Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

An investigation into the factors contributing to the growth-check of conifer regeneration on Northern… Montigny, Louise E. M. de 1992

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

Item Metadata


831-ubc_1992_fall_de_montigny_louise.pdf [ 2.95MB ]
JSON: 831-1.0075172.json
JSON-LD: 831-1.0075172-ld.json
RDF/XML (Pretty): 831-1.0075172-rdf.xml
RDF/JSON: 831-1.0075172-rdf.json
Turtle: 831-1.0075172-turtle.txt
N-Triples: 831-1.0075172-rdf-ntriples.txt
Original Record: 831-1.0075172-source.json
Full Text

Full Text

AN INVESTIGATION INTO THE FACTORS CONTRIBUTING TO THEGROWTH-CHECK OF CONIFER REGENERATIONON NORTHERN VANCOUVER ISLANDbyLOUISE de MONTIGNYB.S.F., University of British Columbia, 1983M.F.S., Yale School of Forestry, 1985A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENT FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIES(Department of Forest Sciences)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1992Louise de Montigny, 1992In 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 tJJ. JtZd2iThe University of British ColumbiaVancouver, CanadaDate 0JL/(, 79’2DE-6 (2188)AbstractConifer plantations established on cutovers in the CWHb1zone on Northern Vancouver Island grow well initially, butcoincident with the reinvasion of salal (Gaultheria shallonPursh.) on sites formerly dominated by western red cedar andwestern hemlock (CH phase) the trees become chiorotic andgrowth stagnates. These symptoms are not seen on sitesdominated by western hemlock and amabalis fir (HA phase).This study examined some site and soil chemical factors whichcould be responsible for differences in forest productivityby: 1) documenting physical differences between the CH and HAphases; 2) documenting morphological and chemical differencesbetween the organic horizons found in forest floors of CH andHA phases using classical wet chemistry techniques and ‘3Cnuclear magnetic resonance (NMR) spectroscopy; 3) determiningseasonal trends in free phenolic acid concentrations of soilsunder salal on CH cutovers by high speed centrifugation andHigh Performance Liquid Chromatography, and 4) determining ifsolutions of phenolic acids at field concentrations or ofsalal leachates have a negative effect on conifer seedgermination, growth and short term 32phosphorus uptake.The HA phase was found to occur on higher topographiciipositions making them drier and more susceptible to windthrowevents than the CH phase which occur on adjacent, lower slopepositions. This windthrow process appears to rejuvenate thesite by mixing organic horizons with mineral soil, increasingaeration, breaking hardpans, and generally improving siteconditions.Six distinct humus horizons were identified on the basisof origin and on degree of decomposition. The proportion ofhumus horizon types on the CH reflected ecosystem maturity andlack of disturbance; that of the HA indicated repetitivewindthrow events. The CH humus horizons tended to have higherconcentrations of K, Ca, Mn, available S, lipids, and totaland labile polysaccharides, as well as a higher pH. The HAhumus horizons were found to be higher in available N and Pand tended to have a lower C/N ratio for the more well—humified horizons. Tannin signals from ‘3C NMR spectroscopywere found in the Fm horizons of both CH and HA, but theintensity was greater for the CH. The source of tanninsappears to be salal, as strong tannin peaks were identified insalal roots, leaves, flowers, berries and litter. Tannins areknown to inhibit decomposition and mineralization processes.Concentrations of phenolic acids originating fromangiosperms (presumably salal) were significantly higher insummer months, coincident with greater physiological activityof salal. Phenolic acids are known to cause root membranedysfunctioning in some situations. The germination values ofiiiseeds and the biomass of seedlings of Sitka spruce, westernhemlock and western red cedar tended to be lower giventreatments of either a phenolic acid solution at fieldconcentrations or a salal leachate solution compared to acontrol of distilled water. The uptake of 32P by the excisedroots of Sitka spruce and western hemlock were higher for thephenolic acid and salal leachate solutions than for thecontrol, indicating P was probably limiting even after only 12weeks. Uptake of 32P by fine roots of mature western red cedarand western hemlock was significantly reduced by the phenolicacid and salal leachate solutions.In conclusion, this study provides evidence to suggestthat the growth—check of conifer regeneration involves anumber of interacting factors. These include: 1) the presenceof hardpans and compacted mineral soils which contribute toannual periods of anaerobic soil conditions; 2) the presenceof large pockets of nutrient poor woody humus, and 3) thepresence of tannins, lipids and phenolic acids in active humushorizons which may be contributing to decreased decomposition,mineralization and nutrient uptake in trees.ivTable of ContentsAbstract iiTable of Contents vListofTablesList of Pigi.ires xiAcknowledgements xiiiCHAPTER I. INTRODUCTION 11.0 THE PROBLEM 12.0 PRELIMINARY APPROACHES AND DIFFICULTIES 42.1 Choosing to work with phenolic acids . . 52.2 Monitoring the release ofphenolic acids using XAD Resin . . 52.3 Sampling Forest Floor by BulkSampling 102.4 Extraction and CharacterizationofPolyphenolics 113.0 FINAL APPROACH AND HYPOTHESES 123.1 Study Objectives 123.2 Overall Hypothesis 133.3 Individual Hypotheses 133.3.1 Soilclassification andcharacterizationof CH and HA sites 133.3.2 Chemicalcharacterizationof humus horizonsusing carbon 13nuclearmagnetic resonance 143.3.3 Allelopathicpotential of salal 14CHAPTER II. THE EFFECTS OF ERICACEOUS PLANTS ON FORESTPRODUCTIVITY: A LITERATURE REVIEW . . . . 151.0 INTRODUCTION 152.0 DISTRIBUTION OF ERICACEOUS PLANTSIN THE NORTHERN HEMISPHERE 153.0 THE HEATHER CHECK SYNDROME 184.0 EFFECTS OF ERICACEOUS PLANTS 214.1 Allelopathy 21v4.2 Soil Acidification and Paludifjcation4.3 Humus Decomposition50 MANAGEMENT OF SITES DOMINATED BYERICACEOUS SPECIES6 . 0 StJI.1IIA..RyCHAPTER III. SOIL CLASSIFICATION ANDCHARACTERIZATION OF CH AND HASITES1.0 INTRODUCTION2 . 0 METHODS2.1 Soil and Vegetation Descriptionand Collection2.2 Laboratory Methods2.2.1 Nutrient analysis2.2.2 Polysaccharides and cellulose2.2.3 Lipids2.2.4 Bound phenolic acids2.2.5 Statistical analysis3.0 RESULTS AND DISCUSSION3.1 Site Description3.2 Humus Classification and Variability3.2.1 Humushorizons3.2.2 Humusprofjles3.2.3 Soil variability3.3 Nutrient Concentration of Humus Horizons3.3.1 Nutrientconcentration ofwoody horizons3.3.2 Nutrientconcentration ofnon—woody horizons3.3.3 Nutrientconcentrationdifferencesbetween sites3.4 Organic Composition of Humus Horizons3 . 4 . 1 Lipids3.4.2 Polysaccharides and cellulose3.4.3 Bound phenolic acids4.0 SUMMARY5.0 CONCLUSION5.1 Soil Classification5.2 Chemical Differences in Humus Horizonsbetween CH and HA• . 2729• . 31• • 33• . 35• 35• 3636• 38• 38• 39• 39• 40• 41• 41• 4145• 45• 475561• . • 61• • • 63• . • 6477• • • 77• . • 81• • • 84• • • 90• • • 95• • • 9596viCHAPTER IV. CHEMICAL CHARACTERIZATION OFHUMUS HORIZONS USING CARBON-13NUCLEAR MAGNETIC RESONANCESPECTROSCOPY . . . . 971.0 INTRODUCTION2.0 METHODS2.1 Sample Preparation2.2 NMR Spectroscopy2.3 Spectral Analysis3.0 RESULTS AND DISCUSSION3.1 Carbon-13 CPMAS NNRCharacterization of HumusHorizons3.1.1 Woody horizons3.1.2 Non—woody horizons3.2 Dipolar-dephased Carbon-13 CPMASNMR Characterization of HumusHorizons3.3 Carbon-13 CPMAS NNRCharacterization of LitterInputs3.4 Site Differences4 . 0 SUIIIvIA.IY5.0 CONCLUSIONCHAPTER V. ALLELOPATHIC POTENTIAL OF SALAL1.0 INTRODUCTION2.0 METHODSSeed germinationSeedling growthSeedling root bioassay .Mature root bioassay• . . 97• . 98• . . 98• • . 98• . . 99• • 100• . 100101107• . 110114117119120121• . • 121• . . 123• . . 123• . . 123• • . 124• . . 126• . . 127• • . 128• • • 128• • . 130131• . . 132• . . 133• . 134• . . 134142• . . 142• • . 146• • • 147• • • 150• • • 1532.1 Seasonal Phenolic Acid Concentration2.1.1 Soil sampling2.1.2 Sample preparation2.1.3 HPLC analysis2.1.4 Statistical analysis • •2.2 Effects of Phenolics on Conifers2.2.1 Seed germination2.2.2 Seedling growing conditions2.2.3 Seedling root bioassay .2.2.4 Mature root bioassay . .2.2.5 Statistical analyses • .3.0 RESULTS AND DISCUSSION3.1 Seasonal Phenolic Acid Concentration3.2 Effects of Phenolics on Conifers . . 0 StJ1M.ARYvii5.0 CONCLUSION5.1 Seasonal Phenolic Acid ConcentrationsUnderSalal5.2 Effects of Phenolics on ConifersCHAPTER VI. OVERALL DISCUSSIONCHAPTER VII. SUMMARY MID CONCLUSION1 . 02 . 0 CONCLUSIONLITERTURE CITED 180155155156157173173178viiiList of TablesCHAPTER III3.1. Site Location and characteristics . . . 