UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ecological specificity of growth promoting bacteria for interior spruce (picea glauca x picea engelmanii)? O’Neill, Gregory Arthur 1991

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

Item Metadata


831-ubc_1992_spring_oneill_gregory_arthur.pdf [ 4.19MB ]
JSON: 831-1.0075215.json
JSON-LD: 831-1.0075215-ld.json
RDF/XML (Pretty): 831-1.0075215-rdf.xml
RDF/JSON: 831-1.0075215-rdf.json
Turtle: 831-1.0075215-turtle.txt
N-Triples: 831-1.0075215-rdf-ntriples.txt
Original Record: 831-1.0075215-source.json
Full Text

Full Text

ECOLOGICAL SPECIFICITY OF GROWTH PROMOTINGBACTERIA FOR INTERIOR SPRUCE(PICEA GLAUCA X PICEA ENGELMANII)?BYGREGORY ARTHUR O'NEILLB.Sc. UNIVERSITY OF BRITISH COLUMBIA 1986A THESIS SUBMITTED IN PARTIAL FULFILLMENTOF THE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEINTHE FACULTY OF GRADUATE STUDIESDEPARTMENT OF FOREST SCIENCESWe accept this thesis as conforming to the required standardTHE UNIVERSITY OF BRITISH COLUMBIANOVEMBER 1991© GREGORY ARTHUR O'NEILLIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.Department of Forest ScienceThe University of British ColumbiaVancouver, CanadaDateDE-6 (2/88)iiABSTRACTSoil, rhizobacteria and interior spruce seed originating from two disparate ecosystems were used toexamine the effect of rhizobacterial inoculation and the role of coexistence between rhizobacteria,seed provenances and soil sources on germination and spruce seedling growth in two experiments.Statistically significant enhancement of germination due to inoculation with bacteria was rare.Germination of seed inoculated with coexistent bacteria was significantly lower than germination ofseed inoculated with non-coexistent bacteria.Inoculation of seed with bacteria resulted in significant enhancement of seedling growth in bothexperiments. Maximum shoot and root dry weight increases of 53% and 67%, respectively, wereobserved. The effect of inoculation on seedling growth varied greatly with seed provenance and soilsource.Coexistent bacteria (i.e. originating from the same location as the target seed or soil) were not moreeffective growth promoters than non-coexistent bacteria. However, uninoculated seedlings grown incoexistent soil had 27% and 35% heavier shoot and root dry weights, respectively, than uninoculatedseedlings grown in non-coexistent soil. The shoot and root biomass stimulation decreased to 17%and 23%, respectively, when coexistent pasteurized soil was used, suggesting that both biotic andabiotic soil factors may have contributed to seed-soil coexistence specificity.Novel findings in these experiments include the detection of: significant bacterial plant growthpromotion of interior spruce; plant growth promotion by a Staphlococcus species; and adaptiverelationships between seed and soil factors.i i iTABLE OF CONTENTSpageAbstractTable of Contents 	 iiiList of TablesList of Figures 	 viAcknowledgements 	 vii1.02.0INTRODUCTION 	LITERATURE REVIEW 	1.32.1 The Rhizosphere 	 32.2 Plant Growth Promoting Rhizobacteria 	 42.3 PGPR in Agriculture	42.4 PGPR in Arboreal Species 	 62.5 Mechanisms of Action of PGPR 	 62.5.1	 Production of Phytohormones 	 72.5.2	 Inhibition of Deleterious Rhizobacteria 	 82.5.3	 Increased Nutrient Availability 	 .92.5.4	 Nitrogen Fixation 	 92.6 Bacteria-Host Specificity 	 113.0 MATERIALS AND METHODS 	 153.1 Soil and Seedling Collection 	 153.2 Bacterial Isolation 	 163.3 Bacterial Storage 	 .173.4 Acetylene Reduction Assay 	 173.5 Strain Selection and Inocula Preparation 	 173.6 Experiment 1 	 183.6.1	 Seed Sowing and Inoculation 	 183.6.2	 Experimental Design and Seedling Culture 	 183.6.3	 Germination Survey, Harvests andStatistical Analysis 	 193.7 Experiment 2 	 .204.0 RESULTS 234.1 Soil Analysis and Bacterial Isolation 	 234.2 Experiment 1 	 234.2.1	 Germination Responses	234.2.2	 Growth Responses	294.3 Experiment 2 	 354.3.1	 Germination Responses 	 354.3.1.1 Coexistence Specificities 	 354.3.2	 Growth Responses 	 394.3.2.1 Coexistence Specificities 	 45ivpage5.0 DISCUSSION 	 48	5.1	 Germination 	 485.1.1 Coexistence Specificities 	 51	5.2	 Seedling Growth 	 525.2.1 Effects of Bacterial Inoculation 	 525.2.1.1 Host Specificity	  .535.2.1.2 Coexistence Specificities 	 545.2.1.3 Age of the Host Plant	  585.2.1.4 Endorhizosphere Bacteria	 595.2.1.5 Abundance of the Inoculanton the Host Plant 	 625.2.2 Effect of Seed Provenance, Soil Sourceand Soil Pasteurization 	 625.2.3 Mechanisms of Action	 645.2.4 Reproducibility of Effects	 656.0 CONCLUSION 	 677.0 SUMMARY 	 688.0 LITERATURE CITED	 69Appendix 1 	 79Appendix 2 	 80Appendix 3 	 81Appendix 4 	 82Appendix 5 	 83Appendix 6 	 84Appendix 7 	 85Appendix 8 	 86Appendix 9 	 87vLIST OF TABLESpageTable 1.	 Characteristics of pasteurized and unpasteurized Mackenzie andSalmon Arm soil.	 24Table 2a.	 List of Mackenzie bacterial strains used in Experiment 1. 	 25Table 2b.	 List of Salmon Arm bacterial strains used in Experiment 1.	 26Table 3.	 Effect of bacterial inoculation with Mackenzie and Salmon Armisolates on germination capacity and germination value ofMackenzie and Salmon Arm seedlots in Experiment 1.	 28Table 4a.	Experiment 1 treatment means for Mackenzie seedlings. 	 33Table 4b.	 Experiment 1 treatment means for Salmon Arm seedlings.	 34Table 5.	 Mean germination capacity and germination value of 56 treatments ofExperiment 2.	 37Table 6.	 Coexistent and non-coexistent factor combination treatment means forgermination capacity and germination value in Experiment 2. 	 38Table 7.	 Growth parameter means of 56 treatments in Experiment 2. 	 41Table 8	 Coexistent and non-coexistent factor combination treatment meansfor growth parameters in Experiment 2.	 46viLIST OF FIGURESpageFig. 1. Effect of Mackenzie and Salmon Arm bacterial isolates on the germinationcapacity and germination value of their respective seedlots.	 27Fig. 2. Effect of inoculation with Mackenzie and Salmon Arm bacterial strains inshoot dry weight of their respective seedlots. 	 30Fig. 3. Effect of inoculation with Mackenzie strains on growth parameters. 	 31Fig. 4. Effect of inoculation with Salmon Arm strains on growth parameters.	 32Fig. 5. Effect of seed provenance, soil source and pasteurization on germinationcapacity	 36Fig. 6. Effect of inoculation with putative PGPR on seedling shoot dry weight inall seed, soil and pasteurization treatment combinations in Experiment 2. 	 40Fig. 7. Shoot dry weight of uninoculated control seedlings in all seed, soiland pasteurization treatments. 	 42Fig. 8. Shoot dry weight promotion due to bacterial inoculation in coexistentbacteria-seed-soil treatments in Experiments 1 and 2. 	 44Fig. 9. Shoot dry weight of treatments comprised of coexistent and non-coexistentfactor combinations and control treatments in Experiment 2.	 47vi iACKNOWLEDGEMENTSI express my gratitude to Dr. C. Chanway for his many hours of enthusiastic guidance, support andeditorial assistance in this work. Laboratory facilities were generously supplied by Dr. B. Sutton ofBritish Columbia Research Corporation. I am indebted to Dr. P. Axelrood and R. Radley whoprovided valuable laboratory direction.Drs. B. Holl, V. LeMay, D. Lester and G. Thdrien, extended thoughtful insights, and Dr. M. Carlson,G. Kiss, A. Percrastins and J. Cruikshank of the B.C. Ministry of Forests supplied the seed and theuse of the collection sites. Technical assistance was provided by K. Clark, S. Grimes, E. Klein, S.McInnis, J. McPhee, B. Thompson, P. Warne and J. Zwierink.Financial assistance was provided by the B.C. Science Council in the form of a G.R.E.A.T. grant.11.0 INTRODUCTIONPoor seedling growth after out-planting on highly stressful or competitive sites is one of the majorcauses of failed or poorly growing conifer plantations in British Columbia. The use of seedlingswhich are in optimal physiological condition has been seen as a solution to this problem (Grossnickleet al. 1991). Bacterial inoculants may influence a seedling's physiological condition (Sang et al.1988) and have been shown to enhance shoot and root growth of many agricultural and of somearboreal species in laboratory, greenhouse and field trials. If similar effects were obtained withconifers, bacterial inoculation could predispose planted seedlings to be better competitors for light,nutrients or moisture, and thereby possibly improve their survival and growth.Bacterial inoculation of seeds has also been shown to influence germination rate and capacity (thepercent of seeds which germinate) of some plant species. If such technology was developed forconifer nurseries, it could obviate or minimize expensive over-sowing and subsequent thinningoperations, and help to produce a uniformly-sized crop. In addition, bacteria capable of enhancingthe rate of germination could help nursery-growers using multiple seedlots to stagger or synchronizematuration as desired, and perhaps permit the use of a wider range of seedlots in regions which havea short growing season.While several factors may influence plant reactions to rhizosphere bacteria (rhizobacteria), thegenotypic match between plants and bacteria has been shown to influence the nature of thisresponse (Chanway et al. 1988a,b, 1991b). Genotypic specificity may develop over time betweencoexistent plants and naturally-occurring rhizosphere bacteria through adaptation of bacterialpopulations to plant hosts (Holl 1983; Chanway et al. 1991b).Therefore, the ability of rhizobacteria isolated from naturally-regenerating interior spruce [Piceaglauca (Moench) Voss. x Picea engelmanii Parry ex Engelm.] seedlings to enhance germination andgrowth of interior spruce was tested under greenhouse and growth chamber conditions as a2preliminary evaluation of their potential for nursery and field applications. In particular, the role ofcoexistence in specificity between conifer provenances, soil and growth promoting bacteria in thegermination and growth of interior spruce seedlings was examined.Specifically, the objectives of the experiments were to determine if(1) interior spruce seed inoculated with rhizobacteria germinate faster or more completely, andproduce larger seedlings than uninoculated seed;(2) treatments of coexistent factor combinations (bacteria-seed, bacteria-soil, seed-soil, bacteria-seed-soil) result in more rapid or complete germination, or larger seedlings, than treatments ofnon-coexistent combinations of the same factors.32.0 LITERATURE REVIEW	2.1	 The RhizosphereLorenz Hiltner (1904) was the first to recognize the potential importance of the intense microbialactivity on and around root systems to plant growth. He called this area the rhizosphere. Bacteriacomprise the most common class of rhizosphere micro-organism (Rovira and Davey 1974) and canattain populations of up to 3 x 10 9 cells per gram of rhizosphere soil (Rouatt and Katznelson 1961).Bacteria isolated from this area have been termed ectorhizosphere bacteria; those isolated fromwithin surface-sterilized roots have been termed endorhizosphere bacteria (Lalande et al. 1989).(The terms ectorhizobacteria and endorhizobacteria will be used in this paper to describe bacteriaisolated from these two root regions, and the general term rhizobacteria to describe all rhizospherebacteria.)Rhizosphere bacteria can exhibit considerable influence on plant nutrient availability through theactivities of various intra- and extra-cellular enzymes (e.g. lipase, phosphatase and nitrogenase).They may also suppress plant pathogens through competition or antibiotic production, as well asproducing active phytohormones. Since all soil-borne nutrients received by the plant must passthrough the rhizosphere, it is not surprising that rhizobacteria may affect plant growth.The organic compounds contained in senescent root tissue and in root exudates and secretionsprovide substrate for the growth of heterotrophic soil microbes in the rhizosphere. Radio-isotopelabelling experiments have indicated that up to 40% of cereal (Whipps and Lynch 1986) and 50% ofconifer (Reid and Mexal 1977; Perry et al. 1987) net primary production can be exuded into therhizosphere. The allocation of such an astonishingly large quantity of photosynthates below-groundfurther reflects the potential importance of rhizosphere microbial ecology to plant growth.	2.2	Plant Growth Promoting RhizobacteriaPlant growth promoting micro-organisms have been studied intensively because of their potential4impact on agricultural and forest productivity (Gaskins et al. 1985; Schroth and Weinhold 1986;Chanway et al. 1991a). The term 'plant growth promoting rhizobacteria' (PGPR) has been used todescribe soil bacteria which, when applied to seed, tubers or roots, are able to colonize roots andstimulate plant growth (Kloepper and Schroth 1978). Many strains that belong to commonlyoccurring genera of soil bacteria, such as Arthrobacter, Azospirillum, Azotobacter, Bacillus,Pseudomonas and Serratia have been found in association with plant roots and to promote plantgrowth (Brown 1974; Gaskins et al. 1985).Recent success in growth enhancement of agricultural (Kloepper and Schroth 1981; Kapulnik andOkon 1983; Elad et al. 1987; Chanway et al. 1988a,b; Reddy and Rahe 1989) and tree species(Akhromeiko and Shestakova 1958; Gardner et al. 1984; Strobel and Nachmias 1985; Pandey et al.1986; Caesar and Burr 1987; Chanway et al. 1991b; Chanway and Holl 1991) through inoculationwith PGPR has stimulated a renewed interest in rhizosphere biology. However, variability of theplant growth response remains a major impediment to the implementation of PGPR technology inagriculture and forestry (Kloepper et al. 1989). When rhizosphere synecology and the mechanismsby which PGPR stimulate plant growth are better understood, the likelihood of being able to selectand manage more effective PGPR strains will be greatly increased.2.3 PGPR in AgricultureThe first attempts to improve plant growth by 'bacterization' (coating of seeds with bacteria beforeplanting) were made in Russia in the 1940's with strains of Azotobacter and Bacillus (Allison 1947;Mishustin and Naumova 1962) which were capable of in vitro nitrogen fixation and phosphatesolubilization, respectively. It was claimed that inoculation with these strains could result in yieldincreases in the order of 10% in fifty to seventy percent of the crops to which they were applied(Cooper 1959). Unfortunately, lack of statistical analysis and poor reproducibility have precludeduseful interpretation of these studies (Mishustin 1970).Experimentation with PGPR in the western world was first reported in 1963, when inoculation with5asymbiotic N-fixing Bacillus and Clostridium rhizobacteria was shown to stimulate growth of tomato(Lycopersicon esculentum L.), maize (Zea mays L.) and wheat (Triticum sp.) (Rovira 1963).Subsequently, representatives of these and other genera of bacteria have been used experimentallyas PGPR for agricultural crops. Howie and Echandi (1983) and Kloepper and Schroth (1981)reported significant (all uses of the word 'significant' in this thesis imply statistically significant)increases in the weight of potato (Solanum tuberosum L.) inoculated with antibiotic producingstrains of Pseudomonas. Growth, emergence and vigor of canola (Brassica campestris L.) were alsoenhanced by strains of Pseudomonas, Arthrobacter and Serratia in greenhouse and field trials(Kloepper et al. 1988). Growth stimulation of perennial ryegrass (Lolium perenne L.), crestedwheatgrass (Agropyron cristatum L.), white clover (Trifolium repens L.) (Chanway et al. 1988a; Hollet al. 1988) and spring wheat (Triticum aestivum L.) (Chanway et al. 1988b) were achieved usingBacillus inocula. In some cases, seedling emergence was also stimulated (Holl et al. 1988; Chanwayand Nelson 1990). Other strains have promoted growth of radish (Raphanus sativus L.) (Kloepperand Schroth 1978), tomato, pepper (Capsicum annuum L.), melon (Cucumis melon L.), bean(Phaseolus vulgaris L.), tobacco (Nicotiana tabaccum L.), cucumber (Cucumis sativus L.) (Elad et al.1987) and ornamental plants (Yeun and Schroth 1986).The nature and magnitude of the growth response vary considerably. Examples of plant growthstimulation relative to uninoculated controls include: grain yield (11%) (Kapulnik et al. 1983); height(38%) (Reddy and Rahe 1989); shoot dry weight (56%) (Elad et al. 1987); number of roots (42%) (Tienet al. 1979); root dry weight (40%) (Hussain and Vancura 1970); root surface area (18%) (Kapulnikand Okon 1983) and root length (29%) (Pandey et al. 1986). Variation in the growth response alsooccurs between experiments, inoculants and target plants, and not all growth responses are positive.For example, the fluorescent Pseudomonas strain 599NR inhibited shoot and root growth of sweetorange (Citrus sinenis Osbeck) 39% and 41%, respectively, compared with uninoculated controls, butpromoted shoot and root growth of lemon (Citrus jambhiri Lush.) 38% and 21%, respectively(Gardner et al. 1984).62.4 PGPR in Arboreal SpeciesPlant growth promotion of several arboreal species following PGPR inoculation has also beenobserved. Enhanced dry weight of oak (Quercus sp.) and ash (Fraxinus sp.) seedlings (13 and 26%,respectively) after inoculation with Azotobacter chroococcum was the first report of an arborealPGPR (Akhromeiko and Shestakova 1958). Stimulation of almond (Prunus sp.) root stock (Strobeland Nachmias 1985), apple (Malus sp.) seedlings and rootstock (Caesar and Burr 1987), rough lemon(Citrus jambhiri Lush.) (Gardner et al. 1984) and eucalyptus (Eucalyptus camaldulensis Dehn.)