Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Plant growth promoting rhizobacteria (PGPR) for interior spruce (Picea engelmannii x P. glauca) seedlings Shishido, Masahiro 1997

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

Item Metadata

Download

Media
831-ubc_1997-251594.pdf [ 18.22MB ]
Metadata
JSON: 831-1.0088248.json
JSON-LD: 831-1.0088248-ld.json
RDF/XML (Pretty): 831-1.0088248-rdf.xml
RDF/JSON: 831-1.0088248-rdf.json
Turtle: 831-1.0088248-turtle.txt
N-Triples: 831-1.0088248-rdf-ntriples.txt
Original Record: 831-1.0088248-source.json
Full Text
831-1.0088248-fulltext.txt
Citation
831-1.0088248.ris

Full Text

PLANT GROWTH PROMOTING RH3ZOBACTERIA (PGPR) FOR INTERIOR SPRUCE (Picea engelmannii x P. glauca) SEEDLINGS by MASAHIRO SfflSHTDO B.Sc., Chiba University, 1983 M.Sc., The University of Arizona, 1987 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Forest Sciences We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA August 1997 ©Masahiro Shishido, 1997 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that; the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Foirec, f £ci en C g £ The University of British Columbia Vancouver, Canada Date gef)te.^b*r SI , /f<tj DE-6 (2/88) ABSTRACT The relationship between interior spruce (Picea engelmannii Parry x P. glauca (Moench) Voss) and plant growth promoting rhizobacteria (PGPR) was studied under controlled environments and in the field. Large, statistically significant biomass increases were detected in spruce seedlings after PGPR inoculation, but seedling growth responses were variable. Co-adaptation involving host plants, PGPR and forest floor soils did not appear to explain such variability. Synergistic effects of PGPR and mycorrhizal fungi on seedling growth were detected, but growth promotion also occurred in the absence of mycorrhizae, which suggests that bacteria x mycorrbizae interactions were unrelated to seedling growth response variability. PGPR colonization of seedling tissues was assessed using immunofluorescent microscopy and dilution plate counts. When three Bacillus and three Pseudomonas strains were inoculated onto spruce seedling roots under gnotobiotic conditions, only Bacillus Pw2R and Pseudomonas Sm3RN were detected inside stem vascular tissues four months later. A field experiment was performed to evaluate differences in root colonization and seedling growth promotion between these endophytic and non-endophytic PGPR. Relative growth rates of spruce seedlings in the field suggested that once induced in the greenhouse, seedling growth promotion persisted under field conditions for at least four months. However, endophytic PGPR offered no apparent advantage over non-endophytes as growth promoters. Mechanistic studies of plant growth promotion in sterile microcosms suggested that strains L6-16R, Pw2R, S20R, Sm3RN and Sw5RN did not depend on the presence of other deleterious microorganisms to promote seedling growth, but that strain Ss2RN might. However, addition of sterilized forest soil extracts facilitated seedling growth promotion by most PGPR. These results suggest that abiotic soil compounds may act as precursor substrates for PGPR production of plant growth stimulating substances, possibly, but not restricted to, phytohormones. Patchy distribution of such precursors in soil could cause seedling growth variability in response to PGPR inoculation. Finally, PGPR were observed to change soil microbial community population sizes ii and carbon substrate utilization patterns. Seedlings buffered these effects in only one of two forest soils evaluated, indicating that the origin of the soil microbial community is important in determining microfloral responses to PGPR introduction. iii TABLE OF CONTENTS Page ABSTRACT ii TABLE OF CONTENTS iv LIST OF TABLES viii LIST OF FIGURES xv LIST OF ABBREVIATIONS xxiii ACKNOWLEDGEMENTS xxv CHAPTER 1 General Introduction 1 CHAPTER 2 Specificity of PGPR Strains for Spruce and Pine 8 2.1 Introduction and Literature Review 8 2.2 Materials and Methods 12 2.2.1 Screening of PGPR strains for interior spruce and lodgepole pine seedlings... 12 2.2.1.1 Sampling of seedlings and soils 12 2.2.1.2 Isolation and storage of bacteria 12 2.2.1.3 Screening of plant growth promoting capability 15 2.2.1.4 The first screening 15 2.2.1.5 The second screening 16 2.2.1.6 Selection of bacterial isolates used in further experimentation 17 2.2.2 Experiment 1: Host ecotype specificity as a component of PGPR efficacy 17 2.2.2.1 Microorganisms for Experiment 1 17 2.2.2.2 Seeds and seedling growth for Experiment 1 17 2.2.2.3 Design and statistical analysis for Experiment 1 19 2.2.3 Experiment 2: Host genus specificity as a component of PGPR efficacy 20 2.2.3.1 Microorganisms for Experiment 2 20 2.2.3.2 Seeds and seedling growth for Experiment 2 20 2.2.3.3 Bacterial root colonization assessment for Experiment 2 21 2.2.3.4 Design and statistical analysis for Experiment 2 21 2.2.4 Experiment 3: PGPR-soil specificity 22 2.2.4.1 Microorganisms for Experiment 3 22 2.2.4.2 Seeds and seedling growth for Experiment 3 23 2.2.4.3 Bacterial root colonization assessment for Experiment 3 23 2.2.4.4 Design and statistical analysis for Experiment 3 23 2.3 Results 25 2.3.1 Screening of PGPR for interior spruce and lodgepole pine seedlings 25 2.3.2 Experiment 1: Host ecotype specificity as a component of PGPR efficacy..... 25 2.3.2.1 Seedling emergence 29 2.3.2.2 Seedling growth 29 iv 2.3.3 Experiment 2: Host genus specificity as a component of PGPR efficacy 34 2.3.3.1 Seedling emergence 34 2.3.3.2 Seedling growth 34 2.3.3.3 Inoculum colonization in the rhizosphere 34 2.3.4 Experiment 3: Soil specificity as a component of PGPR efficacy 40 2.3.4.1 Seedling emergence 40 2.3.4.2 Seedling growth 40 2.3.4.3 Inoculum colonization in the rhizosphere 45 2.4 Discussion 47 2.4.1 Host ecotype specificity effects on plant growth promotion by bacteria 47 2.4.2 Host species specificity effects on plant growth promotion by bacteria 48 2.4.3 Soil origin effects on plant growth promotion by bacteria 50 CHAPTER 3 Spruce and Pine Seedling Growth and Mycorrhizal Infection after Inoculation with PGPR 52 3.1 Introduction Literature Review 52 3.2 Materials and Methods 55 3.2.1 Experiment 4: Seedling growth and mycorrhizal infection after inoculation with PGPR 55 3.2.1.1 Microorganisms 55 3.2.1.2 Seeds and soils 55 3.2.1.3 Seedling growth and bacterial inoculation 56 3.2.1.4 Ectomycorrhizae and bacterial root colonization assessment 56 3.2.1.5 Experimental design and statistical analysis 56 3.3 Results. 57 3.3.1 Effects of PGPR inoculation on mycorrhizal formation 57 3.3.2 Seedling growth promotion by PGPR inoculation and mycorrhizal status 59 3.3.3 Bacterial root colonization and mycorrhizal status 65 3.4 Discussion 69 CHAPTER 4 Root Colonization by Spruce PGPR 73 4.1 Introduction and Literature Review 73 4.2 Materials and Methods 79 4.2.1 Experiment 5: Colonization of spruce by PGPR 79 4.2.1.1 Microorganisms 79 4.2.1.2 Seeds and seedling culture '.. 79 4.2.1.3 Antibody production 80 4.2.1.4 Evaluation of antibody cross-reactivity 81 4.2.1.5 Immunofluorescent microscopy 81 4.2.1.6 Recovery of inocula from seedlings 82 4.3 Results 83 4.3.1 Immunofluorescent microscopy 83 4.3.2 Recovery of inocula from the rhizosphere, root and stem tissues 93 4.3.3 Carbon substrate utilization profiles of PGPR in vitro 93 4.4 Discussion 97 v CHAPTER 5 PGPR Colonization and Growth Promotion of inoculated Spruce Seedlings in a Field Test 101 5.1 Introduction and Literature Review 101 5.2 Materials and Methods 104 5.2.2 Experiment 6: Growth performance of PGPR-inoculated spruce under greenhouse and field 104 5.2.2.1 Seedling preparation 104 5.2.2.2 Evaluation of root colonization and seedling performance of PGPR-inoculated spruce grown in the field 105 5.2.2.3 Statistical analyses of seedling performance and colonization by PGPR stains 109 5.3 Results I l l 5.3.1 Plant growth responses to PGPR inoculation under greenhouse and field conditions I l l 5.3.2 Recovery of PGPR strain from root systems 115 5.3.3 Relationships between seedling performance and recovery of PGPR strains... 115 5.4 Discussion 121 CHAPTER 6 Mechanisms of Plant Growth Promotion by Spruce PGPR 125 6.1 Introduction and Literature Review 125 I. Direct plant growth promotion 126 (i) Phytohormone production 126 (ii) Facilitation of nutrient uptake 127 n. Indirect plant growth promotion 128 (i) Suppression of deleterious microorganisms 128 (ii) Enhancing symbioses between plant and other microorganisms 131 m. Effects of PGPR on indigenous forest soil microorganisms 132 6.2 Materials and Methods 135 6.2.1 Experiment 7: Short-term assessment of plant growth promoting mechanism.... 135 6.2.1.1 Seeds and soils 135 6.2.1.2 Seedling medium and bacterial inoculum in a microcosm 135 6.2.1.3 Evaluation of PGPR population sizes 136 6.2.1.4 Measurement of forest soil microbial communities 137 6.2.1.5 Experimental design 139 6.3 Results 140 6.3.1 Soil nutrient levels 140 6.3.2 Effects of PGPR inoculation on soil microbial population size 140 6.3.3 Effects of PGPR inoculation on soil microbial community activity 141 6.3.4 Determination of direct and indirect mechanisms by the PGPR strains 146 6.3.5 Recovery of PGPR in microcosm... 153 6.3.6 Correlations of seedling biomass and populations of soil microorganisms 153 6.4 Discussion 157 CHAPTER 7 Conclusions 163 LITERATURE CITED 170 vi APPENDIX 1 Habitat characteristics of isolated PGPR strains 187 APPENDLX 2 Bacterial species identification using fatty acid metyl-ester (GC-FAME) and Biolog™ systems 189 APPENDIX 3 Antibiotic susceptibilities of PGPR strains using Sensi-Discs 194 APPENDIX 4 Statistical analyses for data presented in Chapter 2 196 APPENDLX 5 Statistical analyses for data presented in Chapter 3 270 APPENDLX 6 Statistical analyses for data presented in Chapter 5 277 APPENDLX 7 Statistical analyses for data presented in Chapter 6 290 APPENDLX 8 Regression equation of optical density (Aeoo) and mean generation time of PGPR strains 319 APPENDLX 9 Euclidean distance matrix for Biolog™ plate responses to forest soil extracts treated with PGPR and spruce seedlings, and the Shepard diagram used for multidimensional scaling of the Euclidean distances 321 vii LIST OF TABLES Page Table 2.1 General characteristics of interior spruce and lodgepole pine seedling collection sites 13 Table 2.2 Chemical properties of forest floor soils collected from Mackenzie spruce, Salmon Arm spruce, and Williams Lake spruce and pine stands 24 Table 2.3 Host seedling performance three months after inoculation with selected bacterial isolates in the first screening 26 Table 2.4 Host seedling performance two months after inoculation with the selected bacterial isolates in the second screening 27 Table 3.1 Number of non-mycorrhizal and mycorrhizal seedlings in response to addition of pine and spruce stand soil from Mackenzie, Williams Lake, or Salmon Arm 58 Table 3.2 Number of non-mycorrhizal and mycorrhizal seedlings in response to PGPR inoculation treatments 60 Table 3.3 The effect of PGPR inoculation on seedling height and biomass 61 Table 4.1 Cross reactivity of polyclonal antisera and PGPR strains used as antigens in an ELISA. Values presented are optical densities @ 450nm 84 Table 4.2 Recovery of PGPR from the rhizosphere and internal root and stem tissues of spruce 5 months after seedlings were inoculated 94 Table 4.3 Carbon substrate utilization profiles of PGPR strains as indicated in Biolog™ assays 95 Table 5.1 Site description of the four field sites 106 Table 5.2 Chemical properties of top soils collected from the four field sites 107 Table 5.3 Correlation coefficients (r)1 for seedling survival, biomass and bacterial colonization of roots 120 Table 6.1 Viable population sizes of microorganisms (mean and standard error, log cfu g"1 soil) isolated from two different soils treated with spruce 142 Table 6.2 Root/shoot ratio of spruce seedlings after treatment with PGPR and forest.... 152 viii Table 6.3 Pearson correlation coefficient (/*) between seedling biomass and soil microorganism variables (n= 12) 156 Table A l . 1 Habitat characteristics of isolated PGPR strains 188 Table A2.1 Species identification of the 6 bacterial strains used in this thesis according to gas chromatography of cellular fatty acid methyl-esters (GC-FAME) 190 Table A2.2 Species identification of the 6 bacterial strains used in this thesis according to Biolog™ 192 Table A3.1 Intrinsic antibiotic resistances of PGPR strains Sm3, Ss2, Sw5, L6, Pw2 and S20, and their rifamycin (R) and nalidixic acid (N) resistant derivatives 195 Table A4.1 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 in Screening 1 197 Table A4.2 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 in Screening 1 198 Table A4.3 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 in Screening 1 199 Table A4.4 ANOVA for pine seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 in Screening 1 200 Table A4.5 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 in the presence of forest floor soil in Screening 2.... 201 Table A4.6 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 without forest floor soil in Screening 2 202 Table A4.7 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 in the presence of forest floor soil in Screening 2..... 203 Table A4.8 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 without forest floor soil in Screening 2 204 Table A4.9 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 in the presence of forest floor soil in Screening 2.... 205 Table A4.10 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 without forest floor soil in Screening 2 206 Table A4.11 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 in the presence of forest floor soil in Screening 2 207 ix Table A4.12 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 without forest floor soil in Screening 2 208 Table A4.13 Descriptive statistics for ecotype effects on spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5 209 Table A4.14 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3, Ss2 orSw5 210 Table A4.15 Descriptive statistics for ecotype effects on spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5 213 Table A4.16 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Sm3, Ss2 orSw5 214 Table A4.17 Descriptive statistics for ecotype effects on spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5 217 Table A4.18 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sm3, Ss2orSw5 218 Table A4.19 Descriptive statistics for ecotype effects on spruce seedlings from the Mackenzie, Salmon Arm and Williams Lake seedlots (pooled) inoculated with Pseudomonas strain Sm3, Ss2 or Sw5 221 Table A4.20 ANOVA and orthogonal contrasts of host ecotype effects on spruce seedlings inoculated with Pseudomonas strain Sm3, Ss2 or Sw5 222 Table A4.21 Descriptive statistics for pine ecotype effects using seedlings from Fort St. John, Kamloops, or Williams Lake seedlots inoculated with Bacillus strain Pw2 225 Table A4.22 ANOVA and orthogonal contrasts of pine ecotype effects on seedlings inoculated with Bacillus strain Pw2 227 Table A4.23 Descriptive statistics for host genus effects on performance of spruce or pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R 230 Table A4.24 ANOVA and orthogonal contrasts of host genus effects on performance of spruce or pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R 233 Table A4.25 Descriptive statistics for host genus effects on performance of spruce and pine seedlings (pooled) inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R 239 Table A4.26 ANOVA and orthogonal contrasts of host genus effects on performance of spruce and pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R 241 Table A4.27 Descriptive statistics for soil type effects on spruce seedlings in the presence of soil from the Mackenzie spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R '.. 244 Table A4.28 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings in the presence of soil from the Mackenzie spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 247 Table A4.29 Descriptive statistics for soil type effects on spruce seedlings in the presence of soil from the Salmon Arm spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 250 Table A4.30 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings in the presence of soil from the Salmon Arm spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 252 Table A4.31 Descriptive statistics for soil ecotype effects on spruce seedlings treated with soil from the Williams Lake spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 255 Table A4.32 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Williams Lake spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 257 Table A4.33 Descriptive statistics for soil type effects on spruce seedlings treated with soil from the Williams Lake pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RNorPw2R 260 Table A4.34 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Williams Lake pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 262 Table A4.35 Descriptive statistics for soil type effects on spruce seedlings treated with soil from the Mackenzie spruce, Salmon Arm spruce, Williams Lake spruce or pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 265 xi Table A4.36 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Mackenzie spruce, Salmon Arm spruce, Williams Lake spurce or pine stand inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R 267 Table A5.1 ANOVA for spruce seedling height and biomass (regardless of mycorrhizal status) after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 271 Table A5.2 ANOVA for non-mycorrhizal spruce seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 272 Table A5.3 ANOVA for mycorrhizal spruce seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 273 Table A5.4 ANOVA for pine seedling height and biomass (regardless of mycorrhizal status) after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 274 Table A5.5 ANOVA for non-mycorrhizal pine seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 275 Table A5.6 ANOVA for mycorrhizal pine seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites 276 Table A6.1 Descriptive statistics for spruce seedling biomass before and after outplanting 278 Table A6.2 Descriptive statistics for relative growth rate (RGR) during the period of field growth 281 Table A6.3 ANOVA for seedling growth in the greenhouse after PGPR inoculation 284 Table A6.4 ANOVA for shoot relative growth rate (SRGR) during the period of growth in the field 285 Table A6.5 ANOVA for root relative growth rate (RRGR) during the period of growth in the field 286 Table A6.6 Descriptive statistics for spruce seedling shoot damage (rank) after growth in the field 287 xii Table A6.7 Descriptive statistics for inoculum population sizes reisolated from external and internal root tissues of spruce seedlings before and after the field trials.... 288 Table A7.1 Descriptive statistics for spruce seedling biomass grown under gnotobiotic conditions (sterile sand) in a growth chamber after inoculation with PGPR 291 Table A7.2 Descriptive statistics for spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site 292 Table A7.3 Descriptive statistics for spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing site 293 Table A7.4 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown under gnotobiotic conditions (sterile sand) 294 Table A7.5 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown with soil extract collected from the Smithers Shoe-house site 295 Table A7.6 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown with soil extract collected from the Williams Lake landing site 296 Table A7.7 ANOVA of shoot and root biomass of spruce seedlings inoculated with PGPR and grown without forest soil extract 297 Table A7.8 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house and Williams Like landing sites 298 Table A7.9 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site 299 Table A7.10 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing site 300 Table A7.11 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site 301 Table A7.12 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing 303 Table A7.13 PGPR survival in microcosms treated with two forest soils with or without spruce seedlings 305 Table A7.14 ANOVA of PGPR population sizes recovered from microcosms treated with Smithers and Williams Lake soil and spruce seedlings 306 xiii Table A7.15 ANOVA of PGPR population sizes recovered from microcosms treated with Srnithers and Williams Lake soil and spruce seedlings 307 Table A7.16 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings 309 Table A7.17 ANOVA of soil microbial populations detected in Smithers soil after inoculation with PGPR in the presence or absence of spruce seedlings 311 Table A7.18 ANOVA of soil microbial populations detected in Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings 312 Table A7.19 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the presence of spruce seedlings 313 Table A7.20 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the absence of spruce seedlings 314 Table A7.21 ANOVA of soil microbial populations detected in Smithers soil after inoculation with PGPR in the presence or absence of spruce seedlings 315 Table A7.22 ANOVA of soil microbial populations detected in Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings 317 Table A8.1 Regression equation of optical density (Aeoo) and mean generation time of six PGPR strains 320 Table A9.1 Euclidean distances for Biolog™ plate responses to forest soil extracts treated with PGPR and spruce seedlings 322 xiv LIST OF FIGURES Page Fig. 2.1 Effect of ecotype specificity between conifers and PGPR strains on seedling emergence value, (a): Percent change from the uninoculated control for each bacterial treatment. -I indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean 30 Fig. 2.2 Effect of ecotype specificity between conifers and PGPR strains on shoot height 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment. -I indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean 31 Fig. 2.3 Effect of ecotype specificity between conifers and PGPR strains on shoot biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment. -I indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean 32 Fig. 2.4 Effect of ecotype specificity between conifers and PGPR strains on root biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment. -I indicates a coexistent host ecotype-bacterial inoculum combination; * indicates the mean value signifcantly differs from the uninoculated control (P<0.\); Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean 33 xv Fig. 2.5 Effect of host genus specificity between conifers and PGPR strains on seedling emergence value, (a): Percent change from the uninoculated control for each bacterial treatment; -l indicates a coexistent host genus-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean 35 Fig. 2.6 Effect of host genus specificity between conifers and PGPR strains on shoot height 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment; -l indicates a coexistent host genus-bacterial inoculum combination; ** and * indicate the mean values differ significantly from uninoculated controls at P<0.05 and P<0.1, respectively; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean.... 36 Fig. 2.7 Effect of host genus specificity between conifers and PGPR strains on shoot biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment; -l indicates a coexistent host genus-bacterial inoculum combination; * indicates the mean value differs significantly from uninoculated controls at P<0.1; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean 37 Fig. 2.8 Effect of host genus specificity between conifers and PGPR strains on root biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment: -l indicates a coexistent host genus-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean 38 Fig. 2.9 Effect of host genus specificity between conifers and PGPR strains on rhizosphere colonization 14 weeks after sowing and inoculation, (a): Size of recovered population of each inoculum: bars indicate means and standard errors; 4- indicates a coexistent host genus-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains 39 xvi Fig. 2.10 Effect of forest soil and PGPR inoculation on seedling emergence value, (a): Percent change from the uninoculated control of each bacterial treatment. -I indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean 41 Fig. 2.11 Effect of soil forest soil and PGPR inoculation on shoot height, (a): Percent change from the uninoculated control of each bacterial treatment, i indicates a coexistent combination, i.e., common origin, between soil and inoculum; ** and * indicate the mean values differ significantly (P<0.05 and P<0.1, respectively) from uninoculated controls; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean 42 Fig. 2.12 Effect of forest soil and PGPR inoculation on shoot biomass. (a): Percent change from the uninoculated control of each bacterial treatment. X indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean 43 Fig. 2.13 Effect of forest soil and PGPR inoculation on root biomass. (a): Percent change from the uninoculated control of each bacterial treatment. 4^  indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean 44 xvii Fig. 2.14 Effect of forest soils and PGPR inoculation on rhizosphere colonization 15 weeks after sowing and inoculation, (a): Size of recovered population of each inoculum. Bars indicnate means and standard errors. -I: coexistent combination between a forest soil and inoculum; MZ: Mackenzie site soil, SA: Salmon Arm site soil, WL: Williams Lake site soil, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains 46 Fig. 3.1 Correlation between seedling biomass and the number (log transformed) of mycorrhizal root tips detected in root systems after inoculation with.sterile phosphate buffer (uninoculated control) for (a) spruce and (b) pine seedlings. #, Wilcoxina sp. (E-strain); • ; Thelephora sp.; A, others (Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings. P=sigriificance level associated with each r 62 Fig. 3.2 Correlation between spruce seedling biomass and the number (log transformed) of mycorrhizal root tips detected in seedling root systems after inoculation with PGPR strains. • , Wilcoxina sp. (E-strain); • ; Thelephora sp.; A, others (Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings, insignificance level associated with each r 63 Fig. 3.3 Correlation between pine seedling biomass and the number (log transformed) of mycorrhizal root tips detected in seedling root systems after inoculation with PGPR strains. • , Wilcoxina sp. (E-strain); • ; Thelephora sp.; A, others (Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings. P=significance level associated with each r 64 Fig. 3.4 Treatment means (and standard errors) for (a) shoot and (b) root biomass of non-mycorrhizal and mycorrhizal spruce seedlings inoculated with a PGPR strain or sterile phosphate buffer (control). ** and * indicate significant difference from control at P<0.05 and P<0.1, respectively. See Table 3.2 for n... 66 Fig. 3.5 Treatment means (and standard errors) for (a) shoot and (b) root biomass of non-mycorrhizal and mycorrhizal pine seedlings inoculated with a PGPR strain or sterile phosphate buffer (control). ** and * indicate significant difference from control at T^O.05 and P<0.1, respectively. See Table 3.2 for n 67 Fig. 3.6 Rhizosphere colonization of (a) spruce and (b) pine seedlings by Pseudomonas strains Sm3RN, Ss2RN and Sw5RN (means and standard errors; n=7-17) 68 xviii Fig. 4.1 Immunofluorescent antibody staining of Bacillus strain L6-16R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Images were produced using phase contrast microscope under (a) fluorescent light (495 nm) and (b) visible light. Bacillus appears as tiny, fluorescent green cells associated with root surface and hairs. The particle seen in the bottom, center of each micrograph is a Turface aggregate 85 Fig. 4.2 Immunofluorescent antibody staining of Pseudomonas strain Sm3RN in a longitudinal section of a 4-month old spruce root inoculated at sowing. Images were produced using phase contrast microscope under (a) fluorescent light (495 nm) and (b) visible light. Pseudomonas appears as green fluorescent cells associated with root surface and hairs 86 Fig. 4.3 Immunofluorescent antibody staining of Bacillus strain S20R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as tiny, fluorescent green cells associated with root surface and hairs 87 Fig. 4.4 Immunofluorescent antibody staining of Pseudomonas strain Ss2RN in a longitudinal section of the central axis of a 5-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent green cells associated with root surface and hairs 88 Fig. 4.5 Immunofluorescent antibody staining of Pseudomonas strain Sw5RN in a longitudinal section of the central axis of a 5-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent green cells associated with root surface and hairs 89 Fig. 4.6 Immunofluorescent antibody staining of Bacillus strain Pw2R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as fluorescent green cells inside root tissues near the base of a lateral root 90 Fig. 4.7 Immunofluorescent antibody staining of Pseudomonas strain Sm3RN in a longitudinal section of the stem of a 4-month old spruce seedling that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent green cells inside vascular tissues 91 xix Fig. 4.8 Immunofluorescent antibody staining of Bacillus strain Pw2R in a longitudinal section of the stem of a 4-month old spruce seedling that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as fluorescent green cells inside vascular tissues 92 Fig. 5.1 Seedling arrangement in field trials at (a) the Smithers Shoe-house site and (b) the Williams Lake Regular site. Four-month-old seedlings were planted 1 m apart each other according to a randomized complete block design (n = 27) 108 Fig. 5.2 Seedling shoot injury was rated visually using a four-rank scale based on the percentage of dead needles. Category "0" seedlings had 90 - 100% needle mortality; category "1" had 50 - 90% needle mortality; category "2" had 10 -50% needle mortality and category "3" had less than 10% needle mortality 110 Fig. 5.3 (a) Shoot and (b) root biomass of spruce seedlings (Mackenzie and Williams Lake ecotypes) 16 weeks after inoculation with PGPR strains in the greenhouse. Error bars indicate the standard errors of the means. ***, **, and * indicate the mean is significantly different from uninoculated control at P <0.01, PO.05, P<0.1, respectively (Fisher's protected LSD) 112 Fig. 5.4 Relative growth rates of (a) shoot and (b) root biomass of seedlings outplanted at 4 field sites. Error bars indicate the standard errors of the means. ***, **, and * indicate the mean is significantly different from uninoculated control at PO.01, P<0.05, P<0.1, respectively (Fisher's protected LSD) 113 Fig. 5.5 Shoot injury rank of each PGPR treatment at each field site. Error bars indicate the standard errors of the means. Injury ranks were: dead = 0, heavily injured (more than 50% of dead needles) = 1, partly injured (less than 50% of dead needles) = 2, and no detectable injury = 3 114 Fig. 5.6 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Smithers Blunt Creek site for 4 months 116 Fig. 5.7 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Smithers Shoe-house site for 4 months 117 Fig. 5.8 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Williams Lake regular cut-block site for 4 months 118 Fig. 5.9 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Williams Lake landing site for 4 months 119 Fig. 6.1 Examples of GN Biolog™ plate responses, i.e., carbon substrate utilization patterns of soil microbial communities after 48h incubation. Smithers and -2 Williams Lake site soil extracts (1.7 x 10 dilution) were used 138 xx Fig. 6.2 Number of Biolog™ substrates used by the two forest soil extracts after inoculation with PGPR strains in the presence and absence of spruce seedlings... 143 Fig. 6.3 Shannon diversity index (FT) for Biolog™ plate responses to the two forest soil extracts after inoculation with PGPR strains in the presence and absence of spruce seedlings 144 Fig. 6.4 Multidimensional scaling of carbon substrate utilization of two forest soils with or without spruce seedlings after inoculation with PGPR. L6:L6-16R; Pw2:Pw2R; S20:S20R; Sm3:Sm3RN; Ss2:Ss2RN; Sw5:Sw5RN; and C:uninoculated control. •:Smithers soil with seedling; •:Smithers soil without seedling; OiWilliams Lake soil with seedling; and •:Williams Lake soil without seedling 145 Fig. 6.5 (a) Shoot and (b) root biomass of spruce seedlings grown under gnotobiotic conditions after inoculation with PGPR (n=20). Error bars indicate the standard errors of the means 147 Fig. 6.6 (a) Shoot and (b) root biomass of spruce seedlings after PGPR inoculation with or without soil extract originated from the Smithers Shoe-house site (n=18). Error bars=the standard errors of the means; * indicates that the mean value differs from that of uninoculated control with P<0.1 (two tailed) 148 Fig. 6.7 (a) Shoot and (b) root biomass of spruce seedlings after PGPR inoculation with or without soil extract originated from the Williams Lake landing site (n=18). Error bars=the standard errors of the means; *** and * indicate that the mean values differ from that of uninoculated control withPO.Ol and P<0.1, respectively 149 Fig. 6.8 Percent change of shoot and root biomass from uninoculated control after inoculation with Bacillus PGPR strains (L6-16R, Pw2R, and S20R) with or without soil extracts collected from Smithers and Williams Lake 150 Fig. 6.9 Percent change of shoot and root biomass from uninoculated control after inoculation with Pseudomonas PGPR strains (Sm3RN, Ss2RN, Sw5RN) with or without soil extracts collected from Smithers and Williams Lake 151 Fig. 6.10 Recovery of PGPR from spruce grown with or without forest soil extracts. Error bars indicate standard errors of the means (n=4) 154 Fig. 6.11 Recovery of the inoculated PGPR strains with or without spruce seedlings after amendment with forest soil extracts. Error bars indicate standard errors of the means (n=4) 155 xxi Fig. 7.1 Percent changes of seedling biomass with each PGPR inoculation from uninoculated control during this research project. Scr: screening; Exp: Experiment; (SBC): Smithers, Blunt Creek site; (SSH): Smithers, Shoe-house site; (WLR): Williams Lake, Regular cut-block site; (WLL): Williams Lake, Landing site; (Sst): sterile Smithers soil added; (Wst): sterile Williams Lake soil added; (Snst): non-sterile Smithers soil added; and (Wnst): non-sterile Williams Lake soil added 164 Fig A9.1 Shepard diagram of distances versus input dissimilarities obtained from the carbon substrate utilization patterns of 28 soil microbial communities. Each number represents a pair of the 378 combinations listed in Table A9.1 324 xxii LIST OF ABBREVIATIONS Variables: ANOVA = Analysis of variance; BGCZ = Biogeoclimatic zone; cfu = Colony forming unit; DCA = Detrended correspondence analysis; DF = Degrees of freedom; DRB = Deleterious rhizobacteria; DRMO = Deleterious rhizosphere microorganisms; ELISA = Enzyme-linked immunosorbent assay; ESSF = Engelmann spruce and subalpine fir; FITC = Fluorescein isothiocyanate; GV = Germination value; IAA = Indole 3-acetic acid; IF = Immunofluorescence; IFAS = Immunofluorescent antibody staining; IgG = Immunoglobluin G; ISR = Induced systemic resistance; KB A - King's B agar; KBB = King's B broth; LSD = Least significant difference; MDS = Multidimensional scaling; MHB = Mycorrhization helper bacteria; MS = Mean squares; n = Sample size; P = Probability value of type I error; xxiii PAR = Photosynthetically active radiation; PBS = Phosphate buffered saline; PC A = Principal components analysis; PCR = Polymerase chain reaction; PGPR = Plant growth promoting rhizobacteria; PPD = /?-phenylenediamine; r = Pearson's correlation; 2 r - Coefficient of determination between two variables; SBS = Sub-boreal spruce; SD = Standard deviation; SE = Standard error of the mean; SEM = Scanning electron microscopy; SEV = Seedling emergence value; SPB = Sterile phosphate buffer; SS = Sum of squares; TEM = Transmission electron microscopy; TSA = Tryptic soy agar; TSB = Tryptic soy broth. xxiv A C K N O W L E D G E M E N T S I would like to express my heartfelt appreciation to all of the people who provided me not only their thoughtful and valuable advice but also tangible and intangible supports during the course of my research. Thank you: Chris Chanway, Brian Holl, Bill Ramey, Colette Breuil, Xeumei Li, Rob Guy, Lynne Tolland, John Markham, Laura Lazo, Dermat Flanagun, Hugues Massicotte, Linda Tackaberry, Sandra Jungwirth, Liz Bent, Daniel Petersen, Murali Srinivasan, Bob Copeman, Seane Trehearne, Bart van der Kamp, Karl Klinka, Salim Silim, Gordon Kayahara, Pal Varga, Mike Brady, Bob Brett, Sonya Budge, Dave Coates, Yoshi Wada, Aki Nemoto, Tatsu and Yumi Ueki, Makoto Ando, Barbara Kishchuk, Gordon Weetman, Hamish Kimmins, Cindy Prescott, Lisa Zabek, Daniel Mailly, Leanne McKinnon, Sandra Millard, Anliang Zhong, Jeff and Hiroko Glaubitz, Kevin and Masami Fox, Rodney and Debbie Keenan, Katsura and Kazuko Nakano, Hiromi and Marc Pingrey, Alice and Jay Hansen, Yoko and Takashi Hamamura, Maruyama family, Shirase family, Shamoto family, Nakajima family, Saito family, Kazuhiro and Yukiko Yagi, Junji Takahashi, Tatsuji Takahashi, Wataru Iida, Etsuko Adachi and the staff of Foundation for Advanced Studies on International Development (FASID), and Yuki Shishido. xxv CHAPTER 1 GENERAL INTRODUCTION Forest resources are of great economic importance in Canada. In 1995, forestry contributed over $30 billion to Canada's balance of trade — more than agriculture, mining and energy combined. Almost one half of this total was generated from British Columbia's forests, rendering it the single most important source of revenue for the province. One of the most important commercial species (or group of species) in British Columbia is interior spruce, a hybrid of white (Picea glauca (Moench) Voss) and Engelmann spruce (Picea engelmannii Parry), which is characterized by relatively slow growth rates, shade tolerance and a preference for moist, nutrient rich soils (Klinka et al. 1990). While interior spruce comprised 19% of the annual cut in British Columbia in 1994-95, it accounted for over 40% of the 250 million seedlings planted province-wide during the year 1994-95, a figure which is typical of the past several years. This means that overlOO million spruce seedlings are planted each year in B.C., but plantations particularly in boreal, montane and subapline environments may still be problematic due to seedling transplant shock, manifest by restricted shoot growth after outplanting (Sutton 1980; Sutton and Tinus 1983). Such reduced shoot growth after planting may occur to facilitate vigorous root growth so that adequate root-soil contact can be established for water and nutrient extraction. Because inadequate root development is a common characteristic of failing outplanted seedlings, vigorous root growth soon after outplanting is thought to be essential for successful seedling and plantation establishment (Burdett etal. 1984; Vyse 1981). In addition to transplant shock and poor root growth, spruce seedling performance may also be restricted by its intrinsically slow growth rate, as well as its low tolerance to direct sunlight (Chapin et al. 1986; Smirnoff et al. 1984) which is abundant at cut-over sites. These factors render spruce seedlings vulnerable to overtopping by competing vegetation such as fireweed (Epilobium angustifolium L.), and by plantation failure (Newton and Comeau 1990). Even though the average conifer seeding survival rate two years after planting is estimated to be 85%, 1 slow growing spruce plantations contribute to the 1.29 million ha of forest lands that are classified as not satisfactorily restocked (B.C. Ministry of Forests, 1993-94 Annual Report). Mechanical site preparation, fertilization and control of competing vegetation can improve plantation performance, but associated costs are substantial. An inexpensive, environmentally benign alternative for enhancing productivity of newly established plantations involves nursery inoculation of seedlings with root-associated plant growth stimulating microorganisms. In forestry, this has generally been restricted to inoculation with mycorrhizal fungi, a practice that can enhance seedling survival or growth, particularly on sites where appropriate fungal symbionts are not present (Kropp et al. 1985; Miller et al. 1992; Sinclair and Marx 1982). While promising results have been observed in many trials, seedlings do not always respond adequately to inoculation with mycorrhizal fungi (Colinas etal. 1994b; Kropp and Langlois 1990; Loopstra et al. 1988). Notwithstanding the traditional focus on mycorrhizal aspects of conifer seedling root microbiology, in nature, plant roots, mycorrhizae, and associated rhizosphere soil are colonized heavily by asymbiotic soil microorganisms. Of these, bacteria are often most abundant (Rovira and Davey 1974). The rhizosphere, which was first described by Hiltner (1904) as the area of intense microbial activity around legume roots, is now defined more broadly, and includes the zone of soil surrounding plant roots and influenced by their metabolism (see Curl and Truelove 1986; Garbaye 1991; Gaskins etal. 1985; Lynch 1990, 1994; Rovira and Davey 1974; Whipps and Lynch 1986 for reviews). It has been delineated into three regions: the root surface or rhizoplane, the surrounding soil or the rhizosphere, and the internal root tissue, sometimes referred to (inappropriately!) as the endorhizosphere (Kloepper et al. 1992b). It is generally agreed that rhizodeposition, i.e., the loss of organic matter from roots as they grow through the soil, provides a significant source of nutrients for microorganisms living in the vicinity of plant roots (Whipps 1990). From a microbial perspective, root exudates, which are soluble, low molecular weight compounds that leak from plant roots and that are easily oxidized by soil microorganisms (Richards 1987), are particularly important nutritional components of 2 plant rhizodeposits. Organic compounds contained in root exudates and in senescent root tissues provide substrates for the growth of heterotrophic soil microorganisms in the rhizosphere in notable quantities. Radio-isotope labeling experiments have indicated that up to 40% of cereal (Whipps and Lynch 1986) and 50% of conifer (Perry et al. 1987; Reid and Mexal 1977) net primary production can be exuded into the rhizosphere. This results in deposition of up to 2.9 and 7.5 tonnes-ha 1 of organic carbon compounds per year in cultivated wheat (Triticum aestivum L.) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco.) forest ecosystems, respectively (Whipps 1990). The allocation of such a large amount of photosynthate below ground reflects the potential importance of rhizosphere ecology in plant growth. Indeed, Ryan and Jagendorf (1995) suggested that plants may actually cultivate populations of beneficial, asymbiotic microorganisms on their root systems through root exudation and rhizodeposition. The large amount of readily available, easily oxidizable nutrients in soil surrounding roots results in the well-known rhizosphere effect. Simply stated, it is the increase in the number of microorganisms in the rhizosphere, particularly bacteria, relative to non-rhizosphere soil (Rovira and Davey 1974). The rhizosphere effect can be quantified by determining the ratio of microbial numbers per unit mass of rhizosphere soil (R) to the numbers per unit mass of non-rhizosphere soil (S) (Katznelson 1946). This proportion, termed the R:S ratio, typically ranges from 10-50:1 for bacteria (Richards 1987), but is smaller for other soil organisms. Bacteria can attain populations of up to 3 x 109 cells-g"1 of soil in the rhizosphere (Rouatt and Katznelson 1961). In general, it is agreed that the rhizosphere effect is a dynamic process which is driven by root exudation and the release of other organic compounds (i.e., rhizodeposition). Therefore, it is inevitable that the rhizosphere effect is influenced by several factors including (1) identity, age and developmental stage of the plant species; (2) fertility, moisture level and physical properties of the soil; (3) environmental conditions such as light and temperature; (4) cultural practices including foliar application of chemicals; (5) soil microbial interactions; and (6) the presence of neighbouring plant species (Curl and Truelove 1986). 3 Rhizosphere bacteria are well-known to affect plant growth both positively and negatively (Kloepper 1993; Nehl et al. 1996). The term "plant growth promoting rhizobacteria" (PGPR) is used to describe strains of naturally-occurring soil bacteria that have the capability to significantly stimulate plant growth (Kloepper and Schoth 1978), sometimes doubling root and/or shoot biomass within a few weeks of inoculation (Kloepper et al. 1989). PGPR commonly belong to the genera Arthrobacter, Azospirillum, Azotobacter, Bacillus, Pseudomonas, or Serratia (Brown 1974; Gaskins et al. 1985). Because of their potential impact on crop productivity, PGPR have been studied extensively in agriculture (Burr and Caesar 1984; Gaskins et al. 1985). Crop biomass gains in response to bacterial inoculation have been observed on the yield of wheat (Kapulnik et al. 1985), height of onion (Allium cepa L.) (Reddy and Rahe 1989), root number of millet (Panicum miliaceum L.) (Tien et al. 1979) and shoot and root dry weight of many other crops (Kloepper et al. 1989). PGPR have also been shown to promote the growth of deciduous fruit species including apple (Malus sylvestris (L.) Mill.) and citrus (Citrus spp.) seedlings (Caesar and Burr 1987; Gardner et al. 1984). It has been speculated that PGPR elicit greatest effects during the early growth stages of most plants, but on all growth stages of short-season vegetables (Burr and Caesar 1984). In contrast to our knowledge of PGPR in agriculture and horticulture, there were only a few reports describing PGPR effects on conifer species before 1990; growth of Scots pine (Pinus sylvestris L.) was increased by inoculation with an unidentified coryneform bacterial isolate or with the supernatant of its growth medium (Pokojska-Burdziej 1982), and container-grown Douglas-fir showed significant increases in height and diameter after inoculation with a mixed suspension of forest floor microorganisms (Parker and Dangerfield 1975). More recently, the potential utility of pure culture conifer inoculation has been investigated with isolates of the bacterial genera Bacillus, Pseudomonas, and Arthrobacter. These isolates have been shown to induce significant growth stimulation of Douglas-fir (Chanway and Holl 1992), hybrid spruce (Picea glauca x engelmannii) (Chanway and Holl 1993a; O'Neill et al. 1992), lodgepole pine (Pinus contorta Dougl. var. latifolia) (Chanway and Holl 1994b; Holl and Chanway 1992), and western hemlock (Tsuga heterophylla (Raf.) Sarg.) (Chanway 1995) under both controlled (growth chambers) and uncontrolled (field) conditions. Results of a co-inoculation study with PGPR (Bacilluspolymyxa) and mycorrhizal fungi (Wilcoxina mikolae) suggest that growth promotion of lodgepole pine seedlings is not dependent on mycorrhizal status of the seedlings (Chanway and Holl 1991). Though not entirely understood, several mechanisms have been proposed to explain how PGPR stimulate plant growth. These mechanisms include suppression or neutralization of plant growth-inhibiting soil microorganisms (Kloepper et al. 1980a; van Peer and Schippers 1989) and compounds (Glick 1994a), increased availability of nutrients (Powell etal. 1980), production of plant growth regulating phytohormones (Kloepper et al. 1989) and stimulation of mycorrhizal infection (Garbaye 1994; Linderman 1988). It is difficult to determine with certainty which mechanism causes growth promotion for any given PGPR especially considering that many bacterial strains possess in vitro attributes consistent with more than one of these proposed mechanisms (Chanway 1997). Not unlike plant growth responses to inoculation with mycorrhizal fungi, however, numerous studies indicate that PGPR inoculation effects are also variable (Kloepper et al. 1989). Such inconsistency of growth promotion is an impediment to the commercial implementation of PGPR technology. One possible factor contributing to inconsistent seedling growth responses is variability in root colonization by PGPR. It is generally accepted that bacteria must be competent root colonizers to be effective PGPR (Schroth and Weinhold, 1986), but if threshold PGPR population sizes are required to elicit positive plant growth responses, they have not been determined. Nevertheless, interference or competition from other indigenous microorganisms may alter root colonization by PGPR and indirectly contribute to inconsistency of inoculation effects. Variable seedling growth responses to PGPR inoculation may also be related to other biotic effects such as host plant x PGPR strain specificity. In some cases, growth promoting rhizobacteria have been shown to be effective with only certain plant genotypes or species (Chanway et al. 1991b), while other PGPR exert broad spectrum plant growth promotion effects 5 (Bashan et al. 1989; O'Neill et al. 1992). The degree of such specificity needs to be defined before particular PGPR strains and host plants may be considered for use in a larger scale inoculation program. Other biotic effects in the rhizosphere such as metabolism of PGPR-produced plant growth promoting compounds or their precursors by neighbouring microbes as well as abiotic factors such as suppression of PGPR growth or activity due to environmental extremes may also be important factors in seedling growth response variability. It should also be noted that some PGPR are capable of colonizing internal root tissues, including the root cortex and vascular tissues (Lalande et al. 1989; van Peer and Schippers 1989). We are only now beginning to understand how such bacteria gain entry into root tissues and which tissues are colonized after initial entry (Mahafee et al. 1994; Quadt-Hallman and Kloepper 1996). Furthermore, some root endophytic PGPR have been isolated from internal shoot tissues of inoculated plants, indicating their ability to spread systemically within host plants (Quadt-Hallman and Kloepper 1996). It is tempting to speculate that endophytic PGPR may be more consistent in seedling growth promotion efficacy compared to external root colonizers due to possible added protection from environmental extremes within root tissues and the more intimate association of PGPR with the host plants compared to that of rhizosphere associations. If this is true, then endophytic PGPR may be able to circumvent at least some of the variability that has been observed with enhanced root tissue colonizing PGPR, however, such a hypothesis requires careful testing. In theory then, it should be possible to exploit mutualistic relationships between interior spruce and associated rhizosphere bacteria to aid in the growth of spruce soon after outplanting, particularly at less productive sites (Chanway and Holl 1993b). However, information on these plant-microbe relationships is lacking in several key areas. In order to address some of the deficiencies in our knowledge of PGPR for conifers and possible applications in reforestation, a series of experiments was conducted to further elucidate the relationship between PGPR strains, spruce and pine seedlings and the soils in which they grow. The general objectives of this work were to isolate PGPR that may be useful for conifer plantation forestry and to identify some 6 possible factors that could contribute to seedling growth response variability after treatment with PGPR. The specific research objectives and corresponding thesis chapters devoted to each objective are: Objective 1. To investigate specificity involving conifer host plants, PGPR strains and the soils from which they originated in relation to seedling growth promotion and growth response variability (Chapter 2); Objective 2. To assess the interaction of PGPR and mycorrhizal fungi in relation to growth promotion of spruce and pine seedlings (Chapter 3); Objective 3. To evaluate internal root colonization of spruce seedlings by PGPR and to determine internal seedling tissue(s) that endophytic PGPR colonize (Chapter 4); Objective 4. To determine if PGPR growth promotion efficacy on spruce seedlings is related to external or internal root colonization under field conditions (Chapter 5); Objective 5. To determine if the mechanism by which PGPR stimulate spruce seedling involves other microorganisms (Chapter 6); and Objective 6. To assess interactions between PGPR, spruce seedlings and indigenous forest soil communities in relation to possible negative effects of the introduced bacteria (Chapter 6). 