links among nodes), clustering                                                  1 A version of this chapter has been published in Beiler, K.J., Durall, D.M., Simard, S.W., Maxwell, S.A., Kretzer, A.M. (2010) Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts. New Phytologist 185: 543-553. Tables and figures are reprinted with permission.  42coefficients (i.e., the accessibility of links to nodes), path lengths (i.e., the number of link steps separating nodes) and the contribution of nodes to network connectivity with respect to their topological position or centrality (Glossary) (Albert et al., 2000; Bray, 2003; Southworth et al., 2005).  With MNs, architecture depends largely on the taxonomic, spatial, and temporal scales investigated (Southworth et al., 2005; Selosse et al., 2006).  For example, whether one uses plants, fungi or both as nodes can lead to different structural interpretations (Southworth et al., 2005).  In this study, I describe MN architecture by representing trees as nodes and mycorrhizal fungi as links (phytocentric network) because it is most relevant for understanding MN influences on forest dynamics.  The architecture of biological networks often follows one of three models: regular, random or scale-free (Glossary).  In both regular and random networks, links tend to distribute equally among nodes, but the topology of regular networks is generally more cliquish and harder to traverse than random networks.  In scale-free networks, some nodes (i.e., hubs) are highly linked and more central to the network, resulting in a skewed node degree distribution (Albert et al., 2000; Bray, 2003).  Scale-free networks are both cliquish and easily traversed, and tend to have a ?small world property?, where most nodes can be accessed from every other node by a small number of hops or steps (i.e., a small path length).  They also tend to be more robust to perturbations than regular or random networks.  For example, the random deletion of a node would usually have little effect on the overall connectivity of the network, unless hubs were specifically targeted for removal (Albert et al., 2000; Bray, 2003).  Ectomycorrhizal fungi are diverse in their form, function and ecology (Smith & Read, 2008), and the networking capabilities of individual species or genets are expected to vary accordingly.  For example, the host-specificity of a fungal species and the size, longevity, morphology and continuity of its individual genets are all factors that influence the architecture and function of the resulting MN.  The physical traits of Rhizopogon vesiculosus and R. vinicolor (Basidiomycota, Villosuli-group sensu Kretzer et al., 2003) genets, combined with their known benefits to hosts in accessing water and nutrients, make them ideal candidates for studies of mycorrhizal networking.  They have been consistently reported as dominant members of ectomycorrhizal communities throughout Douglas-fir  43forest development (Chu-Chou & Grace, 1981; Jones et al., 1997; Twieg et al., 2007) and have been shown to increase seedling establishment, growth, and resistance towards root pathogens and drought (Bowen, 1968).  They produce durable tuberculate mycorrhizas and form rhizomorphs estimated to span from meters (re. R. vinicolor) to several decametres (re. R. vesiculosus) in length (Kretzer et al., 2004; Kretzer et al., 2005).  Previous studies have shown that rhizomorph-forming fungi, including species of Rhizopogon, are particularly adept at translocating nutrients and water (Agerer, 2001).  The old-growth interior Douglas-fir forests (Pseudotsuga menziesii var. glauca) where I studied MN architecture undergo gap-phase regeneration, resulting in self-perpetuating, multi-cohort, climax forests (Vyse et al., 2006).  The regenerative capacity of these forests have decreased over the last decade, however, due to increasing temperatures, summer moisture deficits, and disturbances (Griesbauer, 2008; Klenner et al., 2008; Mbogga et al., 2009).  Understanding MN architecture may provide insight into how MNs affect forest regeneration, thus improving predictions of forest dynamics and resilience to disturbance.  I used multi-locus microsatellite DNA markers to discriminate among individuals of interior Douglas-fir and genets of two ectomycorrhizal fungal species, Rhizopogon vesiculosus and R. vinicolor.  I considered the presence of a single Rhizopogon genet on roots of two different trees as a network link, and then used this approach to characterize the architecture of a MN.  My aims were to: determine if R. vesiculosus and R. vinicolor colonize multiple trees or tree cohorts; describe the size and spatial pattern of the Rhizopogon spp. genets; compare the networking qualities of R. vesiculosus and R. vinicolor; and determine the architecture of the MN.  Based on previous work showing that R. vesiculosus forms larger genets than R. vinicolor (Kretzer et al., 2004; Kretzer et al., 2005), I predicted that R. vesiculosus would link more trees than R. vinicolor.  I also predicted that the greater connectivity of large trees compared to small trees would result in a scale-free network where large trees serve as hubs.      443.2 Materials and methods 3.2.1 Plots and sampling This study took place in the multi-storied, multi-cohort interior Douglas-fir (P. menziesii var. glauca (Beissn.) Franco) forest described in chapter 2.  I conducted the study in a 30 x 30 m plot located on a mesic, permesotrophic (zonal) site in the Thompson Dry, Cool Interior Douglas-fir biogeoclimatic variant (IDFdk2; 51?51?7??N latitude, 120?31?46??W longitude) (Lloyd et al., 1990).  Elevation of the study plot was 1,035 m above sea level, with 10-40% slope and southeast aspect.  The composition and percent cover of understory plant species was estimated from 1 m2 blocks (n = 36) sampled in 5 m intervals over a systematic grid covering the 30 x 30 m plot area.  The plant community was dominated by red-stemmed feather moss [Pleurozium schreberi (Brid.), 25% cover] and pinegrass [Calamagrostis rubescens (Buckley), 8% cover], with interspersed shrubs such as common juniper [Juniperus communis (Thunb.), 16% cover], snowberry [Symphoricarpos albus (L.), 11% cover], prickly rose [Rosa acicularis (Lindl.), 5% cover], soopolallie [Shepherdia Canadensis (Nutt.), 5% cover] and Saskatoon berry [Amelanchier alnifolia (Nutt.), 4% cover].  The rest of the forest floor was covered with litter (34% cover), rock (10% cover) and coarse woody debris (9% cover).  Soil properties and tree rooting depth were described in two soil pits excavated to approximately 60 cm depth.  The forest floor was classified as a Hemimor (1-5 cm depth) (Green et al., 1993).  Tree rooting depth was 35 cm, with most fine roots in the upper 15 cm.  Mineral soils were Orthic Dystric Brunisols (Canadian System of Soil Classification, 1998) and surface horizons were of sandy loam texture.    The plot contained 67 live trees (722 stems ha-1) with a basal area of 22.3 m2 ha-1 (Appendix B).  The height and diameter at breast height (dbh; measured at 1.3 m height), and spatial location of every tree in the plot were recorded.  Tree heights were measured with a Laser Vertex hypsometer (Haglof Inc., Madison MS, USA).  A subsample of trees in and around the plot was cored with an increment borer to establish age-size relationships.  Trees were grouped into four cohorts (age groups) based on correlations between age and stem diameter and the diameter distribution of the whole stand.  Tree locations, adjusted for tree diameter and slope position, were mapped using an Impulse laser and digital compass (Laser Technology Inc., Centennial CO, USA) from a central geo-referenced (Garmin Ltd., Olathe,  45KS, USA) location.  The relative height, diameter, and distribution of trees in the plot were illustrated using Stand Visualization System (SVS, V3.36; USDA Forest Service, PNWRS, WA) (Appendix B).  The extent of tree root systems (maximum root length) was estimated based on the distance between the center of a tree (distance minus half the tree?s dbh) and the location where Rhizopogon spp. tubercles were encountered on its roots.  I sampled tuberculate Rhizopogon-like (Basidiomycota; R. vesiculosus & R. vinicolor sensu Kretzer et al., 2003) mycorrhizas (when encountered; n = 401) from four sides of every tree in the plot, within the dripline or between trees where canopy cover was sparse.  I used a dispersed, non-random sampling approach to maximize the probability of sampling roots from every tree in the plot (see Hewitt et al., 2007).  The location of each sample, derived from its distance and azimuth to two or more neighboring trees, was plotted in ArcMap (ArcGIS V9.1) as shown in Figure 3.1 (for clarity, each Rhizopogon species and interior Douglas-fir root spans are illustrated as separate GIS layers in Appendix B).  The depth (cm) of each tubercle sample, measured from the surface of the forest floor, was recorded and used to compare the mean depths of occurrence of R. vesiculosus and R. vinicolor genets.  Fresh needle or cambium tissue was collected from the 67 trees within the plot and from an additional 64 border trees to provide reference DNA for identifying tree roots.  Reference DNA was collected from border trees if their height was greater than their distance from the plot boundary, ensuring that all trees with roots potentially inside the plot were identified.  After collection, mycorrhiza and tree tissue samples were placed in a refrigerated cooler for transport and stored at 4?C for no more than two weeks before further processing.  Tuberculate mycorrhizas were washed in deionized H2O and their identity confirmed as Rhizopogon spp. based on morphological characters under a stereomicroscope at 20x magnification.  Tree cambial tissue was placed immediately on dry ice and extracted within 24 hours of collection.  3.2.2 Molecular analysis A fragment of mycorrhizal root tip (approximately 0.1 g of tree and fungal material) was sampled randomly from the interior of Rhizopogon spp. tubercles and subject to molecular processing (including DNA extraction, multiplex-PCR amplification of microsatellite loci  46and fragment analysis) as described in chapter 2 methods.  I used primers developed by Kretzer et al. (2004) to genotype R. vesiculosus at the microsatellite loci Rv02, Rv15, Rv46, Rve1.21, Rv1.34, Rve2.10, Rve2.44, Rve2.77 and Rve3.21; and R. vinicolor at loci Rv02, Rv15, Rv46, Rv53, Rv1.34, Rve2.10, Rve2.14, Rve2.77, and Rve3.21.  Of these, Rve2.10 and Rve2.14 were used to differentiate between the two species.  A DNA sequencer (3130XL genetic analyzer, Applied Biosystems, Foster City, USA) and Gene Mapper software (V4.0, Applied Biosystems) were used to analyze microsatellite loci.  Interior Douglas-fir tree DNA isolated from Rhizopogon tubercles were genotyped at microsatellite loci PmOSU_1C3, PmOSU_1F9 and PmOSU_2D4 using primers developed by Slavov et al. (2004).  Two or more samples were considered to represent an individual tree or fungus if they had matching alleles at all microsatellite loci analyzed.  Every tree, including border trees that were not encountered as roots in the plot, had a unique genotype based on the targeted microsatellite loci.  More than 30 tree genotypes were compared between needle, bud, root and cambium tissue sources to confirm consistency among tissue types.  For Rhizopogon, additional tubercle samples were collected in the vicinity of the study site to determine allele frequencies and to test the power of genet delineations within the site (these samples represented an additional 37 R. vesiculosus and 33 R. vinicolor genets) (Table 3.1).  The estimated probability that two unrelated individuals could have identical multilocus genotypes by chance (probability of identity) was 1.4 x 10-6 for interior Douglas-fir trees (n = 131), 2.7 x 10-5 for R. vesiculosus (n = 51) and 6.4 x 10-6 for R. vinicolor (n = 46).  All population genetic statistics were calculated using GenAlEx software version 6.1 available at http://www.anu.edu.au/BoZo/GenAlEx/.       3.2.3 Data analysis and statistics All statistical tests were performed using JMP Software (V6, SAS Institute Inc., Cary, USA) unless otherwise noted, with ? = 0.05.  Descriptive statistics are reported as a range with mean and standard deviation (SD).  The number of trees colonized by individual genets was compared between R. vesiculosus and R. vinicolor using a Wilcoxon rank-sum test.  The maximum width and geometric area of Rhizopogon genets were calculated using ArcMap (ArcGIS V9.1) and compared between Rhizopogon spp. using two-sample t-tests.  The mean depth of occurrence of genets was compared between Rhizopogon spp. using one-way  47ANOVA with subsampling in the R statistical computing environment available at http://www.r-project.org.  Associations between tree cohort class, height, diameter, fungal genet frequency and tree node degree were tested using Spearman?s rank correlation tests.  The mean node degree and centrality of trees were compared between cohorts using ANOVA.     Mycorrhizal network architecture was modeled using Pajek version 2.00 (Batagelj & Mvar, Ljubljana, Slovenia) as per de Nooy et al., (2005).  Network measures were simplified by the exclusion of loops (a tree linked to itself through a single fungal genet) and multiple links between nodes (two or more trees linked through more than one genet) unless otherwise noted.  The network model was classified as regular, random or scale-free based primarily on the degree distribution, centrality and clustering coefficients of nodes (Bray, 2003; Southworth et al., 2005).  The spatial orientation of tree nodes was integrated with graph-theoretical measures for network analysis and was used to produce a spatially-explicit model of network topology.  To assess the attack tolerance of the empirical MN, contrasts were made to hypothetical models following a series of node and link removals.  Removal trials included the most highly linked tree, the most linking fungal genet, all trees from the oldest cohort, and a random selection of trees equal to the number of trees in the oldest cohort.    3.3 Results 3.3.1 Characteristics of trees and fungi involved in the network A total of 56 interior Douglas-fir genotypes encountered as roots forming mycorrhizas with Rhizopogon spp. were matched to reference tree boles based on microsatellite DNA analysis (Table 3.2).  They included 45 of the 67 trees inside the 30 x 30 m plot, with an additional 11 genotypes matching tree boles outside the plot (Fig. 3.1).  I found that up to 19 trees, and trees from all age classes, were linked together by a single Rhizopogon genet.  All 44 of the colonized trees from the youngest two cohorts shared one or more Rhizopogon genet(s) with trees from older cohorts (Appendix B).  In addition, young trees from which Rhizopogon spp. were not sampled were primarily distributed within the area occupied by established fungal genets (Fig. 3.1).  I found no genotype-level associations between specific Rhizopogon spp. genets and tree cohort classes or tree genotypes.    48Of the 401 Rhizopogon spp. tubercles collected, 338 were matched to both a tree and fungal genet (Appendix B).  I encountered a total of 14 R. vesiculosus genets and 13 R. vinicolor genets (Fig. 3.1).  Nine genets from each species were found on roots from more than one tree and were thus considered a link.  Among linking genets, Rhizopogon vesiculosus genets were linked with a greater number of host trees (2-19 trees, mean = 10.2 ? 6.6; n = 9) than R. vinicolor genets (3-10 trees, mean = 4.4 ? 2.2; n = 9) (P = 0.03).  Likewise, R. vesiculosus genets had a larger span (0-20.9 m, mean = 13.9 ? 5.4 m) than R. vinicolor genets (0-12.1 m, mean = 5.4 ? 3.7 m) (P < 0.01) and covered a larger geometric area (R. vesiculosus: 0-135.3 m2, mean = 35.9 ? 42.8 m2; versus R. vinicolor: 0-10.0 m2, mean = 3.4 ? 3.7 m2; P < 0.01) (Fig. 3.1).  I found that R. vesiculosus genets also occurred at greater mean depths (1-34 cm, overall mean = 10.8 ? 1.1 cm; n = 14) than R. vinicolor (1-18 cm, overall mean = 8.7 ? 2.0 cm; n = 13) (F1,259 = 6.52, P < 0.01).  3.3.2 Mycorrhizal network architecture Fifty-five of the 56 tree genotypes identified from Rhizopogon spp. tubercles were linked to one or more trees in the plot (Appendix B).  The maximum distance between any two trees in the network (43.2 m) was traversed through only two fungal links, and the maximum path length between any two trees in the network was three fungal links, regardless of the trees? spatial locations or size (representing a small-world property, Table 3.3; Fig. 3.2).  The degree to which a tree was linked with other trees (i.e., node degree) was positively correlated with its cohort class (r = 0.57, P < 0.01), height (r = 0.58, P < 0.01), diameter (r = 0.60, P < 0.01) or maximum root length (r = 0.78, P < 0.01).  Additionally, there was a strong positive association between the node degree of a tree and the number of Rhizopogon spp. genets colonizing it (r = 0.91, P < 0.01).  Though trees from all cohorts were linked, large mature trees acted as hubs with a higher degree of connectivity (ANOVA comparing node degree among cohorts: F3,77 = 9.94, P < 0.01) and more central position in the MN (ANOVA comparing node centrality among cohorts: F3,77 = 6.70, P < 0.01) (Table 3.2).  This resulted in a skewed distribution of node degrees, characteristic of a scale-free model, with the connectivity of large trees well above average (Fig. 3.3).  The tree with the highest node degree (k = 47) and centrality (0.07) in the MN was a mature tree (94 years-old) located 4.2 m outside the plot boundary (marked with an arrow in Figs. 3.1 & 3.2; see also tree no. 23 in  49Appendix B).  This tree was colonized by eight R. vesiculosus genets and three R. vinicolor genets in the plot.  The degree of node connectivity in the empirical model was robust to hypothetical removals of the most highly connected tree, the most linking fungal genet, and 20% of networking trees selected at random.  In contrast, removing trees from the oldest cohort (also 20% of networking trees) substantially reduced MN connectivity (Figs. 3.4 & 3.5).     3.4 Discussion 3.4.1 Mycorrhizal networks link multiple tree cohorts I uncovered an extensive network that linked trees of all ages in an uneven-aged old-growth forest, where 62% of interior Douglas-fir trees from the two youngest cohorts were established within the network of veteran trees.  The MN was comprised of R. vesiculosus and R. vinicolor, each with unique horizontal and vertical spatial patterning in the soil.  The finding that multiple tree cohorts rather than a single age class were included in the MN implies that Rhizopogon fungi have a diverse energy source that is secure over space and time.  While linked trees collectively supply C to the fungus, young trees in turn gain access to an established fungal inoculum source (Finlay & Read, 1986b; Nara, 2006).  This study demonstrates that the mycorrhizal symbiosis is not just between two or more organisms, but is a complex assemblage of fungal and plant individuals that spans multiple generations (Selosse et al., 2006).    3.4.2 Fungal genet parameters The maximum span of genets in this study was 20.1 m for R. vesiculosus and 12.1 m for R. vinicolor, supporting the findings of Kretzer et al. (2004; 2005) that R. vesiculosus has larger genets (? 13.4 m) than R. vinicolor (2.0 m maximum) in 40- to 80-year-old second-growth coastal Douglas-fir forests in Oregon.  The span of R. vesiculosus genets was also in the range reported for other rhizomorph-forming species such as Suillus bovinus  (17.5 m; Dahlberg & Stenlid, 1994) and S. variegatus (27 m; Dahlberg, 1997) in a >100 year old Pinus sylvestris forest.  Studies of ectomycorrhizal fungi have reported genet sizes ranging from ? 1 m span (Gherbi et al., 1999) to ? 40 m (Bonello et al., 1998), and often attribute these size differences to the life history traits of the fungi.  For example, species forming  50small genets are thought to be short-lived, early-successional species dispersed primarily through sexual reproduction (e.g., spores), while large genet sizes are considered to represent late-stage, perennial species spreading primarily through vegetative growth (Dahlberg & Stenlid, 1994; Bergemann & Miller, 2002).  However, most of these studies were based on the occurrence of sporocarps, which may misrepresent the size and distribution of fungal mycelia belowground (Gardes & Bruns, 1996; Zhou et al., 2001).  Moreover, associations between species presence, genet size, and stand successional stage are not always consistent (Guidot et al., 2001; Redecker et al., 2001; Dunham et al., 2003; Twieg et al., 2007).  Thus, further research is needed to determine whether the size differences I observed between R. vesiculosus and R. vinicolor are due to differences in genet longevity, growth rate, or mode of dispersal.    The number of host trees linked by a single genet was greater for R. vesiculosus (mean = 10.2) than R. vinicolor (mean = 4.4), which is consistent with their size differences.  The potential for linkages between interior Douglas-fir trees and R. vinicolor genets in this study was similar to that reported by Lian et al. (2006) for Tricholoma matsutake in an 85-year-old Pinus densiflora forest (4.5 pine trees per genet).  However, the average number of host trees linked through associations with R. vesiculosus genets was more than twice that of T. matsutake or R. vinicolor.  Moreover, Rhizopogon genets linked recent natural regeneration with old trees, with important implications for old-growth forest dynamics.  By contrast, T. matsutake linked similar-aged trees in an even-aged forest.  I found that R. vesiculosus tubercles occurred at greater mean depths than R. vinicolor, showing evidence of vertical niche differentiation, with implications for fundamental processes such as nutrient uptake (Dickie et al., 2002b; Neville et al., 2002; Tedersoo et al., 2003).  For example, a greater depth range may provide R. vesiculosus with greater access to water and nutrients deeper in the soil during dry periods (Baier et al., 2006; Genney et al., 2006), thereby promoting the water stress tolerance of its hosts.     I represented Rhizopogon genets as continuous links across space, though it is possible genets may be fractioned due to clonal propagation or disruption over time.  The continuity of Rhizopogon genets was supported by a concurrent study using fine-scaled sampling of 20  51cm3 soil blocks in contiguous transects, where I found the mycelia of both species to colonize and form continuous connections between multiple trees within a 2 m span (see Chapter 2 this dissertation).  Previous studies have shown that strand-forming fungi similar to Rhizopogon provide continuous transport pathways across ecologically significant distances (Brownlee et al., 1983; Finlay & Read, 1986b) and that severed linkages in the hyphal network are able to re-connect via anastamosis (Bebber et al., 2007; Fricker et al., 2008; Rotheray et al., 2008).  There was also a high degree of linkage redundancy between trees in my study.  For example, most trees were colonized more than once by individual Rhizopogon genets, and multiple links between trees (the same trees linked through two or more genets) represented 22% of all linkages in the network (Tables 3.2 & 3.3, Appendix B).  3.4.3 Network parameters I found that the node degree of a tree was positively correlated with its size and cohort class.  This is not surprising given that the probability of a tree encountering fungal genets increases with the extent of its root system, but to my knowledge, this has never been empirically demonstrated.  Accordingly, the association between tree size/cohort class and connectivity resulted in a scale-free network as I had predicted.  This highlights the importance of large mature trees in the architecture of the MN, where they accounted for most of the connectivity and centrality among nodes in the network.  In addition, the network was easily traversed both spatially and topologically (i.e., path lengths were small relative to the number of nodes), giving the network a small-world property.  These results are in contrast to those of Southworth et al. (2005), who described a random network in Garry oak forests of California.  The difference in results may be due to different approaches in defining a link.  I defined a link as a single fungal genet found on two different trees, whereas Southworth et al. defined a link as the presence of a morphotype (near species-level identification based on the morphology of a mycorrhiza) on a tree, such that the node degree of a tree was equal to its morphotype richness.  Stand structure also differed between these studies: the interior Douglas-fir forest in my study was dense and had a skewed diameter distribution, with a few large, old hub trees and many young trees, whereas Southworth et al.?s savanna forest had fewer, larger, widely spaced trees that were interspersed with Ceanothus shrubland.     52In this study, the tree with the highest node degree was directly linked to 47 other trees through its association with eight R. vesiculosus genets and three R. vinicolor genets in the plot.  This corresponded to 84% of potential linkages between this tree and all other trees encountered in the plot, and was three times higher than the mean node degree among trees.  The influence of this tree as a network component, despite only a portion of its roots being sampled (the bole of the tree was located 4.2 m outside the plot), suggests it would be an even stronger hub at a larger spatial scale.  Overall, MNs are likely even more extensive and complex than what I describe here because I examined only two of up to 65 ectomycorrhizal fungal species previously described in interior Douglas-fir forests of BC (Twieg et al., 2007).  3.4.4 Implications for network functioning and forest management The scale-free architecture and small-world property of the MN suggests that resources could be efficiently shuttled to expanding mycelial fronts, including those associated with regenerating seedlings (Bray, 2003).  For example, carbon and water resources have been shown to move from multiple plants into a common mycelium (Leake, 2001; Wu et al., 2001; Querejeta et al., 2003).  When a young seedling connects into this common mycelium, as shown in this study, it has direct access to these resources, potentially facilitating its establishment.  This premise is supported by field studies showing greater establishment of germinants when linked into the MN of larger trees (Nara, 2006; McGuire, 2007), and by those showing belowground resource transfer among plants facilitated by mycorrhizal fungi (Simard et al., 1997; Lerat et al., 2002; Teste et al., 2009).  Even if resource transfer does not occur between trees, seedlings would benefit from colonization by Rhizopogon genets already being supplied with C by other trees in the network.  In turn, Rhizopogon genets would benefit from colonizing a range of hosts because their large carbon demands are unlikely to be met by young understory trees alone.  This is implied by the results of H?gberg et al. (1999), where ectomycorrhizal fungi specific to pine hosts acquired 87-100% of their carbon from overstory trees and relatively little from understorey trees.             The architecture of the network also suggests it is a relatively robust system, where it would be resilient against random perturbations (Albert et al., 2000; Bray, 2003) but susceptible to targeted removals of hub trees (e.g., by high-grade harvesting, insects or disease).   53Theoretical trials where 20% of trees were removed from the empirical MN suggested the capacity for resource transfer and forest regeneration facilitation through MNs would be substantially more reduced by the targeted removal of mature trees compared to the same number of trees removed at random.  Because large trees sustained a greater number of fungal genets than small trees, the removal of hub trees may also have a large effect on the genetic diversity of R. vesiculosus and R. vinicolor populations.  In summary, I found that most trees in a multi-cohort old growth forest were linked in a scale-free MN, with large trees serving as hubs with implications for understory regeneration and functional continuity in the stand.  These findings support a management approach that conserves large trees or groups of trees and their mycorrhizal fungal associates to ensure that old-growth interior Douglas-fir forests remain resilient and self-regenerative following disturbances.  