(Mohammad and Prasad 1988) growth by bacterial inoculants has also been reported.Growth of coniferous species may also be stimulated by PGPR. Shoot length of Scots Pinegerminants (Pinus sylvestris L.) was increased after inoculation with Coryneform bacteria ortreatment with the supernatant of the Coryneform growth media (Pokojska-Burdziej 1982). Shootand root dry weight, height, root surface area and root collar diameter of lodgepole pine (Pinuscontorta Dougl.), and shoot biomass, root collar diameter and root surface area of Douglas-fir[Pseudotsuga menziesii (Mirb.) Franco] seedlings were increased by bacterial inoculation (Chanwayand Holl 1991a,b; Chanway et al. 1991a). Container-grown Douglas-fir inoculated with a mixedsuspension of forest floor bacteria showed significantly greater stem height and diameter thanuninoculated controls when grown under nutrient limited conditions (Parker and Dangerfield 1975).2.5 Mechanisms of Action of PGPRThe mechanism(s) by which PGPR stimulate plant growth have yet to be conclusively determined.However, four mechanisms have been frequently postulated. These include: (1) production ofphytohormones; (2) inhibition of deleterious rhizobacteria (DRB) and plant pathogens; (3) increasednutrient availability, and (4) nitrogen fixation. Most PGPR researchers recognize that a singlestrain may possess more than one plant growth promoting attribute (Curl and Truelove 1986; Holl etal. 1988) and that these may interact with other biotic (Bowen and Theodorou 1979; Garbaye andBowen 1987; McAfee and Fortin 1988) or abiotic (McArthur et al. 1988) factors in the soil. As aconsequence of the numerous factors influencing the biology of the rhizosphere, Schroth and7Weinhold (1986) termed investigations in this area 'a masochist's delight'.2.5.1 Production of PhytohormonesPhytohormones that are essential for plant morphogenesis, such as auxins, giberellins andcytokinins, are produced by several genera of rhizosphere bacteria (Brown and Burlingham 1968;Eklund 1970; Hussain and Vancura 1970; Brown 1972; Lynch 1976; and Tien et al. 1979). Plantgrowth promotion by Azotobacter paspali (Barea and Brown 1974; Brown 1976), Azospirillumbrasilense (Tien et al. 1979), Bacillus megaterium (Katznelson and Cole 1965), B. polymyxa (Holl etal. 1988), B. subtilus (Brown et al. 1968) and Pseudomonas species (Katznelson and Cole 1965;Eklund 1970; Hussain and Vancura 1970) has been attributed to their ability to producephytohormones. Venkateswarlu and Rao (1983) correlated root growth stimulation and an increasein the number of lateral roots and root hairs after bacterial inoculation with the magnitude of invitro auxin production by several strains of Azospirillum brasilense.However, most evidence for the involvement of phytohormones in PGPR activity is indirect, and hasbeen derived from experiments in which the effects of bacterial inoculation are mimicked byexogenous application of phytohormones. For example, similar growth effects were observed afterinoculation of tomato with either gibberellic acid-producing Azotobacter or with synthetic gibberellicacid (Brown et al. 1968), and by inoculating wheat with various phytohormone-producingrhizobacteria or with synthetic phytohormones (Brown 1972). In addition, growth of small plantscan be increased by adding plant growth substances and live or heat-killed bacteria to the soil(Jackson et al. 1964; Gaskins and Hubbell 1979).Though indirect, these observations suggest that bacterial production of plant growth substancesmay contribute to PGPR activity. Despite the minute concentration (i.e. nanomolar or picomolar) ofgrowth regulators produced by rhizobacteria, they are absorbed in the region of root-hairdevelopment (Riviere 1963; Libbert and Silhengst 1970; Brown 1972). Their production in therhizosphere in synchrony with the development of new tissue may explain their effectiveness in8altering plant growth (Gaskins et al. 1985). Production of phytohormones as a mechanism of plantgrowth promotion by rhizobacteria continues to receive considerable attention; however, moreconclusive evidence in support of this mechanism may await development of techniques by whichbacterial phytohormone production in situ can be effectively monitered.2.5.2 Inhibition of Deleterious RhizobacteriaPlant growth promotion by PGPR may also occur through the inhibition of minor plant pathogenstermed 'deleterious rhizobacteria' (DRB). DRB colonize roots and reduce plant growth withoutcausing symptoms of disease (Salt 1979; Suslow and Schroth 1982). For example, Rovira (1972)found that root hair number and length, both considered to be extremely important for plant growthin phosphorus-limited soils, were reduced in the presence of many strains of rhizobacteria.Similarly, Bowen and Rovira (1961) found root-growth inhibiting microorganisms in soils collectedfrom a stand of Monterey pine (Pinus radiates) and from three agricultural crops.Postulated mechanisms by which PGPR inhibit proliferation of DRB include niche exclusion throughcompetition for root binding sites (Burr and Caesar 1984), production of compounds that are toxic toDRB, such as antibiotics or hydrogen cyanide (Weller 1988), and production of siderophores, whichchelate soil Fe3+, thereby limiting its availability to, and subsequent growth of, DRB (Kloepper et al.1980; Curl and Truelove 1986).In some cases, supporting evidence for the DRB-inhibition hypothesis is fairly convincing. Kloepperand Schroth (1981) found that inoculation of potato seed pieces with antibiotic-producing strains ofPseudomonas caused significant growth increases (300 - 500%) in total plant weight of potato, butinoculation with non-antibiotic producing strains had no effect. Suslow and Schroth (1982) showedthat co-inoculation of sugar beet seed with strains of PGPR and DRB resulted in inhibition of rootcolonization by DRB and increased plant growth compared to inoculation with DRB alone. Growthpromotion did not occur when experiments were conducted in sterile non-soil media or whenautoclaved field soils were used, presumably because DRB were not present (Suslow 1982).9However, several studies have shown that plant growth promotion with members of the genusBacillus and other Pseudomonads can occur under both sterile and non-sterile conditions (Lifshitz etal. 1987; Holl et al. 1988; Chanway and Nelson 1990; Chanway et al. 1989; Chanway and Holl1991a,b). Therefore, PGPR activity does not appear to be strictly related to DRB inhibition.2.5.3 Increased Nutrient AvailabilityAside from enhanced nutrient uptake resulting from the larger root systems associated withinoculated seedlings, most nutrient-related mechanistic hypotheses have focused on phosphorusavailability. Phosphorus may become more available to plants through the production of organicacids by rhizobacteria, which reduce the local pH and thereby solubilize otherwise insolublephosphorus compounds and other soil minerals (Mishustin and Naumova 1962; Bajpai and SundaraRao 1971). Alternatively, bacterial production of phosphatase can solubilize organic sources ofphosphorus directly. Reviews of the effects of bacteria on the mineral uptake by plants (Katznelson1965; Barber 1978) reveal an accumulation of conflicting evidence (Gaskins et al. 1985). However,after reviewing all available evidence, Tinker (1984) concluded that bacterial solubilization ofphosphate is probably of minor importance in the growth response of plants inoculated with PGPR,and his contention has not been challenged.2.5.4 Nitrogen FixationMembers of various genera of rhizosphere bacteria, including Azotobacter, Azospirillum, Bacillus,Beijerinckia, Clostridium, Desulfovibrio, Klebsiella and Pseudomonas are capable of fixingatmospheric nitrogen. While symbiotic nitrogen-fixation accounts for most of the combined nitrogeninput into forested ecosystems (Kimmins 1987), asymbiotic N-fixation by free-living bacteria maycontribute significantly to the long-term productivity of agricultural (Gaskins et al. 1985) and forestecosystems (Dawson 1983; Marschner 1986, p. 189; Kimmins 1987).Early estimates of asymbiotic nitrogen input of 199 kg N/ha/y for bushlands (Jaiyebo and Moore1963), 165 kg N/ha/y for lowland forest (Greenland and Nye 1959) and up to 313 kg N/ha/y for some1 0agricultural crops (Evans and Barber 1977) were erroneously high (Davey and Wollum 1984). Theirover-estimation has been attributed to several factors, including faulty assay techniques (VanBerkum and Bohlool 1980), inaccurate extrapolation of fixation rates obtained from short-termassays (Brown 1982), inappropriate sample collection, and a lack of consideration of diurnal andseasonal fluctuations in nitrogen-fixation rates (Sims and Dunigan 1984).Current estimates of asymbiotic nitrogen input are usually below 30 kg N/ha/y, and most are below 5kg Nlhaly (Evans and Barber 1977; Davey and Wollum 1984). However, relatively few accurateestimates are available. According to Sprent (1979, p. 114), the paucity of information on theecological importance of asymbiotic nitrogen fixation "reflects lack of information, rather than lack ofimportance".Relatively low rates of nitrogen fixation notwithstanding, inoculation of seeds and plants withnitrogen-fixing bacteria has resulted in significant yield increases of several agricultural species(Smith et al. 1976; Rennie and Larson 1979; Kapulnik et al. 1981; Schank et al. 1981; Chanway et al.1988a,b), and in root growth increases of oak (Quercus serrata) (Pandey et al. 1986) and lodgepolepine (Chanway et al. 1991a) seedlings. However, results of inoculation experiments using nitrogen-fixing PGPR with 15N dilution techniques or with nitrogen rich media (which should suppressnitrogenase activity) suggest that nitrogen fixation is of secondary (Okon et al. 1983; Kapulnik et al.1985; Chanway and Holl 1991a) or no importance (O'Hara et al. 1981; Brown 1982) in the plantgrowth response, and increases in plant growth by diazotrophic PGPR are often attributed tobacterial production of phytohormones (Barea and Brown 1974; Holl et al. 1988). In retrospect, thisconclusion is not surprising if Barber and Lynch (1977) were correct in asserting that "if all thecarbon released by the roots were available only to known nitrogen-fixing [rhizobacteria], and if allthe nitrogenases of the bacteria functioned at their maximum rates, then only 15% of the N contentof temperate cereals could be provided in this way". Therefore, asymbiotic root-associated nitrogenfixation has all but been dismissed as a primary mechanism by which diazotrophic PGPR operate.1 12.6	 Bacteria-Host SpecificityQualitative and quantitative differences in root exudates exist between plant species (Rovira andDavey 1974; Curl and Truelove 1986) and between cultivars and genotypes of the same species(Baldani and Dobereiner 1980). These differences, coupled with the reliance of rhizosphere microbeson root exudates for organic nutrients, may result in the proliferation of microbial populations thatare specific to plant species or to genotypes within species (Neal et al. 1973; Burr and Caesar 1984;Chanway et al. 1991b).Chanway et al. (1991b) proposed that specificity between plants and growth promoting micro-organisms can occur at either of two stages of these associations: during infection of the root systemto form root nodules or mycorrhizas (i.e. infection specificity), or during subsequent growth of theinfected plant host (i.e. growth response specificity). Where the relationship is not symbiotic, butmicrobial association with the host is required (e.g. PGPR), specific colonization of the rhizospheremay occur.Infection (or colonization) specificity may be determined by a cell wall recognition mechanism, inwhich plant lectins (specific plant glycoproteins which adhere to unique carbohydrates on the cellwall of bacteria) operate in a manner similar to that of antigens in immunological reactions (Sumner1990; Chanway et al. 1991b). The possible involvement of lectins in the specificity observed inRhizobium -legume associations has been recognized for nearly two decades (Bohlool and Schmidt1974) and was recently postulated in the adsorption-recognition process between plants and PGPRby Okon and Kapulnik (1986). More recently, infection specificity in Rhizobium -legume associationswas also shown to involve biochemical signals secreted by plant roots which activate nodulationgenes in specific Rhizobium strains (Long 1989). Growth specificity may result from bacterialproduction of compounds of the type, or in an amount that specifically affects growth of individual orrelated groups (i.e. ecotypes) of plants.12Chanway et al. (1991b) also suggest that the development of plant-specific rhizosphere microfloramay arise in either of two ways. Pre-existing genetic variability among resident soil bacteria maypredispose particular bacterial strains to experience a competitive advantage over other strains inthe rhizosphere of a particular plant genotype or species. Consequently, the populations of thosestrains would increase and possibly dominate in the rhizosphere due to superior fitness when inassociation with that plant.An alternative and perhaps less likely mechanism would involve the adaptation of particularbacterial strains to the host plant. This could occur as a result of the generation of genetic variationin the rhizosphere bacteria population through point mutations and/or various forms of geneticrecombination (conjugation, transduction or transformation), with subsequent selection of superiorbacterial genotypes in the rhizosphere. Genotypic specificity between plants and microbes maytherefore result from small genetic differences between host plants that affect root exudation andconsequently the size and nature of the bacterial population that proliferates in the rhizosphere.Some strains of PGPR are capable of promoting the growth of a number of plant species (Elad et al.1987; Holl et al. 1988; Bashan et al. 1989). However, PGPR are not universally effective anddifferences in growth promotion between PGPR-plant combinations are well documented (Rovira1963; Gardner et al. 1984,1985; Chanway et al. 1989). The basis of these differences is notunderstood, but host-plant genetics (Burr and Caesar 1984; Chanway et al. 1989) and the history ofcoexistence between bacterial and plant genotypes (Chanway et al. 1988a,b) are important.The occurrence of specific relationships between strains of associative nitrogen-fixing bacteria andplant genotypes is well known (Baldani and Dobereiner 1980; Holl 1983). Chanway et al. (1988a)tested the hypothesis that genotype specific plant growth promotion by PGPR may develop betweencoexistent plant genotypes and associative PGPR (or rhizobacterial populations). Using physicallycontacting (i.e. coexistent) pairs of white clover and perennial ryegrass plants, and strains of Bacillusisolated from the roots of the white clover, they tested the growth-promoting ability of the bacteria13using clones of the coexistent and non-coexistent 'parental plants'. They discovered that as theexperimental environment became more 'familiar' by growing the clover predominantly with (1) non-coexistent Bacillus and ryegrass, then with (2) coexistent Bacillus but not ryegrass, and finally with(3) coexistent Bacillus and ryegrass, the yield of the legume component of the species mixtureincreased from condition (1) to condition (3). Furthermore, no inoculation reponse was detectedwhen plants were inoculated with non-coexistent Bacillus strains.Perhaps the most pointed display of a specific PGPR-plant relationship emerged from theexperiments by Rennie and Larson (1979) involving the inoculation of disomic chromosomesubstitution lines of wheat with a diazotrophic Bacillus isolated from a parental wheat cultivar.(Disomic chromosome substitution lines contain 20 pairs of indigenous chromosomes plus one pairfrom a donor line, allowing for the study of the effects of the 'donated' pair in an otherwise constantgenetic background.) Using this system they were able to attribute significant plant growthincreases and nitrogen accumulation following bacterial inoculation to the presence of a singlechromosome in the wheat genome. Their results emphasize how relatively small changes in plantgenotype can have substantial effects on growth promotion by rhizosphere bacteria.These findings prompted Chanway et al. (1988b) to argue that if beneficial microbes are positivelyselected over time in the rhizosphere, then the probability of finding a positive effect on plantperformance due to inoculation with coexistent bacterial strains should be greater than if plants areinoculated with strains to which they have not been previously exposed. To test this hypothesis,Chanway et al. (1988b) isolated Bacillus strains from the rhizosphere of spring wheat cultivar'Katepwa' which was growing in a field that had been cropped continuously with this cultivar for thepreceeding five years and to other wheat cultivars for the preceeding 22 years. When inoculatedonto cv. 'Katepwa', a related cultivar to which the field had also been cropped, and an unrelatedMexican cultivar, six of seven Bacillus isolates promoted growth of cv. 'Katepwa', but none promotedgrowth of the other two cultivars. These results indicated that cultivar-specific adaptation ofrhizosphere bacteria (or the bacterial population) to wheat can occur within a period of five years.1 4Chanway et al. (1989) suggested that some of the variability observed in plant growth promotion byPGPR could be explained by genotypic specificity between plants and inoculant microbes. Theyproposed, therefore, that the probability of securing consistent and effective PGPR could beincreased by using strains which were isolated from (i.e. had coexisted with) the target crop.Though beyond the scope of this review, it is interesting to note that varying degrees of plant-microbe specificity have also been documented involving plants and root nodule bacteria (Mytton etal. 1977; Holl 1983; Florence and Cook 1984) and mycorrhizae (Molina and Trappe 1982; Cline andReid 1982; Kendrick and Berch 1984), and more than three decades ago, Moser (1958) recommendedusing the same provenance of trees and fungi to stimulate optimal mycorrhizal formation and treegrowth.It was with these ideas in mind that I undertook a search for interior spruce PGPR. The term'coexistent' has been used to indicate a common origin of organisms (rhizosphere bacteria, interiorspruce seedlings and seed) and soil (i.e. collected from the same micro-site as were the seed,seedlings and bacteria). The term 'coexistence specificity' will refer to the seedling growth responsethat results from testing coexistent organisms and soil compared with non-coexistent (i.e. collectedfrom an alternative site) organisms and soil.The identification of bacteria-plant or bacteria-soil coexistence specificity in bacterial promotion ofgermination or growth (i.e. increased germination or growth when the bacterial inoculant and seed,or bacterial inoculant and soil, have the same geographic origin), would make the isolation of ageneral PGPR unlikely, and would potentially limit the effective range of seedlots or soils that couldbe used with a particular PGPR. Nevertheless, consideration of coexistence specificity, if important,could facilitate the isolation of consistent germination or growth promoters, albeit with a restrictedrange of seedlots or sites.153.0 MATERIALS AND METHODS3.1	 Soil and Seedling CollectionNaturally-established interior spruce (Picea glauca x englemanii) seedlings (1-5 years old) and soilwere collected during the summer of 1989 from the understory of two mature forest sites in centralBritish Columbia. The first of these ecologically disparate sites was located 10 km south ofMackenzie in the Sub-Boreal Spruce biogeoclimatic zone (Krajina et al. 1982), within 1 km of aBritish Columbia Ministry of Forest's (MoF) interior spruce parent-tree stand at Buth Creek(latitude 55° 11', longitude 122° 58', elevation 780 m). The main vegetation in this valley-bottomsite, consisted of interior spruce, black cottonwood (Populus trichocarpa Torr. and Gray ex Hook),coltsfoot (Petasites palmatus [Ait.] Gray) and Pleurozium schreberi ((Brit.) Mitt.).The second site was located 30 km north-west of Salmon Arm in the Engelmann Spruce Sub-alpineFir biogeoclimatic zone, within 50 m of the MoF's interior spruce 'plus tree' #3010 (latitude 51° 04',longitude 119° 26', elevation 1250 m). The main vegetation on this mountain-top site includedinterior spruce, sub-alpine fir [Abies lasiocarpa (Hook.) Nutt.], Canada thistle (Cirsium arvensis),red raspberry (Rubus ideus L.), huckleberry (Vaccinium membranaceum Dougl. ex Hook.) andwestern mountain-ash (Sorb us scopulina Greene).Interior spruce seedlings and their intact root mass contained in forest soil were collected to a depthof 20 cm. Seedlings and soil were placed separately in plastic bags and transferred to the ForestBiotechnology Centre laboratory at the British Columbia Research Corporation in Vancouver wherethey were stored at 4° C. All bacterial isolations took place within seven days of seedling collection.Soil samples within each location were pooled and soil nutrient analysis was conducted usingstandard methodology according to Black (1965).1 63.2	 Bacterial IsolationA total of 25 seedlings from each location were used for bacterial isolations. Seedlings were dividedinto five groups of five in order to maximize the diversity of bacterial strains isolated