7 CHAPTER 2 SPECIFICITY OF PGPR STRAINS FOR SPRUCE AND PINE 2.1 INTRODUCTION AND LITERATURE REVIEW Qualitative and quantitative differences in root exudates between plant species as well as cultivars and genotypes of the same species are known to exist (Baldani and Dobereiner 1980; Curl and Truelove 1986). These differences, coupled with the reliance of rhizosphere microorganisms on root exudates for nutrients, may result in the proliferation of microbial populations that are specific to plant species or to genotypes within species (Burr and Caesar 1984; Chanway et al. 1991b; Neal et al. 1973), and ultimately in PGPR x host plant growth response specificity. Chanway et al. (1991b) recognized two possible levels of plant-microorganism specificity: (i) during infection by the beneficial microorganisms, and (ii) the subsequent effectiveness of the association on plant growth. Where the relationship is not symbiotic but associative (e.g., PGPR), specificity was postulated to occur during microbial colonization. Infection and colonization specificity may be determined by a cellular recognition mechanism. For example, in the Rhizobium-legume symbiosis, it is well-known that plant lectins operate in a manner similar to that of antigens in immunological reactions (Dazzo and Brill 1978; Dazzo and Hubbell 1975). Lectins are glycoproteins specific to some plant species that adhere to the polysaccharide portion on the outer layer of the bacterial cell wall. The idea that lectins are important in Rhizobium-legume specificity has been investigated for more than two decades, and the focus of specificity studies has moved to the involvement of molecular signals between host plant and rhizobia. Long (1989) reviewed evidence that flavonoids and isoflavonoids from host plant roots activate nodulation genes in specific Rhizobium strains. Though colonization specificity of certain saprophytic pseudomonads has been reported (Anderson 1983), the possible involvement of lectins or lectin-like compounds in plant x PGPR interactions requires more study. Agglutinin, a glycoprotein complex released from root surfaces, 8 agglutinates Pseudomonas putida and promotes rapid attachment to sterile root surfaces (Anderson et al. 1988). However, no crop-specific agglutination of fluorescent Pseudomonas isolates was detected between potato (Solanum tuberosum L. cv. Bintje), grass (Lolium multiflorum L. cv. Tiara) and wheat (Triticum aestivum L. cv. Minaret) (Glandorf et al. 1993). Further investigation (Glandorf et al. 1994) suggested that root agglutinins can be involved in short-term adherence of pseudomonads to roots, but that they do not play a decisive role in long-term bacterial colonization. Specificity at effectiveness level may result from bacterial production of compounds that specifically affect growth of the host plant either qualitatively or quantitatively. Symbiotic nitrogen fixation, for example, involves the activity of the enzyme, nitrogenase. Genes directing specific steps in nodulation of a legume by a strain of Rhizobium have been called nod genes. Many nod genes from different Rhizobium species are generally borne on Sym plasmids, which also contain specificity genes, and thereby restrict a strain of Rhizobium to a particular host plant (Paul and Clark 1989). However, the environment plays an important role in these interactions as demonstrated with white clover (Trifolium repens L.) and Rhizobium leguminosarum bv. trifolii (Mytton and Hughes 1984; Mytton and Livesly 1983). Studies of PGPR effectiveness indicate that some strains have the capacity to promote growth of several plant species (Bashan et al. 1989; Holl et al. 1988), but PGPR are not universally effective as differences in growth promotion between various PGPR-plant combinations have been well documented (Chanway et al. 1990; Gardner et al. 1984, 1985). An illustrative example of such specificity involves Azospirillum and cereal crops: Azospirillum brasilense is most frequently isolated from the roots of wheat, whereas Azospirillum lipoferum is commonly isolated from the roots of sorghum and maize. Both bacterial species are often cited as PGPR of the plant species from which they were isolated, but they rarely promote the growth of other species (Sumner 1990). Chanway et al. (1988) found striking evidence of PGPR x plant cultivar specificity with wheat, but studies were conducted under gnotobiotic conditions in the absence of representative 9 field conditions. When such specificity does occur, Chanway et al. (1991b) suggested it may have developed in either of two ways: (i) from pre-existing genetic variability among resident soil bacteria in which particular bacterial strains may have experienced a competitive advantage over other strains in the rhizosphere of a particular plant genotype or species, or (ii) through adaptation of particular bacterial strains to the plant host. The latter mechanism would be facilitated by the generation of genetic variation in the bacterial population arising from point mutations and/or genetic recombination with subsequent selection of superior bacterial genotypes in the rhizosphere. Reviews by Perry et al. (1987, 1989) reinforce the idea that ecological adaptation of below-ground microflora to plant hosts may occur in forest ecosystems. Amaranthus and Perry (1987) demonstrated a substantial influence of including the "correct" (i.e., adapted) soil biota on conifer seedling growth at degraded cut-over sites in southern Oregon and northern California. They found that inoculating Douglas-fir seedlings with soil from a healthy stand and planting at a site with a history of plantation failures significantly improved seedling biomass accumulation and survival. Mycorrhizal fungi have often been implicated as the causal agent in such soil transfer studies, but other soil biota including rhizobacteria may also be involved. O'Neill et al. (1992) demonstrated some degree of adaptation involving seedlings and rhizobacteria from spruce forests. However, growth response specificity was not as apparent as it was in studies with cloned perennials (white clover and ryegrass) or inbreeding annuals (wheat) (Chanway et al. 1988, 1990), and no strong ecotype specificity was found between Douglas-fir seedlings and associated PGPR strains (Chanway and Holl 1992). Since conifer species produce highly heterozygous progenies through outcrossing (Khalil 1986), use of naturally pollinated seeds instead of clonal materials may have prevented detection of fine-scale ecological relationships involving plants and PGPR in these studies. However, a lack of such specificity may be advantageous from a practical point of view, as closely matching seedlots with inoculum strains would not be necessary. 10 Therefore, before applying PGPR technology in forest management, it is necessary to understand the degree of adaptation, if any, between conifer hosts and PGPR strains. To this end, I evaluated the degree of seedling growth specificity in response to "coexistent" and "unrelated" bacterial inocula. The term "coexistent" is used to describe seedlings, bacteria, and soil with a common geographical origin, whereas "unrelated" is used for those originated from disparate locations. The specific hypotheses I tested were: Hypothesis 1. Growth of seedlings belonging to a conifer (spruce and pine) ecotype within a species will be greater after inoculation with coexistent PGPR compared to inoculation with PGPR isolated from a different ecotype. Hypothesis 2. Growth of seedlings belonging to a conifer genus (spruce and pine) will be greater after inoculation with coexistent PGPR compared to inoculation with PGPR isolated from a different conifer genus. Hypothesis 3. Spruce seedling growth promotion will be greatest when grown with forest soil and PGPR with a common origin compared to seedling growth when soil and PGPR are unrelated. 11 2 .2 MATERIALS AND METHODS 2 .2 .1 Screening for PGPR strains for interior spruce and lodgepole pine seedlings 2.2.1.1 Sampling of seedlings and soils. Twenty naturally-regenerating lodgepole pine and interior spruce seedlings (1-3 years of age) were collected during the summer of 1992 from the understorey of each of six different forest stands (three stands per conifer genus) in various biogeoclimatic zones in British Columbia. Physical characteristics of the collection sites are shown in Table 2.1. The root mass of each seedling with adhering soil to a depth of ca. 20 cm was kept intact refrigerated in a cooler for a maximum of three days until dilution plating was performed. Seed of interior spruce and lodgepole pine from these six sites were obtained from the B.C. Ministry of Forests, Surrey Seed Centre, and were used in greenhouse bioassays. 2.2.1.2 Isolation and storage of bacteria. Twenty seedlings collected from each sampling location were divided into four groups of five for bacterial isolation. For dilution plating, pooled root masses of each group of seedlings were shaken vigorously and then cut into 2 cm segments. Root segments (2.0 g) from each group were placed in a flask containing glass beads and 98 mL of 10 mM sterile phosphate buffer (SPB). Flasks were agitated on a rotary shaker (200 rpm) for 20 min. Resultant suspensions were serially diluted and 0.1 mL aliquots of each dilution were spread on three different media, tryptic soy agar (TSA), King's B agar (KBA) (Starr et al. 1981), and combined carbon agar (CCA) (Rennie 1981). All media were supplemented with 100 mg-L 1 cyclohexamide and 30 mg-L"1 benomyl. Plates were incubated aerobically at 28°C for 3-5 days. To isolate root endophytic bacteria, 10 washed root segments per group were surface-sterilized by soaking for 3 min in 2% sodium hypochlorite (NaCIO) followed by rinsing in sterilized water. Root samples were checked for surface contamination by imprinting on 50% strength TSA, and incubating plates for 3 days at 28°C. Uncontaminated roots (ca. 0.2 g) were 12 Table 2.1 General characteristics of interior spruce and lodgepole pine seedling collection sites. Site No. Host species Location (Seedlot No.)1 Latitude Longitude Altitude Biogeoclimatic Zone Stand age (years) 1 Interior spruce Mackenzie (29144) 55°11' 122°58' 780 m Sub-Boreal Spruce > 100 2 Interior spruce Salmon Arm (3010) 51°04' 119°26' 1250 m Montane Spruce > 100 3 Interior spruce Williams Lake (4038) 52°33' 122°06' 884 m Sub-Boreal Spruce > 100 4 Lodgepole pine Fort St. John (2116) 56°30' 121°30' 762 m Boreal White and Black Spruce 40 ~ 100 5 Lodgepole pine Kamloops (2176) 50°55' 120°05' 1402 m Montane Spruce 40-100 6 Lodgepole pine Williams Lake (2633) 52°06* 122°58' 1311 m Sub-Boreal Pine Spruce 40-100 Provided by the British Columiba Ministry of Forest. Table 2.1 Continued. Site No. Mineral soil taxa (Great Group) Mean daily temperature range (°Q Mean annual precipitation (range) (mm) Number of frost free days (range) Annual productivity (range) / 3 , - 1 -k (m ha yr ) 1 Humo-Ferric Podzol -14.6-16.9 500 - 1000 60 - 140 3.5-6.3 2 Humo-Ferric Podzol -10.9-13.3 300 - 500 60 - 140 3.5-6.3 3 Gray Luvisol -14.6-16.9 300 - 500 60 - 140 3.5-6.3 4 Gray Luvisol -24.5 - 16.6 500 - 1000 <60 0.8-3.4 5 Gray Luvisol -10.7-20.8 300 - 500 60 - 140 3.5-6.3 6 Gray Luvisol -13.1-21.3 <300 60 - 140 0.8-3.4 triturated in 10 mM SPB, and resultant suspensions were used for the dilution plating as described above. After incubation, representative bacterial colonies were selected from dilution plates and restreaked onto fresh TSA plates. Surface-sterilized root segments that yielded bacterial growth on imprint plates were used to isolate rhizoplane bacteria: colonies from the root surface were restreaked onto fresh TSA plates for purification and were considered to be rhizoplane bacteria. After verifying the purity of each isolate, bacterial single cell progenies were suspended in tryptic soy broth (TSB) with 20% glycerol, and stored at -80°C. 2.2.1.3 Screening of plant growth promoting capability. A greenhouse bioassay was performed two times to select bacteria which were capable of promoting spruce or pine seedling growth. Of the 216 isolates that were obtained from the rhizosphere of spruce and pine seedlings, only 150 could be recovered after purification and storage at -80°C. In the first screening, these 150 bacterial isolates, including 10 actinomycetes, were tested for PGPR activity using pine and spruce seed with the same geographic origin as the bacteria. In the second screening, 38 spruce and 7 pine isolates which caused a significant increase in plant biomass in the first screening were re-tested. A similar experimental design was employed with the exception that the effect of forest floor soil on growth promotion was also tested. 2.2.1.4 The first screening. Seeds were surface-sterilized as previously described and stratified at 4°C for 28 days. The 3 test was conducted using 164 mL plastic cones (Super Cell 164 cm , 4 cm dia. x 21 cm deep, Ray Leach "Cone-Tainer" Nursery, OR) filled with pasteurized greenhouse soil. Bulk density of the soil was 0.46 g-cm . Chemical analysis performed by Pacific Soil Analysis Inc. (Richmond, B.C.) according to methodology described by Page et al. (1982) indicated: pH=6.9; organic matter=22%; total N=0.58%; available nutrients (pg-g"1): P=279; K=425; Ca=4000; Mg=550; Cu=0.4; Zn=36; Fe=20; Mn=l 16; B=4.7; and S04-S=29. Five seeds were sown per cone. Each 15 bacterial isolate was cultured in an 18 mL culture tube containing 3 mL TSB at 25°C for 4 days on a rotary shaker (120 rpm). Cultures were then diluted five times with SPB, and 1 mL of the resultant suspension was poured directly onto the seeds in cones. The inoculum suspension had at 7 - 1 5 - 1 least 6x10 cfu-mL for bacteria and 3x10 cfu-mL for actinomycetes. Controls received fresh TSB diluted five times with SPB. After inoculation, seeds were immediately covered with 2 mL of forest floor soil (top 20 cm) collected from the same provenance as seed and bacteria. Twenty plastic cones (n=20) were prepared for each treatment and were arranged in a completely randomized design in the greenhouse. Temperature ranged from 14°C to 28°C, and an extended photoperiod of 19 h was achieved through artificial light with a minimum of 130 -2 -1 umol-m -s photosynthetically active radiation (PAR) at bench level. No chemicals or fertilizers were applied throughout the growing period. Seedlings were thinned to the largest single germinant per cone one month after sowing. Seedling emergence was recorded every week after sowing and bacterial inoculation. Three months after sowing, seedlings were harvested and assessed for shoot and root biomass after drying at 70°C for 3 days. In general, the assumptions of ANOVA were not violated. After ANOVA, Fisher's protected Least Significant Difference (LSD) tests were conducted for preplanned comparisons (i.e., each inoculation treatment and uninoculated control) using CoStat Software (Berkeley, CA). Percentage data for seedling emergence were arcsine transformed before statistical analyses. 2.2.1.5 The second screening. The second screening was conducted using similar procedures to the first screening, except for the following modifications: • An additional experimental factor which involved the presence or absence of forest floor soil (2 mL) to the upper surface of cones was imposed; • Each treatment had 18 replicates; • Inoculum cells were washed two times with SPB before inoculation; 16 • Temperature in the greenhouse ranged from 19°C to 34°C. 2.2.1.6 Selection criteria for bacterial isolates used in further experimentation. Isolates were selected after each screening if any of the growth variables (seedling height and shoot and root biomass) showed significant increases over uninoculated control according to Fisher's protected LSD (P<0.1, two-tailed). If two or more isolates had similar effects on seedling growth, were morphologically indistinguishable and were originally isolated from same habitat, they were considered to be the same strain and one was discarded. Most of the selected isolates were more effective seedling growth promoters when seedlings were grown in the presence of forest floor soil. 2 . 2 . 2 Experiment 1: Host ecotype specificity as a component of PGPR efficacy 2.2.2.1 Microorganisms for experiment 1. Bacterial strains Sm3, Ss2, and Sw5 were isolated from the rhizosphere of spruce seedlings collected Mackenzie, Salmon Arm, and Williams Lake sites, respectively. These strains were all identified as Pseudomonas chlororaphis by gas chromatographic analysis of bacterial fatty acids (as methyl-esters) (GC-FAME) using the MIDI (Microbial ID, Inc., Newark, DL) Microbial Identification System at Auburn University described by Kloepper et al. (1992a). These strains were also identified as P. fluorescens using Biolog™ system (Biolog, Inc., Hayward, CA). Strain Pw2, which originated from internal root tissue of a pine seedling from the Williams Lake site, was identified as Bacillus polymyxa by Biolog™, but no matching genera were found by the GC-FAME and MIDI system. Therefore, its species identification should be considered tentative. GC-FAME and Biolog™ similarity indices for these strains are shown in Appendix 2. 2.2.2.2 Seeds and seedling growth for Experiment 1. The effect of host plant ecotype and bacterial strain on seedling growth performance was evaluated in two complete factorial experiments: one with three spruce ecotypes with all possible 17 combinations of the respective rhizosphere isolates, and the other with three pine ecotypes and strain Pw2. Seeds of spruce ecotypes originating from Mackenzie, Salmon Arm, and Williams Lake were inoculated with strain Sm3, Ss2, or Sw5, while pine ecotypes originating from Fort St. John, Kamloops, and Williams Lake were inoculated with strain Pw2. All seeds were obtained from the British Columbia Ministry of Forests Tree Seed Centre, Surrey, B.C. Before use in assays, seeds were surface sterilized (immersion in 2.5% NaCIO for 2 min.) and stratified by rinsing in running tap water overnight followed by storage at 4°C for 30 days. Seedlings were grown in 164 mL plastic cones (Ray Leach "Cone-Tainer" Nursery, OR) filled with steam pasteurized greenhouse soil. Five seeds were sown in a shallow planting hole (ca. 1 cm dia. x ca. 5 mm deep) located in the middle of the soil surface in each container. Inoculum was prepared by growing each Pseudomonas strain in King's B broth and Bacillus strain Pw2 in 50% strength TSB (25°C, 120 rpm) for three days. Cultures were then centrifuged (ca. 3,000 x g for 10 min), rinsed and resuspended in 10 mM sterile phosphate buffer (SPB) (pH 8 8 7 7 -1 7.0) to densities of 2.0 x 10 , 1.7 x 10 , 9.8 x 10 and 7.9 x 10 colony forming units (cfu),mL for strains Sm3, Ss2, Sw5, and Pw2, respectively. For inoculation, 1.0 mL of bacterial suspension was poured directly onto seeds within each container; control seeds received 1.0 mL SPB. Seeds were then covered with ca. 10 mm granite grit (No.l Granite Grit, Imasco, Surrey, B.C.). Containers were watered to saturation and every two days thereafter. Seedlings were grown in the greenhouse under a combination of natural and artificial light -2 -1 which provided a 19 h photoperiod, PAR at the canopy level of 50-160 urnohm *s , and a temperature between 14°C and 32°C. No fertilizer was added to seedling growth medium. Seedling emergence was recorded every week after sowing. A modified germination value (GV) based on Czabator (1962) was used for evaluating the effect of bacterial inoculation on seedling emergence. Czabator's (1962) GV is given by the peak daily germination percentage multiplied by the mean daily germination percentage. This daily based GV was modified by using 18 weekly based seedling emergence percentages and is referred to as seedling emergence value (SEV) in this thesis, where: SEV = (peak weekly seedling emergence %) x (mean weekly seedling emergence %). Seedlings were thinned to the largest single germinant per cone one month after sowing. Spruce seedlings were destructively harvested 14 weeks after sowing and bacterial inoculation. Roots were separated from shoots and seedling biomass was measured after oven-drying (70°C for 3 days). Pine seedlings were treated similarly with the exception that the harvest occurred 10 weeks after sowing. 2.2.2.3 Design and statistical analysis for Experiment J. A complete factorial design for spruce seedling growth generated 12 treatment combinations comprised of four bacterial inoculation treatments (including the uninoculated control) x three spruce ecotypes. Similarly, pine seedling growth was examined using a factorial design with six treatment combinations: two bacterial inoculation treatments (i.e., strain Pw2 and the uninoculated control) x three pine ecotypes. Containers were arranged on a bench in the greenhouse according to a randomized complete block design with 18 blocks (i.e., each block contained all bacteria x conifer ecotype combinations). In general, data did not violate assumptions of ANOVA, therefore, analyses were conducted on untransformed data. A two-way ANOVA for a blocked design was performed on seedling growth data for each conifer genus. Individual treatment means were separated using Fisher's protected LSD tests (P<0.1, two-tailed). The influence of coexistence involving spruce ecotypes and bacteria was tested using preplanned orthogonal contrasts. Statistical analyses were conducted using Systat statistical software (Systat, Inc., Evanston, EL). 19 2.2.3 Experiment 2: Host genus specificity as a component of PGPR efficacy 2.2.3.1 Microorganisms for Experiment 2. Rifamycin-resistant derivatives of strains Sm3, Ss2, Sw5 and Pw2 were generated spontaneously by plating each wild-type strain on solid media containing 100 mg'L 1 rifamycin, and selecting single cell progeny. These mutant strains were similar to wild-type strains in their ability to promote seedling growth, GC-FAME analysis and Biolog™ reactions (Appendix 2), as well as intrinsic antibiotic resistance to vancomycin, streptomycin, kanamycin, tetracycline, penicillin and nalidixic acid (Appendix 3). 2.2.3.2 Seeds and seedling growth for Experiment 2. The effect of conifer genus and bacterial inoculation on seedling growth performance was evaluated in a complete factorial experiment in which spruce and pine seeds were inoculated with strain Sm3R, Ss2R, Sw5R, or Pw2R. Only one ecotype each of pine and spruce (both from Williams Lake) were used, because (i) the pine isolate, Pw2, originated from a pine stand near Williams Lake and was known to be able to stimulate pine seedling growth; (ii) Pseudomonas strains promoted Williams Lake spruce growth; and (iii) these conifer seedlots were found to be free of bacteria inside seeds. The procedures followed for sowing and inoculation in the greenhouse were similar to those described for Experiment 1. However, instead of using steam pasteurized greenhouse soil, seedlings were grown in plastic 164 mL "Ray Leach" cones filled with Sunshine-Mix (Fisons Horticulture Inc., Vancouver, B.C.) soil medium, a mixture of vermiculite, bark, peat moss, quartz, gypsum, perlite, and CaC03 (organic matter=44%; total N=0.72%; pH=5.7; available nutrients (pg-g"1): P=540; K=1260; Ca=9000; Mg=2820; Cu=2; Zn=24; Fe=120; Mn=144; B=4.8; and S04-S=2254). Bulk density of this medium was 0.10 g-cm . Inoculum densities of strains Sm3, Ss2, Sw5, and Pw2 were 3.3 x 108, 2.4 x 108, 5.4 x 107 and 6.0 x 10? cfu-mL"1, respectively. 20 Seedling emergence was recorded every week. Cones were thinned at the beginning of the fifth week, so that the largest seedling in each cone remained. All seedlings were destructively harvested 14 weeks after sowing and bacterial inoculation. Shoot and root dry biomass were measured after oven-drying (70°C for 3 days). 2.2.3.3 Bacterial root colonization assessment for Experiment 2. Before seedling tissues were dried, the distal section (5-6 cm) of the longest root from four randomly selected seedlings in each treatment was assessed for bacterial root colonization. This involved shaking root segments gently to remove loosely adhering soil, after which they were placed in a culture tube containing 10 mL of 10 mM SPB with glass beads and agitated for two minutes. Root washings were then serially diluted and plated in triplicate onto KBA for Pseudomonas strains and 50% strength TSA for strain Pw2R. All agar media were amended with rifamycin (100 mg*L *), cycloheximide (100 mg*L *), and benomyl (30 mg'L *). Root and rhizosphere soil samples were then oven-dried at 70°C to a constant weight. Dilution plates were incubated for up to four days at 30°C before colony number was assessed. Intrinsic resistance to streptomycin, kanamycin, tetracycline and vancomycin (BBL Sensi-Discs 10-30 pg) of representative colonies from dilution plates was compared with that of inoculant strains to reduce the chance of monitoring contaminant strains. 2.2.3.4 Design and statistical analysis for Experiment 2. A complete factorial design for seedling growth generated 10 combinations of five bacterial inoculation treatments (including the uninoculated control) x two conifer genera. Containers were arranged on a bench according to a randomized complete block design with 18 blocks. Because of growth differences between spruce and pine, the distributions of pooled seedling height and biomass data had two distinct peaks, i.e., differed significantly from a normal distribution according to Lilloefors test (Wilkinson et al. 1992). Natural log (In) transformation improved the data distribution, therefore, analyses were conducted on In transformed data. Data for seedling emergence values were not transformed before statistical analysis. 21 For root colonization assessment, the number of colonies that grew on antibiotic-amended agar media was related to the dry weight of roots plus adhering rhizosphere soil. These values were log transformed before data were subjected to ANOVA. The minimum detectable 4 -1 population size was 10 cfxrg soil. Because only a small portion of each root system was sampled for bacterial colonization, samples that yielded no growth on dilution plates were assumed to be uncolonized and were assigned a value of zero for the analysis (Kloepper and Beauchamp 1992). Individual treatment means were separated using Fisher's protected LSD (P<0.1, two-tailed). The possible influence of adaptive relationships involving host genus and rhizosphere bacteria was tested using preplanned orthogonal contrasts. 2.2.4 Experiment 3: PGPR-soil specificity 2.2.4.1 Microorganisms for Experiment 3. Nalidixic acid resistant derivatives of Pseudomonas strains Sm3R, Ss2R, and Sw5R were generated to improve their recovery from environmental samples. This was performed following the same procedure described for generating rifamycin resistant strains (Experiment 2), with the exception that solid media also contained 100 mg-L 1 nalidixic acid. The resulting rifamycin and nalidixic acid resistant strains were named Sm3RN, Ss2RN, and Sw5RN, respectively. Bacillus strain Pw2R was not amenable to generating spontaneous nalidixic acid resistance, therefore, strains Sm3RN, Ss2RN, Sw5RN and Pw2R were used for Experiment 3. The new Pseudomonas mutants were similar to their parental strains in GC-FAME analysis and Biolog™ tests (Appendix 2) as well as intrinsic antibiotic resistance to vancomycin, streptomycin, kanamycin, tetracycline, and penicillin (Appendix 3). 2.2.4.2 Seeds and seedling growth for Experiment 3. The effect of soil and PGPR origin on seedling growth was evaluated in a complete factorial experiment. Spruce seeds (Williams Lake ecotype) were inoculated with strain Sm3RN, Ss2RN, 22 Sw5RN, or Pw2R in the presence of forest floor soil (top 20 cm) collected from the sampling site where each strain originated. Forest soil for Experiment 3 was collected (as described in Section 2.2.1.1) from each forest stand, i.e., Mackenzie spruce, Salmon Arm spruce, Williams Lake spruce and Williams Lake pine. Soil collected from each stand was thoroughly mixed, sieved (2 mm particle size), and stored at 4°C for a maximum of 14 days before use. Chemical properties of these soil samples are presented in Table 2.2. Soil chemical analyses were conducted at Pacific Soil Analysis Inc. according to methodology described Page et al. (1982). Seedling growth conditions including sowing and inoculation in the greenhouse were similar to those described for Experiment 1. However, instead of granite grit, seeds were covered with 2 mL of one of the forest floor soils after bacterial inoculation. Inoculum densities of strains Sm3RN, Ss2RN, Sw5RN, and Pw2R were 2.6 x 108, 3.7 x 108, 1.4 x 108, and 9.1 x 107 cfu-mL \ respectively. All seedlings were destructively harvested 15 weeks after sowing and bacterial inoculation. Shoot and root dry biomass were measured after oven-drying (70°C for 3 days). 2.2.4.3 Bacterial root colonization assessment for Experiment 3. The distal section (5-6 cm) of the longest root from four randomly selected seedlings in each treatment was assessed for bacterial root colonization when seedlings were harvested. The procedure was same as that described in Experiment 2, except that KBA was also amended with nalidixic acid (100 ug*L *) to facilitate recovery of Pseudomonas strains. 2.2.4.4 Design and statistical analysis for Experiment 3. For seedling growth, a complete factorial design generated 20 treatment combinations comprised of five bacterial inoculation treatments (including the uninoculated control) x four forest soils (i.e., origins). Containers were arranged on a bench in a randomized complete block design with 15 blocks. 23 Table 2 . 2 Chemical properties of forest floor soils collected from Mackenzie spruce, Salmon Arm spruce, and Williams Lake spruce and pine stands. Stand type Property Mackenzie Spruce Salmon Arm Spruce Williams Lake Spruce Williams Lake Pine PH' 4.8 4.8 5.6 5.6 Organic matter (%) 14.1 16.9 11.7 23.6 Total nitrogen (%) 0.33 0.32 0.20 0.45 Phosphorus (pg-g *) 40 60 72 42 Potassium (pg-g *) 135 180 180 210 Calcium (pg-g"1) 1650 1350 1600 3000 Magnesium (pg-g *) 170 150 160 445 Copper (pg-g"1) 0.6 1.0 1.0 0.6 Zinc (pg-g"1) 12.0 10.5 17.0 12.0 Iron (pg-g"1) 75 105 55 20 Manganese (pg-g *) 160 130 310 230 Boron (pg-g"1) 0.2 0.6 0.6 1.1 Sulfate-sulfur (pg-g *) 3.3 2.9 2.2 4.4 Oven dry soil: deionized H2O =1:2. 24 In general, seedling height and weight data did not violate assumptions of ANOVA, therefore, analyses were conducted on untransformed data. For root colonization assessment, the number of colonies per gram dry weight of rhizosphere soil was log transformed before 4 -1 calculating population parameters. The minimum detectable population size was 10 cfii'g soil. If no growth occurred on dilution plates, a value of zero was assigned for the analysis (Kloepper and Beauchamp, 1992). After two-way ANOVA for a blocked design, individual treatment means were separated using Fisher's protected LSD (P<0.1, two-tailed). The possible influence of adaptive relationships involving host genus and rhizosphere bacteria was also tested using preplanned orthogonal contrasts. 2.3 RESULTS1 2.3.1 Screening of PGPR for interior spruce and lodgepole pine seedlings. Seedling performance in screening trials after inoculation with the four bacterial isolates selected for further study as PGPR is presented in Tables 2.3 and 2.4. Strains Sm3 and Sw5 were isolated from the rhizosphere of spruce seedlings collected from Mackenzie and Williams Lake sites, respectively. Strain Ss2 was an isolate from the rhizoplane of a spruce seedling collected from the Salmon Arm site, while strain Pw2 originated from internal root tissue of a pine seedling from the Williams Lake site. 2.3.2 Experiment 1: Host ecotype specificity as a component of PGPR efficacy In general, PGPR effects were insignificant when evaluated on an individual conifer ecotype basis due to a large degree of "within treatment" variability and a comparatively small sample size Statistical tables including ANOVA on which Chapter 2 results are based are presented in Appendix 4. 25 Table 2.3 Host seedling performance three months after inoculation with selected bacterial isolates in the first screening. Variable Sm3 (Mackenzie, Sx isolate) Ss2 (Salmon Arm, Sx isolate) Sw5 (Williams Lake, Sx isolate) Pw2 (Williams Lake, PLi 1 isolate) Seedling 64 ± 62 99 ± 1 84 ± 5 89 ± 3 3 emergence (%) (7%)4 (1%) (9%) (4%) Shoot 2.7 ± 0.1 *** 3.3 ± 0.1 4.0 ± 0.1 *** 5.1 ± 0.2 height (mm) (27%) (9%) (22%) (4%) Shoot 22.1 ± 2.3 ** 28.6 ± 2.1 ** 40.1 ± 5.3 *** 57.4 ± 5.7 ** biomass (mg) (65%) (67%) (82%) (39%) Root 14.6 ± 1.6 *** 24.9 ± 2.1 ** 22.4 + 1.9 ** 53.9 ± 5.1 ** biomass (mg) (90%) (61%) (42%) (38%) Sx: interior spruce; PLi: lodgepole pine. Mean + standard error, (n = 20); *** and ** indicate that the treatment was different from the control by Fisher's protected LSD withPO.Ol and P<0.05, respectively. Percentage of seedlings that emerged by the end of the experiment. Percent change from the mean of the uninoculated control. Table 2.4 Host seedling performance two months after inoculation with the selected bacterial isolates in the second screening. (a) In the presence of forest floor soil Variable Sm3 (Mackenzie, Sx1 isolate) Ss2 (Salmon Arm, Sx isolate) Sw5 (Williams Lake, Sx isolate) Pw2 (Williams Lake, PLi 1 isolate) Seedling 82 ± 42 74 ± 6 74 ± 5 61 ± 5 emergence (%) (9%)4 (-11%) (13%) (5%) Shoot 3.1 ± 0.1 * 2.5 ± 0.1 3.3 ± 0.1 *** 6.5 ± 0.2 *** height (cm) (7%) (5%) (15%) (18%) Shoot 21.6 ± 1.7 20.7 ± 1.4 26.3 ± 11.0 *** 64.1 ± 4.8 * biomass (mg) (15%) (4%) (32%) (24%) Root 8.6 ± 0.9 10.7 ± 0.9 * 11.0 ± 1.0 *** 48.9 ± 3.5 ** biomass (mg) (8%) (23%) (33%) (27%) Sx: interior spruce; PLi: lodgepole pine. Mean + standard error, (n = 18); ***, **, and * indicate that the treatment was different from the control by Fisher's protected LSD withPO.01, P<0.05, and P<0.1, respectively. Percentage of seedlings that emerged by the end of the experiment. Percent change from the mean of the uninoculated control. Table 2.4 Continued. (b) Without forest soil Variable Sm3 (Mackenzie, Sx1 isolate) Ss2 (Salmon Arm, Sx isolate) Sw5 (Williams Lake, Sx isolate) Pw2 (Williams Lake, PLi 1 isolate) Seedling 85 ± 42 89 ± 5 54 ± 7 74 ± 6 emergence (%) (9%)4 (-6%) (-23%) (8%) Shoot 3.1 ± 0.1 3.0 ± 0.1 ** 3.2 ± 0.1 6.6 ± 0.2 height (mm) (1%) (10%) (5%) (4%) Shoot 22.9 ± 2.0 24.5 ± 1.3 24.4 ± 2.5 69.2 ± 4.4 biomass (mg) (9%) (13%) (18%) (5%) Root 12.4 ± 0.8 12.3 ± 0.6 9.9 ± 1.1 62.7 ± 3.7 biomass (mg) (8%) (12%) (8%) (10%) Sx: interior spruce; PLi: lodgepole pine. Mean ± standard error, (n = 18); ** indicates that the treatment was different from the control by Fisher's protected LSD with P<0.05. Percentage of seedlings that emerged by the end of the experiment. Percent change from the mean of uninoculated control. (n=18). However, when spruce data were pooled and evaluated in a single analysis, statistically significant (i^ O.OS) gains in spruce seedling height and root biomass of 5% and 16%, respectively, were detected (Appendix 4, Table A4.19). 2.3.2.1 Seedling emergence. Spruce ecotypes showed similar seedling emergence responses to all three Pseudomonas strains Sm3, Ss2, or Sw5 (Fig. 2.1 a). All bacterial inoculations increased mean seedling emergence of the Mackenzie spruce ecotype, decreased the Williams Lake ecotype seedling emergence and did not affect emergence of the Salmon Arm ecotype compared to uninoculated controls. However, none of these differences were statistically significant (Tables A4.13 a and A4.17 a). In addition, orthogonal contrasts involving coexistent and unrelated host ecotype and bacterial inoculum combinations revealed no effect of coexistence on spruce seedling emergence (Fig. 2.1b). Pine seedling emergence did not differ statistically from uninoculated controls after inoculation with Pw2 (Fig. 2.1 a). However, the coexistent ecotype (i.e., Williams Lake) had a 14% increase in seedling emergence value due to treatment with Pw2, but the contrast testing this matched combination was not statistically significant (Table A4.22). 2.3.2.2 Seedling growth. In general, spruce seedling growth (shoot height, shoot biomass and root biomass) was more responsive to bacterial inoculation than pine (Fig. 2.2 a, 2.3 a, and 2.4 a). However, the only statistically significant (P<0.1) effect was the root biomass increase of Williams Lake spruce inoculated with strain Sm3 (Fig. 2.4 a, Table A4.17). Other comparatively large increases in biomass (i.e., > 25%) were not significant due to seedling growth response variability within treatments. Because no interaction was found between PGPR inoculation and host ecotypes for spruce seedling growth (Table A4.20), data from all spruce ecotypes were pooled for performing contrasts of coexistent and unrelated host ecotype-bacterial inoculum combinations. None of the 29 (a) Effect of each bacterial strain 30 20 10 -10 Sx PLi V Sm3 Ss2 Sw5 Inoculum strain Pw2 Sx seed provenance I I Mackenzie • 1 Salmon Arm Uliii Williams Lake PLi seed provenance I I Fort St. John Kamloops Williams Lake (b) Contrasts of coexistence effects Combination Coexistent Unrelated I I Uninoculated Mackenzie Salmon Arm Williams Lake Pooled Sx seed provenance . 2.1 Effect of ecotype specificity between conifers and PGPR strains on seedling emergence value, (a): Percent change from the uninoculated control for each bacterial treatment. -i- indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 30 (a) Effect of each bacterial strain p 10 Sx PLi V Sx seed provenance I I Mackenzie • U Salmon Arm W&M Williams Lake PLi seed provenance I I FortStJohn Kamloops Williams Lake Sm3 Ss2 Sw5 Inoculum strain Pw2 (b) Contrasts of coexistence effects Combination Coexistent Unrelated I I Uninoculated Mackenzie Salmon Arm Williams Lake Pooled Sx seed provenance . 2.2 Effect of ecotype specificity between conifers and PGPR strains on shoot height 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment. <l indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 31 (a) Effect of each bacterial inoculation 40 30 20 10 -10 Sx PLi V 1 Sx seed provenance I 1 Mackenzie • • Salmon Arm W^i Williams Lake PLi seed provenance I I Fort St. John Kamloops Williams Lake Sm3 Ss2 Sw5 Inoculum strain Pw2 (b) Contrasts of coexistence effect Combination fflHi Coexistent K H Unrelated -I Uninoculated Mackenzie Salmon Arm Williams Lake Pooled Sx seed provenance . 2.3 Effect of ecotype specificity between conifers and PGPR strains on shoot biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment. I indicates a coexistent host ecotype-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.\). Error bars indicate the standard errors of the mean. 32 (a) Effect of each bacterial strain Sm3 Ss2 Sw5 Pw2 Inoculum strain Sx seed provenance I 1 Mackenzie • U Salmon Arm HHH Williams Lake PLi seed provenance I 1 Fort StJohn I Kamloops 1 Williams Lake (b) Contrasts of coexistence effects Mackenzie Salmon Arm Williams Lake Pooled Sx seed provenance Combination HHH Coexistent HH^i Unrelated I I Uninoculated . 2.4 Effect of ecotype specificity between conifers and PGPR strains on root biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment, i indicates a coexistent host ecotype-bacterial inoculum combination; * indicates the mean value signifcantly differs from the uninoculated control (P<0.1); Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving spruce ecotypes and PGPR strains previously isolated from spruce rhizospheres. Means designated by different letters within groups of seed provenances are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 33 variables showed statistical differences between coexistent and unrelated combinations, but the inoculated treatments were significantly different from the uninoculated controls due to the increased sample size (Fig. 2.2 b, 2.3 b, and 2.4 b). However, it should be noted that coexistent spruce ecotype-bacterial inoculum combinations resulted in the highest increases in shoot and root biomass. Pine seedling growth was not significantly enhanced by Pw2 inoculation in this experiment (Appendix 4, Table A4.21). 2.3.3 Experiment 2: Host genus specificity as a component of PGPR efficacy. 2.3.3.1 Seedling emergence. Spruce and pine seedling emergence values after bacterial inoculation did not differ from uninoculated controls (Fig. 2.5 a). In addition, orthogonal contrasts indicated no increase in SEV for coexistent conifer genus-bacterial inoculum combinations (Fig. 2.5 b). 2.3.3.2 Seedling growth. Each of the four bacterial strains increased growth of spruce seedlings 14 weeks after sowing and inoculation (Fig. 2.6 a, 2.7 a, and 2.8 a), but only height increases were significant (P<0.05) (Table A4.23). In contrast, height and shoot biomass of pine seedlings significantly decreased after inoculation with Pseudomonas strain Ss2R (P<0.1) (Fig. 2.6 a, 2.7 a, and 2.8 a; Tables A4.23 d and A4.23 f). Other inoculation effects were small and insignificant. No evidence of conifer genus (i.e., interior spruce or lodgepole pine) x inoculum strain specificity was detected (Fig. 2.6 b, 2.7 b, and 2.8 b). 2.3.3.3 Inoculum colonization in the rhizosphere. Recovery of Bacillus strain Pw2R from the spruce rhizosphere was not successful, but all Pseudomonas strains were detected in both spruce and pine rhizospheres 14 weeks after inoculation (Fig. 2.9 a; Tables A4.23 h and A4.23 i). However, recovered populations were 34 (a) Effect of each bacterial strain 25 20 15 10 5 0 -5 -10 -15 -20 V Sm3R Ss2R Sw5R Inoculum strain Pw2R (b) Contrasts of coexistence effects Pooled Combination Coexistent Unrelated I I Uninoculated Host genus 2.5 Effect of host genus specificity between conifers and PGPR strains on seedling emergence value, (a): Percent change from the uninoculated control for each bacterial treatment; <l indicates a coexistent host genus-bacterial inoculum combination; Sx; interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean. 35 (a) Effect of each bacterial strain 15 h 10 5 0 -5 -10 1 ' \ ** V === y === II === 1 Host genus PLi Sm3R Ss2R Sw5R Pw2R Inoculum strain (b) Contrasts of coexistence effects Combination Coexistent Unrelated I I Uninoculated . 2.6 Effect of host genus specificity between conifers and PGPR strains on shoot height 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment; i indicates a coexistent host genus-bacterial inoculum combination; ** and * indicate the mean values differ significantly from uninoculated controls atP<0.05 and P<0.1, respectively; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean. 36 (a) Effect of each bacterial strain 20 g 10 S .8 60 o 0 -10 -20 -30 1 Sm3R Ss2R Sw5R Inoculum strain Pw2R Host genus PLi (b) Contrasts of coexistence effects Combination H H H Coexistent ^ H H Unrelated I I Uninoculated . 2.7 Effect of host genus specificity between conifers and PGPR strains on shoot biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment; -l indicates a coexistent host genus-bacterial inoculum combination; * indicates the mean value differs significantly from uninoculated controls at P<0.1; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean. 37 (a) Effect of each bacterial strain Sm3R Ss2R Sw5R Pw2R Inoculum strain (b) Contrasts of coexistence effects Combination Coexistent Unrelated I I Uninoculated . 2.8 Effect of host genus specificity between conifers and PGPR strains on root biomass 14 weeks after sowing and inoculation, (a): Percent change from the uninoculated control for each bacterial treatment: -i- indicates a coexistent host genus-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. Means designated by different letters within groups of conifer genus are significantly different (P<0.1). Error bars indicate standard errors of the mean. 38 \ (a) Size of recovered population of each strain Sm3R Ss2R Sw5R Pw2R Inoculum strain (b) Contrasts of coexistence effects Combination M M Coexistent I I Unrelated Fig. 2.9 Effect of host genus specificity between conifers and PGPR strains on rhizosphere colonization 14 weeks after sowing and inoculation, (a): Size of recovered population of each inoculum: bars indicate means and standard errors; •l indicates a coexistent host genus-bacterial inoculum combination; Sx: interior spruce and PLi: lodgepole pine, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations involving conifer genera and PGPR strains. 39 substantially smaller (< 1%) than the initial inoculum cell numbers applied. Rhizosphere colonization was slightly higher in coexistent combinations of host genus and bacterial inocula (Fig. 2.9 b). Because only strain Pw2R was related to pine and its recovery was not successful, the orthogonal contrasts reflect primarily rhizosphere colonization by Pseudomonas strains. 2.3.4 Experiment 3: Soil specificity as a component of PGPR efficacy. 2.3.4.1 Seedling emergence. None of the bacterial treatments consistently affected seedling emergence (Fig. 2.10 a), consequently, no coexistence effects were found involving bacterial inoculum and forest floor soils (Fig. 2.10 b). 2.3.4.2 Seedling growth. Seedling growth promotion was detected primarily in the presence of soils collected near Williams Lake, from both spruce and pine stands (Fig. 2.11 a, 2.12a, and 2.13 a). However, only shoot height effects were statistically significant (Tables A4.31 b and A4.33 b). Mackenzie site spruce stand soil decreased shoot growth with all bacterial strains, though growth reductions were not significant (Fig. 2.11 and 2.12; Table A4.27). On the other hand, Salmon Arm soil caused better shoot growth, including for the uninoculated control, than the other two forest soil treatments, but the effects of inoculation with this soil on shoot growth were marginal (Fig. 2.11 a and 2.12 a). Orthogonal contrasts showed no significant effects of coexistence (i.e., a common origin) involving inoculum strains and forest soils (Fig. 2.11 b, 2.12 b, and 2.13 b). However, coexistent soil-bacteria combinations resulted in slightly better seedling growth performance than unrelated combinations. Moreover, seedling growth in the presence of coexistent soil-bacteria combinations was significantly greater than uninoculated controls when the data were pooled among the four forest soil types (Fig. 2.11 b, 2.12 b, and 2.13 b). 40 (a) Effect of each bacterial strain 10 1 8 o e 8> -10 o -20 -30 1 Sm3RN Ss2RN Sw5RN Inoculum strain Pw2R Soil origin/stand type MZ/Sx n m SA/SX WL/Sx Hi WL/PLi (b) Contrasts of coexistence effects 250 -u 3 > 200 § & 150 100 <*> 50 a a _ a _ a a a a a a a a a 1 W 1 MZ/Sx SA/Sx WL/Sx WL/PLi Pooled Soil origin Combination Coexistent i i i l i Unrelated I I Uninoculated .2.10 Effect of forest soil and PGPR inoculation on seedling emergence value, (a): Percent change from the uninoculated control of each bacterial treatment. >l indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 41 (a) Effect of each bacterial strain Sm3RN Ss2RN Sw5RN Inoculum strain Pw2R Soil origin/stand type MZ/Sx ] SA/Sx 1 WL/Sx 1 WL/PLi (b) Contrasts of coexistence effects MZ/Sx SA/Sx WL/Sx WL/PLi Pooled Soil origin .2.11 Effect of soil forest soil and PGPR inoculation on shoot height, (a): Percent change from the uninoculated control of each bacterial treatment. 4 indicates a coexistent combination, i.e., common origin, between soil and inoculum; ** and * indicate the mean values differ significantly (PO.05 and P<0.1, respectively) from uninoculated controls; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 42 (a) Effect of each bacterial strain Sm3RN Ss2RN Sw5RN Pw2R Inoculum strain (b) Contrasts of coexistence effects MZ/Sx SA/Sx WL/Sx WL/PLi Pooled Soil origin . 2.12 Effect of forest soil and PGPR inoculation on shoot biomass. (a): Percent change from the uninoculated control of each bacterial treatment. -I indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 43 (a) Effect of each bacterial strain (b) Contrasts of coexistence effects MZ/Sx SA/Sx WL/Sx WL/PLi Pooled Soil origin Combination Hil l Coexistent BUI Unrelated -I Uninoculated . 2.13 Effect of forest soil and PGPR inoculation on root biomass. (a): Percent change from the uninoculated control of each bacterial treatment. -I indicates a coexistent combination, i.e., common origin, between soil and inoculum; MZ: Mackenzie site, SA: Salmon Arm site, WL: Williams Lake site, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. Means designated by different letters within groups of soil are significantly different (P<0.1). Error bars indicate the standard errors of the mean. 44 2.3.4.3 Inoculum colonization in the rhizosphere. Recovery of the three Pseudomonas strains at the end of the experiment was successful, but Bacillus strain Pw2R was not detected (Fig. 2.14 a). Pseudomonas colonization of the rhizosphere was not affected by soil origin. However, orthogonal contrasts comparing coexistent and unrelated inoculum-soil combinations indicated a strong effect of spruce stand soil from the Salmon Arm site (Fig. 2.14 b) on the colonization of Ss2RN, but not the other two Pseudomonas strains (Fig. 2.14 a). 45 (a) Size of recovered population of each strain Sm3RN Ss2RN Sw5RN Pw2R Inoculum strain (b) Contrasts of coexistence effects Combination IBH Coexistent I 1 Unrelated MZ/Sx SA/Sx WL/Sx WL/PLi Pooled Soil origin Fig. 2.14 Effect of forest soils and PGPR inoculation on rhizosphere colonization 15 weeks after sowing and inoculation, (a): Size of recovered population of each inoculum. Bars indicnate means and standard errors. -I: coexistent combination between a forest soil and inoculum; MZ: Mackenzie site soil, SA: Salmon Arm site soil, WL: Williams Lake site soil, Sx: interior spruce stand and PLi: lodgepole pine stand, (b): Orthogonal contrasts of coexistent, unrelated and uninoculated combinations between soil origins and PGPR strains. 