Additional experimental studies, where MN structure and functioning is assessed before and after selective harvest applications, are needed to test these inferences and to determine the thresholds of MN resilience to disturbance.                    54Table 3.1     Properties of microsatellite DNA loci used to discriminate among individual Pseudotsuga menziesii var. glauca trees and Rhizopogon spp. fungal genets.  Values corresponding to R. vinicolor are given in parentheses following those corresponding to R. vesiculosus; NA = not applicable. Target spp. Locusa Sample numberb Number of alleles HO HE FIS P. menziesii var. glauca PmOSU_1C3 128 31 0.766 0.921 0.169 PmOSU_1F9 131 51 0.715 0.930 0.230 PmOSU_2D4 124 23 0.363 0.876 0.586         R. vesiculosus       (& R. vinicolor) Rve1.21 49 (NA) 7 (NA) 0.449 (NA) 0.639 (NA) 0.297 (NA) Rve2.10 51 (NA) 2 (NA) 0.588 (NA) 0.484 (NA) -0.214 (NA) Rve2.44 49 (NA) 3 (NA) 0.408 (NA) 0.329 (NA) -0.241 (NA) Rv02 8 (12) 2 (4) 0.875 (0.333) 0.492 (0.587) -0.778 (0.432) Rv15 49 (44) 7 (9) 0.551 (0.727) 0.593 (0.768) 0.071 (0.53) Rv46 48 (43) 5 (6) 0.167 (0.698) 0.230 (0.635) 0.276 (-0.098) Rve1.34 50 (47) 4 (3) 0.420 (0.298) 0.363 (0.349) -0.158 (0.148) Rve2.77 50 (47) 4 (6) 0.480 (0.404) 0.459 (0.528) -0.046 (0.234) Rve3.21 50 (45) 9 (9) 0.600 (0.556) 0.732 (0.612) 0.180 (0.092) Rv53 NA (45) NA (6) NA (0.667) NA (0.733) NA (0.090) Rve2.14 NA (46) NA (5) NA (0.174) NA (0.254) NA (0.315) aPCR primers were developed for P. menziesii by Slavov et al. (2005) and Rhizopogon spp. by Kretzer et al. (2003) (see main text).  bInterior Douglas-fir and Rhizopogon spp. samples each represent a unique genotype and were collected in and around the 30 x 30 m study plot within the dry, cool Interior Douglas-fir biogeoclimatic subzone (IDFdk).       55Table 3.2    Size and networking characteristics of tree nodes by cohort class in a 30 x 30 m plot of interior Douglas-fir forest; mean values are reported ? one SD.  cohort 1 cohort 2 cohort 3 cohort 4 Estimated age (years):  ?15 16-50 51-85 ?86 Number of tree boles in 30 x 30 m plot: 6 36 21 4 Number of trees with roots colonized by Rhizopogon in the plot: 1 25 19 11 Height (m):      Mean height: 1.2-1.9 1.6 (?0.3) 1.2-22.8 9.0 (?5.3) 9.6-26.4 20.4 (?4.4) 20.6-31.1 27.4 (?2.7) Diameter at breast height (cm):      Mean dbh: 0.0-1.5 0.7 (?0.68) 0.7-18.6 8.1 (?5.2) 12.9-35.6 24.5 (?6.2) 39.3-56.8 46.4 (?5.3) Maximum observed root lengtha (m):       Mean of maximum lengths: 0.9-0.9 0.9 (?0.0) 0.3-15.6 4.5 (?4.3) 1.4-17.6 7.6 (?5.1) 5.1-22.9 15.1 (?5.9) R. vesiculosus genets per tree:      Mean R. vesiculosus genets per tree: 0-1 0.2 0-3 0.6 0-7 1.9 0-8 2.5 R. vinicolor genets per tree:       Mean R. vinicolor genets per tree: 0-1 0.2 0-2 0.4 0-3 0.7 0-3 1.0 Node degree range:      Mean node degree (ANOVA: P < 0.01): 0-4 0.7 (?1.6) 0-30 5.7 (?7.8) 0-38 16.3 (?13.0) 4-47 19.8 (?13.8) Mean node centrality (ANOVA: P < 0.01): <0.01 <0.01 0.01 0.02 aRoot lengths were estimated based on the occurrence of Rhizopogon spp. mycorrhizas; cohort 1 root data is based on 1 sample only.  56Table 3.3    Attributes of Pseudotsuga menziesii var. glauca trees linked through Rhizopogon spp. genets in a 30 x 30 m plot and characteristics of the resulting mycorrhizal network. number of standing trees in plot: 67 number of standing trees linked: 44 number of tree genotypes from EM roots in plot: 56 mean number of R. vesiculosus genets per tree (?1 SD): 1.23 (?1.55) mean number of R. vinicolor genets per tree (?1 SD): 0.46 (?0.75) node degree range: 0-47 mean node degree (?1 SD): 13.7 (?12.9) total number of links (excluding multiple links): 536 total number of links (including multiple links): 684 network density (excluding multiple links): 0.18 network diameter (measured in link steps): 3 mean path length between linked tree pairs: 1.69 network centralization: 0.44 network clustering coefficient (?1 SD): 0.59 (?0.41)   57 Figure 3.1    The top-down spatially implicit topology of Rhizopogon spp. genets and interior Douglas-fir trees in a 30 x 30 m plot. The plot (square outline) contains 67 trees of mixed age (green shapes, sized relative to each tree?s diameter). Black dots mark Rhizopogon ectomycorrhiza sample locations (n = 401), 338 of which were associated with a specific tree and fungal genet based on microsatellite DNA analysis. Samples representative of each fungal genet are outlined in differing colours. Rhizopogon vesiculosus genets (n = 14) are shaded with a blue background, and R. vinicolor genets (n = 13) with pink. Lines illustrate the linkages between tree roots encountered in Rhizopogon ectomycorrhizas and corresponding source trees aboveground (?root spans?) and are coloured according to tree genotype. An arrow points to the most highly connected tree, linked to 47 other trees through eight R. vesiculosus genets and three R. vinicolor genets inside the plot. Some trees, mycorrhiza samples, and/or genets may be obscured by overlapping features.  58 Figure 3.2    Spatially implicit network model showing interior Douglas-fir trees linked via shared colonization by Rhizopogon spp. genets. Circles represent tree nodes, sized according to tree diameter and coloured with four shades of yellow or green that increase in darkness with increasing age class. Lines represent the Euclidean distances between trees that are linked.  Line width increases with the number of links between tree pairs (i.e., repeated links through multiple fungal genets). An arrow points to the most highly connected tree, linked to 47 other trees through eight R. vesiculosus genets and three R. vinicolor genets inside the plot. Some tree nodes and their links may be obscured by overlapping features.  59    Figure 3.3    The degree distribution of tree nodes linked through a mycorrhizal network, showing the number of trees linked to x number of other trees with roots colonized by Rhizopogon spp. in a 30 x 30 m plot of uneven-aged interior Douglas-fir forest.  Two or more trees were considered linked if they were colonized by the same fungal genet.  Node degrees were higher on average among older trees (cohorts 3-4) than younger trees (cohorts 1-2), leading to a skewed degree distribution characteristic of a scale-free network model.         60 Figure 3.4    Mycorrhizal network topology showing the loss of connectivity between interior Douglas-fir trees linked through Rhizopogon spp. genets with the hypothetical removal of (a) the most highly linked hub tree (k = 47), (b) the most linking fungal genet (colonized 19 trees), (c) 20% of networking trees selected randomly, or (d) trees from the oldest cohort (also 20% of networking trees). Note the rotation by 90? from empirical model orientation.  61 Figure 3.5    Histograms showing the degree distributions of interior Douglas-fir tree nodes linked through Rhizopogon spp. genets following the hypothetical removal of (a) the most highly linked hub tree (k = 47), (b) the most linking fungal genet (linked 19 trees), (c) 20% of networking trees selected randomly, or (d) trees from the oldest cohort (also 20% of networking trees). 62Chapter 4:   The self-organization of ectomycorrhizal networks in xeric and mesic old-growth interior Douglas-fir forest  4.1 Introduction Network analysis is a powerful tool for describing and analyzing the dynamics of complex biological systems, where individuals, species, or species guilds are viewed as nodes linked through their ecological associations (Bascompte, 2009a).  This provides a template for holistically describing the myriad of positive and negative interactions between organisms, or between organisms and their environment that generate the system?s patterns and processes (Levin, 1992; Levin, 2005; Lau et al., 2010).  Examples of network attributes include the diameter (longest of shortest paths), density of links, and the distribution of links among nodes in the network (degree probability distribution, network centralization, and clustering coefficient), which indicate the efficiency at which materials or information are able to traverse the network (Watts & Strogatz, 1998; Barab?si & Albert, 1999; Barab?si, 2009).  Moreover, network analysis provides metrics for identifying key participants, components (e.g., species, functional guilds), or processes that act as binding agents or are direct determinants of the system?s current structural state and modularity (Bascompte, 2009a; Parrott, 2010).  For example, measures of connectivity (node degree, link strength), topological position (node betweenness centrality) and cliquishness (node clustering coefficients) can be used to identify foundational or keystone species, while the overall network centralization, clustering coefficient and degree distributions indicate the redundancy of individual participants or sub-network components (Glossary; de Nooy et al., 2005; Ellison et al., 2005).  Thus, network analysis can be used to predict a system?s robustness and resilience to allogenic or autogenic change, and specific vulnerabilities associated with the loss of critical players or components (Holling, 1973; Albert et al., 2000; Dunne et al., 2002; Montoya et al., 2006).    Much of the complexity of natural systems is due to the nested hierarchical structure of their internal interactions and processes, which span multiple genetic, trophic, spatial, and temporal scales (Levin, 1992; van de Koppel et al., 2006; Whitham et al., 2006; Lau et al., 2010).  This creates an obstacle for developing autonomous conservation goals or  63management criteria for populations, species, or ecosystems; especially when biological or social boundaries are crossed (M?ller, 2005; Crona & Hubacek, 2010).  Using network analysis, nodes and links may be defined according to the scale of interest.  Nodes, components, or network configurations may be incorporated into larger meta-networks at coarser scales without a substantial loss of information regarding underlying variability (de Nooy et al., 2005; Urban, 2005; Simard, 2009).  While there has been increasing attention paid to interactive networks in ecology, most work to date has focused on community trophic dynamics using predator-prey, plant-pathogen, or food-web models (Ings et al., 2009; Parrott, 2010).  In contrast, little attention has been paid to mutualistic network structure at the population level, where nodes represent individuals and links represent intraspecific interactions or shared attributes (Jordano, 1987; Ohtsuki et al., 2006; Beiler et al., 2010).  This focal scale can be used to determine how the networking attributes of individuals change over time, and how phenotypic and phenological heterogeneities within and among  populations contribute to network structure and dynamics at coarser scales (Hewitt et al., 2007; Bascompte, 2009a; Ings et al., 2009; Peters et al., 2009).  Interior Douglas-fir (Pseudotsuga menziesii var. glauca (Beissn.) Franco)  predominates in the dry, cool climate regions of western North America, where it forms single-species, multi-cohort stands (in which age class distributions are multimodal) through gap-phase regeneration following natural mortality or disturbance (Oliver & Larson, 1996; Huggard et al., 2005; LeMay et al., 2009).  Such forests are considered complex self-organizing ecosystems where resource limitations determine the current structural state (Simard, 2009).  For example, precipitation levels below the lower threshold cause transition from forests dominated by interior Douglas-fir to grasslands at lower elevations, and exceeding the upper threshold causes transition to mixed-species forests at higher elevations (Vyse et al., 1990; Meidinger & Pojar, 1991; Huggard et al., 2005).  Seasonal water deficits are the predominant limiting factor for interior Douglas-fir regeneration and growth and can make forests more susceptible to stand-replacing wildfires, outbreaks of insects and disease, or diebacks (Lopushinsky, 1990; Klenner et al., 2008; Littell et al., 2008).  Moreover, projections of future climatic conditions in this region suggest an increased frequency and severity of droughts and abiotic and biotic disturbances (Nitschke & Innes, 2008a; Walker &  64Sydneysmith, 2008; Rodenhuis et al., 2009; Griesbauer et al., 2011).  These problems may be confounded by ?high-grade? harvesting (various selection and diameter limit cutting methods implemented circa 1870s-1970s, where the straightest, largest trees were removed for their economic value) over much of the range of the dry interior Douglas-fir forests (Vyse et al., 1990) by potentially reducing the vigor and adaptive capacity of forests over time through dysgenic selection.  Identifying structures, species, interaction components, or processes that are critical determinants of forest ecosystem structure helps to determine the vulnerabilities and thresholds of the system?s current state and better predict how these forests may respond to climate change (Scheffer et al., 2001; Suding et al., 2004; Ellison et al., 2005; Mueller et al., 2005; Miao et al., 2009).     Ectomycorrhizal fungi (EMF) increase the niche breadth of trees by facilitating water and nutrient uptake and protecting roots from pathogens and extreme temperatures (Amaranthus, 1992; Bruns et al., 2002; Allen et al., 2003).  When multiple tree partners are colonized by a continuous EMF mycelium, an interactive mycorrhizal network (MN) is formed through which energy, water, nutrients, communication signals, or other materials may be transferred (Selosse et al., 2006; Smith & Read, 2008; Song et al., 2010).  With or without the direct transfer of materials between plants, seedlings can gain access to water and nutrients through EMF mycelia that are predominantly subsidized by mature trees (Horton et al., 2005).  As such, MNs influence the spatial and genetic structure of forest communities by altering intra- and interspecific competition between trees and/or by altering local environments towards conditions more or less suitable for tree growth (Amaranthus & Perry, 1994; Dickie et al., 2002a; Booth, 2004; Nara & Hogetsu, 2004; Nara, 2006).  EMF also play important roles in litter decomposition and nutrient cycling (Gadgil & Gadgil, 1971; Entry et al., 1991; Read & Perez-Moreno, 2003; Talbot et al., 2008), and produce sporocarps (e.g., mushrooms, truffles) that are important as non-timber forest products and components of forest food webs (Maser & Maser, 1988; Ashkannejhad & Horton, 2006; Leveau & Preston, 2008).  Thus, EMF not only increase the resilience of individual plants, but provide critical ecosystem services and stabilize forest systems through direct and indirect feedback loops associated with their multifarious, multi-trophic interactions (Molina et al., 1992; Perry, 1995; Simard, 2009; van der Heijden & Horton, 2009).    65Using molecular identification methods, recent field-based studies have begun to address the knowledge gap concerning the architecture of MNs among individual trees and fungal genets (Lian et al., 2006; Nara, 2006; Beiler et al., 2010).  However, it remains unknown how the structural properties of MNs may vary across local environmental gradients (e.g., water and nutrient availability or pH), or in response to disturbances or climate change at coarse spatiotemporal scales; factors which are known to influence the species composition, phenology and productivity of both tree and EMF communities (Rillig et al., 2002; Aitken et al., 2008; Peters et al., 2009).  MNs can be viewed from different perspectives, including the phytocentric view where trees are nodes linked through shared fungal genets, or the mycocentric view where fungal genets are nodes linked through shared host trees (Southworth et al., 2005).  MNs provide a good model for investigating age and size effects on the architecture and self-organization of complex systems over time, because these attributes can be parameterized and explicitly incorporated into network topology as spatially static or contagious processes (Peters & Herrick, 2004).  The overarching objective of this study was to compare the complexity of MNs between sites with differing soil moisture regimes to infer the relative importance of MNs in forest stand structure with increasing site water stress.  This information is useful for estimating the resource thresholds of the system?s current structural state and for facilitating predictions regarding potential regime shifts associated with site water stress under current climate change scenarios.  My objectives were fourfold: (1) to describe the universal emergent structural properties of MNs at the forest stand scale; (2) to contrast network architectures between phytocentric and mycocentric perspectives; (3) to identify the relative contributions of individual trees and Rhizopogon spp. genets in determining the architecture of the MNs; and (4) to contrast the architecture of MNs between forest plots with xeric versus mesic soil moisture regimes.  Based on the mixed-aged structure of the forest stands and Rhizopogon spp. life history traits (see Chapters 2 & 3 of this dissertation), I predicted that size asymmetries among trees and fungal genets would lead to complex, scale-free MN topologies.  Similarly, I predicted that node connectivity and link strength would increase in relation to the size of trees and fungal genets participating in the network, such that large mature trees and R. vesiculosus genets influence network connectivity more than small trees  66and R. vinicolor genets.  I expected MN architectures to differ between phytocentric and mycocentric perspectives.  Therefore, I hypothesized that MNs from the mycocentric perspective would have more regular, cliquish network topologies at the taxonomic and spatial scales investigated due to less spatial overlap between conspecific fungal genets compared to tree roots (owing to the richness of EMF species in pure interior Douglas-fir stands and the spatial segregation hypothesis).  Lastly, I expected greater MN connectivity in xeric stands relative to mesic stands in keeping with the stress-gradient hypothesis (e.g., Bertness & Callaway, 1994; Callaway, 1995).  4.2 Materials and methods 4.2.1 Plots and sampling Six independent 10 x 10 m plots (each separated by ? 150 m) were selected within a multi-storied, multi-cohort old-growth interior Douglas-fir forest (Table 4.1; Fig. 4.1).  These plots were located adjacent to and numbered in correspondence with the six study plots described in chapter 2 of this dissertation (see Fig. 2.1).  To compare MN attributes between xeric and mesic soil moisture regimes, three plots were selected in upper slope positions and three plots in lower slope positions with plant communities indicative of the respective soil moisture regimes as per Lloyd et al. (1990).  These ecosystem attributes are represented by contrasting site series classifications in the Thompson Dry, Cool Interior Douglas-fir biogeoclimatic variant of southern-interior BC, Canada (Pojar et al., 1987).  I used mesoslope position and the relative abundance of Calamagrostis rubescens Buckl. and Pleurozium schreberi Brid. as indicators of soil moisture regime, with xeric plots (plots 1-3) having a greater proportion of C. rubescens in the understory, and mesic plots (plots 4-6) having a greater proportion of P. schreberi (Table 4.1).  The percent cover and frequency of occurrence of understory vascular plant and bryophyte species were estimated using a 30 x 30 m grid covering each plot, where 1 m2 quadrat grid frames (n = 36) were systematically sampled in 5 m intervals across the full extent of the grid (Table 4.2).  Understory vegetation was measured beyond the 10 x 10 m extent where Rhizopogon spp. mycorrhizas were later sampled to better characterize the soil moisture regime experienced by trees with roots potentially in the 10 x 10 m  plots.  The mean richness of vascular plant and bryophyte species encountered in each plot did not significantly differ between xeric and mesic soil moisture regimes based on a Wilcoxon rank- 67sum test in SPSS 16.0 Graduate Student Version for Windows (IBM Corporation, Somers, USA) with ? = 0.05 (W = 9, P = 0.05, n1,2 = 3).  Soils were Orthic Dystric Brunisols with a sandy loam texture and Hemimor humus form based on soil pits located adjacent to plots (Canadian System of Soil Classification 1998).    The diameter at breast height (dbh; cm) of trees was measured using a dbh tape at 1.3 m height on the upslope side of trees.  Tree heights (m) were measured with a Laser Vertex hypsometer (Haglof Inc., Madison MS, USA).  All trees were mapped using an Impulse laser and digital compass (Laser Technology Inc., Centennial CO, USA), and projected to UTM coordinates based on their distance to tree center (including ? dbh), azimuth, and % slope from a central geo-referenced location.  I measured tree ages in years based on growth ring data from a subsample of 62 trees (cored with an increment borer), including one sapling from each site, and grouped trees into five cohort classes based on the modality of stand height and diameter distributions (Table 4.1).  The mean number of tree boles (stems m-2) and basal area of trees (m2 ha-1) were contrasted between plots with xeric versus mesic soil moisture regimes using Wilcoxon rank-sum tests in SPSS to obtain baseline information regarding their potential influences on network measures.  No significant differences in the number of boles (W = 8, P = 0.268, n1,2 = 3) or tree basal area (W = 9, P = 0.513, n1,2 = 3) in 10 x 10 m plots were found between soil moisture regimes.  A chi-square goodness-of-fit test was performed in SPSS to determine if the frequency distribution of tree cohort classes (five classes) represented by mycorrhizal roots sampled in 10 x 10 m plots (as later described) were similar between soil moisture regimes.  Although significant (X2 = 31.37, P < 0.00, df = 4), differences were primarily due to xeric plots having a greater proportion of < 15 year old trees while mesic plots had a greater proportion of 15 ? 30 year old trees, whereas the older age classes did not differ between soil moisture regimes.  Stand characteristics, tree cohort frequency distributions and geographic information are shown in Table 4.1 for each forest plot.    In each of the six 10 x 10 m plots, a total of 200 hours (between five crew members) were spent collecting Rhizopogon spp. mycorrhizas between May and June, 2008.  Rhizopogon spp. mycorrhizas were sampled purposively to obtain samples from a range of tree roots  68representative of the mixed-aged stand structure for modeling purposes (Hewitt et al., 2007).  For this, two or more samples of Rhizopogon spp. mycorrhizas were collected (when encountered) from the four cardinal directions from every tree in each plot at the inside edge of the crown drip line.  This was accompanied by dispersed sampling of plot interspaces where there was no canopy cover.  Samples were placed in 2 ml polypropylene tubes and frozen at -20?C for molecular analysis.  To provide a reference DNA library of tree genotypes for matching with tree roots isolated from Rhizopogon spp. mycorrhiza samples in the 10 x 10 m plots, needle and/or cambium tissue was collected from 646 trees, including all standing trees within the 10 x 10 m plots and surrounding trees with a height greater than their distance to the plot boundary (height was used to estimate potential rooting extent).  To distinguish between trees with boles inside the 10 x 10 m plots and trees located outside plot boundaries but with roots identified from Rhizopogon spp. mycorrhizas inside the plots, I hereafter refer to the latter as ?tree genotypes?.  Tree needles were sampled following bud burst but before needle elongation, and frozen at -20?C for molecular analysis.  Tree cambium was sampled using a 9.53 mm punch tool and placed immediately on dry ice, and DNA was extracted within 24 hours of collection.    4.2.2 Molecular analysis A subsample of approximately 0.1 g tree needle or cambium tissue and fine-root tissue from within Rhizopogon spp. tuberculate mycorrhizas were processed in the lab.  The molecular analysis workflow included DNA extraction, amplification of microsatellite loci in multiplex PCR reactions, and fragment analysis using a capillary sequencer (3130XL genetic analyzer, Applied Biosystems, Foster City, USA) and GeneMapper software (V4.0, Applied Biosystems) as detailed in Beiler et al. (2010).  When cambium DNA was not successfully extracted using the Qiagen DNeasy? plant kit (QIAGEN Inc., Mississauga, Canada) as per Beiler et al. (2010), a Sigma Extract-N-AmpTM Plant PCR kit was used (Sigma-Aldrich Inc., Saint Louis, USA).  For this, cambium tissue was incubated for 10 min. at 95? C in 20 uL of extraction solution followed by the addition of 20 uL dilution solution, and was then diluted 10-fold in a 50:50 mix of extraction and dilution solutions.  Microsatellite DNA loci were amplified by multiplex PCR in 15 ?L reaction volumes, comprised of 3 ?L DNA extract solution, 7.5 ?L Extract-N-Amp PCR ReadyMixTM, 2.6 ?L ultrapure H2O, and microsatellite  69primers in concentrations described in Beiler et al. (2010).  Interior Douglas-fir tree DNA isolated from needles, cambium tissue, and from roots within Rhizopogon tubercles was genotyped at microsatellite loci PmOSU_1C3, PmOSU_1F9 and PmOSU_2D4 using primers developed by Slavov et al. (2004).  I used a systematic approach in multi-plex PCR reactions to first identify R. vesiculosus and R. vinicolor to species using co-specific primers, then applied species-specific primer groupings in subsequent multiplex PCR reactions as per Beiler et al. (2010).  Rhizopogon DNA was identified to species using the microsatellite loci Rve2.10 and Rve2.14; R. vesiculosus was then further genotyped at loci Rv02, Rv15, Rv46, Rve1.21, Rv1.34, Rve2.44, Rve2.77 and Rve3.21; and R. vinicolor at loci Rv02, Rv15, Rv46, Rv53, Rv1.34, Rve2.77, and Rve3.21 using primers developed by Kretzer et al. (2004).  Two or more samples were considered to represent the same individual when they had identical multilocus genotypes based on amplification of all microsatellite loci.  The probability of encountering two unrelated individuals (Rhizopogon spp. genets or trees) with identical multilocus genotypes by chance was sufficiently low (< 10-6) and was calculated with the inclusion of additional samples collected in the vicinity of study plots as reported in Beiler et al. (2010).  4.2.3 Statistics and network modeling Statistical tests were performed using SAS version 9.2 (Statistical Analysis Software, SAS Institute, Cary, USA) except where otherwise noted.  Results are reported as a range with mean ? standard deviation (SD).  A Bonferroni adjustment (alpha / the number of pairwise comparisons) was used to control familywise error rates when determining the significance of tests involving multiple comparisons.  Generalized linear mixed models were obtained (later described in this chapter) using SAS Proc GLMMIX, based on pseudo-likelihood estimation with parameter convergence criterion set at 1E-7 and a singularity tolerance of 1E-12.  Prior to making any plot-level MN comparisons, I tested the assumption that sample sizes were reasonably balanced between soil moisture regimes and between MN components (sub-networks) linked through R. vesiculosus versus R. vinicolor genets.  For this, a MANOVA was used to test the multivariate effects of the factors: Rhizopogon species (R. vesiculosus and R. vinicolor), soil moisture regime (xeric and mesic), and Rhizopogon species x soil moisture regime on the co-dependent variables: y1 = the mean number of  70tuberculate mycorrhizas sampled per plot and y2 = the mean number of Rhizopogon spp. genets among plots using SPSS.  Mulitivariate normality of residuals and homogeneity of error covariance matrices were checked using a Shapiro-Wilk test and Box?s M test, respectively.         Exploratory network analysis was conducted using Pajek version 2.04 (Batagelj & Mvar, 2011, Ljubljana, Slovenia) as per de Nooy et al. (2005).  MNs were initially modeled from the phytocentric perspective, with tree boles as spatially explicit nodes that were considered linked when colonized by the same fungal genet (genets represent hyperlinks).  As such, the cumulative tripartite network motifs between tree pairs linked through the same fungal genet are represented by a single link in the model.  When two trees shared more than one fungal genet, links were given a quantitative value equal to the sum of these multiple links.  Phytocentric MN models were also partitioned into separate components linked through either R. vesiculosus or R. vinicolor; these are referred to as Rhizopogon spp. sub-network components.  