423.2 Vegetation (% presence and % cover onthose plots where present) by standnumber 4 43.3. Physical descriptions of organic soilhorizons on CH and HA sites 463.4. Proportion of humus forms on CH and HAtrench profiles 543.5. Site property values including mean,standard deviation and occurrence (% ofplots) on the CII and HA sites 563.6. Forest floor horizon depths includingmean, standard deviation and occurrence(%)bysite 583.7. Mineral soil horizon depth (cm),standard deviation, and occurrence (% ofplots) by site type . . 603.8. Mean nutrient concentrations andstandard deviation by horizon on oven—dry and ash—free basis 623.9. Mean nutrient concentrations, standarddeviation and number of samples byhorizon and site on oven—dry and ash—free basis 663.10. ANOVA for Horizon and Site (Probabilityof Null Hypothesis) and Squared MultipleR 723.11. Organic composition of humus typesincluding lipids, total polysaccharides,labile polysaccharides and cellulose bysite based on 4 replicates 783.12. Concentration of bound phenolic acids(standard deviation) by humus horizon . 863.13 Concentration of bound phenolic acids (standarddeviation) by humus horizon and site 88ix3.14. Pearson correlation matrix between boundphenolic acid concentration andabundance of salal or other shrubs,presence of wood, and site (CH or HA) 89CHAPTER IV4.1. Relative percentages of carbon inchemical shift regions of horizons bysite. 1034.2. Ratios of calculated total lignin,carbohydrate and aromatic carbon andassociated ratios for horizons by site . 104CHAPTER V5.1. Seasonal pH, water content and phenolicacid concentration means and standarddeviation by month and horizon (non—woody Fm and woody Hw) 1355.2. Germination indices by species,pretreatment and treatment 1435.3. Seedling, shoot, root and total biomass,and uptake of inorganic P by species andtreatment 1485.4. Root biomass and uptake of inorganicphosphorus in nanomoles per oven—dry gof mature root 152xCHAPTER IList of FiguresContinuous root exudate trapping system.Soil coring CH trench profile from 0 to 6 CH trench profile from 6 to 11 CH trench profile from 11 to 16 in3.2a. HA trench profile from 0 to 4 in .3.2b. HA trench profile from 4 to 9 in .3.2c. HA trench profile from 9 to 14 in3.3. a) pH, and b) Water content of Humushorizonsbysite3.4. a) Total carbon concentration, andb) carbon/nitrogen ratio of humushorizons by site3.5. a) Total nitrogen concentration, andb) total sulphur concentration ofhumas horizons by site3.6. Available nutrient concentrations ofhumus by site including a) availablenitrogen, and b) available phosphorus3.7. Available sulphur concentration ofhumus horizons by site3.8. Exchangeable cation concentrations ofhumus horizons by site including a)calcium and b) pottassium3.9. a) Exchangeable magnesium, andb) exchangeable manganese of humushorizons by site3.10. Concentration of lipids in humushorizonsbysite48495051525368697071747576791.1.1.2.CHAPTER III68xi3.11. a) Total polysaccharides, andb) labile polysaccharides by humushorizons by site 823.12. Cellulose concentration byhumushorizonandsite 83CHAPTER IV4.1. Structural units of a) cellulose, b)lignin and c) condensed tannins . . . . 1024.2. Carbon-13 CPMAS NNR spectra of woodyhorizons (Fw, Hrw, Hdw) from CH and HAsites 1054.3. Carbon-13 CPMAS NNR spectra of non-woody horizons (Fm, Hh, IIhi) from CIIandHAsites 1084.4. Dipolar dephased Carbon—13 CPMASspectra of woody horizons (Fw, Hrw,Hw) from CII and HA sites 1124.5. Dipolar dephased Carbon—13 CPMASspectra of non—woody horizons (Fm, Hh,Hhi) from CII and HA sites 1134.6. Carbon—13 CPMAS NNR spectra of littermaterials from a) CH site salallitter, b) CII site coniferous litterand C) HA site coniferous litter . . 1154.7. Carbon-13 CPMAS NNR spectra of salal:a) flowers, b) leaves, and c) roots . 116CHAPTER V5.1. Design of soil centrifuge tubesmodified from a 30 cc syringe 1255.2. Seasonal phenolic acid concentration(ng/g o.d. soil) in Hw horizons undersalal in cutovers 1375.3. Seasonal phenolic acid concentration(ng/g o.d. soil) in Fm horizons undersalal 138xiiAcknowledgementsI am grateful to Drs. Gordon Weetman, Lawrence Lowe,Karel Klinka, Edith Camm and Morag McDonald of U.B.C. for alltheir helpful advice, and to Carol Dyck for technical support.Thank—you to Western Forest Products and especially to Dr.John Barker, Bill Dumont, Steven Joyce and Heather Joneswithout whose field expertise this project would not have beenpossible. I am particularly indebted to Dr. Caroline Prestonfor allowing me the opportunity to work at Pacific ForestryCentre and to all the helpful technical advice there fromKevin McCullough, Ann Van Niekerk, Doug Taylor and Rob Hagel.Finally, a very special thank-you to members of my familyincluding Pol, Pierette, Pierre and Denis de Montigny andRaoul Wiart for the many hours contributed to tedious tasks.This work was supported by the South Moresby ReplacementFund (SMRF), the Salal Cedar-Hemlock Integrated ResearchProject (SCHIRP) and the B.C. Science Council GraduateEngineering and Technology Award (GREAT).xiiiChapter IIntroduction1.0 THE PROBLEMThe growth—check of conifer regeneration on northernVancouver Island cutovers has been a problem intriguing forestersand researchers on the north coast of British Columbia since themid 1970’s. At that time, plantations of Sitka spruce (Piceasitchensis (Bong.)Carr.) and naturally regenerated westernhemlock (Tsuga heterophylla (Raf.) Sarge) and western red cedar(Thuja plicata Donn ex D. Donn), which were established in thelate 1960’s and which had initially grown well, with leadergrowth up to 50 cm, began to show signs of stress. The annualleader growth was 5 to 10 cm and the needles displayed chiorosis,suggestive of nitrogen and phosphorus deficiencies (Weetmanl989a,b). Coincident with the plantation “check” was thereinvasion of the cutovers with salal. Trees planted on1roadsides or landings, or those growing on adjacent cutoverswithout salal, had leader growth of 0.5 in and no apparentchlorosis.Lewis (1982) classified the different sites as “phases” ofthe Thula plicata - Tsuga heterophylla - Gaultheria shallon -Rhytidiadeiphus loreus or “salal — moss” ecosystem. Within thissingle ecosystem association, the phases represented two verydifferent kinds of forest occurring side by side: the cedar-hemlock (CH) phase being the climatic climax community, and thehemlock—amabalis fir (HA) phase being a seral stage occurring onsites with a history of soil disturbance.The CH phase consisted of somewhat open western red cedarand western hemlock stands with a minor component of amabalis firand a dense understorey of salal. These stands were believed tohave been left undisturbed for as long as 1000 years. Followingclearcutting, with or without slashburning, the secondaryvegetation was overwhelmingly dominated by salal. Estimates ofabove—ground biomass of salal were found to increase from 1058 kgha1 two years after cutting and burning, to 4078 kg hat after 8years (Messier and Kimmins, 1990). The growth-check of coniferregeneration occurred on these CH cutovers.The HA phase occurred on sites with a history of sitedisturbance, such as fire, windthrow, or on steeper terrain wheresoil creep and periodic windfall disturbed the soil. Theresultant stands consisted of even—aged, densely stocked westernhemlock and amabalis fir, with relatively clean forest floors2dominated by mosses. Shrub and herb cover were sparse and tendedto favour mineral soil overturns. Planted and naturalregeneration did not show the same symptoms of growth—check andfoliar nutrient deficiencies. The occurrence of communitiesintermediate between the climatic climax CH phase and the seralHA phase lead Lewis (1982) to state that the HA phase had thepotential to develop into the CH phase, given a sufficiently longperiod of time without soil disturbance. Part of the focus ofthis study was therefore to examine some of the chemical andphysical site differences between the CH and HA phases todetermine if these factors may be, in part, responsible for thegrowth-check of conifer regeneration on the CH but not the HAphase.The differences in productivity between cutovers formerlyoccupied by salal—rich cedar—hemlock sites and adjacent salal—free hemlock—amabalis stands have been thought to be due tocompetition by salal. Attempts to eradicate in order to reducethe level of competition by salal have proven to be verydifficult. The thick cuticular wax and deeply rooted, easilypropagated rhizome of salal have made eradication by herbicides,fire, scarification, or clipping almost impossible. Attempts toameliorate the nutrient deficiencies of the trees by nitrogen andphosphorus fertilization have been temporarily successful. Therestoration of rapid cedar growth lasted 5 to 7 years but foliarnutrient concentrations soon returned to pre—fertilizationdeficiency levels. The extent of the problem quickly became3evident, as an estimated 100,000 hectares of these salalecosystems are found throughout the wetter portions of theCoastal Western Hemlock biogeoclimatic zone on Vancouver Island,the coastal mainland and the Queen Charlotte Islands.Experiences in the United