from the roots.Root masses of each group of seedlings were shaken vigorously to dislodge loosely adherent soil, andwere then cut aseptically into 3-5 cm segments. To isolate the ectorhizobacteria, approximately 2.0 gof root segments (0.4 g/plant) from each group of five seedlings were placed in a 250 mL flaskcontaining 20-30 glass beads and 150 mL of 10 mM sterile phosphate buffer (SPB - 1.21 g K2HPO4,0.34 g KH2PO4, 1.0 L distilled water pH 7.0). Flasks were agitated gently on a rotary shaker (100rev/min) for 20 minutes.To obtain the endorhizobacteria, roots treated as described above were removed from the flask, andwere surface sterilized by soaking for five minutes in 70% ethanol and then for ten minutes in 3%HC10 (50% Chlorox bleach). Roots were then rinsed three times in 200 mL sterile distilled water,and blended at high speed for 60 s in a sterile Waring blender containing 20 mL of 10 mM SPB.The root-wash suspensions from unsterilized roots and from the surface-sterilized, blended rootswere diluted serially from 10 -1 to 10-5 in 10 mM SPB, and 0.1 mL aliquots of the dilutions wereplated onto duplicate petri-plates. In order to further increase the diversity of strains recoveredfrom the dilutions, three culture methods were employed: (1) aerobic growth on diazotroph-enrichingcombined carbon media (CCM) (Rennie 1981); (2) anaerobic growth on CCM in anaerobic jars(Baltimore Biological Laboratory, Inc.); and (3) aerobic growth on Pseudomonad-enriching King's Bmedia (King et al. 1954).All media were supplemented with 100 mg/L cyclohexamide and 30 mg/L benomyl (Benlate, W.P.Dupont Inc.) to inhibit the growth of fungi. Plates were held at 28°C for 72 h for aerobic incubation,or for three weeks for anaerobic incubation. After incubation, bacterial colonies of distinctmorphology were isolated by streaking onto new plates of the same media from which the bacteriawere originally cultured. Isolated strains were grown aerobically for 1-4 days, then purified by re-1 7streaking a single isolated colony onto tryptic soy agar (TSA - 20 g Difco tryptic soy agar, 10.0 g agar,1.0 L distilled water). Of the strains which grew on King's B medium, only those which fluorescedunder UV light (300 nm) were considered to be siderophore producers and were purified.3.3	 Bacterial StoragePurified strains were stored at -80°C in order to minimize the potential for genetic mutation whichmay occur with serial re-culturing. This was achieved by culturing strains in tryptic soy broth (TSB- Difco) until turbid and adding 0.5 mL of each culture to 2 mL plastic cryovials containing 0.5 mLTSB in 40% (v/v) glycerol. The resulting suspensions were stirred to facilitate immersion of bacteriainto the glycerol. Suspensions were then held stationary for two hours at room temperature to allowfor glycerol uptake into the cells before storing at -80°C.3.4 Acetylene Reduction AssayTo test for acetylene-reducing capability, isolates were inoculated from frozen cultures into 3 mL ofliquid CCM contained in sterile 5 mL glass vials which were fitted with a rubber seal. Vials wereincubated for 72 h at 28°C on a rotary shaker (120 rev/min). Acetylene was then injected into thevials to a final concentration of 10% (v/v). One mL of gas was withdrawn from each vial 24 h laterand analyzed for C2H4 by flame-ionization gas chromatography, following separation in a stainlesssteel column (0.3 x 180 cm) containing Porapak N (80-100 mesh) at 55°C with N2 carrier gas at aflow rate of 40 mIlmin. Strains registering rates of acetylene reduction ten times greater than the'background' were considered to possess nitrogenase activity.3.5	 Strain Selection and Inocula PreparationTwenty strains from each collection site were selected based on three criteria: the ability to reduceacetylene (an indication of their ability to fix nitrogen); the medium on which the primary isolationwas made (to secure a representative sample of isolates from each type of isolation medium); and thepopulation size of the strain in the rhizosphere of the naturally regenerating seedling, based on thedilution plate from which it was selected (i.e. those present in the largest numbers were selected).1 8The selected bacterial strains were inoculated onto TSA plates from frozen culture and grown for 48h at 28°C. Plate cultures were used to inoculate 150-300 mL of TSB in flasks, which were thenincubated at 28°C for 24-48 h on a rotary shaker (150 rev/min). Bacterial cells were centrifuged(3000 x g for 20 min) and washed by re-suspending the pellet with SPB to the original volume, re-centrifuging and resuspending again. Washed cells were adjusted with SPB to an optical density(OD) (600 nm) intended to give 10 7 colony forming units (cfu) per mL according to previouslyestablished OD/cfu concentration functions for each strain.3.6 Experiment 13.6.1 Seed Sowing and InoculationSeed from the Mackenzie parent-tree stand (seedlot 29144) and the Salmon Arm 'plus tree' wereprovided by the MoF and were stratified by immersion in 250 mL of distilled water for 24 h, followedby surface drying, and storage at 4°C for 30 days. During the first week of April 1990, the twoseedlots were sown, three seed per cell, into plastic cones (Super Cell 160 cm 3 , 4.0 x 21.0 cm, RayLeach 'conetainer' Nursery, Canby, Oregon) filled with a 50:50 mixture of Turface (montmorilloniteclay - Applied Industrial Materials Corporation, Deerfield, Illinios) and their correspondingcoexistent soil (i.e. Mackenzie or Salmon Arm). Soil was sifted through a 1 cm mesh before use.Seed were then drenched with 3 mL of 10 mM SPB which contained 10 7 cfu/mL of one of the twentycoexistent bacterial strains (i.e. Mackenzie seed were sown onto Mackenzie soil and inoculated withMackenzie bacterial strains and Salmon Arm seed were sown onto Salmon Arm soil and inoculatedwith Salmon Arm bacterial strains). Control seed was drenched with 3 mL of 10 mM SPB. Seedswere then covered with 5 mL of 'Forestry Sand' (Target Products Ltd., Vancouver) and wateredlightly.3.6.2 Experimental Design and Seedling CultureTwenty-one treatments (twenty bacterial strains plus the uninoculated control) were tested on eachspruce seedlot (n=20). The 420 'conetainers' of each seedlot (21 treatments x 20'conetainers'/treatment) were arranged in a completely randomized design, and the seedlings were1 9grown in the University of British Columbia Plant Science greenhouse. Daily maximumtemperature ranged from approximately 20-28°C and occasionally reached 35°C. Seedlings werewatered to saturation on alternate days, and daily on the hotter days. Due to extremely slow growthduring the first month after germination, Mackenzie seedlings each received approximately 5 mL ofsoluble fertilizer (650 mg/L 20-8-20 (Plant Products) supplemented with 150 mg/L Fe2(SO4)3) once aweek for four weeks, commencing seven weeks after sowing. An extended photoperiod of 18 h wasachieved with the use of fluorescent lights. Seedlings were thinned to the single largest germinantfive weeks after sowing.3.6.3 Germination Survey, Harvests and Statistical AnalysisThe number of germinants/cell was counted on nine occasions during the active germination period,with the final count taking place 30-45 days after sowing. The germination capacity (GC - finalpercent germination) and germination value (GV), a measure of the speed and completeness ofgermination (Czabator 1962), were calculated for each treatment and compared with the controlusing a two-tailed Least Significant Difference (LSD) at p < 0.05. The germination value (which iscurrently being incorporated into the tree seed registry system of the B.C. Ministry of Forests) givesan overall estimate of a seedlot's germinative quality. To obtain the GV, the germination rate (%germination/days since sowing) was calculated for each cell on each of the nine observation days.The maximum germination rate was then determined, and multiplied by the final germination rateto give the GV.Seedlings were harvested 13 weeks after sowing. Shoot height and root collar diameter weremeasured. Roots were separated from shoots and washed to remove adherent soil. Projected rootsurface area of fresh roots was obtained with the use of a LiCor 3000 surface area meter. The meanof three measurements/root was calculated and multiplied by It to estimate actual root surface area.Shoots and roots were oven-dried for three days at 70°C before shoot, root and total dry weightswere measured. LSDs were calculated as they were for the germination parameters and used tocompare treatment means with that of the control.2 03.7 Experiment 2Following analysis of seedling growth in the first experiment, the three bacterial strains from eachprovenance which elicited the greatest stimulation of shoot dry weight were selected for furtherstudy as putative PGPR in a second inoculation experiment. These six putative PGPR wereidentified to the species level at Auburn University, Georgia, by gas chromatographic analysis ofbacterial fatty acids (as methyl esters) using the MIDI (Microbial ID, Inc.) Microbial IdentificationSystem which has been described by Mertz and Yao (1990).The influence of conifer ecotype, soil source and PGPR on germination and seedling growth wereevaluated in a complete factorial experiment (see Appendix 1). The experimental design permittedevaluation of the influence of bacteria-seed, bacteria-soil, seed-soil and bacteria-seed-soil specificityon germination and seedling growth through the use of contrasts involving control (i.e. uninoculated)treatments, and treatments of coexistent and non-coexistent factor combinations.By utilizing pasteurized and non-pasteurized growing media, the role of biotic and abiotic soil factorsin seed germination and seedling growth were also assessed. One half of the soil-Turface mixtureswere pasteurized by heating 5 kg soil aliquots twice to 100°C for 30 min (24 h between heattreatments) before the 'conetainers' were filled. Fifty-six factorial treatment combinations (n=20)resulted from the use of two seed provenances (Mackenzie and Salmon Arm), two soil sources(Mackenzie and Salmon Arm), two soil types (pasteurized and non-pasteurized) and seveninoculation treatments (three Mackenzie and three Salmon Arm bacterial strains, and a SPBcontrol).Seed were sown and inoculated and seedlings were grown as described for Experiment 1, with thefollowing exceptions: seed were re-inoculated three days after the initial inoculation at sowing;seedlings were grown in two growth chambers (day/night temperatures 24/16°C); a 19 h photoperiodwas used (photosynthetically active radiation was 400-700 umol/m 2/s), and the quantity of fertilizerapplied to the seedlings growing in Mackenzie soil was increased to 13 mL/seedling. Trays of21'conetainers' were rotated within and between growth chambers every two days in an attempt todistribute evenly positional effects among trays. Germination and growth parameters weremeasured as described in Experiment 1.Multi-way analysis of variance was performed on all parameters. After ANOVA, untransformedresiduals were well distributed and no improvement of the homogeneity of the treatment varianceswas observed when data were log, inverse, or square root transformed. Therefore, ANOVA wasconducted using untransformed data. The Least Significant Difference test (two-tailed) at p < 0.05based on the experiment mean square error was used to identify treatments which differedsignificantly from their control.Four sets of contrasts were used to test the general hypothesis that germination and/or growth ofinterior spruce seedlings were greater when coexistent factor combinations (bacteria, seed and soil)were present. The specific sets of contrasts were:Cl - coexistent bacteria, seed and soil vs. non-coexistent bacteria, seed and soilCl - coexistent bacteria, seed and soil vs. controlCl - non-coexistent bacteria, seed and soil vs. controlC2 - coexistent bacteria and seed vs. non-coexistent bacteria and seedC2 - coexistent bacteria and seed vs. controlC2 - non-coexistent bacteria and seed vs. controlC3 - coexistent bacteria and soil vs. non-coexistent bacteria and soilC3 - coexistent bacteria and soil vs. controlC3 - non-coexistent bacteria and soil vs. controlC4 - coexistent seed and soil vs. non-coexistent seed and soil2 2These contrasts were also performed on seedlings grown in pasteurized soil to estimate thecontribution of biotic versus abiotic factors to any observed specificity. The overall comparison errorrate was controlled at a < 0.10 with the use of the Bonferroni procedure (Wilkinson 1988 p. 490).The treatments comprising each contrast are outlined in Table 6 and the treatment numberdesignations are illustrated in Appendix 1.234.0 RESULTS	4.1	 Soil Analysis and Bacterial IsolationThe pooled soil sample from Salmon Arm was richer in all micro- and macro-nutrients and in organicmatter, than that from Mackenzie (Table 1). The only exception was calcium, which was present ata level of 2300 ppm in Mackenzie, but at only 900 ppm in Salmon Arm soil. The two soils alsodiffered substantially in pH (i.e. Salmon Arm soil 4.8; Mackenzie 6.4).Approximately 150 bacterial strains were recovered from the roots of seedlings from each site. Ofthese, 80% originated from the ectorhizosphere, and each of the isolation techniques recoveredapproximately equal numbers of strains. Three of the Mackenzie and six of the Salmon Arm strainswere capable of reducing acetylene (Tables 2a and 2b).	4.2	 Experiment 14.2.1 Germination ResponsesStatistically significant increases in GC or GV due to inoculation were not observed in either seedlot,although strains M4 and M13 both stimulated GC 13% over the control treatment (i.e. GC treatment- GC control) (Fig. la) (Absolute germination values are presented in Table 3). Strain M13 alsoelicited a 23% increase in GV over the control. Inhibition of GC and GV in the Mackenzie seedlotwas frequent and substantial, with GV inhibition greater than 20% occurring with the use of six ofthe strains (Fig. la). Maximum GV inhibition occurred from inoculation with strain M15 (38%).The magnitude of the effects on germination in the Salmon Arm seedlot (Fig. lb) was much smallerthan those in the Mackenzie seedlot. Significant inhibition of GV resulted with the use of strainsS15 (21%) and S20 (21%), and GV inhibition exceeded 10% when strains S8 and S16 were tested.The maximum enhancement of GC and GV was 5% and 8%, and both were achieved with the use ofstrain S2.fa.4 1.0	 Nd'	 0'5	 C+.)CV	 ,"'4	 CI	 •Zrc0 crD c0 ,c1. c01-1coC) 1-1 4.4.41C c**3 Lt N.-4 ..-i ,--i .--;CA N 0 00 N ,-( C).--I .--I r-I 0.1o 0 o 0a.)	 00	 C)7..1	CO1•44	 .--1	 e-4I	 1--1C	 c:::	 0:1	 tir).	 tqN	 C',	 Cg	 in	 m•=t:	 .	 '44.	 c!U	 N N N NI=	 0	 0	 1.0	 inc0 N 0 If),4.1	 v--4	 CV	 CV0 00 0 0 0CI	 CD	 0	 0	 If)C.)	 N	 N	 Cr)	 Cr)0	 In 0	 kt,C)	N	 v-I	 C7)Cg	 CO	 tr)	 VIin0-1 Z	cc) 0cn cn	e. 	 0	 0.0	 6 6 6 C	- 4.)'' e.	 co.	 N	 CI	 c4).ti) al	 N	 14 	 inr-.O PZ	 'zt:	 (4.	 °R	 tz)Q.	 c0	 c0	 NI,	 ,c1,24Cl)alTable 2a. List of Mackenzie bacterial strains used in Experiment 1.Isolation method: (1) CCM, aerobic; (2) CCM, anaerobic; (3)King's B media, aerobic. Habitat: endo - strain isolated fromendorhizosphere; ecto - strain isolated from ectorhizosphere.Ethylene reduction: + - ethylene reducing strain.MackenzieStrain	 Isolation Habitat Ethylene Rank indesignation	 method	 reduction Expt. 1M1 3 endoM2 2 endoM3 1 ecto +M4 3 endoM5 2 ectoM6 1 endoM7 3 endoM8 1 endoM9 1 endoMIO 1 endoMll 2 ectoM12 3 ectoM13 1 ectoM14 3 ecto 1M15 2 endoM16 3 ectoM17 1 ectoM18 2 ecto + 2M19 1 ecto + 3M20 1 ectoStrain identification: M14 - Pseudonionas putida; M18 and M19 -Hydrogenophaga pseudoflava. The top 3 ranking strains fromExperiment 1 with respect to seedling shoot dry weight promotionwere used in Experiment 2.25Table 2b. List of Salmon Arm bacterial strains used in Experiment 1.Isolation method: (1) CCM, aerobic; (2) CCM, anaerobic; (3)King's B media, aerobic. Habitat: endo - strain isolated fromendorhizosphere; ecto - strain isolated from ectorhizosphere.Ethylene reduction: + - ethylene reducing strain.S almon ArmStraindesignationIsolation HabitatmethodEthylene	 Rank inreduction Expt. 1S1 2 endo 3S2 1 endo +S3 1 endo +S4 1 ectoS5 1 endoS6 2 endo +S7 1 endoS8 3 endoS9 1 ectoS10 1 endo 2Sll 3 endoS12 1 endoS13 3 endoS14 2 endoS15 2 endoS16 2 ecto +S17 1 ectoS18 2 ecto +S19 2 ecto +S20 1 ecto 1Strain identification: Si -Pseudomonas putida; S10 - Staphlococcushominis; S20 - Bacillus polymyxa. The top 3 ranking strains fromExperiment 1 with respect to seedling shoot dry weight promotionwere used in Experiment 2.26GC treatment - GC control % change from control (GV)a	 Mackenzie2010-10-20-30Fig. 1. Effect of Mackenzie and Salmon Arm bacterial isolates on the germinationcapacity and germination value of their respective seedlots. Seed wasgrown in soil collected from the same location as was the seed. * -indicates values which differ significantly from the control at p < 0.05.-4011111111111M	 11111 M1 M2 M3 M4 M5 M6 M7 Me M9 M10M11M12M13M14M15M16M17M18M19M20strain♦ germination capacity	 F-71 germination valueb	 Salmon ArmGC treatment - GC control	 % change from control GV272010-10-20-30-4020102010-10-20-30 -30" -40-40 	iiiIIIIIIIII	 1	 1	 I	 t	 ilS1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 s12 S13 S14 S15 S16 s17 s18s19s20strainsIM germination capacity	 F A germination valueTable 3. Effect of bacterial inoculation with Mackenzie andSalmon Arm isolates on germination capacity (GC) andgermination value (GV) of Mackenzie and Salmon Armseedlots in Experiment 1. Ctrl - SPB control; MSE -mean square error; * - indicates treatments whichdiffer significantly from the control at p < 0.05.StrainMackenzieGC GVSalmon ArmStrain	 GC GV(%) (%)M1 72.5 10.8 Si 90.0 23.9M2 53.7 6.9 S2 99.5 27.2M3 68.7 9.3 S3 91.6 25.3M4 77.5 11.6 S4 95.0 27.1M5 63.7 9.5 S5 95.0 26.0M6 51.2 7.3 S6 96.6 25.6M7 57.5 7.2 S7 95.0 24.1M8 52.5 7.5 S8 90.0 22.4M9 57.5 8.5 S9 90.0 23.1M10 58.7 8.8 S10 95.0 25.1Mll 65.0 10.1 S11 93.3 23.6M12 67.5 11.0 S12 91.6 23.4M13 77.5 13.0 S13 96.6 25.9M14 72.5 11.9 S14 96.6 26.8M15 55.0 6.4 * S15 86.6 20.0 *M16 68.7 10.4 S16 88.3 21.3M17 62.5 9.0 S17 91.6 23.6M18 63.7 8.5 S18 96.6 25.4M19 65.0 9.5 S19 93.3 25.5M20 61.2 8.4 S20 88.3 19.8	 *Ctrl 63.8 10.6 Ctrl 95.0 25.3LSD 16.80 3.78 10.80 4.70MSE 734.26 37.28 303.68 57.71282 94.2.2 Growth ResponsesWhile Mackenzie strains inhibited Mackenzie seedling growth (Figs. 2a and 3), most of the SalmonArm rhizobacteria promoted the growth of Salmon Arm seedlings (Figs. 2b and 4). However, effectswere rarely significant. (See Tables 4a and 4b for treatment means of both seedlots.)Although inhibition of growth was widespread and often substantial among Mackenzie inoculants,the only significant decreases were in height with strains M7 and M9 (20% and 21% respectively)and in root dry weight with strain M7 (24%) (Figs. 3a and 3c). Considerable increases occurred inshoot dry weight with strains M14 (19%), M18 (12%) and M19 (16%) (Fig. 2a), and in root dryweight, also with M14 (20%) (Fig. 3c). The only significant growth promotion due to inoculation withMackenzie bacteria was of stem diameter, by strain M18 (10%) (Table 4a).Height growth of Salmon Arm seedlings was greater in all bacterial treatments, relative to thecontrol, and was significantly promoted by strains S1 (12%), S5 (13%), S11 (11%) and S20 (18%) (Fig.4a). Strain S20, the most effective height growth promoter, also produced significant shoot dryweight (27%), root surface area (20%), and total dry weight (21%) increases (Figs. 2b, 4b and 4d). Sixother strains produced non-significant increases in shoot dry weight ranging from 11-18%.Root/shoot dry weight ratio was lower (relative to the control) in all but two of the Salmon Armtreatments (range 3 to -20%) (Table 4b), and was increased in 13 of 20 Mackenzie treatments (range-15 to 42%) (Table 4a). The increase in root/shoot dry weight ratio, however, generally resulted frominhibition of SDW, and not from promotion of root dry weight. Inoculation with strain M15, forexample, resulted in a 42% increase in root/shoot dry weight ratio, which was due to a slightdecrease in root dry weight and a moderate decrease in SDW (Figs. 2a and 3c). The only significanteffects on root/shoot dry weight ratio occurred with strains M15 (42% greater than controls) and S11(20% below controls).30Fig. 2. Effect of inoculation with (a) Mackenzie and (b) Salmon Arm bacterial strains on shoot dryweight of their respective seedlots. Seed was grown in soil collected from the same location aswas the seed. * - indicates treatments which differ significantly from the control at p < 0.05.Mackenziea	 shoot dry weightchange from control30201 00-10-20 M1 M2 M3 M4 M5 M6 M7 M8 M9 M10M11M12M13M14M15M16M17M18M19M20strainb Salmon Armshoot dry weight% change from control3020100-1 0-20SI S2 S3 S4 S5 S6 S7 S8 S9 SIO S11 S12 S13 S14 S15 S16 S17 S18 S19 S20strain-DANNJ0	 0	 O	 0nNMI31X cntar)• z(1)cC3 Etts)2)-0 .0	 1.12sg 0id-	 CtSa) .5cs 4E'u,	 CCE8 'r6 rox.5O 0cu74.