The complimentary mycocentric MN perspective was modeled with fungal genets as nodes linked through shared host trees.  Here, multiple links represent the number of host trees shared between pairs of R. vesiculosus or R. vinicolor.  I chose to model the phytocentric and mycocentric perspectives separately, rather than treating trees and fungi as nodes in a multi-partite affiliation network, to simplify the network model and to make inferences regarding the potential ecological role of MNs.  Multiple links were included for illustrations, but were excluded from all network property measures.  As such, network link densities are equivalent to the mean of normalized node degrees, or the number of links a node has to other nodes relative to the total number of nodes in each network.  MN architectures were classified based on network diameter, density of links, all-degree centralization, and clustering coefficient of the overall networks; and the degree distribution, betweenness centrality, and clustering coefficients of nodes (Bray, 2003; de Nooy et al., 2005).  I then verified MN classifications by contrasting the empirical networks against undirected random Erdos-Renyi configurations in Pajek using average node degree as input.     71Emergent network structural properties including diameter, link density, mean node degree, centralization and clustering coefficients were contrasted between Rhizopogon sp. sub-network components in the phytocentric model and between the full phytocentric and mycocentric models pairwise by plot using Wilcoxon signed-rank tests in SPSS.  The number of trees colonized by fungal genets was contrasted between R. vesiculosus and R. vinicolor in a Wilcoxon rank-sum test in SPSS.  Spearman?s rank correlation coefficient (rho) was used to indicate the strengths of pairwise associations between tree cohort class and tree height, dbh, node degree, node betweenness centrality, and node clustering coefficients, and between tree node degree and clustering coefficients using SPSS.  Tree dbh data were square-root transformed to achieve linearity.  In the mycocentric network, Spearman?s rho was used to indicate the strengths of pairwise associations between the number of mycorrhiza samples per Rhizopogon genet and genet span (i.e., maximum distance between samples of a genet), number of trees associated with genets, genet node degree, betweenness centrality, and clustering coefficients, and between genet node degree and clustering coefficients in SPSS.  For both tree and Rhizopogon spp. genet nodes, betweenness centrality and clustering coefficient data could not be successfully transformed to achieve linearity.  However, in all cases, relationships were monitonically increasing or decreasing.  Therefore, the Spearman?s rank correlation coefficients indicate the strength of relationships.    To further examine the potential influence of genet size asymmetries on the networking characteristics of Rhizopogon spp. (link strengths in the phytocentric models and degree connectivity in mycocentric models), a generalized linear mixed model with a negative binomial distribution and log link function was obtained for the response variable ?number of trees colonized by each genet?.  The model included the fixed effects ?Rhizopogon spp.?, ?number of mycorrhizas sampled per genet across space?, and ?Rhizopogon spp. x number of samples per genet across space?, and the variable ?plot? was included as a random effect.  The model was rerun excluding the interaction term because it was not significant at ? = 0.05 based on -2 Log Pseudo-Likelihood estimation.  The numbers of mycorrhizas sampled per genet were used as a proxy for genet size, rather than the span of samples, because this metric better represented the 3-dimensional extent of genets owing to the dispersed sampling approach and plot edge effects on genet span.         72The emergent structural properties of phytocentric and mycocentric networks, including network diameter, link density, mean node degree, centralization and clustering coefficients, were contrasted between xeric and mesic soil moisture regimes in separate Wilcoxon rank-sum tests using SPSS.  Generalized linear mixed models were obtained to assess the effects of Rhizopogon species and soil moisture regime on the networking attributes of tree subsamples, including the response variables: tree node degree, normalized node degree, node betweenness centrality, and node clustering coefficient.  A negative binomial distribution with log link function was used to fit the count variable ?tree node degree?, and binomial distributions with logit link functions were used to fit the remaining proportional response variables.  Each model included tree dbh as a fixed-effect covariate to account for tree size differences, species, soil moisture regime, interactions between fixed effects, and the variable ?plot? as a random effect.  An alpha = 0.05 was used to remove variables from the model based on -2 Log Pseudo-Likelihood fit statistics.  For this, the 3-way interaction was checked and dropped if not significant.  Then, the significance of 2-way interactions was checked and if not significant, the interaction with the highest p-value was dropped.  The model was then run again and if all 2-way interactions were dropped for a particular class variable, the significance of the class variable was checked to see if it could be dropped.  The process stopped when remaining variables were significant.         4.3 Results 4.3.1 Sampling results A total of 478 R. vesiculosus and 280 R. vinicolor tuberculate mycorrhizas collected within the six 10 x 10 m plots were successfully genotyped at all fungal and tree microsatellite DNA loci targeted.  These corresponded to 28 R. vesiculosus genets, 27 R. vinicolor genets, and 166 tree genotypes that included all tree cohorts.  Among the 81 tree boles located within 10 x 10 m plot boundaries, 65 were found in association with Rhizopogon spp. mycorrhizas, corresponding to 78-89 % of trees in each plot except plot 2 where 50 % of trees were colonized by Rhizopogon spp. (Table 4.3).  Of the aforementioned 65 trees, 55 % were associated with more than one Rhizopogon spp. genet (Table 4.4), which is an underestimate considering plot edge effects (Appendix C).  Conversely, all but one Rhizopogon spp. genet that was sampled more than once colonized two or more trees (of the 166 tree genotypes  73isolated from Rhizopogon sp. mycorrhizas, 165 were linked to other trees through a Rhizopogon sp. genet).  Thus, trees and fungal genets could each be considered as MN nodes or links for complimentary phytocentric and mycocentric models.  A two-way MANOVA with y1 = the mean numbers of Rhizopogon mycorrhizas sampled per plot and y2 = the mean numbers of distinct genets represented by R. vesiculosus/R. vinicolor mycorrhiza samples in plots revealed no significant effects (multivariate main effects or interactions) for Rhizopogon species (F2,7 = 3.257, P = 0.100), soil moisture regime (F2,7 = 1.243, P = 0.345), or Rhizopogon species x soil moisture regime (F2,7 = 0.590, P = 0.943).  The assumptions of multivariate normal error variances and homogeneity of error covariance matrices were met.  4.3.2 Architecture of phytocentric MNs and sub-network components Figure 4.2 illustrates the complex network topology of spatially referenced tree nodes linked through fungal genets in each plot.  The topological attributes of phytocentric MNs are summarized by plot in Table 4.3.  In each plot, the shape and spread of tree node degree distributions resembled that of multimodal tree diameter distributions; with most networking trees having an average degree of connectivity plus a few trees that were highly connected.  The boles of most large trees were located outside plot boundaries (i.e., cohorts 4 & 5, Table 1), and their connectivity was disproportionately reduced by the plot edge effect relative to younger cohorts, suggesting node degree distributions would have a stronger positive skew at larger spatial extents (Fig. 4.2, Appendix C).  This, together with an inverse relationship between node degree and node clustering coefficients (r = -0.528, P < 0.001) (Fig. 4.3), and minimal network centralization (Table 4.3), suggests these MNs have a hierarchically scale-free topology at the forest stand scale.  This deduction was supported by differences between the structural properties of empirical MN models and random Erdos-Renyi model reconfigurations (data not shown).    When the architecture of R. vesiculosus versus R. vinicolor MN components were compared pair-wise by plot using Wilcoxon signed-rank tests, the mean node degree of trees in plots significantly differed (W = 21, P = 0.028, n1,2 = 6), and the density of links (W = 19, P = 0.046, n1,2 = 6) and network clustering coefficients (W = 19, P = 0.046, n1,2 = 6) were marginally significantly different (Fig. 4.4; Table 4.4).  No significant differences were found  74between Rhizopogon spp. sub-network components regarding mean network diameter (W = 9, P = 0.317, n1,2 = 6), betweenness centralization (W = 15, P = 0.116, n1,2 = 6), or degree centralization (W = 11, P = 0.249, n1,2 = 6).  4.3.3 Architecture of mycocentric MNs Figure 4.5 illustrates MN topologies from the mycocentric perspective (non-spatial model), with fungi as nodes linked through shared host trees.  Because there were fewer genets encountered in 10 x 10 m plots than trees, the mycocentric networks appear less complex than the complimentary phytocentric models.  However, fungal nodes had (weakly) positively skewed degree distributions similar to that of the phyotcentric models (Fig. 4.6).  When contrasted pair-wise by plot using Wilcoxon signed-rank tests, the mean node degree and centralization of node degrees differed significantly between phytocentric and mycocentric perspectives; phytocentric networks generally had higher mean node degrees (W = 21, P = 0.028, n1,2 = 6) and less degree centralization (W = -21, P = 0.028, n1,2 = 6) compared to mycocentric network models.  The density of links and network clustering coefficients were similar between phytocentric and mycocentric networks (Tables 4.2 and 4.5 respectively).  The number of trees colonized by R. vesiculosus and R. vinicolor genets differed significantly based on a Wilcoxon rank-sum test (W = 27, P = 0.054, n1,2 = 6), and R. vesiculosus genets generally colonized more trees than R. vinicolor genets.  However, interspecific network linkages between R. vesiculosus and R. vinicolor genets were equally as probable as intraspecific linkages between genets of either species (illustrated in Fig. 4.5).  4.3.4 Factors contributing to mycorrhizal network architecture Among all tree genotypes in the 10 x 10 m plots (n = 182), significant relationships were found between tree cohort class and (dbh)1/2 (r = 0.945, P < 0.001), height (r = 0.845, P < 0.001), node degree (r = 0.252, P = 0.001) and betweenness centrality (r = 0.281, P < 0.001) based on Spearman?s rho.  The relationship between the cohort class and node clustering coefficient of trees was weak but statistically significant (r = -0.018, P > 0.008).  Spearman?s rank correlations between tree cohort class and (dbh)1/2, height, node degree, node betweenness centrality, and node clustering coefficient, and between tree node degree and clustering coefficient, are shown independently for each plot in Table 4.3.  75The higher degree of connectivity among trees in R. vesiculosus-linked MN components compared to R. vinicolor could be explained by a relationship between the number of trees colonized by genets of each species and the number of mycorrhizas sampled per genet across space.  In the generalized linear mixed model with the response variable ?number of trees colonized by each genet?, the ?number of mycorrhizas sampled per genet across space? was the only significant explanatory variable at the ? = 0.05 level (-2 Log Pseudo-Likelihood model fit = 76.40) (Fig. 4.7).  From the mycocentric MN perspective, this pattern led to a positive association between the relative size of genets (proportion of mycorrhizas sampled from 10 x 10 m) and their degree of connectedness to neighboring genets (Table 4.5, Fig. 4.5).  Spearman?s rank correlations between the numbers of mycorrhizas sampled per genet and the genet span, number of trees associated with genets, genet node degree, betweenness centrality, and clustering coefficient, and between genet node degree and clustering coefficients are shown independently for each plot in Table 4.5.  4.3.5 Network properties in xeric versus mesic soil moisture regimes  No significant differences were found between soil moisture regimes regarding the emergent structural properties of MNs, including mean network diameter, link density, mean node degree, centralization, or clustering coefficients for either phytocentric or mycocentric MNs (all with P > 0.05, n = 3 plots per soil moisture regime, in separate Wilcoxon rank-sum tests; data in Tables 4.2 & 4.5).  Conversely, interactions between soil moisture regime and other fixed effects variables in generalized linear mixed models were commonly significant, indicating that soil moisture affects networking attributes of tree nodes (node degree, normalized node degree, betweenness centrality, or clustering coefficient) (Fig. 4.8).  The response variables ?tree node degree?, ?normalized node degree?, and ?node clustering coefficient? each showed a positive relationship with tree dbh, and their values were higher among trees linked through R. vesiculosus compared to R. vinicolor.  The slope for ?tree node degree? on tree dbh was greater for xeric compared to mesic soils.  In support of this finding, relationships between tree cohort class and node degrees appeared stronger among xeric plots compared to mesic plots based on Spearman?s rho (Table 4.3).  In the model with ?normalized node degree? as the response variable, a significant interaction between Rhizopogon species and soil moisture regime was found.  In this model, normalized node  76degrees were higher in xeric plots compared to mesic plots among R. vesiculosus links, but the opposite trend was found for R. vinicolor links (soil moisture regime changed the y-intercepts of Rhizopogon species).  The rates at which normalized node degrees increased with tree dbh were the same among each combination of Rhizopogon species and soil moisture regime.  In the model with ?tree clustering coefficient? as the response variable, the positive trend with tree dbh was influenced by Rhizopogon species and by soil moisture regime, owing to significant 2-way interactions between Rhizopogon species x dbh and soil moisture regime x dbh.  In general, the relationship between tree node clustering coefficients and dbh was stronger for R. vesiculosus compared to R. vinicolor links and in xeric compared to mesic soils, but the strength of these effects were not independent.  In the model with ?node betweenness centrality? as a response variable, no fixed effect variables were significant in the model at ? = 0.05 level.  4.4 Discussion 4.4.1 Self-organized, complex architecture of Rhizopogon spp. MNs To my knowledge, this is the first study to compare the emergent structural properties of multi-trophic mutualism networks between differing environmental conditions, while accounting for population-level size and spatial pattern heterogeneities among symbionts.  Several important findings emerged demonstrating the pervasiveness and complexity of MNs, and their potential role in the self-organization of mixed-aged interior Douglas-fir forests in xeric and mesic soil moisture regimes.  First, interior Douglas-fir trees and Rhizopogon spp. genets formed spatially and structurally complex MNs in all six sample plots.  The phytocentric and mycocentric models provided complimentary network perspectives, but the architecture and underlying topology of these models were similar and typical of hierarchical scale-free network structures.  Second, ?body-size? asymmetries between different tree cohorts and between Rhizopogon spp. genets were largely responsible for the complex topology and nested interconnectedness of MNs, regardless of the spatial orientation of ?neighboring? trees or fungal genets.  Large, mature trees acted as hubs in the phytocentric model and contributed more to MN connectivity than smaller tree cohorts, while R. vesiculosus genets linked more trees through more extensive MNs than R. vinicolor genets.  In the mycocentric model, there was a strong relationship between a genet?s  77connectivity and frequency of encounter across sampled space, which provides a proxy for estimating EMF genet networking characteristics based on the spatial extent and concentration of mycelia.  Lastly, the networking attributes of trees and fungal genets were influenced by soil moisture regime, but the emergent structural properties (i.e., topologies) of the MNs were similar in xeric and mesic plots.  This suggests the overall structure of Rhizopogon spp. MNs in mixed-aged interior Douglas-fir stands is robust to soil water stress.    4.4.2 Architecture of Rhizopogon spp. MNs and species-specific components The functional roles of MNs have been studied in controlled artificial systems (Finlay & Read, 1986b; Wu et al., 2001; He et al., 2004) and in natural ecosystems (Simard et al., 1997a; Lerat et al., 2002; Nara, 2006; Teste et al., 2010), but comparatively less is known about the spatial topology or temporal dynamics of MNs formed in mature, naturally occurring forest stands (Southworth et al., 2005; Lian et al., 2006; Beiler et al., 2010).  As I hypothesized, network links were irregularly distributed amongst trees and fungal genets, leading to MNs with complex, decentralized, hierarchical scale-free network topologies (Almaas & Barab?si, 2006; Barab?si, 2009).  In the phytocentric model, this suggests materials could be shuttled efficiently between numerous trees, including mature veterans and newly established trees, where the roots of those trees intermingle within a MN (Bray, 2003; Leake et al., 2004).  In the mycocentric model, it demonstrates the social ties that exist between different EMF species colonizing the same trees.  Because these fungi share a common energy source, they all stand to benefit if their collective morphological and functional diversity provides their host tree with better access to heterogeneous nutrient pools and a more diverse suite of defense mechanisms (e.g., van der Heijden & Sanders, 2002).    The hierarchical scale-free properties of MNs in this study suggest they are robust systems that would be resilient to random perturbations but could be susceptible to the loss of critical species or network components (Albert et al., 2000; Barab?si, 2009).  For example, the exclusion of R. vinicolor genets would have little effect on the overall connectivity or structure of the phytocentric MNs, but the loss of R. vesiculosus genets would dramatically reduce connectivity; and in Plot 4, effectively break the MN into isolated components.  Together, these species represent a meta-network where R. vinicolor-linked network  78components are nested within larger R. vesiculosus-linked MN components.  This multiplicity contributes to the interconnectedness, cohesiveness, and stability of the MNs by increasing the overall density and redundancy of links between trees, and by spreading the cliquishness of links between tree nodes across multiple spatial and temporal scales (Levin, 2005; Montoya et al., 2006).  4.4.3 Contrasts between phytocentric and mycocentric models My earlier work revealed that genets of R. vesiculosus and R. vinicolor segregate in space somewhat, while a high degree of spatial overlap occurs among the roots of neighboring interior Douglas-fir trees (Chapters 2 & 3).  Because the 10 x 10 m plots used in the current study limited the proportion of Rhizopogon mycelia systems sampled, I expected MNs from the mycocentric perspective to have a more cliquish, regular network topology compared to MNs from the phytocentric perspective.  Rather, I found the phytocentric and mycocentric models to have similar network topologies, despite the different scales of their underlying spatial patterns.    Trees are generally thought to be the largest, longest-lived inhabitants of forest ecosystems, and forest community structure is typically characterized based on the template of tree species composition and spatial patterns.  This, together with practical interest in forest health, silviculture and stability makes phytocentric approaches a logical choice for modeling MNs, and the primary focus of my study.  However, mycocentric MN models provide equally relevant, if less readily applicable, information regarding the structure and functioning of EMF communities (Southworth et al., 2005; Selosse et al., 2006).  The socio-spatial networking topology of EMF species (which fungi are linked through which plants over space and time) can be used to detect or predict the life history traits of EMF species (e.g., host-specificity, dispersal strategies, exploration type, phenology, etc.).  Moreover, the architecture of mycocentric MNs can provide insights into EMF species coexistence and community stability, as well as coevolutionary relationships between EMF species and their host plants, as shown for other mutualistic networks (Bascompte & Jordano, 2007) and by researchers using spatially explicit approaches to study belowground ecology (Ettema & Wardle, 2002; Pickles et al., 2010).  794.4.4 Determinants of mycorrhizal network architecture One of the utilities of network analysis is the ability to identify key determinants of the system?s structure or function (de Nooy et al., 2005; Bascompte, 2009a).  It was reported in Chapter 3 of this dissertation that networking attributes of trees and Rhizopogon spp. genets (tree node degrees and genet link strengths in a phytocentric MN) were positively correlated with their physical size.  Based on this finding, I predicted that large trees and R. vesiculosus genets would contribute more to MN connectivity at the forest stand scale, and the spatial frequency of these features would primarily determine the universal architecture of MNs.  In support of this prediction, a statistically significant relationship between physical size and MN connectivity was found consistently across forest plots in the current study.  With the potential removal of large hub trees or fungal genets such as those produced by R. vesiculosus, connectivity between trees (phytocentric model) or EMF genets (mycocentric model) would be reduced and the emergent structural properties of the MNs changed dramatically (see Chapter 3).  A loss of key nodes or links could have cascading effects throughout the system that ultimately lead to a reduction in MN stability (Dunne et al., 2002; Ellison et al., 2005; Bascompte, 2009a).  For example, if the largest trees were removed from a stand due to selective harvesting or insect attacks, the remaining trees may not adequately meet the carbon demands of EMF species forming large perennial genets such as R. vesiculosus, which in turn could suppress residual tree growth and the regenerative capacity of the stand.           In my comparisons between R. vesiculosus and R. vinicolor components of the phytocentric MN model, topological differences could be attributed to genet size asymmetries between these species (Kretzer et al., 2004; Kretzer et al., 2005; Beiler et al., 2010) as well as differences in their degree of mycelia proliferation (Chapter 2).  In the current study, these patterns were generalized by a positive relationship between the number of times a genet was sampled across space and the number of tree genotypes associated with those genets.  Differences in the sample frequency of genets resulted in part from plot edge effects (the proportion of genets occurring in 10 x 10 m plots varied), and this provided a proxy for assessing the networking capabilities of EMF species based on the size of their genets, given the applied sampling resolution and spatial constraints.  My findings demonstrate that carbon  80cost discrepancies between trees associated with small (e.g., R. vinicolor) or short-lived genets compared to large (e.g., R. vesiculosus) or long-lived genets would be offset by the corresponding number of trees sustaining those genets (i.e., energy demands of genets are dispersed among trees in their social network).   Similarly, with increasing tree sizes, the probability of a tree encountering genets and accumulating links to other trees increased.  Large, mature trees acted as hubs in the network with a higher degree of connectivity relative to younger trees.  The graph topological position of hub trees suggest they play a foundational role in the self-regeneration of these structurally complex forests; for example, by altering local environments towards conditions more suitable for the establishment and growth of young trees (Simard, 2009).  The presence of large trees can influence the ambient temperature and moisture of local environments, modify local edaphic conditions (soil pH, nutrient status, etc.), and sustain rich assemblages of EMF species that provide a diverse inoculum source to regenerating seedlings (Flores & Jurado, 2003; Querejeta et al., 2007; Teste & Simard, 2008).  When seedlings are colonized by EMF forming mycorrhizal networks with veteran trees, they gain access to hydraulically lifted water and patchily distributed nutrients that would otherwise be limiting resources (Leake et al., 2004; Egerton-Warburton et al., 2007; Warren et al., 2008).  The relationship between tree size and connectivity was not as strong as observed for EMF genet sizes, because young trees could become highly connected if colonized by a large, hyper-linking Rhizopogon sp. genet.  This was evident in plot six, where one of the two most highly connected trees, with direct links to 37 other trees (88.1 % of trees in the network), was a sapling belonging to the youngest cohort.        Large hub trees and large fungal genets were important for network connectivity, but it was size heterogeneities among network participants that generated the complex scale-free MN architectures.  From both the phytocentric or mycocentric model perspectives, node and link size asymmetries contributed to the nestedness of connectivity and hierarchical structure of the MNs.  The nestedness of interactions between plant and fungal species in the mycorrhizal symbiosis is a key indicator of self-organization and could influence the diversity, stability, and evolutionary dynamics of forest communities; as demonstrated by earlier studies of  81plant-pollinator mutualistic networks (Bascompte & Jordano, 2007).  MNs formed by EMF species with narrow host specificity could be nested within MNs formed by host-generalist species (or vice versa regarding plant specificity for mycorrhizal types), adding further possibilities for the outcome of inter-specific competitive interactions between trees (Molina et al., 1992; Jordano et al., 2003; Ishida et al., 2007; McGuire, 2007).  This implies a mechanistic link between the structure and functioning of EMF communities belowground and forest stands aboveground.         The ecological implications of body size have garnered much attention with regards to species interactions and ecosystem processes over time and with increasing space and system complexity (Odum, 1969; Lindstedt & Calder, 1981; Peters, 1983; Brown et al., 2002).  However, this variation is seldom accounted for in systems-level studies of mutualism or trophic interaction networks (Otto et al., 2007; Chamberlain & Holland, 2009).  Preferential node attachment is an important mechanism distinguishing scale-free network structures from random networks where every node has an equal probability of attachment (Barab?si & Albert, 1999; Jordano et al., 2003).  In MNs, a node?s probability of attachment comprises both deterministic mechanisms (e.g., host specificity leading to preferential attachment, plant-fungal communication, etc.) and stochastic elements (e.g., dispersal capabilities, size heterogeneities, link strengths, etc.) (Rezende et al., 2007; van der Heijden & Horton, 2009).  The complex MN structures in my study are self-organizing because as trees or fungal genets grow over time, their probability of encountering and forming associations with other organisms increases (Levin, 1992; Turcotte & Rundle, 2002; Levin, 2005).  Thus, the combination of a node?s probability of attachment and the strength of links available to that node could be characterized as the ?interactive interface? of the node, which tends to increase over time and space and with increasing system complexity (Jordano et al., 2003; Barrat et al., 2004).  