Kingdom with Sitka spruce plantedin dense swards of heather (Calluna vulgaris (L.) Hull) broughtsome interesting ecological parallels to light: salal and heatherare both ericaceous shrubs, the symptoms of growth—checkincluding slowed growth and chlorosis are identical, and therapid but temporary response to fertilization is similar(Malcolm, 1987). Evidence for allelopathy by heather (discussedin Chapter 2) has been accumulating, and the question naturallyarose whether salal too, could be allelopathic. To date, littlework has been done on the allelopathic role of salal in theseconiferous ecosystems. Part of the focus of this study wastherefore, to determine if salal could be responsible for theproduction of compounds which could have an allelopathic effecton conifers.2.0 PRELIMINARY APPROACHES AND DIFFICULTIESThis project was intended to be an exploratory studyexamining some of the physical site factors which may beresponsible for differences in conifer regeneration aftercutting, and the possible role of salal as an allelopathiccompetitor to conifer species in plantations. It was intendedfrom the start that results which proved promising would be4pursued, while those that failed would be abandoned. Thefollowing describes some initial work which was felt to beinappropriate for the study, and reasons why they were abandoned.2.1 choosing to Work with Phenolic AcidsMuch of the literature dealing with allelopathic activity ofericaceous plants suggests that both phenolic acids and fattyacids may be the principal allelochemicals involved in allegedconifer—ericaceous allelopathic interaction. However, a gaschromatograph, which would have allowed an examination of fattyacids in these ecosystems, was not available during the course ofthis study. It was therefore decided to limit the study to thatof phenolic acids.2.2 Monitoring the Release of Phenolic Acids Using XAD ResinThe state—of—the—art method for examining root exudation ofphenolic acids was the Continuous Root Exudate Trapping System(CRETS) described by Tang and Young (1982), shown in Figure 1.1.Salal was grown in acid-washed sand in inverted 2.5 litre glassbottles with the bottom removed. A very dilute nutrient solutionwas recycled through the pot using a continuous air stream, andthe solution passed through a column containing XAD-4 resin whichabsorbs hydrophobic and partially hydrophobic compounds, such asphenolic acids. The accumulation of root exudate componentsreleased over a period of time could then be detected, even ifproduced at very low levels.5PLANT---- -Figure 1.1: Continuous root exudate trapping system (From Tangand Young, 1982).ROCkCOARSETEFLON SLEEVESOLIJTI ONAl RTEFLON - - -GLASS WOOLTEFLON SLEEVECONNECTION6Initial analysis of phenolic acids desorbed from the XAD-4resin indicated large concentrations of one acid, but this wasattributed to resin degradation. After more effective clean—up,trace amounts of some phenolic acids were detected, but the mostpolar compounds were absent. Levels were too low to allowquantitative estimates. It was assumed that the results haddemonstrated that salal did not exude phenolic acids. A similarapproach was used to study variability of phenolic acids in soilsolution of forest humus. A coring tool was constructed in whicha stainless steel cylinder could cut and capture a soil corewithin a PVC cylinder, which could then be capped for transferand storage, then easily modified to act as a leaching tubewithout any further transfer or disturbance of the soil core(Figure 1.2). A total of two hundred cored samples werecollected and leached through XAD-4 resin at a field laboratoryset up at Port McNeill. The phenolic acids were desorbed usingmethanol and kept refrigerated until analyzed through HPLC atU.B.C.Measurement of seven separate phenolic acids showed nosystematic relationship to soil phase (CH or HA).Protocatechuic, p—hydroxybenzoic, and vanillic acids werecommonly detected, although at rather low concentrations. Gallicand syringic acids were sometimes found. Interference bycoloured contaminants, manifested by erratic baselines, madequantification of some acids unreliable. Some attempts wereinfluenced by ‘bleeding’ of breakdown products from the XAD-47core cutter3ft pusher—pvc coupler/nutFigure 1.2: Soil coring tool.8resin. This was largely overcome by better column clean—up.The composition of phenolic acids obtained in the CRETS andfield study raised questions about the adequacy of recovery ofall the compounds expected. A careful evaluation of the recoveryof standard compounds, using the published XAD-4 method forseparating phenolic acids from aqueous solutions, indicated thefailure of XAD—4 (even under very acid conditions) toquantitatively recover the phenolic acids present. While therecovery of the less polar acids like vanillic and ferulic werereasonable, losses of the most polar acids were excessive (53%and 89% for protocatechuic and gallic respectively). Examinationof other solid adsorbents, and of variation in pH and eluentsused, failed to overcome this problem.An additional problem with the resin system lay in the factthat XAD resin allows simple phenolic compounds to continuerecycling with the nutrient solution, and thus renders themsusceptible to further microbial degradation. This makes anyestimates of phenolic acid accumulation rather dubious.In light of the problems noted, it was concluded that theXAD approach could not yield reliable quantitative data, andcould well yield misleading results. In relation to publishedstudies of allelopathic effects, a CRETS system can presumably beused to demonstrate the presence of potentially allelopathicagents, but caution must be used in accepting claims forallelopathic effects without rigorous verification of methodreliability. Furthermore, allelopathic effects shown by exudate9fractions, may or may not be due to the specific chemicalsanalyzed, as other active ingredients may also be present.2.3 Sampling Forest Floor by Bulk SamplingThe initial study of free phenolic acids in forest floorsusing resins was based on a bulk sample of forest floor materialcored and kept intact for subsequent leaching. As previouslydescribed the resin method itself was found to be inadequate, butother deficiencies in the methodology were subsequentlyrecognized.Following leaching of the intact soil cores, Munsellcolours, pH, horizon sequence and depths were recorded for allsamples. Although pH rarely varied by more than 0.5 pH units atany one plot, the proportions of horizon types in each corereflected substantial spatial variability. In particular theproportion of decayed wood varied a great deal. Most corescontained two distinct horizon types in varying proportions.It was concluded that demonstrating significant differencesin phenolic acid levels or composition would require a much morehomogenous sample base than that provided by random cores, unlessa prohibitively large number of samples were collected. At thesame time, it was confirmed that rather large sample weightswould be needed for reliable analytical determination ofindividual phenolic acids by HPLC.A preliminary set of grab samples collected from trenchtransects at Port McNeill indicated a dichotomy between woody and10non—woody samples and within each, a variation in degree ofhumification. The differences in plant origin and associatedproperties reflect differences in biochemical processes, and evenmicrosite variations within the forest floor. In order to gainan understanding of these processes, it was therefore deemednecessary to study the behaviour of individual types of materialswhich would be reasonably well characterized. A furtheradvantage of working within defined horizons is that of allowingmore effective communication with other investigators working onsimilar (or dissimilar) materials.2.4 Extraction and Characterization of PolyphenolicsPublished methods used to extract polyphenolics involveshaking overnight in acetone, concentrating, then extracting withethyl acetate to remove waxes and low molecular weight compounds.The aqueous extract is then freeze—dried, and eluted through acolumn of Sephadex LH—20 using combinations of methanol, waterand acetone. Carbon—13 Nuclear Magnetic Resonance Spectroscopyis then used to identify the final product.When this procedure was used, there was a thick suspensionbetween the acetone and ethyl acetate layers, and no cleardemarkation between any of the layers, thus making removal ofwaxes and low molecular weight compounds difficult. Theresulting extract was concentrated and run on Sephadex LH—20resin using 1:1 water:methanol, although a thick waxy layerdeveloped on concentrating. Three coloured bands were separated:11a very fast brownish band; a medium-fast light-yellow band, and aslow, greyish—yellow band. These were individually evaporateddown and freeze dried. The final extract, presumably containingthe tannins, was removed with 70% acetone, evaporated