46 2.4 DISCUSSION 2.4.1 Host ecotype specificity effects on plant growth promotion by bacteria. The overall objective of these experiments was to evaluate the degree of specificity, if any, between spruce and pine ecotypes and coexistent or unrelated PGPR in relation to plant growth promotion. Because pine growth was not enhanced by PGPR inoculation, it was of no use in evaluating this objective. Significant effects of Pseudomonas strains Sm3RN, Ss2RN and Sw5RN on spruce growth facilitated hypothesis testing related to questions of coexistence specificity. When growth of spruce seedlings inoculated with coexistent PGPR was compared with growth of seedlings inoculated with unrelated PGPR and uninoculated controls, coexistent plant-bacteria combinations were found to result in the largest seedlings. However, spruce inoculated with unrelated PGPR were also significantly larger than uninoculated controls, and the difference in seedling growth between spruce treated with coexistent PGPR and unrelated PGPR was small and not significant. The slight advantage detected for coexistent host-PGPR combinations may be related to adaptation or co-adaptation involving these organisms. Plant cultivar or genotype specificity between host plants and rhizobacterial inocula has been observed in studies of agriculturally-important species (Chanway et al. 1991b; Sumner 1990), and in some cases, may comprise a significant component of variation in plant growth responses to PGPR (Burr and Caesar 1984). For example, Chanway et al. (1990) found clear effects on plant growth that were attributable to "adapted" Bacillus-Lolium genotype associations in an unmanaged permanent pasture. However, many other studies of agricultural crops and PGPR report no such specificity (e.g., Bashan et al. 1989). Nonetheless, few such studies of PGPR specificity have been performed with tree species. O'Neill et al. (1992) reported a degree of growth response specificity involving spruce seedlings and soil that had a common geographic origin. However, no evidence of coexistence specificity was observed when seedling biomass was measured after 3 months: stimulation of Mackenzie spruce growth resulted when seed was inoculated with Salmon Arm bacteria and vice versa 47 (O'Neill et al. 1992). In contrast, Chanway and Holl (1993a) reported coexistence specificity of spruce seedling emergence, but not growth, in response to inoculation with indigenous strains of Pseudomonas putida. Ecotype-specific growth responses of Douglas-fir seedlings were also detected 13 months after inoculation with Arthrobacter oxydans or Pseudomonas aureofaciens (Chanway and Holl 1994a). Detecting coexistence specificity with conifers may be difficult because of the highly heterozygous progeny that results from outcrossing. Indeed, by definition interior spruce is a hybrid and will necessarily produce seed with a great degree of genetic variability. Use of such conifer seed instead of clonal material or inbred seed lines commonly used in agricultural crops may have obscured detection of fine-scale ecological relationships that have developed between conifer hosts and rhizosphere bacteria. From a practical perspective, the degree of coexistence specificity seems to be slight, and rarely expressed in the field so that it is unlikely to affect use of PGPR such as the ones described in this thesis as inocula for plantation enhancement. 2.4.2 Host genus specificity effects on plant growth promotion by bacteria. Theoretically, specificity between plants and PGPR could occur at any of several hierarchical taxonomic levels, i.e., genotypes, ecotypes, species, genera, etc. Therefore, in Experiment 2, specificity at a broader level was investigated by evaluating root colonization and seedling growth responses of two different host genera, i.e., spruce (Picea) and pine (Pinus), to inoculation with coexistent or unrelated PGPR. Spruce height and shoot biomass growth promotion was detected, but comparison of coexistent and unrelated conifer host-bacterial inoculum treatment means with uninoculated controls revealed no effect of coexistence on seedling growth. Therefore, it appears that if seedling growth response specificity is at all important in spruce, it could be at broader taxonomic levels than genus. A corollary is that PGPR originally isolated from one genus, such as strain Pw2R from pine, may be of use on other host genus, e.g., spruce. 48 Root-associated microorganisms largely depend on root exudates, secretions and rhizodeposits for nutrients (Richards 1987), and the role of these substrates in establishment and maintenance of rhizosphere populations has long been known (Rovira 1956). The composition and amount of exudates and rhizodeposits are influenced by environmental conditions, but their production is largely under genetic control. Therefore, genetic differences between host plants in the production of rhizodeposits may account for observed differences in rhizosphere microflora among plant taxa, and result in expression of plant-microbe specificity as differential colonization of the rhizosphere (Bachmann and Kinzel 1992; Martin 1971; Neal et al. 1970, 1973). However, evaluation of root colonization in this experiment provided no evidence of such plant-microbe specificity. Pseudomonas invariably colonized the rhizosphere of both spruce and pine, although strain Sw5R was recovered from only one of four samples of each conifer genus. On the other hand, Bacillus occurred at populations that were generally below the detection limits associated 4 -1 with the recovery assay (10 cfu-g soil) as only one pine seedling yielded Pw2R from the rhizosphere. Data from Experiment 2 provide evidence that rhizosphere population size and effects on plant growth are not strictly correlated, though some degree of root colonization is obviously required for plant growth stimulation. For example, strain Sw5RN colonized spruce root systems with relatively low populations, but caused large positive seedling growth responses. In addition, strain Ss2RN was recovered from pine and spruce root samples in equivalent numbers, yet this strain significantly inhibited growth of pine. The comparable Ss2RN population sizes for both pine and spruce likely reflect similarities in rhizosphere conditions between these two genera: However, the contrasting seedling growth responses may relate to innate seedling growth rates so that slower growing genera such as spruce may be more receptive to PGPR than faster growing genera such as pine. 49 2.4.3 Soil origin effects on plant growth promotion by bacteria. The objective of Experiment 3 was to investigate possible effects on PGPR efficacy of soil collected from different regions and forest types. Although no consistent trends on seedling emergence were detected, plant growth promotion by PGPR was altered significantly by the addition of small amounts (<2% v/v) of forest soils. In particular, PGPR were ineffective when seedling growth medium was amended with Mackenzie spruce forest soil, whereas gains of >20% in spruce height and shoot weight were detected in the presence of Williams Lake spruce stand soil. Because of physical and chemical similarities between these soils (Table 2.2) and the small amount added to each container, it is likely that these soils served as an inoculum source of forest soil organisms, and hence, that any effects on seedling growth were probably biological. One can only speculate as to the exact biological processes at work, but they do not appear to involve root colonization by PGPR as rhizosphere populations were similar between forest soil treatments. Because PGPR efficacy may not depend on, but is often enhanced in the presence of certain forest soils, the forest soil biological factor seems to supplement PGPR efficacy rather than cause it. No evidence of "coexistence" specificity involving soil collected from the same sites as seedlots or PGPR was found in this experiment. O'Neill et al. (1992) conducted a similar test of forest soil effects on PGPR efficacy using a greater proportion of forest soil in seedling growing medium (50% v/v), and found that interior spruce seedling growth was greatest when grown in forest soil and inoculated with rhizosphere bacteria originating from the same location as spruce seed. However, the primary factors contributing to this 3-way interaction were seed and soil source. The effect of bacteria was rendered insignificant in unpasteurized soil, indicating that abiotic elements in forest soils were more important in affecting PGPR efficacy than were biotic elements. Organic substrates such as tryptophan, which can be converted to indole-3-acetic acid by soil microorganisms (Martens and Frankenberger 1993), may be responsible for such effects. Because IAA is bioactive in very small concentrations, such substrates could also have been important in Experiment 3. 50 In conclusion, inoculation of spruce with each of the four PGPR strains resulted in statistically significant seedling growth promotion, while pine was less responsive. Depending on its origin, addition of 2% forest soil to seedling growth medium enhanced PGPR efficacy, probably due to biological effects. Large variations in seedling growth responses occurred between screening trials 1 and 2, and could have resulted from the use of culture supernatants in the first screening but not the second. Subsequent experimentation provided little evidence that seedling growth response variability was related to adaptive relationships involving conifer genera, PGPR and forest soils or the PGPR population size in the rhizosphere. Though specificity hypotheses were tested in single experiments that were not repeated, the recurring observation that specificity was not an important component of plant growth response variability was consistent with the findings of others (Chanway and Holl 1992; O'Neill et al. 1992), and plant x PGPR specificity was not considered in the design of subsequent experiments described in this thesis. 51 CHAPTER 3 SPRUCE AND PINE SEEDLING GROWTH AND MYCORRHIZAL INFECTION AFTER INOCULATION WITH PGPR 3.1 INTRODUCTION AND LITERATURE REVIEW One of the most important plant-microbe relationships in forest lands results in the formation of mycorrhizae, the symbiotic fungus-root associations well-known to enhance plant productivity (Harley and Smith 1983; Vogt et al 1991). The rhizosphere of mycorrhizae (i.e., the mycorrhizosphere) is known to support large microbial populations, of which bacteria are likely the most numerous (Rouatt and Katznelson 1961; Linderman 1988; Neal and Bollen 1964). Some of these bacteria can affect the growth and infectivity of mycorrhizal fungi (Bowen and Theodorou 1979; Garbaye and Bowen 1987, 1989; Garbaye 1994). For example, Bowen and Theodorou (1979) showed that bacterial strains isolated from the rhizosphere as well as bulk soil significantly stimulated or inhibited mycelial growth of Rhizopogon luteolus Fr. and Nord. along the root surface of radiata pine (Pinus radiata D. Don) seedlings. Similar bacteria-mycorrhizal fungi interactions were demonstrated by Garbaye and Bowen (1987) when radiata pine seedlings were inoculated with three ectomycorrhizal fungi (Paxillus involutes Batsh. ex Fr., Rhizopogon luteolus, and Hebeloma crustuliniforme Quelet in different sterilized soils and amended with soil microflora. Garbaye and Bowen (1989) subsequently found that a higher proportion of bacteria isolated from mycorrhizae stimulated mycorrhizal development compared to those isolated from bulk soil, and suggested that a symbiotic relationship between mycorrhizal fungi and mycorrhiza-associated bacteria may exist. They coined the term mycorrhization helper bacteria (MHB) to describe such microorganisms. Their results demonstrated that MHB may act as PGPR that promote plant growth indirectly by improving the establishment of mycorrhizae (Fitter and Garbaye 1994; Garbaye 1994). Interestingly, many MHB and PGPR belong to the same taxonomic groups: fluorescent pseudomonads and sporulating bacilli. 52 Some studies of MHB (Duponnois and Garbaye 1991; Duponnois etal, 1993) indicate that certain strains of Pseudomonas and Bacillus isolated from the mantle of Douglas-fir-Zacrar/a laccata Scop, ex Fr. ectomycorrhiza can be strikingly fungus-specific in their ability to enhance mycorrhizal root tip formation. However, none of these mycorrhiza infection-stimulating bacterial strains were capable of promoting Douglas-fir seedling growth directly, i.e., in the absence of the appropriate mycorrhizal fungus. Interestingly, recent work (Frey-Klett et al. 1997) suggests that a significant mycorrhiza helper effect occurs even with MHB populations as low as 30 cfu-g 1 of soil. To date, however, little is known about the mechanism(s) by which MHB enhance mycorrhizal infection. In contrast, several mechanisms by which PGPR stimulate plant growth have been proposed, and are broadly categorized as either direct or indirect (Kloepper 1993). When plant growth promotion by bacteria is direct, PGPR produce a metabolite or compound that is stimulatory to plants (e.g., a phytohormone such as indoleacetic acid), whereas bacteria that elicit plant growth effects indirectly do so by affecting other factors in the rhizosphere which in turn results in plant growth stimulation (e.g., PGPR suppression of indigenous, plant yield-reducing rhizosphere microorganisms). Discussion of PGPR mechanisms will be further developed in Chapter 6. Though MHB may not be considered to be symbiotic to the same degree as mycorrhizal fungi, their ecological niche may be as important as that of symbiotic fungi in forest nutrient cycling, particularly nitrogen. For example, Li et al. (1992) studied N2-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae, and found that water extracts of mycorrhizal root tips enhanced nitrogenase activity of the bacterium, indicating the potential importance of these two organisms in the forest floor nitrogen cycle. Li and Hung (1987) and Amaranthus et al. (1990) also reported N2-fixing soil bacteria belonging to Azospirillum and Clostridium that were associated with Douglas-fir mycorrhizae. Seedling growth increases in response to inoculation with Bacillus and Pseudomonas under greenhouse conditions described in the previous chapter have been observed in other studies with 53 Douglas-fir, interior spruce, and lodgepole pine (Chanway etal. 1991a, 1994; O'Neill etal. 1992), but such studies often exclude evaluation of the number and types of mycorrhizae. If PGPR are really MHB, then variable effects on mycorrhizae formation may be a factor contributing to seedling growth response variability after PGPR inoculation. Therefore, the objective of this study was to evaluate the interaction of mycorrhizal fungi and PGPR in relation to growth promotion of spruce and pine seedlings under controlled conditions. 54 3.2 MATERIALS AND METHODS 3.2.1 Experiment 4: Spruce and pine seedling growth and mycorrhizal infection after inoculation with PGPR 3.2.1.1 Microorganisms. The four PGPR strains described in the previous chapter, i.e., strains Sm3RN, Ss2RN, Sw5RN, and Pw2R, as well as two additional Bacillus PGPR strains, L6-16R (Holl and Chanway 1992) and S20R (O'Neill et al. 1992; Shishido et al. 1996a) were used for assessment of PGPR x mycorrhizae interactions. These additional strains were included to facilitate comparison of PGPR belonging to two common, but physiologically distinct genera, Pseudomonas and Bacillus. In addition, strain L6-16R has been well characterized as a PGPR for both agricultural and forest plant species. Strain L6 was originally isolated from rhizosphere soil containing roots of perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.) (Chanway et al. 1990), whereas S20 was isolated from spruce rhizosphere soil (O'Neill et al. 1992). The identities of' strains L6, L6-16R, S20 and S20R were determined by GC-FAME and Biolog™ as described in Chapter 2. Similarity indices associated from these tests as well as intrinsic resistance to selected antibiotics are presented in Appendices 2 and 3, respectively. 3.2.1.2 Seeds and soils. Interior spruce and lodgepole pine seeds were obtained from B.C. Ministry of Forest Tree Seed Center. Williams Lake seedlots of both species described in Chapter 2 were used. Before use, seeds were surface-sterilized by 2.5% NaCIO for 2 min and stratified by rinsing in mnning tap water overnight followed by storage at 4°C for 30 days. To provide a source of mycorrhizal inoculum for seedlings, forest floor soils (top 20 cm) were collected from three locations in B.C.: near Mackenzie, Williams Lake, and Salmon Arm (see Table 2.1 for locations). At each location, one stand each of interior spruce and lodgepole pine were selected for soil collection. This resulted in six forest floor 'types', which differed in stand history and geographical location, for 1This experiment was performed as part of Experiment 3 in Chapter 2. 55 use in assays. Four areas within each forest stand were randomly selected for soil collection. Soil samples from within each stand were thoroughly mixed, sieved (<2 mm particle size), and stored at 4°C for a maximum of 14 days before assays were initiated. 3.2.1.3 Seedling growth and bacterial inoculation. Procedures for seedling growth and PGPR inoculation were the same as those described in Experiment 3 of Chapter 2. In addition, a set (15) of seedlings was grown without forest soil to check if the pasteurized greenhouse soil contained ectomycorrhizal propagules. Seedlings were destructively harvested 15 weeks after sowing and bacterial inoculation. Roots were separated from shoots and seedling height was measured before shoot biomass was oven- dried (70 C for 3 days). 3.2.1.4 Ectomycorrhizae and bacterial root colonization assessment. Root systems of all seedlings were scanned visually and mycorrhizal root tips were counted and assigned to one of three broad groups: those resembling (i) Wilcoxina sp. (E-strain) for yellow brown to dark brown tips, mostly single to pinnate on spruce and single to dichotomous on pine, (ii) Thelephora sp., for whitish to pale brown tips, mostly single to pinnate on spruce and single to dichotomous on pine, or (iii) others. Identification of mycorrhizae was confirmed by sending representative samples to Dr. H.B. Massicotte (University of Northern B.C.). "Others" included rare or small numbers of Tuber-like, Mycelium radicis atrovirens and Amphinema-like ectomycorrhizae (Shishido et al. 1996a, 1996b). For bacterial root colonization, the distal section (5-6 cm) of the longest root from 4 randomly selected seedlings in each treatment was assessed when seedlings were harvested. The procedure for assessing population size (cfu-g *) was the same as that described in Experiment 3. 3.2.1.5 Experimental design and statistical analysis. A complete factorial design was used to generate 42 bacterial strain x forest floor treatment combinations (seven bacterial treatments, i.e., strains L6-16R, Pw2R, S20R, Sm3RN, Ss2RN, Sw5RN and uninoculated control x six forest floor types, i.e., pine and spruce stands at 56 Mackenzie, Williams Lake, and Salmon Arm) for each conifer species. Containers were arranged on a bench in the greenhouse according to a randomized complete block design with 15 blocks (i.e., each block contained all bacteria x forest soil treatments). %2 analysis was used to determine if there were significant differences in the number of seedlings that developed mycorrhizal root tips between bacteria x forest floor soil treatments, and correlation analysis was performed to determine if there was a significant relationship between the number (log transformed) of mycorrhizal root tips per root system and seedling biomass for each conifer species. In general, seedling height and biomass data did not violate assumptions of ANOVA, therefore, analyses were conducted on untransformed data. A two-way ANOVA for a blocked design was performed on seedling growth data. Mean separations were accomplished using Fisher's protected LSD tests (two tailed). To determine if bacterial promotion of seedling growth was related to the presence of mycorrhizae, data were also separated into two sets, one for seedlings that had developed mycorrhizal roots and the other for those that had not, and an ANOVA was performed on each set separately. Mean separation within each of these data sets was accomplished as described above. The biomass of mycorrhizal and non-mycorrhizal seedlings, averaged over all inoculation and forest soil treatments, was also compared using Student's t test. 3.3 RESULTS1 3.3.1 Effects of PGPR inoculation on mycorrhizal formation. Infective mycorrhizal propagules were present in all six forest soils. Seedlings grown in greenhouse soil alone did not develop mycorrhizae. Except for spruce seedlings treated with the Salmon Arm pine stand soil, there were no significant differences in the mycorrhizal status of conifer seedlings treated with different forest floor soil types (Table 3.1). Pearson % for spruce and pine seedlings grouped by soil type was 22.7 (PO.05) and 10.5 (P=0.395), respectively. ^All statistical tables for Chapter 3 are presented in Appendix 5. 57 Table 3.1 Number of non-mycorrhizal and mycorrhizal seedlings in response to addition of pine and spruce stand soil from Mackenzie, Williams Lake, or Salmon Arm. Mackenzie Salmon Arm Williams Lake Total Pine stand Spruce stand Pine stand Spruce stand Pine stand Spruce stand Spruce seedlings Non-mycorrhizal 67 75 85 65 81 76 449 Mycorrhizal1 38 30 18 40 24 28 178 Dead 0 0 2 0 0 1 3 Pine seedlings Non-mycorrhizal 59 65 73 69 71 60 397 Mycorrhizal1 45 39 32 36 34 45 231 Dead 1 1 0 0 0 0 2 Mycorrhizal seedlings had at least one mycorrhizal root tip. No mycorrhizae were detected on seedlings grown without forest soil. The frequency of mycorrhizal infection was unaffected by bacterial inoculation in both conifer 2 species: Pearson % grouped by inoculation treatment was 5.12 (P=0.954) and 13.9 (P=0.308) for 2 spruce and pine seedlings, respectively (Table 3.2). % analyses also indicated that none of the bacterial treatments showed significant differences in mycorrhizal seedling frequencies within each forest soil type. 3.3.2 Seedling growth promotion by PGPR inoculation and mycorrhizal status. In contrast to their influence on formation of mycorrhizae, PGPR inoculation generally showed a stimulatory effect on seedling growth of both species (Table 3.3). The comparatively low number of replicates within each bacterial inoculum x forest soil treatment combination precluded detection of statistically significant differences from uninoculated controls within individual forest soil treatments, as in Chapter 2. However, when averaged across all forest soil types, seedling biomass, in particular that of roots, was significantly enhanced by inoculation with all PGPR strains, except for pine with Sm3RN. Uninoculated control spruce seedling biomass showed a low, but statistically significant correlation with the number of mycorrhizal root tips, however this trend was not detected in pine (Fig. 3.1). The correlation of mycorrhizal root tip numbers to spruce seedling biomass was obscured in the presences of strains Pw2R, S20R, or Sw5RN (Fig. 3.2), but strain Sm3RN enhanced it. In addition, pine seedlings treated with Sm3RN also showed this correlation (Fig. 3.1b and Fig. 3.3). To evaluate the influence of mycorrhizae in seedling growth promotion by bacteria, root and shoot biomass were segregated into two groups, one for mycorrhizal seedlings and the other for non-mycorrhizal seedlings, so that the effect of PGPR inoculation could be analyzed within each group. Mycorrhizal infection appeared to be related to spruce seedling growth because mycorrhizal spruce seedlings were 14.5% (P<0.01) heavier than non-mycorrhizal spruce seedlings. No such difference was detected in pine: mycorrhizal pine seedlings were 2% heavier 59 Table 3.2 N u m b e r o f n o n - m y c o r r h i z a l a n d m y c o r r h i z a l seedl ings i n r e s p o n s e t o P G P R i n o c u l a t i o n t rea tments . U n i n o c u l a t e d L 6 - 1 6 R P w 2 R S 2 0 R S m 3 R N S s 2 R N S w 5 R N c o n t r o l T o t a l Spruce seedlings N o n - m y c o r r h i z a l 6 6 M y c o r r h i z a l 1 2 4 65 6 7 6 4 61 6 2 2 5 23 2 5 2 8 2 7 6 4 2 6 4 4 9 1 7 8 Pine seedlings N o n - m y c o r r h i z a l 6 8 M y c o r r h i z a l 1 2 2 5 2 3 8 5 7 33 52 3 7 5 6 3 3 53 3 7 5 9 31 3 9 7 2 3 1 1 M y c o r r h i z a l s e e d l i n g s h a d at least o n e m y c o r r h i z a l r o o t t ip . N o m y c o r r h i z a e w e r e d e t e c t e d o n s e e d l i n g s g r o w n w i t h o u t fores t so i l . T h e r e w e r e n o s ta t i s t i ca l ly s i gn i f i cant d i f f e rences d u e t o P G P R i n o c u l a t i o n t r e a t m e n t b y P e a r s o n % test: % = 5 .12 ( P = 0 . 9 5 4 ) f o r s p r u c e s e e d l i n g s a n d % 2 = 1 3 . 8 9 ( P = 0 . 3 0 8 ) f o r p i n e seedl ings . Table 3.3 The effect of PGPR inoculation on seedling height and biomass. Spruce seedlings Pine seedlings Inoculum Seedling height Shoot biomass Root biomass SeedUng height Shoot biomass Root biomass (mm) (mg) (mg) (mm) (mg) (mg) L6-16R 46.51 ± 1.2 74.8 ± 2.8 73.7 ± 2.2 ** 57.4 ± 0.9 95.0 ± 2.4 104.7 ± 2.4 ** Pw2R 46.6 ± 1.0 77.2 ± 2.7 75.8 ± 2.2 *** 56.8 ± 0.8 98.7 ± 2.4 104.4 ± 2.8 ** S20R 47.8 ± 1.0 73.4 ± 2.5 74.0 ± 2.0 ** 58.3 ± 0.9 100.3 ± 2.4 112.4 ± 3.0 *** Sm3RN 47.6 ± 0.9 77.2 ± 2.4 78.8 ± 2.2 *** 59.9 ± 1.0 99.4 ± 2.8 101.1 ± 2.6 Ss2RN 48.0 ± 1.1 76.4 ± 2.8 79.2 ± 2.8 *** 58.6 ± 0.9 100.2 ± 2.6 103.7 ± 2.8 ** Sw5RN 47.8 ± 1.1 75.5 ± 2.8 79.1 ± 2.7 *** 58.6 ± 0.9 99.6 ± 2.7 102.3 ± 2.9 * Uninoculated control 45.1 ± 1.0 69.2 ± 2.5 66.5 ± 1.9 55.6 ± 0.9 91.3 ± 2 . 5 95.5 ± 2.4 ! Mean±S.E. , n=90. ***, ** and * indicate significant difference from uninoculated control atPO.Ol, PO.05 andPO.l, respectively. 3 (a) Spruce r=0.426 (P=0.030) c o • 4 3 O s N 1 o l-l 4> J P 3 1 1 1 1 1 0 50 100 150 200 250 300 (b) Pine r= -0.185 (P=0.319) 2 -I I I i I j 0 50 100 150 200 250 300 350 Seedling biomass (mg) Fig. 3.1 Correlation between seedling biomass and the number (log transformed) of mycorrhizal root tips detected in root systems after inoculation with sterile phosphate buffer (uninoculated control) for (a) spruce and (b) pine seedlings. • , Wilcoxina sp. (E-strain); • : Thelephora sp.; A, others (Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings. P=significance level associated with each r. 62 L6-16R r = 0.369 (P = 0.076) 50 150 250 •s* 15 2 N 1 U o 6 0 5 Pw2R •••• v.-r = -0.065(^  = 0.759) 50 150 250 50 150 250 Seedling biomass (mg) Sm3RN r = 0.482 (P = 0.017) 150 250 150 250 1 -Sw5RN #• = 0.220 (i> = 0.271) 50 150 250 Seedling biomass (mg) Fig. 3.2 Correlation between spruce seedling biomass and the number (log transformed) of mycorrhizal root tips detected in seedling root systems after inoculation with PGPR strains. #, Wilcoxina sp. (E-strain); • : Thelephora sp.; • , others (Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings, ^ significance level associated with each r. 63 2 -•a •*-» 0 s •a 1 o o >> a o 60 o 2 -2 -1 -50 150 250 350 Seedling biomass (mg) L6-16R r=0.2U (P=0.206) • • • 2 -• • . • i % • A 1 • • • " » " A • • l i t 50 150 250 350 •S* Pw2R o s r=0.192(7J=0.247) "S ;orrhi2 2 -A •• T of myc 1 -• • • • •• • • A 1 1 1 1 3 50 150 250 350 60 O Sm3RN r=0.280 (P=0.093) A t f % • i i i i_ 50 150 250 350 Ss2RN r= -0.173 (P=0.335) _ l I 50 150 250 350 Sw5RN r= -0.227 (P=0.176) • Am 50 150 250 350 Seedling biomass (mg) Fig. 3.3 Correlation between pine seedling biomass and the number (log transformed) of mycorrhizal root tips detected in seedling root systems after inoculation with PGPR strains. • , Wilcoxina sp. (E-strain); • : Thelephora sp.; • , others {Amphinema sp., Mycelium radicis atrovirens, or Tuber sp.) refer to the fungal genera that formed mycorrhizae. r (Pearson correlation coefficient) was calculated using pooled data for all mycorrhizal types on seedlings, insignificance level associated with each r. 64 than non-mycorrhizal seedlings (P=0.236). Spruce seedling growth promotion by PGPR was detected to a similar extent in both mycorrhizal and non-mycorrhizal seedling groups (Fig. 3.4). This trend was also observed for pine treated with Bacillus strains L6-16R, Pw2R, and S20R (Fig. 3.5). However, Sm3RN-treated pine shoot biomass was significantly enhanced only when seedlings were mycorrhizal. In contrast, strain Sw5RN inoculation promoted pine seedling growth only when the seedlings were non-mycorrhizal. In general, statistically significant differences from uninoculated controls in both spruce and pine were more often detected in non-mycorrhizal seedlings (Fig. 3.4 and 3.5). 3.3.3 Bacterial root colonization and mycorrhizal status. Root colonization by Bacillus was below detection limits; only strain Pw2R was detected in a dilution plate of a mycorrhizal spruce root washing. Meanwhile, notwithstanding the small amount of root tissue used for assessment of bacterial colonization, only 10% of spruce and 11% of pine root segments had Pseudomonas population sizes below the assay detection limit. Average Pseudomonas rhizosphere population sizes were 6.3 x 103 - 1.3 x 105 cfu-g'1 rhizosphere soil. Though no significant differences in bacterial population sizes were detected between strains or conifer species, two consistent trends were discernible. Strain Sw5RN tended to colonize root systems with fewer cells than strains Sm3RN and Ss2RN, and mycorrhizal root tips tended to support slightly higher populations of Pseudomonas than non-mycorrhizal root tips (Fig. 3.6). 65 (a) Shoot 100 -Non-mycorrhizal Mycorrhizal Fig. 3.4 Treatment means (and standard errors) for (a) shoot and (b) root biomass of non-mycorrhizal and mycorrhizal spruce seedlings inoculated with a PGPR strain or sterile phosphate buffer (control). ** and * indicate significant difference from control at P<0.05 andPO.l, respectively. See Table 3.2 for n. 66 (a) Shoot Non-mycorrhizal Mycorrhizal . 3.5 Treatment means (and standard errors) for (a) shoot and (b) root biomass of non-mycorrhizal and mycorrhizal pine seedlings inoculated with a PGPR strain or sterile phosphate buffer (control). ** and * indicate significant difference from control at P<0.05 and P<0.1, respectively. See Table 3.2 for n. 67 (a) Spruce (b)Pine o CO s 5 CO t>0 O Seedlings: I I Non-mycorrhizal H H 1 Mycorrhizal Sm3RN Ss2RN Sw5RN PGPR strain Fig. 3.6 Rhizosphere colonization of (a) spruce and (b) pine seedlings by Pseudomonas strains Sm3RN, Ss2RN, and Sw5RN (means and standard errors; n=7-17). 68 3.4 DISCUSSION Because seedlings grown solely in greenhouse soil were uniformly non-mycorrhizal, addition of forest soil was necessary for mycorrhizal root tip formation. Visual examination of seedling root systems treated with forest soil indicated that mycorrhizae had developed on 29% of spruce and 37% of pine seedlings. Most mycorrhizae were formed by Wilcoxina-like sp. (E-strain) (93% for spruce and 64% for pine), followed by a small fraction of Thelephora-like fungi, comprising 4% of spruce and 18% of pine mycorrhizae. While Wilcoxina sp. have been shown to be common primary colonizers of containerized lodgepole pine and white spruce (Danielson and Visser 1990), it was surprising that comparatively few other mycorrhizal types were detected. The relatively small number of mycorrhizal seedlings and the preponderance of Wilcoxina sp. (E-strain) mycorrhizae can likely be attributed to the small amount of forest soil used in assays (2 mL) and the comparatively short seedling growth period, which may have precluded detection of less abundant or more fastidious, slower root colonizing mycorrhizal species. However, this degree of mycorrhizal infection was desirable, because it facilitated analysis of PGPR effects on separate mycorrhizal and non-mycorrhizal seedling pools (see below). The relationship between growth promotion of conifer seedlings and the abundance of mycorrhizae in root systems was of particular interest in this experiment. The frequency of mycorrhizal infection for both spruce and pine seedlings was unaffected by inoculation with any of PGPR strains, but mycorrhizal root tips tended to support slightly higher populations of Pseudomonas than non-mycorrhizal roots. These observations suggest that the processes of mycorrhizal formation and bacterial root colonization were not coupled, but that mycorrhizae may provide additional colonization sites or altered/enhanced root exudation in the mycorrhizosphere (Linderman 1988). In addition, ANOVA revealed no interaction between forest soil type and PGPR inoculation on seedling biomass, indicating that PGPR effects were similar in the presence of different forest soils. Therefore, the primary effect of each forest soil in this assay was to induce mycorrhizal root 69 tip formation. Bacterial inoculation had no significant effect on the mycorrhizal status of spruce and pine seedlings, but seedling growth was stimulated by most PGPR strains. (The relatively small number of seedlings available for this analysis likely contributed to the loss of statistical significance of bacterial inoculation effects on mycorrhizal seedlings.) A similar result was obtained for mycorrhizal lodgepole pine after co-inoculation with B. polymyxa strain L6 and Wilcoxina mikolae, but growth promotion of non-mycorrhizal seedlings was slight and insignificant, possibly due to the severe nitrogen limitation imposed on seedlings in that study (Chanway and Holl 1991). In my study, spruce and pine seedling growth promotion by Bacillus was equivalent for mycorrhizal and non-mycorrhizal seedlings, which suggests that Bacillus stimulated conifer seedling growth through a mechanism that was unrelated to mycorrhizal fixngi. Biomass of control spruce seedlings was positively correlated with the number of mycorrhizal root tips (r=0.426; P=0.03). In addition, mycorrhizal spruce seedlings were 16% (P=0.026) heavier than non-mycorrhizal spruce, but no such effect was detected on pine. The lack of relationship between pine biomass and frequency of mycorrhizal infection may have, in part, resulted from the artificial conditions under which seedlings grew (see above). Alternatively, the difference between spruce and pine in apparent response to the presence of mycorrhizae may result from the successional status of these species. Pine is thought to be an early successional conifer species adapted to relatively rapid initial growth rates, whereas spruce a later successional species with lower initial growth rates (Owens and Molder 1984a, b). Therefore, pine may be less dependent on symbiotic fiingi during early development and growth stages than spruce. Because forest soils had differential effects on seedling growth but not on the abundance of mycorrhizal root tips of pine seedlings, they likely contained plant growth altering agents other than mycorrhizal fiingi, e.g., actinomycetes. Perry etal. (1989) and Colinas etal. (1994a, b) have observed similar effects in soil transfer studies with outplanted Douglas-fir seedlings. Inoculation of spruce with strain Sm3RN increased the strength of the correlation between mycorrhizal root tip numbers and seedling biomass of uninoculated control. The same treatment caused a similar effect but of reduced magnitude in pine, suggesting that this strain enhances the 70 mycorrhizal dependence of host seedlings. Bacteria that reside on or in mycorrhizal tissues have been shown to stimulate root infection by certain mycorrhizal fungi (Bowen and Theodorou 1979; Garbaye and Bowen 1987; Garbaye and Bowen 1989), and such effects have been cited as one of the mechanisms by which bacteria influence growth of mycorrhizal conifer seedlings (Garbaye 1994). However, there was no significant difference in the number of mycorrhizal seedlings treated with SPB or Sm3RN. Therefore, stimulation of pine seedling growth by strain Sm3RN appears to be facilitated by mycorrhizae, but not at the level of infection by mycorrhizal fiingi. McAfee and Fortin (1988) also observed a stimulatory effect of a soil suspension, presumably containing bacteria, on shoot length of mycorrhizal jack pine (Pinus banksiana Lamb.) and American larch (Larix laricina (Du Roi) K. Koch) in the absence of an effect on mycorrhizal root formation. Seedling biomass, however, was either unaffected or depressed by the soil suspension. The results of conifer seedling performance in this experiment suggest that a synergistic, metabolic interaction involving existing mycorrhizae and bacteria may occur, and that a product of this interaction could enhance plant growth. Such a product could be a metabolite, whose production results from the direct interaction of the organisms (e.g., bacteria produce a precursor of a plant growth stimulating compound that is subsequently transformed or activated by the mycorrhizal fungus). Alternatively, a plant growth promoting compound, perhaps with phytohormonal or antibiotic properties, might be produced by bacteria and transported to the host by the mycorrhizal fungus. Similarly, if bacterial enzymes such as phosphatase or metabolites such as phosphate solubilizing organic acids enhanced nutrient availability (e.g., phosphate) in the mycorrhizosphere, the fungus might facilitate uptake of the additional nutrient, resulting in seedling growth promotion (Linderman 1988). The inability of strain Sw5RN to promote growth of mycorrhizal pine seedlings, therefore, suggests that pine mycorrhizae utilize or neutralize metabolites produced by the strain that are involved in seedling growth promotion, or alternatively, that mycorrhizae may inhibit bacterial production of seedling growth promoting compounds. 71 In this chapter, I presented results that suggest certain bacteria may specifically stimulate mycorrhizal seedling performance via a mechanism unrelated to simple enhancement of fungal infection. My experiments also indicate that synergistic interactions between PGPR and mycorrhizae are not required for seedling growth promotion. The experiments described in this chapter were not repeated, so conclusions regarding the behavior of individual strains (e.g., Sm3RN and Sw5RN) should be regarded as tentative. However, the finding that PGPR can stimulate conifer seedling growth irrespective of seedling mycorrhizal status was supported by results of two other studies I have performed. The first was of similar design to the one described in this chapter, but with different PGPR strains (Shishido et al. 1993), and the second utilized the six PGPR strains described in this chapter in a microcosm system similar to that described in chapter 6 of this thesis. Therefore, the hypothesis that PGPR stimulate conifer seedling growth through a mechanism dependent on mycorrhizal fimgi cannot be accepted, rendering it unlikely that mycorrhizal fungi contribute significantly to spruce seedling growth response variability after treatment with strain L6-16R, Pw2R, S20R, Sm3RN, Ss2RN or Sw5RN. 72 CHAPTER 4 ROOT COLONIZATION BY SPRUCE PGPR 4.1 INTRODUCTION AND LITERATURE REVIEW Root colonization by PGPR is an obvious prerequisite for plant growth promotion (Hoflich et al. 1994). Notwithstanding numerous studies of PGPR colonization dynamics, no threshold population sizes of PGPR below which efficacy is reduced or lost, have been reported. This is not surprising considering the complexity of biotic and abiotic interactions involved. For example, we know that root exudates and rhizodeposits vary significantly between plant genotypes, ecotypes and species (Hedges and Messens 1990; Smith 1976). If one accepts the view that root exudation and rhizodeposition drive the rhizosphere effect, then qualitative and quantitative differences in these compounds resulting from genetic differentiation and environmental influences undoubtedly result in differences in root-associated microflora (Campbell and Greaves 1990; Garland 1996b). This would very likely include root colonization by PGPR. In addition, root colonization is often only crudely assessed using dilution plate assays, but such techniques yield little information regarding the specific root microsites colonized by bacteria. Differences in microsite colonization represent another potentially important factor that could influence PGPR efficacy, particularly if bacteria are able to enter root tissues (Shishido et al. 1995). It has been demonstrated in a number of studies that certain PGPR strains may proliferate not only on and around plant roots, but also inside root tissues. For example, Baldani et al. (1983) found evidence for the involvement of internal root colonization by PGPR using growth promoting Azospirillum strains previously isolated from surface sterilized roots of wheat. The number of Azospirillum cells detected within wheat roots was well correlated f/=0.92) with total nitrogen accumulation, while no relationship was observed between the number of Azospirillum cells in the rhizosphere and total nitrogen accumulation. Pseudomonads have also been shown to colonize internal root tissues and promote tomato growth by displacing indigenous deleterious rhizobacteria (van Peer and Schippers 1989). Van 73 Peer et al. (1990) subsequently found that Pseudomonas strains isolated from the surface of hydroponically grown tomato (Lycopersicon esculentum Mill.) roots were biochemically distinct from those originating from inside root tissues, and that in general, endophytic isolates were more effective colonizers of internal root tissues than were isolates derived from the root surface. Some Bacillus strains have also been detected inside plant tissues. Lalande et al. (1989) found that population sizes of endophytic growth promoting Bacillus isolates of maize (Zea mays L.) exceeded those in the rhizosphere. Also, endophytic Bacillus strains have been demonstrated to be effective biocontrol agents (Pleban et al. 1995). B. cereus, originally isolated from surface sterilized Sinapis arvensis L., was inoculated onto germinating cotton (Gossypium indicum Lam.) seeds and clearly detected in vascular tissues 72 days later using radioactive labeling. Coincidentally, disease incidence caused by Rhizoctonia solani Kiihn was reduced by 51%. Similarly, Hall and Davis (1990) demonstrated that a strain of Bacillus subtilis antagonistic to Verticillium dahliae Kleb., moved into newly formed silver maple (Acer saccharinum L.) and sugar maple (A. saccharum Marsh) xylem tissue two years after inoculation. In conifers, Hallaksela et al. (1991) found that Bacillus was a predominant bacterial endophyte in young (one- and five-year-old) Norway spruce (Picea abies (L.) Karst), and that isolates were either pectolytic (B. pumilus) or cellulolytic (B. subtilis). Moreover, O'Neill et al. (1992) demonstrated that five of their six most effective spruce PGPR strains originated from within root tissue of naturally-regenerating spruce seedlings. From a theoretical perspective, the occurrence of beneficial endophytic microorganisms is sensible. Law and Lewis (1983) reviewed several types of mutualistic symbioses, and suggested that microorganisms which colonize internal root tissues may be buffered against environmental extremes more so than their external root tissue-colonizing counterparts. The foregoing discussion of endophytic PGPR provided some evidence to support this idea from the perspective of microbial colonization of plant tissues. Additional benefits accruing to host plants from endophytic PGPR compared to rhizosphere colonizing PGPR remain to be demonstrated. 74 From a practical viewpoint, the possibility of enhancing plant growth promotion efficacy and/or reproducibility with endophytic microorganisms offers obvious benefits in industries that depend on some aspect of enhanced plant productivity. However, from an ecological perspective, the development of such relationship is most intriguing, as these may be precursors to, or stages in the evolution of new mutualistic symbioses. Alternatively, they may represent the end-point in symbioses that are morphologically less expressive than root nodules or mycorrhizae, but with intimate physiological relationships that have developed through co-adaptive or co-evolutionary processes. Such relationships may influence plant physiology in ways that have not yet been elucidated (Misaghi and Donndelinger 1990), and could include greater plant growth promotion efficacy compared to rhizosphere colonizing PGPR. The most common technique for evaluation of endophytic bacterial populations has been dilution plating on artificial media using antibiotic-resistant derivative strains (Kloepper and Beauchamp 1992; Kluepfel 1993). Though rapid and inexpensive, this approach is indirect and unable to detect the specific location of bacterial cells in plant tissues. Moreover, some studies indicate that this technique might underestimate the actual population sizes of target bacteria due to variable recovery of antibiotic-resistant derivatives on media (Kluepfel 1993; Mclnroy et al. 1996). Therefore, confirming internal plant tissue colonization by endophytic PGPR requires technical approaches more sophisticated than simple dilution plating. Patriquin and Doberenier (1978) used tetrazolium dye to locate endophytic diazotrophs in cortical cells of several members of the gramineae such as maize, sorghum {Sorghum vulgare L . ) and wheat under a light microscope. However, such redox dyes generally respond to any chemical that is oxidized in plant tissues and results in the formation of NADH. Therefore, plant particles such as starch grains are also stained and are indistinguishable from bacterial cells (Patiquin and Doberenier 1978). Alternatively, Bell etal. (1995) used acridine orange dye to estimate the number of endophytic bacteria extracted from grapevine (Vitis vinifera L.) xylem with epifluorescent microscopy, and found consistently higher population sizes of xylem colonizing bacteria than was indicated through conventional dilution plating. However, acridine orange also 75 stains mucopolysaccharides (Culling 1974), again rendering it is difficult to distinguish bacterial cells from plant organelles. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) have also been extensively used to detect and localize endophytic bacteria in planta (Dong et al. 1994; Hurek et al. 1994; Levanony et al. 1989; Quadt-Hallmann and Kloepper 1996; Ruppel et al. 1992; Sardi et al. 1992). Problems with electron microscopy seem to occur during sample tissue preparation. First, specimens are sectioned and risk bacterial contamination from outer plant tissue. Though epoxy resin-embedded-specimens for TEM are probably less vulnerable to this problem, the resin itself may cause leakage of endophytic bacterial cells from the apoplastic fluid of the cut surface (Dong et al. 1994). Alternatively, laser scanning confocal microscopy allows three-dimensional image observation without preparing thin tissue sections, and should provide a powerful tool for in planta observation of endophytic bacteria (Schloter et al. 1993, 1995). Rapidly advancing molecular technology has also been applied to evaluation of rhizosphere bacteria in soil (Kloepper and Beauchamp 1992). For example, Steffan and Atlas (1988) demonstrated that bacterial DNA probes for particular nucleic acid sequences for in situ hybridization possessed very low detection limits, e.g., 100 bacterial cells in 100 grams of sediment, when used in conjunction with the polymerase chain reaction (PCR). Undoubtedly, this technique is a potentially useful tool to detect endophytic bacteria due to its high degree of sensitivity. However, to date, most applications of nucleic acid hybridization using environmental samples have involved probing immobilized DNA sequences fixed to nitrocellulose or nylon membranes (Pickup 1991), rendering in planta detection indirect. For example, Hurek et al. (1994) used the Western blot technique and PCR to confirm the presence of bacterial DNA from roots, shoots and root bases of rice (Oryza sativa L.) after inoculation with Azoarcus sp. Immunological detection methods, on the other hand, have become increasingly important in microbial ecology for the tracking of specific microorganisms and for community analysis (Schloter et al. 1995). Among numerous applications of immunological techniques, immuno-fluorescence straining (IF) has been used for the detection and localization of various 76 microorganisms including viruses, bacteria, fimgi and protozoa (Schloter et al. 1995), and this technique has been successfully applied in detection and identification of plant endophytic bacteria, particularly pathogens, of agricultural crops (Alvarez et al. 1991; Kikumoto and Sakamoto 1967; Lin et al. 1987). However, there has been little work so far on conifers. It is well-known that polyclonal antisera contain a variety of antibodies directed against both typical as well as less common epitopes of the target cells. This variety in antibody composition often results in non-specific cross-reactions, and frequently necessitates production of highly specific (and labour intensive!) monoclonal antibodies, particularly for inplanta localization of target antigens in natural ecosystems (Lin et al. 1987; Schloter et al. 1995; Quadt-Hallmann and Kloepper 1996). Although highly specific antibodies to target bacteria are usually desirable, it is also known that monoclonal antibodies often lower sensitivity to secondary antibodies that conjugate to fluorochome enzymes, e.g., fluorescein isothiocyanate (FITC) and tetramethyl-rhodamine isothiocyanate (TRITC), because of the reduced number of epitopes associated with the antibodies (De Boer 1990). This problem can be solved with the double staining method, using two different monoclonal antibodies on a single target strain (De Boer 1990). Alternatively, where experimental biological factors are carefully controlled, i.e., under gnotobiotic conditions, and only one or a few microorganisms are being studied simultaneously, polyclonal antisera should theoretically be useful for in planta detection of endophytic bacteria. In Chapter 2,1 evaluated the degree of specificity involving conifer hosts, selected PGPR strains and soils, and concluded that such specificity did not contribute significantly to seedling growth response variability associated with bacterial inoculation treatments. In addition, external root (i.e., rhizosphere) colonization by PGPR was evaluated, but the possibility that some of these strains colonized internal root tissues was not. Moreover, Chapter 3 provided evidence that plant growth promoting efficacy was not dependent on the mycorrhizal status of spruce seedlings, which suggests that mycorrhizal fungi are not involved in the seedling growth response variability to PGPR inoculation. 77 H e n c e , a c c o r d i n g t o O b j e c t i v e 3 o f m y thes i s ( p a g e 7) , I u s e d JJF m i c r o s c o p y w i t h p o l y c l o n a l an t i s e ra as w e l l as d i l u t i o n p l a t i n g t o eva lua te t h e c a p a b i l i t y o f s ix P G P R strains t o c o l o n i z e i n t e r n a l t i s sues o f s p r u c e g r o w n u n d e r g n o t o b i o t i c c o n d i t i o n s . I n a d d i t i o n , c a r b o n subs t ra te u t i l i z a t i o n pa t te rns o f these b a c t e r i a l s trains w e r e d e t e r m i n e d u s i n g t h e B i o l o g ™ s y s t e m t o fac i l i ta te c o m p a r i s o n s o f s trains c a p a b l e o f e n t e r i n g p l an t t i s sues w i t h t h o s e o f s tra ins u n a b l e t o c o l o n i z e i n t e r n a l s p r u c e t i ssues . 7 8 4.2 MATERIALS AND METHODS 4.2.1 Experiment 5: Colonization of spruce by PGPR 4.2.1. 1 Microorganisms. The six PGPR strains described in the previous chapter, i.e., strains L6-16R, S20R, Pw2R, Sm3RN, Ss2RN, and Sw5RN were used for assessment of internal root colonization. Biolog™ (Biolog Inc., Hayward, CA) was used to determine carbon substrate utilization of each bacterial strain (Shishido et al. 1995). In addition to Biolog™ tests, pectolytic activity was determined according to the protocol described by Klement et al. (1990) using PEC-YA medium (Starr et al 1977). To visualize zones of clearing after bacterial inoculation and incubation for 48 h at 28°C, PEC-YA plates were flooded with 1% (w-v *) hexadecyl-trimethylammonium bromide. 4.2.1.2 Seeds and seedling culture. Salmon Arm spruce seeds (see Chapter 2) were used because these seeds possessed a high germination value and no internal bacterial colonization was detected after surface sterilization, aseptic trituration and dilution plating. For stratification, surface sterilized seeds were stored for 30 days at 4°C on moist, sterile gauze in sealed petri-plates. At the end of the stratification period, the effectiveness of this surface-sterilization procedure was confirmed by placing seeds on 10% strength TSA plates amended with 1% triphenyltetrazolium chloride for 3 days. Each germinating seed with no bacterial contamination was transferred aseptically to a plant culture tube (25 mm x 150 mm, Sigma) containing moist (ca. 40% vv *), autoclaved Turface (montmorillonite clay aggregate) (Applied Industrial Corp., Deerfield, IL). Each seed was subsequently inoculated with 1.0 mL of a suspension containing one of the PGPR strains and 5 ug of rifamycin, and covered with surrounding Turface in the tube. Bacterial inocula were prepared as described in Chapter 2 6 1 8 1 to densities of ca. 10 cfu-mL for Bacillus strains and ca. 10 cfu-mL for Pseudomonas strains. Each bacterial treatment had 20 replicate tubes, at least eight of which yielded seedlings. The inoculation procedure was repeated two weeks later for all tubes containing seedlings. 