This property is similar in form to the ecological link strengths in mutualistic and trophic networks, which Williams et al. (2002) referred to as the ?sphere of potential influence? of a species.  In addition to phenotypic traits such as tree or EMF genet sizes, the interactive interface of species also includes their phenology (e.g., seasonality, longevity, dispersal mechanisms, etc.), physiology (e.g., growth rates, nutrient transport mechanisms, enzymatic capabilities, etc.), ecology (e.g., host-specificity, competitive dynamics, etc.), and  82other evolutionary traits (Rezende et al., 2007; van der Heijden & Horton, 2009).  Thus, a better understanding of EMF species autecology will help bridge the gaps between the architecture of MNs and their ecological and evolutionary significance in complex systems.  4.4.5 Network architecture in xeric and mesic soil moisture regimes I expected greater MN connectivity in xeric stands relative to mesic stands in keeping with the stress-gradient hypothesis.  The stress-gradient hypothesis infers that net effects of community interactions shift in importance from competition to facilitation with increasing environmental stress (Bertness & Callaway, 1994; Callaway, 1995; Brooker et al., 2008).  Studies have found that EMF community structure and dynamics can be sensitive to soil moisture, temperature and nutrient status, which could feed back to influence the growth and regeneration potential of forest trees with climate change (Erland & Taylor, 2002; Rillig et al., 2002).  Recently, Bingham reported the increased importance of MN regeneration facilitation in drier soils compared to moist soils (PhD thesis, 2011).  In my study, soil moisture regime affected the connectivity and node clustering coefficients of trees linked through Rhizopogon spp. genets.  The node degree of trees increased more sharply with increasing tree dbh in xeric plots compared to mesic plots, suggesting large trees played a greater role in the interconnectivity of MNs relative to younger cohorts in xeric plots.  Moreover, the effects of Rhizopogon spp. and soil moisture regime on tree normalized node degrees and node clustering coefficients were not independent.  Nonetheless, the emergent MNs were complex and highly interconnected in every plot sampled, and I found no significant differences in emergent MN structural properties when contrasted between xeric and mesic soil moisture regimes.  This suggests the structural thresholds of these MNs extend beyond the soil moisture constraints represented in this study, and points to R. vesiculosus and R. vinicolor being well adapted for a broad range of environmental conditions present in these forests and robust to soil water deficiencies (Pimm, 1984).  It would be insightful to examine the structure of Rhizopogon spp. MNs in wetter and drier soil moisture conditions across the broader geographic range of P. menziesii var glauca (e.g., where P. menziesii var glauca occurs as a component in wet cedar-hemlock forests in the north and in dry mixed conifer ?sky islands? in the south), and to contrast MN structure between interior and coastal variants of P. menziesii.     83In water-limited interior Douglas-fir forests, the potential for hydraulically lifted water from deep-rooted mature trees to be redistributed through MNs to shallow-rooted seedlings could be of paramount importance to the self-maintenance of mixed-cohort stand structures (Querejeta et al., 2003; Egerton-Warburton et al., 2007; Meinzer et al., 2007; Schoonmaker et al., 2007; Warren et al., 2008).  My results, together with studies demonstrating the transfer of water through rhizomorph-forming fungi (Duddridge et al., 1980; Brownlee et al., 1983; Egerton-Warburton et al., 2007), suggest that R. vesiculosus and R. vinicolor genets could contribute towards the persistence of mixed-aged interior Douglas-fir forest stand structures by ameliorating seedling water stress (Dawson, 1993; Allen et al., 2003; Brooks et al., 2006; Plamboeck et al., 2007) (see also Chapters 1 & 2).  This supposition is consistent with the evolutionary history of Rhizopogon spp. and hypotheses regarding the ecological advantages of hypogeous fruiting habit in xeric environments (Thiers, 1984; Lilleskov et al., 2009).  Further evidence was provided by Parke et al. (1983) in microcosms, where R. vinicolor was found to increase the water use efficiency of individual P. menziesii var. menziesii trees and to speed their recovery following periods of water stress.  Although MNs were structurally similar and complex in all plots sampled, it remains unknown how the effects of Rhizopogon spp. MNs on tree regeneration facilitation may change over time or across different soil moisture regimes as predicted by the stress-gradient hypothesis (Bertness & Callaway, 1994; Sthultz et al., 2007; Brooker et al., 2008).  This could be tested experimentally by out-planting local seeds or seedlings in xeric and mesic forests with mesh barriers used to restrict the passage of roots and/or mycelia including/excluding rhizomorphs.      4.4.6 Summary  This study demonstrates the utility of network analysis for describing complex ecological systems, deconstructing these systems to identify key parts and processes, and for comparing the integrity of these systems between differing points along an environmental gradient.  Whether viewed from phytocentric or mycocentric perspectives, Rhizopogon spp. genets and interior Douglas-fir trees were highly interconnected through spatially and socially complex MNs.  MN topologies were characteristic of hierarchical scale-free networks, suggesting the bonds between these organisms are cohesive and robust against the random loss of participants, but could be susceptible to the targeted loss of large trees and large fungal  84genets.  The MNs self-organized endogenously via a positive relationship between the physical size of tree and fungal symbionts and their probability of association (resulting in modularity of their individual interactive interfaces), and by the frequency distribution of body sizes and spatial patterns of each population.  This highlights the bi-directional influence of MNs, where the structure and functioning of EMF communities can influence that of forest stands, and vice versa.  Soil moisture regime affected the networking attributes of tree nodes, but the overall architecture of the MNs were similar in mesic and xeric forest stands.  The results of this study suggest Rhizopogon spp. links between trees are resilient to soil water stress, and could become increasingly important for regeneration facilitation of mixed-aged interior Douglas-fir forests if climate change leads to more frequent and severe drought episodes.     85Table 4.1    Physical characteristics of the six 10 x 10 m study plots.  Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6 Degrees Latitude 50.851515 50.852863 50.853947  50.852333 50.854138 50.855897 Degrees Longitude -120.522422 -120.522762 -120.522563  -120.518997 -120.519682 -120.518665 Elevation (m):  1,155 1,146 1,131  996 1,060 1,076 Slope aspect: 86?E 86?E 84?E  88?E 132?SE 104?E          Number trees/ number tree genotypes isolated  from Rhizopogon spp. mycorrhizas in plot        Cohort 1 (age ? 15 yrs.):  4/3 7/1 10/12  3/1 2/2 2/2  Cohort 2 (age 16 ? 45 yrs.): 3/7 0/1 2/4  6/18 3/11 9/16  Cohort 3 (age 46 ? 70 yrs.): 6/12 5/8 1/6  4/6 4/8 7/19  Cohort 4 (age 71 ? 95 yrs.): 0/4 0/2 2/7  0/1 0/4 0/ 2  Cohort 5 (age ? 95 yrs.): 0/1 0/1 1/2  0/1 0/2 0/2  Total: 13/26 12/13 16/31  13/27 9/27 18/41         Tree basal area (m2 ha-1): 2.19 1.85 4.70  3.40 2.38 4.56 Tree height (m) range:      Mean ? SD (m): 0.3 ? 24.5 12.8 ? 7.1 0.1 ? 29.8 6.0 ? 8.9 0.2 ? 24.8 8.8 ? 9.3  1.8 ? 23.1 14.1 ? 8.4 1.7 ? 31.1 17.6 ? 7.9 1.2 ? 26.3 15.3 ? 8.6 Tree dbh1 (cm) range:       Mean ? SD (cm): 0.0 ? 65.7 19.0 ? 14.9 0.0 ? 88.0 11.5 ? 20.8 0.0 ? 52.8 17.2 ? 16.3  0.2 ? 110.0 18.7 ? 20.3 0.5 ? 47.4 22.0 ? 13.4 0.0 ? 96.1 22.8 ? 19.4         Number of understory vascular plant species: 16 15 16  19 19 14 Mean % Calamagrostis rubescens cover: 0.138 0.403 0.084  0.119 0.069 0.094 Mean % Pleurozium schreberi cover: 0.121 0.028 0.025  0.356 0.217 0.213 Ratio of mean % cover grass:moss: 0.011 0.143 0.034  0.003 0.003 0.004 1diameter at ?breast? height, measured at 1.3 m height on upslope side of tree  86Table 4.2    The percent cover and frequency of forest floor substrates and understory vascular plant and bryophyte species among 1 m2 quadrat grid frames (n = 36 per plot) sampled in 5 m intervals within 30 x 30 m  plots of Pseudotsuga menziesii var. glauca forest with either mesic or xeric soil moisture regimes.  Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6  Cover Occur Cover Occur Cover Occur  Cover Occur Cover Occur Cover Occur CWD (coarse woody debris) 1.39 0.06 NA NA 2.67 0.27  2.92 0.25 1.78 0.19 3.58 0.36 Duff (needle litter) 49.99 0.94 26.29 0.68 3.64 0.06  32.22 0.61 22.94 0.67 56.25 0.92 Woody debris (coarse litter) NA NA 0.21 0.06 NA NA  8.89 0.31 18.28 0.97 6.69 0.58 Rock (?15 cm circumradius) 2.92 0.36 1.21 0.21 3.61 0.39  2.11 0.14 4.75 0.47 0.89 0.31 Psuedotsuga menziesii var. glauca 0.42 0.03 0.44 0.03 1.33 0.12  0.69 0.08 0.67 0.08 1.53 0.08 Amelanchier alnifolia NA NA 3.47 0.53 0.33 0.06  1.36 0.31 1.53 0.39 NA NA Juniperus communis 1.11 0.08 13.74 0.41 NA NA  NA NA 1.31 0.08 NA NA Rosa acicularis 0.22 0.08 0.24 0.12 0.27 0.09  0.78 0.39 2.42 0.53 1.58 0.42 Shepherdia Canadensis NA NA 0.59 0.06 1.21 0.03  0.28 0.03 0.42 0.08 1.03 0.08 Spiraea betulifolia 4.86 0.86 2.53 0.50 3.76 0.55  2.69 0.56 1.92 0.42 2.44 0.56 Arcotostaphylos uva-ursi 0.25 0.19 0.38 0.18 3.12 0.45  0.08 0.06 NA NA NA NA Symphoricarpos albus 0.14 0.03 NA NA NA NA  0.28 0.06 3.17 0.28 NA NA Allium cernuum 0.08 0.06 0.03 0.03 0.88 0.42  0.11 0.06 0.19 0.14 NA NA Arnica cordifolia 3.40 0.56 NA NA 2.15 0.55  2.28 0.33 1.5 0.5 2.39 0.75 Aster consipcuus NA NA 1.09 0.12 NA NA  NA NA NA NA NA NA Berberis aquifolium 0.14 0.06 0.12 0.09 NA NA  0.28 0.14 1.75 0.44 1.67 0.53 Castilleja miniata NA NA NA NA 0.06 0.03  NA NA NA NA NA NA Calamagrostis rubescens 13.75 0.97 40.32 0.94 8.39 0.55  11.89 0.83 6.94 0.89 9.36 0.92 Chimaphila umbellate NA NA NA NA NA NA  NA NA NA NA 0.61 0.06 Fragaria virginiana 0.10 0.11 0.18 0.12 0.06 0.06  0.64 0.42 0.47 0.28 0.25 0.06 Goodyera oblongifolia 0.03 0.03 NA NA NA NA  0.06 0.06 0.19 0.11 1.28 0.31 Achillea millefolium 0.11 0.11 0.53 0.26 0.30 0.21  0.03 0.03 NA NA NA NA Antennaria racemosa 0.08 0.03 NA NA 1.12 0.27  NA NA NA NA 0.14 0.03 Linnaea borealis NA NA NA NA NA NA  1.22 0.28 NA NA 0.78 0.25 Smilacina racemosa NA NA NA NA NA NA  0.50 0.25 NA NA NA NA Aguilegia formosa NA NA NA NA NA NA  NA NA 0.11 0.11 0.08 0.06 Vicia Americana 2.06 0.44 NA NA 0.03 0.03  0.17 0.14 0.03 0.03 4.83 0.89 Viola species NA NA 1.76 0.09 0.15 0.09  3.06 0.72 0.03 0.03 NA NA Peltigera aphthosa 0.36 0.19 0.85 0.29 2.33 0.45  0.42 0.19 3.28 0.69 NA NA Pleurozium schreberi 12.10 0.56 2.82 0.21 2.45 0.30  35.61 0.72 21.67 0.86 21.25 0.81  87Table 4.3    Attributes of mycorrhizal network formed between mixed-aged interior Douglas-fir trees linked through Rhizopogon vesiculosus and R. vinicolor fungal genets.  Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6         Number of nodes (trees linked in MN): 26 13 31  27 27 41 Number of ?links? between tree pairs: 258 66 283  254 177 434 Network diameter: 3 2 3  2 3 3 All-degree centralization: 0.23 0.31 0.39  0.33 0.45 0.42 Betweenness centralization: 0.06 0.01 0.18  0.05 0.15 0.05 Number R. vesiculosus/R. vinicolor      mycorrhizas sampled in 10 x 10 m: 69/7 29/39 79/48  59/53 112/12 130/121 Number R. vesiculosus genets per tree:      Mean ? SD: 0 ? 2 0.92 0 ? 2 0.50 0 ? 2 0.81  0 ? 4 1.15 0 ? 2 1.22 0 ? 4 1.33 Number R. vinicolor genets per tree:      Mean ? SD: 0 ? 1 0.15 0 ? 3 0.67 0 ? 3 0.75  0 ? 1 0.85 0 ? 2 0.56 0 ? 5 1.39 Proportion of trees in 10 x 10 m colonized by Rhizopogon spp. (%): 0.846 0.500 0.875  0.846 0.778 0.889 Node degree range:      Mean node degree ? SD: 0 ? 24 18.00 ? 8.49 0 ? 12 6.95 ? 4.94 0 ? 29 17.15 ? 8.92  0 ? 26 17.52 ? 7.34 0 ? 24 12.21 ? 7.03 0 ? 37 20.19 ? 10.10 Normalized node degree range (%):      Link density ? SD: 0 ? 0.857 0.643 ? 0.303 0 ? 0.667 0.386 ? 0.274 0 ? 0.906 0.536 ? 0.279  0 ? 0.929 0.626 ? 0.262 0 ? 0.857 0.436 ? 0.251 0 ? 0.881 0.481 ? 0.240 Node betweenness centrality (%) range:      Mean node centrality (%) ? SD: 0 ? 0.064 0.006 ? 0.016 0 ? 0.013 0.004 ? 0.006 0 ? 0.187 0.011 ? 0.033  0 ? 0.057 0.009 ? 0.015 0 ? 0.166 0.017 ? 0.035 0 ? 0.062 0.011 ? 0.017 Node clustering coefficient (0-1) range:      Mean cc?? SD (0-1): 0 ? 1 0.844 ? 0.365 0 ? 1 0.627 ? 0.432 0 ? 1 0.843 ? 0.282  0 ? 1 0.847 ? 0.252 0 ? 1 0.828 ? 0.271 0 ? 1 0.798 ? 0.259    88 Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6 Spearman?s rho1:             Tree cohort vs. (dbh)1/2 0.959* 0.979* 0.983*  0.867* 0.940* 0.909*      Tree cohort vs. height 0.931* 0.918* 0.895*  0.673* 0.913* 0.861*      Tree cohort vs. node degree: 0.498* 0.710* 0.327  0.410 0.523* 0.025      Tree cohort vs. node centrality: 0.279 0.331 0.389  0.374 0.494* 0.064      Tree cohort vs. node cc?: 0.225 0.700* -0.285  0.025 -0.322 -0.094      Tree node degree vs. node cc?: 0.146 0.452 -0.350  -0.483* -0.449 -0.436* 1(*) correlation is significant at P ? 0.008 (Bonferroni adjusted ?) level                  89Table 4.4   The frequency distribution of interior Douglas-fir tree cohorts linked through either Rhizopogon vesiculosus or R. vinicolor genets in mycorrhizal networks, including all trees with roots found in Rhizopogon spp. mycorrhizas inside 10 x 10 m plots.   Xeric soil moisture regime  Mesic soil moisture regime Number of trees in each cohort linked  through R. vesiculosus (VESI) genets: Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6         Cohort 1: 2 0 9  1 0 2  Cohort 2: 6 0 3  13 11 12  Cohort 3: 11 8 5  6 8 15  Cohort 4: 4 1 7  1 4 2  Cohort 5: 1 1 2  1 2 2  Total: 24 10 26  22 25 33 % of trees in 100 m2 linked by VESI: 76.92% 41.67% 68.75%  76.92% 55.56% 72.22% % of networking trees linked by VESI: 92.31% 76.92% 83.87%  81.48% 92.59% 80.49%          Number of trees in each cohort linked  through R. vinicolor (VINI) genets:       Cohort 1: 0 1 7  0 2 1  Cohort 2: 2 1 1  12 1 9  Cohort 3: 1 6 2  5 3 14  Cohort 4: 1 2 5  0 2 1  Cohort 5: 1 1 2  0 0 1  Total: 5 11 17  17 8 26 % of trees in 100 m2 linked by VINI: 7.69% 41.67% 50.00%  69.23% 44.44% 66.67% % of networking trees linked by VINI: 19.23% 84.62% 54.84%  62.96% 29.63% 63.41%  90Table 4.5    Attributes of mycorrhizal networks from the mycocentric perspective, with Rhizopogon vesiculosus and R. vinicolor genets as nodes linked through shared host trees in 10 x 10 m plots of xeric and mesic mixed-aged interior Douglas-fir forest.  Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6 Number of R. vesiculosus/R. vinicolor nodes (fungal genets): 4/2 2/4 4/5  4/2 7/4 7/10 Network diameter: 3 2 2  2 3 3 Number ?links? between genets: 3 7 18  3 13 28 Number ?loops? between genets: 1 3 3  8 9 25 All-degree centralization: 0.50 0.50 0.54  0.40 0.49 0.55 Betweenness centralization: 0.44 0.20 0.19  0.14 0.32 0.33         Number trees per R. vesiculosus genet:      Mean ? SD: 1 ? 24 7 ? 1.34 1 ? 10 5.5 ? 6.36 1 ? 22 7.25 ? 9.91  2 ? 19 7.25 ? 7.89 1 ? 15 6.29 ? 5.62 1 ? 22 7.29 ? 7.91 Number trees per R. vinicolor genet:      Mean ? SD: 1 ? 5 3 ? 2.83 1 ? 10 3.75 ? 4.27 1 ? 13 4.8 ? 5.22  4 ? 16 10 ? 8.49 1 ? 3 2.25 ? 0.96 1 ? 18 4.4 ? 5.30 Node degree range:      Mean node degree ? SD: 0 ? 3 1.33 ? 0.94 2 ? 5 3.33 ? 1.25 2 ? 8 4.67 ? 2.05  2 ? 5 3.67 ? 1.11 2 ? 8 4.00 ? 2.09 1 ? 14 6.24 ? 3.42 Normalized node degree (%) range:      Network link density ? SD (%): 0.00 ? 0.600 0.267 ? 0.189 0.400 ? 1.00 0.667 ? 0.249 0.250 ? 1.00 0.583 ? 0.257  0.400 ? 1.00 0.733 ? 0.221 0.200 ? 0.800 0.400 ? 0.209 0.063 ? 0.875 0.390 ? 0.214 Node betweenness centrality (%) range:      Mean node centrality ? SD (%): 0.00 ? 0.500 0.133 ? 0.197 0.00 ? 0.250 0.083 ? 0.118 0.00 ? 0.232 0.060 ? 0.094  0.00 ? 0.183 0.067 ? 0.083 0.00 ? 0.370 0.079 ? 0.135 0.00 ? 0.358 0.043 ? 0.089 Node clustering coefficient (0-1) range:      Mean node cc? ? SD (0-1): 0.00 ? 0.00 0.00 ? 0.00 0.50 ? 1 0.83 ? 0.24 0.46 ? 1 0.83 ? 0.22  0.60 ? 1 0.84 ? 0.18 0.39 ? 1 0.80 ? 0.24 0.00 ? 1 0.77 ? 0.28     91 Xeric soil moisture regime  Mesic soil moisture regime  Plot 1 Plot 2 Plot 3  Plot 4 Plot 5 Plot 6 Spearman?s rho1:             Number mycorrhizas sampled/genet      vs. span of genets: 0.941* 0.900 0.983*  0.754 0.956* 0.910*      Number mycorrhizas sampled/genet      vs. number of tree genotypes/genet: 1.00* 0.939* 0.996*  0.893 0.907* 0.990*      Number mycorrhizas sampled/genet      vs. node degree: 0.742 0.677 0.652  0.582 0.957* 0.900*      Number mycorrhizas sampled/genet      vs. betweenness centrality: 0.359 0.840 0.868*  0.751 0.881* 0.824*      Number mycorrhizas sampled/genet      vs. clustering coefficient: NA -0.121 -0.868*  -0.751 0.891* -0.570      Node degree vs. clustering      coefficient NA 0.500 -0.875*  -0.953* -0.862* -0.601* 1(*) correlation is significant at P ? 0.008 (Bonferroni adjusted ?) level        92 Figure 4.1    Photographs showing mixed-aged stand structures of xeric (a-c) and mesic (d-f) study plots.  93 Figure 4.2    Socio-spatial network models depicting the architecture of mycorrhizal networks, with interior Douglas-fir tree nodes linked through Rhizopogon spp. genets in 100 m2 plots with xeric (a-c) or mesic (d-f) soil moisture regimes. Nodes (circles) are sized relative to tree diameter and darken from yellow to forest green with increasing age class. Models are scaled relative to plots where Rhizopogon spp. mycorrhizas were sampled (squares with dashed lines).  94 Figure 4.3    Scatterplot showing the monitonically negative relationship between the node degree of trees (number of neighboring trees each tree node was liked to) and the network clustering coefficients of those trees, in mycorrhizal networks formed between mixed-aged interior Douglas-fir trees and Rhizopogon spp. genets in six plots with xeric (filled shapes; Spearman?s rho = -0.469, P < 0.01) or mesic (open shapes; Spearman?s rho = -0.763, P < 0.01) soil moisture regimes. 95       96 Figure 4.4    Rhizopogon vesiculosus (i) and R. vinicolor (ii) sub-network components of mycorrhizal networks formed among Pseudotsuga menziesii var. glauca trees in six 10 x 10 m plots with either xeric (a-c) or mesic (d-f) soil moisture regimes. The degree of connectivity and density of links between trees differed significantly between R. vesiculosus and R. vinicolor linked network components (P < 0.05, Wilcoxon signed-rank tests, see main text) with R. vesiculosus forming more links between more trees than R. vinicolor. 97 Figure 4.5    Mycorrhizal networks depicting Rhizopogon vesiculosus and R. vinicolor genets as nodes (circles filled with blue or pink, respectively) linked through shared interior Douglas-fir host trees in plots with xeric (a-c) or mesic (d-f) soil moisture regimes. Nodes are sized relative to the proportion of Rhizopogon spp. mycorrhiza samples in 10 x 10 m plots representing each genet; lines represent tree links between nodes, with line width corresponding to the number of host trees shared between genets. The lower right plot (f) is shown at 75% scale relative to others. All but one genet were linked through shared hosts, and this connectivity was not influenced by fungal species or soil moisture regime.  98Node degree (k) of Rhizopogon spp. genets0 2 4 6 8 10 12 14Frequency024681012R. vesiculosus genetsR. vinicolor genets  Figure 4.6    In mycorrhizal networks modeled with Rhizopogon spp. genets as nodes linked through shared interior Douglas-fir trees, the frequency distribution of genet node degrees (the number connections each node has to other nodes) was positively skewed, with a few genets of each species having higher than average node degrees.                     99              Figure 4.7    The number of trees colonized by R. vesiculosus genets (filled circles) and R. vinicolor genets (open circles) in mycorrhizal networks increased in relation to the number of times genets were sampled within 10 x 10 m plots, which provides a proxy for the link strength of fungal genets sampled within similar spatial extents (see main text).            100 Figure 4.8    Generalized linear mixed model results showing the predicted values for response variables (a) tree node degree, (b) normalized node degree, and (c) node clustering coefficient in relation to increasing tree DBH in cm, for trees linked in mycorrhizal networks through Rhizopogon vesiculosus (?VES?) or R. vinicolor (?VIN?) in plots with xeric (?DRY?) or mesic (?MOIS?) soil moisture regimes (see main text).   101Chapter 5:   Conclusion  5.1  Contributions to current knowledge 5.1.1 Summary This is the first study to describe the spatial and graph-theoretical architecture of stand-scale MNs in the field.  To the best of my knowledge, it is the first to contrast the physical attributes and socio-spatial topologies of complex ecological networks between differing points along a soil moisture gradient.  Intensive sampling across nested, hierarchal spatial extents was used to describe the belowground spatial patterns of trees and fungi, and the attributes of their collective MNs, in high resolution.  This provided a basis for comparing MN features within and across multiple, ecologically relevant spatial scales.  MN features were contrasted between xeric and mesic soil moisture regimes as a proxy for potential changes in site water stress anticipated with climate change.  5.1.2 Chapter 2 synopsis: research conclusions and implications Before the dawn of molecular identification methods, the cryptic nature of EMF had proven a formidable barrier to cataloging or categorizing EMF species and their autecology (Horton & Bruns, 2001; Zhou et al., 2001; Peay et al., 2008).  In Chapter 2, I used molecular markers to identify Rhizopogon spp. hyphae, rhizomorphs and tuberculate mycorrhizas that were intensively sampled in systematic excavations of the forest floor and upper mineral soils.  The fine-scaled 3-dimensional spatial patterns and pervasiveness, microsite associations, and exploration strategies of R. vesiculosus and R. vinicolor mycelia were described in detail.  These attributes were contrasted between species and between different soil moisture regimes to ascertain potential mechanisms explaining the coexistence of anatomically and morphologically similar sister species.    Rhizopogon vesiculosus genets occurred deeper, had more spatially prolific mycelia, and colonized more tree roots than R. vinicolor genets.  Both species were particularly prevalent in microsites associated with moisture retention such as the F and H forest floor layers, coarse woody debris and the lower surfaces of rocks.  However, the occurrence of R. vesiculosus shifted in the presence of R. vinicolor towards deeper mineral soil horizons  102where R. vinicolor was absent, suggesting competition and resource foraging strategy are important for niche partitioning between these species.  Amarasekare (2003) outlined four principal axes of niche differentiation, including resources, natural enemies, space, and time; and noted that spatial niche differentiation is largely determined by species competitive strategies (investment tradeoffs between competitive strength versus evasion of competition, and the scales at which competition occurs) (see also Bolker & Pacala, 1999).  Although controlled experiments would be needed to elucidate direct or indirect mechanisms behind this competition, the result is a shared preference model of niche differentiation (Wisheu, 1998).  The spatial patterns of EMF mycelia may also relate to functional traits (Allen & Allen, 1992; Agerer, 2001; Agerer, 2006); for example, the spatially extensive and pervasive growth form of R. vesiculosus compared to R. vinicolor could reflect tradeoffs between the energy demands associated with spatial proliferation compared to the production of enzymatic defenses (Ogawa, 1985; Andrews, 1992; Read, 1992).  This work demonstrates that spatial niche differentiation is an important mechanism shaping the composition and structure of EMF communities, especially regarding cryptic species with seemingly redundant forms and functions.      I found that R. vesiculosus and R. vinicolor were associated with moist microsites within plots, and their mycelia were generally more pervasive in mesic compared to xeric plots.  These results were in contrast to my prediction that both species would be more prolific in xeric sites compared to mesic sites in dry interior Douglas-fir forests owing to their drought-adapted morphology, combined with increased carbon investment from trees in stressful environments as predicted by the stress-gradient hypothesis (Callaway et al., 2002; Espeland & Rice, 2007; Maestre et al., 2009).  The results of my study suggest the drought-adapted morphology and autecology of Rhizopogon spp. allow them to persist in xeric conditions, and may be important for ameliorating water-stress in interior Douglas-fir trees, but there was no evidence of positive feedbacks between the symbionts with increasing site water stress.   The results of this study also provided evidence that EMF mycelia can provide continuous linkages between the roots of different trees.  Roots from multiple trees shared a common Rhizopogon spp. mycelium in 8% of all 20 cm3 soil blocks sampled, and up to 11 trees were linked by a single continuous mycelium (which is a conservative estimate due to sampling  103constraints).  This information addresses a crucial assumption made in Chapters 3 and 4 regarding the structure of MNs; and is a prerequisite to understanding the ecological significance of MN functioning (Amaranthus & Perry, 1994; Simard & Durall, 2004; Peter, 2006).  For example, MN effects on the establishment, species diversity, and spatial patterns of plant communities and reciprocal effects plants have on belowground fungal communities are difficult to quantify without a spatiotemporal framework for MN linkages in situ (Francis & Read, 1995; Allen et al., 2003).  Because energy requirements are an intrinsic component of a species? life history traits, I emphasize that ectomycorrhizal root networking characteristics (spatiotemporal continuity, pervasiveness, and plasticity of a mycelium?s root colonization patterns) are an integral and distinguishing feature of mycelia exploration types proposed by Agerer (2001).  5.1.3 Chapter 3 synopsis: research conclusions and implications The architecture of MNs has implications for forest regeneration and community dynamics, patch size requirements for maintaining ecosystem functioning and biological diversity, and forest resilience to disturbance or stochastic events (Perry et al., 1990; Amaranthus & Perry, 1994; Simard, 2009).  Based on previous work showing that R. vesiculosus forms larger genets than R. vinicolor (Kretzer et al., 2004; Kretzer et al., 2005), I predicted that R. vesiculosus would link more trees than R. vinicolor.  I found that nearly 70% of trees in the plot, including all cohorts, were linked belowground through one or more Rhizopogon spp. genets.  Rhizopogon vesiculosus genets on average were larger, occurred deeper in the soil profile, and linked more trees than R. vinicolor genets.  I also confirmed the prediction that large trees would sustain more fungal genets, and thus be more connected in the network than smaller trees, resulting in a scale-free network where large trees serve as hubs.    Regenerating seedlings were linked into the extensive MN of mature veteran trees, and seedling establishment was likely facilitated by these linkages through increased access to water, nutrients, and surplus carbohydrates (Finlay & Read, 1986b; Nara, 2006; Teste et al., 2009; Teste et al., 2010).  The potential for such transfers was supported by the physical properties of the Rhizopogon species, which formed spatially extensive genets (I observed R. vesiculosus genets spanning up to 21 m) and vessel-like hyphal structures (i.e., rhizomorphs)  104(Brownlee et al., 1983; Agerer, 2006).  By associating with both young and mature tree cohorts, R. vesiculosus and R. vinicolor secure both current and future sources of energy (see e.g., Jonsson et al., 1999).    Correspondences between the size of genets and the number of trees they colonize could reflect tradeoffs between r- and K- selected life history traits (MacArthur & Wilson, 1967).  