%)	 •_z0 CDg.5 (130 160 20 cito VU)b.!) 4-1C) ••l 71:10 cl)a) (1)w gc.,cd •a) 4_3„s43 cvl Ea)•,-. 0 •I • hlbjD tc-)o (1)Cy+)0.)41 3 .8CDO	 0-ozC8E0 0 O13• U.Th'24J0320 0 0Cd 00Cn0 >,0)CI)00 t4.C) . 135230 0C)"d5+60cr-aC8QQEL0033Table 4a. Experiment 1 treatment means for Mackenzie seedlings. TRT - strain used for inoculation; SDW-shoot dry weight; HT - height; DIAM - stem diameter; RSA - root surface area; RDW - root dryweight; R/S - root/shoot dry weight ratio; TOTDW - total seedling dry weight; LSD - leastsignificant difference; MSE - mean square error. * - indicates values which differ significantly fromcontrol at p < 0.05.Mackenzie ProvenanceTRT SDW(mg)HT(mm)DIAM(mm)RSA(cm2)RDW(mg)TOTDW(mg)Ml 22.4 24.1 0.514 6.67 15.1 0.76 37.5M2 23.1 25.2 0.531 6.95 14.2 0.67 37.3M3 22.0 24.1 0.535 6.69 14.4 0.76 36.5M4 20.4 23.8 0.528 6.98 14.9 0.87 35.3M5 23.1 24.9 0.542 6.42 14.8 0.87 37.9M6 19.8 23.2 0.506 5.56 12.2 0.66 32.1M7 19.6 21.5 * 0.536 5.55 11.7 * 0.65 31.4M8 24.7 24.6 0.527 6.66 13.8 0.62 38.5M9 18.6 21.3 * 0.504 5.71 12.6 0.80 31.2M10 25.2 26.4 0.524 7.36 17.3 0.77 42.5Mll 22.6 24.5 0.547 6.40 13.4 0.61 36.0M12 25.2 26.3 0.565 7.48 15.7 0.73 40.9M13 20.8 25.2 0.525 6.90 14.0 0.70 34.8M14 28.7 27.6 0.573 8.31 18.6 0.75 47.4M15 19.8 23.6 0.500 6.34 14.4 0.97 * 34.1M16 20.1 23.8 0.515 6.67 14.0 0.87 34.1M17 20.9 24.2 0.527 6.69 15.5 0.85 36.4M18 27.2 27.7 0.589 * 7.52 16.6 0.66 43.8M19 28.0 27.1 0.532 7.16 15.6 0.58 43.6M20 23.9 26.2 0.541 7.53 15.9 0.76 39.9control 24.2 26.8 0.534 7.16 15.6 0.69 39.8LSD 7.30 4.05 0.0542 1.672 3.69 0.237 10.3MSE 138.78 42.90 0.0076 7.275 35.35 0.146 276.93 4Table 4b. Experiment 1 treatment means for Salmon Arm seedlings. TRT - strain used for inoculation;SDW -shoot dry weight; HT - height; DIAM - stem diameter; RSA - root surface area; RDW - rootdry weight; R/S - root/shoot dry weight ratio; TOTDW - total seedling dry weight; LSD - leastsignificant difference; MSE - mean square error. * - indicates values which differ significantly fromcontrol at p < 0.05.Salmon Arm ProvenanceTRT SDW(mg)HT(mm)DIAM(mm)RSA(cm2) RDW(mg)R/S TOTDW(mg)Si 64.4 33.8 * 0.842 16.81 37.5 0.61 101.9S2 58.7 32.2 0.835 15.00 35.8 0.67 94.5S3 55.5 32.2 0.845 14.66 35.6 0.67 91.1S4 56.6 31.6 0.812 16.30 36.5 0.66 93.2S5 62.7 34.0 * 0.853 16.21 37.2 0.60 99.9S6 57.9 31.7 0.826 14.21 31.7 0.63 89.7S7 62.5 33.1 0.864 17.50 38.5 0.64 101.0S8 54.9 30.3 0.810 15.71 33.4 0.65 88.3S9 62.4 30.9 0.860 16.47 35.6 0.61 98.1S10 66.5 * 32.8 0.866 16.06 36.8 0.57 103.3511 63.7 33.5 * 0.827 15.08 32.2 0.52 95.9S12 55.8 31.6 0.804 14.46 31.8 0.60 87.6S13 56.7 31.9 0.818 15.21 33.5 0.64 90.3S14 58.8 30.9 0.831 16.57 35.9 0.66 94.7S15 55.0 32.3 0.844 14.40 32.7 0.64 87.8S16 58.7 32.0 0.836 15.16 34.0 0.60 92.8S17 56.5 31.6 0.818 15.25 33.7 0.71 90.2S18 59.8 33.1 0.841 16.34 34.7 0.60 94.5S19 58.7 32.3 0.824 15.67 34.4 0.62 93.1S20 71.3 * 35.5 * 0.862 18.38 * 39.6 0.56 110.9 *control 56.3 30.2 0.805 15.32 35.5 0.67 91.8LSD 7.40 3.28 0.073 2.661 5.46 0.127 15.26MSE 142.54 28.01 0.014 18.429 77.49 0.042 606.19354.3	 Experiment 24.3.1 Germination ResponsesThe three most effective SDW promoting strains from each location in the first experiment wereselected for re-testing in the second experiment. The most effective SDW promoter from Mackenziewas strain M14 and was identified as Pseudomonas putida. The second and third best strains wereM18 and M19, and both were identified as Hydrogenophaga pseudoflava. Both strains M18 and M19were capable of reducing acetylene. The most effective SDW promoting strains from Salmon Armtested in the first experiment were (in decreasing order of effectiveness) S20, S10 and Si. Thesewere identified as Bacillus polymyxa, Staphlococcus hominis, and Pseudomonas putida, respectively.Soil source and pasteurization had a significant effect on germination capacity, which was 3.6%greater in Mackenzie than in Salmon Arm soil, and 6.4% greater in pasteurized than inunpasteurized soil (Fig. 5a). (See Table 5 for absolute germination values and Appendix 2 forANOVA.) A significant seed x bacteria interaction for GC was also detected, although no strainproduced a GC response significantly different from its corresponding control (Fig. 5c).Germination value also varied significantly between seed provenances and soil types (Fig. 5b). GVwas higher in the Salmon Arm (8.3) than in the Mackenzie (6.3) provenance, but lower in the SalmonArm (7.0) than in the Mackenzie (7.7) soil. Although the bacteria x seed provenance interaction wastechnically non-significant (p = 0.107), the effect of most of the inoculants on GV depended greatly onthe seedlot used. For example, strains M14, S10 and S20 all promoted GV by greater than 15%, buthad an inhibitory or negligible effect on the Salmon Arm seedlot (Fig. 5d). Coexistence SpecificitiesIn most of the factor combinations which were tested, coexistence had little effect on germination.The germination capacity of treatments comprised of coexistent bacteria and unpasteurized soil wasslightly, but significantly lower (5.8%) than those comprised of non-coexistent factor combinations(Table 6 C3a). Coexistence of seed and pasteurized soil also resulted in a significantly lower-1136ei v-a ,a0 vCOWtd CVcp 4 gcyVtr,-0 5,e >aa:6. 0 0ucoO03	 W,csE udoon .4,c El 6oo co0 fj •—•1:1.15-1fiu)„„ 0t0 z0 -0O g0 • 0 u°0 owtcn Q.coCu 'C al>o g 0R. ,0"0) gc 0744 2 trl0 0,II0.) W 03i 0WLIDCcoCEio0	 co	 fD	 cv	 00coCEul 0	 117- 6CcoEioE12C112roOC0OCco-c22U,F'x3 7Table 5. Mean germination capacity (GC) and germination value (GV) of 56 treatments in Experiment 2. SEED- seed provenance; SOIL - soil source; PAST - pasteurization treatment; BACT - putative PGPR strainused; M - Mackenzie; S - Salmon Arm; unpast - unpasteurized; past - pasteurized; MSE - mean squareerror; LSD - Least Significant Difference. * - indicates values which differ significantly from control atp < 0.05. Horizontal lines group treatments with their appropriate control. Shaded regions containtreatments in unpasteurized soil.TRT SEED*SOIL* PAST* BACT	 GC GV(%)iiiiPast M14 '	 60.0	 5.4 '. M	 M	 =past M18 -	 - 61.7 '.. 6.1 ' M	 M	 •• unpast, M19	 •-.... 65.0 	 4.0 M	 M	 , Unpast • CONTROL .......10.0 -  8.2M	 M	 unpast 820	 ' - 75.0 - 7.5 .•:	6.	 M	 M	 unpast 810	 78,8	 8.3	:::::7 ::	 :: - M	 ' ' M	 • unpast 81 ' '	 ' 65,0 	 5.9'''':'8	 • M	 M	 past	 M14	 80.0	 8.19	 M	 M	 past	 M18	 73.3	 7.1	10	 M	 M	 past	 M19	 75.0	 6.7	11	 M	 M	 past CONTROL	 66.7	 5.4	12	 M	 M	 past	 S20	 81.7	 8.0	13	 M	 M	 past	 S10	 80.0	 7.4	14	 M	 M	 past	 S1	 83.3	 8.5  *	45.: ::::: 'l M	 S 7 .:	 .: 11.1.03„St ::M14	 CT!	 P4:".,::	31;6'	 : M	 -	 ::$ 	 :: *11:41$:: M10:::::::	 .1.. 	 5 .9 :'''47.:,: :: 	::: M :	 : ::.S. :	1.iiipa:::M19.	 ::::	 .	 .0:::: ::: 5.0	40	 :-::M ::	 :'::$ :::'	 ::iiilpitst.::CONTRQL::: .	 : , ''50.7:::.. :,. 4:0:..2	 S	 ,: M	 invest CONTROL':::: :" .75,0. ' l', 9.0: N:: . : . S M:15 S M36 S M37 S M38 S M39 S M40 S M41 S M42 S M:43	 : S S44 S S.45 S S...., S S	 .37 : S S48	 , S S39 S S50 S S51 S S52 S S53 S S54 S S55 S S56 S SLSD	 17.03 2.84MSE	 755.38 21.06..	 :::	 : M	 . $	 : .	 : : unpast 820 ::::: : '	 .::: 65,0 .:':::i: 	 8,1,	 :::::29 -	 ; : M	 :S	 . unpast : $10 : :::':	 ::: 61:7 :':. :': 5.5:.21 	 M	 :	 Sunpast St	 :::'. • : 	  ' 63;3:: :: 5.1 22	 M	 S	 past	 M14	 76.7	 6.023	 M	 S	 past	 M18	 58.3 * 3.7 *24	 M	 S	 past	 M19	 65.0	 4.925	 M	 S	 past	 CONTROL	 80.0	 7.726	 M	 S	 past	 S20	 75.0	 6.727	 M	 S	 past	 S10	 71.7	 6.128	 M	 S	 past	 S1	 70.0	 6.39 ::: : $	 :: : M	 '	 unpast M14	 76.7 " :'::: 9,309 :	 :; . 5	 ::::: .::: M	 : unpast M1& .. : . : : :. :	 ::00.3::: :;:: 7.8 :'•*:1.	 .$ :: : ::. : M	 unpast M19 :: ::..   	 66:7.: : H:::74: .3	 S	 Unpast 820 • :: ::::	 ::81.7 :.::.::: :: 10.2 : ;unpast 810	 ::::=.65.0 :::::::::: 7.9 - -''unpast Si	 i 81.7: ' , 9.2past	 M14	 71.7	 8.2past	 M18	 78.3	 9.0past	 M19	 85.0	 10.2past	 CONTROL	 83.3	 9.5past	 S20	 76.7	 8.1past	 S10	 66.7	 6.8past	 S1	 80.0	 8.6unpast M14	 :: 71,7	 7.5unpast M18	 66,7	 7.9unpast M19	 70.0	 8.4unpast CONTROL : :	 66,7	 7,7unpast S20	 70,0	 7.9unpast S10	 63.3	 ::: 8.2 : :unpast Si	 : 80,0	 • 9.5 :.:past	 M14	 80.0	 8.8past	 M18	80.0	 9.1past	 M19	 80.0	 8.7past	 CONTROL	 66.7	 7.3past	 S20	 66.7	 7.0past	 S10	 69.7	 6.3past	 S1	 _80.0	 8.015b_ co ,c,b -	4. C) C 4a'	 "a'	 "a'	 4_,c	wo s 	w	 w	 w	4., V) -4) V) 	4.) V) 	44 V) 	4.,V)	CV  CV	 CV	 CV	 CV8s 8 28— 28— B cucV.' 	I7 F	8SC a)SC	 S,82	 u08 808.1j0	 f?).121	 0?4.)380	 0WMMM MMM •-..-M MM .--4-4(DM MMM MMM m0> C) d, mNm C) t- mt- > Nwm•cr t-0,a, vmd: Mm0 Mt-t1iM t-:t-:M cit.:cc;A:oxoco cD 0 t--:t clictit-xmxmt--:t-:N u6co04U00c1-4mWat-:mcoNwtoNO.-4ccimt-:towwqco.'1cot-4N-coMopcoowwC.)CI'1LPI--eqNNtoNMAN c.im4NNNht-NNt.miNNNNNcii,-4tomU)OU)'1..t—I'tti0.-'1...	00 	 N	m	 U)	 oN	M "	 CO 4	t--	 r4	M 	 ... W	 W	m- 	 in 0-	M	 ulo	 ul.oo -lao	N 	 Ilicg	 ..d•	t.--:	 Nt...:	 m4	N	 "ctm	 m.....44	..1,1,-	 NO	tr 	 .	i	..t	 m	Cl 	 -M	 -.TM	4 	 tx:C 	 ,t7	d:,.4 	 M NClNNt....-(C en-	u)csi	cqcsi-	NNCI N2c4' m wzm t- cicl	. ,.., 	 L.	 ..N1.0	 .. cv„,v -14' a)" c,i" 4 oi c.,iu:' 4 oi	,--4 	 M	 .. r-I M	 .. r-.1 Cf"),c'rimui °o-lo- c'rio" mm"1,-;oN .1,-(esi '1,-4N omcla,4,4 01444 cla,4 4,,...CO	 CO . . m.. ,,,,.t31m0-	 14 44m4-	 ej	 t-- cl- 	t-'C	cl"M.'Cl-	 *-4, .1....	 .1. -.1,Cftfi".m .5n C	 Cl.-,.A	 tri,-4Mm	 mcl.0t11C •	 P	 N0,-0 mm mmricc	 criri4O O44c/co "cl a	 --,	 MN	 NmaC.41.,	0	Mt--atic6. miO	 c5(.6,--IN..nv—IN`s) cv,-4 m.bx	 t-- in a	 in o) cosi. cm m04	c.,,-	"g	 't.,--4	 ,..4,-,	 ,,......c0	 ..co	 .c-3MO	 vit..--.03-m vit--...ce Mtn 	mco	Nccl	 C	 am,--4.1,. am-4 cvm. v.„„5.0w o	,4.0-4..v.	 .48.64 .4111".ar v.,..-4-C g-0Ezt,8F+ d 75s . G.,...m4g m0	 t _d4, d, .,..	 8.E	 oPcsw4., 	4.3..4,	 430	 ,,,	 4„.cvvc cv	 cv	 cc..w	"a—a'z	 'a'	 c	 c	 w4..	 4.,BE.4 sse	 s	 s	 s	 c)Yo c 7:1	 88 	 B8—	 8— S8	a j ty 	 yti)	 0	 U7	 0	 t/Icitg14P 'PY-b VYb 1	 5C --	8,9s8 000 85,9 8g0cd.co. uc. uc. .c-c12 ,2c	 cu,r3B c'	 4	 3	 4., ,..q0.	 in.,ci.) . cdcu	 !—•a	_o	CnE-IC.) cu'.6 C.1..ti.a,oCwcl. els_aZaau../3_,,Ii)au	aa)0.,aZ4H 0 N M .1. 0E-I  E5 0 0 0 3 9germination capacity and germination value, compared with non-coexistent seed and pasteurizedsoil factor combinations (Table 6 C4b).4.3.2 Growth ResponsesThe results and discussion of growth parameters in Experiment 2 will concentrate on SDW due tothe high correlation between this parameter and the others (height - 0.72; diameter - 0.80; rootsurface area - 0.73; root dry weight - 0.79; total dry weight - 0.95), and the importance of SDW inseedling growth and survival.Due to the strong 4-way treatment interaction (bacterial strain x seed provenance x soil source xpasteurization treatment) (Appendix 3), the effect of each of the six bacterial strains on SDW ispresented separately for the eight seed x soil x pasteurization combinations (Fig. 6). (Effects on theother growth variables are illustrated in Appendices 4-9, and means of the 56 treatments arepresented in Table 7.) Uninoculated control means are illustrated separately (Fig. 7), in order toemphasize the effects of the 3 other main effects (seed provenance, soil source and soil pasteurizationtype), and to best distinguish their effects from the effects of the bacteria on seedling growth.Four of the six selected bacterial strains caused statistically significant growth promotion (Fig. 6),however, the growth response depended on seed and soil source, and on soil pasteurization. A strainwhich was effective with a particular factor combination often inhibited SDW in other factorcombinations. For example, strains S1 and S10 increased SDW of Mackenzie seedlings by 32% and53%, respectively, (over uninoculated controls) in pasteurized Salmon Arm soil (Fig. 6b); but thesetwo strains inhibited SDW of the same seed provenance by 12% and 19%, respectively, when grownin pasteurized Mackenzie soil (Fig. 6a). Similarly, strain M19 promoted SDW of Salmon Armseedlings by 37% when grown in pasteurized Mackenzie soil (Fig. 6c), but growth of Mackenzieseedlings when seed was inoculated with the same bacteria and grown in the same media wasinhibited by 12% (Fig. 6a).0 0N0Ocs,CD_osCAS8EF5E-5-o40c.2 44x .42C)O• IV• §-g a);)•Zu..204)'L7• 4., ic0.E; 3t'-0 4-, o"to .b..ocn ,n 8c.„," 5tta,a> 5tf32CL) bp(14 : 't7)>/1	 1.13Day• a) 3O-5 -2O coIC) A •Eg4.)0-.6)0 (I)i++4.2°• .09 UCtlcu	 :-8ez 4Wtriabti4"4 1Table 7. Growth parameter means of 56 treatments in Experiment 2. TRT - treatment number designation;SEED - seed provenance; SOIL - soil source; PAST - pasteurization treatment; BACT - putative PGPRstrain used; SDW - shoot dry weight; HT - height; DIAM - stem diameter; RSA - root surface area; RDW- root dry weight; R/S - root/shoot dry weight ratio; TOTDW - seedling total dry weight. past -pasteurized; unpast - unpasteurized; LSD - Least Significant Difference; MSE - mean squares error. * -indicates values which differ significantly from control at p < 0.05. Horizontal lines group treatmentswith their appropriate control. Shaded regions contain treatments in unpasteurized soil.TRT SEED*SOIL*PAST *BACT	 SDW HT	 DIAM RSA RDW R/S TOTWT(mg)	 (mm)	 (mm)	 (cm 4)	 (mg)	 (mg)1234567MMMMMMMMMMMMMMunpastunpastunpastunpastunpastunpast.unpastM14M18MI9CONTROLS20SIOSi28.327.729.628.930.335.128.825.0027.9428.8028.0528.2528.8527.680.5510.4890.5770.5510.5480.5800,5596.,748.666.948.28 M M past M14 33.2 26.70 0.551 * 8.0 22.1 0.693 55.49 M M past M18 31.5 * 23.70 0.557 * 8.3 22.2 0.732 53.710 M M past M19 38.6 21.75 * 0.574 9.9 25.4 0.675 64.111 M M past CONTROL 43.9 25.65 0.648 11.6 28.1 0.664 72.112 M M past S20 35.6 21.20 * 0.599 9.8 25.4 0.793 61.013 M M past S10 35.6 22.85 0.554 * 9.2 21.4 0.613 57.114 M M past Si 38.7 23.55 0.608 10.2 24.0 0,647 62.715 M S unpast M14 42.5 26.40 0.633 11.7 32.7 0,795 76.316 M S unpast M18 43.5 30.36 0.653 11.9 30.1 0.701 73.717 M S unpast M19 40.3 26.78 0,638 10.8 27.4 0.748 67.718 M S unpast CONTROL 45.5 28.63 0.639 11.2 31.0 0.669 76.619 M S unpast S20 50.8 29.05 0,692 12.0 32.4 0.669 83,320 M S unpast SlO 40.9 27.45 0.652 10.5 28.9 0730 69.9'21 M S unpast 51 50.0 30,50 0,704 14.2 35.0 0.706 85.222 M S past M14 39.1 24.80 0.655 18.4 39.3 1.007 78.523 M S past M18 39.0 25.70 0.635 18.7 39.4 0.973 78.524 M S past M19 49.7 26.45 0.718 22.2 47.9 1.021 97.725 M S past CONTROL 43.2 26.75 0.699 19.5 38.9 0.934 82.226 M S past S20 34.9 25.45 0.652 17.5 38.3 1.107 * 73.327 M S past S10 66.1 * 33.75 * 0.828 * 30.1 * 64.7 * 1.034 130.9 *28 M S past Si 57 0 * 32.42 * 0.732 26.8 * 60.1 * 1,144 * 117.2 *'29 S M unpast M14 31.4 24.80 0.414 7.8 19.2 0.658 * 50.630 S • M unpast MI8 25.8 23.33 0.618 8.3 20.1 0.804 46.131 S M unpast M19 25.6 23.35 0.592 9.9 20.1 0.766 * 47.032 S M unpast CONTROL 22.8 22.84 0.570 8.7 19.3 0.976 42.233 S M unpast S20 24.7 22.78 0.629 8.2 19.3 0,809 44.134 S M u npast S10 30.8 25.38 0.625 8.7 22.1 0.841 * 53.735 S M unpast SI :30.1 23.75 0.617 8,9 21.1 Q.768 5L636 S M past M14 34.0 20.90 0.681 12.8 28.1 0.811 61.437 S M past M18 39.8 23.70 0.679 14.0 33.2 0.863 73.238 S M past M19 46.2 * 23.84 0.699 15.1 35.7 0.906 82.1 *39 S M past CONTROL 33.8 20.65 0.685 12.4 27.0 0.874 60.840 S M past S20 36.7 20.90 0.680 12.9 28.8 0.838 65.741 S M past S10 45.5 * 23.90 0.750 15.4 33.2 0.772 78.742 S M past S1 38.4 23.05 0.717 13.7 30.6 0.893 69,143 S S unpast M14 61.5 32.40 0.793 19.6 50.6 0.877 112,244 S S unpast MI8 56.9 29.72 0.776 17.6 47.3 0.630 105,345 S S unpast M19 61.7 33.10 0.778 18.5 50.5 0,834 112,446 S S unpast CONTROL 59.5 29.94 0.770 17.0 50.9 0.852 110.547 S S unpast S20 55.8 31.95 0.766 17.2 46.1 0.829 102.048 S S unpast SIO 57.0 30.29 0.'788 16.3 46.2 0.825 103.349 S a unpast SI 61.7 32.55 0.813 18.2 47.9 0,805 109,750 S S past M14 56.9 32.21 0.814 30.6 64.9 1.179 121.9 *51 S S past M18 52.4 29.84 0.788 26.2 54.8 1.135 107.352 S S past M19 51.8 28.89 0.748 26.9 59.4 1.154 111.353 S S past CONTROL 46.9 27.27 0.774 25.4 54.6 1.169 101.654 S S past S20 60.9 30.66 0.840 29.2 62.9 1.082 123.8 *55 S S past S10 54.0 29.60 0.810 25.2 62.2 1.202 116.356 S S past S1 53.9 29.60 0.771 24.2 57.6 1.127 111.6LSD (2-tailed, p < 0.05) 11.19 3.768 0.0810 4.28 9.36 0.1739 19.26MSE 325.88 36.963 0.0171 47.63 228.15 0.0787 965.66.Va)N*:=MW1/5CZCl7NE.ZZc0E15U)u)0 0 0CD	 LO	 ,:t0 0 0 0CO	 C \J	 •r-4 3Given the preceding examples of the extreme dependance of the PGPR effect on particular seedprovenance, soil source and soil pasteurization combinations, strain S10 nonetheless stimulatedSDW (and other growth parameters) in 4 of the 8 seed x soil x pasteurization treatmentcombinations and strain S1 caused promotion of SDW in five of the eight combinations (Fig. 6).However, the promotion of SDW in Experiment 1 due to inoculation by the selected strains was notrepeated in the same coexistent bacteria-seed-soil treatments of Experiment 2 (Fig. 8), althoughheight was generally promoted in both experiments.Despite the variation in PGPR response between the different seed x soil x pasteurization treatmentcombinations, most of the bacteria responded somewhat similarly within each factor combination.Inoculation of Salmon Arm seed in either soil source generally resulted in SDW promotion regardlessof pasteurization, with five of the strains promoting SDW greater than 30% above uninoculatedcontrols in at least one of the four soil source x pasteurization type combinations (Figs. 6c and 6d).Mackenzie seed on the other hand, varied considerably in its response to inoculation. All strainsdepressed SDW of the Mackenzie provenance when grown in pasteurized Mackenzie soil, and, withthe exception of S10, had a negligible effect (<5%) in unpasteurized Mackenzie soil (Fig. 6a). Thesingle exception to the generalization that strains within a seed x soil x pasteurization combinationresponded similarly, occurred with the inoculation of Mackenzie seed in pasteurized Salmon Armsoil: responses ranged from 53% promotion (S10) to 19% inhibition (S20) (Fig. 6b). There was nosignificant effect of inoculation of the same seed x soil combination in unpasteurized soil.Strong interactions were detected involving the three other main effects (seed provenance, soilsource and soil pasteurization) (see control means Fig. 7 and ANOVA - Appendix 3). Mackenzieseedlings were 22% larger than Salmon Arm seedlings when grown in Mackenzie soil, and SalmonArm seedlings were 31% larger than Mackenzie seedlings when grown in Salmon Arm soil. Thesame pattern of growth occurred in pasteurized soil. In general, the SDW of seedlings grown inSalmon Arm soil exceeded that of seedlings grown in Mackenzie soil regardless of seed x0N—U)STaxC C0 O o• Er) 044csi5Oa)EcuELIcuF.rt3a)cncn117C3)a)tncid4.24cr)•c1)0•••0•••cd0"P)0r4.2a)•.