down, andfreeze—dried. The procedure was replicated 4 times.When these extracts were examined using NMR, results made nosense, looking more like degradation products of the Sephadexthan tannins. It was concluded that the published methods werepoorly described. An attempt was made to find a suitable methodfor tannin extraction, but none was found. Interestingly, Moleand Waterman (1987a and b) compared several methods for measuringtannins and concluded none were consistent. This approach wastherefore considered beyond the scope of this study and was notpursued.3.0 FINAL APPROACH AND HYPOTHESES3.1 Study ObjectivesThe final approach of the overall study has two objectives:Oblective 1:To document in detail some of the differences in soil andstand characteristics between CH and HA phases, includingmorphological and chemical differences between the organichorizons found in forest floor horizons, using classical wetchemistry techniques, as well as NMR mass spectrometry.It was anticipated that this would allow the identificationof factors responsible for growth stagnation on CH sites.12Obiective 2:To determine seasonal trends in phenolic acid concentrationsunder vigorous salal on regenerating cutovers on CH sitesand to determine if the concentrations found had anallelopathic effect on conifer seed germination, growth, andshort term phosphorus uptake.3.2 Overall HypothesisThe overall hypothesis of the study is that there aredifferences in chemical characteristics of similar humus horizonsbetween CH and HA sites, and furthermore, these differences maybe the cause of the poor productivity of conifer regeneration oncutover CH sites.3.3 Individual Hypotheses3.3.1. Soil classification and characterization ofCH and HA phases1) That distinct and recognizable humus horizons occurcommonly on both the CH and HA sites, but that therelative abundance of the horizons varies between thesites, and2) That similar horizons from the CH and HA sites aredifferent with respect to chemical compositionincluding total and/or available nutrientconcentrations, lipid and polysaccharide contents, andphenolic acid content.133.3.2 Chemical characterization of humus horizonsusing ‘3C Nuclear Magnetic Resonance1) That the NNR spectra obtained from the different humushorizons are distinct and recognizable, and2) That similar horizons from the CII and HA sites aredifferent with respect to phenolics and tannins.3.3.3 Allelopathic potential of salal1) That the concentration of free phenolic acids foundunder salal in plantations will vary with season andwill be highest during the summer months when salal ismost physiologically active, and2) That solutions using the maximum concentrations of freephenolic acids found in soils under salal and ofleachates from the flowers and berries of salal, willcause reduced seed germination, biomass growth, and 32Puptake in roots of Sitka spruce, western hemlock andwestern red cedar.14Chapter IIThe Effects Of Ericaceous Plants on Forest Productivity:A Literature Review 11.0 INTRODUCTIONThe chlorosis and stagnation of conifer plantationsassociated with salal on northern Vancouver Island was found tobe similar to that reported for other conifer—ericaceousdominated ecosystems (Weetman p1., 1989a,b). This led to thespeculation that perhaps ericaceous plants use similar mechanismsto achieve similar symptoms in conifers. The literature was thenreviewed to gain an understanding of the history and extent ofthe problem in other areas of the world and the results of pastresearch dealing with ericaceous plants.2.0 DISTRIBUTION OF ERICACEOUS PLANTS IN THE NORTHERN HEMISPHEREEricaceous plants are one of the most ecologically successful plants in the Northern Hemisphere, with representatives1 This chapter is based on the publication by de Montigny andWeetman (1990).15dominating vast areas of heathiands. The term “heathland” isused to describe territories in which trees or tall shrubs aresparse or absent, and in which the dominant life—form is that ofthe ericaceous dwarf shrub, as represented by the order Ericales,particularly of the family Ericaceae, which comprises aboutseventy genera and more than 1900 species. Specht (1978) hasenumerated the conunon traits for worldwide heathland communitiesas follows: 1) their evergreen scierophyllous nature; 2) thepresence, but not necessarily the dominance, of the heathfamilies in the stand — Diapensiaceae, Empetracaea, Epacridaceae,Ericaceae, Grubbiaceae, Prionotaceae, Vacciniaceae, and 3) theirecological restriction to soils very low in plant nutrients.These infertile soils may be well-drained (supporting dryheathlands or ‘sand—heath’) or seasonally waterlogged (supportingwet-heathland).The conditions favouring the dominance of heathlands involvea combination of a relatively cool temperate regime, highhumidity throughout most of the year, and rather freely-drainedsoil not conducive to formation of peat. In addition, somefactor, whether climatic, edaphic, biotic or anthropogenic, mustoperate to initially remove, or exclude the development of tallershrubs and trees. This type of environment may be expected inthree main categories of situations (Gimingham, 1972): 1) wherefor any reason forest is excluded in the strongly oceanic regionsof the cool-temperate belt; 2) in certain parts of sub-arcticand sub-antarctic territory, and 3) where adequate humidity16prevails at sub—alpine or low—alpine altitudes on mountains. Ofthe three situations, the first is of particular economic andland—use importance in developed countries.Oceanic heathlands are most widely represented in westernEurope, South Africa, Eastern Canada and U.S.A. In Europe, theheath region belongs essentially to the oceanic and sub—oceanicregions of west Europe, particularly the broad west Europeancoastal plain in countries bordering the North Sea and theEnglish Channel. The most prominent genera are Erica andCalluna. In Canada, the heathlands predominate in Nova Scotia,Newfoundland and coastal British Columbia, while in U.S.A.,mainly along the extreme eastern coastline from Maine to SouthCarolina and the Alaska panhandle; the prominent genera areVaccinium, Gaylussacia, Gaultheria, Rhododendron, Kalmia andArctostaphylos.Heathlands of the cool—temperate oceanic region are believedto have arisen from pre—existing forests following humansettlement and land cultivation. Analytical studies of pollenindicate the expansion of ericaceous pollen in relation to treepollen in Denmark (Jonassen, 1950) and Norway (Kaland, 1986).Historical land use studies indicate that repeated cutting andburning for agricultural land use effectively maintained andperpetuated the heath vegetation over considerable areas ofBritain (Conway, 1947), Sweden and Denmark (Romell, 1952), Norway(Behre, 1986) and Newfoundland (Meades, 1986). Similarly,clearcut logging and slashburning on northern and western17Vancouver Island has stimulated the productivity of salal(Gaultheria shallon Pursh.) which appears to be correlated withthe exclusion of forest tree regeneration (Weetman1989a,b). Salal biomass was found to almost quadruple from 1058kg ha1 at 2 years after cutting and burning, to 4078 kg ha’ at 8years, while below—ground biomass (including roots of Vacciniumspp) were found to increase from 1449 kg ha-’ at 2 years to 8507kg ha-’ at 8 years (Messier and Kimmins, 1991).3.0 THE HEATHER CHECK SYNDROMEThe effect of ericaceous plants on trees was first observedbetween Calluna vulgaris (L.) Hull and Sitka spruce (Piceasitchensis (Bong.) Carr.) in Scotland by Muller (1897) whocommented that although cultivation of the heathiand soil byploughing and harrowing several times in the course of threeyears resulted in satisfactory growth of spruce, this was onlymaintained so long as heather did not re—invade the sites.Prevention of re—invasion by heather allowed the spruce tocontinue to grow but once the heather covered the site again,stagnation of the spruce ensued; this effect was more pronouncedwhere there had been more rapid growth of the spruce followingcultivation of the soil.In early trials of afforestation on the heathiands, it wasnoted that “heather—sensitive” species such as Sitka spruce,Norway spruce (Picea excela Link), Douglas—fir (Pseudotsugamenziessii (Mirb.) Franco), silver fir (Abies amabalis (Dougi.)18Forbes), western hemlock (Tsuga heterophylla (Raf.) Sarg.), andLawson cypress (Chamaecyparis lawsoniana (A. Murr.) Pan.),virtually ceased growth when planted in Calluna swards. Pioneerspecies such as pines and larches, did not suffer this “check” totheir growth (Weatherall, 1953).Fertilizer applications led to the belief that stagnation ofSitka spruce on heather—dominated sites was the result of directcompetition by heather for water and nutrients, especiallynitrogen (N) and phosphorus (P). But, the elimination of heatherdid not always alleviate the checked condition, particularly onsites deficient in N and P. Fertilization of these sites with Nand P resulted in a temporary