For example, although R. vinicolor has been successfully grown on nursery seedlings (Castellano, M, A. & Trappe, J, M., 1985), there are no such reports for R. vesiculosus, which likely relies on carbon subsidies from mature trees or groups of trees to meet its high carbon demands (Fleming, 1983; Fleming, 1984).  In a study by Simard et al. (1997b), R. vesiculosus/vinicolor mycorrhizas were 20 times more abundant on understory seedlings when fungi had access to roots of mature trees, versus seedlings where root systems were isolated by trenching.  Further evidence was provided by Cline et al. (2005), who found that with increasing distance from mature P. menziesii var. menziesii trees, the diversity of EMF species on seedlings decreased while the relative abundance of Rhizopogon spp. on seedlings increased (also see Taylor & Bruns, 1999).  The loss of carbon subsidies and the subsequent increased burden on photosynthate from residual trees could partly explain the regeneration suppression observed in selective cutting operations where EMF communities are maintained (e.g., Staebler, 1956), in contrast to clear-cuts or stand-replacing disturbances where EMF communities are depleted (especially regarding species forming large perennial genets).  However, I also found that mature residual trees support more genetically diverse assemblages of EMF compared to smaller trees, which could help maintain or accelerate the return of the community to old-growth forest characteristics following disturbances (Perry, 1989; Amaranthus & Perry, 1994; Franklin et al., 2002; Cline et al., 2007; Rosenvald & L?hmus, 2008; Simard, 2009).     105I found a strong positive association between tree size and connectivity, such that large trees acted as hubs in the network.  This, together with the high interconnectedness of the stand, resulted in a scale-free network structure with small-world properties.  Because the structure of a network also indicates its robustness (e.g., Albert et al., 2000), it can be inferred that the MN connectivity I describe would be resilient to random perturbations (fine-root turnover, blow-down, or trees randomly targeted by insects or disease) but susceptible to attacks targeting large veteran trees (high-grading, or insects/disease targeting mature trees).  5.1.4 Chapter 4 synopsis: research conclusions and implications To my knowledge, this is the first study to contrast ecological networks, including the attributes of internal biological components and their collective emergent properties, between different levels of a limiting environmental factor.  Just as the understanding of MN structure and functioning depends on knowledge concerning the autecology of biological components, the application of this knowledge requires systems-level comparisons at coarse spatiotemporal scales (Lau et al., 2010; Klironomos et al., 2011).  My objectives in Chapter 4 were to: determine the universal structural properties of MNs across plots, contrast MN structure between phytocentric and mycocentric perspectives, identify key determinants of emergent MN structural properties, and contrast MN properties between xeric and mesic forest sites while accounting for variation in stand characteristics (bole densities, basal areas, tree diameter distributions, etc.) and tree and fungal size asymmetries.  I contrasted MN architectures between soil moisture regimes to test whether potential regeneration facilitation differs with site water stress and as a proxy for potential changes with climate change.    In all six plots, trees and fungal genets were highly interconnected but network links were irregularly distributed amongst trees and fungal genets, with a few individuals having much higher than average numbers of links.  This finding supports results of chapter 3 and my hypothesis that MNs in mixed-aged interior Douglas-fir forests would have complex, scale-free network topologies (Almaas & Barab?si, 2006; Barab?si, 2009).  MN socio-spatial topology had scale-free characteristics whether modeled from the phytocentric or mycocentric perspective, despite the different scales of tree and fungal genet spatial patterns.  The strong positive relationship between the size measures of individual fungal genets or  106trees and their connectivity (node degree) and contribution towards network connectivity (node centrality, link strength) was consistent in all six plots.  Population-level variability in these traits- namely, ?body size? asymmetries between R. vesiculosus and R. vinicolor genets and between different tree cohorts- was the primary determinant of MN structural complexity (i.e., hierarchical scale-free architecture and link nestedness).  The emergent properties of MNs and the networking attributes of trees and fungal genets did not significantly differ between xeric and mesic soil moisture regimes.  The consistent development of complex, hierarchical scale-free MN structures across all plots sampled has important implications for the diversity, stability, and evolutionary dynamics of interior Douglas-fir forest ecosystems.  This topic has been explored at length for plant-animal mutualistic networks (Jordano et al., 2003; Montoya et al., 2006; Bascompte & Jordano, 2007; Bascompte, 2009b; Bastolla et al., 2009; Jordano, 2010) but remains surprisingly absent from the mycorrhizal literature (Perry, 1995; Bever, 2002; Simard, 2009).  In particular, the nestedness of interactions among individual plants and fungal genets suggests the complexity of these systems is self-organizing (Turcotte & Rundle, 2002; Bascompte et al., 2003; Levin, 2005; McGuire, 2007), and implies a mechanistic link between the structure and stability of forest stands and their belowground EMF communities (Molina et al., 1992; Perry, 1995; Ettema & Wardle, 2002; Johnson et al., 2005; Ishida et al., 2007; Simard, 2009).     The relationship between body size and degree of connectedness among trees and fungal genets suggests individuals tend to accumulate network links over time, and in systems with multiple extant cohorts large veterans act as hubs.  From the phytocentric perspective, hub trees and fungal genets (e.g., R. vesiculosus) could play a foundational role in the self-organization of complex stand structures through the competitive and facilitative effects of MNs (Perry, 1989; Simard, 2009).  MN architecture from the mycocentric perspective provides insights into factors influencing the structure, function, and stability of EMF communities (i.e., Ettema & Wardle, 2002; Kiers & van der Heijden, 2006; Selosse et al., 2006) and provides a template for assessing systems-level processes such as carbon economics in the mycorrhizal symbioses (e.g., Koide & Elliott, 1989; Tinker et al., 1994;  107Johnson et al., 1997; Simard et al., 2002; Johnson et al., 2006; van der Heijden & Horton, 2009).  In this regard, the strong correspondence between the size of Rhizopogon spp. genets and the number of trees they colonized suggests genet energy demands are suitably dispersed among trees in their social network.  The hierarchical scale-free topology of MNs suggests they are cohesive and robust against the random loss of participants, but could be susceptible to the targeted loss of large trees and large fungal genets (Albert et al., 2000; Dunne et al., 2002; Montoya et al., 2006).  This study demonstrates the utility of network analysis for identifying foundational or keystone species within an ecosystem (i.e., Dayton, 1972; Holling, 1992) and for predicting the potential consequences of losing these components (Dunne et al., 2002; Ellison et al., 2005; Bascompte, 2009b; Lau et al., 2010).  In chapter 4 I proposed the term ?node preferential attachment? sensu Barab?si & Albert (1999), a mechanism contributing toward scale-free network topologies, be expanded to ?node probability of attachment? in the context of MN self-organization and evolutionary dynamics.  Along with the strength of links available to nodes, this provides a more inclusive framework for quantifying the ?ecological interface? of a networking individual, population, guild, community, etc. based on measurable traits such as morphology, phenology, physiology, ecology and evolutionary traits (see chapter 4 discussion).  Elements of this concept have been tested analytically in various constructs (Jordano et al., 2003; Barrat et al., 2004; Rezende et al., 2007), and empirically by Williams et al. (2002) who describe the ?zone of potential influence? regarding path lengths of associative links between species in multipartite food webs.  In MNs, a node?s probability of attachment comprises both deterministic mechanisms (e.g., host specificity leading to preferential attachment, plant-fungal communication, competitive/facilitative feedbacks) and stochastic processes (e.g., legacy conditions, dispersal limitation, host mortality, size heterogeneities, link strengths) (Rezende et al., 2007; van der Heijden & Horton, 2009).  The ecological interface of trees and fungal genets forms the backbone of MN self-organized complexity because as trees or fungal genets grow over time their probability of encountering and forming associations with other organisms increases.   108Soil moisture regime influenced the networking attributes of tree nodes and Rhizopogon spp. links.  In particular, there was a stronger positive trend between tree cohort class (or dbh) and tree node degrees in xeric plots compared to mesic plots, suggesting the role of large trees in regeneration facilitation is more pronounced under water stress.  However, the emergent socio-spatial architectures of MNs in this study were similar between xeric and mesic soil moisture regimes.  This implies that R. vesiculosus and R. vinicolor genets and their host tree associations are generally robust to soil water deficiencies (i.e., Pimm, 1984).  The potential for hydraulically lifted water from deep-rooted mature trees to be redistributed through MNs to shallow-rooted seedlings could help maintain mixed-cohort stand structures in dry interior Douglas-fir forests (Dawson, 1993; Querejeta et al., 2003; Brooks et al., 2006; Allen, 2007; Egerton-Warburton et al., 2007; Meinzer et al., 2007; Plamboeck et al., 2007; Schoonmaker et al., 2007; Warren et al., 2008).  The resilience of MNs could become increasingly important for regeneration facilitation of mixed-aged interior Douglas-fir forests if the frequency and severity of drought episodes increases as predicted by climate change models (Simard, 2009; Bingham, 2011), although previous studies suggest the outcome of ectomycorrhizal symbioses vary in response to soil water stress at the species level (Theodorou, 1978; Parlad? et al., 2001; Ortega et al., 2004; Kennedy & Peay, 2007).    The particular importance of rhizomorph-forming fungi such as R. vesiculosus and R. vinicolor in tree water acquisition is exemplified by studies demonstrating transfer of water through rhizomorphs in/over ecologically relevant volumes and distances (see e.g., Duddridge et al., 1980; Brownlee et al., 1983; Egerton-Warburton et al., 2007) and direct evidence they increase tree resilience to water stress (Parke et al., 1983; but see Dosskey et al., 1990; Dosskey et al., 1991).  The greater size, pervasiveness, and depth range of R. vesiculosus genets compared to R. vinicolor (as reported in chapters 2 & 3) and greater contribution of this species towards MN connectivity as reported in chapter 4 suggests R. vesiculosus plays a particularly important role in interior Douglas-fir resilience to water stress.  This conjecture is supported by the findings of Querejeta et al. (2007) in an oak savannah ecosystem, who found the impacts of soil water stress on EMF hyphal proliferation were primarily limited to upper forest floor layers exposed to evaporation (0-5 cm; where R. vinicolor commonly occurs).    109After spending many hours searching for R. vesiculosus and R. vinicolor tubercles in the field, my impression is that in the F and H forest floor layers these species are generally most prolific during spring months; and vivacious, plump tubercles become more difficult to find during the warmer, drier months of summer before revitalizing in the fall.  In contrast, vivacious tubercles may still be found throughout the summer in deeper soil horizons and moist microclimates such as buried coarse woody debris and beneath large rocks.  I encountered exploratory rhizomorphs of R. vesiculosus and R. vinicolor throughout all seasons (including beneath snowpack in December) in a concurrent study using sterile sand-filled mesh bags as ?traps? buried at the interface between the H forest floor layer and A mineral soil horizon (unpublished data).  Moreover, a dispersed sampling of Rhizopogon spp. tubercles from plot 5 was performed each spring and fall (May/June and Sept./Oct.) for 2 yrs. following the initial sampling effort (6 total collections over 3 yr.) and over this time, multi-locus genotypes were consistently encountered in the same general positions when mapped using GIS (unpublished data).  However, these studies were limited by small sample sizes and more information regarding the seasonal dynamics of these fungi is needed to resolve the significance of their role in interior Douglas-fir resilience to water stress and stand dynamics in general.  5.2 Project strengths and limitations 5.2.1 Primary strengths of project One of the distinguishing features of this research is the use of high-resolution sampling across hierarchical spatial scales to concurrently describe the intrinsic and emergent structural properties of a complex, multifarious ecological network.  Population-level body size asymmetries among interior Douglas-fir and Rhizopogon spp., and spatial niche partitioning between Rhizopogon sp. populations, added structural complexity (e.g., connectedness, modulation, and flux pathways) and stability to forest stand-scale MNs by diversifying the interactive interfaces of individuals.  This work highlights the mycorrhizal symbiosis as a model system for studying links between the structure, function and stability of complex ecosystems (Dahlberg, 2001; Allen et al., 2003; Peay et al., 2008).     110A primary strength of this study was the use of spatially explicit sampling and multi-locus microsatellite DNA identification methods to map the spatial patterns of fungal mycelia systems in situ.  It is estimated that 60-80% of EMF biomass is allocated to extramatrical mycelia, where active foraging for resources takes place near hyphal apices (Colpaert et al., 1992).  Overall, EMF mycelia represent a significant portion of microbial biomass and play an important role in carbon cycling within temperate forest ecosystems (Fogel & Hunt, 1979; Wallander et al., 2001; H?gberg & H?gberg, 2002; Langley & Hungate, 2003; Godbold et al., 2006).  Yet, few studies have used molecular techniques to study the spatial patterns of fungal mycelia in bulk mineral soils (Zhou et al., 2001; Dickie et al., 2002b; Zhou & Hogetsu, 2002; Landeweert et al., 2003; Guidot et al., 2004; Genney et al., 2006).    To the best of my knowledge, this work represents the first study to test specific hypotheses regarding MN structure at the population level, the first to model stand-scale MNs formed between different tree cohorts in mixed-aged forests, and the first to use a complex systems-level approach to contrast the structural properties of MNs and their biological components between differing points along an environmental gradient.  Moreover, the presentation of MN structure from the ?mycocentric? perspective is particularly underrepresented in the MN literature (Southworth et al., 2005; Selosse et al., 2006).  The results of this study underscore the multidimensionality of the mycorrhizal symbiosis and the potential importance of MNs in forest ecosystem stability.    5.2.2 Project scope and limitations 5.2.2.1 The general problem of complexity: mycorrhizal networks in curtus The MNs I describe are only small windows into more expansive, ?open? systems in terms of space, time, and complexity (Levin, 1992).  I sampled only two species among a genetically diverse and functionally multifarious community of EMF that associate with P. menziesii var. glauca trees (Twieg et al., 2007).  Moreover, the spatial distribution patterns of tree roots and Rhizopogon spp. genets form a continuous matrix of networks spanning multiple scales of space and time; making them inherently difficult to measure with adequate sampling resolution (see Chapters 2-4).  In chapter 4, more than half of the tree roots sampled in 10 x  11110 m plots were matched to tree boles outside plot boundaries.  A tree located over 20 m distance from the center of Plot 6 (#116 in order of distance to center) had roots sharing Rhizopogon spp. genets with 30 other trees inside the plot.  Considering the observed root lengths of these trees, their distance and orientation from plots, tree bole densities surrounding the plots, and measures of connectivity inside the plots, these trees could be directly linked to 250 or more trees through genets of a single EMF species such as R. vesiculosus (see also Chapter 3).  Thus, the spatial extent, efficiency, adaptive capacity (resilience), and robustness (resistance) of MNs are even greater than I describe considering the taxonomic and functional diversity of EMF species associated with interior Douglas-fir trees.  Interestingly, the probability distributions of node connectivity and betweenness centrality were similar when scaled across nested spatial extents, even though the plot-level means of these measures were reduced by plot edge and shape effects (data not shown).  5.2.2.2 Sampling limitations: spatial extent and grain  Owing to the heterogeneity of mycelia spatial patterns among genets in any given EMF community, describing these traits often require tradeoffs between sampling extent and resolution (i.e., grain) which depend on the research objectives or hypotheses to be tested (Dungan et al., 2002; Legendre et al., 2002; Fortin & Dale, 2005).  The cryptic nature of EMF and a paucity of species autecology information further complicate such efforts (Lilleskov et al., 2004; Pickles et al., 2009).  In this study, intensive sampling was conducted across a hierarchy of spatial scales with lag distances ranging from decimetres to hectometres.  Each focal scale was selected to address specific hypothesis, which ultimately lead to certain limitations including the ability to extrapolate results beyond these contexts.  For example, meticulous forest floor excavations in chapter 2 were used to describe the spatial patterns and continuity of Rhizopogon spp. mycelia in high resolution, but excavating the entire extent of genets (found to span up to 21 m in chapter 3) would not have been possible with available resources.  Conversely, the 30 x 30 m of dispersed sampling in chapter 2 enabled a comprehensive census of fungal genets and tree root systems occurring at the forest stand scale, but required liberties regarding the assumption that MN links are spatially continuous across tens of metres (see chapter 3 discussion).  In chapter 4,  112concentrating sampling within 10 x 10 m plots enabled replication on the landscape and increased the likelihood of including only spatially continuous MN links, but grossly underestimated measures of potential MN connectivity because of plot edge and shape effects as discussed in the previous section.  This design could be improved in future studies by the use of circular plots, which would improve estimations of edge effects by controlling for shape effects (the distance of tree boles to plot center, and relationships between tree networking attributes and the proportion of root systems sampled, could be measured without bias introduced by orientation to square plot boundaries).  Despite this study representing one of the most intensively sampled (per unit space) EMF species to date, the number of tree genotypes identified from Rhizopogon spp. mycorrhizas continued to increase with the number of mycorrhizas collected (data not shown).  This suggests a complete census of networking trees was not reached even within focal plots.  5.2.2.3 Temporal limitations Several lines of evidence support the temporal continuity of R. vesiculosus and R. vinicolor.  First, their pervasive growth through soil (chapter 2), expansive genet sizes (chapter 3), and frequency and consistency of their occurrence across space (chapter 4) imply substantial longevity.  Second, exploratory hyphae of both species were encountered year-round using mesh bags (see section 5.1.4 paragraph 7) buried/sampled on a rotational basis in 3 month intervals.  Finally, multi-locus genotypes of both species persisted over several years of sampling in the same location.  However, none of these efforts yielded adequate data to conclusively describe the fine-scaled temporal patterns of individual R. vesiculosus or R. vinicolor genets.  The possibility for the revitalization of desiccated EMF root tips or hyphae over time (see e.g., Durall et al., 1994a) would be analogous to the clonal propagation of hyphal fragments across space, resulting in the overestimation of genet temporal continuity.  Although Rhizopogon spp. tubercles were found year round including beneath snowpack in winter, they were seemingly less vivacious and more difficult to find during the driest weeks of summer, with peaks in productivity during wetter periods in spring and fall (see Chapter 4 synopsis, section 5.1.4).  However, the sampling methods used did not allow empirical testing of temporal trends in Rhizopogon spp. productivity.  Knowledge in this area would  113benefit from future studies investigating Rhizopogon spp. tubercle vitality and extramatrical hyphal exploration at more frequent time intervals during dry summer months (July-Aug.) and beneath winter snowpack (Jan.-March).    Likewise, it remains to be answered whether the vertical partitioning I observed between R. vesiculosus and R. vinicolor mycelia may coincide with phenological differences or temporal changes in soil water availability (e.g., Trappe, 1965; Koide et al., 2007).  The spatial scale at which temporal changes take place may also vary among EMF species (Zhou et al., 2001; Izzo et al., 2005).  Numerous studies have demonstrated that where EMF species or genets persist over time, their patterns of occurrence can be spatially dynamic within a given area (Zhou et al., 2001; Izzo et al., 2005; Courty et al., 2008; Pickles et al., 2010).  Lian et al (2006) found that genets of Tricholoma matsutake effectively migrate across tree roots systems in a continually advancing mycelial front, where genets pass through rather than persist within a given area of soil.    5.3 Applications and inferences This research was undertaken in the dry-belt interior Douglas-fir forests immediately north of Kamloops in the Southern Interior Forest Region of BC, Canada (in areas surrounding UTM: 10 U 674256, 5636450).  Results are directly pertinent to areas where interior Douglas-fir occurs in mixed-aged, pure species stands.  They likely apply equally well to other Douglas-fir stand structures, given the consistency at which Rhizopogon spp. have been found to dominate EMF communities on P. menziesii var. menziesii and P. menziesii var. glauca tree roots.  Only practical applications are discussed, as theoretical applications (implications for cryptic species coexistence, self-organization of complex systems, etc.) were discussed at length in previous sections.    Mycorrhizal network architecture has implications for the self-organization, stability, and productivity of complex forest ecosystems.  Previous studies have found that MNs may be important for regeneration facilitation in dry interior Douglas-fir forests, but the existence of mycorrhizal links between plants was not explicitly demonstrated (Simard et al., 1997a; Teste et al., 2009; Teste et al., 2010).  In the present study, the continuity and  114multiplicity of Rhizopogon spp. links between roots of different tree cohorts was demonstrated and the complex architecture of these networks was described at forest stand scales.  This information provides a template for quantifying ecological ties between the spatial patterns and productivity of Rhizopgon spp. genets and interior Douglas-fir forest stands.    Large trees and large fungal genets acted as hubs for MN connectivity in this study.  In partial retention silvicultural systems, conserving large trees or groups of trees would facilitate understory regeneration more efficiently than retaining only young or middle-aged isolated trees by providing a more diverse legacy of EMF symbionts.  Moreover, prescriptions which include aggregates of trees with intermingling roots systems provide a refuge for diverse EMF communities while also supporting the carbon demands of large, perennial fungal genets such as R. vesiculosus and R. vinicolor.  Among R. vesiculosus genets represented by ? 20 mycorrhiza samples in this study, genets colonized an average of 16 trees each (SD ? 6, n = 12 genets).  Studies have found that tree growth responses can be suppressed by the carbon or water demands of fungal symbionts (Colpaert et al., 1992; Peng et al., 1993), which would be subsidized by mature trees or groups of trees under natural conditions (Simard et al., 1997a; H?gberg et al., 1999; Smith & Read, 2008).  Failure to account for EMF carbon requirements could retard the growth response of residual trees following partial cutting applications, ultimately decrease EMF species diversity, and suppress understory regeneration by altering MN dynamics.  However, it is important to note that the functional roles of MNs with respect to inter-tree competition or facilitation could change depending on site productivity, the distances between trees, and the developmental stage of forest stands.  Experimental studies are needed to test predictions regarding the effects of silvicultural applications or stand dynamics in general on MN structure and vice versa.  This study provides basic MN structural information needed to design such experiments.            115The presence of R. vesiculosus may be a useful indicator of stand-scale connectivity through MNs.  The morphological and ecological traits described in this study suggest R. vesiculosus could play an important role in the resource economics of interior Douglas-fir ecosystems; particularly regarding the potential translocation of water, carbon, and mineral nutrients between trees via the cytoplasmic streaming of materials through rhizomorphs (reviewed by Simard et al., 2002; Allen, 2007).  The substantial frequency and abundance of both living and dead Rhizopogon spp. tuberculate mycorrhizas in study plots suggests they contribute largely to belowground biomass and soil carbon fluxes in these forests (see also McDowell et al., 2001).  These traits, together with other known ecosystem services provided by Rhizopogon spp. (e.g., importance of their sporocarps in the diets of small mammals) and their ease of recognition in the field (tuberculate mycorrhizas, presence in mammal feces) make R. vesiculosus/R. vinicolor ideal candidates for future monitoring and experimental studies.    Both R. vesiculosus and R. vinicolor could be highly invasive if introduced to new Douglas-fir habitats.  Reforestation efforts have shown that Rhizopogon spp. can persist on nursery inoculated P. menziesii var. menziesii seedlings after outplanting in a variety of habitats (R. vinicolor by Castellano, M, A. & Trappe, J, M., 1985; Rhizopogon spp. by Pera et al., 1999).  Although they have a narrow host specificity, Rhizopogon spp. are adept at invading foreign habitats when co-introduced with P. menziesii (Chu-Chou & Grace, 1981; Parlad? et al., 1996) and could outcompete native EMF fungi in these locations.  Practitioners should consider potential impacts on endemic EMF species and Rhizopogon spp. genetic diversity prior to introducing seedlings inoculated with R. vesiculosus and R. vinicolor.   Rhizopogon spp. MNs are resilient to site water stress.  Rhizopogon spp. mycelia were more prolific in mesic compared to xeric plots, and showed selection for microsites associated with moisture retention within plots.  Buried or partially buried coarse woody debris and other moisture-retaining features (e.g., large rocks) could promote the growth of Rhizopogon spp. fungi.  Nonetheless, the continuity of Rhizopogon spp. mycelia across space was similar between xeric and mesic soil moisture regimes, providing a potential pathway for  116the redistribution of hydraulically lifted water from mature trees to young saplings in xeric soil conditions.  