79 All tubes were placed in a growth chamber (Conviron, CMP3244) under a 19 h photoperiod, -2 -1 PAR at floor level of 160 pmol-m -s , and a 25:18°C light-dark temperature regime. Each tube was fertilized with 1.0 mL of a sterile solution containing (mg-L"1): N=20.0, P=5.0, K=10.0, Ca=10.0, Mg=5.0, S=9.6, Fe=0.4, Mn=0.05, Cu=0.02, Zn=0.02, B=0.02, Mo=0.005, and rifamycin=5.0 every 4 weeks after sowing. Seedlings were grown 4-5 months after sowing before root colonization was assessed. 4.2.1.3 Antibody production. Whole cells of each bacterial strain (L6-16R, Pw2R, S20R, Sm3RN, Ss2RN, or Sw5RN) were used to generate antibodies. Bacteria were grown for two days in broth culture as described in Chapter 2 for inoculum preparation and harvested at ca. 3000 g for 10 min. Pellets were washed twice with 10 mM phosphate-buffered saline (PBS, pH 7.4), then resuspended in fresh 8 -1 PBS. Cell densities were adjusted to ca. 10 cfu-mL based on optical density (0.9 ^ 440 units), which resulted in (cfu-mL"1): L6-16R=1.9 x 108; Pw2R=1.4 x 108; S20R=2.3 x 108; Sm3RN=4.3 8 8 8 x 10 ; Ss2RN=1.5 x 10 ; and Sw5RN=5.5 x 10 . Live Pseudomonas and heat-killed (autoclaved for 60 min) Bacillus cells were used as antigens because growth of these Bacillus strains was detected at 37°C, similar to mouse body temperature. Six-week old female mice were used for immunization following an injection scheme similar to that described by Ball et al. (1990). For each of the six PGPR strains, 0.1 mL of bacterial cell suspension (in PBS) was emulsified with an equal volume of Freund's adjuvant and injected subcutaneously on day 1. Intraperitoneal boosts prepared in the same way were adrninistered on days 21, 35, 49 and 67. Three mice were used for each bacterial strain. On day 77 titers were tested, found to be satisfactory and blood samples were collected. Because little difference in titer was found between mice injected with the same antigen, samples from mice treated with the same bacterial strain were pooled, allowed to clot for 1 h at 20°C and centrifuged at 2000 g for 10 min. The resultant supernatant was collected as antiserum and stored at -20°C. 80 4.2.1.4. Evaluation of antibody cross-reactivity. The cross reactivity of each antiserum with heterologous bacterial cells as antigens was tested using an indirect enzyme-linked immunosorbent assay (ELIS A) according to McLaughlin and g Chen (1990). Wells of microtiter plates were coated with 100 uL of each bacterial strain at 10 cells-mL"1 suspended in a coating buffer (50 mM carbonate-bicarbonate, pH 9.6). After drying at 37°C, wells were blocked with 2% milk in 10 mM PBS for 1 h at 37°C, and subsequently washed with PBS 3 times. Wells were then filled with the bacterial antisera (100 uL-well 1), comprising all possible combinations of antigens and antisera (diluted 1:5000 for Bacillus and 1:10000 for Pseudomonas), and incubated for 2 h at 37°C. After washing the wells, 100 uL of goat antimouse IgG (diluted 1:3000) conjugated with horseradish peroxidase (diluted 1:1000) were added and plates were incubated for 1 h at 37°C. After 3 washings with PBS, wells were treated with 100 uL of substrate solution (50 mL citric acid and Na2HP04 buffer at pH 5.0 with 17 mg o-phenylenediamine and 20 uL 30% H2O2). The reaction was stopped after 20 min by adding 100 uL of 1 N H2SO4. The absorbance values were read on a Biotech EL310 autoreader (Mandel Scientific) at 450 nm. 4.2.1.5 Immunofluorescent microscopy. Indirect immunofluorescent antibody staining (IFAS) was performed according to De Boer (1990). Seedlings were carefully removed from plant culture tubes and adhering soil particles were gently dislodged. Shoots and roots were separated and frozen individually in 10 mM PBS. Five seedlings for each bacterial treatment were examined by making thin-sections (approximately 20 sections per shoot or root). Thin-sections of frozen tissues were made by hand with a dissecting blade, and transferred to a glass slide coated with 0.1% poly-l-lysine in 10 mM PBS. After fixing with a few drops of cool 95% ethanol for 10 min, each section was blocked with 2% milk for 1 h at 37°C in a moist chamber and rinsed with 10 mM PBS. A drop of homologous antiserum (diluted 1:250 for Bacillus and 1:500 for Pseudomonas) was applied to each section and slides were incubated at 37°C for 60 min. After rinsing with 10 mM PBS, sections were treated with fluorescein isothiocyanate (FITC) conjugated with goat antimouse IgG (diluted 81 1:100) at 37°C for 60 min in a darkened moist chamber. Slides were then gently washed with distilled water and mounted in 50% glycerol in PBS with 0.1% /?-phenylenediamine (PPD). Slides were observed using an epifluorescent microscope (Zeiss Axiophot) at 495 nm with a high-pressure mercury lamp, a BP 450-490 excitation filter, and an LP 520 barrier filter. In some cases, phase contrast was used simultaneously with epi-illumination to improve resolution at low light intensities. Photo-micrographs were taken with a 35-mm Zeiss MC 80 Microscope Camera with ASA 1600 film (Fuji SuperHG). 4.2.1.6 Recovery of inocula from seedlings. Five-month old seedlings remaining after sample selection for microscopy were used for dilution plating. Three seedlings from each bacterial treatment were gently removed from tubes and separated into roots, stems and needles. To evaluate external root colonization, i.e., in the rhizosphere and on the rhizoplane, each root system was placed in a culture tube containing 10 mL of 10 mM SPB with glass beads and agitated for 1 min. Root washings were then serially diluted and plated onto 50% strength TSA amended with 100 mg-L"1 rifamycin and cycloheximide and 50 mg-L"1 nystatin for Bacillus, and KB A amended with 100 mg-L-1 nalidixic acid, rifamycin and cycloheximide and 50 mg-L 1 nystatin for Pseudomonas. Root washings were also plated on 50% strength TSA without antibiotics to check for contaminants in tubes. To evaluate internal colonization by bacterial inocula, stems and root systems were washed individually and surface-sterilized by immersion in 1.8% NaCIO for 1 min, followed by three rinses in sterile water. Tissues were then aseptically triturated in 10 mL SPB with a mortar and pestle, and serial dilutions of the resulting homogenates were plated (0.1 mL aliquots) on 50% strength TSA (for Bacillus) or KBA (for Pseudomonas) amended with antibiotics as described above. Dilution plates were incubated for up to 4 days at 28°C before colony numbers were assessed 82 4.3 RESULTS 4.3.1 Immunofluorescent microscopy. Antiserum raised for each bacterial strain was very sensitive to the original antigens in the 7 ELISA (Table 4.1). Using 10 antigen cells and crude antisera, Bacillus and Pseudomonas strains were easily detected at serum dilution of 1:5000 and 1:10000, respectively. Cross-reactivities were severe in heterologous combinations of Bacillus antigens and Bacillus antisera, moderate between Pseudomonas antigens and Pseudomonas antisera, and low between Bacillus antigens and Pseudomonas antisera. Although some autofluorescence of plant tissue was detected during IF, the bright green fluorescence associated with bacterial cells conjugated to FITC was easily detected with the epifluorescent microscope (Fig. 4.1, 4.2, 4.3., 4.4, 4.5, and 4.6). However, it was difficult to determine the exact tissues colonized by bacteria due to the possibility that bacterial cells were displaced during sample and slide preparation, e.g., Fig. 4.3, 4.4 and 4.6. Nevertheless, if it is assumed that sample preparation did not contribute significantly to the location of bacterial cells on slides, all strains were detected on external root tissues, including root hairs. Possible displacement of microorganisms during sample preparation could be a serious problem when microscopically evaluating microbial colonization of internal plant tissues. However, if reached in conjunction with surface-sterilization and dilution plating assays, conclusions regarding the endophytic capabilities of bacterial strains should be reliable. Due to the large bacterial populations that developed on and around the root epidermis, bacterial colonization of internal root tissues could not be reliably confirmed microscopically. All strains were detected on the peripheral internal root tissues, but cells could easily have been introduced during sample preparation. However two strains, Pseudomonas Sm3RN and Bacillus Pw2R, were repeatedly detected inside stem tissues of inoculated seedlings (Fig. 4.7 and 4.8), which implies that they penetrated root tissues and were transported systemically. None of the other strains were found in stem tissues, and no PGPR strains were detected in needle tissues. 83 Table 4.1 Cross reactivity of polyclonal antisera and PGPR strains used as antigens in an ELISA. Values presented are optical densities @ 450nm. Antiserum Antigen Bacillus sp. (x 5000) Pseudomonas sp. (x 10000) (10' cells) L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN L6-16R 0.832* 0.405 0.689 0.028 0.006 0.008 Pw2R 0.437 0.373 0.472 0.021 0.008 0.013 S20R 0.747 0.406 0.622 0.007 0.003 0.008 Sm3RN 0.106 0.081 0.151 1.103 0.534 0.508 Ss2RN 0.348 0.206 0.236 0.205 0.843 0.248 Sw5RN 0.235 0.115 0.315 0.385 0.614 1.259 * Homologous antigen-antiserum combinations are underlined. 84 Fig. 4.1 Irnmunofluorescent antibody staining of Bacillus strain L6-16R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Images were produced using phase contrast microscope under (a) fluorescent light (495 nm) and (b) visible light. Bacillus appears as tiny, fluorescent green cells associated with root surface and hairs. The particle seen in the bottom, center of each micrograph is a Turface aggregate. 85 Fig. 4.2 Jjnmunofluorescent antibody stairiing of Pseudomonas strain Sm3RN in a longitudinal section of a 4-month old spruce root inoculated at sowing. Images were produced using phase contrast microscope under (a) fluorescent light (495 nm) and (b) visible light. Pseudomonas appears as green fluorescent cells associated with root surface and hairs. 86 Fig. 4.3 Jjrrmunofluorescent antibody staining of Bacillus strain S20R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as tiny, fluorescent green cells associated with root surface. 87 Fig. 4.4 Immunofluorescent antibody staining of Pseudomonas strain Ss2RN in a longitudinal section of the central axis of a 5-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent green cells associated with root surface and hairs. 88 Fig. 4.5 Immunofluorescent antibody staining of Pseudomonas strain Sw5RN in a longitudinal section of the central axis of a 5-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent blight green cells associated with root hairs. 89 Fig. 4 . 6 Immunofluorescent antibody staining of Bacillus strain Pw2R in a longitudinal section of the central axis of a 4-month old spruce root that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as fluorescent green cells inside root tissues near the base of a lateral root. 90 Fig. 4.7 Immunofluorescent antibody staining of Pseudomonas strain Sm3RN in a longitudinal section of the stem of a 4-month old spruce seedling that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Pseudomonas appears as fluorescent green cells inside vascular tissues. 91 Fig. 4.8 Lnmunofluorescent antibody staining of Bacillus strain Pw2R in a longitudinal section of the stem of a 4-month old spruce seedling that was inoculated at sowing. Image was produced using phase contrast microscope under fluorescent light (495 nm). Bacillus appears as fluorescent green cells inside vascular tissues. 92 4.3.2 Recovery of inocula from the rhizosphere, root and stem tissues. Bacterial population sizes outside and inside roots and inside stems determined by dilution plating are presented in Table 4.2. Six of the 18 seedlings showed some fungal contamination in the rhizosphere, but no contaminants were found inside plant tissues. Strains L6-16R, S20R, Ss2RN and Sw5RN were recovered from internal root tissues of only one (of three) seedlings assayed and from a statistical perspective, did not develop population sizes different from zero. Though it cannot be concluded with certainty, I suspect that incomplete surface sterilization of root segments was responsible for these results. In contrast, internal root colonization was detected in all three seedlings treated with strains Pw2R or Sm3RN. These were also the only two strains that were recovered from stem tissues using this assay. 4.3.3 Carbon substrate utilization profiles of PGPR in vitro. Most differences in carbon substrate utilization among these PGPR strains seemed to reflect genus level characteristics (Table 4.3). However, notwithstanding genus level differences, D-sorbitol and D-galacturonic acid utilization were similar among the two stem-endophyte strains, Pw2R and Sm3RN. In contrast to the random pattern of utilization of the monomer, galacturonic acid, among the six PGPR strains, none of Pseudomonas strains had in vitro pectolytic activity while all of the Bacillus strains possessed it. 93 Table 4.2 Recovery of PGPR from the rhizosphere and internal root and stem tissues of spruce 5 months after seedlings were inoculated. Inoculum Rhizosphere (log cfu-g root) Internal tissue (log cfu-g 1 tissue) Root Stem L6-16R 5.6 ±0.3 1.7 ± 1.7 0 Pw2R 5.8 ±0.5 4.4 ±2 .2 5.0 ±0 .2 S20R 5.6 ±0 .4 1.7 ± 1.7 0 Sm3RN 7.6 ±0.1 3.9 ± 1.9 4.6 ±0.3 Ss2RN 7.5 ±0.1 2.0 ±2 .0 0 Sw5RN 7.1 ±0 .2 2.0 ±2 .0 0 94 Table 4.3 Carbon substrate utilization profiles of PGPR strains as indicated in Biolog assays. Biolog test reaction 1 Carbon substrate -• L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Dextrin + + + - - -Glycogen + + + - - -Tween 40 - - ± + + + N-Acetyl-D-glucosamine - - - + + + D-Arabitol - - - + + + Cellobiose + + + - - -L-Fucose - ± - - - -Gentiobiose + + + - -w-Inositol - - - + ± + a-D-Lactose + + + - - -Maltose + + + ± ± -D-Melibiose + + + - - -0-Methyl-D-Glucoside + + + - - -D-Raffinose + + + - - -Sucrose + + + - + + Lactulose + + + - - -D-Psicose ± + ± + ± + D-Sorbitol - + - + ± ± Turanose - + + - - -Acetic acid - ± - + ± + D-Galacturonic acid - + - + - + a-Hydroxybutyric acid - - - + + P-Hydroxybutyric acid - - - + + + a-Ketoglutaric acid - - - + + + a-Ketovaleric acid - + - ± - ± Propionic acid - ± - + ± + Succinic acid + + - + + + Succinamic acid — — — + ± 95 Table 4.3 Continued. Biolog test reaction Carbon substrate1 L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Alaninamide - - - ± ± + D-Alanine - - - + + + L-Alanine - - - + + + L-Alanylglycine - - - + + + L-Asparagine - ± - + + + L-Glutamic acid - - + + + L-Pyroglutamic acid - - + + + L-Serine - - + + + 2,3-Butanediol - ± - - - -Glycerol ± + + + + + Inosine + - + + + + Thymidine - + - - - -Uridine - ± ± ± + + Glucose-6-phosphate - + - - - -Pectolytic activity + + + - - -Only substrates that are common between GN and GP Biolog plates and that were utilized differently by the strains are listed. +: positive; - : negative; +: weakly positive. Determined in separate assays according to the protocol described by Klement et al. (1990). 96 4.4 DISCUSSION The primary objective of the experiments described in this chapter was to evaluate the capability of six PGPR strains to colonize external and internal spruce tissues using IF AS and surface sterilization-dilution plating assays. Although the procedures I employed are commonly used to evaluate endophytic colonization of plant tissues, unfortunately both techniques have limitations that restrict the degree of confidence with which conclusions can be reached. For example, mild sterilization procedures may result in incomplete root surface disinfestation, but more rigorous protocols can sterilize internal root tissues as well (Kloepper and Beauchamp 1992). Microscopic techniques render samples vulnerable to relocation of bacteria during tissue preparation, because sample specimens are cut or thin-sectioned before observation, possibly introducing bacterial contaminants from tissue surfaces (see Section 4.1). Indeed, all six strains I used were occasionally detected inside roots whether evaluated microscopically or by dilution plate counts using surface sterilized seedling segments. However, only Bacillus strain Pw2R and Pseudomonas strain Sm3RN were consistently found to be endophytic using both assays. Moreover, these were the only two strains that were also detected in vascular tissues of stem sections, suggesting that they were transported systemically within tissues of the host plant. While it can not be concluded with certainty, these results suggest strongly that strains Pw2R and Sm3RN are able to colonize internal spruce tissues. Bacterial movement has been detected apoplastically in intercellular spaces (Patriquin and Dobereiner 1978; Hurek et al. 1994). Other studies have also detected systemic colonization of endophytes as bacteria were recovered from within plant shoot tissues including stems and flowers after root inoculation (Lamb et al. 1996; Misaghi and Donnelinger 1990). Systemic colonization of the vascular system from root cortical tissues is probably limited by the endodermis (Kloepper et al. 1992b); however, some endophytic bacteria are apparently able to break this barrier. This was apparently the case for strains Pw2R and Sm3RN in this study. Endophytic colonization might be facilitated by the presence of a large bacterial population during 97 seed germination because bacterial cells, once inside cortical tissue, could pass through the endodermal barrier via the undifferentiated cells of root tips. Endophytic colonization of bacteria in "healthy" plant tissues is well established (Chanway 1996; Kloepper and Beauchamp 1992; Manion 1991), and the processes and routes by which bacteria enter plant tissues are active areas of investigation (Huang 1986; Quadt-Hallmann and Kloepper 1996; Sprent and de Faria 1988). Some studies suggest that internal colonization is facilitated by naturally occurring openings in plant tissues, e.g., stomata and lenticels or wounds made mechanically or by other organisms (Hallaksela et al. 1991; Huang 1986; Mahaffee et al. 1994; Sprent and de Faria 1988), but the precise mechanisms remain to be demonstrated. Since microorganisms other than PGPR were excluded from most plant tubes, the effects of other soil organisms can be excluded as facilitators of endophyte colonization. However, lateral root development is known to result in significant disruption of existing root tissues and in the development of potential entry points for microorganisms (Mahaffee et al. 1994). In an earlier study, strain Pw2R was detected inside root tissues of four-week old pine seedlings only after lateral roots had developed (Shishido et al. 1995). While only a few Bacillus cells were detected around the base of lateral roots (Fig. 4.6), this observation was made almost 5 months after inoculation, which presumably was significantly later than the timing of the initial bacterial ingress. In addition, wounds caused both naturally and accidentally during handling may have provided bacterial entry sites. Sprent and de Faria (1988) suggested that the major infection sites of heterotrophic N2 fixing microorganisms are naturally occurring wounds, root hairs, and epidermal conjunctions. Root surface openings including lateral root emergence and wounds may play a role in root colonization by endophytic bacteria, but their occurrence does not explain the differential abilities of the six PGPR strains to enter plant tissues: non-endophytic strains were also present at the sites of lateral root emergence but did not enter root tissues. Biolog™ profiles of the six PGPR strains revealed interesting contrasts of in vitro carbon substrate utilization patterns, but the only substrates that were differentially utilized between endophytes and non-endophytes were D-98 sorbitol and D-galacturonic acid. In both cases, strains Pw2R and Sm3RN were able to metabolize these substrates while the other four strains could not, or could do so only weakly. In addition, the differential ability to utilize D-sorbitol and D-galacturonic acid was also detected in their respective wild type strains. The biological significance of these differences in substrate utilization is not obvious, but it is possible that such differences are related to the differential ability of these strains to colonize different root microsites. For example, Mavingui et al. (1992) found that Bacillus polymyxa strains isolated from the rhizoplane of wheat were generally capable of metabolizing sorbitol whereas rhizosphere and non-rhizosphere isolates were not. They hypothesized that intense competition for oxygen would occur on the root surface owing to root respiration, which would result in selection pressure for bacteria capable of anaerobic growth on highly reduced substrates, such as sorbitol. The oxygen-depletion argument could easily be extended to internal root tissues as well. However, this hypothesis does not explain endophyte colonization with Sm3RN, as it is aerobic. In addition, the non-endophytes Ss2RN and Sw5RN also utilized D-sorbitol, though weakly. To facilitate root colonization, root endophytic bacteria may also possess the ability to metabolize structural components of plant cells. In particular, the ability to metabolize pectin (polygalacturonic acid), the primary component of the middle lamellae of plant cell walls, has been proposed to at least partly explain why bacterial root endophytes are often found in the root cortex intercellularly (Baldani and Dobereiner 1980). All of the Bacillus strains possessed pectolytic activity in vitro, while none of the Pseudomonas strains did. Interestingly, however, only the endophytic strains Pw2R and Sm3RN strongly metabolized D-galacturonic acid, the primary monomelic component of pectin (Paul and Clark 1989). The lack of pectolytic activity associated with Pseudomonas strains supports the results of van Peer et al. (1990). They found that tomato rhizobacteria, including endophytic Pseudomonas strains, generally had little or no pectolytic activity, but bacterial utilization of D-galacturonic acid was not evaluated. Although these observations demonstrate the possibility that differences exist in enzymatic activity between 99 endophytic and non-endophytic bacteria, little is known about the origin and regulation of such enzymes. The results in this chapter suggest that two of the six spruce PGPR strains, Pw2R and Sm3RN, are endophytic and can be transported systemically. In addition, these strains possess some physiological characteristics that are distinct from their non-endophytic counterparts. Whether or not the capacity to utilize D-sorbitol and D-galacturonic acid contributes to the ability to colonize specific root microsites could not be ascertained from my data and requires further evaluation. 100 CHAPTER 5 PGPR COLONIZATION AND GROWTH PROMOTION OF INOCULATED SPRUCE SEEDLINGS IN A FIELD TEST 5.1 INTRODUCTION AND LITERATURE REVIEW For agricultural and forestry applications, the ultimate evaluation of PGPR must be made in the field under environmental conditions similar to those that target plants are typically exposed. Suslow and Schroth (1982) emphasized the importance of colonization persistence in the rhizosphere during the growing season of an annual agricultural crop such as sugar beet, however, the objectives of agricultural management often differ from those of forestry. For agricultural crops, PGPR inoculation is expected to deliver short term, significant increases in plant yield through growth promotion or biocontrol. In forestry, where rotation times are generally measured in decades, PGPR inoculation would not necessarily be expected to have a direct positive influence on timber production. Instead, the goal of forest seedling inoculation with PGPR is to improve survival and establishment of outplanted seedlings, often through enhancing root system development, though reduction of disease incidence in the early years after outplanting is also desirable. Therefore, the effects of PGPR inoculation should persist beyond the first field season, ideally until host seedlings are well established in the field. However, PGPR population sizes often decline rapidly in the rhizosphere. For example, Kloepper et al. (1980b) observed decreases in potato rhizosphere PGPR population sizes from >104 cfu-cnr1 root to <103 cfu-cnr1 root after halting irrigation. For conifers, Holl and Chanway (1992) demonstrated the ability of strain L6-16R to colonize the rhizosphere of lodgepole pine one month after inoculation, but observed large monthly population declines thereafter. On the other hand, in a growth chamber study Bacillus strains L6-16R and Pw2R strains established large populations (i.e., >5 x 107 cfu-g"1 fresh root tissue) in the lodgepole pine rhizosphere 2 weeks after inoculation, with only slight declines in population size 4 weeks after inoculation 101 (Shishido et al. 1995). In general, though, very little is known about root colonization population dynamics of PGPR under field conditions at reforestation sites. As I described in Chapter 4, strains Pw2R and Sm3RN appear to be capable of colonizing internal spruce seedling root and stem tissues. The observation that nonpathogenic bacteria colonize internal plant tissues is not new (Wennstrom 1994; Wilson 1995), and Mclnroy and Kloepper (1995) demonstrated that internal plant microhabitats may be colonized quite heavily by plant-associated bacteria. Indigenous, endophytic bacterial population sizes >4 log cfu-g"1 fresh tissue were detected in corn and cotton roots for most of the growing season. However, work is only beginning on the possible practical applications of endophytic PGPR, for which persistence and growth promotion efficacy may be enhanced compared to external root colonists because of a more uniform and protective environment for microorganisms. The best known microorganisms that colonize roots and associated rhizosphere soil with great persistence are those that enter into host plant tissues and establish mutualistic symbioses. In the case of bacteria, examples include the nodule forming ^ -fixers, i.e., Frankia and Rhizobium as well as some endophytic diazotrophs, e.g., Acetobacter and Azospirillium that cause no morphological differentiation of their host plants (Bashan and Levanony 1990; Li and MacRae 1991, 1992; Dong et al. 1994). Other endophytic bacteria including members of Bacillus and Pseudomonas also well known to act as biological control agents either directly, by inhibiting plant pathogenic microorganisms such as Pythium ultimum Trow and Rhizoctonia solani Kiihn (Hagedorn et al. 1989, 1993) or indirectly, by inducing plant systemic resistance to fungal and bacterial pathogens such z&Fusarium oxysporum Schlechtend., Colletotrichum orbiculare (Berk. & Mont.) Arx and Pseudomonas syringae (Wei et al. 1991, 1996). The possible practical advantages associated with endophytic PGPR certainly warrant further investigation. In addition to practical applications for plant growth promoting bacterial endophytes, it is intriguing from an ecological perspective that plants harbour growth-altering, possibly mutualistic, microorganisms within their own tissues. Notwithstanding semantic confusion regarding the terms symbiosis, cooperation and mutualism, the latter is generally used to describe "interspecific 102 interactions that benefit both species by providing some kind of service that its partner cannot provide for itself and receive some kind of reward in return (Bronstein 1994)." However, these are extremely diverse interactions, ranging from obligate, species-specific associations that undoubtedly have a long co-evolutionary history, to loose, less predictable relationships in which mutual benefits may be difficult to detect (endophytic PGPR?). There is little doubt that certain bacterial inocula possess the potential to increase growth of conifer seedlings significantly under field conditions, however, PGPR efficacy in such inoculation trials has been variable (Chanway and Holl 1993b, 1994b). Competition from indigenous soil microorganisms can reduce or eliminate the population of introduced beneficial microorganisms and may explain a significant component of such seedling growth response variability (Kloepper et al. 1989). However, if PGPR strains can proliferate within the plant tissues, colonization and growth promotion may be enhanced for the reasons already described, whether or not such plant-microbe relationships represent advanced stages in the development of a mutualistic association. The objective of the experiment described in this chapter was to evaluate differences if any, in the persistence of root colonization and plant growth promotion between external and internal root colonizing PGPR strains in the field. 103 5.2 MATERIALS AND METHODS 5.2.2 Experiment 6: Growth performance of PGPR-inoculated spruce under greenhouse and field conditions 5.2.2.1 Seedling preparation. Seeds of two interior spruce ecotypes (Mackenzie and Williams Lake, see Table 2.1 for more information) were used to grow seedlings for this experiment. Seeds were stratified as described in Chapter 2 before sowing in plastic containers (RLC-3 Fir Cells, 2.5 cm dia x 12 cm deep, 45 cm3, Stuewe & Sons, Inc., Covallis, OR) filled with autoclaved Sunshine mix (see section 2.2.3.2 for the detailed chemical analysis). Four seeds were planted in each container. PGPR inoculum was prepared with 50% strength TSB for the Bacillus strains or full-strength KBB for Pseudomonas as described in Chapter 2. Resulting inoculum densities were (cfu-mL"1): L6-16R=2.0 x 105; Pw2R=3.3 x 105; S20R=1.1 x 105; Sm3RN=1.8 x 107; Ss2RN=2.2 x 107; and Sw5RN=4.2 x 106. Containers were inoculated with 1.0 mL of one of the strains. Control containers received 1.0 mL of SPB. A second inoculation was performed one week later following the same protocol. Cell densities for the second inoculation were (cftrmL"1): L6-16R=2.9 x 106; Pw2R=4.9 x 107; S20R=9.8 x 106; Sm3RN=6.6 x 108; Ss2RN=9.9 x 108; and Sw5RN=8.5 x 108. Each inoculation treatment had 180 containers (90 seedlings per spruce ecotype) which were contained in 30 cm x 60 cm racks (RL200 Tray, Stuewe & Sons, Inc., OR). Racks were placed on the greenhouse bench ca. 30 cm apart and were rotated every week. Seedlings were thinned to the single largest germinant per container after the second inoculation. Seedlings were watered every 3 days and fertilized every two weeks with a solution containing the following nutrients (lag-seedling"1): N=1000, P=250, K=500, Ca=500, Mg=25, S=48, Fe=2, Mn=0.25, Cu=0.1, Zn=0.1, B=0.1, Mo=0.025 (Van den Driessche 1989). Other conditions for seedling growth were similar to those described in Chapter 2. Fifteen weeks after sowing (and two weeks before the outplanting), external and internal root colonization were assessed on 8 randomly selected seedlings per bacterial treatment (4 seedlings 104 per spruce ecotype). Procedures for inoculum recovery were similar to those described in experiment 5 (Chapter 4) with the exceptions that internal stem colonization was not examined, and only the primary roots including the apical meristem were sampled. 5.2.2.2 Evaluation of root colonization and seedling performance of PGPR-inoculated spruce grown in the field. PGPR-inoculated seedlings were four-months-old at the time of outplanting. Two field sites located near Smithers, B.C. (Blunt Creek site and Shoe-house site), and two near Williams Lake, B.C. (Regular cut-block site and Landing site) were used for this experiment. Site descriptions including soil physical and chemical properties are presented in Tables 5.1 and 5.2. Seedlings were matched to the region of their origin. Seedlings were not subjected to dormancy inducing treatments before planting because they were to be harvested before fall frosts would normally occur in these regions. Before outplanting at each site during the first week of June 1995, 74 seedlings of uniform height were selected from each PGPR treatment. Twenty of these were used for determining preplanting seedling biomass and the remainders (54) were used for planting (27 at each of two sites near either Smithers or Williams Lake). At each site, existing vegetation was cleared and a randomized complete block design (n=27) was arranged with seven inoculation treatments (Fig. 5.1). After gently removing root plugs from plastic containers, seedlings were planted at 1 m spacing with care to reduce the possibility of cross-contamination between PGPR treatments, e.g., treatments were established one at time at each site, and planting tools were disinfected before starting a new treatment. Four months after outplanting (late September or early October 1995), seedlings were excavated, taking care to include the whole root system. Seedlings were placed in individual plastic bags for cold storage (4°C) and were processed in the laboratory within two weeks of harvesting. External and internal root colonization by each strain was assessed on four seedlings in each treatment using the dilution plate assay described in Chapter 4. All seedlings in each 105 Table 5.1 Site description of the four field sites. Site Property Smithers Blunt Creek Smithers Shoe-house Williams Lake Regular cut-block Williams Lake Landing Latitude 55°10' 55°55' 52°37' 52°37' Longitude 127°02' 127°34' 121°45' 121°45' Elevation (m) 1200 730 950 950 Slope (%) 5-10 0-5 0-5 0-5 Landform glacial till fluvial veneer over till colluvial veneer colluvial veneer over till over till BGCZ unit ESSF SBS SBS SBS Soil classification (Canada) Humo-Ferric Podzol Humo-Ferric Podzol Gray luvisol Gray luvisol Soil texture NA 1 Loam Clay loam Loam Coarse fragments (>2mm) NA 57% 61% 48% Humus form Mor NA Mor NA Forest floor thickness (cm) 4.0 (1.5-9.0) NA 3.6(1.0- 9.5) NA Disturbance type scalp, wheel ruts landing scalp, wheel ruts landing Not available 106 Table 5.2 Chemical properties of top soils collected from the four field sites. Site Property Smithers Blunt Creek Smithers Shoe-house Williams Lake Regular cut-block Williams Lake Landing pH 1 (H20) 5.1 6.3 5.3 6.3 Electrical conductivity 0.44 0.48 0.42 0.48 (dS-m"1) Total carbon (%) 32.2 8.9 12.0 2.0 Total nitrogen (%) 0.79 0.12 0.24 0.06 Phosphorus (pg-g *) 57 52 71 52 Potassium (pg-g *) 380 190 290 190 Calcium (pg-g1) 6100 2900 2300 1500 Magnesium (pg-g"1) 980 390 480 290 Copper (pg-g"1) 0.7 2.3 3.3 3.8 Zinc (pg-g"1) 5.9 2.9 3.2 2.5 Iron (pg-g"1) 35 205 155 220 Manganese (pg-g 120 225 190 115 Boron (pg-g"1) 1.5 0.6 0.7 0.3 Sulfate-sulfur (pg-g ) 15.1 3.2 4.0 6.5 Oven dry soil: deionized FL:0=1 : 2. 107 Seedling arrangement in field trials at (a) the Smithers Shoe-house site and (b) the Williams Lake Regular site (b). Four-month-old seedlings were planted 1 m apart from each other according to a randomized complete block design (n = 27). 108 treatment were visually assessed for shoot injury using a four-rank scale based on the percentage of dead needles. Category "0" seedlings had 90 - 100% needle mortality; category "1" had 50 -90% needle mortality; category "2" had 10 - 50% needle mortality and category "3" had less than 10% needle mortality (Fig. 5.2). Root systems were then washed free of soil and separated from shoots. Root and shoot dry weights were assessed after drying at 70°C for 4 days. 5.2.2.3 Statistical analyses of seedling performance and root colonization by PGPR strains. Shoot injury ranks were analyzed using the nonparametric Kruskal-Wallis test. Shoot and root growth responses to PGPR treatment in the greenhouse were assessed using ANOVA and Fisher's protected LSD for each spruce provenance. Relative growth rates (RGR) of seedlings in the field were calculated according to the relationship: RGR = W'l'(&WI6T) = d(ln W)ldT, where Wis the seedling biomass and Tis the time interval between harvests (Hunt 1978). Biomass of dead seedlings was treated as unchanged. Data from all sites were not pooled because of differences in environmental conditions among the planting sites. To assess the site effect on seedling performance and root colonization, Pearson correlation coefficients were calculated for each variable using arithmetic means. For shoot injury data, Spearman's rank-order correlation coefficient was used because data possessed a non-normal distribution. Percentage data for seedling survival and root colonization data were arcsine and log transformed, respectively, before statistical analyses were performed. 109 Fig. 5.2 Seedling shoot injury was rated visually using a four-rank scale based on the percentage of dead needles. Category "0" seedlings had 90 - 100% needle mortality; category "1" had 50 - 90% needle mortality; category "2" had 10 - 50% needle mortality and category "3" had less than 10% needle mortality. 110 5.3 RESULTS1 5.3.1 Plant growth responses to PGPR inoculation under greenhouse and field conditions. Inoculation with all strains except Sm3RN significantly enhanced seedling growth (either shoot or root) of both Mackenzie and Williams Lake ecotypes at the end of the greenhouse phase of the experiment (Fig. 5.3). Strain L6-16R was most effective in both shoot and root growth promotion for both ecotypes, causing biomass increases ranging from 29% to 35% (Appendix 6, Table A6.1). However, it should be noted that to avoid cross-contamination of inoculum strains, seedlings were not randomly positioned on the greenhouse bench, but pooled within racks according to inoculation treatments. Racks were rotated weekly to reduce positional effects. Therefore, strictly speaking, the ANOVA assumption that treatments were randomly applied was violated, but this likely did not affect the experimental outcome. Relative growth rates of seedlings at each of the four field sites are presented in Fig. 5.4. In general, shoot growth was more responsive to bacterial treatment than root growth. Strain Ss2RN significantly increased either shoot or root growth rates at all sites. In contrast, seedling RGR was not enhanced by S20R at any of the field sites. The effect of other strains varied by site so that strains Pw2R and Sm3RN, for example, were ineffective at the Smithers Blunt Creek site but were effective RGR promoters at the other three sites. Interestingly, Blunt Creek is the only site in the Engelmann spruce and subalpine fir (ESSF) biogeoclimatic (BGC) zone; the other sites are in the sub-boreal spruce (SBS) BGC zone. Seedlings planted at landing sites (Smithers Shoe-house and Williams Lake Landing) were more likely to benefit from PGPR inoculation than those at less compacted, prepared sites, i.e., the soil surface was mechanically scarified to remove slash and litter at the Smithers Blunt Creek site and Williams Lake Regular cut-block site (Fig. 5.4). Shoot injury ranks showed large variations among inoculation treatments (Figure 5.5). Seedlings inoculated with strains L6-16R, Pw2R, Ss2RN and Sw5RN had less shoot injury than 1 All statistical tables for Chapter 5 are presented in Appendix 6. I l l 180 Mackenzie Williams Lake Seed provenance .5.3 (a) Shoot and (b) root biomass of spruce seedlings (Mackenzie and Williams Lake ecotypes) 16 weeks after inoculation with PGPR strains in the greenhouse. Error bars indicate the standard errors of the means. ***, **, and * indicate the mean is significantly different from uninoculated control at P <0.01, PO.05, P<0.1, respectively (Fisher's protected LSD). 3.0 2.0 1.0 0.0 -1.0 (b) Root Smithers Blunt creek Smithers Shoe-house Williams Lake Reg. cut-block Williams Lake Landing L6-16R Pw2R S20R Sm3RN _ S s 2 R N 1^ Sw5RN I 1 Control Field site Fig. 5.4 Relative growth rates of (a) shoot and (b) root biomass of seedlings outplanted at 4 field sites. Error bars indicate the standard errors of the means. ***, **, and * indicate the mean is significantly different from uninoculated control at P<0.01, P<0.05, P<0.1, respectively (Fisher's protected LSD). Pi » 2 If o o C/3 Kruskal-Wallis test P = 0.065 Kruskal-Wallis test P = 0.073 Kruskal-Wallis test P = 0.214 Kruskal-Wallis test P = 0.047 Non-endophyte Endophyte I I Uninoculated > <^  Smimers Blunt creek Smithers Shoe-house Williams Lake Reg. cut-block Williams Lake Landing Fig. 5.5 Shoot injury rank of each PGPR treatment at each field site. Error bars indicate the standard errors of the means. Injury ranks were: dead = 0, heavily injured (more than 50% of dead needles) = 1, partly injured (less than 50% of dead needles) = 2, and no detectable injury = 3. uninoculated controls at every site. Shoot injury of strain S20R-treated seedlings did not differ from controls, and strain Sm3RN had variable effects among the four sites. 5.3.2 Recovery of PGPR strains from root systems. Four months after inoculation (i.e., at the end of the greenhouse growth period), Bacillus and Pseudomonas strains were detected with > 4.2 and > 5.6 log cfu-g"1 rhizosphere soil, respectively (Fig. 5.6-5.9 and Appendix 6, Table A6.7). In general, however, external root colonization levels declined during growth in the field. Population levels of Bacillus strains occasionally dropped below assay detection limits (104), which resulted in low mean values and large standard errors (Fig. 5.6a, 5.7a, 5.8a and 5.9a). Of the six strains evaluated, Pseudomonas strain Sw5RN was recovered with the highest external root populations at every site, while Bacillus strain S20R had the lowest population sizes at all but the Williams Lake Regular cut-block site. Internal root colonization by Pseudomonas strain Sm3RN was detected in seedlings sampled at each of the four sites. Bacillus strain Pw2R was detected inside seedlings from three of the four sites. Pseudomonas strains Ss2RN and Sw5RN were occasionally detected in homogenates from surface sterilized roots. 5.3.3 Relationships between seedling performance and recovery of PGPR strains The correlation matrix between seedling survival, growth and bacterial root colonization using the mean values of inoculation treatments from each site indicated that seedling survival was correlated with seedling injury, but not with seedling growth rate (Table 5.3). However, seedling injury was correlated with the seedling growth rates, both in the greenhouse and the field, but not with the PGPR population sizes. Only weak correlations were detected in seedling growth rates in between the greenhouse and the field. 115 (a) External root tissue (b) Internal root tissue ON O &, O N O OO o L6-16R Pw2R S20R V - V Sm3RN Ss2RN O - ** Sw5RN May 31 Sept. 25 May 31 Sept. 25 Fig. 5.6 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Smithers Blunt Creek site for 4 months. (a) Eexternal root tissue (b) Internal root tissue 1 5 ~ 4 o O o J 3 L6-16R Pw2R S20R V - V Sm3RN Ss2RN Sw5RN June 1 Sept. 26 June 1 Sept. 26 Fig. 5.7 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Smithers Shoe-house site for 4 months. (a) External root tissue (b) Internal root tissue oo o 2 5 <u A 0. to O H A 'oo <H -o 3 00 o i s > < 0"° L6-16R 0 - 0 Pw2R S20R V - V Sm3RN Ss2RN 0-0 Sw5RN June 8 Oct. 1 June 8 Oct. 1 Fig. 5.8 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Williams Lake Regular cut-block site for 4 months. (a) External root tissue (b) Internal root tissue o co a 5 a, CO o N A 2 4 'an o 3 oD o L6-16R Pw2R S20R V - V Sm3RN ^""^ Ss2RN Sw5RN June 7 Oct. 1 June 7 Oct. 1 Fig. 5.9 PGPR recovery from (a) external and (b) internal root tissues of 8-month-old spruce seedlings grown at the Williams Lake Landing site for 4 months. Table 5. 3 Correlation coefficients (r) for seedling survival, biomass and bacterial colonization of roots. Seedling Shoot Shoot Root Shoot Root Rhizosphere Rhizosphere Survival Injury Growth Rate Growth Rate Growth Rate Growth Rate Colonization Colonization Percent Rank (Greenhouse) (Greenhouse) (Field) (Field) (Greenhouse) (Field) (%SS)2 (SIR)3 (SGG) (RGG) (SGF) (RGF) (RCG)4 (RCF)4 n = 28 n = 28 n = 28 n = 28 n = 28 n = 28 n = 24 n = 24 (%SS)2 1.000 (SIR)3 0.599 *** 1.000 (SGG) 0.254 0.612 *** 1.000 (RGG) 0.223 0.637 *** 0.875 *** 1.000 (SGF) 0.252 0.651 *** 0.393 ** 0.456 ** 1.000 (RGF) -0.003 0.633 *** 0.317 * 0.403 ** 0.587 ** 1.000 (RCG)4 0.218 0.121 -0.338 -0.400 * 0.126 0.297 1.000 (RCF)4 0.239 0.191 -0.203 -0.202 -0.029 -0.071 0.319 1.000 Spearman correlation coefficient for SIR; Pearson correlation coefficient for other variables. 2 1/2 (%SS) = arcsine [% of (survived seedlings)/(planted seedlings)] 3 (SIR) = Shoot Injury Rank: dead = 0; heavily injured = 1; partly injured = 2; not injured = 3. 4 (RCG) and (RCF): data were log transformed before the correlation calculation. ***, ** and * indicate rejection of null hypothesis at P < 0.001, P <0.05 and P < 0.1, respectively. 5.5 DISCUSSION Spruce seedling growth was sigriificantly enhanced by all PGPR strains in the greenhouse, except Sm3RN. Although the growth rates observed in the greenhouse were reduced after outplanting, growth enhancement by PGPR inoculation was detected after outplanting except for seedlings treated with strain S20R. Growth promotion in the field was reflected primarily in the relative shoot growth rates, and less so in relative root growth rates. This may be due to the combination of strong light intensity available in the field in addition to technical difficulties associated with harvesting root samples, i.e., incomplete root system recovery from the sites. If one assumes that the root plug was a significant source of supplementary nutrients and possibly moisture (e.g., Burdett et al. 1984; Nilsson and Grander 1995), then the increased light intensity associated with field sites may have favoured shoot growth over root growth during the field phase of this experiment, increasing the likelihood that growth promotion would be expressed primarily in above ground performance. It is interesting that PGPR were more effective at both landing sites (Smithers Shoe-house and Williams Lake Landing) compared to the regularly prepared (scarified) cut-block sites (Smithers Blunt Creek and Williams Lake Regular cut-block). Surface organic matter may impede the growth of planted seedlings by reducing the available soil-root contact area in duff layers for nutrients and water (McMinn and Hedin 1990), or by increasing the incidence of low soil temperature. Dobbs and McMinn (1977) indicated that, because of its high thermal diffusivity, mineral soil facilitates larger heat re-radiation during the night and reduces the incidence of frost damage. Though it is difficult to pinpoint exact environmental factors that differentiated the seedling growth response to PGPR inoculation between these "landing" and "regular" sites, PGPR inoculation may be more useful in soils that have been degraded physically, such as landing sites compared to relatively undisturbed environments. In the only other similar studies performed to date, Chanway and Holl (1993b, 1994b) also observed greater PGPR efficacy on interior spruce and lodgepole pine at sites of lower productivity and hypothesized that such inocula may be most useful at sites with harsh seedling growth conditions. 121 Another objective of this experiment was to determine if root colonization was related to growth and survival of spruce seedlings in the field. By the end of the greenhouse growth period, external root population sizes of Bacillus and Pseudomonas strains were ca. 104 and 106 cfu-g"1 soil, respectively. Since the inoculum densities of the Bacillus and Pseudomonas strains were ca. 105 and 107 cfu-plant-cone"1, respectively, approximately 1/10 of the initial inoculum density remained after the greenhouse phase of this experiment. Similar root zone declines in PGPR population have been observed in studies on agricultural crop plants, particularly under water stress condition (Kloepper etal. 1980b; Bashan etal. 1991; Juhnke etal. 1987, 1989; Reddy and Rahe 1989; Kumar and Dube 1992), as well as for strain L6-16R in the rhizosphere of lodgepole pine (Chanway and Holl 1992). PGPR population sizes associated with spruce roots decreased a further 10-100 fold during the 4-month period of field growth, suggesting that field conditions were severe, even in the rhizosphere of host seedlings. However, there appear to be differences in population dynamics among these strains: Pseudomonas strains Sw5RN and Sm3RN maintained their population sizes better than the others, while populations of Pseudomonas strain Ss2RN declined drastically at most of the sites. The reason for the rapid decline of the Ss2RN population is not clear, but it is possible that this strain was more susceptible to factors absent in the greenhouse than the other two pseudomonads, such as low water potentials and large temperature fluctuations. Since strain Ss2 was originally isolated from the rhizoplane of a spruce root (see Chapter 2 and Appendix 1), it may be adapted to a narrower range of growth conditions than the other two Pseudomonas strains, which were rhizosphere isolates. On the other hand, the population sizes of Bacillus strains recovered from the rhizosphere varied with sites. Bacillus are well-known to sporulate under adverse environmental conditions (Roszak and Colwell 1987), and the rapid decline in vegetative cell population soon after soil inoculation has been documented (Siala et al. 1974; van Elas et al. 1986; Roszak and Colwell 1987). In general, these colonization data clearly support the view by Kloepper (1993) that "one cannot assume from a report on a single strain of any genus or species that all strains of the taxon are capable of colonizing roots." 122 Of the strains I evaluated, the two putative endophytic strains showed greater persistence of seedling growth promotion in the field relative to the greenhouse than other strains at the Williams Lake landing site. However, this was not observed at the other sites. Therefore, the hypothesis that "endophytic PGPR strains are protected from adverse environmental conditions in the host tissues and improve the persistence of plant growth promotion efficacy" was not supported at least for the short term after the outplanting. However, population sizes of Pw2R and Sm3RN recovered from the internal root tissues were almost always smaller than the corresponding rhizosphere population, which suggest that, notwithstanding their endophytic capabilities, these two strains are still primarily external root colonists. Though one can estimate the size of bacterial populations which are potentially viable in natural environmental samples, it is very difficult technically to distinguish between active and dormant bacterial cells. In addition, Mclnroy et al. (1996) recently reported the possibility of "antibiotic masking" — the temporary loss of the antibiotic-resistant phenotype associated with rifampicin-resistant derivatives of endophytic bacteria. Preliminary tests with Pw2R revealed reduced recovery of this strain on antibiotic amended TSA plates, but not complete growth inhibition of the antibiotic-resistant derivative. Moreover, the recovery of the bacteria from the rhizosphere soil on antibiotic-amended media was also reduced. Therefore, even though internal root tissue population sizes of Pw2R may have been underestimated, such an error apparently occurred for rhizosphere samples as well. Whether or not the recovered populations of Pw2R and Sm3RN from internal root tissues were metabolically active cannot be ascertained, but the ability of endophytic strains to diversify their habitat (i.e., living on and in plant tissues) in contrast to non-endophytes under field conditions could provide a significant advantage in terms of survival and persistence. A significant correlation was detected between seedling survival (%) and shoot injury (rank), but not between seedling survival and seedling growth rates either in the greenhouse or the field (Table 5.3). This is not surprising because poor survival may not be expressed in relation to growth for some time after the first year in the field (Burdett et al. 1984; Vyse 1981). Since there 123 are significant correlations between shoot injury and seedling growth rates for both the greenhouse and field, seedling growth soon after outplanting is still an important factor for seedling survival in later years. Some studies have indicated the importance of outplanted seedling size in relation to soil moisture stress at planting sites. For example, Hines and Long (1986) reported that smaller Engelmann spruce seedlings, based on diameter of the root collar, exhibited significantly higher drought stress 4 weeks after outplanting as well as lower survival rates during the first and second growing seasons. Van den Driessche (1992) also observed significant correlations between shoot dry weight of seedlings in the nursery before planting and seedling growth at xeric (v|/ = -1.9 Mpa) sites (r=0.71, 0.73 and 0.89 for Douglas fir, lodgepole pine and white spruce, respectively). The results of my experiment agree with these previous studies in terms of the importance of seedling size and subsequent performance in the field. However the unique aspect of my study is that seedling sizes were increased by PGPR. In conclusion, PGPR-inoculated spruce seedlings frequently outperformed uninoculated controls under field conditions. Though endophytic PGPR inoculation did not necessarily lead to better persistence of enhanced seeding growth rates in the field, their survival in internal root tissues was confirmed after the first season of outplanting. Since the population sizes of most of PGPR in the rhizosphere generally declined in the greenhouse and the field, it seems that once induced in the greenhouse, seedling growth promotion by PGPR persists after outplanting, and may subsequently enhance spruce seedling survival under field conditions. However, additional field experimentation for longer periods with such "pre-inoculated" seedlings is required to determine how long growth promotion persists after outplanting. 