growth response, but it soon becameapparent that further fertilization was necessary. Thus, theCalluna check was thought to be caused by more than just anutrient deficiency (Malcolm, 1975).Braathe (1950) suggested unfavourable soil conditions,excessive soil acidity, deficiency of trace elements in the soil,deficiency of mycorrhizal fungi and severe competition fromCalluna as being responsible for the inhibition of growth ofspruce on heathlands. He found it very difficult to believe thatthe soil could change so markedly in two to three years, or thatCalluna at the time of invasion, could monopolize all thenutrients so completely. He therefore concluded that Callunahad a biological effect on spruce and suggested that a substancewas produced which in some way inhibits the growth of spruce.Circumstantial evidence for allelopathy by the Ericaceae was19gathered in Scotland where it was noted that species sensitive toheather competition did not develop the branched mycorrhizal rootsystems typical of the normal condition of actively growingplants. This led to the speculation that some factor closelyassociated with the Calluna plant prevented the development ofthe ectomycorrhizal association in the trees. Handley (1963)found that aqueous extracts of mor from stands of vigorousCalluna could inhibit the growth of a range of mycorrhizaeforming fungus, whereas extracts from other soils of similaracidity and nutritional status did not. Some fungi, includingthose that formed mycorrhizal associations with pine, were notinhibited. The inhibition was less pronounced where the morsample came from shaded Calluna or was less acid. There seemedto be some variation in the resistance of fungi to the inhibitoryfactor, suggesting that some strains of fungus could form ecto—mycorrhizal associations with trees such as pines, that couldapparently tolerate the Calluna competition. The disappearanceof the factor inhibitory to the mycorrhizae when the Calluna wassuppressed, suggested that it must be continuously produced tomaintain an inhibitory level.Handley’s work did not achieve immediate acceptance, and aslate as 1970 researchers were still suggesting direct competitionof Calluna was inhibiting the growth of trees (Bjorkman, 1970).However, the ability of “heather—sensitive” species to competewith other vegetation, some of which is more demanding in itswater and nutrient demands than Calluna, would seem to relegate20direct competition to a somewhat lesser role.According to Read (1984), toxicity of heathiand soils occursas a result of the ability of Calluna to modify the soilenvironment in its favour. Interactions between the high organicacid content of humus, low pH, and low base status producephytotoxicity sufficient to exclude or debilitate most would-becompetitors. The success of ericaceous species therefore must beexamined in terms of all interacting effects.4.0 EFFECTS OF ERICACEOUS PLANTS4.1 AllelopathyThe word allelopathy was coined by Molisch (1937) todescribe the chemical interactions among all plants (microbes andhigher plants), including stimulator as well as inhibitoryinfluences. Typically, and in this case, the word is used todescribe only the harmful effects of one higher plant uponanother, since allelopathy translates literally as “mutualsuffering” (Putnam and Tang, 1986). It must be rememberedhowever, that many cases of allelopathy involve microbes eitherdirectly or indirectly, and some chemicals found to inhibitgrowth of the some species at certain concentrations maystimulate the growth of the same or different species at lowerconcentrations (Rice, 1984).Some of the earliest observations of allelopathy concernedharmful effects of crops upon other crops or weeds. Theophrastus(ca. 300 B.C.) observed that chick pea (Cicer arientum)21“exhausts” the ground and destroys the weeds. Pliny (PliniuaSecundis, 1 A.D.) not only reported that a number of cropsincluding chick pea, barley (Hordeum vulgare), and bitter vetch(Vicea ervilia) “scorch up” cornland, but he also recognizedtoxicity of walnut (Julans recfia) trees. He attributed thetoxicity of plants to their scents or juices and indicated thatbracken fern (Pteridium aguilinum) might even be controlled bybreaking young stalks and allowing “the juice trickling down outof the fern to itself kill the roots”.Two likely classes of compounds implicated in allelopathy ofericaceous plants are phenolic and aliphatic compounds. Harborneand Williams (1973) found that simple phenols were abundant inthe Ericaceae, particularly hydroxybenzoic, cinnamic, gentisic,vanillic, p—coumaric acid, and caffeic acids. Towers(1966) found p-hydroxybenzoic, o-pyrocatechuic, gentisic,protocatechuic, vanillic, syringic, p—coumaric, caffeic, ferulicand sinapic acids in hydrolyzates of ethanolic extracts of salal.Of the common flavonols, quercetin is found to occur in allspecies, while both kaempferol and myrcetin are of more limitedoccurrence. Myrcetin is found in more woody members of thefamily such as Rhododendroideae, Ericoideae, Vaccinoideae, andCassiopeae, while kaempferol occurs predominantly in the moreherbaceous members such as Pyroloideae and Monotropoideae(Harborne and Williams, 1973).Aliphatic compounds are also potential allelochemicalsbecause the fat content of ericaceous plants is extremely high,22and fatty acids have been shown to be a major lipid storageproduct (Tschager 1982). The chain length distributionof alkanes in the epicuticular wax of many Ericacae and closelyassociated Epacridaceae vary from C23 to C35, with C31 the majoralkane, and the odd—carbon chains exceeding the even—carbons(Salasoo, 1981; 1983a and b; 1987).The high concentrations of some organic acids in theEricaceae and/or the allelopathic interaction between ericaceousand non—ericaceous species has been examined for Erica scopariaL. and . australis L. in Spain (Ballester 1977;Carballeira,1980; Carballeira and Cuervo, 1980); forArctostaphylos glauca Lindl. and A. cflandulosa Lindl. in theCalifornia chaparral (Muller .,1968; Chou and Muller, 1972);for Calluna vulgaris in Scotland (Handley, 1963; Robinson, 1972;Jalal et al., 1982; Jalal and Read l983a and b); for Kalmiaangustifolia L. in Newfoundland (Mallik, 1987); and Emetrumhermaphroditum in Sweden (Zackrisson and Nilsson, 1989).The occurrence of phenolic or aliphatic compounds in livingtissue does not prove an allelopathic effect exists, or even thatthese compounds are released into the soil environment.Furthermore, the release of plant—produced phytotoxins andcorrelated toxic qualities of the environment does not mean thatthe original toxic product acts, in unchanged condition, as theallelopathic agent. Some factors of the environment that affectretention or alteration of allelochemicals include redoxpotential of the soil, its fixation on clay or humus, the23presence of metallic ions for chelation reactions, and thecomposition of the soil solution and atmosphere (Vaughan1983; Haider and Martin, 1975; Huang 1977). For example,levels of phenolic acids have been shown to vary seasonally insoils under Erica australis L. (Carballeira and Cuervo, 1980) andCalluna vulcaris (Jalal and Read, 1983a and b) ranging from verylow levels in late summer months to maximum levels of 0.12 mM inearly summer months. The seasonal nature of phenolic acidconcentration was felt to be associated with accumulation duringcold, wet winter months. At the outset of spring, increasingaeration and temperature lead to more rapid breakdown byincreased microbial and fungal growth and metabolism withincreased production of phenolics at the roots (Jalal1982). The fatty acids (hydroxyalkanoic acids) were found inquantities comparable with or greater than those of the aromaticmoieties. They felt that lipids reaching the soil wouldeventually be degraded by the /3 oxidation pathway into decanoic,nonanoic and octanoic acids along with lower fatty acids.The mechanisms by which organic acids act as allelochemicalsis still unknown, but some work has been done to determine thelikely mechanisms. Takijima (1964) found increasing toxicity ofaliphatic acids with increasing carbon chain length accompaniedby an increasing affinity for lipids. Similarly, Glass (1973)found the degree of inhibition of ion uptake in roots wascorrelated well with the lipid solubilities of phenolic acidstested. Further studies by Glass (1973; 1974; 1975; 1976) and24Glass and Dunlop (1974) showed that a likely mechanism forreduced ion uptake by phenolic acids is that the phenolic acidspartition themselves between the aqueous medium and the lipidcomponent of the cell according to their lipid solubilities. Themembrane then becomes permeable to both anions and cations. Thisresultant loss of ions rapidly depolarizes the membrane potentialby increasing the permeability coefficients of the ions and byreducing the imbalance of ion concentrations across the cellmembrane. The dysfunctioning of the plasma membrane then leadsto the failure of cells