-.0"menO•ai4.)0(XiOD2COE2OZccU_c3-O O 0 0 074 5pasteurization combination. The only exception was Mackenzie seed sown in pasteurized Mackenziesoil; resultant seedling SDW exceeded very slightly (2%) the SDW of the same seed in pasteurizedSalmon Arm soil. Pasteurization resulted in profoundly larger seedlings when Mackenzie soil wastested (57% for Mackenzie seed and 49% for Salmon Arm seed), however, it had the opposite effecton seedling growth in Salmon Arm soil (3% inhibition of Mackenzie seedling SDW and 25%inhibition of Salmon Arm seedling SDW) (Fig. 7). Coexistence SpecificitiesThe SDW of seedlings in treatments comprised of coexistent three-factor combinations (i.e.Mackenzie seed inoculated with Mackenzie bacteria and grown in Mackenzie soil, and Salmon Armseed inoculated with Salmon Arm bacteria and grown in Salmon Arm soil) was significantly greaterthan the SDW of uninoculated control seedlings (coexistent and non-coexistent) and seedlings intreatments which were not comprised of coexistent seed, soil and bacteria combinations (i.e. themean of all other inoculated treatments) (Table 8 C l a and Fig. 9a). Seedlings in treatmentscomprised of coexistent bacteria-seed-pasteurized soil, bacteria-seed and bacteria-soil did not differin SDW from seedlings with the same non-coexistent factor combinations, nor from the controlseedlings. (Inoculated seedlings grown in unpasteurized soil were contrasted only with controlsgrown in unpasteurized soil, and inoculated seedlings grown in pasteurized soil were contrasted onlywith controls grown in pasteurized soil.)However, treatments comprised of coexistent combinations of seed and soil (Mackenzie seed grownin Mackenzie soil and Salmon Arm seed grown in Salmon Arm soil) resulted in seedlings which hada significantly and substantially larger SDW (27%) in comparison with treatments comprised of non-coexistent combinations of the same factors (Fig. 9a and Table 8 C4 a). The same contrast (coexistentseed-soil versus non-coexistent seed-soil) of seedlings grown in pasteurized soil did not indicate asignificant influence of pasteurization, although the SDW advantage due to the presence ofcoexistent combinations was reduced to 17% (compared with non-coexistent factor combinations)(Fig. 9b).46o	A	 4mo	 IW	 644o	 OA	 GOA OSA4 	 mvt...a) .-4.-4r.	 Ne.- Tt-Wacclol [-..-4r4 rWot-M mq101010Ncrie•im .46°6 e-:m	momm t•vm mom mc...muloto .toom - O.-.O ON1	 061.coto Nt-:Di gc.ie to-4I '	 t...w.,	 wt...., t.,., -.t,	 wt.t.,,, woo,-	 1Nmt.-.4;	 P	 04 K8P Pg8 48 PR	 '425g0 up4S n su,.`= 	 1	 t-t-- elelm: N N 00 N R6606 oclo 000 60	 ....5	 0101011 v ..lolot )01	 qg. 00c; 06= 4"	.JaAJD0 1,	--NO	414	 4 NAls 44 A VIA OAE0 .0	 et	 To-400 0No NO0 12).-4142t1Wm V:CsPAR t'lv!W OWN t".-10W 0.-40 We-0t.'qC/IW '1 N471 7W 0? WM'''NI l' g gnm ARM RI RR gin	 g0 44n? me44 vgn 4g.0 i0001 .N	 tn Neht- 000) 0r-o t....-1	 1-40N0 OON NC-N 00Plsi	i 	.-l , MVNWN Nt-W VWW NO	m q,loi	 t-...4 CEO t-:OC: ClitZeZ WW._	..4.-4,4.-4	 .-4.-4,44	 .-1,-1.-1 	 ,-1.-1	 ”41-1.-4.-4	 .-4.-t.-4	 ..4.-1.-•	 .-81-t11!	NiAlg-iNov .-4.-40 .-INO VO	 .-o0OW t..tn O TWO ONgi-t	 1	c-mko locotooulw w%om cmc! ww wo wm tow wo6606 660 666 60	1 mmo	 mmo mwocc! o ulm tom t....o cociot- t°-,CC6000 066 666 oo3.	 A6446	 44	 ANO40N -ir .4 0V.-401 0N.-4 oo.-4 or..Mtn t•:W t.:WC-: tn t"-t.: WO gr-crIP: illwq WNq V: t"'t'n 1400 VIWO cowilo WoiNNNN NNN NNN NN	NNNN sNNN NNN NNwwoE Ei	 NA410	 . 64ocqN.-4NIri	et	 W N t-t--	 V AC N t-C--,: ol	 c-r-	 mr- m.-,	 1	 4ctc-i c4gi c7W4 gAg c°0!refc	 g	 4cpcpRt24 E147,4:g :71: 1 g e64vm	 4444 444 44:4 42LA 2. .6).5. I.E 4 u	 N12i	 0 r-iIP	 LoOS	V.oi-	w6N;I Ali'	V-r.4.m 0E8	M	 v,..	 oi 	 e	 N	 N$.,00	 M	 .0	 10O.m 	 	 tc,r	 tc,..,-0,..r.	 04m.	 ,r.. v 	 m mm E	,a	, .0m6	 v 	 'v.'''.ooNty	 co4	 m	 Lo.ko	 qgv!.'4'	 mr,11E6.101	I	co'	 Ice laNc,i'	 r=m,1c	&	 vNt.	 v..,,,	 c`l	 . tn.N t.,	 00 ,...rt--C X :E	 ,..1.	 Iti.-i	 .-ile	N.	 VIM	 ClV.OW,	 N	 MM MM. 1.40'	 t- '713 5=	 6N	 gg 6g;0N.V	V:M q4as	,0,17m.NEI gl 	 e.	 -1,gR Rg	 NANNN.4. C-^ 	e6C=4	 Or:	 t=,-;	 .4t---	 lep.N	 N.N	 N01.4;0c.	 V ,-44	 .-4N	 N.-4	 .NZ.0	co- co- to 	to- 6	 0- te,- cz	IOU	 NAM NSIMo..J. 	v.v.	 .c.),. N n41,,,Ci.4,-C	 C-0 ''	 Lo m ^ m w -	 ,v; .4." oi 	 N 4 oi el ..or 6Nx•?.°	-4.-1g	 -...-tV,	 -.-tgl	 --...m	 .... el.	.. ,,,E p rft-.cs0 C6t-.C6 !76.,-06 WM	°Oik6M °1111') °Kw- mmr4.,-Icvo "4."-IN "4.,-Im Lom8.t,	 ..tw.v. aw. csttc. ,,...- 	 ch.a-;. ch.a.7 cka,-; .etg	'-4 U)	 .1. 16. 4 ...tkri. 4 v.-4	 m,-4,4,-, co,-1,-1 co..4,-4 .-4N-6	 -61:3(.... co , 	 1	 cz 6 	 013	 0	 075C b.s o	c..	 M	 4/_,ri-EcD4J 44 4.-CCC	 4C'	 .4..E	4a'	4 c=,	 ...	1 		 c	.,,c	 44.gmv	 4I2 8	 -I	 t	 2 0	22 8 2	 2	 211R11 lv 1*p	 .11	 1111 11 11	 11, 0	18E; 187 118 -O It	 Ig8I ItT, tt7 It5;11	'PYY'R 30-h VY-b 'RY	 VYYT4 10b VY-b 0.4 	 44	 85g8 855 85E 85	 1	 8558 850 850 855.2.-	 '1.ccv vco ucc, .c	vcc. ucv vCcA c..0.6 -ft	 .2E014,-.az	 4	 reTViq	 m 1	 -o	_	 -0T	 Tc!b,,, -t T	g 	 1	 t	 24).,,	 4	 i	 4t	 taz -.8 	 u	 t	 w	 cl.)o	 o	 o	 o	 o	 oc6F3a	 ..c	 .1:1	 ..o	 2 x	 az	 x	 24).13H	0 .	 N	 M	 V	 0 50	 N	 M	 VE-1 	0	 0	 0	 0	 0	 1	 0	 C.)	 0b pasteurizedIII coexistent I A non-coexistent control tv controlunpasteurizedabacteria soil	 seed-soilbacteria-seed soli	 bacteria-seednon-coexistent	 El control	 r-Na controlF AINN coexistentshoot dry weight (mg)50shoot dry weight (mg)bacteria-seed soil	 bacteria-seed seed-sonFig. 9. Shoot dry weight of treatments comprised of coexistent and non-coexistent factorcombinations and control treatments in (a) unpasteurized and (b) pasteurized soil inExperiment 2. Blank control bar represents non-coexistent control for bacteria-seed-soil contrasts. Slashed control bar represents coexistent control for bacterin-seod-soilcontrasts. Within n set of contrasts, means having no letter or the same letter do notdiffer significantly at p < 0.10.474 85.0 DISCUSSION5.1	 GerminationWhen applied to seed in pure culture, some bacterial strains have promoted the germination ofseveral important agricultural species (Hussain and Vancura 1970; Holl et al. 1988; Kloepper et al.1988; Chanway et al. 1989; Chanway and Nelson 1990). Chanway et al. (1991a) recentlydemonstrated that the germination rate of white spruce [Picea glauca (Moench) Voss.], and thegermination capacity of lodgepole pine can be increased in response to inoculation with Bacilluspolymyxa in a controlled environment. Germination capacity of white spruce was also stimulated inan outdoor nursery experiment with the same bacterial inoculum (O'Neill et al. 1991).Most of the bacterial strains from both locations inhibited germination when inoculated onto theseed (Fig. 1). Although these effects were generally not statistically significant, the magnitude waswas substantial in several cases. Inhibition of germination of maize following inoculation withbacterial supernatant was reported by Hussain and Vancura (1970) for one of the strains theytested. However, germination was subsequently promoted by a lower concentration of the samesupernatant. This observation and the large quantity of several phytohormones in the supernatantprovide anecdotal evidence for the involvement of phytohormones in this process. By examiningphytohormone production of the strains tested in the present experiment and the effect ofinoculation with supernatant dilutions of these strains on germination, the influence of bacterialphytohormones on spruce germination could be assessed.Numerous factors (mainly abiotic) regulate germination to ensure that the environment in whichgermination occurs is optimal (Kramer and Kozlowski 1979). The inhibition of germination observedhere suggests that rhizosphere bacteria may be included within the repertoire of methods whichplants use to maximize their fitness.4 9Despite the dirth of statistically significant positive germination effects, inoculation with several ofthe Mackenzie and Salmon Arm strains substantially altered both the germination capacity and thegermination value, particularly in the Mackenzie seedlot. In Experiment 1, GC of Mackenzie seedwas stimulated 14% compared with uninoculated controls after inoculation with strains M4 andM13, and GV increased by 23%, also in response to strain M13 (Fig. la). In Experiment 2, strainS20 increased the GC and GV of Mackenzie seed an average of 5.8% and 18%, respectively, overuninoculated controls in the four soil x pasteurization combinations (Figs. 5c and 5d).The mechanisms by which such promotive effects occur have yet to be elucidated, but germinationenhancement by soil bacteria has been postulated to involve solubilization of organic phosphate (Hollet al. 1988) and production of plant growth regulators (Hussain and Vancura 1970). The latterauthors observed enhanced germination capacity of maize (Zea mays L.) by up to 40% afterinoculation with either growth regulator-producing bacteria or their supernatants. Kramer andKozlowski (1979) suggested that the ratio of growth inhibitors (mainly abscisic acid) to growthpromoters (gibberellins and cytokinins) in seeds is largely responsible for regulating seed dormancy.Alternatively, Elad et al. (1987) related bacterial promotion of tobacco, radish and cucumber seedgermination to the ability of inoculum bacteria to suppress Fusarium, a common fungal pathogen ofagricultural and conifer species. However, some fungi have been found to enhance germination,possibly through inactivation of germination inhibitors within the seed coat (Jones and Waid 1963).Storage temperature, seed moisture content and seed age are also known to influence germination(Kramer and Kozlowski 1979). Mackenzie and Salmon Arm seedlots were stored at differentlocations and were different ages, therefore, it was not surprising that significant differences weredetected in the GC and the GV between seedlots (Fig. 5b and Appendix 2). In addition, variation ingermination between populations of interior spruce (Khalil 1986) and differences in the percent offilled seed of the two seedlots may also have contributed to the observed differences between the twoprovenances.5 0However, causes of the observed differences in germination parameters associated with soil sourceand soil pasteurization are not so obvious (Fig. 5a and Appendix 2). Allelochemicals, such as organiccyanides, terpenes or phenolic acids, which are leached from living and dead leaves, or are exudedfrom plant roots of many species can inhibit germination and growth (Rice 1974). Their presenceand activity may have differentially influenced germination in the two soils because different plantspecies were found associated with each of the soil collection sites.The significant enhancement of germination capacity after soil pasteurization suggests thatmicrobial inhibitors of germination may exist within these soils. This concurs with the work of Jonesand Waid (1963), in which greater germination capacity and germination rate were observed insterile vermiculite than in unsterile soil. The preponderance of germination-inhibiting overgermination-promoting bacteria in Experiment 1 implies that rhizobacteria are at least partiallyresponsible for this phenomenon.The utility of a bacterial germination promoter in a conifer nursery would require that the strain beable to elicit consistently a positive response under a variety of environmental conditions. Despitethe relatively similar environmental conditions of Experiments 1 and 2, the reproducibility of theeffects of individual strains on germination between experiments was poor: of the six strains testedin both experiments, the strongest promoter of GC in Experiment 1 (M14 - 9% greater than control)was the second strongest GC inhibitor in Experiment 2 (-5%). Similarly, the second strongestpromoter in Experiment 1 (2%) was the strongest GC inhibitor in Experiment 2 (M19 - -15%). Itappears therefore, that the factors influencing the expression of germination promotion byrhizobacterial inoculation must be more thoroughly understood before this technology can besuccessfully implemented in conifer nurseries.Analysis of the effects of phosphorus fertilization on germination, seedling performance assaysrelated to organic phosphate metabolism and growth regulator production of Mackenzie and SalmonArm strains, and the use of bacterial mutants that are deficient in phosphatase or hormone51production could help to elucidate the mechanisms by which germination is enhanced by theserhizobacteria. Inactivation of germination inhibitors by rhizobacteria, as has been reported forfungi, should also be explored.5.1.1 Coexistence SpecificitiesThe GC of coexistent bacteria-soil treatments was significantly lower (6%) than that of non-coexistent treatments (Table 6 C3a). While the biological significance of this difference isquestionable, it may suggest that the expression of germination regulation by rhizobacteria mayhave evolved through a period of coexistence between the rhizobacteria and soil.In general, the presence of coexistent factor combinations (seed provenance, soil source and bacteria)did not substantially affect seedling germination (Table 6). However, the presence of coexistentcombinations of seed and pasteurized soil resulted in a substantial inhibition of GC and GV relativeto non-coexistent combinations (Table 6 C4b). The same specificity was not observed inunpasteurized soil (Table 6 C4a). The influence of pasteurization on seed germination is difficult toexplain and effects may have resulted from altered soil nutrient availability (e.g. the nearly three-fold increase in manganese in Salmon Arm soil), synthesis of inhibitory compounds, or simply fromthe removal of beneficial micro-organisms that buffer seed from the abiotic environment.The lack of a stimulatory effect on germination when coexistent factor combinations were testedagainst uninoculated controls and treatments comprised of non-coexistent factor combinations doesnot mean that germination-promoting bacteria are non-specific with regard to seed and soil (i.e. thatgermination promoters are effective on a range of seed genotypes). On the contrary, a significantbacteria x seed interaction for GC was detected: strain S10 promoted the GV of Mackenzie seed 15%,but inhibited that of Salmon Arm seed 12%, while strain M19 had the opposite effect (Mackenzieseed was inhibited 8%, but Salmon Arm seed was promoted 4%) (Fig. 5c). This suggests thatbacteria-seed specificity may influence germination, but that such specificity is unrelated tocoexistence.52Nonetheless, bacteria-seed, bacteria-soil or seed-soil specificities resulting from coexistence mayindeed exist, but may have been undetected in the present experiments because of the absence of thenecessary genetic differences between the rhizobacteria, seed or soil biota of the two sites. In otherwords, we may have wrongly assumed that the geographically disparate sites which were utilizedwould provide rhizobacteria, seed or soil biota possessing the appropriate qualitative or quantitativegenetic differences required to detect coexistence specificities.Similarly, coexistence specificities may have been missed if their expression is dependent upon someenvironmental factor which differed between experimental and natural conditions. If growthregulating substances produced by soil bacteria in nature are assumed to cause enhancedgermination, production of these substances may be altered substantially under the experimentalconditions that were used. This, in turn, may have affected seedling emergence in a way that doesnot reflect the natural course of events.5.2	 Seedling Growth5.2.1 Effects of Bacterial InoculationBacterial inoculation had a substantial effect on seedling growth in some treatments. Shoot dryweight increases of up to 27% in Experiment 1 and 53% in Experiment 2, as well as equivalent orlarger increases in root surface area, root dry weight and total dry weight relative to uninoculatedcontrols were observed. These findings represent the first report of bacterial growth promotion ofspruce and indicate that bacteria capable of stimulating early spruce seedling growth, in controlledconditions at least, can be secured by isolating rhizobacteria from naturally-regenerating coniferseedlings.However, bacterial inoculation also resulted in the inhibition of shoot dry weight accumulation(maximum inhibition was 28% relative to uninoculated controls) and the effect of a particular strainvaried greatly between seed x soil x pasteurization treatments. The successful acquisition of PGPRin these experiments and the variation between treatments may be related to several factors,5 3including: (1) host specificity; (2) coexistence specificity; (3) the use of endorhizosphere bacteria; (4)the use of bacteria isolated from young plants, and (5) the natural abundance of selected bacteria inthe rhizosphere. Host SpecificityVariability of host plant growth responses to inoculation with growth promoting rhizosphere bacteriais not uncommon and its cause remains elusive (Kloepper et al. 1989). Although not always the case(Kloepper et al. 1988; Bashan et al. 1989), plant cultivar or genotype specificity (i.e. differentialgrowth responses of differing cultivars or genotypes due to inoculation with a particular microbe)may comprise a substantial component of the variation in plant growth responses to PGPR (Burrand Caesar 1984; Chanway et al. 1988b,1989). Differential growth responses of the Mackenzie andSalmon Arm seed provenances to the putative PGPR (as indicated by the significant seed x bacteriainteraction (Appendix 3)) demonstrate the occurrence of provenance specificity in these spruce-rhizobacteria associations.Host specificity may also exist at the species level (i.e. species specificity) (Baldani and Dobereiner1980; Gardner et al. 1984). Azospirillum brasilense is most frequently isolated from the roots ofwheat, while Azospirillum lipoferum is most frequently isolated from the roots of sorghum andmaize. Both bacterial species are often cited as PGPR of the plant species from which they wereisolated, but rarely do they promote the growth of other species (Sumner 1990).Inoculation of Douglas-fir, lodgepole pine and interior spruce with Bacillus strains isolated from, andcapable of promoting the growth of, perennial ryegrass and white clover, resulted in little or nogrowth promotion of spruce and Douglas-fir (Chanway et al. 1991a; O'Neill et al. 1991), while thestimulatory effects of inoculation on lodgepole pine were short-lasting unless the seedlings were re-inoculated (Chanway et al. 1991a). Therefore, the successful acquisition of interior spruce PGPR inthe present experiments may have been related to the use of bacterial inoculants isolated from thetarget species, interior spruce.545.2.1.2 Coexistence SpecificitiesIn contrast to germination capacity and value, the presence of coexistent bacteria, seed and soilfactor combinations resulted in seedlings with significantly greater SDW than seedlings in non-coexistent factor combinations (Fig. 9 and Table 8 C l a). However, analysis of the three 2-factorcombinations (bacteria-seed, bacteria-soil, seed-soil) comprising the 3-factor combination indicatedthat the seed-soil-bacteria coexistence specificity was almost entirely due to the use of coexistentseed and soil (Table 8 C3 a ,C4a ,C5a).The SDW of seedlings grown in coexistent soil was significantly larger (27%) than that of seedlingsgrown in non-coexistent soil. However, the difference in SDW between coexistent and non-coexistenttreatments decreased to 17% when the soils were pasteurized, suggesting that both biotic and abioticelements may have been involved in the manifestation of coexistence specificity.This trend was also reflected when seedling height, root dry weight and seedling total dry weightwere analyzed (Table 8 C4a and C4b). For example, the root dry weight of seedlings associated withcoexistent seed and soil factor combinations was 35% greater than those in non-coexistentcombinations, and this advantage decreased to 23% after pasteurization.In addition to bacteria and mycorrhizae, other microflora (fungi, actinomycetes and algae) andmicrofauna (protozoa, nematodes, mites and insects) inhabit the rhizosphere, and can influenceplant growth (Curl and Truelove 1986). In theory, these could also affect coexistence specificities.However, the less specific nutritional demands and the greater ability of the microfauna to adapt todiffering environments are likely to make them less dependent than the microflora on specific hosts.Hence, their involvement in seed-soil coexistence specificity is less probable than that of the moreabundant microflora.Abiotic soil factors which could potentially facilitate the expression of coexistence specificity betweenseed and soil include soil physical properties, the inorganic micro- or macro-nutrients, and organic5 5compounds, such as hormones, vitamins, amino acids or enzymes. In these experiments notabledifferences in the levels of many of the inorganic elements between the two soils were detected (Fig.1) and may have contributed to the abiotic component of the observed seed-soil coexistencespecificity.Although most of the seed-soil coexistence specificity effect appears to have been due to abioticfactors (inferred from the decrease in the relative shoot dry weight difference between coexistent andnon-coexistent seed-soil combinations from 27% to 17% upon soil pasteurization), biotic elements areoften directly responsible for the quality and quantity of many abiotic factors, such as organic andinorganic compounds, and soil physical properties, particularly in the rhizosphere. For example,siderophores produced by bacteria are known to influence available ferric iron levels in therhizosphere (Powell et al. 1982) and many species of rhizobacteria secrete various phytohormones(Whightman et al. 1980). Therefore, the overall influence of biotic factors in seed-soil coexistencespecificity, and consequently, in the potential benefit to plant growth through manipulation of therhizosphere with coexistent soil micro-organisms, may be under-estimated from the degree withwhich soil pasteurization affected this specificity.It is tempting to ascribe the decrease in seedling growth associated with seed-soil coexistencespecificity following soil pasteurization solely to the removal of the indigenous microflora, butalterations in soil nutrients and other soil properties following soil pasteurization may also havecontributed. Mulder (1979) notes that soil pasteurization can result in manganese and ammoniumtoxicity in some agricultural species. Consequently, if these nutrients or other soil properties whichwere altered by soil pasteurization were responsible for the manifestation of the abiotic component ofseed-soil specificity, then they too may have contributed to its reduction following pasteurization.No differences were observed in the physical characteristics of the soil, and most of the soil nutrientlevels were altered only slightly following soil pasteurization. However, the level of availablemanganese almost tripled after pasteurization of Salmon Arm soil (Table 1), and seedlings grown inpasteurized Salmon Arm soil had potentially toxic levels of manganese in their tissue (data not56shown) (Brady 1974 p. 486). Therefore, soil pasteurization effects may be confounded withmanganese toxicity and these should be interpreted with caution.By contributing to competitive ability, enhanced plant growth rates can increase the likelihood ofdominance and reproductive success within a plant population, and thereby influence the structureof plant communities (Sarukhan et al. 1984). The significant seedling growth advantage associatedwith coexistent seed-soil combinations compared with non-coexistent combinations suggests thatseed-soil adaptation could be an important determinant of plant population distribution byincreasing community resistance to invasion by non-coexistent genotypes within a species.Chanway et al. (1988a,b) have shown that a history of coexistence between plants and microbes caninfluence the plant response to bacterial inoculation. However, bacteria-seed coexistence specificitynot was detected in the present experiments. Coexistence specificity between these bacterial strainsand the spruce provenances may not have been detected due to the heterozygous nature of interiorspruce which results from its tendency (and the tendency of most forest tree species) to outcross(Zobel and Talbert 1984 p. 52). Only when vegetative clones have been used has microbe-plantcoexistence specificity been observed (Chanway et al. 1988a,1990). Therefore, clonalexperimentation should be conducted with spruce to confirm these results.Notwithstanding the lack of bacteria-seed coexistence specificity, significant adaptive relationshipswere observed between spruce provenances and soil sources. The manifestation of adaptiverelationships in the complex environment of soil (as opposed to in less complex non-soil media)attests to the influence these adaptive relationships may have on plant growth. Buffering thisobservation, however, is the fact that the seed x soil interaction accounted for only 2% of the totalexperimental variation (Appendix 3); clearly, factors other than seed-soil specificity had considerableinfluence on plant growth.57In support of this argument is the general observation from provenance tests of many forest treespecies that local provenances are often not among the best performing provenances at a given site.This would imply that seed adaptation to climate is much stronger than adaptation to soil. By re-testing the effect of seed-soil coexistence on plant growth using a larger number of seed/soil sourcesthe strength of the inference of this relationship to other seed and soil sources would be increased.Additionally, by growing each provenance in each of the soils and at each location, the contributionof climatic and soil effects to seed-soil specificity could be partitioned.The contribution of biotic elements to seed-soil coexistence specificity could also have significantsilvicultural implications. The benefit of inoculating seedlings at planting with soil possessing the'proper' microbes was demonstrated by Amaranthus and Perry (1987). They increased Douglas-firseedling survival by 50% on an unsuccessfully reforested clear-cut by inoculating seedlings atplanting with a small amount of soil from a young Douglas-fir plantation. Incorporation of theconcept of seed-soil coexistence specificity into silvicultural practice could result in a simple methodof improving the efficacy of soil inoculants.Bacteria-soil coexistence also had a negligible effect on SDW relative to uninoculated controlseedlings and to seedlings associated with non-coexistent factor combinations (Fig. 9 and Table 8C3a). These results may indicate that plant-microbe and soil-microbe coexistence specificity is notimportant in these ecosystems, and that no advantage would accrue through the use of 'adapted'PGPR for interior spruce.Alternatively, coexistence specificity between these factors may exist, but may not have beendetected due to the over-riding influence of soil factors or the heterozygosity of spruce seed(discussed above). The failure to impose realistic environmental conditions on the experimentalsystem (Harley and Smith 1983), or the alteration of soil microbial populations as a result of storageof the soil for six months may also have obscured the detection of these specificities.5 Age of the Host PlantGrowth promotion of spruce seedlings in the present study may in part also have been related to theuse of young seedlings for the isolation of bacterial inoculants. The variation in the quality andquantity of root exudates between plant species and genotypes described above also exists betweenplants of differing ages and stages of development (Rovira 1959; Alexander 1977 p. 428). Vancuraand Hanzlikova (1972) found that general exudate components (sugars, amino acids, organic acids,lipids, enzymes, etc.), and the specific compounds comprising these broad groupings, varied betweenseed and seedling exudates. Similar differences between seedlings and mature plants have also beenobserved (Vancura and Hovadik 1965). For example, Smith (1970) reported that carbohydrates inthe root exudate of 3-week-old sugar maple (Acer saccharum March.) seedlings were more diverseand abundant than those of 55-year-old trees.Given the dependence of rhizosphere microbes on root exudates for their supply of organic nutrients(Rovira 1969) and the variation in nutritional requirements between bacterial species (Lochhead andChase 1943; Boyd 1984), it is not surprising that the kinds and numbers of rhizosphere organismschange with plant growth and development (Riviere 1960; Parkinson et al. 1963; Burr and Caesar1984). Consequently, bacterial inoculants isolated from, and adapted to, young seedlings may showincreased survival, proliferation, and efficacy when used as PGPR.This hypothesis is supported by the soil inoculation experiments of Amaranthus and Perry (1987).In their work, Douglas-fir seedlings which were inoculated at planting with soil from a young coniferplantation exhibited dramatic increases in survival (50%) and basal area (200%) compared withcontrols, while survival decreased slightly when inoculated with soil from a mature forest.Therefore, growth promotion of spruce seedlings in the present study may have been related to theuse of bacterial inoculants that were isolated from seedlings less than 5 years of age.5 Endorhizosphere BacteriaEndorhizobacteria accounted for five of six of the most effective Salmon Arm SDW promoters inExperiment 1 (Table 2b and Fig. 2b), and for the two SDW promoters which were most consistent inExperiment 2 (Fig. 6). This sugests that bacteria originating within the root may be more likely topromote plant growth than bacteria originating on the root, and would appear to contradict thegeneral assumption that the lack of plant growth promotion by PGPR can be attributed to the failureof inoculum to thrive in the ectorhizosphere (Gaskins et al. 1984). However, Reddy and Rahe (1989)recently reported that growth promotion of onion by Bacillus subtilus was not correlated withinoculum survival in the ectorhizosphere. They observed the greatest stimulation of shoot and rootdry weight due inoculation during the final week of their study when the population of the markedinoculant was lowest (95 cfu's/plant) and suggested that growth promotion may have resulted fromPGPR-related manipulation of the indigenous rhizosphere microflora (i.e. DRB and/or pathogenswere suppressed). Holl and Chanway (unpublished data) were also unable to correlate growthstimulation of pine after inoculation with Bacillus polymyxa and rhizosphere colonization by theinoculant when measured at the same time, but colonization four weeks after inoculation wascorrelated with the seedling growth response eight weeks after inoculation.Alternatively, some species of bacteria may promote plant growth from within the endorhizosphere;such a strategy may help to explain the apparent incongruencies regarding plant growth promotionand PGPR colonization of the ectorhizosphere. For example, members of the genus Azospirillum arealso frequently cited as being PGPR (Baldani and Dobereiner 1980; Patriquin et al. 1983; Umalia-Garcia et al. 1980), and may be more abundant within cereal roots than in the ectorhizosphere(Baldani and Dobereiner 1980). In addition to Azospirilla, plant growth promoting Bacilli have alsobeen found within the root (Larson and Neal 1978), and their populations have been demonstrated toexceed those in the ectorhizosphere (Lalande et al. 1989).Electron microscopy has revealed that endorhizosphere colonization occurs between live root corticalcells and within dead cortical cells (Umalia-Garcia et al. 1980; Bashan and Levanony 1988). Due to6 0root respiration, oxygen tension within root tissues is lower than ambient (personal communication,Dr. H. Weger, University of Regina). This could facilitate the growth of bacteria which are micro-aerophilic (i.e. Arthrobacter and Azospirillum) or facultatively anaerobic (i.e. Bacillus, Klebsiella,Serratia and Staphlococcus) and reduce the number of aerobic bacteria within the root.Endorhizosphere localization of bacteria may also facilitate an efficient exchange of materialsbetween the two organisms (Sumner 1990).In addition, it has been suggested that colonization of the endorhizosphere would be conducive tobacterial nitrogen fixation because nitrogenase, the enzyme responsible for nitrogen fixation, isoxygen-labile (Sumner 1990). For example, Pohlman and McColl (1982) found that inoculation ofbarley with an unidentified nitrogen-fixing bacterium isolated from the rhizoplane of the same cropenhanced nitrogenase activity of excised roots 10-fold. This activity persisted despite washing andsterilizing the roots, suggesting that the diazotroph had colonized the root interior.Yield increases of cereal crops due to inoculation with rhizobacteria isolated from surface-sterilizedroots have been observed by several authors (Lalande et al. 1989; Lethbridge and Davidson 1983;O'Hara et al. 1981). Baldani et al. (1983) provided convincing evidence for the involvement ofendorhizosphere colonization by PGPR using growth promoting Azospirillum strains isolated fromsurface sterilized roots of wheat. The number of Azospirillum cells detected within wheat roots wasstrongly correlated (r = 0.92) with total nitrogen accumulation, but no relationship was observedbetween the number of ectorhizosphere Azospirillum cells and total nitrogen accumulation.Staphlococcus hominis was the most effective and consistent growth promoter in Experiment 2, andhas not previously been reported as a PGPR. Its absence from the literature may be due to theinfrequency with which endorhizosphere and conifer PGPR are characterized. Staphloccoci areclosely related to Bacillus (Kloos and Jorgensen 1985) and are found occasionally in the soil(Alexander 1977 p. 26). The anaerobic nature of some Staphloccoci strains may facilitate theircolonization in the endorhizosphere.6 1Bacillus polymyxa, the most effective Salmon Arm SDW promoter in Experiment 1, was isolatedfrom the ectorhizosphere, although endorhizosphere-inhabiting Bacilli have also been detected(Larson and Neal 1978). In a rare enumeration study of endo- and ectorhizosphere bacteria, 88% ofthe strains within the endorhizosphere of maize were identified as Bacilli, some of whichsignificantly promoted maize growth (Lalande et al. 1989). Rennie and Larson (1979) also obtainedsignificant growth promotion of wheat with the use of a Bacillus isolated from the endorhizosphereof wheat.Members of the genus Bacillus, particularly Bacillus polymyxa, have been shown to promote avariety of yield parameters of several crop and forest species, including white clover, crestedwheatgrass (Holl et al. 1988), maize (Lalande et al. 1989), sorghum (Broadbent et al. 1977), potato(Burr et al. 1978), onion (Allium fistulosum L.) (Reddy and Rahe 1989), spring wheat (Chanway et al.1988b), lodgepole pine and Douglas-fir (Chanway et al. 1991a). While no attempt was made in theseexperiments to specifically isolate endorhizobacteria, cutting the roots into segments, sometimes asshort as 5 mm, may have resulted in their exudation from the root interior into the root wash mediaduring bacterial isolation.Of the six most effective seedling growth promoting strains selected from Experiment 1, two wereidentified as Pseudomonas putida and two were Hydrogenophaga pseudoflava, formerlyPseudomonas pseudoflava (Willems et al. 1989) (Tables 2a and 2b). The ability of pseudomonads toenhance plant growth has been widely attributed to their effectiveness in ectorhizospherecolonization (Burr and Caesar 1984). However, they also appear to be aggressive endorhizospherecolonizers, as members of this genus were the second most abundant within the maize rootsexamined by Lalande et al. (1989), and Lynch (1980) encountered pseudomonads in the intercellularspaces of maize root cortex. Furthermore, the most abundant group of bacteria isolated by Gardneret al. (1982) from the xylem of rough lemon roots were pseudomonads, including Pseudomonasputida. Numerous other genera, including Bacillus, were also isolated from the xylem. Theprevalence of these two genera in the endorhizosphere, and their incidence as PGPR in this and62other studies, suggests that the ability to colonize the endorhizosphere may also be related to theability to stimulate plant growth. Abundance of the Inoculant on the Host PlantCertain rhizobacteria, particularly those that contribute to the fitness of their host, may have aselective advantage in the rhizosphere over those which do not benefit the plant, and with time, maycomprise a significant component of the rhizobacterial population (Chanway et al. 1991b). If plantgrowth promoting ability of bacteria is related to rhizosphere colonization ability (Suslow andSchroth 1982; Bashan 1986), and if the most abundant strains of bacteria in the rhizosphere are themost successful colonizers, then selection of the most abundant rhizobacteria may have assisted inthe acquisition of growth promoting strains in the present experiments.This proposition, however, has not been critically assessed (Kloepper et al. 1989) and supportingevidence is lacking (Chanway et al. 1991b). Reddy and Rahe (1989) were unable to relate growthpromotion of onion in the field by a Bacillus PGPR to survival of the inoculum in the rhizosphere,and recent work with maize PGPR confirmed that the rhizosphere population of growth promotingAzospirillum is small relative to the total maize rhizobacterial population (Mubyana 1990). Inaddition, Lalande et al. (1989) obtained the best growth promotion of maize with Serratialiquefaciens, a species which represented only 2% of the ectorhizobacteria which were isolated.Correlation of the abilities of bacteria to colonize roots and to promote plant growth may thereforedepend on the mechanism by which plant growth is stimulated, and hence, on the particularbacterial strain in question.5.2.2 Effects of Seed Provenance, Soil Source and Soil PasteurizationAlthough the primary purpose of these investigations was to explore the effects of bacterialinoculants and coexistent factor combinations on seedling growth, some inferences regarding theeffects of seed provenance, soil source and soil pasteurization on seedling growth can be made fromthe performance of uninoculated control seedlings (Fig. 7). First, that seedlings from Salmon Arm6 