124 CHAPTER 6 MECHANISMS OF PLANT GROWTH PROMOTION BY SPRUCE PGPR 6.1 INTRODUCTION AND LITERATURE REVIEW In the previous chapters, neither coexistence specificity, interactions with mycorrhizal fiingi or root microsite colonization were found to explain the variable effects of PGPR on plant growth. However, it appears that once PGPR induce stimulation of host plant growth, the effect persists even with substantial changes in environmental conditions, i.e., from greenhouse to forest. Therefore, the key to determining the cause of these variable seedling growth responses may he in the mechanisms by which these PGPR stimulate plant growth. Unlike symbiotic rhizobia, mechanisms of plant growth promotion by PGPR vary greatly, and have been broadly categorized into two groups: direct and indirect (Kloepper et al. 1989). While mechanisms are difficult to demonstrate conclusively, there appear to be at least two major types associated with each category. Direct plant growth promotion may involve: (i) production of plant growth regulating substances or phytohormones1, or (ii) increased availability of plant growth-limiting nutrients. Indirect plant growth promotion may involve (i) suppression of deleterious microorganisms including plant pathogens, as well as (ii) enhancement of mutualism between host plants and symbiotic microorganisms such as mycorrhizal fiingi or root-nodule bacteria (Kloepper et al. 1989). However, many PGPR are known to possess more than a single plant-growth promoting attribute, rendering the task of elucidating growth promotion mechanisms difficult. By definition, a phytohormone is an organic compound synthesized in one part of a plant and translocated to another part, where, in very low concentrations it causes a physiological response (Salisbury and Ross 1992). However, I use the term "phytohormone" in this thesis to describe identical chemical compounds that microorganisms synthesize and which affect plant growth in the same way plant produced phytohormones do. 125 I. Direct Plant Growth Promotion (i) Phytohormone production. Bacterial production of phytohormones has been implicated as the mechanism by which many PGPR, including strains of Azospirillum (Tien et al. 1979), Azotobacter (Brown 1972), Bacillus (Holl et al. 1988), and Pseudomonas (Burr and Caesar 1984; Dubeikovsky et al. 1993) stimulate plant growth. Auxins, or various derivatives of indole-3-acetic acid (IAA) have most commonly been suggested to be the causative plant-growth altering agent, as these phytohormones are well-known to affect plant cell elongation (Frankenberger and Arshad 1995). However, rhizobacteria are known to produce several other types of plant growth altering substances including gibberellins (by Azospirillum and Azotobacter), cytokinins ( Agrobacterium and Azotobacter), ethylene (Bacillus, Klebsiella, Pseudomonas, and Serratia), and abscisic acid (Azospirillum) (Frankenberger and Arshad 1995), and their possible role in PGPR mechanisms must be considered. Most phytohormonal effects are concentration-dependent, and such compounds often inhibit plant growth when present in high quantities e.g., the herbicide dichlorophenoxyacetic acid (2,4-D). Therefore, another possible PGPR mechanism is mediation of phytohormone production so that plant growth inhibiting concentrations are not reached in the vicinity of the plant (Glick 1995). For example, Glick et al. (1994a, 1994b) demonstrated that Pseudomonasputida GR12-2 hydrolyzes 1-aminocyclopropane-l-carboxylate (ACC), the precursor molecule of ethylene in plants, thereby lowering the level of endogenous ethylene production by the plant and reducing associated inhibitory effects on root elongation. However, it is generally agreed that both plant growth responses to PGPR as well as bacterial production of phytohormones are influenced by the environment. Moreover, to elucidate the mechanism of growth promotion, it is important to distinguish the producer of plant growth substances, i.e., bacteria versus plants, because both organisms can synthesize such compounds -7 (Gaudin et al. 1994). Since phytohormones are active in extremely minute quantities (e.g., 10" -126 10" M of IAA can stimulate root elongation), conclusive evidence to support this type of mechanism for PGPR awaits the development of techniques for monitoring small amounts of plant growth substances and distinguishing them from those produced by host plants. (ii) Facilitation of nutrient uptake. Aside from indirect effects through phytohormone — induced root elongation and/or branching, plant nutrient availability may be directly increased by rhizobacteria that produce mineral solubilizing metabolites such as phosphatase enzymes, organic acids or metal chelating siderophores. Although it has been clearly demonstrated that various rhizobacteria can dissolve poorly soluble phosphate compounds in vitro by producing organic acids, there is little direct evidence indicating that such effects actually occur in soil (Tinker 1984). Therefore, increased root phosphate due to bacterial solubilization processes in soil may be quite modest in comparison with the stimulatory effects of bacteria on uptake of nutrients by roots themselves. This view is supported by the findings that some organic acid exudates such as citric acid (Gardner et al. 1983; Gardner and Boundy 1983) and piscidic acid (Ae et al. 1990) from the roots themselves facilitate the release of phosphorus from Fe-P compounds. Iron is another mineral with restricted availability to plants due to its poor solubility. For example, at low pH the ferric ion (Fe3+) is only marginally available in phosphorus compounds such as Fe(OH)2H2P04 (Bonn et al. 1985). Siderophores are low-molecular-weight iron-chelating agents produced by various microorganisms that can significantly increase the availability of iron in soil to plants (Powell etal. 1980). Fluorescent pseudomonads are often abundant in the rhizosphere and can produce the siderophores pseudobactin and pyoverdine, characterized by yellow-green pigmentation, under Fe-limited conditions (Bossier et al. 1988). However, there is little evidence that these bacterial siderophores serve as a direct source of iron for plants. For example, Kloepper et al. (1980a) found no plant growth promoting effect after amending soil with ferric pseudobactin, while Becker et al. (1985) observed an inhibitory effect on plant growth by pseudobactin. Therefore, the significance of bacterial siderophores in PGPR mechanisms is likely more related to biocontrol activity (see below) than Fe acquisition. 127 Growth promotion almost inevitably is reflected by enhanced root growth and/or development as well as by healthier plants compared to uninoculated controls, rendering mechanisms related to increased nutrient uptake difficult to distinguish from other possible causes of plant growth promotion. In contrast, PGPR mechanisms related to enhanced nitrogen supply through N2 fixation are more easily discernible. Members of various rhizobacteria including Azotobacter, Azospirillum, Bacillus, Beijerinckia, Clostridium, Enterobacter, Klebsiella and Pseudomonas are capable of fixing atmospheric N2 (Richards, 1987). However, most of these free living diazotrophs fix only limited amounts of nitrogen that rarely meet their own needs as well as those of their "host" plants (Glick 1995). Results of inoculation trials using ^ -fixing PGPR with 1 5 N dilution techniques also suggest that atmospheric ^ -fixation is of secondary importance in plant growth promotion (Chanway and Holl 1991), and few studies have described N2-fixation as a major mechanism of PGPR in recent years (e.g., Dong et al. 1994). II. Indirect Plant Growth Promotion (i) Suppression of deleterious microorganisms. Because the rhizosphere provides the front-line defense for roots against attack by pathogens, microorganisms that can grow in the rhizosphere and suppress deleterious microorganisms are ideal for use as biocontrol agents. Bacterial biocontrol agents can improve plant growth by suppressing either major or minor pathogens (Kloepper 1993). Major pathogens produce recognizable root or vascular diseases with obvious symptoms, while minor ones are parasites or saprophytes that damage mainly juvenile tissue such as root hairs and tips and cortical cells without obvious symptoms of injury (Weller 1988). Within the category of minor pathogens, Schippers et al. (1987) distinguished the parasitizing minor pathogens from non-parasitizing deleterious rhizosphere microorganisms (DRMO). DRMO include deleterious rhizobacteria (DRB) (Suslow and Schroth 1982) and deleterious fungi. The causes of deleterious effects on plant growth have not yet been fully elucidated (Williamson 1990), but microbial production of 128 volatile metabolites such as hydrogen cyanide (HCN) may be involved (Alstom and Burns 1989). Inhibition of DRMO and pathogens by PGPR may also occur through niche exclusion resulting from competition for root colonization sites, production of non-volatile compounds that are toxic to DRMO and pathogens, e.g., antibiotics and siderophores, as well as induced systemic resistance (Kloepper 1993). Considering the diversity of rhizosphere microflora, it is likely that the full spectrum of potentially effective bacterial strains with biocontrol activity has barely been explored (Weller 1988). Currently, Pseudomonas and Bacillus are two major genera receiving attention as biocontrol agents. Pseudomonas strain 2-79 is a well-characterized example of an antibiotic producing PGPR. This strain inhibits Gaeumannomyces graminis (Sacc.) von Arx and Olivier var. tritici Walker, the pathogen that causes take-all disease of wheat (Weller et al. 1988), by producing a phenazine antibiotic at a concentration <1 (j.g-mL"1 (Brisbane and Rovira 1988). Tn5 mutants that lacked the ability to produce this antibiotic were ineffective against G. graminis (Thomashow and Weller 1988). Hebbar et al. (1992) also found that pseudomonads comprised 88% of the Gram-negative bacteria isolated from the maize rhizosphere that were inhibitory to Fusarium moniliforme. However, their in vitro assay results suggested that antifungal compounds distinct from siderophores were responsible for the antibiotic effects. Bacteria belonging to the genus Bacillus are also known for antibiotic production and biocontrol of plant pathogens. A well-known example is Bacillus subtilis strain A13, which was originally isolated from the mycelium of Sclerotium rolfsii (Broadbent et al. 1977). This strain inhibits in vitro growth of several soil borne plant pathogens such as Fusarium oxysporum (Yuen et al. 1985) and Rhizoctonia solani (Merriman et al. 1974), and has improved the growth of many plant species. Since 1983, Bacillus subtilis A13 has been sold as a treatment for peanut (Arachis hypogaea L.) under the name QUANTUM-4000 (Gustafson, Inc., Dallas, TX). Ohno et al. (1992) also found that B. subtilis strain RB14 produced the antifungal-antibiotic iturin, which is an effective growth inhibitor of various phytopathogens. Interestingly, strain RB14 also 129 produces a lipopeptide biosurfactant that improves the antibiotic efficiency on the surface of the target fungus. Some PGPR strains produce extracellular siderophores that bind ferric iron, thereby effectively sequestering it from pathogens and controlling soil borne diseases (Kloepper et al. 1980a, 1980b). Because bacterial membrane-receptor proteins specifically recognize and take up the siderophore-Fe complex, Fe availability for other rhizosphere microorganisms including DRMO is lowered (Hemming 1986). Many studies indicate that bacterial siderophore production is one way biocontrol PGPR elicit their effects (Kloepper 1993). However, it should be noted that even if a PGPR strain produces a siderophore, it may not be the primary mode of action, as previously indicated in the case of the Pseudomonas strain 2-79. In addition to DRMO suppression by antibiotic and siderophore production, simple niche exclusion through competition for nutrients on the root surface may also be the primary mechanism by which certain biocontrol PGPR act. For example, Stephens et al. (1993) demonstrated that pseudomonads, which were capable of producing siderophores and antibiotics, inhibited growth of Pythium ultimum. However, inhibition of fungal growth was found to result from the ability of bacteria to rapidly metabolize seed exudates, and thus effectively sequestering these nutrients from the pathogens. Furthermore, signal transduction mechanisms for activating plant defenses against microbial infection have also been investigated in the context of induced systemic resistance (ISR) by PGPR in recent years (Wei et al. 1991, 1996; Liu et al. 1995; Hoffland et al. 1996). Hoffland et al. (1996) demonstrated that Pseudomonas fluorescens-mediated ISR resulted in broad spectrum systemic resistance against pathogens. However, neither necrosis nor pathogenesis-related proteins were detected, which prompted the conclusion that this response was different from the classic hypersensitive-type reaction. In addition, since ISR against Rhizoctonia soloni was not detected in radish (Raphanus sativus L.) after treatment with P. fluorescens, they suspected that induction of systemic resistance was only possible when a basic level of genetic resistance was already present, which radish was postulated to lack. 130 (ii) Enhancing symbioses between plant and other microorganisms. The rhizosphere of a single plant may accommodate more than hundreds of millions of microorganisms in an environment which undoubtedly fosters complex interactions. Another mechanism by which PGPR may indirectly exert a positive effect on plant performance is through positive interactions with mutually symbiotic fungi and bacteria, resulting in enhanced root nodule and mycorrhizae formation or functioning. There are several examples of soil bacteria including strains of Pseudomonas and Bacillus that enhance root nodule size and number (Grimes and Mount 1984; Li and Alexander 1988; Yahalom et al. 1990). A recent example was provided by Srinivasan et al. (1996) who inoculated bean (Phaseolus vulgaris L.) with Bacillus and Rhizobium and observed increased nodulation and root growth likely due to phytohormone production. Li and Alexander (1988), on the other hand, improved alfalfa (Medicago sativa L.) and soybean (Glycine max Merrill) nodulation by co-inoculating antibiotic-resistant rhizobia and antibiotic-producing Pseudomonas and Bacillus strains. Mycorrhiza-stimulating bacteria have also been described in many studies and may prove to be useful, particularly for tree seedlings used in plantation forestry. However, the mycorrhizal status of seedlings is rarely reported in studies of PGPR, prompting Linderman (1988) and Fitter and Garbaye (1994) to suggest that variable plant growth responses to PGPR inoculation may result in part from microbial interactions between inoculant bacteria and mycorrhizal fungi. Although it is unlikely that such a mechanism is central to the growth promotion of spruce seedlings described in Chapter 3 (i.e., PGPR effects were mostly independent of the mycorrhizal status of seedlings) and Chapter 5 (i.e., seedling growth promotion was detected under conditions in which mycorrhizal infection would be rare, e.g., in pasteurized soil), the degree of growth promotion often increased in the presence of other soil biota. Therefore, it was of interest to evaluate interactions of PGPR with other, non-mycorrhizal soil microorganisms as a possible indirect mechanism of plant growth promotion. 131 III. Effect of PGPR on Indigenous Forest Soil Microorganisms It is known that introduced microorganisms may alter the indigenous microflora of soil-based ecosystems (Gilbert et al. 1993, 1996), therefore the effects of PGPR inoculation should not only be studied in relation to plant and mycorrhizal performance, but from the perspective of the entire soil ecosystem. The effects of PGPR inoculation on plant performance as well as PGPR population dynamics have been studied (Kloepper and Beauchamp 1992; Glick 1995), but few studies have investigated the effects of PGPR on indigenous microbial communities, and vice versa. Kloepper and Schroth (1981) found that inoculation of Pseudomonas fluorescens strains on potato reduced populations of Gram-positive bacteria in the rhizosphere. In contrast, Yuen and Schroth (1986) found no effect of P. fluorescens strain E6 on the population densities of bacteria, actinomycetes or fluorescent pseudomonads in the rhizosphere of ornamental plant species including carnation (Dianthus caryophyllus L.), stock (Matthiola incana R. Br.), and sunflower (Helianthus annuus L.) in spite of significant plant growth promotion by the strain. Introduction of Bacillus cereus UW85 changed metabolic characteristics of microbial communities in the soybean rhizosphere, even when the introduced strain itself did not persist in the root zone (Gilbert et al. 1993, 1996). It is unclear whether such changes are benign with respect to the functioning of soil communities, but the possibility that introduced microorganisms antagonize certain groups of indigenous organisms may lead to re-structuring of the local soil microbial community. Microbial ecology has been traditionally studied at the process level, e.g., total numbers of microorganisms, biomass, respiration rates, and enzyme activity (Kennedy and Smith 1995). However, from a functional perspective, nutrient utilization by the soil community may be a useful index for characterization of soil ecosystems. The Biolog™ system (Biolog Inc., Hayward, CA) can measure the ability of pure or mixed microbial cultures to oxidize 95 different carbon substrates. This technique has also been used to differentiate natural microbial communities by analyzing substrate utilization of associated soil extracts (Garland and Mills 1991; Garland 1996b; Zak et al. 1994). Garland and Mills (1991) first introduced the use of community-level carbon 132 source utilization patterns for comparison of microbial communities from different habitats, i.e., freshwater, saltwater, estuaries, rhizosphere and soil. They reasoned that because Biolog™ profiles are based on differences in carbon source utilization between samples, this technique provided a functional basis on which to distinguish natural microbial communities. Zak et al. (1994) supported this concept by reporting distinctive profiles of soil samples from six desert plant communities. Although the method's rapidity and simplicity have made it easier to interpret community profiles in various natural ecosystems, there is still no general agreement on standard procedures. In particular, soil extract inoculum density in relation to colour development in Biolog™ plate wells and the choice of multivariate analyses need to be standardized. Recent studies suggest that soil extract inoculum density used for Biolog™ analyses may be an important factor influencing colour development in wells (Haack et al. 1995). The degree of substrate oxidation was a function of soil inoculum density, but the pattern of substrate oxidation rates in a plate was highly reproducible for a simple microbial community if inoculum density was controlled (Haack et al. 1995). They suggested that the method was most useful when community kinetic profiles on given substrates were used to describe community composition. There is little doubt that kinetic profiles provide a more precise description of microbial community composition, but Garland (1996a) argued that even samples of identical inoculum density vary 10-20% in average well colour development and supported its usefulness for qualitative discernment of microbial communities. In addition, some form of multivariate analysis is required to summarize the large amount of information yielded by each Biolog™ assay. The number of variables is usually reduced using either a correlation or distance matrix derived from well reactions containing the 95 carbon substrates. The correlation approach includes principal component analysis (PCA) (Garland and Mills 1991; Zak et al. 1994; Haack et al. 1995; Garland 1996a; Pfender et al. 1996) and detrended correspondence analysis (DCA) (Garland 1996b), while the distance approach utilizes cluster analysis (Zak et al. 1994; Ellis et al. 1995; Janisiewicz 1996) and multidimensional scaling 133 (MDS). The latter is the method used by the MicroLog program (Biolog, Inc., Hayward, CA), originally designed for Biolog™ plate analysis to identify a unknown microorganism. Although diagrammatic results appear similar between PCA and MDS, their statistical bases are not: the former aims at finding a few new variables that represent other all variables, while the latter focuses on finding the best suite of objects to describe the proximity index calculated from all the variables. In other words, the main focus of PCA is on variables and their linear combinations or correlation matrix, which facilitates the reduction of highly correlated and thus redundant variables. In contrast, MDS looks for relationships between the objects or individuals by using a proximity index, typically the Euclidean distance between variables if the units of variables are comparable. Thus, both PCA and MDS use Euclidean space for their expression, but MDS is based on distances between points, whereas PCA uses the correlations between variables, i.e., angles between vectors (Dillon and Goldstein 1984). Hence, if the goal is to find a few variables that determine the characteristics of the microbial community in question, i.e., carbon substrate sources that differentiate microbial communities, then the correlation approach is appropriate, particularly when differences in carbon utilization patterns between the subject communities are large with low multicollinearity between substrates. If, on the other hand, the goal is to find relative distances between individual communities to interpret their positions from others, then the distance approach is more appropriate. This is particularly useful when the pattern of carbon substrate utilization between the subject communities are similar, because the distance is already a single dimension and thus uninfluenced by a strong first component. In view of the numerous complex mechanisms by which PGPR may stimulate spruce seedling growth as well as the potential interactions between PGPR and other soil microorganisms, the objectives of this chapter were to (i) determine whether or not the mechanism by which spruce PGPR promote seedling growth was dependent on factor(s) in the rhizosphere other than host roots, i.e., was growth promotion direct or indirect, and (ii) evaluate interactions between PGPR and indigenous soil microbial communities originated from forest soils in relation to potential negative effects of introduced microorganisms on native ones. 134 6.2 MATERIALS AND METHODS 6.2.1 Experiment 7: Short-term assessment of plant growth promoting mechanisms 6.2.1.1 Seeds and soils. Interior spruce seeds, supplied by the B.C. Ministry of Forests, Tree Seed Centre (Surrey, B.C.), were collected from a site near Houston, B.C. (54°N, 126°W) in 1990, and stored at -17°C. Germination and moisture content were 96.0% and 7.9%, respectively. Before use in assays, seeds were surface sterilized by immersion in 2.5% NaCIO for 2 min, rinsed several times in sterile water and stratified at 4°C for 30 days. No microorganisms were detected in the seed interior after surface sterilization. "Indigenous" forest soil microbial communities originated from two clear-cut sites near Smithers and Williams Lake as described in Chapter 5. Ten soil samples were randomly collected from each clear-cut, bulked (by site), mixed thoroughly, sieved (2 mm particle size), and stored in sealed plastic bags at 4°C. Soil water content, pH and nutrient status were determined according to standard protocol described by Page et al. (1982). 6.2.1.2 Seedling, medium and bacterial inoculum in a microcosm. Industrial sand (Target Products, Burnaby, B.C.) was washed with tap water and oven-dried at 70°C before 25 mL (37 g) were apportioned to individual glass tubes (25 mm x 150 mm). Tubes were covered with a plastic cap, autoclaved for 60 min and cooled to room temperature before use in experiments. To generate "indigenous" soil microbial communities originated from two different forest types, bulked soil samples from each site (Smithers and Williams Lake) were mixed to reduce the effect of localized microbial colonization within soil samples and suspended separately in 10 volumes of 10 mM SPB containing 5 mL of glass beads (3 mm dia.) by placing on a rotary shaker for 20 min. Each soil suspension was then sieved (500 pm particle size) to remove soil macrofauna and glass beads. Filtered soil extracts were then vigorously stirred as 5.0 mL (=0.5 g 135 dry soil) were removed and added to each tube. In each inoculation treatment, 18 replicate-tubes that received each soil extract were autoclaved again for 60 min. Stratified spruce seeds were sown aseptically on autoclaved sand in a plastic box (28 cm x 16 cm x 10 cm) enclosed in an autoclave bag. Two weeks after sowing, individual seedlings of uniform height were randomly selected and transplanted in the top 2 cm of the sand in glass tubes using a sterile glass rod. Several replicate-tubes per inoculation treatment were left unplanted for each soil type. All tubes were then inoculated with one of the six PGPR strains (L6-16R, Pw2R, S20R, Sm3RN, Ss2RN, and Sw5RN) or with SPB as uninoculated control (n=18 per bacterial strain for tubes with seedlings; n=4 per strain for tubes without seedlings). Inoculum was generated as described in Chapter 2. All tubes were placed in a growth chamber (Conviron, CMP3244, Winnipeg, MB) under a 19 h photoperiod, PAR at floor level of 160 umol-m"2^"1, and a 25:18°C light: dark temperature regime. Three weeks after transplanting and every 4 t h week thereafter, seedlings were fertilized aseptically with 1.0 mL of sterile nutrient solution which contained (ug-g *): N=20.0, P=5.0, K=10.0, Ca=10.0, Mg=5.0, S=9.6, Fe=0.4, Mn=0.05, Cu=0.02, Zn=0.02, B=0.02, and Mo=0.005. Seedlings were harvested destructively 18 weeks after treatments were established. Roots were cut from shoots and the biomass was measured after oven-drying (70°C for 3 days). 6.2.1.3 Evaluation of PGPR population sizes. PGPR colonization was assessed in all four tubes containing treatments without spruce seedlings, as well as on four randomly selected tubes in each treatment with seedlings. This was done by adding 25 mL of SPB to each tube, agitating tubes vigorously, performing serial dilutions of the resultant soil suspensions, and plating onto both 50% strength TSA fox Bacillus and KBA for Pseudomonas. Both types of media were amended with 100 mg-L"1 of the appropriate antibiotics (i.e., rifamycin for Bacillus and nalidixic acid plus rifamycin for Pseudomonas), 100 mg-L"1 of cycloheximide and 50 mg-L"1 Nystatin. Dilution plates were incubated up to four days at 28°C before colony number was assessed. Intrinsic resistance of representative colonies to 136 streptomycin, kanamycin, tetracycline and vancomycin (BBL Sensi-Discs 5-30 pg, Becton Dickinson Co., Cockeysville, MD) was compared with that of inoculum strains to reduce the chance that non-target populations were being monitored. The minimum detectable population size using this assay was 6 x 102 cfu-g"1 soil. 6.2.1.4 Measurement offorest soil microbial communities. Four replicate-tubes containing a spruce seedling and four tubes without seedlings were used for analysis of PGPR survival in soil and to evaluate PGPR effects on forest soil community substrate utilization. Population sizes of viable microorganisms (fungi, actinomycetes, and bacteria) in the microcosms were determined before and after growing spruce seedlings by standard dilution plate counts on 50% strength TSA amended with 100 mg-L-1 of cycloheximide and 50 mg • L" 1 Nystatin for total bacteria and actinomycetes, and potato dextrose agar (PDA) amended with rifamycin and streptomycin (100 mg • L"1) for fungi. Plates were incubated at 28°C for 4 days before colonies were counted. Bacterial colonies observed with filamentous hyphae under the light-microscope were counted as actinomycetes. The number of colonies on a plate was expressed per unit of dry weight of amended forest soil. These values were log transformed and subjected to ANOVA. Although heteroscedasticities were detected in several samples of microbial populations, the ANOVA was assumed to be robust because of the equal number of replicates between the treatments (Zar 1984). Soil suspensions associated with each PGPR x forest soil extract x spruce seedling treatment were pooled and mixed thoroughly. GN-Biolog™ microplates were inoculated with 140 pL per well of the pooled soil suspensions at the initial dilution (1.7 x 10"2 wv 1 ) . Plates were incubated at 28°C and colour development (absorbance) of each well (see Fig. 6.1 for an example) was measured by a microplate reader (Multiskan 310C, Titertek, Mississauga, ON) at 595 nm every 24 h for 96 h. This produced 4 sets of data, one from each 24 h period. 137 Fig. 6.1 Examples of GN Biolog™ plate responses, i.e., carbon substrate utilization patterns of soil microbial communities after 48h incubation. Smithers and Williams Lake site soil extracts (1.7 x 10 dilution) were used. 138 To select a database for multivariate analyses, Shannon's diversity index (FT) for colour development (Zak et al. 1994) was determined for each database. Nominal data for these indices were transformed by taking the difference in absorbance between the control well and the subject well as >0.22=positive, <0.22=negative. The threshold absorbance (0.22) was previously determined by visual estimation and absorbance of uninoculated control plates. Fungal growth was detected in some wells after 72 h incubation. Before applying multivariate analyses, data were standardized by subtracting the absorbance of the control well, followed by dividing by the average absorbance for the 95 wells in each plate to correct for differences in overall activity of inoculum between plates (Garland and Mills 1991). Principal component analysis (PCA) of the soil microbial utilization of 95 carbon substrates revealed high multicollinearity between dependent variables. The first and second components accounted for 82.5% and 4.3%, respectively, of the total variability defined by the carbon source utilization patterns. Therefore, multidimensional scaling (MDS) in 2 dimensions was chosen for this analysis after calculating the Euclidean distances of all 378 well pairs for proximity between the standardized absorbance. Because treatment clusters were observed to depend on soil types and the presence or absence of seedlings, a discriminant analysis of the two dimensional scales was performed using Wilk's Lambda for these factors. 6.2.1.5 Experimental design. Tubes were arranged according to a completely random design for seedling growth and microbial population evaluation. There were 28 treatment combinations for seedling growth comprising 2 "indigenous" soil types (i.e., Smithers and Williams Lake soil) x 2 soil sterility levels (i.e., sterilized and unsterilized) x 7 inoculation treatments (i.e., Bacillus strains L6-16R, Pw2R, and S20R; Pseudomonas strains Sm3RN, Ss2RN, and Sw5RN, and SPB-inoculated controls) with n=18. In addition, there were 7 inoculation treatments for seedling growth in sterile sand medium (n=20). In general, seedling weight data did not violate assumptions of ANOVA, therefore, analyses were conducted on untransformed data. Means were separated used pre-planned contrasts according to Fisher's protected LSD tests (two-tailed). 139 There were 36 treatment combinations for PGPR population assessment generated from 2 soil types x 2 soil sterility levels x 6 inoculation treatments x 2 seedling treatments (i.e., tubes with and without seedlings) with n=4. In addition, there were 7 inoculation treatments for PGPR populations in sterile sand medium (n=4). The number of colonies on a plate was expressed on forest soil dry weight basis and was log transformed before ANOVA. Because entire root systems were sampled for PGPR colonization, samples that yielded no growth on dilution plates were assigned the minimum detectable population size associated with this assay (Kloepper and Beauchamp 1992). All statistical analyses were conducted using Systat (Systat, Inc. Evanston, no. 6.3 RESULTS1 6.3.1 Soil nutrient levels. Nutrient analyses indicated that Smithers soil had higher water (12.5% vs. 8.5% w-w"1), total carbon (8.9% vs. 2.0%) and nitrogen (0.12% vs. 0.06%), and available Ca (2900 vs. 1500 ug-g"1), Mg (390 vs. 290 ug-g"1), Zn (2.9 vs. 2.5 ug-g"1), Mn (225 vs. 115 ug-g"1) and B (0.6 vs. 0.3 ug-g *) contents, but lower Fe (205 vs. 220 ug-g *) and SO4-S (3.2 vs. 6.5 ug-g *) contents than Williams Lake soil. There were no differences in pH (6.3), available P (52 ug-g *) and available K (190 ug-g ^ levels. 6.3.2 Effects of PGPR inoculation on soil microbial population size. Before transplanting spruce seedlings into the microcosms, population sizes of viable microorganisms of the Smithers and Williams Lake soil treatments were (mean ± S.E. log cfu-g"1 soil): bacteria=6.89 ± 0.11 and 7.12 ± 0.28; actinomycetes=5.65 ± 0.58 and 5.41 + 0.32; fungi-6.89 + 0.11 and 7.12 ± 0.28, respectively. All statistical tables relevant to Chapter 6 are presented in Appendix 7. 140 T h e p o p u l a t i o n o f p r o k a r y o t e s ( a c t i n o m y c e t e s a n d bac te r i a ) w a s s i gn i f i cant ly g r e a t e r i n t u b e s that c o n t a i n e d seed l ings c o m p a r e d t o s o i l a l o n e ( m e a n ± S . E . l o g c f u - g " 1 s o i l : f o r a c t i n o m y c e t e s = 4 . 3 3 ± 0 . 0 6 4 w i t h seed l ings versus 3 .78 ± 0 . 0 9 8 w i t h o u t seed l ings ; f o r b a c t e r i a = 5 . 8 5 ± 0 . 0 4 4 w i t h seed l ings versus 5 .63 ± 0 . 0 4 3 w i t h o u t s e e d l i n g ; P O . 0 0 1 f o r b o t h o r g a n i s m s b y S t u d e n t ' s t test) , b u t th i s effect w a s n o t e v i d e n t f o r f i i n g i : 4 . 4 2 ± 0 .051 l o g c m - g " 1 s o i l w i t h s eed l ing s a n d 4 . 4 0 ± 0 . 0 3 9 l o g c f u - g " 1 s o i l w i t h o u t seedl ings . H o w e v e r , P G P R i n o c u l a t i o n h a d a d i f ferent ia l effect o n t h e a b u n d a n c e o f p r o k a r y o t i c o r g a n i s m s : n o s ign i f i cant i n o c u l a t i o n effects w e r e d e t e c t e d o n a c t i n o m y c e t e p o p u l a t i o n s izes , b u t a l l P G P R e x c e p t s t ra in S w 5 R N c a u s e d s ign i f i cant c h a n g e s i n b a c t e r i a l p o p u l a t i o n s ( T a b l e 6 .1) . T h e s e c h a n g e s w e r e p a r t i c u l a r l y e v i d e n t w h e n e v a l u a t e d o n S m i t h e r s s o i l ex t rac t s i n t h e a b s e n c e o f s p r u c e seed l ings , b u t n o g e n e r a l t r e n d c o u l d b e d i s c e r n e d (i.e., t h r e e s tra ins e n h a n c e d fores t s o i l b a c t e r i a l p o p u l a t i o n s a n d t h r e e r e d u c e d t h e m , t w o n o n s i g n i f i c a n t l y ) . L i k e a c t i n o m y c e t e s , f u n g a l p o p u l a t i o n s w e r e g e n e r a l l y u n a f f e c t e d b y b a c t e r i a l i n o c u l a t i o n . 6.3.3 Effects of PGPR inoculation on soil microbial community activity S o i l m i c r o b i a l ac t iv i ty , as i n d i c a t e d b y t h e t o t a l n u m b e r o f substrates u s e d o n B i o l o g ™ p l a te s w a s s imi l a r a m o n g s o i l t y p e s , s p r u c e s e e d l i n g a n d P G P R t r e a t m e n t s ( F i g . 6 .2 ) . I n a d d i t i o n , af ter 4 8 h i n c u b a t i o n , these ac t iv i t ie s b e c a m e stable as i n d i c a t e d b y the d i v e r s i t y i n d i c e s ( H ' ) b a s e d o n c o l o u r d e v e l o p m e n t ( F i g . 6 .3) . T h e r e f o r e , t h e B i o l o g ™ d a t a o b t a i n e d at 4 8 h i n c u b a t i o n w a s a s s u m e d t o r e p r e s e n t t h e m i c r o b i a l a c t i v i t y o f these soi ls . E u c l i d e a n d i s t ances b e t w e e n e a c h p a i r o f s t a n d a r d i z e d a b s o r b a n c e s ( A p p e n d i x 9 , T a b l e A 9 . 1 ) w e r e p l o t t e d w i t h M D S i n 2 d i m e n s i o n s ( F i g . 6 .4) . A f t e r 14 i t e ra t ions t o m i n i m i z e t h e K r u s k a l stress b y m o n o t o n i c M D S , t h e stress o f t h e f ina l c o n f i g u r a t i o n w a s 0 .151 a n d t h e p r o p o r t i o n o f 2 t o t a l v a r i a n c e s (r ) w a s 0 . 8 9 5 . T h e s m o o t h s l o p e o f t h e S h e p a r d d i a g r a m ( A p p e n d i x 9 , F i g . A 9 . 1 ) i n d i c a t e d that t h e d i s t ances b e t w e e n ob jec t s i n c r e a s e d g r a d u a l l y as d i s s i m i l a r i t y i n c r e a s e d . M D S o f t h e B i o l o g ™ substrate u t i l i z a t i o n pa t te rns i n d i c a t e d that S m i t h e r s a n d W i l l i a m s L a k e s o i l 141 Table 6.1 Viable population sizes of microorganisms (mean and standard error, log cfu-g" soil) isolated from two different soils treated with spruce seedlings and PGPR. Uninoculated Control L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Smithers soil Fungi Plant coexist 4.87 ±0.12 4.83 ± 0.04 4.74 ±0.09 4.36 ±0.21 4.64 ±0.12 4.65 ± 0.09 4.53 ±0.13 Soil only 4.27 ± 0.08 4.63 ±0.04 * 4.79 ±0.08 ** 4.60 ± 0.08 4.62 ±0.12 4.48 ± 0.06 4.16 ±0.29 Actinomycetes Plant coexist 4.62 ±0.08 4.35 ±0.09 4.90 ±0.10 4.65 ± 0.20 4.45 ±0.15 4.77 ±0.12 4.61 ±0.15 Soil only 3.62 ±0.50 3.10 ±0.33 3.97 ±0.12 4.17 ±0.05 4.15 ±0.54 3.72 ±0.33 3.53 ±0.25 Bacteria Plant coexist 5.80 ±0.16 5.82 ±0.08 5.80 ±0.14 5.98 ±0.08 6.11 ±0.08 * 5.78 ±0.10 6.16 ±0.06 * Soil only 5.61 ±0.08 5.31 ±0.04 ** 5.92 ±0.11 ** 6.07 ± 0.08 *** 5.96 ±0.10 *** 5.44 ± 0.08 5.58 ±0.05 Williams Lake soil Fungi Plant coexist 4.08 ±0.13 4.37 ±0.09 4.07 ±0.13 4.19 ±0.19 4.27 ±0.15 4.22 ±0.16 3.98 ±0.24 Soil only 4.23 ± 0.07 4.41 ±0.03 4.34 ±0.13 4.32 ±0.17 4.18 ±0.17 4.27 ±0.15 4.33 ±0.09 Actinomvcetes Plant coexist 3.65 ±0.31 4.08 ±0.12 4.12 ±0.13 * 4.37 ±0.21 * 4.47 ± 0.09 3.97 ±0.12 3.68 ±0.31 Soil only 3.35 ±0.34 3.47 ±0.43 3.43 ±0.40 3.81 ±0.35 4.20 ±0.17 3.60 ± 0.28 3.85 ±0.37 Bacteria Plant coexist 6.15 ±0.07 5.73 ±0.17 * 5.59 ±0.14 ** 5.41 ±0.18 ** 6.16 ±0.05 5.63 ± 0.26 ** 5.81 ±0.20 Soil only 5.76 ±0.18 5.50 ±0.14 5.29 ±0.10 *** 5.63 ±0.03 5.43 ± 0.09 ** 5.26 ±0.05 *** 6.07 ±0.07 * ***, ** and * indicate that the mean value differs from me uninoculated control according to Fisher's protected LSD at.P<0.01,0.05, and 0.1, respectively (a) Smithers soil without seedling (b) Smithers soil with seedling 24h 48h 72h 96h 24h 48h 72h 96h Fig. 6.2 Number of Biolog™ substrates used by the two forest soil extracts after inoculation with PGPR strains in the presence and absence of spruce seedlings. (a) Smithers soil without seedling (b) Smithers soil with seedling 24h 48h 72h 96h (c) Williams Lake soil without seedling 2.0 £ 1 . 5 CO "a 1.0 0.5 0.0 -— = J — ^ 4 i 24h 48h 72h Incubation time 96h 24h 48h 72h (d) Williams Lake soil with seedling 48h 72h Incubation time 96h o L6-16R D Pw2R A S20R V Sm3RN Ss2RN Sw5RN + Control 6.3 Shannon diversity index (FT) for Biolog™ plate responses to the two forest soil extracts after inoculation with PGPR strains in the presence and absence of spruce seedlings. Q 1.5 1.0 • S20 " S s 2 • S20 0.5 0.0 • Pw2 • Sm3 •Sm3 • L6 OSm3 • c Sm3 Ss2» # #S20 C " QSw5 L6 • Sw5 O c •c O Sw5 • L6 • Sw5 -0.5 Pw20 • Pw2 • D P w 2 Ss2 S20O -1.0 S s 2 0 -1.5 Q L 6 i i i i i i > 1 -2.0 -1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0 DIMENSION-1 Fig. 6.4 Multidimensional scaling of carbon substrate utilization of two forest soils with or without spruce seedlings after inoculation with PGPR. L6:L6-16R; Pw2:Pw2R; S20:S20R; Sm3:Sm3RN; Ss2:Ss2RN; Sw5:Sw5RN; and C:uninoculated control. • : Smithers soil with seedling; • : Smithers soil without seedling; 0:Williams Lake soil with seedling; and • : Williams Lake soil without seedling. 145 cornmunities were different (P=0.025). MDS also revealed a significant influence of spruce seedlings in each soil type (PO.001). However, bacterial inoculation had differential effects on soil communities in the presence and absence of seedlings. In general, PGPR had the greatest effect on soil 'functional diversity' in the absence of seedlings. These effects were greatly buffered by the presence of a seedling in Smithers soil, but not in Williams Lake soil (Fig. 6.4). The mean distance (± S.E.) between uninoculated control and PGPR treatments in Smithers soil with seedlings was 0.53 ± 0.077; without seedlings, it was 0.85 ± 0.13 (P=0.102). For Williams Lake soil, the mean difference between controls and PGPR treatments in the presence of seedlings was 1.47 ± 0.33, and 1.00 ± 0.14 (P=0.195) in the absence of seedlings. 6.3.4 Determination of direct and indirect mechanisms by the PGPR strains. None of the PGPR strains significantly enhanced biomass accumulation of spruce seedlings grown in sterile sand (Fig. 6.5). However, addition of forest soil extracts, whether sterilized or not, frequently facilitated plant growth promotion by these strains (Fig. 6.6 - 6.9). Because there was a significant interaction for root biomass (P=0.018) involving soil extracts and PGPR inoculation (Table A7.8), PGPR effects were analyzed separately for each soil extract. All PGPR strains except Pseudomonas Ss2RN caused statistically significant shoot growth promotion in the presence of sterile Williams Lake soil, while Pseudomonas strains Sm3RN and Sw5RN were the most effective PGPR, primarily as root growth promoters, with non-sterile Williams Lake soil (Fig. 6.7). Although not statistically significant, Bacillus strains L6-16R and Pw2R enhanced root biomass in non-sterile Smithers soil by 20.5% and 16.0%, respectively (Fig. 6.8) Amendment of either non-sterile soil extract, particularly that from Smithers soil, caused substantial reductions in seedling biomass. Shoot growth reductions were greater than those associated with roots, therefore root/shoot ratios in treatments involving unsterilized soil extracts were higher than those with sterilized extracts (Table 6.2). 146 (a) Shoot L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control (b) Root L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control 6.5 (a) Shoot and (b) root biomass of spruce seedlings grown under gnotobiotic conditions after inoculation with PGPR (n=20). Error bars indicate the standard errors of the means. 147 (a) Shoot L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control (b) Root L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control 6.6 (a) Shoot and (b) root biomass of spruce seedlings after PGPR inoculation with or without soil extract originated from the Smithers Shoe-house site (n=18). Error bars=the standard errors of the means; * indicates that the mean value differs from that of uninoculated control with P<0.1 (two tailed). 148 (a) Shoot L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control (b)Root L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN Control . 6.7 (a) Shoot and (b) root biomass of spruce seedlings after PGPR inoculation with or without soil extract originated from the Williams Lake landing site (n=18). Error bars=the standard errors of the means; *** and * indicate that the mean values differ from that of uninoculated control with P<0.01 and P<0.1, respectively. 149 (a) L6-16R (b) Pw2R (c) S20R 30 h 25 20 Shoot Root (Legend) I I Without soil extract 30 -25 -20 15 £ 10 5 0 -5 Shoot Root With Smithers soil extract (sterilized) With Smithers soil extract (unsterilized) 30 -25 -20 '15 o 10 5 0 -5 Shoot Root With Williams Lake soil extract (sterilized) With Williams Lake soil extract (unsterilized) Fig. 6.8 Percent change of shoot and root biomass from uninoculated control after inoculation with Bacillus PGPR strains (L6-16R, Pw2R, and S20R) with or without soil extracts collected from Smithers and Williams Lake. (a) Sm3RN (b) Ss2RN (c) Sw5RN 60 30 25 20 15 S 10 5 0 -5 Shoot Root Shoot Root Shoot Root (Legend) I I Without soil extract With Smithers soil extract (sterilized) With Smithers soil extract (unsterilized) With Williams Lake soil extract (sterilized) With Williams Lake soil extract (unsterilized) Fig. 6.9 Percent change of shoot and root biomass from uninoculated control after inoculation with Pseudomonas PGPR strains (Sm3RN, Ss2RN, Sw5RN) with or without soil extracts collected from Smithers and Williams Lake. Table 6.2 Root/shoot ratio of spruce seedlings after treatment with PGPR and forest soil extracts. Strain No forest soil (sterile sand) Smithers soil extract Williams Lake soil extract Sterilized Sterilized Sterilized L6-16R 0.96 ± 0.05 0.62 ± 0.03* 0.56 ±0.03 Pw2R 0.98 ± 0.07 0.57 ± 0.04 0.58 ± 0.03 S20R 0.97 ± 0.06 0.52 ± 0.02 0.61 ± 0.04 Sm3RN 0.99 ± 0.05 0.49 ± 0.02 0.59 ± 0.03 Ss2RN 1.02 ± 0.07 0.55 ± 0.04 0.59 ± 0.04 Sw5RN 0.92 ± 0.04 0.58 ± 0.03 0.59 ± 0.02 Uninoculated control 0.97 ± 0.05 0.53 ± 0.02 Unsterilized 0.62 ± 0.03 Unsterilized L6-16R 0.74 ± 0.05 0.58 ± 0.03 Pw2R 0.71 ±0.05 0.64 ± 0.02 S20R 0.64 + 0.06 0.61 ±0.03 Sm3RN 0.61 ± 0.03 0.70 ± 0.04 Ss2RN 0.72 ± 0.04 0.64 ± 0.03 Sw5RN 0.70 ± 0.07 0.68 ± 0.03 Uninoculated control 0.68 ± 0.04 0.62 ± 0.03 indicates the mean value differs significantly from uninoculated control (P <0.1). 152 6.3.5 Recovery of PGPR in microcosm. PGPR population sizes recovered from spruce treated with the forest soil extracts were generally smaller than those associated with spruce grown in sand only (Fig. 6.10). In general, unsterilized soil extracts were consistently associated with slightly smaller PGPR populations compared to treatments involving sterilized extracts. However, Pseudomonas strain Ss2RN was recovered with the lowest populations of all six strains after treatment with any of the forest soil extracts. No significant effect (P<0.1) of spruce seedling presence on the population size of recovered PGPR strains was detected (Fig. 6.11). 6.3.6 Correlations of seedling biomass and populations of soil microorganisms. Seedling biomass was negatively correlated with the population size of viable fiingi and actinomycetes, possibly due to competition for nutrients (Table 6.3). Seedling biomass was not correlated with the population of viable bacteria or PGPR at the end of the experiment. Euclidean distances based on carbon substrate utilization by soil microorganisms between each PGPR treatment and uninoculated controls were correlated with the seedling biomass but not with the populations of viable bacteria including PGPR. 153 0 L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN (Legend) I I Without soil extract With Smithers soil extract (sterilized) With Smithers soil extract (unsterilized) With Williams Lake soil extract (sterilized) With Williams Lake soil extract (unsterilized) Fig. 6.10 Recovery of PGPR from spruce grown with or without forest soil extracts. Error bars indicate standard errors of the means (n=4). (a) Smithers soil (b) Williams Lake soil L6-16R Pw2R S20R Sm3RN Ss2RN Sw5RN (Inoculum) .6.11 Recovery of the inoculated PGPR strains with or without spruce seedlings after amendment with forest soil extracts (unsterilized). Error bars indicate standard errors of the means (n=4). 155 Table 6.3 Pearson correlation coefficient (r) between seedling biomass and soil microorganism variables (n=12). Shoot biomass Root biomass Viable fungi Viable actinomycetes 1 Viable bacteria1 Re-isolated PGPR inoculum 1 Shoot biomass Root biomass Viable fungi Viable actinomycetes Viable bacteria Re-isolated PGPR inoculum 2 Distance from control 1.000 0.918 *** -0.740 *** -0.410 0.050 0.709 ** 1.000 -0.818 *** -0.740 *** -0.670 ** -0.300 0.230 0.697 ** 1.000 0.710 ** 0.350 -0.060 -0.515 * 1.000 0.350 -0.150 -0.420 1.000 0.600 -0.050 ** 1.000 0.016 1 Population with dilution plate counts (log cfu-g"1 soil). 2 Euclidean distance between controls and PGPR inoculation treatments. ***,**, and *: Significantly different by ANOVA of linear regression (two tailed) at PO.01, P< 0.05, and P< 0.1, respectively. 