to maintain proper mineral nutrition.This could then lead to the inefficiency of the energy systems ofrespiration and photosynthesis in plants, which demand precisemembrane organization, charge separation, and the work ofmembrane—associated proteins.A somewhat similar effect on growth, as occurs with membranedysfunctioning, could occur with impairment of the establishmentof a mycorrhizal association by trees growing on these nutrientdeficient, acidic sites. Several studies have shown thefungitoxic effect of Calluna on mycorrhizal associations ofcompeting species. Handley’s (1963) experiments showed thatCalluna, or its ericoid mycorrhizae, were responsible for therelease of a factor inhibitory to the formation of ectotrophicmycorrhizal associations of competing trees. The prevention ofthe mycorrhizal association was due, not to lack of fungalinoculum, but to the prevention of growth and inoculation by thefungus. Robinson (1972) confirmed Handley’s observations. Fatty25acids of intermediate chain length (C6-C14) are potent inhibitorsof growth and respiration of micro—organisms (Franke andSchillinger, 1944; Wyss 1945). Both spore germinationand mycelial growth are affected. The sensitivity ofectomycorrhizal fungi like Suillus variegetus to octanoic andnonanoic acid (Pederson, 1970) might lead to their exclusion fromsoils and explain the widespread inhibition of the fungi reportedby Handley (1963).The inability to properly absorb nutrients at the membrane,or to form mycorrhizal associations, can then have far—reachingimplications for overall plant growth. Alterations of themineral content of plants subjected to nonspecific allelopathicconditions has been shown in many investigations (Rice, 1984),but it is difficult to generalize about changes in mineralcontent incurred from allelopathic interference. Phosphoruscontents are frequently reduced, while nitrogen, potassium andmagnesium uptake may be increased or decreased. Consistent withthese findings of nutrient imbalance is that tree species“sensitive” to ericaceous plants often show symptoms of N and Pdeficiency (Malcolm, 1975; Weetman j., 1989a,b). Fertilizerapplications have been shown to overcome allelochemical inducedgrowth suppression in laboratory experiments, as well as withfield experiments using “sensitive” trees (Malcolm, 1975).Soil or substrate fertility can affect the toxicity andrate of breakdown of phenolics; higher fertility leads to a morerapid breakdown and less toxic conditions. This was shown in a26study by Stowe and Osborn (1980) where phenolic toxicity appearedto depend intimately on nutrient concentrations; the phenolicacids were uniformly and significantly inhibitory only at lownutrient concentrations. Since phenolics are more likely to beproduced in a plant under stress (Rice, 1984) it appears thatallelopathy with phenolics is more likely in nutrient poor soils.4.2 Soil Acidification and PaludificationHeathland flora is generally indicative of oligotrophic,acidic soils, but more significantly, ericaceous plants mayactually contribute to the process of soil acidification, whichmay be one reason why they are capable of invading and dominatingmore complex vegetation types. Pollen analysis of heath soils byDimbleby (1962) has shown that the rise of dominance of Callunais closely linked with increasing soil acidification, thedisappearance of deep—burrowing earthworms and the subsequentaccumulation of raw humus. Similarly, Webley (1952)have shown marked reductions in bacteria and increases in fungiwhen a fixed Ammophila sand dune community was succeeded by adune heath dominated by Calluna. Grubb (1969), found astrong correlation between the size of Calluna bushes and thesoil pH beneath their centres; also between distance from thecentre and pH both at the soil surface and below. More directevidence for soil acidification by ericaceous plants was shown ina study with Calluna and Rhododendron, which dominate soils ofvery low pH in the range 3—4. When grown in sand with mineral27nutrient solution at pH 4.5, Calluna and Rhododendron acidifiedthe medium to below pH 4.0 in 8 weeks, and to 3.5 in a subsequent2 month period (Read, 1984). The increase in acidity was felt tobe due to production of organic acids, particularly from theoxidation of long-chain fatty acids, and to the depletion of thesoil base status.The presence of well developed heath vegetation isassociated with soil podsolization, in which sesquioxides areeluviated to lower mineral horizons (Soil Survey Staff, 1975).The solubility of iron, aluminum and manganese are greater underthe more acidic conditions, and the organic acids associated withheath vegetation then chelate with the metal ions and are leacheddown the soil profile to the lower horizons. In many casesiron-podsols form, in which the iron is deposited in the form ofa thin, hard iron pan (placic horizon or fragipan) leading to theconsequent rise in the perched water table, and ultimately inwidespread paludification (Damman, 1965; McKeague 1968).The formation of pans under heathiands has been noted in northEngland and Scotland under Calluna (Gimingham, 1972),inNewfoundland under Kalmia (McKeague .]., 1968), on westernVancouver Island under Gaultheria (Carter, 1988), and insoutheast Alaska under Vaccinium and Menziesia (Ugolini and Mann,1979). If such sites are left undisturbed, the high water tablecan lead to reductions in forest productivity and changes invegetation through accumulation of forest humus resulting inwetter, colder soils, and a reduction in tree rooting depths.284.3 Humus DecompositionAs previously discussed, oxidation products of fatty acidsare fungitoxic and can therefore affect decomposition in soilsrich in lipids and other fatty acids. A further effect of theapparent accumulation of organic acids under ericaceous plantsmay be the tanning effect of polyphenolic compounds on humus.Tanning is the process by which proteins are made resistant todecomposition through bonding with a polyphenolic molecule suchas hydrolysable tannins (tannic acid or gallotannins) orcondensed tannins (proanthocyanidins). Hydrogen bonds are formedbetween the hydroxyl groups of the tannin molecules and thecarbonyl groups of the protein—amide linkages and covalent bondsare formed between quinone residues and free amino groups inamino acids. Both of these reactions modify protein structure.The efficiency of tanning is associated with the moleculardimensions of the tanning agent, since the tannin has to form astable cross link with the protein molecule. Tannin moleculesbelow the critical size cannot form these cross links, and thoseabove the size will combine only at easily accessible outer sitesproducing a case—hardening or surface combination. It appearsthat the tannins formed between pH 3 and 5 are of about the rightmolecular size to afford adequate protection to the protein.Outside of this pH range, the molecular size is probably toogreat and the protein is not adequately protected (Gustaven,1956). These results suggest that acid conditions, such as those29associated with the mor sites, will favour the formation ofstable protein—tannin complexes so that mineralization of theprotein is delayed. Since as much as half of total nitrogen (N)in humic substances can be accounted for as amino acid—N(Stevenson, 1982), this delay in mineralization may result in Ndeficiencies for growing plants.Tannins can also slow the decomposition of nonproteins suchas cellulose and hemicellulose because the tannin—proteincomplexes coat and permeate cell walls, making them considerablyresistant to microbial attack (Benoit and Starkey, 1968a and b).Tannins have been found to be most tightly bound to cellulose atpH 1 to 5, below the isoelectric points of the proteins, and tendto dissociate above pH 5, above the isoelectric point (Benoit andStarkey, l968a).Tannins also affect decomposition through inactivation ofcertain enzymes important to the process of decomposition oflarge molecular weight compounds such as proteins, cellulose,hemicellulose and other polysaccharides and lipids. The degree ofcomplexing of an enzyme by tannin will be affected by thechemical properties of the tannins, the ratio of tannin toenzyme, the presence of substances that regenerate the enzymeactivity, the composition of the enzyme-protein, and the pH ofthe liquid in which they are contained (Benoit and Starkey,1968b)The mor humus forms associated with ericaceous species arecharacteristically acidic, deep and not conducive to rapid30decomposition. Not surprisingly then, these same sites show verylow rates of mineralization of nitrogen, phosphorus and sometimessulphur, consequently, there are overall deficiencies of theseimportant nutrients. It may be then, that the litter ofericaceous shrubs are rich in polyphenolics and long—chain fattyacids, which when released into the soil greatly reduce themineralization of organic nitrogen and phosphorus.5.0 MANAGEMENT OF SITES DOMINATED BY