3attained a SDW 5% greater than Mackenzie seedlings is not surprising. Much of the geneticvariation within forest tree species for growth traits resides between provenances (Zobel and Talbert1984 p. 62), and in white spruce, provenances, as well as trees within provenances, are significantsources of genetic variation (Khalil 1986).Second, seedlings in Salmon Arm soil generally grew much better than seedlings in Mackenzie soil(Fig. 7) regardless of pasteurization or seed provenance. This was also predictable because theSalmon Arm soil had a greater cation exchange capacity, organic matter concentration, totalnitrogen and available nutrient content (Fig. 1). The only exception to this generalization wascalcium, but Salmon Arm seedlings did not indicate calcium deficiency (data not shown) according toBallard and Carter 1985. Furthermore, seedlings in Salmon Arm soil did not display twisted anddeformed leaves, nor dead or dying meristematic tissues characteristic of calcium deficiency(Salisbury and Ross 1985).Finally, and most interestingly, pasteurization of the Mackenzie soil resulted in a dramatic increasein the SDW of both Mackenzie (57%) and Salmon Arm (49%) seedlings, while pasteurization ofSalmon Arm soil inhibited the SDW of seedlings from Mackenzie (3%) and Salmon Arm (25%). Thiswould suggest that the Mackenzie soil harboured a larger number of deleterious soil micro-organisms than the Salmon Arm soil, or that the effects of the deleterious micro-organisms in theMackenzie soil were more profound than those of the beneficial ones. Although effects of soilpasteurization on seedling growth are usually attributed to changes in populations of soil microbes,changes in the physical properties and the observed changes in the nutrient concentrations of thesoils could also have affected seedling growth. In particular, tripling of the manganese concentrationin the Salmon Arm soil following pasteurization (Table 1), and potentially toxic levels of manganese(Ballard and Carter 1985) in Salmon Arm seedlings grown in Salmon Arm soil, could have beenresponsible for the growth inhibition of these seedlings. Wider soil sampling (e.g. more than onesample/site) and assessment of soil bacteria populations in pasteurized and unpasteurized soils couldhave shed more light on this observation.6 45.2.3 Mechanisms of ActionAlthough the mechanism(s) of action of the inoculants was not investigated in these experiments,some speculation can be made on the basis of the bacterial species involved and the experimentaldesign utilized. Strains M14 and S1 were identified as Pseudomonas putida, and strains M18 andM19 as Hydrogenophaga pseudoflava (previously Pseudomonas pseudoflava) (Tables 2a and 2b).Pseudomonad PGPR activity is often related to their ability to inhibit deleterious rhizobacteriathrough the production of Fe 3+-chelating siderophores and other antibiotics (Powell et al. 1982;Kloepper et al. 1980).If this mechanism had operated in these experiments, then growth promotion should have beenobserved only in the unpasteurized soil treatments, where DRB would have reduced the growth ofcontrol seedlings. In pasteurized soil where DRB would have been absent, PGPR inoculation shouldhave had no effect because the growth of control seedlings would not have been reduced. However,in only two of the 16 seed x soil x inoculant treatments involving these four strains was growthpromotion greater in unpasteurized than in pasteurized treatments (excluding two treatments inwhich promotion was negligible) (Fig. 6). Greater growth promotion in unpasteurized versuspasteurized soil was also rare in treatments which received the Bacillus polymyxa or theStaphlococcus hominis inocula and suggests that antibiosis did not contribute to the observed growthpromotion by any of the strains. However, microbes were probably re-introduced into pasteurizedmedia from the atmosphere and tap water. Therefore, this interpretation should be made withcaution.Bacterial solubilization of phosphate was not examined in these experiments, although foliarnutrient concentrations of the 56 seed x soil x pasteurization x bacteria treatments showed littleevidence of enhanced phosphorus uptake among inoculated seedlings relative to uninoculatedcontrols (data not shown). The largest increase in foliar phosphorous content was due to bacterialinoculation with strain M14, but was only 5.6% greater than controls.6 5The SDW of seedlings inoculated with N-fixing and non-N-fixing strains was contrasted and resultsindicated that the diazotrophic inoculants as a group were not more effective in promoting shoot dryweight gain than non-diazotrophs (data not shown). When examined individually, the six SalmonArm diazotrophs in Experiment 1 ranked among the poorest growth promoters. However, in thenitrogen depauperite Mackenzie soil, the three Mackenzie diazotrophs tested in Experiment 1ranked as the second and third most effective growth promoters. Low levels of available nitrogenwere cited by Brown (1982) as a requirement for asymbiotic bacterial nitrogen fixation to contributesignificantly to the nitrogen capital of ecosystems. Consequently, the poor nitrogen status of theMackenzie soil may have facilitated nitrogen fixation by these two strains. However, thecontribution of N-fixation by PGPR to plant growth is questionable, and most recent estimates of thequantity of nitrogen fixed in the rhizosphere by PGPR are too low to account for the observed plantgrowth promotion (Kapulnik et al. 1985; Sumner 1990). The failure of nitrogen fixation byassociative rhizobacteria to contribute significantly to plant growth was also demonstrated withconifers. Chanway and Holl (1991) determined that asymbiotic nitrogen fixation by a diazotrophicBacillus PGPR inoculated onto lodgepole pine contributed only 4% of the seedling's foliar nitrogen.The significance of bacterial phytohormone production to the observed growth promotion can not bedetermined from these experiments.5.2.4 Reproducibility of EffectsAs has been previously documented (Kloepper et al. 1989), substantial growth response variabilitybetween experiments was detected in this work. All putative PGPR selected from Experiment 1,with the exception of strains M18 and Si, inhibited seedling growth in Experiment 2 when similarseed and soil sources were used (Fig. 8). Strains M18 and S1 promoted SDW in both experiments,although promotion in Experiment 2 was less than 5%. Storage of the soil for six months betweenexperiments may have altered the microbial population of the soil, and thereby modified the plantresponse to bacterial inoculation (in addition to affecting bacteria-seed and bacteria-soil coexistencespecificities discussed above).6 6Adsorption of different strains of Azospirillum brazilense to several crops was strongly affected bythe growth phase of the inoculum (Kapulnik et al. 1985). Consequently, inadequate detail to theprecise growth phase of the strains in the two experiments may also have contributed to thedifferent responses to inocula between experiments. Finally, environmental conditions, soil nutrientstatus and inoculum density can influence the success of bacterial inoculation (Sumner 1990).Therefore, minor differences in inoculation and fertilization, and dissimilar environmentalconditions in the greenhouse used in Experiment 1 and in the growth chamber used in Experiment2, may also have imparted different inoculation responses in the two experiments.676.0 CONCLUSIONSInterior spruce PGPR were isolated in these experiments using a relatively cheap and simple' natural plant enrichment technique'. Attempts to secure PGPR by selecting rhizobacteriapossessing in vitro attributes presumed to be beneficial to plant growth are expensive and time-consuming, and may not be more effective than the methods used in the present experiments (e.g.Kloepper et al 1988).The isolation of a plant growth promoting strain of Staphlococcus hominis is the first report of aPGPR within this genus. The infrequency with which members of this genus are found in the soilmay support the use of a 'natural plant enrichment technique', and may signify the importance of'minor' bacterial species in bacterial plant growth promotion. Additionally, the disproportionatesuccess of endorhizosphere bacteria over ectorhizosphere bacteria in promoting plant growth in theseexperiments suggests that more attention be paid to endorhizosphere-colonizing PGPR.Evidence for coexistence specificity involving bacteria and seed or bacteria and soil was not detected.Specific responses unrelated to coexistence were observed between these factors. Consequently,methods of isolating PGPR based coexistence between seed or soil and the bacteria may not provebeneficial.Evolved specificities between organisms have been detected in the relatively simple environment ofnon-soil media by others. However, the detection of significant coexistent specificity between seedand soil amidst the numerous interactions of biotic and abiotic soil elements, attests to theimportance this phenomenon may have in nature. Seed-soil coexistence specificity could be animportant determinant of plant growth and competitive ability, and, therefore, of plant distribution.687.0 SUMMARY1. PGPR for interior spruce can be isolated by using a natural plant enrichment technique.2. The six most effective PGPR strains were identified as Pseudomonas putida (2x),Hydrogenophaga pseudoflava (2x), Bacillus polymyxa, and Staphlococcus hominis.3. PGPR activity depended on specific seed, soil, and pasteurization treatments.4. The growth response to PGPR inoculation varied between the two experiments.5. A disproportionately larger number of PGPR strains were isolated from the endorhizospherethan from the ectorhizosphere.6. Seedling growth was significantly greater when coexistent seed, soil and bacteria were used,compared with seedlings associated with non-coexistent seed, soil and bacteria.7.	 Most of the growth advantage when coexistent seed, soil and bacteria were present could beattributed to the seed x soil specificity.698.0 LITERATURE CITEDAkhromeiko, A.I. and V.A. Shestakova. 1958. The influence of rhizosphere microorganisms on theuptake and secretion of phosphorus and sulphur by the roots of arboreal seedlings. In:Proceedings of the 2nd U.N. International Conference on the Peaceful Uses of Atomic Energy.Geneva. pp. 193-199.Alexander, M. 1977. Introduction to Soil Microbiology. Second edition. John Wiley and Sons, Inc.Toronto. 467 pp.Allison, F.E. 1947. Azotobacter inoculation of crops. I. Historical. Soil Sci. 64:413-429.Amaranthus, M.P. and D.A. Perry. 1987. Effect of soil transfer on ectomycorrhiza formation and thesurvival and growth of conifer seedlings on old, non-reforested clear-cuts. Can. J. For. Res.17:944-950.Bajpai, P.D. and W.V.B. Sundara Rao. 1971. Phosphorus solubilizing bacteria. Soil Sci. Plant Nutr.17:46-53.Baldani, V.L.D. and J. Dobereiner. 1980. Host plant specificity in the infection of cereals withAzospirillum spp. Soil Boil. Biochem. 12:433-439.Baldani. V.L.D., J.I. Baldani and J. Dobereiner. 1983. Effect of Azospirillum inoculation on on rootinfection and nitrogen incorporation in wheat. Can. J. Microbiol. 29:924-929.Ballard, T.M. and R.E. Carter. 1985. Evaluating Forest Stand Nutrition Status. Land managementreport no. 20, Ministry of Forests, British Columbia. 60 pp.Barber, D.A. 1978. Nutrient uptake. In: Interactions Between Non -Pathogenic Soil Microorganismsand Plants. Y.R. Dommergues and S.V. Krupa (editors). Elsevier Scientific. Amsterdam.pp. 131-162.Barber, D.A. and J.M. Lynch. 1977. Microbial growth in the rhizosphere. Soil Biol. Biochem. 9:305-308.Barea, J.M. and M.E. Brown. 1974. Effects on plant growth produced by Azotobacter paspali relatedto synthesis of plant growth regulating substances. J. Appl. Bacteriol. 40:583-593.Bashan, Y. 1986. Enhancement of wheat root colonization and plant development by Azospirillumbrasilense Cd. following temporary depression of rhizosphere microflora. Appl. Env.Microbiol. 51:1067-1071.Bashan, Y. and H. Levanony. 1988. Interaction between Azospirillum brasilense Cd and wheat rootcells during early stages of root colonization. In: Azospirillum IV: Genetics, Physiology,Ecology. W. Klingmuller (editor). Springer Verlag. pp. 166-173.Bashan, Y., Y. Ream, H. Levanony and A. Sade. 1989. Non-specific responses in plant growth, yieldand root colonization of non-cereal crop plants to inoculation with Azospirillum barasilenseCd. Can. J. Bot. 67:1317-1324.Black, C.A. 1965. Methods of Soil Analysis. Part 2. American Society of Agronomy. Madison, NY.1572 pp.Bohlool, B.B. and E.L. Schmidt. 1974. Lectins: a possible basis for specificity in the Rhizobiumleguminosarum symbiosis. Science 185:269-271.7 0Bowen, G. and A.D. Rovira. 1961. The effects of microorganisms on plant growth. I. Development ofroots and root hairs in sand and agar. Plant Soil 15:166-188.Bowen, G.D. and C. Theodorou. 1979. Interactions between bacteria and ectomycorrhizal fungi.Soil Boil. Biochem. 11:119-126.Boyd, R.F. General Microbiology. 1984. Times Mirror/Mosbey College Publishing. Toronto. 807 pp.Brady, N.C. 1974. The Nature and Properties of Soils, 8th edition. McMillan Publishing Company,Inc. New York. 639 pp.Broadbent, P., K.F. Baker, N. Franks and J. Holland. 1977. Effect of Bacillus spp. on increasedgrowth of seedlings in steamed and in nontreated soil. Phytopathology 67:1027-1034.Brown M.E., R.M. Jackson and S.K. Burlingham. 1968. Effects produced on tomato plants,Lycopersicon esculentum, by seed or root treatment with gibberellic acid and indole-3-aceticacid. J. Exp. Bot. 19:544-552.Brown, M.E. 1972. Plant growth substances produced by microorganisms of soil and rhizosphere.J. Appl. Bacteriol. 35:443-451.Brown, M.E. 1974. Seed and root bacterization. Ann. Rev. Phytopath. 12:181-197.Brown, M.E. 1976. Role of Azotobacter paspali in association with Paspalum notatum. J. Appl.Bacteriol. 40:341-348.Brown, M.E. 1982. Nitrogen Fixation by Free-living Bacteria Associated with Plants - Fact orFiction? In: Bacteria and Plants. M.E. Rhodes-Roberts and F.A. Skinner (editors). AcademicPress. New York. pp. 25-41.Brown, M.E. and S.K. Burlingham. 1968. Production of plant growth substances by Azotobacterchroococcum. J. Gen. Microbiol. 53:135-144.Burr, T. and A. Caesar. 1984. Beneficial plant bacteria. CRC Critical Reviews in Plant Science 2:1-20.Burr, T.J., M.N. Schroth and T. Suslow. 1978. Increased potato yields by treatment of seedpieceswith non-specific strains of Pseudomonas fluorescens and putida. Phytopathology 68:1377-1383.Caesar, A.J. and Burr T.J. 1987. Growth promotion of apple seedlings and rootstock by specificstrains of bacteria. Phytopathology 77:1583-1588.Chanway, C.P. and F.B. Holl. 1991a. Biomass increase and associative nitrogen fixation ofmycorrhizal Pinus contorta seedlings inoculated with a plant growth promoting Bacillusstrain. Can. J. Bot. 69:507-511.Chanway, C.P. and F.B. Holl. 1991b. Influence of soil biota on Douglas-fir (Pseudotsuga menziesii(Mirb.) Franco) seedling growth: the role of rhizosphere bacteria. New Phytologist(submitted).Chanway, C.P. and L.M. Nelson. 1990. Field and laboratory studies of Triticum aestivum L.inoculated with co-existent growth-promoting Bacillus strains. Soil Biol. Biochem. 22:789-795.Chanway, C.P., F.B. Holl and R. Turkington. 1988a. Genotypic coadaptation in plant growthpromotion of forage species by Bacillus polymyxa. Plant Soil 106:281-284.7 1Chanway, C.P., F.B. Holl and R. Turkington. 1990. Specificity of association between Bacillusisolates and genotypes of Lolium perenne and Trifolium repens from a grass - legumepasture. Can. J. Bot. 68:1126-1130.Chanway, C.P., L.M. Nelson and F.B. Holl. 1988b. Cultivar-specific growth promotion of springwheat by coexistent Bacillus species. Can. J. Microbiol. 34:925-929.Chanway, C.P., R. Radley and F.B. Holl. 1991a. Inoculation of conifer seed with plant growthpromoting Bacillus strains causes increased seedling emergence and biomass. Soil Biol.Biochem. 23:575-580.Chanway, C.P., R. Turkington and F.B. Holl. 1991b. Ecological implications of specificity betweenplants and rhizosphere micro-organisms. Adv. Ecol. Res. 21:121-169.Chanway, C.P., R.K. Hynes, L.M. Nelson. 1989. Plant growth promoting rhizobacteria: effects ongrowth and nitrogen fixation of lentil and pea. Soil Biol. Biochem. 31:511-517.Cline, M.L. and C.P.P. Reid. 1982. Seed source and mycorrhizal fungus effects on growth ofcontainerized Pinus contorta and Pinus ponderosa seedlings. Forest Science 28:237-250.Cooper, R. 1959. Bacterial fertilizers in the Soviet Union. Soils and Fertility 22:327-333.Curl, E.A. and B. Truelove. 1986. The Rhizosphere. Springer-Verlag, Berlin, Heidelberg. 288 pp.Czabator, F.J. 1962. Germination Value: An index combining speed and completeness of Pine seedgermination. Forest Science 8:386-396.Davey, C.B. and A.B. Wollum II. 1984. Nitrogen fixation systems in forest plantations. In:Nutrition of Plantation Forests. G.D. Bowen and E.K.S. Nambiar (editors). Academic Press.London. pp. 361-377.Dawson, J.O. 1983. Dinitrogen fixation in forest ecosystems. Can. J. Microbiol. 29:979-992.Eklund, E. 1970. Secondary effects of some pseudomonads in the rhizoplane of peat growncucumber plants. Acta Agric. Scandinavica, suppl. 17:1-57.Elad, Y., I. Chet and R. Baker. 1987. Increased growth response of plants induced by rhizobacteriaantagonistic to soil-borne pathogenic fungi. Plant Soil 98:325-330.Evans, H.J. and L.E. Barber. 1977. Biological nitrogen fixation for food and fiber production.Science 197:332-339.Florence, L.Z. and F.D. Cook. 1984. Asymbiotic N2-fixing bacteria associated with three borealconifers. Can. J. For. Res. 14:595-597.Garbaye, J. and G.D. Bowen. 1987. Effect of different microflora on the success of ectomycorrhizalinoculation of Pinus radiata. Can. J. For. Res. 17:941-944.Gardner, J.M., A.W. Feldman and R. Zablotowicz. 1982. Identity and Behavior of xylem-residingbacteria in rough lemon roots of Florida citrus trees. Appl. Envir. Microbiol. 43:1335-1342.Gardner, J.M., J.L. Chandler and A.W. Feldman. 1984. Growth promotion and inhibition byantibiotic-producing fluorescent pseudomonads on citrus roots. Plant Soil 77:103-113.Gardner, J.M., J.L. Chandler, and A.W. Feldman. 1985. Growth responses and vascular plugging ofcitrus inoculated with rhizobacteria and xylem-resident bacteria. Plant Soil 86:333-345.7 2Gaskins, M.H. and D.H. Hubbell. 1979. Response of non-leguminous plants to root inoculation withfree-living diazotrophic bacteria. In: The Soil-Root Interface. J.R. Harley and R.S. Russell(editors). Academic Press. New York. 448 pp.Gaskins, M.H., S.L. Albrecht and D.H. Hubbell. 1985. Rhizosphere bacteria and their use toincrease plant productivity: a review. Agriculture, Ecosystems and Environment 12:99-116.Gaskins, M.H., S.L. Albrecht and J.R. Milam. 1984. Survival of root-associated bacteria in therhizoplane. In: Advances in Nitrogen Fixation Research. Fifth International Symposium onN2 fixation. C. Vegger and W.E. Newton (editors). Martinus Nijhoff Publishers. TheNetherlands. pp. 51-54.Greenland, D.J. and P.H. Nye. 1959. Increases in the carbon and nitrogen contents of tropical soilsunder natural fallows. J. Soil Sci. 10:284-299.Grossnickle, S.C., J.T. Arnott, J.E. Major and T.J. Tschaplinski. 1991. Influence of dormancyinduction treatments on western hemlock seedlings. I. Seedling development and stockquality assessment. Can. J. For. Res. 21:164-174.Harley, J.L. and S.E. Smith. 1983. Mycorrhizal Symbiosis. Academic Press. London. 