6.4 DISCUSSION Seedling biomass results indicate that, except for strain Ss2RN, these PGPR strains affected plant growth similarly under the assay conditions described in this chapter (Fig. 6.6 and 6.7). The mechanism by which plant growth was stimulated appears to be more complex than encompassed by the direct and indirect mechanisms described in the introduction. Seedling growth promotion was detected in the presence of both sterilized and unsterilized forest soil extracts, leading to the conclusion that growth promotion by these strains was direct because it did not depend on the presence of other microorganisms (i.e., PGPR were effective in the presence of sterilized extracts). However, PGPR inoculation did not enhance plant growth in the absence of sterile forest soil extracts, which suggests that growth promotion requires an additional, presumably abiotic factor present in these forest soil extracts. It is interesting that Pseudomonas Ss2RN was the least effective growth promoter under these assay conditions, even in the presence of unsterilized forest soil extracts, because this strain was very effective as a spruce PGPR under field conditions (Chapter 5). Such variability of plant growth responses to microbial inoculation between under laboratory and field conditions has been observed repeatedly (Kloepper et al. 1989; Molina et al. 1992). Since Ss2RN was the only strain that displayed such variability between field and laboratory performance, its growth promoting mechanism likely differs from those of the other strains. In fact, it can be argued that strain Ss2RN works on plant growth promotion mainly through a mechanism related to controlling DRMO that were excluded from the soil extracts used in laboratory assays, though direct evidence for this hypothesis is lacking. This strain's copious production of a fluorescent pigment on King's A and B agar, likely a siderophore, indirectly supports this contention, because iron-chelating siderophores produced by fluorescent Pseudomonas species have been linked to biocontrol activity (Schroth and Hancock 1982; O'Sullivan and O'Gara 1992). It is unlikely that mechanisms involving direct facilitation of nutrient availability explain the effects of strains L6-16R, Pw2R, S20, Sm3RN, and Sw5RN on spruce seedlings. For example, none of the spruce seedlings inoculated with these bacteria supported rhizosphere acetylene 157 reduction activity after 48 h incubation with 10% acetylene (data are not shown). This finding is in agreement with the conclusions in other studies (Chanway and Holl 1991; Hong et al. 1991b) that N 2 fixation is at best, of secondary importance in growth promotion mechanisms. Furthermore, all mineral nutrients including phosphorus were provided in soluble forms rendering mineral solubilization an unlikely mechanism of growth promotion. It should be noted that previous work with Bacillus polymyxa L6 suggested that production of plant growth substances similar to IAA were likely involved in bacterial promotion of root growth (Holl et al. 1988). Interestingly, strain Pw2 also produces large amounts of IAA in vitro (Srinivasan et al. 1996), therefore involvement of phytohormone related substances in seedling growth promotion by strains L6-16R and Pw2R can not be ruled out as a possible mechanism. The lack of seedling growth responses in sterile sand culture with no forest soil extract suggests that these PGPR strains may require chemical compounds as precursors of such plant-growth stimulating compounds. For example, precursors of auxins, e.g., L-tryptophan have been detected in root exudates, and IAA synthesis from L-tryptophan in soil appears to be linked with the co-metabolism of L-tryptophan by L-aminotransferases (Martens and Frankenberger 1993). Furthermore, tryptophan availability is extremely limited in soil (Frankenberger and Arshad 1995), which emphasizes the potential importance of rhizosphere sources. The importance of L-tryptophan in Douglas-fir seedling growth was demonstrated by Frankenberger and Poth (1987) by applying a minute amount of this compound to Douglas-fir (Pseudotsuga menziesii) seeds and inoculating with an ectomycorrhizal fungus (Pisolithus tinctorius). They found no significant differences in seedling growth due to mycorrhizal inoculation treatments, but minute amounts of l -6 8 L-tryptophan (0.34-34 pg-kg" soil = 10" -10" ML-tryptophan) resulted in the greatest seedling growth responses (24%-28% root elongation), indicating its physiological rather than nutritional role. In addition, the most effective treatment combination was fungal inoculum plus L-tryptophan, which stimulated root biomass growth more than shoot biomass. A change in carbon allocation in favor of root biomass was also found in my experimental treatment that included unsterilized forest soil extracts. However, this shift in carbon allocation to 158 roots occurred as total biomass decreased, unlike the mycorrhizae-Douglas-fir system cited above (Frankenberger and Poth 1987). Increases of root/shoot ratio have also been observed as availability of N, P, S, and Fe decrease (Ingestad 1979; Cannell and Dewar 1994). Such decreases may be expected in treatments with unsterilized forest soil extracts as introduced microorganisms could have increased competition for mineral nutrients. Indeed, seedling biomass was negatively correlated with the population sizes of fungi and actinomycetes in the microcosms (Table 6.3). In contrast, significant correlations of Euclidean distances based on carbon substrate utilization patterns of PGPR treatments and uninoculated controls were found with shoot and root biomass, but not with population sizes of viable bacteria, including PGPR. Therefore, plant mediated effects appear to be more important than bacterial effects in causing changes to soil communities in microcosms with seedlings, i.e., the rhizosphere effect. This reduced inoculation effect on the soil microbial community is consistent with the hypothesis that PGPR operate via a direct growth promotion mechanism. The recovery of PGPR strains from the rhizosphere did not correlate with seedling growth responses to inoculation treatments. This may be explained by the time lag between the host plant responses and bacterial metabolism of precursors in the system, followed by declining PGPR population sizes due to decreasing amounts of nutrients in the microcosm. Although it is generally agreed that rhizosphere colonization by PGPR is necessary for growth promotion, proliferation of inoculum is not always detected. For example, significant growth promotion of onion was observed in the field even though only low levels of root colonization by PGPR were detected (Reddy and Rahe 1989). In addition, electron microscopic studies of canola seedling root in the presence of a growth promoting pseudomonad indicated that bacterial adherence to the seed coat before germination was sufficient to enhance root elongation (Hong et al. 1991a). Alternatively, the different growth promotion responses to bacterial inoculation between tests in sterile sand and in forest soil-amended sand medium could be explained by PGPR induction of phytohormone production by the host plant itself. It is known that some bacterial genes are capable of modifying hormonal balances in plants (Gaudin et al. 1994), therefore results obtained 159 by mimicking plant responses to PGPR inoculation by exogenous application of phytohormones (Brown 1972; Holl et al. 1988) could be explained either by production of phytohormones by bacteria or the plant itself. In this case, only a threshold quantity of the signal for phytohormone production would be required for a plant growth response, and one would not predict a quantitative relationship between bacterial colonization and plant growth responses. Plant pathogenic bacteria represent a valuable model for this hypothesis. Agrobacterium species have been studied for almost a century, and are known to induce production of novel metabolites, the opines, in plants (Smith and Townsend 1907). While A. tumefaciens gives rise to crown galls, the related species A. rhizogenes induces proliferation of roots, resulting in a conditions known as "hairy root" disease. These morphological changes induced by Agrobacterium in plants result from the expression of certain genes passing from the bacteria to the plant. The genes, or T-DNA, carried by the bacterial Ti-plasmid are integrated into the genome of the host plant and expressed (Winans 1992). This phenomenon has been extensively utilized to create transgenic plants in recent years. In addition, it is also possible that a signal transduction mechanism induces plant hormone production, analogous to ISR (see Section 6.1). Information regarding this type of signal transduction resulting in PGPR-mediated phytohormone production is scarce, but by analogy with bacterial plant pathogens, e.g., Agrobacterium species and Pseudomonas syringae, it is possible that PGPR strains may use the host plant's signal regulation system, perhaps in conjunction with a chemical co-factor in the soil. Although distinguishing the source of phytohormones in vivo, i.e., host plant versus PGPR, is technically difficult, further research in molecular signals in plant-PGPR communication is needed to elucidate the mechanisms of these PGPR. Experiment 7 described in this chapter was not repeated, therefore results are best used to generate hypotheses rather than hard conclusions. I hypothesize that the mechanism by which strains L6-16R, Pw2R, S20R, Sm3RN and Sw5RN stimulate plant growth involves a direct pathway, but with a dependence on soil-borne substance(s) supplemented by the forest soil extract. Such precursor or activator compound(s) may serve as an energy source for several soil 160 organisms, and therefore may not be consistently available in soil. If the mechanisms of bacterial growth promotion are dependent on such soil factor(s), their patchy distribution may explain a significant part of variable seedling responses to PGPR inoculation. The second objective of this chapter was to evaluate interactions between PGPR and indigenous forest soil communities in relation to possible negative effects of the introduced microorganisms. The size of the PGPR populations in the microcosms at the end of the study was similar between soils and treatments with and without seedlings. In general, less than 1% of the initial inoculum population size was recovered from microcosms 18 weeks after treatments were established. Although this represented a large proportional decrease, soil PGPR populations were still between 103 and 105 cfu-g"1, and comprised ca. 10% of the total soil bacterial populations in microcosms. Similar rates of population decline have been observed in other studies (Reddy and Rahe 1989; Holl and Chanway 1992). Notwithstanding the decline in PGPR population sizes, sizable soil community responses were detected in response to PGPR inoculation that depended on the origin of the soil and the presence of seedlings. For example, PGPR inoculation had little significant effect on Smithers soil microbial population sizes (Table 6.1) or carbon substrate utilization patterns (Fig. 6.4) when spruce seedlings were present in the microcosms. In contrast, four of the six PGPR strains (L6-16R, Pw2R, S20R, and Sm3RN) had significant effects on Smithers soil bacterial and fungal abundance in the absence of spruce seedlings (Table 6.1). PGPR inoculation effects on the Smithers soil community in the absence of seedlings were also evident in carbon substrate utilization patterns (Fig. 6.4). These results suggest that the presence of a spruce seedling buffered the Smithers soil community from perturbations caused by PGPR introduction, which were otherwise manifest by increases or decreases in bacterial and fungal population sizes as well as large shifts in carbon substrate utilization patterns. In contrast to Smithers soil communities, Williams Lake soil communities seemed susceptible to perturbation by bacterial inoculation whether or not spruce seedlings were present in microcosms. Williams Lake soil bacterial populations were significantly altered by four of the six 161 PGPR strains in the presence and absence of seedlings (Table 6.1). Furthermore, PGPR inoculation caused large shifts in carbon substrate utilization profiles in all microcosms with or without seedlings (Fig. 6.4). These results suggest that the origin of the soil microbial community can be more important than the presence or absence of seedlings in determining microfloral responses to PGPR introductions. In addition to the general trends observed with soils and seedlings, some PGPR strain specific effects on quantitative and qualitative aspects of soil microbial communities were also detected. For example, Bacillus strain L6-16R did not significantly affect the number of soil microorganisms associated with the seedlings grown in the Williams Lake soil, but the carbon utilization pattern of the microbial community in the same system was substantially different from that of the uninoculated control. A similar trend was detected in the treatment with Bacillus strain S20R inoculated in the Williams Lake soil without seedlings. Such strain specific effects are not surprising considering the large differences that different strains of the same bacterial species may exhibit on plant growth (Kloepper 1993). A final noteworthy point relates to the conditions under which this experiment was performed. Soil communities in each microcosm originated from a 0.5 g soil sample containing .r approximately 107 bacterial and fungal cells. PGPR treatment involved addition of 106 to 108 bacterial cells in a small volume (25 ml sand). Under such conditions, biotic interactions would be expected to be intense resulting in treatment effects that are comparatively easy to detect. However, even under these artificial conditions, Smithers and Williams Lake indigenous soil microbial communities showed a great degree of resilience as population declines of indigenous microorganisms were generally at least an order of magnitude smaller than those of PGPR by the end of the experiment. Furthermore, in the case of Smithers soil with seedlings, almost no effect on substrate utilization could be detected. Because similar inoculum densities would be employed in forest plantations, I hypothesize that PGPR will exert fewer effects or effects of less magnitude than those detected in my compact microcosms, but field tests with these organisms at reforestation sites will be required to confirm this assertion. 162 CHAPTER 7 CONCLUSIONS From the preceding chapters, it is clear that inoculation of interior spruce and lodgepole pine seeds with strains L6, Pw2, S20, Sm3, Ss2 and Sw5 can produce statistically significant increases in seedling growth and may prove to be useful in forest regeneration. However, significant seedling growth response variability was detected even during the initial screening assays (Chapter 2) which were conducted under similar experimental conditions. Growth response variability of spruce seedlings between assays conducted under different environmental conditions was even greater (Fig. 7.1). Such inconsistency is an impediment to the practical use of PGPR technology, and part of my research was a quest to determine the cause(s) of variable spruce seedling growth responses to bacterial inoculation. In this concluding chapter, the experiments presented in previous chapters are discussed with respect to PGPR x conifer interactions according to my specific objectives presented in Chapter 1. Objective 1. To investigate specificity involving conifer host plants (ecotypes and genus), PGPR strains and the soils from which they originated in relation to seedling growth promotion and growth response variability. Specificity of PGPR is a possible factor contributing to inconsistent seedling growth responses. Indeed, spruce biomass accumulation was greatest when conifer ecotypes were inoculated with PGPR originating from the same geographical area as spruce seed (Chapter 2). However, "coexistent" combinations of host ecotypes, PGPR strains and forest soils did not result in seedlings that were significantly larger than those in "unrelated" factor combinations. Hence, it was concluded that co-adaptation involving conifer ecotypes, PGPR and forest soils do not contribute significantly to variable conifer seedling growth responses to PGPR inoculation. Use of wild conifer seed, which is highly heterozygous may have obscured detection of fine scale PGPR-plant relationships, but from a practical perspective, a low degree of specificity among the 163 —I I I I I I I I I I I I I I Scrl Scr2 Expl Exp2 Exp3 Exp4 Exp6 Exp6 Exp6 Exp6 Exp7 Exp7 Exp7 Exp7 (SBC) (SSH) (WLR) (WLL) (Sst) (Wst) (Snst) (Wnst) n=20 n=36 n=54 n=18 n=60 n=90 n=27 n=27 n=27 n=27 n=18 n=18 n=18 n=18 EXPERIMENT . 7.1 Percent changes of spruce seedling biomass from uninoculated control after PGPR inoculation on an experiment basis during this research project. Scr: screening; Exp: Experiment; (SBC): Smithers, Blunt Creek site; (SSH): Smithers, Shoe-house site; (WLR): Williams Lake, Regular cut-block site; (WLL): Williams Lake, Landing site; (Sst): sterile Smithers soil added; (Wst): sterile Williams Lake soil added; (Snst): non-sterile Smithers soil added; and (Wnst): non-sterile Williams Lake soil added. three components is desirable to avoid cumbersome matching of PGPR strains with conifer ecotypes. Objective 2. To assess the interaction of PGPR and mycorrhizal fungi in relation to growth promotion of spruce and pine seedlings. Historically, mycorrhizal fungi have been the most intensively studied root-associated microorganisms in forestry. Though bacteria are likely the most numerous root-associated organisms and the interactions of mycorrhizal fiingi and associated soil bacteria are well-known, they have only recently been investigated for their potential as seeding growth promoting organisms in forestry. Therefore, I carefully evaluated the mycorrhizal status of seedlings after PGPR inoculation to determine if bacteria x symbiotic fungus interactions contribute to variable seedling growth responses (Chapter 3). My results suggest that PGPR can enhance the performance of both non-mycorrhizal and mycorrhizal conifer seedlings, and do so through a mechanism other than simple enhancement of fungal infection. A synergistic effect between PGPR strains and mycorrhizal fungi may occur, but it is certainly not a prerequisite for conifer seedling growth promotion. Therefore, bacterial dependence on mycorrhizae does not explain a significant component of the conifer seedling growth response variability after inoculation with the PGPR strains described in this thesis. Objective 3. To evaluate internal root colonization of spruce seedlings by PGPR and to determine internal seedling tissue(s) that endophytic PGPR colonize. Root colonization by PGPR is a logical prerequisite for plant growth promotion. A number of studies have demonstrated that certain PGPR strains may proliferate not only on and around plant roots but also inside root tissues. The possibility of enhancing plant growth promotion efficacy and/or reproducibility with endophytic microorganisms offers obvious benefits to industries that depend on some aspect of enhanced plant productivity. From an ecological perspective, the 165 development of such relationships is intriguing as these could represent stages in the evolution of new mutualistic symbioses. Results from immunofluorescent microscopy (Chapter 4) suggest that two of the PGPR strains I worked with, Bacillus strain Pw2R and Pseudomonas strain Sm3RN, are vascular tissue endophytes and possess some physiological characteristics that are different from the other non-endophytic strains. Having provided strong evidence that these two strains were distinct from the other strains investigated in their ability to colonize internal seedling tissues, I then evaluated the hypothesis that endophytic PGPR were more effective seedling growth promoters than those restricted to colonizing external root tissues (objective 4). Objective 4. To determine if PGPR plant growth promotion efficacy on spruce seedlings is related to external and internal root colonization under field conditions. My field study, using pre-inoculated spruce seedlings, confirmed the ability of PGPR to colonize external and, in two cases, internal root tissues after one growing season in the field (Chapter 5). However, inoculation treatments of endophytic PGPR strains did not result in better seedling growth in the field compared to external root colonists. Since the rhizosphere population sizes of most PGPR declined under field conditions, it seems that continued proliferation of the beneficial bacterial population is not a necessary condition for PGPR plant growth promotion. Though associated microbial populations were not assessed in my experiment, PGPR may have been replaced through some ecological interactions, of which niche exclusion is a reasonable possibility. The results of the field trial imply that PGPR population declines are not related to growth response variability of spruce seedlings. Therefore, once induced in the nursery, seedling growth promotion by PGPR may persist after outplanting, and contribute to enhancing seedling performance in the field. Moreover, the decline in PGPR populations in the field indirectly suggests that disturbance of indigenous microbial communities due to introduction of these strains would be small. 166 Objective 5. To determine if the mechanism by which PGPR stimulate spruce seedling growth involves other microorganisms. The results of the microcosm experiment (Chapter 6) indicate that the mechanism by which strains L6-16R, Pw2R, S20R, Sm3RN and Sw5RN stimulate plant growth involves a direct pathway, but with dependence on a soil-borne substance (or substances) supplemented by the forest soil extract. Such compounds could serve as precursor or activator molecules for the production of plant growth altering substances. However, because they may also serve as nutrients for soil organisms, they may not be consistently available in soil. If the mechanism by which bacteria stimulate plant growth involves such a soil organic factor, its heterogeneous distribution in soil could explain variable seedling responses to PGPR inoculation. This hypothesis was not tested as the soil extract experiment was conducted only once, and hence, the growth response variability between assays in which soil extracts were added was not evaluated. However, if exogenous soil organic factors were required to reduce variability, such a treatment could easily be provided in a nursery. Objective 6. To assess interactions between PGPR and indigenous forest soil communities in relation to possible negative effects of the introduced bacteria. Indigenous soil microbial communities showed some resilience to PGPR inoculation, as population declines of indigenous microorganisms were generally smaller than those of PGPR by the end of Experiment 7. Notwithstanding the decline in PGPR population sizes, sizable soil community responses to PGPR that depended on the origin of the soil and the presence of seedlings were detected (Chapter 6). Seedlings may buffer indigenous soil communities from perturbations caused by PGPR introduction, but the origin of the soil microbial community may be more influential than seedlings in determining microfloral responses to PGPR introduction. 167 Future Research Needs The experimentation described in this thesis shows that growth of conifer seedlings, particularly spruce, can be enhanced by inoculation with PGPR. However, growth promotion was inconsistent, and I was unable to identify a single cause of such variability. Nevertheless, my results repeatedly suggest that both abiotic and biotic soil factors exert a significant influence on PGPR efficacy. It would be fruitful to investigate the importance of soil extracts in repeated PGPR assays, and ultimately, to identify the component(s) of soil extracts that affects PGPR efficacy. Longer-term field studies are also required to evaluate persistence of both PGPR populations and efficacy of plant growth promotion. Preliminary results suggest that site quality (e.g., compacted landing versus regular sites) may influence PGPR efficacy, so field tests should be conducted at sites where different environmental stresses are known to limit conifer seedling growth. In addition, I hypothesize that PGPR will likely exert fewer effects or effects of less magnitude on indigenous soil microbial communities at reforestation sites than those detected in microcosms, but field tests will be needed to evaluate this prediction. While some studies suggest that stimulation of root infection by mycorrhizal fungi is central to the mechanism by which seedling growth is enhanced, my research indicates that the influence of mycorrhizal fungi on PGPR efficacy is of secondary importance. However, this conclusion is based on data from short-term, controlled environment studies, therefore, field tests should also include careful evaluation of mycorrhizae formation. Such studies should also include endophytic PGPR as potential long term benefits that might accrue from this more intimate plant x PGPR relationship. Such information would also improve our understanding of non-pathogenic bacterial endophytes from ecological and evolutionary perspectives. PGPR inoculation may prove to be an inexpensive, environmentally-benign technique for enhancing forest plantation productivity, but the mechanisms of growth promotion and causes of growth response variability need to be understood before significant commercial advancement will 168 occur. Hopefully, further research will result in a better understanding of PGPR ecology and their effects on seedling growth, and ultimately, in improved forest regeneration and sustainable forestry. 169 LITERATURE CITED Ae, N., Arihara, J., Kensuke, O., Yoshihara, T., and Johansen, C. 1990. Phosphorus uptake by pigeon pea and its role in cropping system of the Indian subcontinent. Science 247: 477-480. Alstom, S., and Burns, R.G. 1989. Cyanide production by rhizobacteria as a possible mechanism of plant growth inhibition. Biol. Fertil. Soils 7: 232-238. Alvarez, A M . , Benedict, A.A., Mizumoto, C.Y., Pollard, L.W., and Civerolo, E L . 1991. Analysis of Xanthomonas campestris pv citri and Xanthomonas campestris pv citrumelo with monoclonal antibodies. Phytopathology 81: 857-864. Amaranthus, M.P., Li, C.Y., and Perry, D.A. 1990. Influence of vegetation type and madrone soil inoculum on associative nitrogen fixation in Douglas-fir rhizospheres. Can. J. For. Res. 20: 368-371. Amaranthus, M.P., and Perry, D.A. 1987. Effect of soil transfer on ectomycorrhiza fromation and the survival and growth of conifer seedlings on old, nonreforested clear-cuts. Can. J. For. Res. 17: 944-950. Anderson, A.J. 1983. Isolation from root and shoot surfaces of agglutinins that show specificity for saprophytic pseudomonads. Can. J. Bot. 61: 3438-3443. Anderson, A.J., Habibzadegah-Tari, P., and Tepper, C.S. 1988. Molecular studies on the role of a root surface agglutinin in adherence and colonization by Pseudomonas putida. Appl. Environ. Microbiol. 54: 375-380. Bachmann, G., and Kinzel, H. 1992. Physiological and ecological aspects of the interactions between plant roots and rhizosphere soil. Soil Biol. Biochem. 24:543-552. Baldani, V.L.D., and Dobereiner, J. 1980. Host plant specificity in the infection of cereals with Azospirillum spp. Soil Biol. Biochem. 12: 433-439. Baldani, V.L.D., Baldani, J.I., and Dobereiner, J. 1983. Effect of Azospirillum inoculation on root infection and nitrogen incorporation in wheat. Can. J. Microbiol. 29: 924-929. Ball, E,M., De Boer, S.H., and Schaad, N.W. 1990. Polyclonal antibodies. In: Serological methods for detection and identification of viral and bacterial plant pathogens. Hampton, R. O., Ball, E. M. and De Boer, S. H. (eds.). American Phytopathological Society, St. Paul, MN. pp. 33-54. Bashan, Y., and Levanony, H. 1990. Current status of Azospirillum inoculation technology: Azospirillum as a challenge for agriculture. Can. J. Microbiol. 36: 591-608. Bashan, Y., Levanony, H., and Whitmoyer, R.E. 1991. Root surface colonization of non-cereal crop plants by pleomorphic Azospirillum brasilense Cd. J. Gen. Microbiol. 137: 187-196. 170 Bashan, Y., Ream, Y. Levanony, H., and Sade, A. 1989. Non-specific responses in plant growth, yield and root colonization of non-cereal crop plants to inoculation with Azospirillum barasilense Cd. Can. J. Bot. 67: 1317-1324. Becker, J.O., Hedges, R.W., and Messens, E. 1985. Inhibitory effect of pseudobactin on the uptake of iron by higher plants. Appl. Environ. Microbiol. 49: 1090-1093. Bell, C.R., Dickie, G.A., Harvey, W.L.G., and Chan, J.W.Y.F. 1995. Endophytic bacteria in grapevine. Can. J. Microbiol. 41: 46-53. Bonn, H.L., McNeal, B.L., and O'Connor, G.A. 1985. Soil Chemistry. John Wiley & Sons, Inc. New York, NY. 341pp. Bossier, P., Hofte, M., and Verstraete, W. 1988. Ecological significance of siderophores in soil. Adv. Microb. Ecol. 10: 385-414. Bowen, G.D., and Theodorou, C. 1979 Interactions between bacteria and ectomycorrhizal fungi. Soil Biol. Biochem. 11: 119-126. Brisbane, P.G., and Rovira, A.D. 1988. Mechanisms of inhibition of Gauemannomyces graminis var. tritici by fluorescent pseudomonads. Plant Pathol. 37: 104-111. Broadbent, P., Baker, K.F., Franks, N., and Holland, J. 1977. Effect of Bacillus spp. on increased growth of seedlings in steamed and in nontreated soil. Phytopahology 67: 1027-1034. Bronstein, J.L. 1994. Our current understanding of mutualism. Quart. Rev. Biol. 69: 31-51. Brown, M.E. 1972. Plant growth substances produced by microorganisms of soil and rhizosphere. J. Appl. Bacteriol. 35: 443-451. Burdett, A.N., Herring, L.J., and Thompson, CF. 1984. Early growth of planted spruce. Can. J. For. Res. 14: 644-651. Burr, T.J., and Caesar, A. 1984. Beneficial plant bacteria. CRC Critical Reviews in Plant Science 2: 1-20. Caesar, A., and Burr, T.J. 1987. Growth promotion of apple seedlings and rootstock by specific strains of bacteria. Phytopathology 77: 1583-1588. Campbell, R., and Greaves, M.P. 1990. Anatomy and community structure of the rhizosphere. In: The rhizosphere. Lynch, J.M. (ed.). John Wiley & Sons, Inc. New York. NY. pp.11-34. Cannell, M.G.R., and Dewar, R.C. 1994. Carbon allocation in trees: a review of concepts for modelling. Adv. Ecol. Res. 25: 59-104. Chanway, CP. 1995. Differential response of western hemlock from low and high elevations to inoculation with plant-growth promoting Bacillus polymyxa. Soil Biol. Biochem. 27: 767-775. 171 Chanway, CP. 1997. Inoculation of tree roots with plant growth promoting soil bacteria: an emerging technology for reforestation. For. Sci. 43: 99-112. Chanway, CP. 1996. Endophytes: they're not just fungi! Can. J. Bot. 74: 321-322. Chanway, CP., and Holl, F.B. 1991. Biomass increase and associative nitrogen fixation of mycorrhizal Pinus contorta seedlings inoculated with a plant growth promoting Bacillus strains. Can. J. Bot. 69: 507-511. Chanway, CP., and Holl, F.B. 1992. Influence of soil biota on Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) seedling growth: The role of rhizosphere bacteria. Can. J. Bot. 70: 1025-1031. Chanway, CP., and Holl, F.B. 1993a. Ecotypic specificity of spruce emergence-stimulating Pseudomonasputida. For. Sci. 39: 520-527. Chanway, CP., and Holl, F.B. 1993b. First year field perfomance of spruce seedlings inoculated with plant growth promoting rhizobacteria. Can. J. Microbiol. 39: 1084-1088. Chanway, CP., and Holl, F.B. 1994a. Ecological growth response specificity of two Douglas-fir ecotypes inoculated with coexistent beneficial rhizosphere bacteria. Can. J. Bot. 72: 582-586. Chanway, CP., and Holl, F.B. 1994b. Growth of outplanted lodgepole pine seedlings one year after inoculation with growth promoting rhizobacteria. For. Sci. 40: 238-246. Chanway, CP., Nelson, L.M., and Holl, F.B. 1988. Cultivar specific growth promotion of spring wheat (Triticum aestivum L.) by coexistent Bacillus species. Can. J. Microbiol. 34: 925-929. Chanway, CP., Holl, F.B., and Turkington, R. 1990. Specificity of association between Bacillus isolates and genotypes of Lolium perenne L. and Trifolium repens L. from a grass/legume pasture. Can. J. Bot. 68: 1126-1130. Chanway, CP., Radley, R.A., and Holl, F.B. 1991a. Inoculation of conifer seed with plant growth promoting Bacillus strains causes increased seedling emergence and biomass. Soil Biol. Biochem. 23: 575-580. Chanway, CP., Turkington, R., and Holl, F.B. 1991b. Ecological implications of specificity between plants and rhizosphere micro-organisms. Adv. Ecol. Res. 21: 121-169. Chanway, CP., Shishido, M., Holl, F.B. 1994. Root-endophytic and rhizosphere plant growth-promoting rhizobacteria for conifer seedlings. In: Ryder, M.H., Stephens P.M., Bowen, G.D.(eds.) Improving plant productivity with rhizosphere bacteria. Glen Osmond: CSIRO Division of Soils, pp.72-74. Chapin, HJ, F.S., Van Cleve, K., and Tryon, P.R. 1986. Relationship of ion absorption to growth rate in taiga trees. Oecologica 69: 238-242. 172 Colinas, C , Molina, R., Trappe, J., and Perry, D. 1994a. Ectomycorrhizas and rhizosphere microorganisms of seedlings of Pseudotsuga menziesii (Mirb.) Franco planted on a degraded site and inoculated with forest soils pretreated with selective biocides. New Phytol. 127: 529-537. Colinas, C , Perry, D., Molina, R , and Amaranthus, M. 1994b. Survival and growth of Pseudotsuga menziesii seedlings inoculated with biocide-treated soils at planting in a degraded clearcut. Can. J. For. Res. 24: 1741-1749. Culling, C.F.A. 1974. Modern microscopy. Elementary theory and practice. Butterworth & Co., London. 148p. Curl, E.A., and Truelove, B. 1986. The Rhizosphere. Springer-Verlag, Berlin, Heidelberg. 288pp. Czabator, F.J. 1962. Germination value: An index combining speed and completeness of Pine seed germination. For. Sci. 8: 386-396. Danielson, R.M. 1991. Temporal changes and effects of amendments on the occurrence of sheathing (ecto-)mycorrhizas of conifers growing in oil sands tailings and coal spoil. Agric. Ecosyst. Environ. 35: 261-281. Danielson, R.M., and Visser, S. 1990. The mycorrhizal and nodulation status of container-grown trees and shrubs reared in commercial nurseries. Can. J. For. Res. 20: 609-614. Dazzo, F.B., and Brill, W.J. 1978. Regulation by fixed nitrogen of host-symbiont recognition in the Rhizobium-cXowet symbiosis. Plant Physiol. 62: 18-21. Dazzo, F.B., and Hubbell, D.H. 1975. Cross-reactive antigens and lectin as determinants of symbiotic specificity in the Rhizobium-dover association. Appl. Microbiol. 30: 1017-1033. De Boer, S.H. 1990. Immunofluoresence for bacteria. In: Serological methods for detection and identification of viral and bacterial plant pathogens. Hampton, R. O., Ball, E. M. and De Boer, S. H. (eds.). American Phytopathological Society, St. Paul, MN. pp. 295-298. Dillon, W.R., and Goldstein, M. 1984. Multivariate analysis, methods and applications. John Wiley & Sons, Inc. New York, NY. 587pp. Dobbs, R.C., and McMinn, R.G. 1977. Effects of scalping on soil temperature and growth of white spruce seedlings. In: 6th B.C. Soil Science Workshop Report, B.C. Min. Agr., Victoria, pp. 66-73. Dong, Z., Canny, M.J., McCully, M.E., Roboredo, M.R., Cabadilla, C.F., Ortega, E., and Rodes, R. 1994. A nitrogen-fixing endophyte of sugarcane stems. Plant Physiol. 105: 1139-1147. Dubeikovsky, A.N., Mordukhova, E.A., Kochetkov, V.V., Polikarpova, F.Y., and Boronin, A.M. 1993. Growth promotion of blackcurrant softwood cuttings by recombinant strain Pseudomonas fluorescens BSP53a synthesizing an increased amount of indole-3-acetic acid. Soil Biol. Biochem. 25: 1277-1281. 173 Duponnois, R., and Garbaye, J. 1991. Effect of dual inoculation of Douglas fir with the ectomycorrhizal fungus Laccaria laccata and mycorrhization helper bacteria (MHB) in two bare-root forest nurseries. Plant Soil 138: 169-176. Duponnois, R., Garbaye, J.,Bouchard, D., and Churin, J.L. 1993. The fungus specificity of mycorrhization helper bacteria (MHBs) used as an alternative to soil fumigation for ectomycorrhizal inoculation of bare-root Douglas-fir planting stocks with Laccaria laccata. Plant Soil 157: 257-262. Ellis, R.J. Thompson, LP, and Bailey, M.J. 1995. Metabolic profiling as a means of characterizing plant-associated microbial communities. FEMS Microbiology Ecology 16: 9-18. Fitter, A.H., and Garbaye, J. 1994 Interactions between mycorrhizal fungi and other soil organisms. Plant Soil 159: 123-132. Frankenberger, Jr., W.T., and Arshad, M. 1995. Phytohormones in soils. Marcel Dekker, Inc. New York, NY. 503pp. Frankenberger, Jr., W.T., and Poth, M. 1987. Biosynthesis of indole-3-acetic acid by the pine ectomycorrhizal fungus Pisolithus tinctorius. Appl. Environ. Microbiol. 53: 2908-2913. Frey-Klett, P., Pierrat, J.C., and Garbaye, J. 1997. Location and survival of mycorrhizal helper Pseudomonas fluorescens during establishment of ectomycorrhizal symbiosis between Laccaria bicolor and Douglas fir. Appl. Environ. Microbiol. 63: 139-144. Garbaye, J. 1991. Biological interactions in the rhizosphere. Experientia 47: 370-375. Garbaye, J. 1994. Helper bacteria: a new dimension to the mycorrhizal symbiosis. New Phytol. 128: 197-210. Garbaye, J., and Bowen, G.D. 1987. Effect of different microflora on the success of ectomycorrhizal inoculation of Pinus radiata. Can. J. For. Res. 17: 941-943. Garbaye, J., and Bowen, G.D. 1989. Stimulation of ectomycorrhizal infection of Pinus radiata by some microorganisms associated with the mantle of ectomycorrhizas. New Phytol. 112: 383-388. Gardner, W.K., Barber, D.A., and Parkery, D.G. 1983. The acquisition of phosphorus by Lupinus albus L. HI. The probable mechanism by which phosphorus movement in the soil/root interface is enhanced. Plant Soil 70: 107-124. Gardner, W.K., and Boundy, K.A. 1983. The acquistion of phosphorus by Lupinus albus L. IV. The effect of interplanting wheat and white lupin on the growth and mineral composition of the two species. Plant Soil 70: 391-402. Gardner, J.M., Chandler, J.L., and Feldmas, A.W. 1984. Growth promotion and inhibition by antibiotic-producing fluorescent pseudomonads on citrus roots. Plant Soil 77: 103-113. 174 Gardner, J.M., Chandler, J.L., and Feldmas, A.W.. 1985. Growth responses and vascular plugging of citrus inoculated with rhizobacteria and xylem-resident bacteria. Plant Soil 86: 333-345. Garland, J.L. 1996a. Analytical approaches to the characterization of samples of microbial communities using patterns of potential C source utilization. Soil Biol. Biochem. 28: 213-221. Garland, J.L. 1996b. Patterns of potential C source utilization by rhizosphere communities. Soil Biol. Biochem. 28: 223-230. Garland, J.L., and Mills, A.L. 1991. Classification and characterization of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization. Appl. Environ. Microbiol. 57: 2351-2359. Gaskins, M.H., Albrecht, S.L., and Hubbell, D.H. 1985. Rhizosphere bacteria and their use to increase plant productivity: a review. Agric. Ecosyst. Environ. 12: 99-116. Gaudin, V., Vrain, T., and Jouanin, L. 1994. Bacterial genes modifying hormonal balances in plants. Plant Physiol. Biochem. 32: 11-29. Gilbert, G.S., Clayton, M.K., Handelsman, J., and Parke, J.L. 1996. Use of cluster and discriminant analyses to compare rhizosphere bacterial communities following biological perturbation. Microb. Ecol. 32: 123-147. Gilbert, G.S., Parke, J.L., Clayton, M.K., and Handelsman, J. 1993. Effects of introduced bacterium on bacterial communities on roots. Ecology 74: 840-854. Glandorf, D.C.M., Peter, L.G.L., van der Sluis, I., Bakker, P.A.H., and Schippers, B. 1993. Crop specificity of rhizosphere pseudomonads and the involvement of root agglutinins. Soil Biol. Biochem. 25: 981-989. Glandorf, D.C.M., van der Sluis, I., Anderson, A.J., Bakker, P.A.H., and Schippers, B. 1994. Agglutination, adherence, and root colonization by fluorescent pseudomonads. Appl. Environ. Microbiol. 60: 1726-1733. Glick, B.R. 1995. The enhancement of plant growth by free-living bacteria. Can. J. Microbiol. 41: 109-117. Glick, B.R., Jacobson, C.B., Schwarze, M.M.K., and Pasternak, J.J. 1994a. Does the enzyme 1-aminocyclopropane-l-carboxylate deaminase play a role in plant growh-promotion by Pseudomonasputida GR12-2? In: Improving plant productivity with rhizosphere bacteria. Ryder, M.H., Stephens P.M., Bowen, G.D.feds.). Glen Osmond: CSIRO Division of Soils. pp.150-152. Glick, B.R., Jacobson, C.B., Schwarze, M.M.K., and Pasternak, J.J. 1994b. 1-aminocyclopropane-l-carboxylate deaminase mutants of the plant growth-promoting rhizobacterium Pseudomonas putida GR12-2 do not stimulate canola root elongation. Can. J. Microbiol. 40: 911-915. 175 Grimes, H.D., and Mount, M.S. 1984. Influence of Pseudomonas putida on nodulation of Phaseolus vulgaris. Soil Biol. Biochem. 16: 27-30. Haack, S.K., Garchow, H., Klug, M.J., and Forney, L.J. 1995. Analysis of factors affecting the accuracy, reproducibility, and interpretation of microbial community carbon source utilization patterns. Appl. Environ. Microbiol. 61: 1458-1468. Hagedorn, C , Gould, W.D., and Bardinelli, T.R. 1989. Rhizobacteria of cotton and their repression of seedling disease pathogens. Appl. Environ. Microbiol. 55: 2793-2797. Hagedorn, C , Gould, W.D., and Bardinelli, T.R. 1993. Field evaluations of bacterial inoculants to control seedling disease pathogens on cotton. Plant Dis. 77: 278-282. Hall, T. J., and Davis, W.E.E. 1990. Survival of Bacillus subtilis in silver and sugar maple seedlings over a two year period. Plant Dis. 74: 608-609. Hallaksela, A-M., Vaisanen, O., and Salkinoja-Salonen, M. 1991. Identification of Bacillus species isolated from Picea abies by physiological tests, phage typing and fatty acid analysis. Scand. J. For. Res. 6: 365-377. Harley, J.L., and Smith, S.E. 1983. Mycorrhizal symbioses. Academic Press, London. 483pp. Hebbar, K.P., Davey, A.G., and Dart, P.J. 1992. Rhizobacteria of maize antagonistic to Fusarium moniliforme, a soil-borne fungal pathogen: isolation and identification. Soil Biol. Biochem. 24: 979-987. Hedges, R.W., and Messens, E. 1990. Genetic aspects of rhizosphere interactions. In: The rhizosphere. Lynch, J.M. (ed.). John Wiley & Sons, Inc. New York. NY. pp. 129-176. Hemming, B.C. 1986. Microbial-iron interactions in the plant rhizosphere. An overview. J. Plant Nutr. 9: 505-521. Hiltner, L. 1904. Ueber neuere Erfahrungen und Probleme auf dem Gebiet der Boden-bakteriologie und unter besonderer Berucksichtigung der Griindingung und Brache. Arb. Deut. Landw. Ges. 98: 59-78. Hines, F.D., and Long, J.N. 1986. First- and second-year survival of containerized Engelmann spruce in relation to initial seedling size. Can. J. For. Res. 16: 668-670. Hoffland, E., Hakulinen, J., and van Pelt, J.A. 1996. Comparison of systemic resistance induced by avirulent and nonpathogenic Pseudomonas species. Phytopathology 86: 757-762. Hdflich, G., Wiehe, W., and Kuhn, G. 1994. Plant growth stimulation by inoculation with symbiotic and associative rhizosphere microorganisms. Experienta 50: 897-905. Holl, F.B., and Chanway, CP. 1992. Rhizosphere colonization and seedling growth promotion of lodgepole pine by Bacilluspolymyxa. Can. J. Microbiol. 38: 303-308. 176 Holl, F.B., Chanway, CP., Turkington, R., and Radley, R.A. 1988. Response of crested wheatgrass, perennial ryegrass and white clover to inoculation with Bacillus polymyxa. Soil Biol. Biochem. 20: 19-24. Hong, Y., Glick, B.R., and Pasternak, J.J. 1991a. Plant - microbial interaction under gnotobiotic conditions: a scanning electron microscope study. Curr. Microbiol. 23: 111-114. Hong, Y., Pasternak, J.J., and Gkck, B.R. 1991b. Biological consequences of plasmid transformation of the plant growth promoting rhizobacterium Pseudomonas putida GR12-2. Can. J. Microbiol. 37: 796-799. Huang, J.-S. 1986. Ultrastructure of bacterial penetration in plants. Ann. Rev. Phytopathol. 24: 141-157. Hunt, R. 1978. Plant growth analysis. Edward Arnold Limited, London. 67p. Hurek, T., Reinhold-Hurek, B., van Montagu, M., and Kellenbaerger, E. 1994. Root colonization and systemic spreading of Azoarcus sp. Strain BH72 in grasses. J. Bacteriol. 176: 1913-1923. Ingestad, T. 1979. Nitrogen stress in birch seedlings. II. N, K, P, Ca and Mg nutrition. Physiol. Plant. 45: 149-157. Janisiewicz, W. 1996. Ecological diversity, niche overlap, and coexistence of antagonists used in developing mixtures for biocontrol of postharvest diseases of apples. Phytopathology 86: 473-479. Juhnke, M.E., Mathre, D.E., and Sands, D.C. 1987. Identification and characterization of rhizosphere-competent bacteria of wheat. Appl. Environ. Microbiol. 53: 2793-2799. Juhnke, M.E., Mathre, D.E., and Sands, D.C. 1989. Relationship between bacterial seed inoculum density and rhizosphere colonization of spring wheat. Soil Biol. Biochem. 21: 591-595. Kapulnik, Y., Okon, Y., and Henis, Y. 1985. Changes in root morphology of wheat caused by Azospirillum inoculation. Can. J. Microbiol. 31: 881-887. Katznelson, H. 1946. The rhizosphere effect of mangels on certain groups of microorganisms. Soil Sci. 62: 343-354. Kennedy, A.C. and Smith, K.L. 1995. Soil microbial diversity and the sustainability of agricultural soils. Plant Soil 170: 75-86. Khalil, M.A.K. 1986. Variation in seed quality and some juvenile characteristics of white spruce. Silvae Genetica 3 5: 78-85. Kikumoto, T., and Sakamoto, M. 1967. Ecological studies on the soft rot bacteria of vagetables. Annu. Phytopathol. Soc. Jpn. 33: 181-186. Klement, Z., Rudolph, K., and Sands, D.C. 1990. Methods in phytobacteriology. Akademiai Kiado, Budapest. 177 Klinka, K., Feller, M.C., Green, R.N., Meidinger, D.V., Pojar, J. and Worrall, J. 1990. Ecological principles: applications. In: Regenerating British Columbia's Forests. Lavender, D.P., Parish, R., Johnson, C M . , Montgomery, G., Vyse, A., Willis, R.A., and Winston, D. (eds.). UBC Press, Vancouver, B.C. pp.55-72. Klinka, K., Green, R.N., Trowbridge, R.L., and Lowe, L.E. 1981. Taxonomic Classification of Humus Forms in Ecosystems of British Columbia. Land Management Report No. 8. British Columbia Ministry of Forests. Victoria, B.C. 54pp. Kloepper, J.W. 1993. Plant growth-promoting rhizobacteria as biological control agents. In: Soil microbial ecology-applications in agricultural and environmental management. Metting, F.B., Jr. (ed.). Marcel Dekker, New York. NY. pp.255-274 Kloepper, J.W., and Beauchamp, C.J. 1992. A review of issues related to measuring colonization of plant roots by bacteria. Can. J. Microbiol. 38: 1219-1232. Kloepper, J.W., and Schroth, M.N. 1978. Plant growth promoting rhizobacteria on radishes. In: Proc. Int. Conf. Plant Pathog. Bact. 4th. Vol. 2. Angers, France, pp. 879-882. Kloepper, J.W., and Schroth, M.N.1981. Relationship of in vitro antibiosis of plant growth-promoting rhizobacteria to plant growth and the displacement of root microflora. Phytopathology 71:1020-1024. Kloepper, J.W., Leong, L., Teintze, M., and Schroth, M.N. 1980a. Enhanced plant growth by siderophores produced by PGPR. Nature 286: 885-886. Kloepper, J.W., Schroth, M.N., and Miller, T.D. 1980b. Effects of rhizosphere colonization by plant growth-promoting rhizobacteria on potato plant development and yield. Phytopathology 70: 1078-1082. Kloepper, J.W., Lifshitz, R, and Zablotowicz, R.M. 1989. Free-living bacterial inocula for enhancing crop productivity. TJJBTECH. 7: 39-44. Kloepper, J.W., Mclnroy, J.A., and Bowen, K.L. 1992a. Comparative identificatin by fatty acid analysis of soil, rhizosphere, and geocarposphere bacteria of peanut (Arachis hypogaea L.). Plant Soil 139: 85-90. Kloepper, J.W., Schippers, B., and Bakker, P.A.H.M. 1992b. Proposed elimination of the term endorhizosphere. Phytopathology 82: 726-727. Kluepfel, D.A. 1993. The behavior and tracking of bacteria in the rhizosphere. Ann. Rev. Phytopathol. 31: 441-472. Kropp, B.R., and Langlois, C.G. 1990. Ectomycorrhizae in reforestation. Can.J. For. Res. 20: 438-451. Kropp, B.R., Castellano, M.A., and Trappe, J.M. 1985. Performance of outplanted western hemlock (Tsuga hetrophylla (Raf) Sarg.) seedlings inoculated with Cenococcum geophilum. Tree Plant. Notes 36(4): 13-16. 178 Kumar, B.S.D., and Dube, H.C. 1992. Seed bacterization with a fluorescent Pseudomonas for enhanced plant growth, yield and disease control. Soil Biol. Biochem. 24: 539-542. Lalande, R., Bissonnette, N., Coutlee, D., and Antoun, H. 1989. Identification of rhizobacteria from maize and determination of their plant-growth promoting potential. Plant Soil 115:7-11. Lamb, T.G., Tonkyn, D.W., and Kluepfel, D.A. 1996. Movement of Pseudomonas aureofaciens from the rhizosphere to aerial plant tissue. Can. J. Microbiol. 42: 1112-1120. Law, R., and Lewis, D.H. 1983. Biotic environments and the maintenance of sex - some evidence from mutualistic symbioses. Biol. J. Linn. Soc. 20: 249-276. Levanony, H., Bashan, Y., Romano, B., and Klein, E. 1989. Ultrastructural localization and identification of Azospirillum brasilence Cd on and within wheat root by immunogold labelling. Plant Soil 117: 207-218. Li, C.Y., and Hung, L.L. 1987. Nitrogen-fixing (acetylene-reducing) bacteria associated with ectomycorrhizae of Douglas-fir. Plant Soil 98: 425-428. Li, C.Y., Massicote, H.B., and Moore, L.V.H. 1992. Nitrogen-fixing Bacillus sp. associated with Douglas-fir tuberculate ectomycorrhizae. Plant Soil. 140: 35-40. Li, D.-M., and Alexander, M. 1988. Co-inoculation with antibiotic-producing bacteria to increase colonization and nodulation by rhizobia. Plant Soil 108: 211-219. Li, R.P., and MacRae, LC. 1991. Specific association of diazotrophic acetobacters with sugarcane. Soil Biol. Biochem. 23: 999-1002. Li, R.P., and MacRae, I.C. 1992. Specific identification and enumeration of Acetobacter diazotrophicus in sugarcane. Soil Biol. Biochem. 24: 413-419. Lin, CP., Chen, T.A., Wells, J.M., and van der Zwet, T. 1987. Identification and detection of Erwinia amylovora with monoclonal antibodies. Phytopathology 77: 376-380. Linderman, R. G. 1988. Mycorrhizal interactions with the rhizosphere microflora; the mycorrhizosphere effect. Phytopathology 78: 366-371. Liu, L., Kloepper, J.W., and Tuzun, S. 1995. Induction of systemic resistance in cucumber against Fusarium wilt by plant growth-promoting rhizobacteria. Phytopathology 85: 695-698. Long, S.R. 1989. Rhizobium-Legpme nodulation: Life together in the underworld. Cell 56: 203-214. Loopstra, E.M., Shaw, C.G., and Sidle, R.C. 1988. Ectomycorrhizal inoculation fails to improve performance of Sitka spruce seedlings on clearcuts in southeastern Alaska. West. J. Appl. For. 3: 110-112. Ludwig, J.A., and Reynolds, J.F. 1988. Statistical ecology, a primer on methods and computing. John Wiley & Sons, New York, NY. 337pp. 179 Lynch, J.M. 1990. The rhizosphere. John Wiley & Sons, Inc. New York. NY. 458pp. Lynch, J.M. 1994. The rhizosphere — form and function. Appl. Soil. Ecol. 1:193-198. Mahaffee, W.F., Kloepper, J.W., van Vuurde, J.W.L., and van den Brink, M. 1994. Endophytic colonization of Phaseolus valgaris by Pseudomonas fluorescens strain 89B-27 and Enterobacter asburiae strain JM 22. In: Improving plant productivity with rhizosphere bacteria. Ryder, M.H., Stephens, P.M., and Bowen G.D. (eds.). CSISRO Division of Soils, Glen Osmond, Australia. 180p. Manion, P.D. 1991. Tree disease concepts (Second ed.). Prentice-Hall, Englewood Cliffs, NJ. 402pp. Martens, D.A. and Frankenberger, W.T. Jr. 1993. Metabolism of trptophan in soil. Soil Biol. Biochem. 25: 1679-1687. Martin, J.K. 1971. Influence of plant species and age on the rhizosphere microflora. Australian J. Biol. Sci. 24: 1143-1150. Mavingui, P., Laguerre, G., Berge, O., and Heulin, T. 1992. Genotypic and phenotypic variability of Bacillus polymyxa in soil and in the rhizosphere of wheat. Appl. Environ. Microbiol. 58: 1894-1903. McAfee, B.J., and J.A Fortin. 1988. Comparative effects of the soil microflora on ectomycorrhizal inoculation of conifer seedlings. New Phytol. 108: 443-449. Mclnroy, J.A., and Kloepper, J.W. 1995. Population dynamics of endophytic bacteria in field-grown sweet corn and cotten. Can. J. Microbiol. 41: 895-901. Mclnroy, J.A., Musson, G., Wei, G., and Kloepper, J.W. 1996. Masking of antibiotic-resistance upon recovery of endophytic bacteria. Plant Soil 186: 213-218. McLaughlin, R.J., and Chen, T.A. 1990. ELISA methods for plant pathogenic prokaryotes. In: Serological methods for detection and identification of viral and bacterial plant pathogens. Hampton, R. O., Ball, E. M. and De Boer, S. H. (eds.). American Phytopathological Society, St. Paul, MN. pp. 197-204. McMinn, R.G., and Hedin, LB. 1990. Site preparation: mechanical and manual. In: Regenerating British Columbia's Forests. Lavender, D.P., Parish, R., Johnson, C M . , Montgomery, G., Vyse, A , Willis, R.A., and Winston, D. (eds.). UBC Press, Vancouver, B.C. pp. 150-163. Merriman, P.R., Price, R.D., and Baker, K.F. 1974. The effect of inoculation of seed with antagonists ofRhizoctonia solani on the growth of wheat. Aust. J. Agric. Res. 25: 213-218. Miller, M., McGonigle, T., and Addy, H. 1992. An economic approach to evaluate the role of mycorrhizas in managed ecosystems. Plant Soil 159: 27-35. Misaghi, I.J., and Donndelinger, C.R. 1990. Endophytic bacteria in symptom-free cotton plants. Phytopathology 80: 808-811. 180 Molina, R., Massicotte, H., and Trappe, J. M., 1992. Specificity phenomena in mycorrhizal symbioses: community-ecological consequences and practical implications. In: Mycorrhizal functioning: an Integrative plant-fungal process. Allen, M. F. (Ed.). Chapman and Hall, New York, NY, pp.357-423. Mytton, L.R., and Hughes, D.M. 1984. Inoculation of white clover with different strains of Rhizobium trifolii on mineral hill soil. J. Agric. Sci. 102: 455-459. Mytton, L.R., and Livesly, C.J. 1983. Specific and general effectiveness of Rhizobium trifolii populations from two different agricultural locations. Plant Soil 73: 299-305. Neal, J.R., Jr. and Bollen, W.B. 1964 Rhizosphere microflora associated with mycorrhizae of Douglas fir. Can. J. Microbiol. 10, 259-265. Neal, J.R., Jr., Atkinson, T.G., and Larson, R.I. 1970. Changes in the rhizosphere microflora of spring wheat induced by disomic substitution of a chromosome. Can. J. Microbiol. 