ERICACEOUS SPECIESFollowing clearcutting or burning without promptregeneration of conifers, it is common for old growth forestswith ericaceous plants already in the understorey, to becomeheath plant dominated cutovers. Some examples are: 1) Kalmiabarrens in Newfoundland following cutting of black spruce; 2)Vaccinium, Kalmia, and Ledum heaths in Nova Scotia or in theboreal forests following cutting of jack pine or black spruce onlow productivity soils, and 3) Gaultheria dominated cutoversfollowing clearcutting of old growth western red cedar forests oncoastal British Columbia.Once established it is very difficult to eradicate the heathvegetation. Herbicides are not usually effective or licensed foruse. Planting trees into dense heath cover is not feasible.Burning usually stimulates further heath plant sprouting andrenews their vigour. In some cases the dominance of heath plantscan be very long term and represent a permanent exclusion offorest cover, as seen in British heathiands and in Newfoundland31and Nova Scotia. For boreal and westcoast heathlands, evidencesuggests slow invasion by trees and eventual forestreestablishment.Evidence from coastal western hemlock forests in Alaskasuggests that periodic natural windthrow which uproots old treesand buries the humus layer, is an important factor in maintainingsoil fertility (Bowers, 1987). In some British heathlands wherepure Sitka spruce stands suffer growth check and nitrogenshortages, mixed spruce and pine or spruce and larch stands showvigorous growth and no shortage of nitrogen (Malcolm, 1975). Todate, there is no North American evidence for this “nurse crop”or mixed species effect.Reforestation of heathiands usually requires ploughing oruse of backhoes to physically turn over or rip out the heathplants, usually followed by nitrogen and phosphorus fertilizationprescribed on a site specific basis. Current work withchiorotic, slow growing Sitka spruce, western hemlock andwestern red cedar regeneration on Gaultheria dominated cutoversindicates that nitrogen and phosphorus provide an immediate buttemporary release of growth check, whether applied at time ofplanting or later.European experience with attempts to reforest establishedheathiands indicate it is difficult, slow and expensive ascompared to preventative measures. Therefore, the most appropriate actions are those designed to avoid establishment of heathplants such as prevention of fires on naturally regenerated32cutovers; seed bed preparation for prompt natural regeneration;and rapid planting, supplemented where necessary by fertilization.6.0 SUMMARYFrom the review of the literature, it is evident that theproblems of chiorosis and stagnation of conifer plantationsassociated with ericaceous plants, such as salal, are similarthroughout the northern hemisphere. The presence of ericaceousplants can greatly affect forest productivity through a number ofcomplex interacting mechanisms.Ericaceous plants have been found to be associated withorganic acids such as phenolic acids, polyphenolics and fattyacids in soils which can be phytotoxic to plants, probablythrough their effects on root membrane permeabilities ormycorrhizal infections, leading to a failure of cells to maintainadequate mineral nutrition, and the inefficiency of the energysystems of respiration and photosynthesis. The phenoliccompounds can polymerize and chelate with iron and aluminum insoils forming organo—metallic complexes which migrate andprecipitate in lower mineral horizons leading to podsolizationand possibly the formation of iron pans, thereby impedingdrainage and rooting. The polyphenolics can also form tannin—likecompounds with proteins and enzymes resulting in decreaseddecomposition and mineralization of organic material and theaccumulation of raw humus. The fatty acids are reduced by j3—33oxidation into octanoic, nonanoic and decanoic acids, which canbe both phytotoxic and fungitoxic. The overall result is adecrease in the productivity of forests.Land management practices which allow the establishment ofheathlands, results in expensive site specific efforts atrehabilitation, which are rarely fully effective. Therefore, themost appropriate actions are those designed to avoidestablishment of heath plants such as prevention of fires onnaturally regenerated cutovers, seed bed preparation for promptnatural regeneration, and rapid planting, supplemented wherenecessary by fertilization.Based on this literature review, it is evident that salalmay be using allelopathic mechanisms such as the production ofphenolic acids, polyphenolics such as tannins, or fatty acidswhich reduces tree growth and leads to an overall decrease inforest productivity. This hypothesis is investigated in thefollowing chapters.34Chapter IIISoil Classification and Characterizationof CH and HA Sites1.0 INTRODUCTIONThe growth—check of conifer regeneration on CH but not HAphases of the Thwia plicata- Tsuga heterophylla - Gaultheriashallon— Rhytidiadeiphus loreus or “salal — moss” ecosystem isthought to be related to the reinvasion of the cutover by salal.However, site conditions which encourage the growth of salal onCH but not HA may in itself contribute to the differences instand productivity. The objectives of this study were todocument some of the differences in physical and chemicalproperties of mineral soil and forest floor humus horizons and tosee if these properties differed between the CH and HA phases.It was anticipated that this could identify the factorsresponsible for the poor growth performance of coniferregeneration on the CH phase relative to the HA phase.352.0 METHODS2.1. Soil and Vegetation Description and CollectionFive sites each of the CH and HA phases were located nearPort McNeill on Western Forest Products Tree Farm License 6. Thesites were located such that an HA sites was adjacent to a CHsite and were thus paired. The sites were numbered such that oddnumbers were CII sites and even numbers were HA sites, thus pairedsites would be numbered 1 and 2, 3 and 4, et cetera.Preliminary investigations of the forest floor horizons onCII and HA sites indicated that the organic materials were diversein origin and state of decomposition which could cause inherentdifferences in chemical composition, biological activity andphysical properties of the various forest floor horizons.Traditional methods of forest floor sampling involving the use ofbulk samples could consequently cause results to be misleading orinsignificant, unless prohibitively large numbers of samples werecollected. It was therefore decided to work within defined humushorizons which would provide a more homogenous sample base thanby random bulk sampling, and would also allow more effectivecommunication with other investigators working on similarmaterials.Forest floor horizons were initially examined along two soiltransects through typical CII (Rupert 206) and HA (Rupert 400)sites. The transects were run 15 m in a direction in which therewere no obvious changes in elevation. Salal and other brush wasremoved and the transect dug to a width of about 1 m and a depth36of about 1 to 2 m. The forest floor and mineral soil horizonswere examined macroscopically for colour and texture, thenrepresentative samples of forest floor were returned to thelaboratory for further analysis. Air dry samples were againexamined for colour and texture, then oven dried and ashed.Obvious similarities and differences which could be used forquick and easy field identification were felt to be important tothe classification. The publication by Green (1991) onhumus form classification indicated ways of reporting moreinformative horizon designations and the approach was adopted.Once the horizon classification was determined, samples werecollected for further analysis.Within each of the five previously chosen CH and HA stands,a 30 m transect was run through an area that appeared homogeneousin terms of topography and vegetation. Along this transect, 10plots were located 3 m apart, and soil pits were dug through theforest floor and mineral soil to a depth where roots were absent,usually to a root restricting layer. Vegetation within a 2 mradius was noted (as described in the previous section). Forestfloor and mineral soil horizons were classified and measured fordepth, with estimates made of abundance and size of roots and thecoarse fragment content of the mineral soil at depths of 0 to 10cm and 10 to 20 cm.From each site, one or two samples best representing eachhumus type were collected in plastic bags and kept in arefrigerated truck until transported back to the laboratory where37they were stored at 4°C until processed. Each sample was sievedat field moisture through first an 80 mm sieve to remove largeroots and wood pieces, then through a 40 mm sieve to break upsmall pieces of wood and remove fine roots. Samples were halved,with one half frozen at -30°C for phenolic acid determination,and the other half air dried, ground to