The extensive mycelial networks formed by these fungi could help ameliorate the negative impacts of site water stress on tree growth and regeneration, which is predicted to increase in severity with climate change.  Additional information regarding the phenology of Rhizopogon spp. fungi is needed to clarify their role in promoting tree resilience to inra- and inter-annual water stress.      Mature interior Douglas-fir trees can develop extensive lateral roots and a large degree of overlap occurs between the root systems of neighboring trees.  Measures of rooting extents in this study were anecdotal because only roots in association with Rhizopogon spp. mycorrhizas identified with multi-locus microsatellite DNA analysis were measured.  Nonetheless, this data provides a proxy for modeling relationships between tree age/bole size and lateral root extents with minimal disruption.  The maximum distance between mycorrhizal roots and source boles observed was 23 m (chapters 3 & 4), which to the best of my knowledge is the greatest span reported for Douglas-fir (var. menziesii or glauca) lateral roots (Stone & Kalisz, 1991); although studies reporting root topological lengths or biomass have indicated the frequent occurrence of long, rope-like laterals (Hengst, 1958; McMinn, 1963; Eis, 1974; Fogel, 1983; Eis, 1987; Kuiper & Coutts, 1992).  The lateral root extents observed were in the range of grafted roots reported by Lanner (1961) in P. menziesii var. menziesii forests (50 m combined distance between living stumps and ?host? boles); and the distance between tree boles and EMF samples matched using microsatellite DNA analysis by Saari et al. (2005) in a Pinus sylvestris L. var. scotica forest (18 m span).  Extended lateral root growth could explain the persistence of EMF species found ? 16 m distance from forest edges following clear-cut harvests in interior Douglas-fir (Hagerman et al., 1999) and coastal Douglas-fir (Cline et al., 2005; Cline et al., 2007) forests with no alternate host plants present.  The extent of Douglas-fir root systems is considered to be generally proportional to crown radius (Hermann & Lavender, 1990; Roering et al., 2003), but slope-dependent asymmetries and long, rope-like lateral roots are common and notable exceptions (Hengst, 1958; McMinn, 1963; Smith, 1964; Eis, 1974; Eis, 1987; Kuiper & Coutts, 1992).  The potential importance of far-reaching lateral roots for MN associations and resource acquisition should not be overlooked.  1175.4 Future research directions 5.4.1 Cryptic species coexistence In chapter 2, mechanisms leading to spatial niche differentiation between the sister species R. vesiculosus and R. vinicolor were inferred based on the fine-scaled spatial patterns of their mycelia within replicated plots.  Competitive dynamics between EMF species are inherently difficult to measure; they can be influenced by many interacting abiotic and biotic factors which change over space and time (Bowen, 1994; Kennedy et al., 2007; Kennedy, 2010).  It is intriguing that genets of R. vesiculosus and R. vinicolor are consistently encountered at similar frequencies across space despite their significantly different spatial extents (Kretzer et al., 2004; Kretzer et al., 2005; chapters 3 & 4 this dissertation).  Perhaps a valid hypothesis to be tested is whether either of these species ever occurs without the other in natural conditions.  Interactions between these species could be clarified with experiments designed to identify direct or indirect mechanisms (facilitation, interference or exploitative competition, 3rd party associations).  Another avenue for further research relevant to the role R. vesiculosus/R. vinicolor play in interior Douglas-fir resource acquisition is to elucidate whether the spatial partitioning observed between these species corresponds with the materials on which they forage.  For this, the comparison of stable C and N isotope signatures between Rhizopogon spp. mycelia and different soil layers/substrates could be particularly insightful (Hobbie, 2005; E. Hobbie, pers. comm.).  5.4.2 Mycorrhizal network structure and phenology For logistical reasons, studies investigating MN characteristics are universally limited in scope.  As discussed throughout this manuscript, there remains much to be learned regarding the architecture of MNs and their variability across scales of space, time, and complexity.  In particular, information regarding the temporal dynamics of MNs has not kept pace with investigations into their architecture and functioning, yet is needed to infer the ecological significance of these traits.  The following are several key knowledge gaps concerning MNs formed between R. vesiculosus/R. vinicolor fungi and P. menziesii trees; these are not intended to be all-inclusive: (1) the longevity of individual Rhizopogon spp. tubercles; (2) the maximum age and spatial extents reached by Rhizopogon spp. genets; (3) the maximum MN connectivity of interior Douglas-fir trees estimated with systematic sampling (4) contrasts  118between MN architectures in P. menziesii var. glauca versus P. menziesii var. menziesii Douglas-fir forests; and (5) the contribution of Rhizopogon spp. towards tree root colonization and MN connectivity in xeric versus mesic soils relative to other EMF species present (i.e., changes to MN architecture and component strengths in a EMF community context).  Additional experimental studies assessing MN structure and functioning before and after disturbance or selective harvesting are needed to test inferences made in Chapters 3 and 4 of this dissertation pertaining to the effects of losing large trees and fungal genets on MN connectivity and functioning, and potential corresponding effects on plant and fungal community dynamics.  These studies should include varying selective harvest levels and retention patterns (e.g., basal area and/or spatial clustering of retained trees) to help elucidate the disturbance thresholds of MN patterns and processes.    5.4.3 Mycorrhizal networking within individual tuberculate mycorrhizas As a negative control during the molecular processing of Rhizopogon spp. tubercles, 5-7 root tips and one portion of hyphal ?rind? were subsampled from six tubercles per species to test the hypothesis that tubercles may be formed by R. vesiculosus and R. vinicolor concurrently (thus confirming that root tips sampled from tubercles are representative of genet locations).  Samples were selected from locations where R. vesiculosus and R. vinicolor were found to spatially overlap, but without exception subsamples matched a single R. vesiculosus or R. vinicolor genotype.  Surprisingly, one R. vinicolor tubercle contained two tree root genotypes with subsample replication.  Because of this finding, three root tips were occasionally subsampled from randomly selected tubercles during subsequent molecular processing (ultimately, 24 of these tubercles corresponded to R. vesiculosus and 5 to R. vinicolor) and of these, one tubercle from each species again contained two root genotypes.  In this case, each tubercle contained a tree genotype that was not matched to a bole aboveground.  For this reason and for lack of within-tubercle replication in general, these samples were all excluded from further analysis.  A study is currently underway to test the validity of these results and determine the frequency and proportions at which multiple trees may intermingle within individual tubercles.    1195.4.4 Kin selection via mycorrhizal networks The influences of MNs on regeneration facilitation or inter-tree competition following years of mast seed production (see Simard, 2009) might indirectly result in maternal genome kin selection in complex multi-cohort forests.  Kin selection refers to the selection for particular alleles that are indirectly associated with the relatedness of individuals (Hamilton, 1964; Kelly, 1996; Hartl & Clark, 1997; Queller & Strassman, 2002).  Hormonal signaling between fungal and plant cells may be genetically controlled to favour selection by kin plants of specific fungal individuals that link them in a MN.  Alternatively, fungal links could provide a direct pathway for the interplant transfer of phytochemicals involved in kin recognition between plants (e.g., Dudley & File, 2007), just as with water and nutrient transfers.   Recent studies have demonstrated that MNs play a role in seedling regeneration facilitation in interior Douglas-fir forests (Teste & Simard, 2008; Teste et al., 2009; Teste et al., 2010; Bingham, 2011).  However, these experiments were conducted using nursery and seed grown seedlings or transplants from off-site.  Solitary cone-producing trees (in seed-tree retention plots following harvest or ?tree islands? along the forest-grassland interface) could be used to test the ?kinship effect?, where seeds that are related or not related are planted in the vicinity of the parent tree.  This approach has the advantage of being easily incorporated into factorial designs used in previous studies testing MN and distance effects on regeneration facilitation (Teste et al., 2009; Bingham, 2011).  There is still a paucity of information regarding the relative contribution of MNs towards inter-tree competition versus facilitation, and even less regarding interactions with space, time, and endogenous or exogenous processes (e.g., shifts in competition:facilitation related to stand spatial patterns, complexity, or seral stages, EMF community composition and diversity, natural or artificial disturbances, and other climatic and edaphic factors).  Testing these hypotheses under natural conditions poses many challenges, but the ?kinship effect? could be included as a factor in long-term experiments seeking to quantify MN competition:facilitation effects over time.      120Bibliography   Agerer, R. 2001. Exploration types of ectomycorrhizae. Mycorrhiza 11(2): 107-114. Agerer, R. 2006. Fungal relationships and structural identity of their ectomycorrhizae. Mycological Progress 5(2): 67-107. Agerer, R, A G?ttlein. 2003. Correlations between projection area of ectomycorrhizae and H2O extractable nutrients in organic soil layers. Mycological Progress 2(1): 45-51. Aitken, SN, S Yeaman, JA Holliday, T Wang, S Curtis-McLane. 2008. Adaptation, migration or extirpation: climate change outcomes for tree populations. Evolutionary Applications 1(1): 95-111. Albert, R, J Hawoong, A-L Barab?si. 2000. Error and attack tolerance of complex networks. Nature 406(6794). Alexander, RR. 1988. Forest vegetation on national forests in the Rocky Mountain and Intermountain regions: habitat types and community types. General Technical Report RM-162, 162, UF Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO, 47 p. Allen, MF. 2007. Mycorrhizal fungi: highways for water and nutrients in arid soils. Vadose Zone Journal 6(2): 291-297. Allen, MF, EB Allen 1992. Mycorrhizae and plant community development: mechanisms and patterns. In: GC Carroll, DT Wicklow, eds. The Fungal Community: Its Organization and Role in the Ecosystem. Marcel Dekker, New York, NY. 455-480. Allen, MF, W Swenson, JI Querejeta, LM Egerton-Warburton, KK Treseder. 2003. Ecology of mycorrhizae: a conceptual framework for complex interactions among plants and fungi. Annual Review of Phytopathology 41(1): 271-303. Almaas, E, A-L Barab?si 2006. Power laws in biological networks. In: EV Koonin, YI Wolf, GP Karev, eds. Power laws, scale-free networks and genome biology. Springer US. 1-11. Amaranthus, M, D Perry. 1994. The functioning of ectomycorrhizal fungi in the field: linkages in space and time. Plant and Soil 159(1): 133-140. Amaranthus, MP 1992. Mycorrhizas, forest disturbance and regeneration in the Pacific Northwestern United States. In: DJ Read, DH Lewis, AH Fitter, IJ Alexander, eds. Mycorrhizas in ecosystems. Cambridge University Press, Cambridge, UK. 202-207. Amaranthus, MP, DS Parrish, DA Perry. 1989. Decaying logs as moisture reservoirs after drought and wildfire. In: I Alexander, ed. Watershed 89. USDA-FS Region 10, Juneau, AK. 191-194.  Amarasekare, P. 2003. Competitive coexistence in spatially structured environments: a synthesis. Ecology Letters 6(12): 1109-1122. Andrews, JH 1992. Fungal life-history strategies. In: GC Carroll, DT Wicklow, eds. The Fungal Community: Its Organization and Role in the Ecosystem. Marcel Dekker, New York, NY. 119-146. Ashkannejhad, S, TR Horton. 2006. Ectomycorrhizal ecology under primary succession on coastal sand dunes: interactions involving Pinus contorta, suilloid fungi and deer. New Phytologist 169(2): 345-354. Aubry, KB, MP Amaranthus, CB Halpern, JD White, BL Woodard, CE Peterson, . . . AJ Horton. 1999. Evaluating the effects of varying levels and patterns of green-tree  121retention: Exerimental design of the DEMO study. Northwest Science 73(Special Issue): 12-26. Avis, PG, DJ McLaughlin, BC Dentinger, PB Reich. 2003. Long-term increase in nitrogen supply alters above- and below-ground ectomycorrhizal communities and increases the dominance of Russula spp. in a temperate oak savanna. New Phytologist 160(1): 239-253. Bai, S-L, G-L Li, Y Liu, R Kasten Dumroese, R-H Lv. 2009. Ostryopsis davidiana seedlings inoculated with ectomycorrhizal fungi facilitate formation of mycorrhizae on Pinus tabulaeformis seedlings. Mycorrhiza 19(6): 425-434. Baier, R, J Ingenhaag, H Blaschke, A G?ttlein, R Agerer. 2006. Vertical distribution of an ectomycorrhizal community in upper soil horizons of a young Norway Spruce (Picea abies [L.] Karst.) stand of the Bavarian Limestone Alps. Mycorrhiza 16: 197-206. Bailey, JD, JC Tappeiner. 1998. Effects of thinning on structural development in 40- to 100-year-old Douglas-fir stands in western Oregon. Forest Ecology and Management 108(1-2): 99-113. Barab?si, A-L. 2009. Scale-free networks: a decade and beyond. Science 325(5939): 412-413. Barab?si, A-L, R Albert. 1999. Emergence of scaling in random networks. Science 286: 509-512. Barrat, A, M Barthalemy, R Pastor-Satorras, A Vespignani. 2004. The architecture of complex weighted networks. Proceedings of the National Academy of Sciences of the United States of America 101(11): 3747-3752. Bascompte, J. 2009a. Disentangling the web of life. Science 325(5939): 416-419. Bascompte, J. 2009b. Mutualistic networks. Frontiers in Ecology and the Environment 7(8): 429-436. Bascompte, J. 2010. Structure and dynamics of ecological networks. Science 329(765). Bascompte, J, P Jordano. 2007. Plant-animal mutualistic networks: the architecture of biodiversity. Annual Review of Ecology, Evolution, and Systematics 38(1): 567-593. Bascompte, J, P Jordano, CJ Melian, JM Olesen. 2003. The nested assembly of plant-animal mutualistic networks. Proceedings of the National Academy of Sciences of the United States of America 100(16): 9383-9387. Bastolla, U, MA Fortuna, A Pascual-Garcia, A Ferrera, B Luque, J Bascompte. 2009. The architecture of mutualistic networks minimizes competition and increases biodiversity. Nature 458(7241): 1018-1020. Batagelj, V, A Mvar. 2011. Pajek: program for analysis and visualization of large networks. Ljublijana, Slovenia. Bebber, DP, J Hynes, PR Darrah, L Boddy, MD Fricker. 2007. Biological solutions to transport network design. Proceedings of the Royal Society B: Biological Sciences 274(1623): 2307-2315. Beiler, KJ, DM Durall, SW Simard, SA Maxwell, AM Kretzer. 2010. Architecture of the wood-wide web: Rhizopogon spp. genets link multiple Douglas-fir cohorts. New Phytologist 185(2): 543-553. Bending, GD, DJ Read. 1995. The structure and function of the vegetative mycelium of ectomycorrhizal plants. VI. Activities of nutrient mobilizing enzymes in birch litter colonized by Paxillus involutus (Fr.) Fr. New Phytologist 130(3): 411-417.  122Bergemann, SE, SL Miller. 2002. Size, distribution, and persistence of genets in local populations of the late-stage ectomycorrhizal basidiomycete, Russula brevipes. New Phytologist 156: 313-320. Bertness, MD, R Callaway. 1994. Positive interactions in communities. Trends in Ecology & Evolution 9(5): 191-193. Bever, JD. 2002. Negative feedback within a mutualism: host-specific growth of mycorrhizal fungi reduces plant benefit. Proceedings of the Royal Society of London. Series B: Biological Sciences 269(1509): 2595-2601. Bever, JD, KM Westover, J Antonovics. 1997. Incorporating the soil community into plant population dynamics: the utility of the feedback approach. Journal of Ecology 85(5): 561-573. Bingham, MA. 2011. The role of ectomycorrhizal networks in plant-to-plant facilitation across a climatic moisture regime. University of British Columbia, Vancouver. Blackwell, M. 2000. Terrestrial life--fungal from the start? Science 289(5486): 1884-1885. Boddy, L, J Wood, E Redman, J Hynes, MD Fricker. 2010. Fungal network responses to grazing. Fungal Genetics and Biology 47(6): 522-530. Bolker, BM, SW Pacala. 1999. Spatial moment equations for plant competition: understanding spatial strategies and the advantages of short dispersal. The American naturalist 153(6): 575-602. Bonello, P, TD Bruns, M Gardes. 1998. Genetic structure of a natural population of the ectomycorrhizal fungus Suillus pungens. New Phytologist 138: 533-542. Bonfante, P, I-A Anca. 2009. Plants, mycorrhizal fungi, and bacteria: a network of interactions. Annual Review of Microbiology 63(1): 363-383. Booth, MG. 2004. Mycorrhizal networks mediate overstorey-understorey competition in a temperate forest. Ecology Letters 7(7): 538-546. Bowen, G. 1994. The ecology of ectomycorrhiza formation and functioning. Plant and Soil 159(1): 61-67. Bowen, GD 1968. The roles of mycorrhizae and root nodules in tree nutrition. In:  Collected Papers Forest. Sch. Univ. New England, Armidale, New South Wales. Bray, D. 2003. Molecular networks: the top-down view. Science 301: 1864-1865. Brooker, RW, FT Maestre, RM Callaway, CL Lortie, LA Cavieres, G Kunstler, . . . R Michalet. 2008. Facilitation in plant communities: the past, the present, and the future. Journal of Ecology 96(1): 18-34. Brooks, JR, FC Meinzer, JM Warren, J-C Domec, ROB Coulombe. 2006. Hydraulic redistribution in a Douglas-fir forest: lessons from system manipulations. Plant, Cell & Environment 29(1): 138-150. Brown, JH, VK Gupta, B-L Li, BT Milne, C Restrepo, GB West. 2002. The fractal nature of nature: power laws, ecological complexity and biodiversity. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 357(1421): 619-626. Brownlee, C, J Duddridge, A Malibari, D Read. 1983. The structure and function of mycelial systems of ectomycorrhizal roots with special reference to their role in forming inter-plant connections and providing pathways for assimilate and water transport. Plant and Soil 71(1): 433-443. Bruns, T. 1995. Thoughts on the processes that maintain local species diversity of ectomycorrhizal fungi. Plant and Soil 170(1): 63-73.  123Bruns, TD, MI Bidartondo, DL Taylor. 2002. Host specificity in ectomycorrhizal communities: what do the exceptions tell us? Integrative and Comparative Biology 42(2): 352-359. Bruns, TD, KG Peay, PJ Boynton, LC Grubisha, NA Hynson, NH Nguyen, NP Rosenstock. 2009. Inoculum potential of Rhizopogon spores increases with time over the first 4 yr of a 99-yr spore burial experiment. The New phytologist 181(2): 463-470. Cairney, JWG, RM Burke. 1996. Physiological heterogeneity within fungal mycelia: an important concept for a functional understanding of the ectomycorrhizal symbiosis. New Phytologist 134(4): 685-695. Callaway, R. 1995. Positive interactions among plants. The Botanical Review 61(4): 306-349. Callaway, RM, RW Brooker, P Choler, Z Kikvidze, CJ Lortie, R Michalet, . . . BJ Cook. 2002. Positive interactions among alpine plants increase with stress. Nature 417(6891): 844-848. Case, MJ, DL Peterson. 2005. Fine-scale variability in growthclimate relationships of Douglas-fir, North Cascade Range, Washington. Canadian Journal of Forest Research 35: 2743-2755. Castellano, M, A., J Trappe, M. 1985. Ectomycorrhizal formation and plantation performance of Douglas-fir nursery stock inoculated with Rhizopogon spores. Canadian Journal of Forest Research 15(4). Castellano, MA 1996. Outplanting performance of mycorrhizal inoculated seedlings. In: KG Mukerji, ed. Concepts in mycorrhizal research: handbook of vegetation science. Kluwer, Dordrecht, Netherlands. 223-301. Castellano, MA, JM Trappe. 1985. Mycorrhizal associations of five species of Monotropoideae in Oregon. Mycologia 77(3): 499-502. Cavender-Bares, J, A Izzo, R Robinson, C Lovelock. 2009. Changes in ectomycorrhizal community structure on two containerized oak hosts across an experimental hydrologic gradient. Mycorrhiza 19(3): 133-142. C?zares, E, JE Smith. 1996. Occurrence of vesicular-arbuscular mycorrhizae in Pseudotsuga menziesii and Tsuga heterophylla seedlings grown in Oregon Coast Range soils. Mycorrhiza 6(1): 65-67. Chamberlain, SA, JN Holland. 2009. Body size predicts degree in ant?plant mutualistic networks. Functional Ecology 23(1): 196-202. Chu-Chou, M, LJ Grace. 1981. Mycorrhizal fungi of Pseudotsuga menziesii in the north island of New Zealand. Soil Biology & Biochemistry 13: 247-249. Claridge, AW, JM Trappe, SJ Cork, DL Claridge. 1999. Mycophagy by small mammals in the coniferous forests of North America: nutritional value of sporocarps of Rhizopogon vinicolor, a common hypogeous fungus. Journal of Comparative Physiology B: Biochemical, Systemic, and Environmental Physiology 169(3): 172-178. Cline, E, B Vinyard, R Edmonds. 2007. Spatial effects of retention trees on mycorrhizas and biomass of Douglas-fir seedlings. Canadian Journal of Forest Research 37: 430-438.  124Cline, ET, JF Ammirati, RL Edmonds. 2005. Does proximity to mature trees influence ectomycorrhizal fungus communities of Douglas-fir seedlings? New Phytologist 166(3): 993-1009. Colgan III, W, AW Claridge. 2002. Mycorrhizal effectiveness of Rhizopogon spores recovered from faecal pellets of small forest-dwelling mammals. Mycological Research 106(3): 314-320. Colpaert, JV, JA Van Assche, K Luijtens. 1992. The growth of the extramatrical mycelium of ectomycorrhizal fungi and the growth response of Pinus sylvestris L. New Phytologist 120(1): 127-135. Courty, P-E, A Franc, J-C Pierrat, J Garbaye. 2008. Temporal changes in the ectomycorrhizal community in two soil horizons of a temperate oak forest. Applied Environmental Microbiology 74(18): 5792-5801. Crona, B, K Hubacek. 2010. The right connections: how do social networks lubricate the machinery of natural resource governance? Ecology and Society 15(4). Culver, DC 1992. Introduction to the theory of species interactions. In: GC Carroll, DT Wicklow, eds. The Fungal Community: Its Organization and Role in the Ecosystem. Marcel Dekker, New York, NY. 229-242. Curtis, RO. 1998. "Selective cutting" in Douglas-fir: history revisited. Journal of Forestry 96: 40-46. Dahlberg, A. 1997. Population ecology of Suillus variegatus in old Swedish Scots pine forests. Mycological Research 101(1): 47-54. Dahlberg, A. 2001. Community ecology of ectomycorrhizal fungi: an advancing interdisciplinary field. New Phytologist 150(3): 555-562. Dahlberg, A, J Stenlid. 1994. Size, distribution and biomass of genets in populations of Suillus hovinus (L.: Fr.) Roussel revealed by somatic incompatibility. New Phytologist 128: 225-234. Dale, MRT, MJ Fortin. 2010. From graphs to spatial graphs. Annual Review of Ecology, Evolution, and Systematics 41(1): 21-38. Dawson, TE. 1993. Hydraulic lift and water use by plants: implications for water balance, performance and plant-plant interactions. Oecologia 95(4): 565-574. Dayton, PK 1972. Toward an understanding of community resilience and the potential effects of enrichments to the benthos at McMurdo Sound. In: BC Parker, ed. Proceedings of the colloquium on conservation problems in Antarctica. Allen Press, Lawrence, KS. de Nooy, W, A Mrvar, V Batagelj. 2005. Exploratory social network analysis with pajek. Cambridge University Press, New York, NY. Dehlin, H, DA Peltzer, VJ Allison, GW Yeates, M-C Nilsson, DA Wardle. 2008. Tree seedling performance and below-ground properties in stands of invasive and native tree species. New Zealand Journal of Ecology 32(1): 67-79. Deslippe, JR, SW Simard. 2011. Below-ground carbon transfer among Betula nana may increase with warming in Arctic tundra. New Phytologist: no-no. Dickie, IA, RT Koide, KC Steiner. 2002a. Influences of established trees on mycorrhizas, nutrition, and growth of Quercus rubra seedlings. Ecological Monographs 72(4): 505-521. Dickie, IA, B Xu, RT Koide. 2002b. Vertical niche differentiation of ectomycorrhizal hyphae in soil as shown by T-RFLP analysis. New Phytologist 156(3): 527-535.  125Dosskey, MG, L Boersma, RG Linderman. 1991. Role for the photosynthate demand of ectomycorrhizas in the response of Douglas fir seedlings to drying soil. New Phytologist 117(2): 8. Dosskey, MG, RG Linderman, L Boersma. 1990. Carbon-sink stimulation of photosynthesis in Douglas fir seedlings by some ectomycorrhizas. New Phytologist 115(2): 269-274. Duddridge, JA, A Malibari, DJ Read. 1980. Structure and function of mycorrhizal rhizomorphs with special reference to their role in water transport. Nature 287: 834-836. Dudley, SA, AL File. 2007. Kin recognition in an annual plant. Biology Letters 3(4): 435-438. Dungan, JL, JN Perry, MRT Dale, P Legendre, S Citron-Pousty, MJ Fortin, . . . MS Rosenberg. 2002. A balanced view of scale in spatial statistical analysis. Ecography 25(5): 626-640. Dunham, S, K-H Larsson, J Spatafora. 2007. Species richness and community composition of mat-forming ectomycorrhizal fungi in old- and second-growth Douglas-fir forests of the HJ Andrews Experimental Forest, Oregon, USA. Mycorrhiza 17(8): 633-645. Dunham, SM, A Kretzer, ME Pfrender. 2003. Characterization of Pacific golden chanterelle (Cantharellus formosus) genet size using co-dominant microsatellite markers. Molecular Ecology 12: 1607-1618. Dunne, JA, RJ Williams, ND Martinez. 2002. Network structure and biodiversity loss in food webs: robustness increases with connectance. Ecology Letters 5(4): 558-567. Durall, DM, JD Marshall, MD Jones, R Crawford, JM Trappe. 1994a. Morphological changes and photosynthate allocation in ageing Hebeloma crustuliniforme (Bull.) Quel. and Laccaria bicolor (Maire) Orton mycorrhizas of Pinus ponderosa Dougl. ex. Laws. New Phytologist 127(4): 719-724. Durall, DM, AW Todd, JM Trappe. 1994b. Decomposition of 14C-labelled substrates by ectomycorrhizal fungi in association with Douglas fir. New Phytologist 127(4): 725-729. Egerton-Warburton, LM, JI Querejeta, MF Allen. 2007. Common mycorrhizal networks provide a potential pathway for the transfer of hydraulically lifted water between plants. Journal of Experimental Botany 58(6): 1473-1483. Ehrlich, PR, AH Ehrlich. 1981. Extinction: the causes and consequences of the disappearance of species. Random House, NY, New York. Eis, S. 1974. Root system morphology of western hemlock, western red cedar, and Douglas-fir. Canadian Journal of Forest Research 4(1): 28-38. Eis, S. 1987. Root systems of older immature hemlock, cedar, and Douglas-fir. Canadian Journal of Forest Research 17(11): 1348-1354. Ellison, AM, MS Bank, BD Clinton, EA Colburn, K Elliott, CR Ford, . . . JR Webster. 2005. Loss of foundation species: consequences for the structure and dynamics of forested ecosystems. Frontiers in Ecology and the Environment 3(9): 479-486. Entry, JA, CL Rose, K Cromack. 1991. Microbial biomass and nutrient concentrations in hyphal mats of the ectomycorrhizal fungus Hysterangium setchellii in a coniferous forest soil. Soil Biology and Biochemistry 24(5): 447-453.  126Erland, S, AFS Taylor 2002. Diversity of ectomycorrhizal fungal communities in relation to the abiotic environment. In: MGA van der Heijden, IR Sanders, eds. Mycorrhizal Ecology. Springer-Verglag Berlin, Heidelberg, Germany. 163-200. Espeland, EK, KJ Rice. 2007. Facilitation across stress gradients: the importance of local adaptation. Ecology 88(9): 2404-2409. Ettema, CH, DA Wardle. 2002. Spatial soil ecology. Trends in Ecology & Evolution 17(4): 177-183. Farrar, JL. 1995. Trees in Canada. Canadian Forest Service, Markham, Ontario. Ferguson, DE, CE Carlson. 1991. Natural regeneration of interior douglas-fir in the northern Rocky Mountains. In: DM Baumgartner, JE Lotan, eds. Interior Douglas-Fir: The Species and Its Management. Washington State University, Spokane, WA. 239-246.  