16: 153-158. Neal, J.R., Jr. Larson, R.I., and Atkinson, T.G. 1973. Changes in rhizosphere populations of selected physiological groups of bacteria related to substitution of specific pairs of chromosomes in spring wheat. Plant Soil 39: 209-212. Nehl, D.B., Allen, S.J., and Brown, J.F. 1996. Deleterious rhizosphere bacteria: an integrating perspective. Appl. Soil Ecol.: 1-20. Newton, M., and Comeau, P.G. 1990. Control of competing vegetation. In: Regenerating British Columbia's Forests. Lavender, D.P., Parish, R., Johnson, C M . , Montgomery, G., Vyse, A., Willis, R.A., and Winston, D. (eds.). UBC Press, Vancouver, B.C. pp.256-265. Nilsson, U., and Grander, G. 1995. Effects of regeneration methods on drought damage to newly planted Norway spruce seedlings. Can. J. For. Res. 25: 790-802. O'Neill, G.A., Chanway, CP., Axelrood, P.E., Radley, R.A., and Holl, F.B. 1992. An assessment of spruce growth response specificity after inoculation with coexistent rhizosphere bacteria. Can. J. Bot. 70: 2347-2353. O'Sullivan, D.J., and O'Gara, F. 1992. Traits of fluorescent Pseudomonas spp. involved in suppression of plant root pathogens. Microbiol, reviews 56: 662-676. Ohno, A., Ano, T., and Shoda, M. 1992. Production of a lipopeptide antibiotic surfactin with recombinant Bacillus subtilis. Biotechnol. letters 14: 1165-1168. Owens, J.N., and Molder, M. 1984a. The reproductive cycle of interior spruce. Ministry of Forests, Province of British Columbia, Victoria, B.C. 31pp. Owens, J.N., and Molder, M. 1984b. The reproductive cycle of lodgepole pine. Ministry of Forests, Province of British Columbia, Victoria, B.C. 30pp. Page, R.H., Miller, R.H., and Keeney, D.R. 1982. Methods of soil analysis (Part 2). Agronomy Society of America Inc., and Soil Science Society of America Inc., Madison, WI. 1159pp. 181 Parker, A.K., and Dangerfield, J.A. 1975. Influence of bacterial inoculations on growth of containerized Douglas-fir seedlings. Can. For. Service Bulletin. Bi-monthly research notes 31:13-14. Patriquin, D.G., and Ddbereiner, J. 1978. Light microscopy observations of tetrazolium-reducing bacteria in the endorhizosphere of maize and other grasses in Brazil. Can J. Microbiol. 24: 734-742. Paul, E.A., and Clark, F.E. 1989. Soil Microbiology and Biochemistry. Academic Press, Inc. London. 275pp. Perry, D.A., Amaranthus, M.P., Borchers, J.G., Borchers, S.L., andBrainerd, R.E. 1989. Bootstrapping in ecosystems. Bioscience 39: 230-237. Perry, D.A., Molina, R., and Amaranthus, M.P. 1987. Mycorrhizae, mycorrhizospheres, and reforestation: current knowledge and research needs. Can. J. For. Res. 17: 929-940. Pfender, W.F., Fieland, V.P., Ganio, L.M., and Seidler, R.J. 1996. Microbial community structure and activity in wheat straw after inoculation with biological control organims. Appl. Soil Ecol. 3: 69-78. Pickup, R.W. 1991. Development of molecular methods for the detection of specific bacteria in the environment. J. Gen. Microbiol. 137: 1009-1019. Pleban, S., Ingel, F., and Chet, I. 1995. Control of Rhizoctonia solani and Sclerotium rolfsii in the greenhouse using endophytic Bacillus spp. Euro. J. Plant Path. 101: 665-672. Pokojska-Burdziej, A. 1982. The effect of microorganisms, microbial metabolites and plant growth regulators on the growth of pine seedlings (Pinus sylvestris L.) Pol. J. Soil Sci. 15: 137-143. Powell, P.E., Cline, G.R., Reid, C.P.P., and Szaniszlo, P.J. 1980. Occurrence of hydroxamate siderophore iron chelators in soils. Nature 287: 833-834. Quadt-Hallmann, A., and Kloepper, J.W. 1996. Immunological detection and localization of the cotten endophyte Enterobacter asburiae JM22 in different plant species. Can. J. Microbiol. 42: 1144-1154. Reddy, M.S., and Rahe, J.E. 1989. Growth effects associated with seed bacterization not correlated with populations of Bacillus subtilis inoculant in onion seedling rhizospheres. Soil Biol. Biochem. 21: 373-378. Reid, C.P.P., and Mexal, J.G. 1977. Water stress effects on root exudation by lodgepole pine. Soil Biol. Biochem. 9: 417-422. Rennie, R.J. 1981. A single medium for the isolation of acetylene-reducing (dinitrogen-fixing) bacteria from soils. Can. J. Microbiol. 27:8-14. Richard, B.N. 1987. The Microbiology of Terrestrial Ecosystems. John Wiley & Sons, Inc., New York. NY. 399pp. 182 Roszak, D.B., and Colwell, R.R. 1987. Survival strategies of bacteria in the natural environment. Microbiol. Reviews 51: 365-379. Rouatt, J.W. and Katznelson, H. 1961. A study of bacteria on the root surface and in the rhizosphere soil of crop plants. J. Appl. Bact. 24, 164-171. Rovira, A.D. 1956. Plant root excretions in relation to the rhizosphere effect. I. The nature of root exudate from oats and peas. Plant Soil 7: 178-194. Rovira, A.D., and Davey, C.B. 1974. Biology of the rhizosphere. In: The Plant Root and Its Environment: Proceedings, Carson, E.W. (ed.). Univ. Virginia Press. Charlottesville, pp. 153-204. Ruppel, S., Hecht-Buchholz, C , Remus, R., Ortmann, U., and Schmelzer, R. 1992. Settlement of the diazotrophic, phytoeffective bacterial strain Pantoea agglomerans on and within winter wheat: An investigation using ELISA and transmission electron microscopy. Plant Soil 145: 261-273. Ryan, C , and Jagendorf, A. 1995. Self defense by plants. In: Self-defense by plants: induction and signalling pathways. Ryan, C.A., Lamb, C J. , Jagendorf, A.T., and Klattukudy, P.E. (eds.). Proc. Natl. Acad. Soc. U.S.A. p.4075. Ryglewicz, P.T., and Andersen, CP. 1994. Mycorrhizae alter quality and quantity of carbon allocated below ground. Nature 369: 58-60. Salisbury, F.B., and Ross, C.W. 1992. Plant Physiology (4th edition). Wadsworth Publishing Co., Belmont, CA. 682pp. Sardi, P., Saracchi, M., Quaroni, S., Petrolini, B., Borgonovi, G.E., and Merli, S. 1992. Isolation of endophytic Streptomyces strains from surface-sterilized roots. Appl. Environ. Microbiol. 58: 2691-2693. Schippers, B., Bakker, A.W., and Bakker, P.A.H.M. 1987. Interactions of deleterious and beneficial rhizosphere microorganisms and the effect of cropping practices. Annu. Rev. Phytopathol. 25: 339-358. Schloter, M., APmus, B., and Hartmann, A. 1995. The use of immunological methods to detect and identify bacteria in the environment. Biotech. Adv. 13: 75-90. Schloter, M., Borlinghaus, R., Bode, W., and Hartmann, A. 1993. Direct identification, and localization of Azospirillum in the rhizosphere of wheat using fluorescence-labelled monoclonal antibodies and confocal scanning laser microscopy. J. Microscopy 177: 173-177. Schroth, M.N., and Hancock, J. 1982. Disease suppressive soil and root-colonizing bacteria. Science 216: 376-1381. Schroth, M.N., and Weinhold, A.R. 1986. Root-colonizing bacteria and plant health. Hortscience 21: 1295-1298. 183 Shishido, M., Massicotte, H.B., and Chanway, CP. 1993. Influence of root endophytic and rhizosphere bacteria on growth and mycorrhizal formation of hybrid spruce and lodgepole pine seedlings. 9th N. Am. Conf. Mycor. Guelph, Ontario. Shishido, M., Loeb, B.M., and Chanway, CP. 1995. Rhizosphere and internal root colonization of lodgepole pine by two seedling growth-promoting Bacillus strains originating from different root microsites. Can J. Microbiol. 41: 701-713. Shishido, M., Massicotte, H.B., and Chanway, CP. 1996a. Effect of plant growth promoting Bacillus strains on pine and spruce seedling growth and mycorrhizal infection. Ann. Bot. 77: 433-441. Shishido, M., Petersen, D.J., Massicotte, H.B., and Chanway, CP. 1996b. Pine and spruce seedling growth and mycorrhizal infection after inoculation with plant growth promoting Pseudomonas strains. FEMS Microbiol. Ecol. 21: 109-119. Siala, A., Hill, L.R., and Gray, T.R.G. 1974. Populations of spore-forming bacteria in an acid forest soil, with special reference to Bacillus subtilis. J. Gen. Microbiol. 81: 183-190. Sinclair, W.A., and Marx, D.H. 1982. Evaluation of plant response to inoculation. A. Host variables. In: Methods and principles of mycorrhizal research. Schenck, N.C.Ced.). The American Phytophathological Society, St. Paul, MN. pp. 165-174. Smirnoff, N., Todd, P., and Stewart, G.R. 1984. The occurrence of nitrate reduction in the leaves of woody plants. Ann. Bot. 54: 363-374. Smith, H.H., and Townsend, CO. 1907. A plant tumor of bacterial origin. Science 25: 671-673. Smith, W.H. 1976. Character and significance of forest tree root exudates. Ecology 57: 324-331. Sprent, J.L, and de Faria, S.M. 1988. Mechanisms of infection of plants by nitrogen fixing organisms. Plant Soil 110: 157-165. Srinivasan, M., Petersen, D.J., and Holl, F.B. 1996. Influence of indoleacetic-acid-producing Bacillus isolates on the nodulation of Phaseolus vulgaris by Rhizobium etli under gnotobiotic conditions. Can. J. Microbiol. 42: 1006-1014. Steffan, R.J., and Atlas, R.M. 1988. DNA amplification to enhance detection of genetically engineered bacteria in environmental samples. Appl. Environ. Microbiol. 54: 2185-2191. Starr, M.P. Chatterjee, A.K., Starr, P.B., and Buchanan, G.E. 1977. Enzymatic degradation of polygalacturonic acid by Yersinia and Klebsiella species in relation to clinical laboratory procedures. J. Clin. Microbiol. 6: 379-386. Starr, M.P., Stolp, H., Triiper, H.G., Balows, A , Schlegel, H.G. 1981. The prokaryotes, a handbook on habitats, isolation, and identification of bacteria. Springer-Verlag, Berlin Heidelberg. 2284p. Stephens, P.M. Crowley, J.J., and O'Connel, C. 1993. Selection of pseudomonad strains inhibiting Pytium ultimum on sugarbeet seeds in soil. Soil Biol. Biochem. 25: 1283-1288. 184 Sumner, M.E. 1990. Crop responses to Azospirillum inoculation. Adv. Soil Sci. 12: 53-123. Suslow, T.V., and Schroth, M.N. 1982. Rhizobacteria of sugar beets: effect of seed application and root colonization on yield. Phytopathology 72: 199-206. Sutton, R.F. 1980. Planting stock quality, root growth capacity, and field performance of three boreal conifers. N.Z. J. For. Sci. 10: 54-71. Sutton, R.F., and Tinus, R.W. 1983. Root and root system terminology. For. Sci. Monogr. 24. The Society of American Foresters. Washington, D.C. 137pp. Thomashow, L.S., and Weller, D.M. 1988. Role of a phenazine antibiotic in suppression of Gaeumannomycesgraminis var. tritici. J. Bacterid. 170: 3499-3508. Tien, T.M. Gaskins,M.H., and Hubbell, D.H. 1979. Plant growth substances produced by Azospirillum brasilense and their effect on the growth of pearl millet. Appl. Environ. Microbiol. 37: 1016-1024. Tinker, P.B. 1984. The role of the microorganisms in mediating and facilitating the uptake of plant nutrients from soil. Plant Soil 76: 77-91. Van den Driessche, R. 1992. Changes in drought resistance and root growth capacity of container seedlings in response to nursery drought, nitrogen, and potassium treatments. Can. J. For. Res. 22: 740-749. Van ElsasJ.D., Dijkstra, A.F., Govaert, J.M., and van Veen, J.A. 1986. Survival of Pseudomonas fluorescens and Bacillus subfiles introduced into two soils of different texture in field microplots. FEMS Microbiol. Ecol. 38: 151-160. Van Peer, R., and Schippers, B. 1989. Plant growth responses to bacterization with selected Pseudomonas spp. strains and rhizosphere microbial development in hydroponic cultures. Can. J. Microbiol. 35: 456-463. Van Peer, R., Punte, H.L., De Weger, L.A., and Schippers, B. 1990. Characterization of root surface and endorhizosphere pseudomonads in relation to their colonization of roots. Appl. Environ. Microbiol. 56: 2462-2470. Vogt, K.A., Publicover, D.A., and Vogt, D.J. 1991. A critique of the role of ectomycorrhizas in forest ecology. Agric. Ecosyst. Environ. 35: 171-190. Vyse, A. 1981. Growth of young spruce plantations in Interior British Columbia. For. Chron. 57: 174-180. Wei, G., Kloepper, J.W., and Tuzun, S. 1991. Induction of systemic resistance of cucumber to Colletotrichum orbiculare by selected strains of plant growth-promoting rhizobacteria. Phytopathology 81: 1508-1512. Wei, G., Kloepper, J.W., and Tuzun, S. 1996. Induced systemic resistance to cucumber diseases and increased plant growth-promoting rhizobacteria under field conditions. Phytopathology 86: 221-224. 185 Wennstrom, A. 1994. Endophyte: the misuse of an old term. Oikos 71: 535-536. Weller, D.M. 1988. Biological control of soilborne plant pathogens in the rhizosphere with bacteria. Annu. Rev. Phytopathol. 26: 379-407. Weller, D.M., Howie, W.J., and Cook, R.J. 1988. Relationship between in vitro inhibition of Gaeumannomyces graminis var. triad and suppression of take-all of wheat by fluorescent pseudomonads. Phytopathology 78: 1094-1100. Whipps, J.M. 1990. Carbon economy. In: The rhizosphere. Lynch, J.M. (ed.). John Wiley & Sons, Inc. New York. NY. pp.59-97. Whipps, J.M., and Lynch, J.M. 1986. The influence of the rhizosphere on crop productivity. Adv. Microbiol. Ecol. 9: 187-244. Williamson, G.B. 1990. Allelopathy, Koch's postulates, and the neck riddle. In: Perspectives on plant competition. Grace, J.B., and Tilman, D. (eds.). Academic Press, Inc. San Diego, CA. pp. 143-162. Wilson, D. 1995. Endophyte: the evolution of a term, and clarification of its use and definition. Oikos 73:274-276. Winans, S.C. 1992. Two-way chemical signaling in Agrobacterium - plant interactions. Microbiol. Rev. 56: 12-31. Yahalom, E., Okon, Y., and Dovrat, A. 1990. Possible mode of action of Azospirillum brasilense strain Cd on the root morphology and nodule formation in burr medic (Medicago polymorpha). Can. J. Microbiol. 36: 10-14. Yuen, G.Y., and Schroth, M.N. 1986. Interactions of Pseudomonas fluorescens strain E6 with ornamental plants and its effect on the composition of root-colonizing microflora. Phytopathology 76: 176-180. Yuen, G.Y., Schroth, M.N., and McCain, A.H. 1985. Reduction of Fusarium wilt of carnation with suppressive soils and antagonistic bacteria. Plant Dis. 69: 1071-75. Zak, J.C., Willing, M.R., Moorhead, D.L., and Wildman, H.G. 1994. Functional diversity of microbial communities: a quantitative approach. Soil Biol. Biochem. 26: 1101-1108. Zar, J.H. 1984. Biostatistical analysis (second edition). Prentice-Hall, Inc., Englewood Cliffs, NJ. 718pp. 186 APPENDIX 1 Habitat characteristics of isolated PGPR strains 187 Table A l . l Habitat characteristics of isolated PGPR strains. Strain Habitat characteristics Sm3 Rhizosphere of a spruce seedling (<3 years old) from Mackenzie spruce stand1. Sm3R Spontaneous rifamycin resistant mutant of Sm3. Sm3RN Spontaneous rifamycin and nalidixic acid resistant mutant of Sm3. Ss2 Rhizosplane of a spruce seedling (<3 years old) from Salmon Arm spruce stand. Ss2R Spontaneous rifamycin resistant mutant of Ss2. Ss2RN Spontaneous rifamycin and nalidixic acid resistant mutant of Ss2. Sw5 Rhizosphere of a spruce seedling (<3 years old) from Williams Lake spruce stand. Sw5R Spontaneous rifamycin resistant mutant of Sw5. Sw5RN Spontaneous rifamycin and nalidixic acid resistant mutant of Sw5. 2 L6 Rhizosphere soil from a permanent pasture containing perennial ryegrass and white clover roots. L6-16R Spontaneous rifamycin resistant mutant of L6. Pw2 Homogenate of surface sterilized pine seedling (<3 years old) root tissue from Williams Lake pine stand. Pw2R Spontaneous rifamycin resistant mutant of Pw2. 3 S20 Rhizosphere of a spruce seedling (<3 years old) from Salmon Arm spruce stand. S20R Spontaneous rifamycin resistant mutant of S20. See Chapter 2 for stand locations. 2 Chanway etal. 1990. 3 O'Neill etal. 1992. 188 A P P E N D I X 2 Bacterial species identification using fatty acid methyl-ester ( G C - F A M E ) and Biolog™ systems 189 Table A2.1 Species identification of the 6 bacterial strains used in this thesis according to gas chromatography of cellular fatty acid methyl-esters (GC-FAME). Strain Species identification Similarity index Database library Sm3 Pseudomonas chlororaphis 0.745 TSBA 1 Rev. 3.80 Pseudomonas putida A 0.723 CLIN Rev. 3.80 Pseudomonas fluorescens 0.684 CLIN Rev. 3.80 Sm3R Pseudomonas chlororaphis 0.872 TSBARev. 3.80 Pseudomonas putida B 0.676 TSBARev. 3.80 Pseudomonas fluorescens 0.676 CLIN Rev. 3.80 Sm3RN Pseudomonas chlororaphis 0.906 TSBARev. 3.80 Pseudomonas putida A 0.711 CLIN Rev. 3.80 Pseudomonas fluorescens 0.680 CLIN Rev. 3.80 Ss2 Pseudomonas chlororaphis 0.571 TSBARev. 3.80 Pseudomonas fluorescens C 0.477 TSBARev. 3.80 Pseudomonas putida B 0.574 CLIN Rev. 3.80 Ss2R Pseudomonas fluorescens C 0.907 TSBARev. 3.80 Pseudomonas chlororaphis 0.863 TSBARev. 3.80 Pseudomonas putida B 0.529 TSBARev. 3.80 Ss2RN Pseudomonas fluorescens B 0.913 TSBARev. 3.80 Pseudomonas chlororaphis 0.863 TSBARev. 3.80 Pseudomonas putida B 0.583 TSBARev. 3.80 Sw5 Pseudomonas chlororaphis 0.588 TSBARev. 3.80 Pseudomonas putida B 0.532 CLIN Rev. 3.80 Pseudomonas fluorescens 0.496 CLIN Rev. 3.80 Sw5R Pseudomonas chlororaphis 0.814 TSBARev. 3.80 Pseudomonas fluorescens C 0.766 TSBARev. 3.80 Pseudomonas putida B 0.703 TSBARev. 3.80 Sw5RN Pseudomonas chlororaphis 0.530 TSBARev. 3.80 Pseudomonas fluorescens C 0.458 TSBARev. 3.80 Pseudomonas putida B 0.488 CLIN Rev. 3.80 190 Table A2.1 Continued. Strain Species identification Similarity index Database library L6 Bacillus polymyxa 0.842 TSB A Rev. 3.30 L6-16R Bacillus polymyxa 0.693 TSB A Rev. 3.80 Pw2 Bacillus pabuli 0.254 TSBARev. 3.30 Pw2R No match - TSBARev. 3.30 S20 Bacillus polymyxa 0.326 TSBARev. 3.30 S20R Bacillus polymyxa 0.564 TSBARev. 3.80 Environmental bacteria library. Clinical bacteria library. 191 Table A2.2 Species identification of the 6 bacterial strains used in this thesis according to Biolog™. Strain Species identification Similarity index (Bio-number of Microlog database) Database library Sm3 Pseudomonas fluorescens B 0.762 (1326-2547-1123-7776-1775-7777-5577-7154) Microlog 2, 3.01 Sm3R Pseudomonas fluorescens G 0.777 (0326-2547-1122-7776-1775-7775-1177-7050) Microlog 2, 3.01 Sm3RN Pseudomonas fluorescens G 0.764 (0326-2547-1122-7776-1775-7775-1177-7050) Microlog 2, 3.50 Ss2 Pseudomonas fluorescens B 0.672 (0326-2503-0063-7662-1777-6775-7557-7150) Microlog 2, 3.01 Ss2R Pseudomonas fluorescens F 0.822 (0326-2547-1163-7666-0575-6775-7177-7154) Microlog 2, 3.01 Ss2RN Pseudomonas fluorescens F 0.577 (0326-2547-1163-7666-0575-6775-7177-7154) Microlog 2, 3.50 Sw5 Pseudomonas fluorescens B 0.774 (3726-2547-3063-7776-1777-7775-7577-7150) Microlog 2, 3.01 Sw5R Pseudomonas fluorescens F 0.687 (0326-2543-1163-7676-1775-5775-7577-7154) Microlog 2, 3.01 Sw5RN Pseudomonas fluorescens F 0.459 (0326-2543-1163-7676-1775-5775-7577-7154) Microlog 2, 3.50 192 Table A2.2 Continued. Strain Species identification Similarity index (Bio-number of Microlog database) Database library L6 Bacillus polymyxa 0.839 (3601 -5676-7736-5663-2200-0310-0001-5610) Microlog 2, 3.01 L6-16R Bacillus polymyxa 0.677 (1700-5656-7736-4623-2200-0210-0001-5010) Microlog 2, 3.50 Pw2 Bacillus polymyxa 0.680 (1701-5676-7566-5737-7301-0110-0001-4212) Microlog 2, 3.01 Pw2R Bacillus polymyxa 0.784 (1741-5776-7766-5667-7301-3332-0203-6616) Microlog 2, 3.50 S20 Bacillus polymyxa 0.653 (1700-5676-7724-5663-2200-0010-0001 -4010) Microlog 2, 3.01 S20R Bacillus polymyxa 0.752 (1721-5656-7734-5623-3200-0300-0001-5210) Microlog 2, 3.50 193 APPENDLX 3 Antibiotic susceptibilities of PGPR strains using Sensi-Discs 194 Table A3.1 Intrinsic antibiotic resistances of PGPR strains Sm3, Ss2, Sw5, L6, Pw2 and S20, and their rifamycin (R) and nalidixic acid (N) resistant derivatives. Strain ' Antibiotics Vm Pn Tc Sm Km Rm NA Sm3 - - + ± + ± ± Sm3R - - + ± + - ± Sm3RN - - + + + - -Ss2 - - + ± + - + Ss2R - - + ± + - ± Ss2RN - - + ± + - -Sw5 - - + ± + ± ± Sw5R - - + ± + - ± Sw5RN - - + ± + - -L6 + + + + + + + L6-16R + + + + + - + Pw2 + + + - + + + Pw2R + + + - + - + S20 + + + + + + + S20R + + + + - + Antibiotic susceptibility test discs (BBL Sensi-Disc). Vm: Vancomycin (30 pg); Pn: Penicillin (10 units); Tc: Tetracycline (30 pg); Sm: Streptomycin (10 pg); Km: Kanamycin (30 pg); Rn: Rifampin (5 pg); and NA: Nalidixic acid (30 pg). '+' indicates growth inhibiton (> 3 mm zone width) on 1/2 TSA after 3 days aerobic incubation at 28°C. '±' and '- ' indicate variable and no inhibition (< 3 mm) zones, respectively. 195 APPENDIX 4 Statistical analyses of data presented in Chapter 196 Table A4.1 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 in Screening 1. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 7964.843 148683.841 26 513 306.344 289.832 1.057 0.389 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 30308.612 16459.011 26 404 127.254 40.740 3.124 0.000 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 18311.689 50610.389 26 404 704.296 125.273 5.622 0.000 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 8164.508 22616.647 26 404 314.019 55.982 5.609 0.000 197 Table A4.2 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 in Screening 1. ANOVA MODEL: Y = u, + INOC, where Y: effect, u,: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 3871.423 48661.500 30 589 129.047 82.617 1.561 0.301 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 3319.787 16458.600 30 589 110.659 27.943 3.960 0.000 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 46847.567 151778.850 30 589 1561.585 257.689 6.059 0.000 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 17232.667 62325.500 30 589 574.422 105.815 5.428 0.000 198 Table A4.3 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 in Screening 1. ANOVA MODEL: Y = \x + INOC, where Y: effect, u.: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 10475.873 226444.521 38 741 275.681 305.593 0.902 0.640 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 4481.172 30265.950 38 741 117.926 40.844 2.887 0.000 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 42211.971 203899.700 38 741 1110.841 275.168 4.037 0.000 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 11035.361 56607.350 38 741 290.404 76.393 3.801 0.000 199 Table A4.4 ANOVA for pine seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 in Screening 1. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC 8482.098 29 292.486 1.2673 0.161 ERROR 131547.129 570 230.784 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC 2419.339 29 83.425 1.597 0.026 ERROR 29497.457 565 52.208 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC 27201.412 29 937.979 1.983 0.002 ERROR 267249.774 565 473.008 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC 31873.402 29 1099.082 2.938 0.000 ERROR 211319.314 565 374.016 200 Table A4.5 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 in the presence of forest floor soil in Screening 2. ANOVA MODEL: Y = \i + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 4411.861 50682.277 7 136 630.266 372.664 1.691 0.116 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 215.529 1907.751 7 131 30.789 314.563 2.114 0.046 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 727.724 5208.764 7 131 103.961 39.762 2.614 0.015 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 71.132 1393.098 7 131 10.162 10.634 0.955 0.466 201 Table A4.6 ANOVA for spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3 without forest floor soil in Screening 2. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 2418.878 44726.906 7 136 345.554 328.874 1.051 0.399 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 177.136 2542.222 7 134 25.305 18.972 1.334 0.239 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 649.997 7047.333 7 134 92.856 52.592 1.766 0.099 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 297.767 1964.091 7 134 42.538 14.657 2.902 0.007 202 T a b l e A4.7 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 in the presence of forest floor soil in Screening 2. ANOVA MODEL: Y = u. + INOC, where Y: effect, p.: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 8129.660 140299.004 18 323 451.647 434.362 1.039 0.414 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 257.096 3549.277 18 318 14.283 11.161 1.279 0.198 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 1293.479 12146.562 18 318 71.859 38.197 1.888 0.017 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 361.613 3411.199 18 318 20.089 10.727 1.872 0.018 203 Table A4.8 ANOVA for spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Ss2 without forest floor soil in Screening 2. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC 3180.188 18 176.677 0.778 0.725 ERROR 73300.313 323 226.935 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC 274.555 18 15.253 1.512 0.083 ERROR 3257.222 323 10.084 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC 1327.801 18 73.766 2.381 0.001 ERROR 10006.222 323 30.979 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio p INOC 82.146 18 4.564 0.488 0.963 ERROR 3022.333 323 9.357 204 Table A4.9 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 in the presence of forest floor soil in Screening 2. ANOVA MODEL: Y = p. + INOC, where Y: effect, \x: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 4252.528 84696.528 13 238 327.117 355.867 0.919 0.5337 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 1025.798 4238.055 13 238 78.906 17.806 4.431 0.000 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 2688.984 11017.000 13 238 206.845 46.289 4.468 0.000 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio p INOC ERROR 404.746 2411.111 13 238 31.134 10.131 3.073 0.000 205 Table A4.10 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sw5 without forest floor soil in Screening 2. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and fNOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 5427.552 82098.327 13 238 417.503 344.951 1.210 0.272 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 350.542 4069.947 13 235 26.964 17.319 1.557 0.099 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 1585.983 13173.173 13 235 121.998 56.056 2.176 0.112 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 379.017 2998.725 13 235 29.155 12.761 2.284 0.007 206 Table A4 . l l ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 in the presence of forest floor soil in Screening 2. ANOVA MODEL: Y = \i + INOC, where Y: effect, p.: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 6494.726 49367.500 7 136 927.818 362.996 2.555 0.017 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 236.989 2779.839 7 133 37.713 20.901 1.804 0.091 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 635.698 7384.954 7 133 90.8145 55.526 1.635 0.131 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 69.125 1601.016 7 133 9.875 12.038 0.820 0.572 207 Table A4.12 ANOVA for spruce seedlings from the Williams Lake seedlot inoculated with Bacillus strain Pw2 without forest floor soil in Screening 2. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. (a) Seedling emergence (arcsine transformed). SOURCE SS DF MS F-ratio P INOC ERROR 870.196 44417.349 7 136 124.313 326.598 0.381 0.912 (b) Shoot height (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 170.391 2685.098 7 133 24.342 20.189 1.206 0.304 (c) Shoot biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 614.777 7711.052 7 133 87.825 57.978 1.514 0.167 (d) Root biomass (untransformed). SOURCE SS DF MS F-ratio P INOC ERROR 262.459 2741.029 7 133 37.494 20.609 1.819 0.089 208 Table A4.13 Descriptive statistics for ecotype effects on spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. (a) Seedling emergence (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3 18 275 58 ns Ss2 18 284 84 ns Sw5 18 272 88 ns Uninoculated control 18 233 83 1 Fisher's protected least significant difference. indicate significant difference the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3 18 37.5 4.6 ns Ss2 18 37.6 4.9 ns Sw5 18 36.9 4.6 ns Uninoculated control 18 35.8 6.5 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 33.2 10.3 ns Ss2 18 31.5 10.3 ns Sw5 18 32.8 11.6 ns Uninoculated control 18 26.1 11.2 (d) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 27.0 7.2 ns Ss2 18 25.2 9.9 ns Sw5 18 23.7 8.5 ns Uninoculated control 18 20.6 9.4 209 Table A4.14 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Mackenzie seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 79818.594 17 4695.211 0.694 0.794 INOC 27403.628 3 9134.543 1.350 0.269 ERROR 345147.392 51 6767.596 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 92.593 1 92.593 0.014 0.907 H0 2 15924.036 1 15924.036 2.353 0.131 HO3 24128.874 1 24128.874 3.565 0.065 ERROR 345147.392 51 6767.596 (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 672.903 17 39.583 1.736 0.066 INOC 36.819 3 12.273 0.538 0.658 ERROR 1162.931 51 22.803 210 Table A4.14 Continued. (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 0.926 1 0.926 0.041 0.841 H0 2 26.694 1 26.694 1.171 0.284 HO3 25.037 1 25.037 1.098 0.300 ERROR 1162.931 51 22.803 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 3389.125 17 199.360 2.183 0.016 INOC 584.819 3 194.940 2.134 0.107 ERROR 4657.931 51 91.332 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 12.676 1 12.676 0.139 0.711 H0 2 455.111 1 455.111 4.983 0.030 HO3 444.083 1 444.083 4.862 0.032 ERROR 4657.931 51 91.332 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1554.069 17 91.416 1.250 0.262 INOC 401.375 3 133.792 1.830 0.153 ERROR 3728.875 51 73.115 211 Table A4.14 Continued. (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 80.083 1 80.083 1.095 0.300 H 0 2 373.778 1 373.778 5.112 0.028 H 0 3 178.898 1 178.898 2.447 0.124 ERROR 3728.875 51 73.115 212 Table A4.15 Descriptive statistics for ecotype effects on spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3 18 346 34 ns Ss2 18 345 27 ns Sw5 18 329 58 ns Uninoculated control 18 345 27 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3 18 38.5 5.0 ns Ss2 18 37.1 6.4 ns Sw5 18 37.9 5.2 ns Uninoculated control 18 35.4 4.3 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 30.7 12.5 ns Ss2 18 30.8 13.1 ns Sw5 18 28.5 9.1 ns Uninoculated control 18 28.6 7.3 (d) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 27.2 9.1 ns Ss2 18 28.6 13.2 ns Sw5 18 24.7 7.6 ns Uninoculated control 18 25.3 11.1 213 Table A4.16 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Salmon Arm seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. ANOVA MODEL: Y = \x + BLOCK + INOC, where Y: effect, p;: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 22267.574 17 1309.857 0.829 0.654 INOC 3877.551 3 1292.517 0.818 0.490 ERROR 80612.245 51 1580.632 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 755.858 1 755.858 0.478 0.492 H 0 2 0.000 1 0.000 0.000 1.000 HO3 755.858 1 755.858 0.478 0.492 ERROR 80612.245 51 1580.632 (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 928.500 17 54.618 2.868 0.002 INOC 95.833 3 31.944 1.678 0.183 ERROR 971.167 51 19.042 214 Table A4.16 Continued. (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 14.815 1 14.815 0.778 0.382 H 0 2 25.000 1 25.000 1.313 0.257 H 0 3 92.593 1 92.593 4.862 0.032 ERROR 971.167 51 19.042 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 3525.611 17 207.389 2.413 0.008 INOC 88.944 3 29.648 0.345 0.793 ERROR 4384.056 51 85.962 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 16.333 1 16.333 0.190 0.665 H 0 2 44.444 1 44.444 0.517 0.475 H 0 3 13.370 1 13.370 0.156 0.695 ERROR 4384.056 51 85.962 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 3868.944 17 227.585 3.215 0.001 INOC 170.944 3 56.981 0.805 0.497 ERROR 3610.056 51 70.785 215 Table A4.16 Continued. (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 83.565 1 83.565 1.181 0.282 H 0 2 96.694 1 96.694 1.366 0.248 H 0 3 4.898 1 4.898 0.069 0.794 ERROR 3610.056 51 70.785 216 Table A4.17 Descriptive statistics for ecotype effects on spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3 18 193 62 ns Ss2 18 194 103 ns Sw5 18 190 88 ns Uninoculated control 18 220 88 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3 18 39.6 5.6 ns Ss2 18 36.5 5.3 ns Sw5 18 38.2 5.0 ns Uninoculated control 18 36.3 5.1 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 27.3 8.6 ns Ss2 18 25.8 7.0 ns Sw5 18 28.6 9.6 ns Uninoculated control 18 24.7 7.9 (d) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 18 24.7 8.6 * Ss2 18 21.1 6.1 ns Sw5 18 21.9 7.7 ns Uninoculated control 18 18.7 5.7 217 Table A4.18 ANOVA and orthogonal contrasts of ecotype effects on spruce seedlings from the Williams Lake seedlot inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 65770.975 17 3868.881 0.456 0.961 INOC 10430.839 3 3476.946 0.409 0.747 ERROR 433140.590 51 8492.953 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 188.964 1 188.964 0.022 0.882 H0 2 8185.941 1 8185.941 0.964 0.331 HO3 8231.293 1 8231.293 0.969 0.330 ERROR 433140.590 51 8492.953 (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 637.569 17 37.504 1.543 0.117 INOC 129.042 3 43.014 1.770 0.165 ERROR 1239.708 51 24.308 218 Table A4.18 Continued. (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 0.148 1 0.148 0.006 0.938 H0 2 30.250 1 30.250 1.244 0.270 HO3 35.593 1 35.593 1.464 0.232 ERROR 1239.708 51 24.308 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1612.569 17 94.857 1.555 0.113 INOC 156.708 3 52.236 0.856 0.470 ERROR 3112.042 51 61.020 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 52.083 1 52.083 0.854 0.360 H0 2 136.111 1 136.111 2.231 0.141 HO3 39.120 1 39.120 0.641 0.427 ERROR 3112.042 51 61.020 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1233.111 17 72.536 1.667 0.081 INOC 324.778 3 108.259 2.488 0.071 ERROR 2219.222 51 43.514 219 Table A4.18 Continued. (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 10.704 1 10.704 0.246 0.622 H 0 2 93.444 1 93.444 2.147 0.149 HO3 208.333 1 208.333 4.788 0.033 ERROR 2219.222 51 43.514 220 Table A4.19 Descriptive statistics for ecotype effects on spruce seedlings from the Mackenzie, Salmon Arm and Williams Lake seedlots (pooled) inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3 54 271 82 ns Ss2 54 275 99 ns Sw5 54 263 96 ns Uninoculated control 54 266 90 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3 54 38.5 5.1 ** Ss2 54 37.1 5.5 ns Sw5 54 37.7 4.9 * Uninoculated control 54 35.9 5.3 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 54 30.4 10.3 ns Ss2 54 29.4 10.6 ns Sw5 54 30.0 10.2 ns Uninoculated control 54 26.4 8.9 (d) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3 54 26.3 8.3 ** Ss2 54 25.0 10.4 ** Sw5 54 23.4 7.9 ns Uninoculated control 54 21.5 9.3 221 Table A4.20 ANOVA and orthogonal contrasts of host ecotype effects on spruce seedlings inoculated with Pseudomonas strain Sm3, Ss2 or Sw5. ANOVA MODEL: Y = p + BLOCK + INOC + ECOT + ECOT x INOC, where Y: effect, p: constant, INOC: inoculation with PGPR including uninoculated controls, and ECOT: seedling ecotype (i.e., Mackenzie, Salmon Arm, or Williams Lake). Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated controls). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated controls). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 42649.282 17 2508.781 0.477 0.961 INOC 4092.971 3 1364.324 0.259 0.855 ECOT 727309.146 2 363654.573 69.102 0.000 ECOT xINOC 37619.048 6 6269.841 1.191 0.313 ERROR 984108.088 187 5262.610 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 5.669 1 5.669 0.001 0.974 H 0 2 425.170 1 425.170 0.081 0.777 H 0 3 459.184 1 459.184 0.087 0.768 ERROR 984108.088 187 5262.610 222 Table A4.20 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1279.833 17 75.284 3.249 0.000 INOC 206.259 3 68.753 2.967 0.033 ECOT 18.861 2 9.431 0.407 0.666 ECOT x INOC 55.435 6 9.239 0.399 0.879 ERROR 4332.944 187 23.171 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 2.086 1 2.086 0.090 0.764 H 0 2 81.815 1 81.815 3.531 0.062 H 0 3 141.346 1 141.346 6.100 0.014 ERROR 4332.944 187 23.171 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 5306.204 17 312.130 3.796 0.000 INOC 513.296 3 171.099 2.081 0.104 ECOT 697.898 2 348.949 4.244 0.016 ECOTxINOC 317.176 6 52.863 0.643 0.696 ERROR 15375.130 187 82.220 223 Table A4.20 Continued. (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 73.198 1 73.198 0.890 0.347 H 0 2 524.481 1 524.481 6.379 0.012 H 0 3 320.012 1 320.012 3.892 0.050 ERROR 15375.130 187 82.220 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 3694.968 17 217.351 3.247 0.000 INOC 678.273 3 226.091 3.377 0.020 ECOT 850.843 2 425.421 6.354 0.002 ECOT x INOC 218.824 6 36.471 0.545 0.774 ERROR 12519.310 187 66.948 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 73.198 1 73.198 1.093 0.297 H 0 2 502.676 1 502.676 7.508 0.007 H 0 3 300.444 1 300.444 4.488 0.035 ERROR 12519.310 187 66.948 224 Table A4.21 Descriptive statistics for pine ecotype effects using seedlings from Fort St. John, Kamloops, or Williams Lake seedlots inoculated with Bacillus strain Pw2. (a) Seedling emergence value (untransformed). Seed provenace Treatment n Mean S.D. Student f1 Fort St. John Pw2 18 313 77 ns Uninoculated 18 298 79 Kamloops Pw2 18 260 86 ns Uninoculated 18 271 74 Williams Lake Pw2 18 270 83 ns Uninoculated 18 237 97 1 Student's t test (2-tailed) comparing Pw2-inoculated and uninoculated treatments. (b) Shoot height (untransformed). Seed provenace Treatment n Mean (mm) S.D. Student t Fort St. John Pw2 18 34.3 4.5 ns Uninoculated 18 34.6 4.5 Kamloops Pw2 18 33.0 4.0 ns Uninoculated 18 31.5 4.9 Williams Lake Pw2 17 34.5 4.5 ns Uninoculated 18 34.1 3.9 225 Table A4.21 Continued. (c) Shoot biomass (untransformed). Seed provenace Treatment n Mean (mg) S.D. Student/ Fort St. John Pw2 18 28.6 8.6 ns Uninoculated 18 29.2 6.8 Kamloops Pw2 18 27.6 6.2 ns Uninoculated 18 26.2 9.6 Williams Lake Pw2 17 28.6 7.8 ns Uninoculated 18 27.8 5.9 (d) Root biomass (untransformed). Seed provenace Treatment n Mean (mg) S.D. Student t Fort St. John Pw2 18 26.0 6.8 ns Uninoculated 18 26.4 13.3 Kamloops Pw2 18 24.3 5.9 ns Uninoculated 18 22.8 7.1 Williams Lake Pw2 17 24.8 7.0 ns Uninoculated 18 24.6 6.3 226 Table A4.22 ANOVA and orthogonal contrasts of pine ecotype effects on seedlings inoculated with Bacillus strain Pw2. ANOVA MODEL: Y = u, + BLOCK + INOC + ECOT + ECOT x INOC, where Y: effect, p.: constant, INOC: inoculation with PGPR including uninoculated controls (i.e., inoculated or uninoculated), and ECOT: seedling ecotype (i.e., Mackenzie, Salmon Arm, or Williams Lake). Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). H O 2 : (Coexistent plant-bacteria combinations) vs. (Uninoculated controls). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated controls). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 144363.190 17 8491.952 1.292 0.218 INOC 3826.531 1 3826.531 0.582 0.448 ECOT 53677.249 2 26838.624 4.083 0.020 INOC x ECOT 8650.794 2 4325.397 0.658 0.520 ERROR 558709.373 85 6573.051 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 3176.493 1 3176.493 0.483 0.489 H 0 2 9529.478 1 9529.478 1.450 0.232 H 0 3 45.351 1 45.351 0.007 0.934 ERROR 558709.373 85 6573.051 227 Table A4.22 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 158.491 17 9.323 0.437 0.972 INOC 7.787 1 7.787 0.365 0.547 ECOT 107.463 2 53.731 2.519 0.087 INOC x ECOT 15.241 2 7.620 0.357 0.701 ERROR 1813.343 85 21.333 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 8.898 1 8.898 0.417 0.520 H 0 2 1.778 1 1.778 0.083 0.774 H 0 3 6.125 1 6.125 0.287 0.593 ERROR 1813.343 85 21.333 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 978.503 17 57.559 1.002 0.465 INOC 11.568 1 11.568 0.201 0.655 ECOT 70.225 2 35.112 0.611 0.545 INOC x ECOT 16.953 2 8.477 0.148 0.863 ERROR 4827.102 84 57.465 228 Table A4.22 Continued. (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 5.574 1 5.574 0.097 0.756 H0 2 9.235 1 9.235 0.161 0.690 HO3 4.014 1 4.014 0.070 0.792 ERROR 4827.102 84 57.465 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 961.150 17 56.538 0.828 0.657 INOC 5.006 1 5.006 0.073 0.787 ECOT 128.570 2 64.285 0.941 0.394 INOC x ECOT 16.846 2 8.423 0.123 0.884 ERROR 5737.187 84 68.300 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 1.321 1 1.321 0.019 0.890 H 0 2 0.308 1 0.308 0.005 0.947 H 0 3 5.556 1 5.556 0.081 0.776 ERROR 5737.187 84 68.300 229 Table A4.23 Descriptive statistics for host genus effects on performance of spruce or pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R. (a) Spruce seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3R 18 208 127 ns Ss2R 18 220 130 ns Sw5R 18 245 117 ns Pw2R 18 204 142 ns Uninoculated control 18 220 106 1 Fisher's protected least significant difference. indicate significant difference the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Pine spruce seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD Sm3R 18 300 86 ns Ss2R 18 331 78 ns Sw5R 18 324 76 ns Pw2R 18 324 76 ns Uninoculated control 18 346 47 (c) Spruce shoot height (In transformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3R 18 3.26 0.10 ** Ss2R 16 3.27 0.07 ** Sw5R 18 3.25 0.12 ** Pw2R 16 3.27 0.09 ** Uninoculated control 18 3.18 0.09 230 Table A4.23 Continued. (d) Pine shoot height (In transformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3R 18 3.70 0.19 ns Ss2R 18 3.59 0.14 * Sw5R 17 3.70 0.15 ns Pw2R 18 3.73 0.13 ns Uninoculated control 18 3.68 0.13 (e) Spruce shoot biomass (In transformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3R 18 3.16 0.25 ns Ss2R 16 3.19 0.27 ns Sw5R 18 3.23 0.24 ns Pw2R 16 3.25 0.19 ns Uninoculated control 18 3.09 0.23 (f) Pine shoot biomass (In transformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3R 18 3.59 0.42 ns Ss2R 18 3.31 0.34 * Sw5R 17 3.55 0.33 ns Pw2R 18 3.57 0.29 ns Uninoculated control 18 3.58 0.28 231 Table A4.23 Continued. (g) Spruce root biomass (In transformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3R 18 3.04 0.27 ns Ss2R 16 3.10 0.32 ns Sw5R 18 3.10 0.29 ns Pw2R 16 3.08 0.25 ns Uninoculated control 18 2.99 0.26 ns (h) Pine root biomass (In transformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3R 18 3.40 0.27 ns Ss2R 18 3.42 0.31 ns Sw5R 17 3.41 0.28 ns Pw2R 18 3.52 0.27 ns Uninoculated control 18 3.42 0.25 (i) Spruce rhizosphere colonization (log transformed). Treatment n Mean (cfu-g sou) S.D. Sm3R 4 3.90 2.62 Ss2R 4 4.23 2.88 Sw5R 4 1.36 2.72 Pw2R 4 0 0 (j) Pine rhizosphere colonization (log transformed). Treatment n Mean (cfu-g soil) S.D. Sm3R 4 3.06 3.56 Ss2R 4 3.61 4.16 Sw5R 4 1.21 2.42 Pw2R 4 1.37 2.74 232 Table A4.24 ANOVA and orthogonal contrasts of host genus effects on performance of spruce or pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R. ANOVA MODEL: Y = p. + BLOCK + INOC, where Y: effect, \x: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for spruce seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 191657.411 17 11273.965 0.676 0.815 INOC 18674.857 4 4668.714 0.280 0.890 ERROR 1133731.192 68 16672.518 (b) Orthogonal contrasts for spruce seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 5446.745 1 5446.745 0.327 0.569 H 0 2 292.417 1 292.417 0.018 0.895 H 0 3 2143.401 1 2143.401 0.129 0.721 ERROR 1133731.192 68 16672.518 233 Table A4.24 Continued. (c) ANOVA for pine seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 116825.487 17 6872.087 1.344 0.194 INOC 20051.235 4 5012.809 0.980 0.424 ERROR 347799.464 68 5114.698 (d) Orthogonal contrasts for pine seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 466.578 1 466.578 0.091 0.764 H 0 2 4374.178 1 4374.178 0.855 0.358 H 0 3 10527.182 1 10527.182 2.058 0.156 ERROR 347799.464 68 5114.698 (e) ANOVA for spruce shoot height (In transformed). SOURCE SS DF MS F-ratio P BLOCK 0.216 17 0.013 1.519 0.117 INOC 0.106 4 0.026 3.169 0.019 ERROR 0.535 64 0.008 234 Table A4.24 Continued. (f) Orthogonal contrasts for spruce shoot height (In transformed). SOURCE SS DF MS F-ratio P HOi 0.006 1 0.006 0.676 0.414 H 0 2 0.080 1 0.080 9.547 0.003 H 0 3 0.082 1 0.082 9.785 0.003 ERROR 0.535 64 0.008 (g) ANOVA for pine shoot height (In transformed). SOURCE SS DF MS F-ratio P BLOCK 0.628 17 0.037 2.048 0.020 INOC 0.210 4 0.052 2.908 0.028 ERROR 1.209 67 0.018 (h) Orthogonal contrasts for pine shoot height (In transformed). SOURCE SS DF MS F-ratio P HOi 0.060 1 0.060 3.337 0.072 H 0 2 0.027 1 0.027 1.516 0.223 H 0 3 0.002 1 0.002 0.105 0.747 ERROR 1.209 67 0.018 235 Table A4.24 Continued. (i) ANOVA for spruce shoot biomass (In transformed). SOURCE SS DF MS F-ratio P BLOCK 1.610 17 0.095 2.013 0.023 INOC 0.298 4 0.074 1.583 0.190 ERROR 3.010 64 0.047 (j) Orthogonal contrasts for spruce shoot biomass (1 n transformed). SOURCE SS DF MS F-ratio P HOi 0.055 1 0.055 1.168 0.284 H 0 2 0.130 1 0.130 2.758 0.102 H 0 3 0.231 1 0.231 4.901 0.030 ERROR 3.010 64 0.047 (k) ANOVA for pine shoot biomass (In transformed). SOURCE SS DF MS F-ratio P BLOCK 3.421 17 0.201 2.253 0.010 INOC 0.962 4 0.241 2.693 0.038 ERROR 5.984 67 0.089 236 Table A4.24 Continued. (1) Orthogonal contrasts for pine shoot biomass (In transformed). SOURCE SS DF MS F-ratio P HOi 0.109 1 0.109 1.224 0.273 H02 0.000 1 0.000 0.004 0.952 H03 0.124 1 0.124 1.392 0.242 ERROR 5.984 67 0.089 (m) ANOVA for spruce root biomass (In transformed). SOURCE SS DF MS F-ratio P BLOCK 1.855 17 0.109 1.597 0.091 INOC 0.204 4 0.051 0.748 0.563 ERROR 4.372 64 0.068 (n) Orthogonal contrasts for spruce root biomass (In transformed). SOURCE SS DF MS F-ratio P HOi 0.004 1 0.004 0.064 0.801 H 0 2 0.125 1 0.125 1.826 0.181 H 0 3 0.112 1 0.112 1.637 0.205 ERROR 4.372 64 0.068 (0) ANOVA for pine root biomass (1 n transformed). SOURCE SS DF MS F-ratio P BLOCK 1.769 17 0.104 1.548 0.105 INOC 0.184 4 0.046 0.683 0.606 ERROR 4.504 67 0.067 237 Table A4.24 Continued. (p) Orthogonal contrasts for pine root biomass (In transformed). SOURCE SS DF MS F-ratio P HOi 0.171 1 0.171 2.539 0.116 H 0 2 0.102 1 0.102 1.515 0.223 H 0 3 0.001 1 0.001 0.008 0.928 ERROR 4.504 67 0.067 238 Table A4.25 Descriptive statistics for host genus effects on performance of spruce and pine seedlings (pooled) inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R. (a) Seedling emergence value (untransformed). Treatment n Mean SD. Fisher's LSD 1 Sm3R 36 254 117 ns Ss2R 36 275 120 ns Sw5R 36 285 105 ns Pw2R 36 264 127 ns Uninoculated control 36 283 103 1 Fisher's protected least significant difference. indicate significant difference the uninouclated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (In transformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3R 36 3.48 0.27 2 Ss2R 34 3.44 0.20 -Sw5R 35 3.47 0.27 -Pw2R 34 3.52 0.26 -Uninoculated control 36 3.43 0.27 Interaction between the treatments of inoculation and host genus was found. (c) Shoot biomass (In transformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3R 36 3.37 0.40 ns Ss2R 34 3.25 0.31 ns Sw5R 35 3.39 0.33 ns Pw2R 34 3.42 0.30 ns Uninoculated control 36 3.34 0.35 239 Table A4.25 Continued. (d) Root biomass (In transformed). Treatment n Mean (mg) SD. Fisher's LSD Sm3R 36 3.22 0.32 ns Ss2R 34 3.27 0.35 ns Sw5R 35 3.25 0.32 ns Pw2R 34 3.32 0.34 ns Uninoculated control 36 3.20 0.33 240 Table A4.26 ANOVA and orthogonal contrasts of host genus effects on performance of spruce and pine seedlings inoculated with strain Sm3R, Ss2R, Sw5R or Pw2R. ANOVA MODEL: Y = p + BLOCK + INOC + SPP + INOC x SPP, where Y: effect, p: constant, INOC: inoculation with PGPR including uninoculated controls, and SPP: host genus (i.e., interior spruce and lodgepole pine). Orthogonal contrast, hypothesis tests: HOi: (Coexistent plant-bacteria combinations) vs. (Unrelated plant-bacteria combinations). HO2: (Coexistent plant-bacteria combinations) vs. (Uninoculated controls). HO3: (Unrelated plant-bacteria combinations) vs. (Uninoculated controls). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 133764.390 17 7868.494 0.727 0.772 INOC 24518.778 4 6129.695 0.566 0.688 HOST 501110.428 1 501110.428 46.291 0.000 INOC x HOST 14207.313 4 3551.828 0.328 0.859 ERROR 1656249.163 153 10825.158 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 4550.817 1 4550.817 0.420 0.518 H0 2 1633.357 1 1633.357 0.151 0.698 HO3 10151.095 1 10151.095 0.938 0.334 ERROR 1656249.163 153 10825.158 241 Table A4.26 Continued. (c) ANOVA for shoot height (In transformed). SOURCE SS DF MS F-ratio P BLOCK 0.357 17 0.021 1.392 0.148 INOC 0.140 4 0.035 2.318 0.060 HOST 8.299 1 8.299 550.489 0.000 INOC x HOST 0.146 4 0.036 2.418 0.051 ERROR 2.231 148 0.015 (e) ANOVA for shoot biomass (In transformed). SOURCE SS DF MS F-ratio P BLOCK 1.571 17 0.092 1.098 0.361 INOC 0.563 4 0.141 1.672 0.160 HOST 4.957 1 4.957 58.906 0.000 INOC x HOST 0.618 4 0.155 1.837 0.125 ERROR 12.455 148 0.084 (f) Orthogonal contrasts for shoot biomass (In transformed). SOURCE SS DF MS F-ratio P HOi 0.007 1 0.007 0.084 0.772 H 0 2 0.042 1 0.042 0.501 0.480 H 0 3 0.016 1 0.016 0.187 0.666 ERROR 12.455 148 0.084 242 Table A4.26 Continued. (g) ANOVA for root biomass (In transformed). SOURCE SS DF MS F-ratio P BLOCK 1.399 17 0.082 1.097 0.362 INOC 0.204 4 0.051 0.680 0.607 HOST 6.110 1 6.110 81.464 0.000 INOC x HOST 0.121 4 0.030 0.403 0.806 ERROR 11.101 148 0.075 (h) Orthogonal contrasts for root biomass (In transformed). SOURCE SS DF MS F-ratio P HOi 0.080 1 0.080 1.065 0.304 H 0 2 0.206 1 0.206 2.746 0.100 H 0 3 0.036 1 0.036 0.484 0.488 ERROR 11.101 148 0.075 243 Table A4.27 Descriptive statistics for soil type effects on spruce seedlings in the presence of soil from the Mackenzie spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3RN 15 215 86 ns Ss2RN 15 211 89 ns Sw5RN 15 208 93 ns Pw2R 15 199 99 ns Uninoculated control 15 215 86 1 Fisher's protected least significant difference. jft 9|C )JC )JC if* "^I^ l^ ^ indicate significant difference the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3RN 15 45.5 9.3 ns Ss2RN 15 46.1 11.0 ns Sw5RN 15 43.9 13.4 ns Pw2R 15 40.7 8.2 ns Uninoculated control 15 46.7 9.3 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 68.3 24.3 ns Ss2RN 15 65.7 27.4 ns Sw5RN 15 66.1 31.5 ns Pw2R 15 64.5 16.8 ns Uninoculated control 15 71.9 24.1 244 Table A4.27 Continued. (d) Root biomass (untransformed). Treatment n Mean(mg) S.D. Fisher's LSD Sm3RN 15 76.1 18.6 ns Ss2RN 15 75.2 25.8 ns Sw5RN 15 72.3 20.9 ns Pw2R 15 66.9 18.4 ns Uninoculated control 15 66.1 15.7 (e) Rhizosphere colonization (log transformed) in the presence of the Mackenzie spruce stand soil. Treatment n Mean (cfu-g soil) S.D. Sm3RN 4 3.61 2.44 Ss2RN 4 3.55 2.44 Sw5RN 4 4.37 0.80 Pw2R 4 0 0 (f) Rhizosphere colonization (log transformed) in the presence of the Salmon Arm spruce stand soil. Treatment n Mean (cfu-g ^oil) S.D. Sm3RN 4 4.47 0.66 Ss2RN 4 5.26 0.54 Sw5RN 4 4.64 0.28 Pw2R 4 0 0 245 Table A4.27 Continued. (g) Rhizosphere colonization (log transformed) in the presence of the Williams Lake spruce stand soil. Treatment n Mean (cfu-g soil) S.D. Sm3RN 4 4.93 0.64 Ss2RN 4 5.09 1.10 Sw5RN 4 3.56 2.40 Pw2R 4 1.04 2.08 (h) Rhizosphere colonization (log transformed) in the presence of the Williams Lake pine stand soil. Treatment n Mean (cfu-g 1soiI) S.D. Sm3RN 4 3.62 2.44 Ss2RN 4 5.19 0.74 Sw5RN 4 2.24 2.60 Pw2R 4 0 0 246 Table A4.28 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings in the presence of soil from the Mackenzie spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN orPw2R. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p.: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent soil type-bacteria combinations) vs. (Unrelated soil type-bacteria combinations). HO2: (Coexistent soil type-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated soil type-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 250462.963 14 17890.212 3.073 0.001 INOC 2835.648 4 708.912 0.122 0.974 ERROR 325983.796 56 5821.139 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 964.506 1 964.506 0.166 0.686 H 0 2 0.000 1 0.000 0.000 1.000 HO3 964.506 1 964.506 0.166 0.686 ERROR 325983.796 56 5821.139 247 Table A4.28 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1548.187 14 110.585 1.027 0.442 INOC 341.520 4 85.380 0.793 0.535 ERROR 6032.480 56 107.723 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 39.200 1 39.200 0.364 0.549 H 0 2 10.800 1 10.800 0.100 0.753 H 0 3 105.800 1 105.800 0.982 0.326 ERROR 6032.480 56 107.723 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 9998.747 14 714.196 1.149 0.339 INOC 510.480 4 127.620 0.205 0.934 ERROR 34800.720 56 621.441 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 89.606 1 89.606 0.144 0.706 H 0 2 100.833 1 100.833 0.162 0.689 H 0 3 473.689 1 473.689 0.762 0.386 ERROR 34800.720 56 621.441 248 Table A4.28 Continued. (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 3204.347 14 228.882 0.507 0.919 INOC 1279.813 4 319.953 0.709 0.589 ERROR 25275.787 56 451.35 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 238.050 1 238.050 0.527 0.471 H 0 2 750.000 1 750.000 1.662 0.203 H 0 3 328.050 1 328.050 0.727 0.398 ERROR 25275.787 56 451.353 249 Table A4.29 Descriptive statistics for soil type effects on spruce seedlings in the presence of soil from the Salmon Arm spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD 1 Sm3RN 15 213 90 ns Ss2RN 15 197 95 ns Sw5RN 15 233 87 ns Pw2R 15 224 79 ns Uninoculated control 15 207 84 1 Fisher's protected least significant difference. indicate significant difference the uninouclated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) SD. Fisher's LSD Sm3RN 15 48.8 7.6 ns Ss2RN 15 52.7 11.2 ns Sw5RN 15 51.3 10.1 ns Pw2R 15 52.7 7.7 ns Uninoculated control 15 50.5 10.5 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 81.8 21.9 ns Ss2RN 15 87.7 30.0 ns Sw5RN 15 85.5 27.9 ns Pw2R 15 84.3 21.8 ns Uninoculated control 15 75.5 24.9 250 Table A4.29 Continued. (g) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 83.6 25.4 ns Ss2RN 15 85.4 30.2 ns Sw5RN 15 84.5 25.2 ns Pw2R 15 76.5 10.6 ns Uninoculated control 15 71.7 21.0 251 Table A4.30 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings in the presence of soil from the Salmon Arm spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN orPw2R. ANOVA MODEL: Y = u. + BLOCK + INOC, where Y: effect, u,: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent soil type-bacteria combinations) vs. (Unrelated soil type-bacteria combinations). HO2: (Coexistent soil type-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated soil type-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 69247.685 14 4946.263 0.597 0.856 INOC 11956.019 4 2989.005 0.361 0.835 ERROR 463912.037 56 8284.144 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 7561.728 1 7561.728 0.913 0.343 H 0 2 708.912 1 708.912 0.086 0.771 HO3 2953.800 1 2953.800 0.357 0.553 ERROR 463912.037 56 8284.144 252 Table A4.30 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 2095.600 14 149.686 1.975 0.037 INOC 157.867 4 39.467 0.521 0.721 ERROR 4244.533 56 75.795 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 33.800 1 33.800 0.446 0.507 H 0 2 34.133 1 34.133 0.450 0.505 H 0 3 24.500 1 24.500 0.323 0.572 ERROR 4244.533 56 75.795 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 15184.987 14 1084.642 1.984 0.036 INOC 1317.253 4 329.313 0.602 0.663 ERROR 30619.547 56 546.778 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 162.450 1 162.450 0.297 0.588 H 0 2 1116.300 1 1116.300 2.042 0.159 H 0 3 1210.320 1 1210.320 2.214 0.142 ERROR 30619.547 56 546.778 253 Table A4.30 Continued. (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 14015.067 14 1001.076 2.295 0.014 INOC 2144.667 4 536.167 1.229 0.309 ERROR 24428.933 56 436.231 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 168.200 1 168.200 0.386 0.537 H0 2 1414.533 1 1414.533 3.243 0.077 HO3 1568.000 1 1568.000 3.594 0.063 ERROR 24428.933 56 436.231 254 Table A4.31 Descriptive statistics for soil ecotype effects on spruce seedlings treated with soil from the Williams Lake spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN orPw2R. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD Sm3RN 15 221 91 ns Ss2RN 15 196 118 ns Sw5RN 15 185 93 ns Pw2R 15 211 93 ns Uninoculated control 15 192 104 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninoculated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3RN 15 52.5 8.8 ** Ss2RN 15 50.5 8.3 ** Sw5RN 15 50.7 10.2 ** Pw2R 15 47.9 8.4 ns Uninoculated control 15 42.8 8.0 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 80.9 21.1 ns Ss2RN 15 80.1 22.4 ns Sw5RN 15 75.7 24.0 ns Pw2R 15 76.3 24.5 ns Uninoculated control 15 64.1 21.0 255 Table A4.31 Continued. (g) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 83.0 26.1 ns Ss2RN 15 87.6 29.8 ns Sw5RN 15 83.7 26.9 ns Pw2R 15 79.4 21.1 ns Uninoculated control 15 69.1 18.0 256 Table A4.32 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Williams Lake spruce stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent soil type-bacteria combinations) vs. (Unrelated soil type-bacteria combinations). HO2: (Coexistent soil type—bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated soil type-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 89270.833 14 6376.488 0.567 0.879 INOC 13113.426 4 3278.356 0.291 0.882 ERROR 629942.130 56 11248.967 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 6773.245 1 6773.245 0.602 0.441 H 0 2 361.690 1 361.690 0.032 0.858 H 0 3 3481.867 1 3481.867 0.310 0.580 ERROR 629942.130 56 11248.967 257 Table A4.32 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 809.515 14 57.823 0.702 0.762 INOC 836.365 4 209.091 2.539 0.050 ERROR 4528.785 55 82.342 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 3.601 1 3.601 0.044 0.835 H0 2 472.033 1 472.033 5.733 0.020 H0 3 605.808 1 605.808 7.357 0.009 ERROR 4528.785 55 82.342 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 5367.539 14 383.396 0.702 0.763 INOC 2650.692 4 662.673 1.213 0.316 ERROR 30039.108 55 546.166 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 125.945 1 125.945 0.231 0.633 H0 2 997.633 1 997.633 1.827 0.182 HO3 2476.394 1 2476.394 4.534 0.038 ERROR 30039.108 55 546.166 258 Table A4.32 Continued. (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 8192.343 14 585.167 0.952 0.512 INOC 2941.862 4 735.465 1.197 0.323 ERROR 33802.338 55 614.588 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 1.521 1 1.521 0.002 0.961 H 0 2 1613.333 1 1613.333 2.625 0.111 H 0 3 2282.762 1 2282.762 3.714 0.059 ERROR 33802.338 55 614.588 259 Table A4.33 Descriptive statistics for soil type effects on spruce seedlings treated with soil from the Williams Lake pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN orPw2R. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD Sm3RN 15 171 100 ns Ss2RN 15 213 91 ns Sw5RN 15 178 98 ns Pw2R 15 217 94 ns Uninoculated control 15 214 95 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninouclated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3RN 15 45.4 6.6 * Ss2RN 15 48.3 6.5 ** Sw5RN 15 43.2 10.0 ns Pw2R 15 47.6 9.3 ** Uninoculated control 15 40.7 7.4 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 74.3 21.9 ns Ss2RN 15 81.3 16.3 ns Sw5RN 15 67.9 25.7 ns Pw2R 15 76.7 26.4 ns Uninoculated control 15 62.5 21.6 260 Table A4.33 Continued. (g) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 15 71.1 16.9 ns Ss2RN 15 76.8 20.0 ns Sw5RN 15 69.5 27.6 ns Pw2R 15 73.1 23.7 ns Uninoculated control 15 61.3 17.3 261 Table A4.34 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Williams Lake pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RNorPw2R. ANOVA MODEL: Y = p. + BLOCK + INOC, where Y: effect, u,: constant, and INOC: inoculation with PGPR including uninoculated control. Orthogonal contrast, hypothesis tests: HOi: (Coexistent soil type-bacteria combinations) vs. (Unrelated soil type-bacteria combinations). HO2: (Coexistent soil type-bacteria combinations) vs. (Uninoculated control). HO3: (Unrelated soil type-bacteria combinations) vs. (Uninoculated control). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 121840.278 14 8702.877 0.939 0.525 INOC 29363.426 4 7340.856 0.792 0.536 ERROR 519247.685 56 9272.280 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 9876.543 1 9876.543 1.065 0.306 H 0 2 57.870 1 57.870 0.006 0.937 HO3 8111.497 1 8111.497 0.875 0.354 ERROR 519247.685 56 9272.280 262 Table A4.34 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1305.947 14 93.282 1.594 0.110 INOC 594.080 4 148.520 2.537 0.050 ERROR 3277.920 56 58.534 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 44.006 1 44.006 0.752 0.390 H 0 2 360.533 1 360.533 6.159 0.016 H 0 3 276.272 1 276.272 4.720 0.034 ERROR 3277.920 56 58.534 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 8636.320 14 616.880 1.265 0.258 INOC 3309.653 4 827.413 1.697 0.164 ERROR 27310.747 56 487.692 (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 53.356 1 53.356 0.109 0.742 H 0 2 1512.300 1 1512.300 3.101 0.084 H 0 3 1626.006 1 1626.006 3.334 0.073 ERROR 27310.747 56 487.692 263 Table A4.34 Continued. (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 7836.587 14 559.756 1.279 0.250 INOC 1993.253 4 498.313 1.138 0.348 ERROR 24511.147 56 437.699 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 4.050 1 4.050 0.009 0.924 H 0 2 1044.300 1 1044.300 2.386 0.128 H 0 3 1411.200 1 1411.200 3.224 0.078 ERROR 24511.147 56 437.699 264 Table A4.35 Descriptive statistics for soil type effects on spruce seedlings treated with soil from the Mackenzie spruce, Salmon Arm spruce Williams Lake spruce and Williams Lake pine stand and inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R. (a) Seedling emergence value (untransformed). Treatment n Mean S.D. Fisher's LSD Sm3RN 60 205 92 ns Ss2RN 60 204 97 ns Sw5RN 60 201 93 ns Pw2R 60 213 91 ns Uninoculated control 60 207 91 Fisher's protected least significant difference. ***, **, and * indicate significant difference from the uninouclated control at P < 0.01, P < 0.05, and P < 0.1, respectively. (b) Shoot height (untransformed). Treatment n Mean (mm) S.D. Fisher's LSD Sm3RN 60 48.0 8.5 ns Ss2RN 60 49.4 9.6 ns Sw5RN 60 47.3 11.4 ns Pw2R 60 47.2 9.2 ns Uriinoculated control 60 45.2 9.5 (c) Shoot biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 60 76.3 22.4 ns Ss2RN 60 78.7 25.4 ns Sw5RN 60 73.8 27.8 ns Pw2R 60 75.5 23.2 ns Uninoculated control 60 68.5 23.0 265 T a b l e A4.35 Continued. (g) Root biomass (untransformed). Treatment n Mean (mg) S.D. Fisher's LSD Sm3RN 60 78.5 22.2 *** Ss2RN 60 81.1 26.5 *** Sw5RN 60 77.5 25.6 ** Pw2R 60 74.0 19.2 * Uninoculated control 60 67.0 18.1 266 Table A4.36 ANOVA and orthogonal contrasts of soil type effects on spruce seedlings treated with soil from the Mackenzie spruce, Salmon Arm spruce, Williams Lake spruce and Williams Lake pine stand inoculated with strain Sm3RN, Ss2RN, Sw5RN or Pw2R. ANOVA MODEL: Y = u. + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p,: constant, INOC: inoculation with PGPR including uninoculated controls, and SOIL: soil type from conifer stand where PGPR were isolated (i.e., Mackenzie spruce, Salmon Arm spruce, Williams Lake spruce and pine). Orthogonal contrast, hypothesis tests: HOi: (Coexistent soil type-bacteria combinations) vs. (Unrelated soil type-bacteria combinations). HO2: (Coexistent soil type-bacteria combinations) vs. (Uninoculated controls). HO3: (Unrelated soil type-bacteria combinations) vs. (Uninoculated controls). (a) ANOVA for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P BLOCK 161313.657 14 11522.404 1.328 0.191 INOC 4340.278 4 1085.069 0.125 0.973 SOIL 13152.488 3 4384.163 0.505 0.679 INOC x SOIL 52928.241 12 4410.687 0.508 0.909 ERROR 2308593.750 266 8678.924 (b) Orthogonal contrasts for seedling emergence value (untransformed). SOURCE SS DF MS F-ratio P HOi 376.760 1 376.760 0.043 0.835 H0 2 361.690 1 361.690 0.042 0.838 HO3 15.070 1 15.070 0.002 0.967 ERROR 2308593.750 266 8678.924 267 Table A4.36 Continued. (c) ANOVA for shoot height (untransformed). SOURCE SS DF MS F-ratio P BLOCK 1019.950 14 72.854 0.846 0.619 INOC 556.242 4 139.060 1.615 0.171 SOIL 2256.968 3 752.323 8.735 0.000 INOC x SOIL 1383.853 12 115.321 1.339 0.196 ERROR 22823.016 265 86.125 (d) Orthogonal contrasts for shoot height (untransformed). SOURCE SS DF MS F-ratio P HOi 102.853 1 102.853 1.194 0.275 H 0 2 468.075 1 468.075 5.435 0.020 H 0 3 266.833 1 266.833 3.098 0.080 ERROR 22823.016 265 86.125 (e) ANOVA for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 6074.306 14 433.879 0.738 0.736 INOC 3511.972 4 877.993 1.493 0.205 SOIL 9593.023 3 3197.674 5.436 0.001 INOC x SOIL 4341.098 12 361.758 0.615 0.829 ERROR 155883.408 265 588.239 268 Table A4.36 Continued. (f) Orthogonal contrasts for shoot biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 78.867 1 78.867 0.134 0.715 H 0 2 2201.633 1 2201.633 3.743 0.054 H 0 3 2356.255 1 2356.255 4.006 0.046 ERROR 155883.408 265 588.239 (g) ANOVA for root biomass (untransformed). SOURCE SS DF MS F-ratio P BLOCK 9052.954 14 646.640 1.296 0.209 INOC 7153.762 4 1788.440 3.585 0.007 SOIL 6947.137 3 2315.712 4.641 0.004 INOC x SOIL 1207.725 12 100.644 0.202 0.998 ERROR 132213.594 265 498.919 (h) Orthogonal contrasts for root biomass (untransformed). SOURCE SS DF MS F-ratio P HOi 252.038 1 252.038 0.505 0.478 H 0 2 4725.075 1 4725.075 9.471 0.002 HO3 4657.571 1 4657.571 9.335 0.002 ERROR 132213.594 265 498.919 269 APPENDIX 5 Statistical analyses of data presented in Chapter 3 270 Table A5.1 ANOVA for spruce seedling height and biomass (regardless of mycorrhizal status) after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = p + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOIL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 1718.135 14 122.724 1.384 0.155 INOC 589.017 6 98.170 1.107 0.357 SOIL 3913.636 5 782.727 8.827 0.000 INOC x SOIL 3487.404 30 116.247 1.311 0.127 ERROR 50632.504 571 88.673 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 11119.512 14 794.251 1.307 0.198 INOC 4244.075 6 707.346 1.164 0.324 SOIL 18105.191 5 3621.038 5.957 0.000 INOC x SOIL 10463.204 30 348.773 0.574 0.968 ERROR 347105.841 571 607.891 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 11628.261 14 830.590 1.772 0.039 INOC 11149.022 6 1858.170 3.963 0.001 SOIL 11190.128 5 2238.026 4.773 0.000 INOC x SOIL 5890.385 30 196.346 0.419 0.998 ERROR 267718.306 571 468.859 271 Table A5.2 ANOVA for non-mycorrhizal spruce seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = u, + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOIL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 2351.479 14 167.963 2.047 0.014 INOC 395.773 6 65.962 0.804 0.567 SOIL 2716.823 5 543.365 6.624 0.000 INOC x SOIL 3608.233 30 120.274 1.466 0.057 ERROR 32239.450 393 82.034 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 14691.770 14 1049.412 1.889 0.026 INOC 3634.049 6 605.675 1.090 0.368 SOIL 14892.370 5 2978.474 5.360 0.000 INOC x SOIL 13876.892 30 462.563 0.832 0.722 ERROR 218377.662 393 555.668 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 15040.590 14 1074.328 2.451 0.003 INOC 6174.876 6 1029.146 2.348 0.031 SOIL 10356.055 5 2071.211 4.725 0.000 INOC x SOIL 8249.684 30 274.989 0.627 0.939 ERROR 172259.941 393 438.320 272 Table A5.3 ANOVA for mycorrhizal spruce seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = p + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOIL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 952.329 14 68.023 0.671 0.799 INOC 300.877 6 50.146 0.495 0.811 SOIL 1196.783 5 239.357 2.361 0.044 INOC x SOIL 2777.275 30 92.576 0.913 0.600 ERROR 12369.433 122 101.389 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 6135.969 14 438.283 0.643 0.825 INOC 1384.005 6 230.667 0.338 0.915 SOIL 6788.777 5 1357.755 1.991 0.085 INOC x SOIL 17645.826 30 588.194 0.862 0.672 ERROR 83205.515 122 682.012 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 4550.793 14 325.057 0.757 0.713 INOC 4987.723 6 831.287 1.935 0.080 SOIL 2622.580 5 524.516 1.221 0.303 INOC x SOIL 13663.834 30 455.461 1.060 0.397 ERROR 52412.526 122 429.611 273 Table A5.4 ANOVA for pine seedling height and biomass (regardless of mycorrhizal status) after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = p + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOIL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 1354.391 14 96.742 1.423 0.137 INOC 1103.674 6 183.946 2.705 0.013 SOIL 1064.544 5 212.909 3.131 0.008 INOC x SOIL 3312.831 30 110.428 1.624 0.020 ERROR 38894.818 572 67.998 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 4533.011 14 323.787 0.559 0.897 INOC 6202.659 6 1033.776 1.786 0.100 SOUL 12384.779 5 2476.956 4.279 0.001 INOC x SOIL 15414.401 30 513.813 0.888 0.641 ERROR 331089.727 572 578.828 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 13716.986 14 979.785 1.472 0.116 INOC 13860.006 6 2310.001 3.470 0.002 SOIL 7315.733 5 1463.147 2.198 0.053 INOC x SOIL 8128.726 30 270.958 0.407 0.998 ERROR 380775.538 572 665.691 274 Table A5.5 ANOVA for non-mycorrhizal pine seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = p + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOUL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 1057.007 14 75.500 1.136 0.325 INOC 587.776 6 97.963 1.473 0.186 SOIL 1064.826 5 212.965 3.203 0.008 INOC x SOIL 2498.375 30 83.279 1.253 0.175 ERROR 22670.834 341 66.483 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 4308.356 14 307.740 0.575 0.884 INOC 7560.067 6 1260.011 2.354 0.031 SOIL 15375.439 5 3075.088 5.745 0.000 INOC x SOIL 14958.172 30 498.606 0.932 0.573 ERROR 182512.386 341 535.227 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 11372.180 14 812.299 1.255 0.234 INOC 8702.166 6 1450.361 2.240 0.039 SOIL 7908.921 5 1581.784 2.443 0.034 INOC x SOIL 9617.732 30 320.591 0.495 0.989 ERROR 220752.974 341 647.369 275 Table A5.6 ANOVA for mycorrhizal pine seedling height and biomass after inoculation with PGPR and addition of forest floor soil collected from the Mackenzie, Salmon Arm or Williams Lake sites. ANOVA MODEL: Y = p. + BLOCK + INOC + SOIL + INOC x SOIL, where Y: effect, p: constant, BLOCK: block, INOC: inoculation with PGPR including uninoculated controls, and SOIL: forest floor soil type. (a) Seedling height SOURCE SS DF MS F-ratio P BLOCK 1001.996 14 71.571 1.076 0.382 INOC 1040.609 6 173.435 2.607 0.019 SOIL 98.384 5 19.677 0.296 0.915 INOC x SOIL 3315.237 30 110.508 1.661 0.024 ERROR 11641.188 175 66.521 (b) Shoot biomass SOURCE SS DF MS F-ratio P BLOCK 14936.248 14 1066.875 1.836 0.037 INOC 7029.390 6 1171.565 2.016 0.066 SOIL 2160.543 5 432.109 0.744 0.592 INOC x SOIL 19321.612 30 644.054 1.108 0.331 ERROR 101697.030 175 581.126 (c) Root biomass. SOURCE SS DF MS F-ratio P BLOCK 9967.935 14 711.995 1.006 0.450 INOC 7704.708 6 1284.118 1.814 0.100 SOIL 3314.854 5 662.971 0.936 0.459 INOC x SOIL 17409.942 30 580.331 0.820 0.735 ERROR 123901.825 175 708.010 276 A P P E N D I X 6 Statistical analyses for data presented in Chapter 5 277 Table A6.1 Descriptive statistics for spruce seedling biomass before and after outplanting. (a) Before outplanting (i.e., at the end of the period of greenhouse growth) at the Smithers sites. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 20 139.4 5.1 150.7 6.5 Pw2R 20 115.9 6.4 136.6 6.9 S20R 20 125.1 6.5 129.4 6.1 Sm3RN 20 101.7 4.8 119.3 5.1 Ss2RN 20 121.1 6.2 132.3 5.9 Sw5RN 20 125.6 6.3 136.1 6.5 Uninoculated control 20 103.1 6.1 116.8 6.3 (b) Before outplanting (i.e., at the end of the period of greenhouse growth) at theWillaims Lake sites. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 20 143.4 5.8 153.7 5.1 Pw2R 20 107.8 5.4 123.6 6.6 S20R 20 107.5 4.7 135.3 6.3 Sm3RN 20 103.8 5.7 125.6 5.2 Ss2RN 20 118.9 5.2 137.8 6.1 Sw5RN 20 118.8 5.9 138.1 6.7 Uninoculated control 20 107.9 4.7 116.0 9.0 278 Table A6.1 Continued. (c) After outplanting (i.e., at the end of the period of field growth) at the Smithers Blunt Creek site. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 26 203.9 8.1 172.9 13.0 Pw2R 26 158.3 7.0 162.5 11.7 S20R 25 155.4 5.4 126.9 8.5 Sm3RN 24 143.8 7.5 132.9 10.4 Ss2RN 25 159.8 8.6 163.8 8.8 Sw5RN 26 187.5 8.4 162.0 11.4 Uninoculated control 24 145.0 7.2 127.2 11.8 (d) After outplanting (i.e., at the end of the period of field growth) at the Smithers Shoe-house site. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 26 239.1 7.1 171.7 9.8 Pw2R 25 194.9 9.5 163.7 13.4 S20R 23 203.3 11.4 161.0 15.1 Sm3RN 26 167.4 6.8 137.5 7.6 Ss2RN 27 222.3 15.7 163.6 12.0 Sw5RN 26 221.6 11.0 156.7 15.0 Uninoculated control 22 160.1 7.2 132.1 6.2 279 Table A6.1 Continued. (e) After outplanting (i.e., at the end of the period of field growth) at the Williams Lake Regular cut-block site. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 25 212.1 8.4 183.6 11.8 Pw2R 25 160.9 5.7 139.4 6.4 S20R 22 163.5 9.4 147.6 13.2 Sm3RN 25 150.9 8.1 145.8 8.5 Ss2RN 27 186.9 6.6 162.8 8.7 Sw5RN 26 178.4 5.9 152.1 6.9 Uninoculated control 24 145.3 4.1 133.7 7.0 (f) After outplanting (Le., at the end of the period of field growth) at the Williams Lake Landing site. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 23 222.7 9.9 210.3 11.7 Pw2R 22 204.8 9.8 162.4 9.4 S20R 21 173.2 10.7 155.6 12.8 Sm3RN 26 183.2 7.9 164.9 13.3 Ss2RN 23 215.7 9.7 208.4 19.1 Sw5RN 20 209.9 12.2 192.2 24.1 Uninoculated control 21 160.7 10.8 134.0 8.6 280 Table A6.2 Descriptive statistics for relative growth rate (RGR) during the period of field growth. (a) Smithers Blunt Creek site. Shoot (mg-g"1-day"1) Root (mg-g ^ -day1) Treatment n Mean S.E. Mean S.E. L6-16R 27 2.972 0.338 0.549 0.624 Pw2R 27 2.371 0.373 0.962 0.544 S20R 27 1.601 0.295 -0.597 0.558 Sm3RN 27 2.476 0.442 0.191 0.687 Ss2RN 27 1.941 0.504 1.672 0.432 Sw5RN 27 3.087 0.397 0.976 0.540 Uninoculated control 27 2.179 0.425 0.015 0.605 (b) Smithers Shoe-house site. Shoot (mg-g ^day"1) Root (mg-g"1-day"1) Treatment n Mean S.E. Mean S.E. L6-16R 27 4.344 0.307 0.708 0.510 Pw2R 27 3.913 0.416 0.962 0.527 S20R 27 3.305 0.470 1.031 0.591 Sm3RN 27 3.941 0.356 0.864 0.448 Ss2RN 27 4.736 0.524 1.519 0.550 Sw5RN 27 4.425 0.438 0.877 0.436 Uninoculated control 27 2.923 0.410 0.701 0.329 281 Table A6.2 Continued. (c) Williams Lake Regular cut-block site. Shoot (mg-g^ -day"1) Root (mg-g'^ day"1) Treatment n Mean S.E. Mean S.E. L6-16R 27 2.996 0.367 1.061 0.495 Pw2R 27 3.100 0.340 0.763 0.395 S20R 27 2.733 0.471 0.122 0.557 Sm3RN 27 2.777 0.415 0.878 0.466 Ss2RN 27 3.789 0.314 1.122 0.473 Sw5RN 27 3.280 0.319 0.550 0.397 Uninoculated control 27 2.231 0.260 0.844 0.422 (d) Williams Lake Landing site. Shoot (mg-g^ -day"1) Root (mg-g^ -day"1) Treatment n Mean S.E. Mean S.E. L6-16R 27 3.115 0.397 2.100 0.414 Pw2R 27 4.359 0.546 1.676 0.436 S20R 27 2.954 0.536 0.542 0.512 Sm3RN 27 4.719 0.389 1.714 0.652 Ss2RN 27 4.262 0.466 2.442 0.672 Sw5RN 27 3.444 0.558 1.310 0.737 Uninoculated control 27 2.399 0.511 0.729 0.407 282 Table A6.3 ANOVA for seedling growth in the greenhouse after PGPR inoculation. ANOVA MODEL: Y = p. + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated controls. (a) Shoot growth of Mackenzie provenance (outplanted at the Smithers site). SOURCE SS DF MS F-ratio P INOC ERROR 21287.386 94690.500 6 133 3547.898 711.959 4.983 0.000 (b) Root growth of Mackenzie provenance (outplanted at the Smithers site). SOURCE SS DF MS F-ratio P INOC ERROR 15667.271 102782.700 6 133 2611.212 772.802 3.379 0.004 (c) Shoot growth of Williams Lake provenance (outplanted at the Williams Lake site). SOURCE SS DF MS F-ratio P INOC ERROR 22351.571 76114.400 6 133 3725.262 572.289 6.509 0.000 (d) Shoot growth of Williams Lake provenance (outplanted at the Williams Lake site). SOURCE SS DF MS F-ratio P INOC ERROR 18384.886 93543.650 6 133 3064.148 703.336 4.357 0.000 283 Table A6.4 ANOVA for shoot relative growth rate (SRGR) during the period of growth in the field. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated controls. (a) Smithers Blunt Creek site. SOURCE SS DF MS F-ratio P BLOCK 110.086 26 4.234 0.970 0.511 INOC 45.899 6 7.650 1.753 0.112 ERROR 680.711 156 4.364 (b) Smithers Shoe-house site. SOURCE SS DF MS F-ratio P BLOCK 199.076 26 7.657 1.761 0.019 INOC 66.720 6 11.120 2.557 0.022 ERROR 678.324 156 4.348 (c) Williams Lake Regular cut-block site. SOURCE SS DF MS F-ratio P BLOCK 132.558 26 5.098 1.566 0.050 INOC 38.394 6 6.399 1.966 0.074 ERROR 507.871 156 3.256 (d) Williams Lake Landing site. SOURCE SS DF MS F-ratio P BLOCK 279.663 26 10.756 1.859 0.011 INOC 118.457 6 19.743 3.412 0.003 ERROR 902.751 156 5.787 284 Table A6.5 ANOVA for root relative growth rate (RRGR) during the period of growth in the field. ANOVA MODEL: Y = p + BLOCK + INOC, where Y: effect, p: constant, and INOC: inoculation with PGPR including uninoculated controls. (a) Smithers Blunt Creek site. SOURCE SS DF MS F-ratio P BLOCK 334.980 26 12.884 1.560 0.052 INOC 90.180 6 15.030 1.820 0.099 ERROR 1288.521 156 8.260 (b) Smithers Shoe-house site. SOURCE SS DF MS F-ratio P BLOCK 389.563 26 14.983 2.940 0.000 INOC 12.525 6 2.087 0.410 0.872 ERROR 795.142 156 5.097 (c) Williams Lake Regular cut-block site. SOURCE SS DF MS F-ratio P BLOCK 209.505 26 8.058 1.506 0.067 INOC 18.731 6 3.122 0.583 0.743 ERROR 834.749 156 5.351 (d) Williams Lake Landing site. SOURCE SS DF MS F-ratio P BLOCK 637.788 26 24.530 4.114 0.000 INOC 77.539 6 12.923 2.167 0.049 ERROR 930.214 156 5.963 285 Table A6.6 Descriptive statistics for spruce seedling shoot damage (rank) after growth in the field. (a) After outplanting (i.e., at the end of the period of field growth) at the Smithers Blunt Creek site. Treatment n Mean S.E. L6-16R 27 2.04 0.17 Pw2R 27 1.81 0.16 S20R 27 1.33 0.15 Sm3RN 27 1.74 0.19 Ss2RN 27 1.85 0.17 Sw5RN 27 1.81 0.18 Uninoculated control 27 1.44 0.19 (b) After outplanting (i.e., at the end of the period of field growth) at the Smithers Shoe-house site. Treatment n Mean S.E. L6-16R 27 1.85 0.16 Pw2R 27 1.93 0.17 S20R 27 1.63 0.20 Sm3RN 27 1.52 0.13 Ss2RN 27 2.00 0.14 Sw5RN 27 1.96 0.16 Uninoculated control 27 1.41 0.19 286 Table A6.6 Continued. (c) After outplanting (i.e., at the end of the period of field growth) at the Williams Lake Regular cut-block site. Treatment n Mean S.E. L6-16R 27 1.89 0.19 Pw2R 27 1.70 0.18 S20R 27 1.44 0.19 Sm3RN 27 1.52 0.15 Ss2RN 27 1.96 0.15 Sw5RN 27 1.81 0.16 Uninoculated control 27 1.48 0.17 (d) After outplanting (i.e., at the end of the period of field growth) at the Williams Lake Landing site. Treatment n Mean S.E. L6-16R 27 1.93 0.21 Pw2R 27 1.85 0.22 S20R 27 1.26 0.18 Sm3RN 27 1.78 0.12 Ss2RN 27 1.85 0.20 Sw5RN 27 1.56 0.23 Uninoculated control 27 1.30 0.18 287 Table A6.7 Descriptive statistics for inoculum population sizes reisolated from external and internal root tissues of spruce seedlings before and after the field trials. (a) Before outplanting (i.e., at the end of the period of greenhouse growth) at the Smithers sites. External root Internal root Treatment n Mean (cfu-g 1 soil) S.E. Mean (cfu-g 1soil) S.E. L6-16R 4 4.25 0.36 0 0 Pw2R 4 4.70 0.09 2.39 1.95 S20R 4 5.01 0.31 0 0 Sm3RN 4 5.56 0.23 4.55 0.19 Ss2RN 4 6.58 0.38 2.17 1.26 Sw5RN 4 5.87 0.35 0 0 (b) Before outplanting (i.e., at the end of the period of greenhouse growth) at the Williams Lake sites. External root Internal root Treatment n Mean (cfu-g 1soil) S.E. Mean (cfb-g 1soil) S.E. L6-16R 4 4.66 0.29 0 0 Pw2R 4 5.12 1.33 3.47 1.16 S20R 4 4.92 0.27 0 0 Sm3RN 4 5.67 0.29 2.37 1.36 Ss2RN 4 6.31 0.33 0 0 Sw5RN 4 5.80 0.28 0 0 (c) After outplanting (i.e., at the end of the period of field growth) at the Smithers Blunt Creek site. External root Internal root Treatment n Mean (cfu-g 1soil) S.E. Mean (cfii-g soil) S.E. L6-16R 4 4.91 0.23 0 0 Pw2R 4 4.00 0.22 1.83 1.07 S20R 4 3.11 1.38 0 0 Sm3RN 4 4.74 1.29 4.17 3.44 Ss2RN 4 4.19 1.60 0 0 Sw5RN 4 5.34 0.05 0 0 288 Table A6.7 Continued. (d) After outplanting (i.e., at the end of the period of field growth) at the Smithers Shoe-house site. External root Internal root Treatment n Mean (cfu-g 1soil) S.E. Mean (cfu-g 1soil) S.E. L6-16R 4 2.53 1.48 0 0 Pw2R 4 4.86 0.14 2.59 0.89 S20R 4 2.42 1.40 0 0 Sm3RN 4 4.27 1.48 0.98 0.98 Ss2RN 4 4.85 0.23 1.54 0.89 Sw5RN 4 5.34 0.09 0 0 (e) After outplanting (i.e., at the end of the period of field growth) at the Williams Lake Regular cut-block site. External root Internal root Treatment n Mean (cfu-g1 soil) S.E. Mean (cfu-g ^ soil) S.E. L6-16R 4 2.51 1.45 0 0 Pw2R 4 3.82 1.28 1.99 1.15 S20R 4 3.48 1.16 0 0 Sm3RN 4 5.60 0.24 4.44 1.21 Ss2RN 4 2.26 1.31 0 0 Sw5RN 4 5.70 0.13 1.91 1.14 (f) After outplanting (i.e., at the end of the period of field growth) at the Williams Lake Landing site. External root Internal root Treatment n Mean (cfu-g 1 soil) S.E. -1 Mean (cfu-g soil) S.E. L6-16R 4 3.44 1.15 0 0 Pw2R 4 3.46 1.16 0 0 S20R 4 2.27 1.33 0 0 Sm3RN 4 3.99 1.34 1.27 1.27 Ss2RN 4 3.13 1.05 0 0 Sw5RN 4 5.26 0.205 0 0 289 APPENDIX 7 Statistical analyses for data presented in Chapter 290 Table A7.1 Descriptive statistics for spruce seedling biomass grown under gnotobiotic conditions (sterile sand) in a growth chamber after inoculation with PGPR. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 20 24.2 0.6 22.9 0.9 Pw2R 20 23.7 0.7 22.9 1.6 S20R 20 23.7 1.1 22.4 1.3 Sm3RN 20 24.8 0.6 24.3 1.2 Ss2RN 20 22.4 0.9 22.3 1.2 Sw5RN 20 25.0 1.1 22.6 1.0 Uninoculated control 20 23.3 0.7 22.4 1.0 291 Table A7.2 Descriptive statistics for spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site. (a) Sterilized soil extract. Shoot Root Treatment n Mean(mg) S.E. Mean(mg) S.E. L6-16R 18 31.3 1.1 19.2 0.9 Pw2R 18 32.0 1.0 17.8 0.9 S20R 18 32.8 1.0 16.9 0.6 Sm3RN 18 34.8 0.9 16.9 0.6 Ss2RN 18 31.5 1.1 16.7 0.7 Sw5RN 18 33.3 1.0 19.1 0.9 Uninoculated control 18 32.3 1.1 17.1 0.7 (b) Nonsterile soil extract. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 18 18.6 0.9 13.4 0.8 Pw2R 18 18.5 0.8 12.9 0.8 S20R 18 18.9 0.9 11.6 0.6 Sm3RN 18 19.1 0.8 11.4 0.5 Ss2RN 18 16.8 0.5 11.9 0.6 Sw5RN 18 17.7 0.8 11.9 0.8 Uninoculated control 18 16.9 0.9 11.1 0.5 292 Table A7.3 Descriptive statistics for spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing site. (a) Sterilized soil extract. Shoot Root Treatment n Mean(mg) S.E. Mean(mg) S.E. L6-16R 18 32.8 0.8 18.2 0.9 Pw2R 18 31.6 0.8 18.3 0.9 S20R 18 31.4 1.0 18.8 0.7 Sm3RN 18 32.1 0.9 18.7 0.6 Ss2RN 18 30.7 0.9 17.7 1.0 Sw5RN 18 31.6 0.8 18.4 0.6 Uninoculated control 18 28.8 0.8 17.5 0.7 (b) Nonsterile soil extract. Shoot Root Treatment n Mean (mg) S.E. Mean (mg) S.E. L6-16R 18 27.4 1.3 15.6 0.7 Pw2R 18 25.7 1.1 16.2 0.7 S20R 18 26.4 1.2 15.7 0.7 Sm3RN 18 27.1 0.9 18.6 1.0 Ss2RN 18 24.8 1.0 15.6 0.7 Sw5RN 18 26.8 1.2 17.7 0.6 Uninoculated control 18 23.4 1.2 14.3 0.7 293 Table A7.4 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown under gnotobiotic conditions (sterile sand). Treatment n Rhizosphere Internal root Mean (cfu-tube *) S.E. Mean (cfu-g tissue) S.E. L6-16R 4 3.87 0.48 0.00 0.00 Pw2R 4 5.58 0.04 4.97 0.60 S20R 4 4.81 0.24 0.00 0.00 Sm3RN 4 6.11 0.04 5.69 0.15 Ss2RN 4 5.42 0.19 0.00 0.00 Sw5RN 4 4.89 0.20 0.00 0.00 294 Table A7.5 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown with soil extract collected from the Smithers Shoe-house site. (a) Sterilized soil extract. Treatment n Rhizosphere Internal root Mean (cfu-tube *) S.E. Mean (cfu-g tissue) S.E. L6-16R 4 3.94 0.69 0.00 0.00 Pw2R 4 4.18 0.16 4.11 0.18 S20R 4 3.53 0.36 0.00 0.00 Sm3RN 4 4.69 0.74 2.80 0.93 Ss2RN 4 2.77 0.18 0.00 0.00 Sw5RN 4 4.84 0.79 0.00 0.00 (b) Non-sterilized soil extract. Treatment n Rhizosphere Internal root Mean (cfu-tube"1) S.E. Mean (cfu-g t^issue) S.E. L6-16R 4 3.70 0.58 0.00 0.00 Pw2R 4 2.98 0.41 0.00 0.00 S20R 4 2.82 0.20 0.00 0.00 Sm3RN 4 4.31 0.10 3.96 0.12 Ss2RN 4 3.08 0.41 0.00 0.00 Sw5RN 4 3.95 0.67 0.00 0.00 295 Table A7.6 Descriptive statistics for PGPR populations re-isolated from spruce seedlings grown with soil extract collected from the Williams Lake landing site. (b) Sterilized soil extract. Treatment n Rhizosphere Internal root Mean (cfu-tube"1) S.E. Mean (cfu-g tissue) S.E. L6-16R 4 2.88 0.31 0.00 0.00 Pw2R 4 3.70 0.42 3.83 0.16 S20R 4 3.32 0.49 0.00 0.00 Sm3RN 4 5.49 0.13 0.00 0.00 Ss2RN 4 2.85 0.15 0.00 0.00 Sw5RN 4 4.93 0.60 0.00 0.00 (b) Non-sterilized soil extract. Treatment n Rhizosphere Internal root Mean (cfu-tube"1) S.E. Mean (cfu-g tissue) S.E. L6-16R 4 2.67 0.19 0.00 0.00 Pw2R 4 3.42 0.55 0.00 0.00 S20R 4 3.11 0.37 0.00 0.00 Sm3RN 4 4.45 0.21 0.00 0.00 Ss2RN 4 2.79 0.11 0.00 0.00 Sw5RN 4 4.31 0.41 0.00 0.00 296 Table A7.7 ANOVA of shoot and root biomass of spruce seedlings inoculated with PGPR and grown without forest soil extract. ANOVA MODEL: Y = p + INOC, where Y: effect, p: constant, and INOC: inoculation of PGPR including uninoculated controls. (a) Shoot biomass SOURCE SS DF MS F-ratio P INOC 94.843 6 15.807 1.134 0.346 ERROR 1854.300 133 13.942 (b) Root biomass SOURCE SS DF MS F-ratio P INOC 58.943 6 9.824 0.346 0.911 ERROR 3774.850 133 28.382 (c) Root/shoot ratio SOURCE SS DF MS F-ratio P INOC 0.108 6 0.018 0.285 0.943 ERROR 8.410 133 0.063 297 Table A7.8 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house and Williams Like landing sites. ANOVA MODEL: Y = u, + INOC + SITE + STERIL + INOC x SITE + INOC x STERTL + SITE x STERIL+ INOC x SITE x STERIL, where Y: effect, p.: constant, INOC: inoculation of PGPR including uninoculated controls, SITE: soil collection site, and STERIL: soil sterilization treatment. (a) Shoot biomass SOURCE SS DF MS F-ratio P INOC 422 069 6 70.345 4.193 0.000 SITE 1324 718 1 1324.718 78.965 0.000 STERIL 12172 302 1 12172.302 725.582 0.000 INOC x SITE 137 019 6 22.836 1.361 0.229 INOC x STERIL 29 582 6 4.930 0.294 0.940 SITE x STERIL 2599 944 1 2599.944 154.981 0.000 INOC x SITE x STERIL 48 936 6 8.156 0.486 0.819 ERROR 7851 127 468 16.776 (b) Root biomass. SOURCE SS DF MS F-ratio P INOC 177 749 6 29.625 3.066 0.006 SITE 699 727 1 699.727 72.409 0.000 STERIL 1810 268 1 1810.268 187.331 0.000 INOC x SITE 150 103 6 25.017 2.589 0.018 INOC x STERIL 37 995 6 6.333 0.655 0.686 SITE x STERIL 416 924 1 416.924 43.144 0.000 INOC x SITE x STERIL 66 535 6 11.089 1.148 0.334 ERROR 4522 510 468 9.663 298 Table A7.9 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site. ANOVA MODEL: Y = p + INOC + STERTL + INOC x STERTL, where Y: effect, p: constant, INOC: inoculation of PGPR including uninoculated controls, and STERTL: soil sterilization treatment. (a) Shoot biomass. SOURCE SS DF MS F-ratio P INOC 172.526 6 28.754 1.900 0.082 STERTL 12958.171 1 12958.171 856.225 0.000 INOC x STERTL 69.496 6 11.583 0.765 0.598 ERROR 3526.239 233 15.134 (b) Root biomass. SOURCE SS DF MS F-ratio P INOC 155.921 6 25.987 2.911 0.009 STERTL 1974.199 1 1974.199 221.116 0.000 INOC x STERTL 32.163 6 5.360 0.600 0.730 ERROR 2080.307 233 8.928 (c) Root/shoot ratio SOURCE SS DF MS F-ratio P INOC 0.398 6 0.066 2.235 0.041 STERTL 1.104 1 1.104 37.206 0.000 INOC x STERTL 0.021 6 0.004 0.120 0.994 ERROR 6.912 233 0.030 299 Table A7.10 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing site. ANOVA MODEL: Y = p + INOC + STERTL + TNOC x STERTL, where Y: effect, p: constant, TNOC: inoculation of PGPR including uninoculated controls, and STERTL: soil sterilization treatment. (a) Shoot biomass. SOURCE SS DF MS F-ratio P TNOC 381.815 6 63.636 3.458 0.003 STERTL 1767.835 1 1767.835 96.058 0.000 TNOC x STERTL 8.738 6 1.456 0.079 0.998 ERROR 4324.889 235 18.404 (b) Root biomass. SOURCE SS DF MS F-ratio P TNOC 172.716 6 28.786 2.770 0.013 STERTL 245.851 1 245.851 23.657 0.000 TNOC x STERTL 73.221 6 12.204 1.174 0.321 ERROR 2442.203 235 10.392 (a) Root/shoot ratio. SOURCE SS DF MS F-ratio P TNOC 0.117 6 0.020 1.128 0.347 STERTL 0.121 1 0.121 6.978 0.009 TNOC x STERTL 0.095 6 0.016 0.910 0.488 ERROR 4.068 235 0.017 300 Table A7.l l ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Smithers Shoe-house site. ANOVA MODEL: Y = p + INOC, where Y: effect, p.: constant, and INOC: inoculation of PGPR including uninoculated controls. (a) Shoot biomass with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 150.964 2201.641 6 117 25.161 18.817 1.337 0.246 (b) Root biomass with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 116.352 1170.382 6 117 19.392 10.003 1.939 0.080 (c) Root/shoot ratio with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 0.195 1.855 6 117 0.033 0.016 2.054 0.064 301 Table A7.l l Continued. (d) Shoot biomass with non-sterilized soil extract. SOURCE SS DF MS F-ratio P INOC 91.418 6 15.236 1.334 0.248 ERROR 1324.598 116 11.419 (e) Root biomass with non-sterilized soil extract. SOURCE SS DF MS F-ratio P INOC 73.002 6 12.167 1.551 0.168 ERROR 909.925 116 7.844 (f) Root/shoot ratio with non-sterilized soil extract. SOURCE SS DF MS F-ratio P INOC 0.224 6 0.037 0.858 0.528 ERROR 5.056 116 0.044 302 Table A7.12 ANOVA of spruce seedling biomass after inoculation with PGPR and soil extracts collected from the Williams Lake landing. ANOVA MODEL: Y = u, + INOC, where Y: effect, u,: constant, and INOC: inoculation of PGPR including uninoculated controls. (a) Shoot biomass with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 170.385 1575.163 6 117 28.397 13.463 2.109 0.057 (b) Root biomass with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 23.893 1263.784 6 117 3.982 10.802 0.369 0.898 (c) Root/shoot ratio with sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 0.040 2.220 6 117 0.007 0.019 0.351 0.908 303 Table A7.12 Continued. (d) Shoot biomass with non-sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 220.627 2749.725 6 118 36.771 23.303 1.578 0.159 (e) Root biomass with non-sterilized soil extract SOURCE SS DF MS F-ratio P INOC ERROR 222.750 1178.418 6 118 37.125 9.987 3.717 0.002 (f) Root/shoot with non-sterilized soil extract. SOURCE SS DF MS F-ratio P INOC ERROR 0.171 1.848 6 118 0.028 0.016 1.815 0.102 304 Table A7.13 PGPR survival in microcosms treated with two forest soils with or without spruce seedlings. (a) Smithers soil. Seedling present Soil only Treatment n Mean (log cfu-g 1 soil) S.E. Mean (log cfu-g"1 soil) S.E. L6-16R 4 4.00 0.58 2.90 0.12 Pw2R 4 3.12 0.20 2.90 0.12 S20R 4 3.28 0.41 3.95 0.46 Sm3RN 4 4.61 0.10 4.76 0.27 Ss2RN 4 3.38 0.41 3.00 0.14 Sw5RN 4 4.25 0.67 4.76 0.43 (b) Williams Lake soil. Seedling present Soil only Treatment n Mean (log cfu-g 1 soil) S.E. Mean (log cfu-g"1 soil) S.E. L6-16R 4 2.97 0.19 2.90 0.12 Pw2R 4 3.72 0.55 3.35 0.57 S20R 4 3.41 0.37 3.24 0.27 Sm3RN 4 4.75 0.21 4.14 0.52 Ss2RN 4 3.09 0.11 3.00 0.14 Sw5RN 4 4.61 0.41 4.11 0.36 305 Table A7.14 ANOVA of PGPR population sizes recovered from microcosms treated with Smithers and Williams Lake soil and spruce seedlings. ANOVA MODEL: Y = p + TNOC + SOIL + PLT + INOC x SOIL + INOC x PLT + SOIL x PLT + INOC x SOIL x PLT, where Y: effect, p: constant, and INOC: inoculation of PGPR including uninoculated controls, SOIL: soil type, i.e., Smithers or Williams Lake, and PLT: presence or absence of spruce seedling. SOURCE SS DF MS F-ratio P INOC 38.239 5 7.648 14.861 0.000 SOIL 0.145 1 0.145 0.281 0.598 PLT 0.469 1 0.469 0.912 0.343 INOC x INOC 2.181 5 0.436 0.848 0.521 INOC x PLT 1.573 5 0.315 0.611 0.692 SOIL x PLT 0.091 1 0.091 0.176 0.676 TNOC x SOIL x PLT 3.039 5 0.608 1.181 0.327 ERROR 37.053 72 0.515 306 Table A7.15 ANOVA of PGPR population sizes recovered from microcosms treated with Smithers and Williams Lake soil and spruce seedlings. ANOVA MODEL: Y = p + SOIL + PLT + SOIL x PLT, where Y: effect, p: constant, SOUL: soil type, i.e., Smithers or Williams Lake, and PLT: presence of spruce seedling. (a) L6-16R. SOURCE SS DF MS F-ratio P SOIL 1.053 1 1.053 2.644 0.130 PLT 1.384 1 1.384 3.476 0.087 SOIL x PLT 1.053 1 1.053 2.644 0.130 ERROR 4.780 12 0.398 (b) Pw2R. SOURCE SS DF MS F-ratio P SOIL 0.028 1 0.028 0.028 0.871 PLT 0.092 1 0.092 0.090 0.769 SOIL x PLT 1.077 1 1.077 1.059 0.324 ERROR 12.209 12 1.017 (c) S20R. SOURCE SS DF MS F-ratio P SOIL 0.394 1 0.394 1.510 0.243 PLT 0.158 1 0.158 0.605 0.452 SOIL x PLT 0.003 1 0.003 0.012 0.914 ERROR 3.128 12 0.261 307 Table A7.15 Continued. (d) Sm3RN. SOURCE SS DF MS F-ratio P SOIL 0.427 1 0.427 1.045 0.327 PLT 0.075 1 0.075 0.184 0.675 SOIL x PLT 0.893 1 0.893 2.186 0.165 ERROR 4.904 12 0.409 (e) Ss2RN. SOURCE SS DF MS F-ratio P SOIL 0.086 1 0.086 0.390 0.544 PLT 0.220 1 0.220 0.999 0.337 SOIL x PLT 0.086 1 0.086 0.390 0.544 ERROR 2.644 12 0.220 (f) Sw5RN. SOURCE SS DF MS F-ratio P SOIL 0.338 1 0.338 0.432 0.523 PLT 0.113 1 0.113 0.144 0.711 SOIL x PLT 0.017 1 0.017 0.022 0.885 ERROR 9.389 12 0.782 308 Table A7.16 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings. ANOVA MODEL: Y = p. + INOC + SOIL + PLT + INOC x SOIL + INOC x PLT + SOIL x PLT + INOC x SOIL x PLT, where Y: effect, p.: constant, INOC: inoculation of PGPR including uninoculated controls, SOIL: soil type, i.e., Smithers or Williams Lake, and PLT: presence of spruce seedling. (a) Viable fiingal population (log transformed). SOURCE SS DF MS F-ratio P INOC 0.916 6 0.153 1.948 0.082 SOIL 3.452 1 3.452 44.034 0.000 PLT 0.005 1 0.005 0.063 0.802 INOC x SOIL 0.369 6 0.062 0.785 0.584 INOC x PLT 0.491 6 0.082 1.045 0.403 SOIL x PLT 0.555 1 0.555 7.634 0.007 INOC x SOIL x PLT 0.698 6 0.116 1.485 0.193 ERROR 6.664 85 0.078 (b) Viable actinomycete population (log transformed). SOURCE SS DF MS F-ratio P INOC 4.342 6 0.724 2.308 0.041 SOIL 2.982 1 2.982 9.514 0.003 PLT 10.838 1 10.838 34.578 0.000 INOC X SOIL 1.944 6 0.324 1.033 0.409 INOC x PLT 1.153 6 0.192 0.613 0.719 SOIL x PLT 1.706 1 1.706 5.745 0.019 INOC x SOUL x PLT 1.303 6 0.217 0.693 0.656 ERROR 26.643 85 0.313 309 Table A7.16 Continued. (c) Viable bacterial population (log transformed). SOURCE SS DF MS F-ratio P INOC 2.257 6 0.376 6.715 0.000 SOIL 0.538 1 0.538 9.605 0.003 PLT 1.384 1 1.384 24.698 0.000 INOC x SOIL 1.807 6 0.301 5.376 0.000 INOC x PLT 1.039 6 0.173 3.089 0.009 SOIL x PLT 0.000 1 0.000 0.000 0.994 TNOC x SOIL x PLT 1.363 6 0.227 4.054 0.001 ERROR 4.763 85 0.056 310 Table A7.17 ANOVA of soil microbial populations detected in Smithers soil after inoculation with PGPR in the presence or absence of spruce seedlings. ANOVA MODEL: Y = p + INOC + PLT + INOC x PLT where Y: effect, p: constant, INOC: inoculation of PGPR including uninoculated controls, and PLT: presence of spruce seedling. (a) Viable fungal population (log transformed). SOURCE SS DF MS F-ratio P INOC 0.988 6 0.165 2.637 0.029 PLT 0.333 1 0.333 5.325 0.026 INOC x PLT 0.930 6 0.155 2.482 0.038 ERROR 2.623 42 0.062 (b) Viable actinomycete population (log transformed). SOURCE SS DF MS F-ratio P INOC 2.840 6 0.473 1.726 0.139 PLT 10.572 1 10.572 38.545 0.000 INOC x PLT 1.431 6 0.238 0.869 0.525 ERROR 11.519 42 0.274 (c) Viable bacterial population (log transformed). SOURCE SS DF MS F-ratio P INOC 1.701 6 0.283 8.093 0.000 PLT 0.694 1 0.694 19.817 0.000 INOC x PLT 0.900 6 0.150 4.285 0.002 ERROR 1.471 42 0.035 311 Table A7.18 ANOVA of soil microbial populations detected in Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings. ANOVA MODEL: Y = p + INOC + PLT + INOC x PLT where Y: effect, p: constant, INOC: inoculation of PGPR including uninoculated controls, and PLT: presence of spruce seedling. (a) Viable fungal population (log transformed). SOURCE SS DF MS F-ratio P INOC 0.297 6 0.050 0.597 0.731 PLT 0.228 1 0.228 2.742 0.105 INOC x PLT 0.260 6 0.043 0.522 0.789 ERROR 3.486 42 0.083 (b) Viable actinomycete population (log transformed). SOURCE SS DF MS F-ratio P INOC 3.445 6 0.574 1.797 0.123 PLT 1.973 1 1.973 6.174 0.017 INOC x PLT 1.026 6 0.171 0.535 0.778 ERROR 13.419 42 0.319 (c) Viable bacterial population (log transformed). SOURCE SS DF MS F-ratio P INOC 2.364 6 0.394 5.027 0.001 PLT 0.690 1 0.690 8.801 0.005 INOC x PLT 1.501 6 0.250 3.192 0.011 ERROR 3.292 42 0.078 312 Table A7.19 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the presence of spruce seedlings. ANOVA MODEL: Y = p + INOC + SOIL + INOC x SOIL where Y: effect, p: constant, INOC: inoculation of PGPR including uninoculated controls, and SOIL: soil type (i.e.,. Smithers or Williams Lake). (a) Viable fungal population (log transformed). SOURCE SS DF MS F-ratio P INOC 0.674 6 0.112 1.393 0.240 SOIL 3.388 1 3.388 41.999 0.000 INOC x SOIL 0.479 6 0.080 0.989 0.445 ERROR 3.388 42 0.081 (b) Viable actinomycete population (log transformed). SOURCE SS DF MS F-ratio P INOC 1.350 6 0.225 1.901 0.103 SOIL 4.599 1 4.599 38.873 0.000 INOC x SOIL 1.807 6 0.301 2.545 0.034 ERROR 4.969 42 0.118 (c) Viable bacterial population (log transformed). SOURCE SS DF MS F-ratio P INOC 1.503 6 0.251 3.244 0.010 SOIL 0.270 1 0.270 3.501 0.068 TNOC x SOUL 1.031 6 0.172 2.224 0.059 ERROR 3.244 42 0.077 313 Table A7.20 ANOVA of soil microbial populations detected in Smithers and Williams Lake soil after inoculation with PGPR in the absence of spruce seedlings. ANOVA MODEL: Y = p + INOC + SOIL + INOC x SOIL where Y: effect, p: constant, INOC: inoculation of PGPR including uninoculated controls, and SOUL: soil type (i.e.,. Smithers or Williams Lake). (a) Viable fungal population (log transformed). SOURCE SS DF MS F-ratio P INOC 0.733 6 0.122 1.887 0.106 SOIL 0.619 1 0.619 9.560 0.004 INOC x SOIL 0.589 6 0.098 1.515 0.197 ERROR 2.721 42 0.065 (b) Viable actinomycete population (log transformed). SOURCE SS DF MS F-ratio P INOC 4.145 6 0.691 1.453 0.218 SOIL 0.089 1 0.089 0.186 0.668 INOCx SOIL 1.440 6 0.240 0.505 0.801 ERROR 19.969 42 0.475 (c) Viable bacterial population (log transformed). SOURCE SS DF MS F-ratio P INOC 1.793 6 0.299 8.265 0.000 SOIL 0.268 1 0.268 7.407 0.009 INOC x SOIL 2.139 6 0.357 9.864 0.000 ERROR 1.518 42 0.036 314 Table A7.21 ANOVA of soil microbial populations detected in Smithers soil after inoculation with PGPR in the presence or absence of spruce seedlings. ANOVA MODEL: Y = p + TNOC, where Y: effect, p: constant, and INOC: inoculation of PGPR including uninoculated controls, (a) Viable fungal population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P INOC 0.736 6 0.123 2.059 0.102 ERROR 1.251 21 0.060 (b) Viable fungal population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P INOC 1.182 6 0.197 3.017 0.028 ERROR 1.372 21 0.065 (c) Viable actinomycete population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P INOC 0.809 6 0.135 1.891 0.130 ERROR 1.498 21 0.071 (d) Viable actinomycete population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P INOC 3.462 6 0.577 1.209 0.340 ERROR 10.021 21 0.477 315 Table A7.21 Continued. (e) Viable heterotrophic bacterial population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P INOC 0.613 6 0.102 2.333 0.070 ERROR 0.919 21 0.044 (f) Viable heterotrophic bacterial population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P INOC 1.989 6 0.331 12.608 0.000 ERROR 0.552 21 0.026 316 Table A7.22 ANOVA of soil microbial populations detected in Williams Lake soil after inoculation with PGPR in the presence or absence of spruce seedlings. ANOVA MODEL: Y = p. + INOC, where Y: effect, p: constant, and INOC: inoculation of PGPR including uninoculated controls. (a) Viable fungal population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P INOC 0.417 6 0.069 0.683 0.665 ERROR 2.137 21 0.102 (b) Viable fungal population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P INOC 0.140 6 0.023 0.364 0.894 ERROR 1.349 21 0.064 (c) Viable actinomycete population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P INOC 2.347 6 0.391 2.366 0.066 ERROR 3.471 21 0.165 (d) Viable actinomycete population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P INOC 2.124 6 0.354 0.747 0.618 ERROR 9.948 21 0.474 317 Table A7.22 Continued. (e) Viable heterotrophic bacterial population (log transformed) in the presence of a spruce seedling. SOURCE SS DF MS F-ratio P TNOC 1.921 6 0.320 2.892 0.032 ERROR 2.325 21 0.111 (f) Viable heterotrophic bacterial population (log transformed) with no seedling present. SOURCE SS DF MS F-ratio P TNOC 1.944 6 0.324 7.040 0.000 ERROR 0.966 21 0.046 318 APPENDIX 8 Regression equation of optical density (A^oo) and mean generation time of PGPR strains 319 Table A8.1 Regression equation of optical density (^ 60o) and mean generation time of six PGPR strains. Strain Regression equation (A^oo) Mean generation time (min.) Bacillus L6-16R 4.7+16.1x^600 (r2 = 0.96) 65 Pw2R 4.5 + 18.7x^600 (r2 = 0.96) 74 S20R 4.7+15.2x^600 (r2 = 0.95) 67 Pseudomonas Sm3RN 6.9+ 8.0 x^600 (r2 = 0.84) 65 Ss2RN 6.1+21.6x^600 (r2 = 0.93) 65 Sw5RN 6.6+10.1 x^ eoo (r2 = 0.98) 102 Valid where 0.02 < ^600 ^ 0.15 for Bacillus strains and 0.03 < ^600 ^ 0-22 for Pseudomonas strains. Cultured with 50% tryptic soy broth and King's B broth (120 rpm, 23°C) for the Bacillus and the Pseudomonas strains, respectively. Generation time = T ln2 / (In - In XQ), where T: time in culture, Xt: cell density (cfu-mL *) at time T, and XQ: initial cell density. 320 APPENDIX 9 Euclidean distance matrix for Biolog™ plate responses to forest soil extracts treated with PGPR and spruce seedlings, and the Shepard diagram used for multidimensional scaling of the Euclidean distances. 321 Table A9.1 Euclidean distances for Biolog plate responses to forest soil extracts treated with PGPR and spruce seedlings. S-C-P S-C-NP S-L6-P S-L6-NP S-Pw2-P S-Pw2-NP S-S20-P S-S20-NP S-Sm3-P S-Sm3-NP S-Ss2-P S-Ss2-NP S-Sw-5P S-Sw5-lSrP S-C-P 0.000 - - - - - - - •. - - - - - -S-C-NP 0.458 0.000 - - - - - - - - - - - -S-L6-P 0.283 0.329 0.000 - - - - - - - - - - -S-L6-NP 0.433 0.326 0.343 0.000 -• - - - - - - - - -S-Pw2-P 0.441 0.414 0.300 0.441 0.000 - - - - - - - - -S-Pw2-NP 0.456 0.371 0.366 0.344 0.472 0.000 - - - - - - - -S-S20-P 0.356 0.361 0.250 0.378 0.365 0.398 0.000 - - - - - - -S-S20-NP 0.531 0.384 0.422 0.390 0.482 0.366 0.415 0.000 - - - - - -S-Sm3-P 0.267 0.402 0.261 0.394 0.423 0.434 0.289 0.490 0.000 - - - - -S-Sm3-NP 0.353 0.402 0.326 0.405 0.456 0.454 0.320 0.471 0.251 0.000 - - - -S-Ss2-P 0.185 0.475 0.279 0.426 0.438 0.411 0.366 0.529 0.260 0.355 0.000 - - -S-Ss2-NP 0.430 0.428 0.392 0.391 0.456 0.403 0.413 0.417 0.445 0.430 0.417 0.000 - -S-Sw-5P 0.331 0.337 0.231 0.366 0.390 0.392 0.207 0.447 0.238 0.311 0.337 0.454 0.000 -S-Sw5-NP 0.618 0.435 0.496 0.433 0.528 0.497 0.501 0.454 0.565 0.520 0.609 0.458 0.523 0.000 W-C-P 0.484 0.617 0.489 0.590 0.559 0.581 0.553 0.678 0.543 0.579 0.486 0.543 0.542 0.767 W-C-NP 0.425 0.410 0.377 0.405 0.459 0.366 0.373 0.444 0.406 0.399 0.426 0.430 0.411 0.522 W-L6-P 0.481 0.471 0.429 0.509 0.501 0.555 0.481 0.638 0.506 0.542 0.528 0.607 0.490 0.629 W-L6-NP 0.455 0.403 0.383 0.392 0.494 0.431 0.395 0.452 0.435 0.426 0.486 0.512 0.404 0.488 W-Pw2-P 0.333 0.399 0.286 0.426 0.422 0.437 0.385 0.521 0.359 0.408 0.365 0.493 0.378 0.555 W-Pw2-NP 0.503 0.431 0.435 0.444 0.475 0.460 0.431 0.456 0.498 0.499 0.548 0.511 0.464 0.487 W-S20-P 0.376 0.366 0.313 0.451 0.446 0.449 0.376 0.517 0.395 0.428 0.407 0.532 0.371 0.583 W-S20-NP 0.503 0.603 0.484 0.526 0.567 0.455 0.490 0.575 0.499 0.528 0.453 0.460 0.519 0.701 W-Sm3-P 0.478 0.645 0.489 0.597 0.568 0.554 0.526 0.653 0.492 0.554 0.449 0.553 0.535 0.778 W-Sm3-NP 0.432 0.521 0.411 0.461 0.478 0.430 0.399 0.514 0.436 0.463 0.419 0.429 0.450 0.648 W-Ss2-P 0.396 0.445 0.350 0.517 0.437 0.493 0.396 0.555 0.431 0.465 0.445 0.553 0.419 0.611 W-Ss2-NP 0.496 0.436 0.389 0.435 0.451 0.429 0.396 0.467 0.458 0.460 0.508 0.523 0.433 0.518 W-Sw5-P 0.478 0.575 0.438 0.556 0.529 0.534 0.534 0.623 0.521 0.567 0.456 0.538 0.519 0.708 W-Sw5-NP 0.500 0.418 0.408 0.433 0.476 0.384 0.393 0.432 0.488 0.486 0.503 0.500 0.432 0.503 S:Smithers soil; W:Williams Lake soil; C:Uninoculated control; L6:L6-16R; Pw2:Pw2R; S20:S20R; Sm3:Sm3RN; Ss2:Ss2RN; Sw5:Sw5RN; RSeedling present; NP: Soil only. Table A9.1 Continued. W-C-P W-C-NP W-L6-P W-L6-NP W-Pw2-P W-Pw2-N W-S20-P W-S20-NP W-Sm3-P W-Sm3-N W-Ss2-P W-Ss2-MP W-Sw5-P W-Sw5-NP W-C-P 0.000 - - - - - - - - - - - -W-C-NP 0.558 0.000 - - - - - - - - - - -W-L6-P 0.583 0.505 0.000 - - - - - - - - - -W-L6-NP 0.576 0.306 0.473 0.000 - - - - - - - - -W-Pw2-P 0.489 0.396 0.348 0.367 0.000 - - - - - - - -W-Pw2-NP 0.644 0.337 0.483 0.261 0.421 0.000 - - - - - - -W-S20-P 0.508 0.422 0.373 0.382 0.214 0.439 0.000 - - - - - -W-S20-NP 0.459 0.448 0.625 0.553 0.538 0.608 0.584 0.000 - - - - -W-Sm3-P 0.359 0.524 0.596 0.604 0.497 0.675 0.549 0.305 0.000 - - - -W-Sm3-MP 0.442 0.409 0.544 0.487 0.460 0.529 0.511 0.273 0.351 0.000 - - -W-Ss2-P 0.562 0.455 0.379 0.411 0.229 0.427 0.246 0.624 0.590 0.544 0.000 - -W-Ss2-NP 0.602 0.324 0.470 0.247 0.376 0.272 0.401 0.573 0.617 0.485 0.393 0.000 -W-Sw5-P 0.338 0.497 0.559 0.537 0.439 0.616 0.480 0.462 0.359 0.466 0.531 0.558 0.000 W-Sw5-NP 0.603 0.303 0.507 0.312 0.421 0.333 0.447 0.496 0.577 0.436 0.462 0.341 0.541 0.000 S:Smithers soil; W:Williams Lake soil; C:Uninoculated control; L6:L6-16R; Pw2:Pw2R; S20:S20R; Sm3:Sm3RN; Ss2:Ss2RN; Sw5:Sw5RN; P:Seedling present; NP: Soil only. Distances 4.0 0.0 1 1 1 1 2 1 2 443 1 1114 4121 1 429322 14364621 14137576311 345789631 1 1 1149996111 11 523798865 4 1 11 2261964613311 1 1 121422221113143121 21 21 1 0.0 0.2 0.4 0.6 0.8 Dissimilarities Fig. A9.1 Shepard diagram of distances versus input dissimilarities obtained from the carbon substrate utilization patterns of 28 soil microbial communities. Each number represents a pair of the 378 combinations listed in Table A9.1. 324 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088248/manifest

Comment

Related Items