483 pp.Hiltner, L. 1904. Ueber neuere Erfahrangen und Probleme auf dem Gebiet der Boden-bakteriologieund under besonderer Berucksichtigung der Grundingung und Brache. Arb. Deut. Landw.Ges. 98:59-78.Holl, F.B. 1983. Plant genetics: manipulation of the host. Can. J. Microbiol. 29:945-953.Holl, F.B., C.P. Chanway, R. Turkington and R.A. Radley. 1988. Response of crested wheatgrass,perennial ryegrass and white clover to inoculation with Bacillus polymyxa. Soil Biol.Biochem. 20:19-24.Howie, W.J. and E. Echandi. 1983. Rhizobacteria: influence of cultivar and soil type on plantgrowth and yield of potato. Soil Biol. Biochem. 15:127-132.Hussain, A. and V. Vancura. 1970. Formation of biologically active substances by rhizobacteria andtheir effect on plant growth. Folia Microbiol. (Prague) 15:468-478.Jackson, R.M., M.E. Brown and S.K. Burlingham. 1964. Similar effects on tomato plants ofAzotobacter inoculation and application of gibberellins. Nature 203:851-852.Jaiyebo, B.O. and A.W. Moore. 1963. Soil nitrogen accretion under different covers in a tropicalrain-forest environment. Nature 197:317-318.Jones, D.R. and J.S. Waid. 1963. Influence of fungal isolates on germination and growth ofperennial ryegrass. Plant Soil 19:139-150.Kapulnik, Y. and Y. Okon. 1983. Benefits of Azospirillum spp. inoculation on wheat: effects on rootdevelopment, mineral uptake, nitrogen fixation and crop yield. In: Azospirillum II. W.Klingmuller (editor). Birkhauser, Basel. pp. 163-170.Kapulnik, Y., S. Sarig, I. Nur and Y. Okon. 1983. Effect of Azospirillurn inoculation on yield of field-grown wheat. Can. J. Microbiol. 29:895-899.Kapulnik, Y., S. Sart, I. Nur, Y. Okon, J. Kigel, and Y. Hennis. 1981. Yield increases in summercereal crops of Israel in fields inoculated with Azospirillum. Exp. Agric. 17:179-188.7 3Kapulnik, Y., Y. Okon and Y. Henis. 1985. Changes in root morphology of wheat caused byAzospirillum inoculation. Can. J. Microbiol. 31:881-887.Katznelson, H. 1965. Nature and importance of the rhizosphere. In: Ecology of Soil-Borne PlantPathogens: Prelude to Biological Control. KF. Baker and W.C. Snyder (editors). Universityof California Press. Berkley, CA. pp. 187-209.Katznelson, H. and S.E. Cole. 1965. Production of gibberelin-like substances by bacteria andactinomycetes. Can. J. Microbiol. 11:733-741.Kendrick, B. and S. Berch. 1984. Mycorrhizae: applications in agriculture and forestry. In:Comprehensive Biotechnology, 3rd edition. M. Moo-Young (editor). Pergammon Press. NewYork. pp. 122-165.Khalil, M.A.K. 1986. Variation in seed quality and some juvenile characters of white spruce. SilvaeGenetica 35:78-85.Kimmins, J.P. 1987. Forest Ecology. McMillan Publishing Company, New York. 531 pp.King, 0.E., M.K. Ward and D.E. Raney. 1954. Two simple media for the demonstration ofpyocyanin and fluorescein. J. Lab. Clin. Med. 44:301-307.Kloepper, J.W. and M.N. Schroth. 1978. Plant growth promoting rhizobacteria on radishes. In:Proc. 4th International Conference on Plant Pathogenic Bacteria, Vol. 2. Station depathologie Vegetale et de Phytobacteriology (editor). Angers, France. pp. 879-882.Kloepper, J.W. and M.N. Schroth. 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., D.J. Hume, F.M. Scher, C. Singleton, B. Tipping, M. LaLiberte, K. Frauley, T.Kutchaw, C. Simonson, R. Lifshitz, I. Zaleska and L. Lee. 1988. Plant growth-promotingrhizobacteria on canola (rapeseed). Plant Disease 72:42-46.Kloepper, J.W., J. Leong, M. Tientze and M.N. Schroth. 1980a. Enhanced plant growth bysiderophores produced by plant growth promoting rhizobacteria. Nature 286:885-886.Kloepper, J.W., R. Lifshitz and R.M. Zablotowicz. 1989. Free-living bacterial inocula for enhancingcrop productivity. Tibtech 7:39-44.Moos, W.E. and J.H. Jorgenson. 1985. Staphlococci. In: Manual of Clinical Microbiology, 4thedition. E.H. Lynette (editor). pp. 143-153.Kramer, P.J. and T.T. Kozlowski. 1979. Physiology of Woody Plants. Academic Press. SanFransisco. 520 pp.Krajina, V.J., K. Klinka and J. Worrall. 1982. Distribution and Ecological Characteristics of Treesand Shrubs of British Columbia. The University of British Columbia Vancouver. 131 pp.Lalande, R., N. Bissonnette, D. Coutlee, H. Antoun. 1989. Identification of rhizobacteria from maizeand determination of their plant-growth promoting potential. Plant Soil 115:7-11.Larson, R.I. and J.L. Neal. 1978. Selective colonization of the rhizosphere of wheat by nitrogen-fixing bacteria. In: Environmental Role of Blue-Green Algae and Asymbiotic Bacteria. U.Granhall (editor). Ecol. Bull. (Stockh.) 26:331-342.7 4Lethbridge, G. and M.S. Davidson. 1983. Root-associated nitrogen-fixing bacteria and their role inthe nitrogen of wheat estimated by I- 5N isotope dilution. Soil Biol. Biochem. 15:365-374.Libbert, E. and P. Silhengst. 1970. Interactions between plants and epiphytic bacteria regardingtheir auxin metabolism. VIII. Transfer of 14C-indoleacetic acid from epiphytic bacteria tocorn coleoptiles. Physiol. Plant. 23:480-487.Libbert, E. and R. Manteuffel. 1970. Interactions between plants and epiphytic bacteria regardingtheir auxin metabolism. VII. The influence of the epiphytic bacteria in the amount ofdiffusable auxin from corn coleoptiles. Physiol. Plant. 23:93-98.Lifshitz, R., J.W. Kloepper, M. Kozlowski, C. Simonson, J. Carlson, E.M. Tipping, I. Zaleska. 1987.Growth promotion of canola (rapeseed) seedlings by a strain Pseudomonas putida undergnotobiotic conditions. Can. J. Microbiol. 33:390-395.Lockhead, A.G. and F.E. Chase. 1943. Qualitative studies of microorganisms. V. Nutritionalrequirements of the predominant bacterial flora. Soil Sci. 55:185-195.Lynch, J.M. 1976. Products of soil microorganisms in relation to plant growth. CRC CriticalReviews in Microbiology 5:67-107.Lynch, J.M. 1982. Interactions between bacteria and plants in the root environment. In: Bacteriaand Plants. M.E. Rhodes-Roberts and F.E. Skinner (editors). Academic Press. New York.pp. 1-23.Marschner, H. 1986. Mineral Nutrition of Higher Plants. Academic Press. San Diego. 674 pp.McAfee, B.J. and J.A. Fortin. 1988. Comparative effects of the soil microflora on ectomycorrhizalinoculation of conifer seedlings. New Phytol. 108:443-449.McArthur, J.V., D.A. Kovacic and M.H. Smith. 1988. Genetic diversity in natural populations of asoil bacterium across a landscape gradient. Proc. Natl. Acad. Sci. 85:9621-9624.Mertz, F.P. and R.C. Yao. 1990. Actinomadura fibrosa sp. nov. isolated from soil. Int. J. Sys.Bacteriol. 40:28-33.Mishustin, E.N. 1970. The importance of non-symbiotic nitrogen-fixing organisms in agriculture.Plant Soil 32:545-554.Mishustin, E.N. and A.N. Naumova. 1962. Bacterial fertilizers, their effectiveness and mode ofaction. Mikrobiologiya 31:543-555.Mohammad, G. and R. Prasad. 1988. Influence of microbial fertilizers on biomass accumulation inpolypotted Eucalyptus camaldulensis seedlings. J. Trop. For. 4:74-77.Molina, R. and J.M. Trappe. 1982. Patterns of ectomycorrhizal host specificity and potential amongPacific Northwest conifers and fungi. For. Sci. 28:423-458.Moser, M. 1958. Die kunstliche Mykorrhizaimpfung an Forstpflanzen. I. Erfahrungen bei derReinkultur von Mykorrhizapilzen. Forstwiss. centralbl. 77:32-40.Mubyana, T. 1990. Maize growth promoting rhizobacteria in Saskatchewan soils. Ph.D. Thesis.University of Saskatchewan. 198 pp.Mulder, D. 1979. Soil Disinfection. Elsevier Scientific. New York. 386 pp.7 5Mytton, L. R., M.H. El-Sherbeeny, and D.A. Lawes. 1977. Symbiotic variability in Vicia Faba. 3.Genetic effects of host plant, Rhizobium strain and of host x strain interaction. Euphytica26:785-791.Neal, J.R., Jr., R.I. Larson, and T.G. Atkinson. 1973. Changes in rhizosphere populations ofselected physiological groups of bacteria related to substitution of specific pairs ofchromosomes in spring wheat. Plant Soil 39:209-212.O'Hara, G.W., M.R. Davey, and J.A. Lucas. 1981. Effect of inoculation of Zea mays withAzospirillum brasilense strains under temperate conditions. Can. J. Microbiol. 27:871-877.O'Neill, GA, R.A. Radley and C.P. Chanway. 1991. Emergence promoting rhizobacteria can inhibitconifer seedling growth under nursery conditions. Biol. Fert. Soils (in press).Okon, Y. and Y. Kapulnik. 1986. Development and function of Azospirillum -inoculated roots. PlantSoil 90:3-16.Okon, Y., P.G. Heytler, and R.W.F. Hardy. 1983. N2-fixing by Azospirillum brasilense and itsincorporation into host Setaris italica. Appl. Env. Microbiol. 46:694-697.Pandey, R.K., R.K. Bahl and P.R.T. Rao. 1986. Growth stimulating effects of nitrogen fixingbacteria on oak seedlings. Indian Forester 112:75-79.Parker, A.K. and J.A. Dangerfield. 1975. Influence of bacterial inoculations on growth ofcontainerized Douglas-fir seedlings. Can. For. Service Bulletin. Bi-monthly research notes31:13-14.Parkinson, D., G.S. Taylor and R. Pearson. 1963. Studies on fungi in the root region. I. Thedevelopment of fungi on new roots. Plant Soil 19:332-349.Patriquin, D.G., J. Dobereiner and D.K. Jain. 1983. Sites and processes of association betweendiazotrophs and grasses. Can. J. Microbiol. 29:900-915.Perry, D.A., R. Molina and M.P. Amaranthus. 1987. Mycorrhizae, mycorrhizospheres, andreforestation: current knowledge and research needs. Can. J. For. Res. 17:929-940.Pohlman, A.A. and J.G. McColl. 1982. Nitrogen fixation in the rhizosphere and rhizoplane of barley.Plant Soil 69:341-352.Pokojska-Burdziej, A. 1982. The effect of microorganisms, microbial metabolites and plant growthregulators on the growth of pine seedlings (Pinus sylvestris L.). Pol. J. Soil Sci. 15:137-143.Powell, P.E., P.J. Szaniszlo, G.R. Cline and C.P.P. Reid. 1982. Hydroxamate siderophores in theiron nutrition of plants. J. Plant Nutrition 5:653-673.Reddy, M.S. and J.E. Rahe. 1989. Growth effects associated with seed bacterization not correlatedwith populations of Bacillus subtilis inoculant in onion seedling rhizospheres. Soil Biol.Biochem. 21:373-378.Reid, C.P.P. and J.G. Mexal. 1977. Water stress effects on root exudation by lodgepole pine. SoilBoil. Biochem. 9:417-422.Rennie, R.J. 1981. A single medium for the isolation of acetylene reducing dinitrogen-fixingbacteria from soils. Can. J. Microbiol. 27:8-14.Rennie, R.J. and R.I. Larson. 1979. Dinitrogen fixation associated with disomic chromosomesubstitution lines of spring wheat. Can. J. Bot. 57:2771-2775.7 6Rice E.L. 1974. Allelopathy. Academic Press. New York. 422 pp.Riviére, J. 1960. Etude de la rhizosphere du blè. Ann. Agron. 11:397-440.Riviere, J. 1963. Rhizosphere et croissant du blë. Ann. Agron. 14:619-622.Rouatt, J.W. and H. Katznelson. 1961. A study of bacteria on the root surfaces and in therhizosphere soil of crop plants. J. Appl. Bacteriol. 24:164-171.Rovira, A.D. 1959. Root excretions in relation to the rhizosphere effect. IV. Influence of plantspecies, age of plant, light, temperature and calcium nutrition on exudation. Plant Soil.11:53-64.Rovira, A.D. 1963. Microbial inoculation of plants, I., Establishment of free-living nitrogen-fixingbacteria in the rhizosphere and their effects on maize, tomato, and wheat. Plant Soil.19:304-314.Rovira, A.D. 1969. Plant root exudates. Bot. Rev. 35:35-57.Rovira, A.D. 1972. Studies on the interactions between plant roots and microorganisms. J. Aust.Inst. Agr. Sci. 38:91-94Rovira, A.D. and C.B. Davey. 1974. Biology of the rhizosphere. In: The Plant Root and ItsEnvironment. E.W. Carson (editor). Univ. Virginia Press. Charlottesville. pp. 153-204.Salisbury, F.B. and C.W. Ross. 1985. Plant Physiology, 3rd edition. Wadsworth Publishing Co.Belmont, CA. 540 pp.Salt, G.A. 1979. The increasing interest in minor pathogens. In: Soil Borne Plant Pathogens. B.Schippers and W. Gams (editors). Academic Press. New York. pp. 289-312.Sarig, S., A. Blum and Y. Okon. 1988. Improvement in the water status and yield of field-growngrain sorghum (Sorghum bicolor) by inoculation with Azospirillum brasilense. J. Agric. Sci.Camb. 110:271-277.Sarukhan J., M. Martinez-Ramos and D. Pinero. 1984. The analysis of demographic variability atthe individual level and its population consequences. In: Perspectives on Plant PopulationEcology. R. Dirzo and J. Sarukhan (editors). Sinauer Associates Ltd. Massachusets. pp. 85-105.Schank, S.C., KL. Weier and L.C. McRae. 1981. Yield and nitrogen studies of a digitgrass afterinoculation with Azospirillum on two Australian soils. Appl. Env. Microbiol. 41:342-345.Schroth, M.N. and A.R. Weinhold. 1986. Root-colonizing bacteria and plant health. Hortsci.21:1295-1298.Sims, G.K and E.P. Dunigan. 1984. Diurnal and seasonal variation in nitrogen activity (C2H2reduction) of rice roots. Soil Biol. Biochem. 16:15-18.Smith, R.L., J.H. Bouton, S.C. Schank, KH. Quesenberry, M.G. Tyler, J.R. Milam, M.H. Gaskinsand R.C. Little. 1976. Nitrogen fixation in grasses inoculated with Spirillum lipoferum.Science 193:1003-1005.Smith, W.H. 1970. Root exudates of seedlings and mature sugar maple. Phytopathology 60:701-703.Sprent, J.I. 1979. The biology of nitrogen-fixing organisms. McGraw (U.K.) Ltd. London. 196 pp.7 7Strobel, G.A. and A. Nachmias. 1985. Agrobacterium rhizogenes promotes the initial growth of barerootstock almond. J. Gen. Microbiol. 131:1245-1249.Sumner, M.E. 1990. Crop responses to Azospirillum inoculation. Adv. Soil Sci. 12:53-123.Suslow, T.V. 1982. Role of root colonizing bacteria in plant growth. In: PhytopathogenicProcaryotes, Vol. 1. M.S. Mount and G.H. Lacy (editors). Academic Press. New York. 187PP.Suslow, T.V. and M.N. Schroth. 1982. Role of deleterious rhizobacteria as minor pathogens inreducing crop growth. Phytopathology 72:111-115.Tien, T.M., M.H. Gaskins, and D.H. Hubbell. 1979. Plant growth substances produced byAzospirillum brasilense and their effect on the growth of pearl millet. Appl. Env. Microbiol.37:1016-1024.Tinker, P.B. 1984. The role of microorganisms in mediating and facilitating the uptake of plantnutrients from soil. Plant Soil 76:77-91.Umalia-Garcia, M., D.H. Hubbell, M.H. Gaskins and F.B. Dazzo. 1980. Association of Azospirillumwith grass roots. Appl. Env. Microbiol. 39:219-226.Van Berkum, P. and B.B. Bohlool. 1980. Evaluation of nitrogen-fixation by bacteria in associationwith roots of tropical grasses. Microbiol. Rev. 44:491-517.Vancura, B. and A. Hovadik. 1965. Root exudates in plants. II. Composition of root exudates ofsome vegetables. Plant Soil 22:21-32.Vancura, V. and Hanzlikova, A. 1972. Root exudates of plants. IV. Differences in chemicalcomposition of seed and seedlings exudates. Plant Soil 36:271-282Venkateswarlu, B. and A.V. Rao. 1983. Response of pearlmillet to inoculation with different strainsof Azospirillum brasilense. Plant Soil. 74:379-386.Weller, D.M. 1988. Biological control of soilborne plant pathogens in the rhizosphere with bacteria.Ann. Rev. Phytopath. 26:379-407.Whightman, F., E.A. Schneider and K.V. Thimann. 1980. Hormonal factors controlling theinitiation and development of lateral roots. Physiol. Plant 49:304-314.Whipps, J.M. and J.M. Lynch. 1986. The influence of the Rhizosphere on crop productivity. Adv.Microbiol. Ecol. 9:187-244.Wilkinson, L. 1988. Systat: The System for Statistics. Systat, Inc. Evanston, IL. pp. 823.Willems, A., J. Busse, M. Goor, B. Pot, E. Falsen, E. Jantzen, B. Hoste, M. Gillis, K. Kersters, G.Auling and J. DeLay. 1989. Hydrogenophaga, a new genus of hydrogen-oxidizing bacteriathat includes Hydrogenophaga flava comb. nov. (formerly Pseudomonas flaua),Hydrogenophaga palleronii (formerly Pseudomonas palleronii), Hydrogenophaga pseudoflava(formerly Pseudomonas pseudoflava and "Pseudomonas carboxydoflava"), andHydrogenoflava taeniospiralis (formerly Pseudomonas taeniospiralis). Int. J. Sys. Bacteriol.39:319-333.Yeun, G.Y. and M.N. Schroth. 1986. Interactions of Pseudomonas fluorscens strain E6 withornamental plants and its effects on the competition of root colonizing microflora.Phytopathology 76:176-180.7 8Zobel, B. and J.T. Talbert. 1984. Applied Forest Tree Improvement. John Wiley and Sons, Inc.Toronto. 505 pp.Cl)8tpCn0.CS1COC'D •Zrt— coer1.4LI"Cr)0Cr)CO9.4 `crCO0CV N—.-g1.0 t.0 CO.Na 9-4 r-4 v-4 0.10-0ryCO r•ICgr.4I.-40.OCl)0 U) Cl)79Appendix 2. Analysis of variance of germination capacity (GC) andgermination value (GV) for Experiment 2. SEED - seed provenance;SOIL - soil source; PAST - pasteurization treatment; BACT -putative PGPR strain used.SOURCE DF SUM-OF-SQUARESGC	 GVPROBABILITYGC	 GVSEED 1 4582 1122 0.014 * 0.000 *SOIL 1 3589 167 0.029 * 0.005 *PAST 1 11570 12 0.000 * 0.456BACT 6 6387 88 0.208 0.651SEED*SOIL 1 35 6 0.830 0.579SEED*PAST 1 1155 33 0.216 0.208SEED*BACT 6 9682 220 0.047 * 0.107SOIL*PAST 1 197 23 0.609 0.295SOIL*BACT 6 4702 51 0.399 0.874PAST*BACT 6 2364 107 0.792 0.535SEED*SOIL*PAST 1 567 10 0.386 0.498SEED*SOIL*BACT 6 4311 134 0.457 0.385SEED*PAST*BACT 6 4363 163 0.449 0.259SOIL*PAST*BACT 6 3984 132 0.510 0.396SEED*SOIL*PAST 6 7392 222 0.135 0.104*BACTERROR 1066 805232 2244680•A-4••4.••481• •	 • •00 0 0 0 CI CO CI 0 •cr CI )1, II!8 8 8 2:1 8 r..!	IR	 gg:z; ;s 0aaciocioaaoaciaciac6• • •	 •00000NCOCI00-0CDt-Ot-11,co 8 gcA gg El '' 8 co,	 ,7,4O Oac)Olooac; O OcSoc>c;t0 M CA N• • •	 •	 •	 • •	 • •0 0 0 CI 0 CO N'. CO CI Cf) ,I N 11, •FFI!0 0 0 N 0 CD 0) 0 4•1 CO 0 CI 0) 01 30 40000CD 9-40C19-400C10coac)O0000ociaoaoci• • •	 •	 • •	 • •	 •0 0 0 10 01 0 C*3 0 en CI 01 0- 0 01 4-I0000-0V4-10000104-4001CR 0 0 " CR 72 C! q It) V. R c) '-' to c!oaccooaoaaooacicio123 • * * •	•	 •	 •0 0 0 4t. CD II CD a 4-I co e.- ,-( a) in coO OONVONV NC-00,0000° 0 CI' ° " CD '-c CR al V. 0 CI 01 " c'60060060o° cioocia• •	 •	 • •	 • •	 •m0000 CD C) 01 r4 CD (1) '-4 N N c,0100C-00)01014-40304-40-..-44 .—IOR 0000 0.100CI01000010ea a aaocSOOciaoaoo• •	 •000N0	 00000VIICO4-401.4.0 0 0 01 O U) 4.0 O et to O N N 01 4-40000010410000104-104-i0oaoaaa aoc66 C ciciao4.cf, .4cr 4•-f •-4 CI N CD 00 N N C) 03 CO et.	 4-4	 0`cr CO	 Cn 0) CI CO CO C) 4-1 CD CO OD LI	 N, 4 'TN 11,	 CD COCO CI 1:1 •ct , 0	 CO 4-1 4 cow co' Cs1 Cr)C.4 Ne° 6.4Lc 4	 9-1	 Fri	 NCO	 en N en en M V' 	 CD V' 01	 ID44-i lf) 0 0 0 CD 4 N..n	 Mmommc.IN	 t-6 6 6 6 6 6 6 6 6 6 0M 0 co .-I CO CO C- Nt. Le) 11) C*4 1-4 In CO C.0	 01	 CI44-i L-- .ct .4*. Nr ,	co c*.) a) o a a) cv 4-1 0).4	 cD	 1-4	 ,-4CI 00 CD CD CO	 0.1 ,..40	 1 N,...,C0 	 6—.1 „.4. 52cr, 	.-C	 o.a. o o N 0-	 9-1	 9-4N V 0,	 en•• 4 	O kr) M CY) CO r, N Cs1 en et, 40 II	 e}' v.)	 C)	 et.C•4 CD ID et	 t- ••4	Le)C")	 Cr) kr) 9-14 CD o?,	t-NN010100,0.01t-	 04	 0CONCO 4-I 0) tD N O t- N M C7) N 1-4 M N C?)0 0 'cf. 01 0 CD 4-4 0 0 0	 4-4 0	 CNI	."4csi 4 0 0 0 0 0 6 6 6 6 6 6 6 61-4	 4-1L.' CO ''): '": el '4' e--- LI c! c%? Cr? In. If? "I'. CD.-t C- .--( 01 1" 0 01 C') CI 03 CI 0 01 N 0	 N	 CICD t- Cn 'or V N C7) CO CD CO N N 0 N	 0111, a) "ai t.-*	 1L, 9-4 N N N CD ,c14 MCD	 0CD	 4-i	 4-IN 0 0 CD LI et. 01 N N 1.4-1 N CO CD 4:0N	 0 CI CO 4-1 N 4•4 N VONO CO 4-4N	 V	 Q) 0 co U) co olco	 0N U) 'd' .4:14 	0) CO	 01 01 04 C') C') 11,	 4-4CO4-4 4-4 4-4 CD 4-I 4-1 CD ••4 CD CD	 CD CO CO CDE-1 E-4 E.' E-4 E-1S ) '(- . )a, a) ccl co C14•4	 U ,,,E-4 U U 7-"-i s'a 'Eli L-1 LI	 cn-•d	 0	 •:4 .-ie ,4 < 0 0	 ;,40E-4	 •ts5	ic72 cAAA*o 	 __1(  si bcp E me" ,1, 1c  alF,' :10 1wc)GLIWG46- n 	 Cn4.141W•-,W4,	 a)o	 WC:1-dC.441W00.<4.1[4[40[z.1	 1:4cn	 co ci, Cl., PCI cn o cn v) m 1:1 En 0 co cn cn	 r 	 *0 0 0000R8 2Of2\\Nc\\14. _ enI0 01083•010(/)V)OO0R Ls,(7)•NO•Cof2•z;7)Cf.•0• CU)of	s22	 5•8N1R8E0 g g0 05i9FiRCTSFO	 0Cl)	 VOa0	 a• Ca)Na)2Ona0cv• C2!0)CIECOCCcr)Va)a)840 08 5PligSqPIRS' ° :"Ec6itCr)a)g(7)0El•ICEOEcr)a)Na)2a)a)O	 O861:3U)O05CC80OzCD0 0CttS8 72\\` 2000RI	 (7)3II0U)0


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items