Finlay, RD, DJ Read. 1986a. The structure and function of the vegetative mycelium of ectomycorrhizal plants. II. The uptake and distribution of phosphorus by mycelial strands interconnecting host plants. New Phytologist 103(1): 157-165. Finlay, RD, DJ Read. 1986b. The structure and function of the vegetative mycelium of ectomyocrrhizal plants I. Translocation of 14C-labelled carbon between plants interconnected by a common mycelium. New Phytologist 103(1): 143-156. Fleming, L. 1983. Succession of mycorrhizal fungi on birch: infection of seedlings planted around mature trees. Plant and Soil 71(1): 263-267. Fleming, LV. 1984. Effects of soil trenching and coring on the formation of ectomycorrhizas on birch seedlings grown around mature trees. New Phytologist 98(1): 143-153. Flores, J, E Jurado. 2003. Are nurse-prot?g? interactions more common among plants from arid environments? Journal of Vegetation Science 14(6): 911-916. Fogel, R. 1983. Root turnover and productivity of coniferous forests. Plant and Soil 71(1): 75-85. Fogel, R, G Hunt. 1979. Fungal and arboreal biomass in a western Oregon Douglas-fir ecosystem: distribution patterns and turnover. Canadian Journal of Forest Research 9(2): 245-256. Fortin, M, MRT Dale. 2005. Spatial analysis: a guide for ecologists. Cambridge University Press, Cambridge, UK. Francis, R, D Read. 1994. The contributions of mycorrhizal fungi to the determination of plant community structure. Plant and Soil 159(1): 11-25. Francis, R, DJ Read. 1995. Mutualism and antagonism in the mycorrhizal symbiosis, with special reference to impacts on plant community structure. Canadian Journal of Botany 73(S1): 1301-1309. Frank, AB. 1885. ?ber die auf W?rzelsymbiose beruhende Ehrn?hrung gewisser B?um durch unterirdische Pilze. Berichte der Deutschen Botanischen Gesellschaft 3: 128-145. Franklin, JF, DR Berg, DA Thornburgh, JC Tappeiner 1997. Alternative silvicultrual approaches to timber harvesting: variable retention harvest systems. In: KA Kohm, JF Franklin, eds. Creating a forestry for the 21st century: the science of ecosystem management. Island Press, Washington, D.C. 111-139. Franklin, JF, TA Spies, RV Pelt, AB Carey, DA Thornburgh, DR Berg, . . . J Chen. 2002. Disturbances and structural development of natural forest ecosystems with  127silvicultural implications, using Douglas-fir forests as an example. Forest Ecology and Management 155(1-3): 399-423. Fransson, PMA, AFS Taylor, RD Finlay. 2001. Elevated atmospheric CO2 alters root symbiont community structure in forest trees. New Phytologist 152(3): 431-442. Freinkel, S. 2007. American Chestnut. University of California Press, Berkeley, CA. Fricker, MD, D Bebber, L Boddy 2008. Mycelial networks: structure and dynamics. In: L Boddy, JC Frankland, P van West, eds. Ecology of Saprotrophic Basidiomycetes. Elsevier Ltd., London, UK. 3-18. Gadgil, RL, PD Gadgil. 1971. Mycorrhiza and litter decomposition. Nature 233(5315): 133-133. Gardes, M, TD Bruns. 1996. Community structure of ectomycorrhizal fungi in a Pinus muricata forest: above- and below-ground views. Canadian Journal of Botany 74(10): 12. Genney, DR, IC Anderson, IJ Alexander. 2006. Fine-scale distribution of pine ectomycorrhizas and their extramatrical mycelium. New Phytologist 170: 381-390. Gherbi, H, C Delaruelle, M-A Selosse, F Martin. 1999. High genetic diversity in a population of the ectomycorrhizal basidiomycete Laccaria amethystine in a 150-year-old beech forest. Molecular Ecology 8: 2003-2013. Godbold, D, M Hoosbeek, M Lukac, M Cotrufo, I Janssens, R Ceulemans, . . . A Peressotti. 2006. Mycorrhizal hyphal turnover as a dominant process for carbon input into soil organic matter. Plant and Soil 281(1): 15-24. Goodman, D. 1975. The theory of diversity-stability relationships in ecology. The Quarterly Review of Biology 50(3): 237-266. Goodman, DM, JA Trofymow. 1998. Distribution of ectomycorrhizas in micro-habitats in mature and old-growth stands of Douglas-fir on southeastern Vancouver Island. Soil Biology and Biochemistry 30(14): 2127-2138. Green, RN, RL Trowbridge, K Klinka. 1993. Towards a taxonomic classification of humus forms. Forest Science 39(1): a0001-z0002. Griesbauer, H. 2008. Regional, ecological and temporal patterns in Douglas-fir climate-growth relationships in the British Columbia Interior. University of Northern British Columbia, Prince George, Canada. Griesbauer, H, SD Green. 2010. Assessing the climatic sensitivity of Douglas-fir at its northern range margins in British Columbia, Canada. Trees - Structure and Function 24(2): 375-389. Griesbauer, HP, DS Green, GA O'Neill. 2011. Using a spatiotemporal climate model to assess population-level Douglas-fir growth sensitivity to climate change across large climatic gradients in British Columbia, Canada. Forest Ecology and Management 261(3): 589-600. Grime, JP. 1998. Benefits of plant diversity to ecosystems: immediate, filter and founder effects. Journal of Ecology 86(6): 902-910. Grubisha, LC, JM Trappe, R Molina, JW Spatafora. 2002. Biology of the ectomycorrhizal genus Rhizopogon. VI. Re-examination of infrageneric relationships inferred from phylogenetic analyses of ITS sequences. Mycologia 94(4): 607-619. Gugger, PF, S Sugita, J Cavender-Bares. 2010. Phylogeography of Douglas-fir based on mitochondrial and chloroplast DNA sequences: testing hypotheses from the fossil record. Molecular Ecology 19(9): 1877-1897.  128Guidot, A, J-C Debaud, A Effosse, R Marmeisse. 2004. Below-ground distribution and persistence of an ectomycorrhizal fungus. New Phytologist 161(2): 539-547. Guidot, A, JC Debaud, R Marmeisse. 2001. Correspondence between genet diversity and spatial distribution of above- and below-ground populations of the ectomycorrhizal fungus Hebeloma cylindrosporum. Molecular Ecology 10(5): 1121-1131. Hagerman, SM, D Durall. 2004. Ectomycorrhizal colonization of greenhouse-grown Douglas-fir (Pseudotsuga menziesii) seedlings by inoculum associated with the roots of refuge plants sampled from a Douglas-fir forest in the southern interior of British Columbia. Canadian Journal of Botany 82(6): 742-751. Hagerman, SM, MD Jones, GE Bradfield, M Gillespie, DM Durall. 1999. Effects of clear-cut logging on the diversity and persistence of ectomycorrhizae at a subalpine forest. Canadian Journal of Forest Research 29(1). Halpern, CB, D McKenzie, SA Evans, DA Maguire. 2005. Initial responses of forest understories to varying levels and patterns of green-tree retention. Ecological Applications 15(1): 175-195. Hamann, A, T Wang. 2006. Potential effects of climate change on ecosystem and tree species distribution in British Columbia. Ecology 87(11): 2773-2786. Hamilton, WD. 1964. The genetical evolution of social behaviour. I. Journal of theoretical biology 7(1): 1-16. Harrington, CA, DL Reukema. 1983. Initial shock and long-term stand development following thinning in a Douglas-fir plantation. Forest Science 29: 33-46. Hartl, DL, AG Clark. 1997. Principles of population genetics. Sinauer Associates Inc., Sunderland, MA. Harvey, AE, MJ Larsen, MF Jurgensen. 1979. Comparative distribution of ectomycorrhizae in soils of three western Montana forest habitat types. Forest Science 25: 350-358. Haystead, A, N Malajczuk, TS Grove. 1988. Underground transfer of nitrogen between pasture plants infected with vesicular-arbuscular mycorrhizal fungi. New Phytologist 108(4): 417-423. He, XO, C Critchley, H Ng, CS Bledsoe. 2004. Reciprocal N (15NH4+ or 15NO3?) transfer between non-N2-fixing Eucalyptus maculata and N2-fixing Casuarina cunninghamiana linked by the ectomycorrhizal fungus Pisolithus sp. New Phytologist 163: 629-640. Hengst, E. 1958. Wurzelstockuntersuchungen an der Douglasie. Arch. Forstwes. 7: 338-351. Hermann, RK. 1982. The genus Pseudotsuga: historical records and nomenclature. 2a, Forest Research Laboratory, Oregon State University, Corvallis, OR,  Hermann, RK, DP Lavender 1990. Pseudotsuga menziesii. In: RM Burns, BH Honkala, eds. Silvics of North America Volume 1: Conifers. USDA Forest Service, Washington. Hewitt, E Judi, Thrush, F Simon, Dayton, K Paul, . . . Erik. 2007. The effect of spatial and temporal heterogeneity on the design and analysis of empirical studies of scale-dependent systems. University of Chicago Press, Chicago, IL, ETATS-UNIS. Hobbie, E 2005. Using isotopic tracers to follow carbon and nitrogen cycling of fungi. In: J Dighton, P Oudemans, J White, eds. The Fungal Community. Marcel Dekker, New York, NY. 361-381.  129Hobbie, E, R Agerer. 2010. Nitrogen isotopes in ectomycorrhizal sporocarps correspond to belowground exploration types. Plant and Soil 327(1): 71-83. Hoeksema, JD, JN Thompson. 2007. Geographic structure in a widespread plant?mycorrhizal interaction: pines and false truffles. Journal of Evolutionary Biology 20(3): 1148-1163. H?gberg, MN, P H?gberg. 2002. Extramatrical ectomycorrhizal mycelium contributes one-third of microbial biomass and produces, together with associated roots, half the dissolved organic carbon in a forest soil. New Phytologist 154(3): 791-795. H?gberg, P, AH Plamboeck, AFS Taylor, PMA Fransson. 1999. Natural 13C abundance reveals trophic status of fungi and host-origin of carbon in mycorrhizal fungi in mixed forests. Proceedings of the National Academy of Sciences 96(15): 8534-8539. Holling, CS. 1973. Resilience and stability of ecological systems. Annual Review of Ecology and Systematics 4: 1-23. Holling, CS. 1992. Cross-scale morphology, geometry, and dynamics of ecosystems. Ecological Monographs 62(4): 447-502. Hope, GD, WR Mitchell, DA Lloyd, WR Erickson, WL Harper, BM Wikeem. 1991. Ecosystems of British Columbia: interior Douglas-fir zone. Special Report Series, Vol. 6, BCMo Forests, Canadian Forest Service, 153-166. Horton, TR, TD Bruns. 1998. Multiple-host fungi are the most frequent and abundant ectomycorrhizal types in a mixed stand of Douglas fir (Pseudotsuga menziesii) and bishop pine (Pinus muricata). New Phytologist 139(2): 331-339. Horton, TR, TD Bruns. 2001. The molecular revolution in ectomycorrhizal ecology: peeking into the black-box. Molecular Ecology 10(8): 1855-1871. Horton, TR, TD Bruns, VT Parker. 1999. Ectomycorrhizal fungi associated with Arctostaphylos contribute to Pseudotsuga menziesii establishement. Canadian Journal of Botany 77(1): 93-102. Horton, TR, R Molina, K Hood. 2005. Douglas-fir ectomycorrhizae in 40- and 400-year-old stands: mycobiont availability to late successional western hemlock. Mycorrhiza 15(6): 393-403. Horton, TR, MGA van der Heijden 2008. The role of symbiosis in seedling establishment and survival. In: MA Leck, VT Parker, RL Simpson, eds. Seedling ecology and evolution. Cambridge University Press, Cambridge, UK. 189-214. Howe, GE 1995. Genetic effects of uneven-aged management. In: KL O'Hara, ed. Uneven-aged management: opportunities, constraints and methodologies. Montana Forest and Conservation Experimental Station, Missoula, MT. Huggard, DJ, A Arsenault, A Vyse, W Klenner. 2005. The Opax mountian silvicultural systems project: preliminary results for managing complex, Dry Interior Douglas-fir forests. BCMoFa Range, Forest Sciences Program,  Hunt, GA, R Fogel. 1983. Fungal hyphal dynamics in a western Oregon Douglas-fir stand. Soil Biology and Biochemistry 15(6): 641-649. Ings, TC, JM Montoya, J Bascompte, N Bl?thgen, L Brown, CF Dormann, . . . G Woodward. 2009. Review: Ecological networks ? beyond food webs. Journal of Animal Ecology 78(1): 253-269. Isaac, LA. 1956. Place of partial cutting in old-growth stands of the Douglas-fir region. No. 16, UF Service, PNW Old Series Research Paper, 1-48.  130Ishida, TA, K Nara, T Hogetsu. 2007. Host effects on ectomycorrhizal fungal communities: insight from eight host species in mixed conifer?broadleaf forests. New Phytologist 174(2): 430-440. Izzo, A, J Agbowo, TD Bruns. 2005. Detection of plot-level changes in ectomycorrhizal communities across years in an old-growth mixed-conifer forest. New Phytologist 166(2): 619-629. Johnson, D, M Ijdo, DR Genney, IC Anderson, IJ Alexander. 2005. How do plants regulate the function, community structure, and diversity of mycorrhizal fungi? Journal of Experimental Botany 56(417): 1751-1760. Johnson, NC, JH Graham, FA Smith. 1997. Functioning of mycorrhizal associations along the mutualism-parasitism continuum. New Phytologist 135(4): 575-586. Johnson, NC, JD Hoeksema, JD Bever, VB Chaudhary, C Gehring, J Klironomos, . . . C Zabinski. 2006. From lilliput to brobdingnag: extending models of mycorrhizal function across scales. BioScience 56(11): 889-900. Jones, CG, JH Lawton, M Shachak. 1994. Organisms as ecosystem engineers. Oikos 69(3): 373-386. Jones, MD, DM Durall, JWG Cairney. 2003. Ectomycorrhizal fungal communities in young forest stands regenerating after clearcut logging. New Phytologist 157(3): 399-422. Jones, MD, DM Durall, SMK Harniman, DC Classen, SW Simard. 1997. Ectomycorrhizal diversity on Betula papyrifera and Pseudotsuga menziesii seedlings grown in the greenhouse or outplanted in single-species and mixed plots in southern British Columbia. Canadian Journal of Forest Research 27(11): 1872-1889. Jones, MD, BD Twieg, V Ward, J Barker, DM Durall, SW Simard. 2010. Functional complementarity of Douglas-fir ectomycorrhizas for extracellular enzyme activity after wildfire or clearcut logging. Functional Ecology 24(5): 1139-1151. Jonsson, L, A Dahlberg, MC Nilsson, O Karen, O Zackrisson. 1999. Continuity of ectomycorrhizal fungi in self-regenerating boreal Pinus sylvestris forests studied by comparing mycobiont diversity on seedlings and mature trees. New Phytologist 142(1): 151-162. Jordano, P. 1987. Patterns of mutualistic interactions in pollination and seed dispersal: connectance, dependence asymmetries, and coevolution. The American naturalist 129(5): 657-677. Jordano, P. 2010. Coevolution in multispecific interactions among free-living species. Evolution: Education and Outreach 3(1): 40-46. Jordano, P, J Bascompte, JM Olesen. 2003. Invariant properties in coevolutionary networks of plant?animal interactions. Ecology Letters 6(1): 69-81. Kelly, JK. 1996. Kin selection in the annual plant Impatiens capensis. The American naturalist 147(6): 899-918. Kennedy, P. 2010. Ectomycorrhizal fungi and interspecific competition: species interactions, community structure, coexistence mechanisms, and future research directions. New Phytologist 187(4): 895-910. Kennedy, P, K Peay. 2007. Different soil moisture conditions change the outcome of the ectomycorrhizal symbiosis between Rhizopogon species and Pinus muricata. Plant and Soil 291(1): 155-165.  131Kennedy, PG, SE Bergemann, S Hortal, TD Bruns. 2007. Determining the outcome of field-based competition between two Rhizopogon species using real-time PCR. Molecular Ecology 16(4): 881-890. Kennedy, PG, TD Bruns. 2005. Priority effects determine the outcome of ectomycorrhizal competition between two Rhizopogon species colonizing Pinus muricata seedlings. New Phytologist 166(2): 631-638. Kiers, TE, MGA van der Heijden. 2006. Mutualistic stability in the arbuscular mycorrhizal symbiosis: exploring hypotheses of evolutionary cooperation. Ecology 87(7): 1627-1636. Klenner, W, R Walton, A Arsenault, L Kremsater. 2008. Dry forests in the southern interior of British Columbia: historic disturbances and implications for restoration and management. Forest Ecology and Management 256(10): 1711-1722. Klironomos, J, M Zobel, M Tibbett, WD Stock, MC Rillig, JL Parrent, . . . JD Bever. 2011. Forces that structure plant communities: quantifying the importance of the mycorrhizal symbiosis. New Phytologist 189: 366-370. Koide, R, G Elliott. 1989. Cost, benefit and efficiency of the vesicular-arbuscular mycorrhizal symbiosis. Functional Ecology 3(2): 252-255. Koide, RT, DL Shumway, B Xu, JN Sharda. 2007. On temporal partitioning of a community of ectomycorrhizal fungi. New Phytologist 174(2): 420-429. Korkama, T, A Pakkanen, T Pennanen. 2006. Ectomycorrhizal community structure varies among Norway spruce (Picea abies) clones. New Phytologist 171(4): 815-824. Kottke, I, A Beiter, M Weiss, I Haug, F Oberwinkler, M Nebel. 2003. Heterobasidiomycetes form symbiotic associations with hepatics: Jungermanniales have sebacinoid mycobionts while Aneura pinguis (Metzgeriales) is associated with a Tulasnella species. Mycological Research 107(8): 957-968. Krauch, H. 1956. Management of Douglas-fir timberland in the southwest. SP USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, CO, 59 p. Krebs, CJ. 1989. Ecological Methodology. Harper & Row, New York, NY. Kretzer, AM, S Dunham, R Molina, JW Spatafora. 2004. Microsatellite markers reveal the below ground distribution of genets in two species of Rhizopogon forming tuberculate ectomycorrhizas on Douglas fir. New Phytologist 161(1): 313-320. Kretzer, AM, S Dunham, R Molina, JW Spatafora. 2005. Patterns of vegetative growth and gene flow in Rhizopogon vinicolor and R. vesiculosus (Boletales, Basidiomycota). Molecular Ecology 14(8): 2259-2268. Kretzer, AM, DL Luoma, R Molina, JW Spatafora. 2003. Taxonomy of the Rhizopogon vinicolor species complex based on analysis of ITS sequences and microsatellite loci. Mycologia 95(3): 480-487. Kuiper, LC, MP Coutts. 1992. Spatial disposition and extension of the structural root system of Douglas-fir. Forest Ecology and Management 47(1-4): 111-125. Kytoviita, MM, M Vestberg, J Tuomi. 2003. A test of mutual aid in common mycorrhizal networks: established vegetation negates benefit in seedlings. Ecology 84(4): 898-906. Landeweert, R, P Leeflang, TW Kuyper, E Hoffland, A Rosling, K Wernars, E Smit. 2003. Molecular identification of ectomycorrhizal mycelium in soil horizons. Appl. Environ. Microbiol. 69(1): 327-333.  132Langley, JA, BA Hungate. 2003. Mycorrhizal controls on belowground litter quality. Ecology 84(9): 2302-2312. Lanner, RM. 1961. Living stumps in the Sierra Nevada. Ecology 42(1): 170-173. Lau, MK, TG Whitham, LJ Lamit, NC Johnson. 2010. Ecological and evolutionary interaction network exploration: addressing the complexity of biological interactions in natural systems with community genetics and statistics. Journal of Intelligent and Fuzzy Systems 7: 19-27. Lawton, JH. 1994. What do species do in ecosystems? Oikos 71(3): 367-374. Lawton, JH, VK Brown 1993. Redundancy in ecosystems. In: E-D Schulze, HA Mooney, eds. Biodiversity and ecosystem function. Springer, Berlin. 255-270. Leake, J, D Johnson, D Donnelly, G Muckle, L Boddy, D Read. 2004. Networks of power and influence: the role of mycorrhizal mycelium in controlling plant communtities and agroecosystem functioning. Can. J. Bot. 82: 1016-1045. Legendre, P, MRT Dale, M-J Fortin, J Gurevitch, M Hohn, D Myers. 2002. The consequences of spatial structure for the design and analysis of ecological field surveys. Ecography 25(5): 601-615. LeMay, V, A Pommerening, P Marshall. 2009. Spatio-temporal structure of multi-storied, multi-aged interior Douglas fir (Pseudotsuga menziesii var. glauca) stands. Journal of Ecology 97(5): 1062-1074. Lerat, S, R Gauci, J Catford, H Vierheilig, Y Pich?, L Lapointe. 2002. 14C transfer between the spring ephemeral Erythronium americanum and sugar maple saplings via arbuscular mycorrhizal fungi in natural stands. Oecologia 132(2): 181-187. Leveau, JHJ, GM Preston. 2008. Bacterial mycophagy: definition and diagnosis of a unique bacterial?fungal interaction. New Phytologist 177(4): 859-876. Levin, SA. 1992. The problem of pattern and scale in ecology: the Robert H. MacArthur award lecture. Ecology 73(6): 1943-1967. Levin, SA. 1998. Ecosystems and the biosphere as complex adaptive systems. Ecosystems 1(5): 431-436. Levin, SA. 2005. Self-organization and the emergence of complexity in ecological systems. BioScience 55(12): 1075-1079. Li, CY, HB Massicote, LVH Moore. 1992. Nitrogen-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae. Plant and Soil 140(1): 35-40. Li, P, WT Adams. 1989. Range-wide patterns of allozyme variation in Douglas-fir (Pseudotsuga menziesii). Canadian Journal of Forest Research 19(2). Lian, C, M Narimatsu, K Nara, T Hogetsu. 2006. Tricholoma matsutake in a natural Pinus densiflora forest: correspondence between above- and below- ground genets, association with multiple host trees and alteration of existing ectomycorrhizal communities. New Phytologist 171: 825-836. Lilleskov, EA, TD Bruns, TE Dawson, FJ Camacho. 2009. Water sources and controls on water-loss rates of epigeous ectomycorrhizal fungal sporocarps during summer drought. The New phytologist 182(2): 483-494. Lilleskov, EA, TD Bruns, TR Horton, D Taylor, P Grogan. 2004. Detection of forest stand-level spatial structure in ectomycorrhizal fungal communities. FEMS microbiology ecology 49(2): 319-332. Lindstedt, SL, WA Calder, III. 1981. Body size, physiological time, and longevity of homeothermic animals. The Quarterly Review of Biology 56(1): 1-16.  133Littell, JS, DL Peterson, M Tjoelker. 2008. Douglas-fir growth in mountain ecosystems: water limites tree growth from stand to region. Ecological Monographs 78(3): 349-368. Lloyd, DA, D Angove, G Hope, C Thompson. 1990. A Guide to Site Identification and Interpretation for the Kamloops Forest Region. British Columbia Ministry of Forests, Victoria, BC. Loarie, SR, PB Duffy, H Hamilton, GP Asner, CB Field, DD Ackerly. 2009. The velocity of climate change. Nature 462(7276): 1052-1055. Lopushinsky, W. 1990. Water relations of interior Douglas-fir. In: DM Baumgartner, JE Lotan, eds. Interior Douglas-fir: the species and its management. Washington State University, Spokane, WA,USA. 35-46.  Loreau, M. 2000. Biodiversity and ecosystem functioning: recent theoretical advances. Oikos 91(1): 3-17. Lorio, PL, VK Howe, CN Martin. 1972. Loblolly pine rooting varies with microrelief on wet sites. Ecology 53(6): 1134-1140. MacArthur, R. 1955. Fluctuations of animal populations and a measure of community stability. Ecology 36(3): 533-536. MacArthur, RH, EO Wilson. 1967. The theory of island biogeography. Princeton University Press, Princeton, NJ, USA. Maestre, FT, RM Callaway, F Valladares, CJ Lortie. 2009. Refining the stress-gradient hypothesis for competition and facilitation in plant communities. Journal of Ecology 97(2): 199-205. Maguire, DA, DB Mainwaring, CB Halpern. 2006. Stand dynamics after variable-retention harvesting in mature Douglas-fir forests of Western North America. Allgemeine Forst und Jagdzeitung 177(6/7): 120-131. Martin, F, S Duplessis, F Ditengou, H Lagrange, C Voiblet, F Lapeyrie. 2001. Developmental cross talking in the ectomycorrhizal symbiosis: signals and communication genes. New Phytologist 151(1): 145-154. Maser, C, J Trappe. 1984. Seen and unseen world of the fallen tree. USDA-FS Pacific Northwest Research Station, Portland, OR,  Maser, CE, Z Maser. 1988. Interactions among squirrels, mycorrhizal fungi, and coniferous forests in Oregon. Great Basin Naturalist 48(3): 358-369. May, RM. 1977. Thresholds and breakpoints in ecosystems with a multiplicity of stable states. Nature 269(5628): 471-477. Mbogga, MS, A Hamann, T Wang. 2009. Historical and projected climate data for natural resource management in western Canada. Agricultural and Forest Meteorology 149(5): 881-890. McCann, KS. 2000. The diversity-stability debate. Nature 405(6783): 228-233. McDowell, NG, NJ Balster, JD Marshall. 2001. Belowground carbon allocation of Rocky Mountain Douglas-fir. Canadian Journal of Forest Research 31(8). McGuire, KL. 2007. Common ectomycorrhizal networks may maintain monodominance in a tropical rain forest. Ecology 88(3): 567-574. McMinn, RG. 1963. Characteristics of Douglas-fir root systems. Canadian Journal of Botany 41(1): 105-122.  134Meidinger, D, J Pojar. 1991. Ecosystems of British Columbia. Special Report Series 6, BCMo Forests,  Meinzer, FC, JM Warren, JRe Brooks. 2007. Species-specific partitioning of soil water resources in an old-growth Douglas-fir/western hemlock forest. Tree Physiology 27(6): 871-880. Miao, S, S Carstenn, M Nungesser, DPC Peters, BT Bestelmeyer, AK Knapp, . . . KM Havstad 2009. Approaches to predicting broad-scale regime shifts using changing pattern-process relationships across scales. In:  Real World Ecology. Springer New York. 47-72. Mikola, J, H Setala. 1998. Relating species diversity to ecosystem functioning: mechanistic backgrounds and experimental approach with a decomposer food web. Oikos 83(1): 180-194. Miller, M, B Emmingham. 2001. Can selection thinning convert even-age Douglas-fir stands to uneven-age structures? Western Journal of Applied Forestry 16: 35-43. Minore, D. 1979. Comparative autecological characteristics of northwestern tree species: a literature review. UF Service, Pacific Northwest Forest and Range Experiment Station, 77. Molina, R, H Massicotte, JM Trappe 1992. Specificity phenomena in mycorrhizal symbioses: community-ecological consequences and practical implications. In: MF Allen, ed. Mycorrhizal functioning: an integrative plant-fungal process. Chapman and Hall, London, U.K. 357-422. Molina, R, JM Trappe. 1994. Biology of the ectomycorrhizal genus, Rhizopogon: I.  Host associations, host-specificity and pure culture syntheses. New Phytologist 126: 653-675. Molina, R, JM Trappe, LC Grubisha, JW Spatafora 1999. Rhizopogon. In:  Ectomycorrhizal Fungi: Key Genera in Profile. Springer Verlag, Berlin. 129-160. Montoya, JM, SL Pimm, RV Sole. 2006. Ecological networks and their fragility. Nature 442(7100): 259-264. Morris, MH, MA P?rez-P?rez, ME Smith, CS Bledsoe. 2009. Influence of host species on ectomycorrhizal communities associated with two co-occurring oaks (Quercus spp.) in a tropical cloud forest. FEMS microbiology ecology 69(2): 274-287. Mueller, RC, CM Scudder, ME Porter, R Talbot Trotter, CA Gehring, TG Whitham. 2005. Differential tree mortality in response to severe drought: evidence for long-term vegetation shifts. Journal of Ecology 93(6): 1085-1093. M?ller, F. 2005. Indicating ecosystem and landscape organisation. Ecological Indicators 5(4): 280-294. Munger, TT. 1950. A look at selective cutting in Douglas-fir. Journal of Forestry 48: 97-99. Naeem, S. 1998. Species redundancy and ecosystem reliability. Conservation Biology 12(1): 39-45. Nara, K. 2006. Ectomycorrhizal networks and seedling establishment during early primary succession. New Phytologist 169: 169-178. Nara, K, T Hogetsu. 2004. Ectomycorrhizal fungi on established shrubs facilitate subsequent seedling establishment of successional plant species. Ecology 85(6): 1700-1707. Neville, J, JL Tessier, I Morrison, J Scarratt, B Canning, JN Klironomos. 2002. Soil depth distribution of ecto- and arbuscular mycorrhizal fungi associated with Populus  135tremuloides within a 3-year-old boreal forest clear-cut. Applied Soil Ecology 19(3): 209-216. Newman, MEJ. 2003. The structure and function of complex networks. SIAM Review 45(2): 167-256. Nilsson, LO, R Giesler, E B??th, H Wallander. 2005. Growth and biomass of mycorrhizal mycelia in coniferous forests along short natural nutrient gradients. New Phytologist 165(2): 613-622. Nitschke, CR, JL Innes. 2008a. Climatic change and fire potential in South-Central British Columbia, Canada. Global Change Biology 14(4): 841-855. Nitschke, CR, JL Innes. 2008b. A tree and climate assessment tool for modelling ecosystem response to climate change. Ecological Modelling 210(3): 263-277. O'Hara, KL. 2002. The historical development of uneven-aged silviculture in North America. Forestry 75(4): 339-346. Odum, EP. 1969. The strategy of ecosystem development. Science 164(3877): 262-270. Ogawa, M. 1985. Ecological characters of ectomycorrhizal fungi and their mycorrhizae: an introduction to the ecology of higher fungi. Japan Agricultural Research Quarterly 18(4): 305-314. Ohtsuki, H, C Hauert, E Lieberman, MA Nowak. 2006. A simple rule for the evolution of cooperation on graphs and social networks. Nature 441(7092): 502-505. Okuyama, T, JN Holland. 2008. Network structural properties mediate the stability of mutualistic communities. Ecology Letters 11(3): 208-216. Oliver, CD, BC Larson. 1996. Forest Stand Dynamics. John Wiley & Sons, Inc., New York. Ortega, U, M Dunabeitia, S Menendez, C Gonzalez-Murua, J Majada. 2004. Effectiveness of mycorrhizal inoculation in the nursery on growth and water relations of Pinus radiata in different water regimes. Tree Physiology 24(1): 65-73. Otto, SB, BC Rall, U Brose. 2007. Allometric degree distributions facilitate food-web stability. Nature 450(7173): 1226-1229. Overpeck, JT, D Rind, R Goldberg. 1990. Climate-induced changes in forest disturbance and vegetation. Nature 343(6253): 51-53. Paine, RT. 1966. Food web complexity and species diversity. The American naturalist 100(910): 65-75. Parke, JL, RG Linderman, CH Black. 1983. The role of ectomycorrhizas in drought tolerance of Douglas-fir seedlings. New Phytologist 95(1): 83-95. Parlad?, J, M Cohen, J Doltra, J Luque, J Pera. 2001. Continuous measurement of stem-diameter growth response of Pinus pinea seedlings mycorrhizal with Rhizopogon roseolus and submitted to two water regimes. Mycorrhiza 11(3): 129-136. Parlad?, J, J Pera, IF Alvarez. 1996. Inoculation of containerized Pseudotsuga menziesii and Pinus pinaster seedlings with spores of five species of ectomycorrhizal fungi. Mycorrhiza 6(4): 237-245. Parrott, L. 2010. Measuring ecological complexity. Ecological Indicators 10(6): 1069-1076. Peay, KG, PG Kennedy, TD Bruns. 2008. Fungal community ecology: a hybrid beast with a molecular master. BioScience 58(9): 799-810. Peng, S, DM Eissenstat, JH Graham, K Williams, NC Hodge. 1993. Growth depression in mycorrhizal citrus at high-phosphorus supply (analysis of carbon costs). Plant Physiology 101(3): 1063-1071.  136Pera, J, IF ?lvarez, A Rinc?n, J Parlad?. 1999. Field performance in northern Spain of Douglas-fir seedlings inoculated with ectomycorrhizal fungi. Mycorrhiza 9(2): 77-84. Perry, DA. 1989. Bootstrapping in ecosystems. BioScience 39(4): 230. Perry, DA. 1995. Self-organizing systems across scales. Trends in Ecology & Evolution 10(6): 241-244. Perry, DA, JG Borchers, SL Borchers, MP Amaranthus. 1990. Species migrations and ecosystem stability during climate change: the belowground connection. Conservation Biology 4(3): 266-274. Peter, M. 2006. Ectomycorrhizal fungi ? fairy rings and the wood-wide web. New Phytologist 171(4): 685-687. Peters, DPC, BT Bestelmeyer, AK Knapp, JE Herrick, HC Monger, KM Havstad 2009. Approaches to predicting broad-scale regime shifts using changing pattern-process relationships across scales. In: S Miao, S Carstenn, M Nungesser, eds. Real World Ecology: Large-Scale and Long-Term Case Studies and Methods. Springer New York. 47-72. Peters, DPC, JE Herrick. 2004. Strategies for ecological extrapolation. Oikos 106(3): 627-636. Peters, RH. 1983. The ecological implications of body size. Cambridge University Press, New York. Peterson, G, CR Allen, CS Holling. 1998. Ecological resilience, biodiversity, and scale. Ecosystems 1(1): 6-18. Peterson, RL, H Massicotte, B. 2004. Exploring structural definitions of mycorrhizas, with emphasis on nutrient-exchange interfaces. Canadian Journal of Botany 82(8): 15. Peterson, RL, HB Massicotte, LH Melville. 2004. Mycorrhizas: anatomy and cell biology. CABI Publishing, Wallingford. Pickles, BJ, D Genney, IJ Alexander, IC Anderson 2009. Spatial ecology of ectomycorrhizas: analytical strategies. In: C Azc?n-Aguilar, JM Barea, S Gianinazzi, V Gianinazzi-Pearson, eds. Mycorrhizas - Functional Processes and Ecological Impact. Springer Berlin Heidelberg. 1-11. Pickles, BJ, DR Genney, JM Potts, JJ Lennon, IC Anderson, IJ Alexander. 2010. Spatial and temporal ecology of Scots pine ectomycorrhizas. New Phytologist 186(3): 755-768. Pimm, SL. 1984. The complexity and stability of ecosystems. Nature 307(5949): 321-326. Plamboeck, A, T Dawson, L Egerton-Warburton, M North, T Bruns, J Querejeta. 2007. Water transfer via ectomycorrhizal fungal hyphae to conifer seedlings. Mycorrhiza 17(5): 439-447. Pojar, J, K Klinka, DV Meidinger. 1987. Biogeoclimatic ecosystem classification in British Columbia. Forest Ecology and Management 22(1-2): 119-154. Proulx, SR, DE Promislow, PC Phillips. 2005. Network thinking in ecology and evolution. Trends in Ecology and Evolution 20(6): 345-353. Queller, DC, JE Strassman. 2002. Kin selection. Current Biology 12(24): R832. Querejeta, J, L Egerton-Warburton, M Allen. 2003. Direct nocturnal water transfer from oaks to their mycorrhizal symbionts during severe soil drying. Oecologia 134(1): 55-64.  137Querejeta, JI, LM Egerton-Warburton, MF Allen. 2007. Hydraulic lift may buffer rhizosphere hyphae against the negative effects of severe soil drying in a California Oak savanna. Soil Biology and Biochemistry 39(2): 409-417. Querejeta, JI, LM Egerton-Warburton, MF Allen. 2009. Topographic position modulates the mycorrhizal response of oak trees to interannual rainfall variability. Ecology 90(3): 649-662. Read, DJ 1992. The mycorrhizal mycelium. In: MF Allen, ed. Mycorrhizal Functioning: An Integrative Plant-Fungal Process. Chapman & Hall, London, UK. 102-133. Read, DJ 2002. Towards ecological relevance- progress and pitfalls in the path towards an understanding of mycorrhizal functions in nature. In: MGA van der Heijden, IR Sanders, eds. Mycorrhizal Ecology. Springer-Verglag Berlin, Heidelberg, Germany. 3-29. Read, DJ, R Boyd 1986. Water relations of mycorrhizal fungi and their host plants. In: BL Ayres P, ed. Water, fungi and plants. Cambridge University Press, Cambridge. 287-303. Read, DJ, JG Duckett, R Francis, R Ligrone, A Russell. 2000. Symbiotic fungal associations in 'lower' land plants. Philosophical Transactions of the Royal Society of London. Series B: Biological Sciences 355(1398): 815-831. Read, DJ, J Perez-Moreno. 2003. Mycorrhizas and nutrient cycling in ecosystems ? a journey towards relevance? New Phytologist 157(3): 475-492. Redecker, D, TM Szaro, RJ Bowman, TD Bruns. 2001. Small genets of Lactarius xanthogalactus, Russula cremoricolor and Amanita francheti in late-stage ectomycorrhizal successions. Mol Ecol 10(4): 1025-1034. Remy, W, TN Taylor, H Hass, H Kerp. 1994. Four hundred-million-year-old vesicular arbuscular mycorrhizae. Proceedings of the National Academy of Sciences 91(25): 11841-11843. Renninger, HJ, FC Meinzer, BL Gartner. 2007. Hydraulic architecture and photosynthetic capacity as constraints on release from suppression in Douglas-fir and western hemlock. Tree Physiology 27(1): 33-42. Rezende, EL, P Jordano, J Bascompte. 2007. Effects of phenotypic complementarity and phylogeny on the nested structure of mutualistic networks. Oikos 116(11): 1919-1929. Rillig, MC, KK Treseder, MF Allen 2002. Global Change and Mycorrhizal Fungi. In: MGA van der Heijden, I Sanders, eds. Mycorrhizal Ecology. Springer-Verlag, Berlin. Rodenhuis, DR, KE Bennett, AT Werner, TQ Murdock, D Bronaugh 2009. Hydro-climatology and future climate impacts in British Columbia.In. Victoria, BC Canada: Pacific Climate Impacts Consortium, University of Victoria. 132. Roering, JJ, KM Schmidt, JD Stock, WE Dietrich, DR Montgomery. 2003. Shallow landsliding, root reinforcement, and the spatial distribution of trees in the Oregon Coast Range. Canadian Geotechnical Journal 40: 237-253. Rosenvald, R, A L?hmus. 2008. For what, when, and where is green-tree retention better than clear-cutting? A review of the biodiversity aspects. Forest Ecology and Management 255(1): 1-15. Rosling, A, R Landeweert, BD Lindahl, K-H Larsson, TW Kuyper, AFS Taylor, RD Finlay. 2003. Vertical distribution of ectomycorrhizal fungal taxa in a podzol soil profile. New Phytologist 159: 775-783.  138Roth, AL, SM Berch. 1992. Ectomycorrhizae of Douglas-fir and western hemlock seedlings outplanted on eastern Vancouver Island. Canadian Journal of Forest Research 22(11): 1646. Rotheray, TD, TH Jones, MD Fricker, L Boddy. 2008. Grazing alters network architecture during interspecific mycelial interactions. Fungal Ecology 1(4): 124-132. Ryker, RA. 1975. A survey of factors affecting regeneration of Rocky Mountain Douglas-Fir. Intermountain Forest and Range Experiment Station, USDA Forest Service, Ogden, UT. Saari, SK, CD Campbell, J Russell, IJ Alexander, IC Anderson. 2005. Pine microsatellite markers allow roots and ectomycorrhizas to be linked to individual trees. New Phytologist 165: 295-304. Sawada, M. 1999. Rookcase: an Excel 97/2000 Visual Basic (VB) add-in for exploring global and local spatial autocorrelation. Bulletin of the Ecological Society of America 80(4): 231-234. Scheffer, M, S Carpenter, JA Foley, C Folke, B Walker. 2001. Catastrophic shifts in ecosystems. Nature 413(6856): 591-596. Schenk, HJ, RB Jackson. 2002. Rooting depths, lateral root spreads and below-ground/above-ground allometries of plants in water-limited ecosystems. Journal of Ecology 90(3): 480-494. Schoonmaker, A, F Teste, S Simard, R Guy. 2007. Tree proximity, soil pathways and common mycorrhizal networks: their influence on the utilization of redistributed water by understory seedlings. Oecologia 154(3): 455-466. Selosse, MA, S Duplessis. 2006. More complexity in the mycorrhizal world. New Phytologist 172(4): 600-604. Selosse, MA, F Richard, X He, SW Simard. 2006. Mycorrhizal networks: des liaisons dangereuses? Trends in Ecology and Evolution 21(11): 621-628. Shafer, SL, PJ Bartlein, RS Thompson. 2001. Potential changes in the distributions of western North America tree and shrub taxa under future climate scenarios. Ecosystems 4(3): 200-215. Shatford, JPA, DE Hibbs, KJ Puettmann. 2007. Conifer regeneration after forest fire in the Klamath-Siskiyous: how much, how soon? Journal of Forestry 105: 139-146. Simard, SW. 2009. The foundational role of mycorrhizal networks in self-organization of interior Douglas-fir forests. Forest Ecology and Management 258(Supplement): S95-S107. Simard, SW, D Durall, M Jones 2002. Carbon and nutrient fluxes within and between mycorrhizal plants. In: MGA van der Heijden, I Sanders, eds. Mycorrhizal ecology. Springer-Verlag, Berlin. 33-74. Simard, SW, DM Durall. 2004. Mycorrhizal networks: a review of their extent, function, and importance. Canadian Journal of Botany-Revue Canadienne De Botanique 82(8): 1140-1165. Simard, SW, DA Perry, MD Jones, DD Myrold, DM Durall, R Molina. 1997a. Net transfer of carbon between ectomycorrhizal tree species in the field. Nature 388(6642): 579-582. Simard, SW, DA Perry, JE Smith, R Molina. 1997b. Effects of soil trenching on occurrence of ectomycorrhizas on Pseudotsuga menziesii seedlings grown in mature  139forests of Betula papyrifera and Pseudotsuga menziesii. New Phytologist 136(2): 327-340. Simon, L, J Bousquet, RC Levesque, M Lalonde. 1993. Origin and diversification of endomycorrhizal fungi and coincidence with vascular land plants. Nature 363(6424): 67-69. Slavov, GT, GT Howe, I Yakovlev, KJ Edwards, KV Krutovskii, GA Tuskan, . . . WT Adams. 2004. Highly variable SSR markers in Douglas-fir: Mendelian inheritance and map locations. Theoretical and Applied Genetics 108: 873-880. Smith, AH, SM Zeller. 1966. A preliminary account of the North American species of Rhizopogon. Memoirs of the New York Botanical Garden 14: 1-178. Smith, JE, KA Johnson, E C?zares. 1998. Vesicular mycorrhizal colonization of seedlings of Pinaceae and Betulaceae after spore inoculation with Glomus intraradices. Mycorrhiza 7(6): 279-285. Smith, JHG. 1964. Root spread can be estimated from crown width of Douglas fir, lodgepole pine, and other British Columbia tree species. The Forestry Chronicle 40(4): 456-473. Smith, SE, DJ Read. 2008. Mycorrhizal Symbiosis. Academic Press, London, U.K. . Smock, LA, CM MacGregor. 1988. Impact of the american chestnut blight on aquatic shredding macroinvertebrates. Journal of the North American Benthological Society 7(3): 212-221. Song, YY, RS Zeng, JF Xu, J Li, X Shen, WG Yihdego. 2010. Interplant communication of tomato plants through underground common mycorrhizal networks. PLoS ONE 5(10): e13324. Southworth, D, X-H He, W Swenson, CS Bledsoe. 2005. Application of network theory to potential mycorrhizal networks. Mycorrhiza 15: 589-595. Staebler, GR. 1956. Evidence of shock following thinning of young Douglas-fir. Journal of Forestry 54(5): 339-340. Stendell, ER, TR Horton, TD Bruns. 1999. Early effects of prescribed fire on the structure of the ectomycorrhizal fungus community in a Sierra Nevada ponderosa pine forest. Mycological Research 103(10): 1353-1359. Sthultz, CM, CA Gehring, TG Whitham. 2007. Shifts from competition to facilitation between a foundation tree and a pioneer shrub across spatial and temporal scales in a semiarid woodland. New Phytologist 173(1): 135-145. Stone, EL, PJ Kalisz. 1991. On the maximum extent of tree roots. Forest Ecology and Management 46: 59-102. Suding, KN, KL Gross, GR Houseman. 2004. Alternative states and positive feedbacks in restoration ecology. Trends in Ecology & Evolution 19(1): 46-53. Sutherland, JP. 1974. Multiple stable points in natural communities. The American naturalist 108(964): 859-873. Taber, RA, WA Taber. 1984. Notes: evidence for ectomycorrhizal fungus-mediated nutrient transfer between Pinus and Tradescantia. Forest Science 30(4): 892-896. Talbot, JM, SD Allison, KK Treseder. 2008. Decomposers in disguise: mycorrhizal fungi as regulators of soil C dynamics in ecosystems under global change. Functional Ecology 22(6): 955-963.  140Taylor, DL, TD Bruns. 1999. Community structure of ectomycorrhizal fungi in a Pinus muricata forest: minimal overlap between the mature forest and resistant propagule communities. Molecular Ecology 8(11): 1837-1850. Tedersoo, L, T Jairus, BM Horton, K Abarenkov, T Suvi, I Saar, U K?ljalg. 2008a. Strong host preference of ectomycorrhizal fungi in a Tasmanian wet sclerophyll forest as revealed by DNA barcoding and taxon-specific primers. New Phytologist 180(2): 479-490. Tedersoo, L, U K?ljalg, N Hallenberg, K-H Larsson. 2003. Fine scale distribution of ectomycorrhizal fungi and roots across substrate layers including coarse woody debris in a mixed forest. New Phytologist 159: 153-165. Tedersoo, L, T Suvi, T Jairus, U K?ljalg. 2008b. Forest microsite effects on community composition of ectomycorrhizal fungi on seedlings of Picea abies and Betula pendula. Environmental Microbiology 10(5): 1189-1201. Teste, F, S Simard. 2008. Mycorrhizal networks and distance from mature trees alter patterns of competition and facilitation in dry Douglas-fir forests. Oecologia 158(2): 193-203. Teste, FP, SW Simard, DM Durall. 2009. Role of mycorrhizal networks and tree proximity in ectomycorrhizal colonization of planted seedlings. Fungal Ecology 2(1): 21-30. Teste, FP, SW Simard, DM Durall, RD Guy, SM Berch. 2010. Net carbon transfer between Pseudotsuga menziesii var. glauca seedlings in the field is influenced by soil disturbance. Journal of Ecology 98(2): 429-439. Theodorou, C. 1978. Soil moisture and the mycorrhizal association of Pinus radiata D. don. Soil Biology and Biochemistry 10(1): 33-37. Thiers, HD. 1984. The secotioid syndrome. American Cinematographer, Hollywood, CA, ETATS-UNIS. Tilman, D. 1982. Resource Competition and Community Structure. Princeton University Press, Princeton, NJ. Tilman, D, JA Downing. 1994. Biodiversity and stability in grasslands. Nature 367(6461): 363-365. Tinker, PB, DM Durall, MD Jones. 1994. Carbon use efficiency in mycorrhizas: theory and sample calculations. New Phytologist 128(1): 115-122. Trappe, JM. 1965. Tuberculate mycorrhizae of Douglas-fir. Forest Science 11: 27-32. Trappe, JM. 1977. Selection of fungi for ectomycorrhizal inoculation in nurseries. Trappe, JM. 2005. A.B. Frank and mycorrhizae: the challenge to evolutionary and ecologic theory. Mycorrhiza 15(4): 277-281. Trappe, JM, DL Luoma 1992. The ties that bind: fungi in ecosystems. In: GC Carroll, DT Wicklow, eds. The Fungal Community: Its Organization and Role in the Ecosystem. Marcel Dekker, New York, NY. 383-401. Turcotte, DL, JB Rundle. 2002. Self-organized complexity in the physical, biological, and social sciences. Proceedings of the National Academy of Sciences of the United States of America 99(Suppl 1): 2463-2465. Twieg, B, D Durall, S Simard, M Jones. 2009. Influence of soil nutrients on ectomycorrhizal communities in a chronosequence of mixed temperate forests. Mycorrhiza 19(5): 305-316. Twieg, BD, DM Durall, SW Simard. 2007. Ectomycorrhizal fungal succession in mixed temperate forests. New Phytologist 176(2): 437-447.  141Urban, DL. 2005. Modeling ecological processes across scales. Ecology 86(8): 1996-2006. van de Koppel, J, AH Altieri, BR Silliman, JF Bruno, MD Bertness. 2006. Scale-dependent interactions and community structure on cobble beaches. Ecology Letters 9(1): 45-50. van der Heijden, MGA, TR Horton. 2009. Socialism in soil? The importance of mycorrhizal fungal networks for facilitation in natural ecosystems. Journal of Ecology 97(6): 1139-1150. van der Heijden, MGA, IR Sanders. 2002. Mycorrhizal Ecology. Springer-Verlag, Berlin. von Rudloff, E. 1975. Volatile leaf oil analysis in chemosystematic studies of North American conifers. Biochemical Systematics and Ecology 2(3-4): 131-167. Vyse, A, C Ferguson, SW Simard, T Kano, P Puttonen. 2006. Growth of Douglas-fir, lodgepole pine, and ponderosa pine seedlings underplanted in a partially-cut, dry Douglas-fir stand in south-central British Columbia. The Forestry Chronicle 82(5): 723-732. Vyse, A, RA Smith, BG Bondar. 1990. Management of interior Douglas-fir stands in British Columbia: past, present, and future. In: DM Baumgartner, JE Lotan, eds. Interior Douglas-fir: the species and its management. Washington State University Extension, Pullman, WA, Spokane, WA, USA. 177-185.  Walker, IJ, R Sydneysmith 2008. British Columbia. In: DS Lemmen, FJ Warren, J Lacroix, E Bush, eds. From impacts to adaptation: Canada in a changing climate 2007. Government of Canada, Ottawa, ON Canada. 329-386. Wallander, H, LO Nilsson, D Hagerberg, E B??th. 2001. Estimation of the biomass and seasonal growth of external mycelium of ectomycorrhizal fungi in the field. New Phytologist 151(3): 753-760. Wang, B, YL Qiu. 2006. Phylogenetic distribution and evolution of mycorrhizas in land plants. Mycorrhiza 16(5): 299-363. Warren, JM, JR Brooks, FC Meinzer, JL Eberhart. 2008. Hydraulic redistribution of water from Pinus ponderosa trees to seedlings: evidence for an ectomycorrhizal pathway. New Phytologist 178(2): 382-394. Watts, DJ, SH Strogatz. 1998. Collective dynamics of 'small-world' networks. Nature 393(6684): 440-442. Whitfield, J. 2007. Fungal roles in soil ecology: underground networking. Nature 449(7159): 136-138. Whitham, TG, JK Bailey, JA Schweitzer, SM Shuster, RK Bangert, CJ LeRoy, . . . SC Wooley. 2006. A framework for community and ecosystem genetics: from genes to ecosystems. Nat Rev Genet 7(7): 510-523. Williams, RJ, EL Berlow, JA Dunne, AL Barab?si, ND Martinez. 2002. Two degrees of separation in complex food webs. Proceedings of the National Academy of Sciences of the United States of America 99(20): 12913-12916. Wisheu, IC. 1998. How organisms partition habitats: different types of community organization can produce identical patterns. Oikos 83(2): 246-258. Worthington, NP, GR Staebler. 1961. Commercial thinning of Douglas-fir in the Pacific Northwest. UF Service, Pacific Northwest Forest and Range Experiment Station,  Wu, B, K Nara, T Hogetsu. 1999. Competition between ectomycorrhizal fungi colonizing Pinus densiflora. Mycorrhiza 9(3): 151-159.  142Wu, B, K Nara, T Hogetsu. 2001. Can 14C-labeled photosynthetic products move between Pinus densiflora seedlings linked by ectomycorrhizal mycelia? New Phytologist 149: 137-146. Yachi, S, M Loreau. 1999. Biodiversity and ecosystem productivity in a fluctuating environment: The insurance hypothesis. Proceedings of the National Academy of Sciences 96(4): 1463-1468. Zak, B. 1971. Characterization and classification of mycorrhizae of Douglas fir  II. Pseudotsuga menziesii + Rhizopogon vinicolor. Canadian Journal of Botany 49: 1079-1084. Zavarin, E, K Snajberk. 1973. Geographic variability of monoterpenes from cortex of Pseudotsuga menziesii. Pure and Applied Chemistry 34: 411-434. Zeide, B. 2001. Thinning and growth: a full turnaround. Journal of Forestry 99: 20-25. Zhou, Z, T Hogetsu. 2002. Subterranean community structure of ectomycorrhizal fungi under Suillus grevillei sporocarps in a Larix kaempferi forest. New Phytologist 154(2): 529-539. Zhou, Z, M Miwa, Y Matsuda, T Hogetsu. 2001. Spatial distribution of the subterranean mycelia and ectomycorrhizae of Suillus grevillei genets. Journal of Plant Research 114(2): 179-185.   143Appendices Appendix A  Chapter 2 supplemental material A.1 Rhizopogon morphology  Supplemental Figure 2.1    Morphology of Rhizopogon vesiculosus/vinicolor, showing tuberculate mycorrhizas (top row) with root tips (a) encased in a fungal ?rind? (b), vessel-like rhizomorphs (c), and truffle-like sporocarps (d). Bottom left photo shows interior Douglas-fir roots and ectomycorrhizal mycelia (incl. Rhizopogon tubercle, arrow) peeled from rock surface. Bottom middle shows immature sporocarp with rhizomorphs attached. Scale bars are 1 mm.  (a) (a) (a) (b) (b) (c) (c) (d) (d)  144A.2 Plot characteristics and sampling  Supplemental Figure 2.2    Picture of subsurface soil following the removal of a large rock, showing a natural cross-section of a Rhizopogon spp. tubercle (arrow) that was plastered to the rock surface prior to removal.  Roots and ectomycorrhizal hyphae (including Rhizopogon-like, brown) permeate the soil.    145 Supplemental Figure 2.3    Characteristic multi-cohort stand structure of study plots in the dry, cool Interior Douglas-fir biogeoclimatic ecological subzone near Kamloops, BC (plot with relatively xeric soil conditions shown, May 2007). 146 Supplemental Figure 2.4    Typical characteristics of study plot soils, showing the litter layer (L), F and H forest floor layers, eluviated A mineral horizon, and modified B mineral horizons according to the Canadian System of Soil Classification.  Supplemental Figure 2.5    Contiguous 20 cm3 soil blocks were excavated systematically from the top down (forest floor layers and mineral horizons separately) in search of Rhizopogon-like hyphae, rhizomorphs, or tuberculate mycorrhizas. 147A.3 Mycelia continuity: auxiliary results  Supplemental Figure 2.6    Spatial occurrence and continuity of R. vesiculosus (a) and R. vinicolor (b) mycelia across 1 x 2 m area, showing mycelia density classes krigged across contiguously sampled 20 cm3 soil blocks, classified separately in the F,H forest floor layers (top panels), Ae mineral horizon (middle panels), and B mineral horizon (bottom panels). Density classes are shown grading from purple to red, representing: absent, scarce, diffuse, patchily dense, and dense mycelia classes respectively. See main text for additional information.1 metre  148Appendix B  Chapter 3 supplemental material B.1 Incidence matrix showing the number of times each Rhizopogon vesiculosus and/or R. vinicolor genet was encountered on the roots of each interior Douglas-fir tree.  asee chapter 3 main text for cohort descriptions; btree boles located outside the 30 x 30 m plot boundaries with roots encountered inside the plot; cexcludes genets (one from each Rhizopogon spp.) representing mycorrhiza samples in which the identity of tree roots was undetermined; [link to data in MS Excel 2011: Table S2.xls] Tree # Cohorta# VES /tree# VIN /treeTotal genets/tree37 1 0 0 038 1 0 0 039 1 0 0 040 1 1 1 1 1 242 1 0 0 043 1 0 0 03 2 4 1 2 1 2 2 44 2 1 4 2 1 2 2 47 2 1 1 1 1 210 2 1 1 0 111 2 1 2 1 3 0 313 2 1 1 0 114 2 1 0 1 115 2 1 0 1 126 2 0 0 027 2 3 3 1 1 228 2 0 0 029 2 0 0 030b 2 1 1 1 1 231b 2 1 0 1 132 2 0 0 033 2 0 0 034 2 0 0 035 2 1 1 0 136 2 0 0 041 2 1 0 1 144 2 1 1 0 146 2 0 0 047 2 0 0 050 2 2 1 1 2 1 353 2 1 1 0 154 2 1 1 1 1 255 2 1 1 1 1 256 2 1 0 1 159 2 1 1 0 160 2 0 0 062 2 1 1 0 169 2 1 0 1 176 2 0 0 077 2 0 0 078 2 0 0 079 2 1 1 0 181 2 1 1 0 188 2 1 1 0 11 3 0 0 02 3 3 1 0 15 3 14 1 1 1 26 3 10 1 1 1 2 2 48 3 1 1 0 19 3 7 1 1 6 1 2 5 4 2 1 7 3 1012 3 4 2 1 6 2 2 417 3 1 2 1 2 1 318 3 1 7 2 1 3 1 419 3 4 2 2 0 220 3 2 1 1 1 4 0 421 3 2 3 2 2 1 345 3 0 0 048 3 0 0 049 3 10 1 1 1 3 1 451 3 3 1 0 152 3 1 1 1 1 4 0 461 3 0 0 074 3 1 1 0 175 3 1 1 0 187 3 3 1 1 1 9 3 1 5 2 789b 3 1 0 1 191b 3 6 1 0 116 4 1 1 7 1 5 4 1 522b 4 1 1 3 4 2 3 2 523b 4 11 3 12 6 6 5 1 2 9 2 5 8 3 1124b 4 1 3 2 0 225 4 1 1 1 1 257 4 1 0 1 164 4 3 1 0 183b 4 1 11 3 2 3 1 485b 4 1 1 1 2 1 386b 4 3 2 2 0 290b 4 1 1 1 2 1 379 10 8 1 1 59 42 1 28 15 1 3 9 7 9 13 10 12 7 1 10 5 1 1 5 Total EM samples> 33819 7 6 1 1 18 18 1 11 7 1 2 4 4 4 4 3 10 3 1 4 3 1 1 5 # colonized trees > 56VES-04VES-03VES-02VES-01VES-10VES-09VES-08VES-07VES-06VES-05VIN-03VIN-02VIN-01VES-13VES-12VES-11# trees/ genet >VIN-12VIN-11VIN-10VIN-09VIN-08VIN-07VIN-06VIN-05VIN-04R. vinicolor  genetsc:Rhizopogon vesiculosus  genetsc:# samples/ genet > 149B.2 Ancillary illustrations  Supplemental Figure 3.1    The relative height and spatial distribution of interior Douglas-fir trees in the 30 x 30 m study plot depicted using Stand Visualization System.         150  151Supplemental Fig. 3.2    The location of interior Douglas-fir trees and Rhizopogon spp. mycorrhiza samples in the 30 x 30 m plot and the spatial distribution of (a) tree root spans, (b) R. vesiculosus genets, and (c) R. vinicolor genets (see chapter 3 main text).  152Appendix C  Chapter 4 supplemental material C.1 Incidence matrices showing associations between Rhizopogon spp. genets and interior Douglas-fir trees in 100 m2 plots with xeric or mesic soil moisture regimes.  Supplemental Table 4.1    Incidence matrices showing the number of associations between Rhizopogon vesiculosus and/or R. vinicolor genets and different interior Douglas-fir tree cohorts in plots with xeric (plots 1-3) or mesic (plots 4-6) soil moisture regimes.  asee chapter 4 main text for tree cohort descriptions; btree boles located outside the 10 x 10 m plot boundaries with roots encountered inside the plot Tree Cohorta VES-1VES-2VES-3VES-4VIN-1VIN-2Tree Cohorta VES-1VES-2VIN-1VIN-2VIN-3VIN-4Tree Cohorta VES-1VES-2VES-3VES-4VIN-1VIN-2VIN-3VIN-4VIN-5T-01 1 T-01 1 1 T-01b 1 1 1T-02 1 4 T-02 1 T-02 1T-03 1 1 T-03 1 T-03 1 2T-04 1 1 T-04 1 T-04 1 1T-05 2 3 T-05 1 T-05 1 3T-06 2 3 1 T-06 1 T-06 1 1 1T-07 2 T-07 1 T-07 1 1T-08b 2 1 T-08b 2 1 T-08b 1 1T-09b 2 1 1 T-09 3 1 2 15 1 T-09b 1 1 1 4T-10b 2 3 T-10 3 1 1 T-10 1T-11b 2 1 T-11 3 4 T-11b 1 1T-12b 2 1 T-12 3 6 1 1 T-12b 1 1T-13 3 4 1 T-13 3 5 1 1 T-13 1 1T-14 3 2 T-14b 3 4 2 1 T-14 1 2 1T-15 3 1 T-15b 3 1 T-15 2 1 1T-16 3 5 T-16b 3 1 5 T-16b 2 1T-17 3 T-17b 4 4 2 T-17b 2 1T-18 3 18 1 T-18b 4 3 T-18 2 1T-19b 3 1 T-19b 5 1 2 T-19b 3 8 2T-20b 3 4 T-20b 3 1T-21b 3 1 T-21b 3 1T-22b 3 2 T-22b 3 1T-23b 3 1 T-23b 3 1T-24b 3 1 T-24b 3 2T-25b 4 2 2 T-25 4 2 4T-26b 4 1 T-26b 4 4 3T-27b 4 1 T-27b 4 5 2 1 1T-28b 4 1 T-28 4 15 2 2 7T-29b 5 4 1 T-29b 4 1T-30b 4 7 1 2T-31b 4 1T-32 5 10 1 1 7 1T-33b 5 1 1Plot 1 Plot 2 Plot 3 153Supplemental Table 4.1 continued  asee chapter 4 main text for tree cohort descriptions; btree boles located outside the 10 x 10 m plot boundaries with roots encountered inside the plot Tree CohortaVES-1VES-2VES-3VES-4VIN-1VIN-2Tree CohortaVES-1VES-2VES-3VES-4VES-5VES-6VES-7VIN-1VIN-2VIN-3VIN-4Tree CohortaVES-1VES-2VES-3VES-4VES-5VES-6VIN-1VIN-2VIN-3VIN-4VIN-5VIN-6VIN-7VIN-8VIN-9T-01 1 T-01 1 1 T-01 1 1T-02 1 T-02 1 1 T-02 1 2 2 2 4 2T-03b 1 1 T-03 2 1 2 1 T-03 2 3 1T-04 2 1 T-04 2 T-04 2T-05 2 1 1 2 T-05 2 T-05 2 1 1T-06 2 5 3 T-06b 2 1 T-06 2 2T-07 2 1 T-07b 2 1 3 1 T-07 2 2T-08 2 1 1 T-08b 2 1 T-08b 2 1T-09 2 1 1 T-09b 2 2 T-09 2 2 1 1T-10b 2 1 T-10b 2 1 T-10 2 3 1 4 1T-11b 2 1 T-11b 2 1 T-11 2 1 3 5T-12b 2 3 T-12b 2 1 T-12 2 2T-13b 2 1 1 T-13b 2 1 1 1 T-13b 2 3 1 2T-14 2 12 1 1 3 T-14b 2 1 T-14b 2 2 1T-15b 2 1 5 T-15b 2 1 T-15b 2 1T-16b 2 1 T-16 3 12 1 T-16b 2 1T-17b 2 1 1 T-17 3 1 7 2 1 T-17b 2 1T-18b 2 1 T-18 3 3 2 T-18b 2 1T-19b 2 1 T-19 3 2 1 1 T-19b 2 1T-20b 2 1 T-20b 3 6 2 T-20 3 5 4 3 11 9T-21b 2 1 T-21b 3 6 1 1 T-21 3 7 1 6 1 1 1T-22 3 5 2 T-22b 3 1 T-22 3 4 1 1T-23 3 2 1 11 T-23b 3 1 T-23 3 7 3 6T-24 3 4 6 10 T-24b 4 1 1 3 1 T-24b 3 1 2T-25 3 1 3 2 T-25b 4 11 3 8 5 3 T-25b 3 2 1 1 1 3T-26b 3 1 1 2 1 T-26b 4 1 T-26 3T-27b 3 1 T-27b 4 1 6 2 T-27 3 7 8 5 1 4 1T-28b 4 1 1 T-28b 5 1 T-28 3 10 5 1T-29b 5 1 T-29b 5 2 1 T-29b 3 1 1 1T-30b 3 1T-31b 3 1T-32b 3 1T-33b 3 1 1T-34b 3 1 5 9T-35b 3 3T-36b 3 1T-37b 3 3 1T-38b 3 1T-39b 3 1T-40b 4 11 3 2 6 1T-41b 4 2T-42b 5 1T-43b 5 3 1Plot 6Plot 4 Plot 5 154C.2 Estimation of the plot edge effect  Sampling area (square meters)0 200 400 600 800 1000Proportional change in variable1.01.52.02.53.0# Rhizopogon spp. mycorrhizas sampled# trees associated with Rhizopogon mycorrhizasmaximum tree node degree (connectivity)mean of tree node degreesmedian of tree node degrees# trees per R. vesiculosus genet# trees per R. vinicolor genet  Supplemental Figure 4.1    Estimation of the plot edge effect on mycorrhizal network attributes, based on tree ?nodes? linked through Rhizopogon spp. genets in a 100 m2 plot, and the proportional increase in networking attributes of trees and fungi from the 100 m2 plot with additional Rhizopogon spp. mycorrhizas sampled in 20 x 20 m and 30 x 30 m buffer areas surrounding the initial plot.  With increasing sampling area: (1) the number of tree genotypes identified from mycorrhizas increased relative to the number of mycorrhizas sampled, (2) R. vesiculosus genets, but not R. vinicolor genets, continued to expand beyond plot borders and colonize additional tree roots, and (3) the maximum and median node degree of trees in the 10 x 10 m plot had a greater proportional increase than mean node degrees, suggesting the node degree distribution of trees in the 10 x 10 m plot became more positively skewed at greater sampling extents.   

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0072536/manifest

Comment

Related Items