pass a 0.4 mm screen andstored at room temperature in sealed plastic containers.2.2 Laboratory Methods2.2.1 Nutrient analysisThe nutrient analysis work was done at the U.B.C. Departmentof Soil Science in the Soil Chemistry Laboratory unless otherwisenoted. All nutrient concentrations are expressed on an oven—dry,ash—free basis. Soil moisture content was measured as thedifference between field moist and oven-dry weight after dryingovernight at 105°C. Ash content was measured on oven—dried soilin a Thermolyne Furnatrol I muffle furnace. The temperature wasset at 200°C for one hour, then raised to 450°C for three hours.Total carbon was measured by a Leco Induction Furnace (model521) and Leco Carbon Analyzer (model 572-200). Total nitrogenwas determined by a semimicro—Kjeldahl procedure (Bremner andMulvaney, 1982).Total sulphur was determined with a Fisher High TemperatureFurnace and Sulphur Analyzer (Models 472 and 475). Availablesulphate S was estimated colorimetrically after HI—reduction of a38CaCl2 extract, as described by Kowalenko and Lowe (1972).Mineralizable and available N were analyzed by Pacific SoilsAnalysis Incorporated using an anaerobic incubation at 30°C for14 days (Waring and Bremner, 1964), then measuring the NH4-N onan autoanalyzer.Available P was analyzed at Pacific Soils AnalysisIncorporated using the P extraction method of Mehlich (1978), asdescribed by Lavkulich (1982).The exchangeable cations calcium (Ca), magnesium (Mg),potassium (K) and manganese were analyzed by Pacific SoilsAnalysis Incorporated using 1 M ammonium acetate extraction at pH7 followed by atomic absorption spectrometry (Lavkulich, 1982).2.2.2 Polysaccharides and celluloseTotal and labile polysaccharides were estimated using thephenol—sulphuric acid procedure of Dubois 1. (1956),following acid hydrolysis. For labile polysaccharides,hydrolysis was carried out by autoclaving for 1 hour at 15 PSI(103 KPa). For total polysaccharides, hydrolysis involved coldtreatment with 72% H2S04, which was subsequently diluted to 0.5 Mand then autoclaved as for labile polysaccharides (Ivarson andSowden, 1962; Cheshire, 1979). Cellulose was calculated as thedifference between the total and labile polysaccharides.2.2.3 LipidsLipids were measured at the U.B.C. Soil Chemistry Laboratory39by shaking 5 g of sample with 75 ml of 1:1 ethanol-benzene for 2hours and suction filtering. The leachate was transferred to atared 250 ml beaker, evaporated in a fume hood, and the residueweighed (Lowe, 1974).2.2.4 Bound phenolic acidsBound phenolic acids were measured at the U.B.C. SoilChemistry Laboratory using an alkaline hydrolysis. Air-driedsoils were shaken overnight in 1 N NaOH. The extract wasfiltered and then acidified to pH 2 to precipitate humic acidsand to fully protonate the phenolic acids. The protonatedphenolic acids were then extracted in diethyl ether, which wasallowed to evaporate, and the residue taken up in 3 ml ofmethanol for HPLC analysis. Prior to injection into the HPLC,samples were filtered through a 0.2 micrometer Prep-Disc Filter(Biorad). Chromatography was done based on the procedure ofDrijber and Lowe (1991) using an acetic acid/acetonitrilegradient and chromatographic conditions as follows:Time % 1% Acetic % AcetonitrileAcid0 92 811 92 816 86 1430 86 1433 40 6035 40 6040 92 843 92 840Flow = 1.5 mi/mmWavelength = 280 nmTemperature = Room temperatureAttenuation = 16 (bound phenolics)Chart speed = 0.5 cm/mmDetector Sensitivity = 0.01Standard = 30 ppm (bound phenolics)Peak Threshold = 22Peak Width = 6Column = ODS Spheri-5 25 cm (Brownlee)Guard Pak = p Bondapak C18 Inserts (Waters)Solvents = Acetic acid Aristar (BDH)Acetonitrile Omnisolv (BDH)Standards = Protocatechuic, p-Hydroxybenzoic, Vanillic, Syringic,Trans—p—coumaric, Ferulic (Sigma)2.2.5 Statistical analysisAll statistical analyses were done using SYSTAT System forStatistics (Wilkinson, 1990). Bartlett’s test was run to ensurehomogeneity of variances, and a log transformation done whennecessary. Statistical tests involved analysis of variance,Tukey’s HSD test, covariance analysis and discriminant analysis.3.0 RESULTS AND DISCUSSION3.1 Site DescriptionSample sites are described in Table 3.1. The elevationsof the site were similar, between 90 and 110 metres. The HAsites tended to be on ridgetops with 0 percent slope and aspect,while the CH sites were immediately adjacent on lower slopes,ranging from 11 to 20% slopes. The CH site therefore, tended tobe wetter than the HA.The site characteristics were taken from inventory mapsbelonging to Western Forest Products. The CH stands werecharacteristically old, with estimates as great as 1000 years41Table3.1:SiteLocationandCharacteristicsSite#LocationForestElev.SlopeAspectAgeHeightDensitySiteType(ft)(%)(deg)(yrs)(m)Class1SCHIRPC(H)85110141+40.1-50NormalMedium2SCHIRPBH988458230.1-40DenseMedium3RupertC(H)672020141+40.1—50NormalMedium4004RupertHB854458230.1-40DenseGood4005RupertC(H)980141+30.1—40NormalMedium2066RupertBH107081-10030.1—40DenseMedium2207MistyCH8811330141+40.1—50NormalMedium1008MistyHB910101—12030.1—40DenseMedium1009MistyCH8518350141+40.1—50NormalMedium20010MistyHB900141+40.1—50NormalMedium200*Indicatestherelativeproportionofspecieswithinthestand:westernredcedar(C),westernhemlock(H)andamabalisfir(B).42since last disturbance (Lewis, 1982). Dominant tree heights wereabove 40 in with the exception of stand 5 (30 m). The HA standswere younger (80 to 120 years old) and denser. Site class wasmedium for all stands except stand 4, which was rated as good.According to the Western Forest Products Inventory map, three ofthe CII stands had a greater proportion of western red cedar thanwestern hemlock, and the other two had equal amounts of each.The HA stands tended to have equal amounts of western hemlock andamabilis fir, with two stands having a predominance of amabilisfir, and the other three of western hemlock. All sites werecharacterized as having moderately well to imperfectly drainedDuric Humo—ferric Podzols arising from a sandy—loam glacial tillwith a blanket and rolling surface expression.Differences in vegetation between CII and HA sites wereobvious, both in the field and statistically. The CH stands weremore open and had abundant understorey vegetation, particularlysalal (occurring on 98% of the plots at a surface area of 66%),while the HA stands were dense with very little understoreyvegetation (Table 3.2). Vaccinium pp. (including 31. alaskaenseHowell and 31. ovalifolium Smith) were present with similarfrequency and abundance on both sites, as were the mossesRhytidiadeiphus loreus (Hedw.) Warnst and Kindbergia oregana(Sull.) Ochyra. In addition to salal, the only other constantvegetation on the CII site was deer fern (Blechnum spicant (L)With) and the moss Hylocomium splendens (Hedw.) B.S.G. The HA,because of its dense stand structure had sparse shrub cover43Table 3.2: vegetation (% presence and % cover on those plotswhere present) by stand number.Stand Number and PhaseCHSpeciesGaultheria shallonVaccinium pp.HA1 3 5 7 9100 100 100 100 9068 19 95 3035__________60 70 604 5 3Menziesia ferruginea 0 10 100 3 2Rubus spectabalis 50 0 2010 0 5Blechnum spicant 80 70 3013 8 17Cornus canadensis 30 40 401 2 5Tiarella trifoliata 0 10 00 1 0Lysichiton americanum 10 30 02 2 0Montia siberica 0 10 00 1 0100 401340 02 010 02 010 301 210 101 130 01 00 00 00 100 12 450 03 050 701 20 00 00 00 030 2011 20 00 00 00 00 00 00 00 070 4033 3550 10010 516 8 100 10 200 1 160 60 603 3 110 0 00 0 00 0 00 0 020 10 03 2 00 0 00 0 00 0 00 0 00 0 00 0 00 0 00 0 050 100 1001 18 2350 100 1007 23 30Hylocomium splendens 100 100 100 100 9046 30 25 54 42Kindberctia orecrana 100 60 60 80 5032 19 19 31 66Rhytidiadeiphus loreus 30 60 60 80 5023 13 40 15 1790 90 10 70 6033 21 10 6 2844including salal, Vaccinium pp. and occasional Rubus spectabilisPursh.3.2 Humus Classification and Variability3.2.1 Hi.unus horizonsIn the context of this paper, the term “humus” is used todescribe any of the organic horizons of the forest floor. Theforest floor was divided into master horizons based on the degreeof decomposition, as described by Green . ., (1991). Litterhorizons (L), which were generally very thin (< 1 cm) andconsisting of the freshly fallen debris of the surroundingvegetation, were removed prior to sample collection of underlyinghorizons.The F and H master horizons were subdivided into 2 broadcategories based on their woody or non—woody nature. Those withwood were given the suffix “w”. The woody horizons included anFw, in which the woody structure held when rubbed between thefingers; an Hrw (residuic), in which the woody structure failedwhen rubbed between two fingers, but consisting of greater than20% woody materials; and an Hw consisting of less than 20% woodymaterial. Simple field tests that further distinguished the twohumic horizons include the appearance of dark coloured, greasymaterials that rubbed out on fingers for Hw but not for Hrw; thereddish colour of Hrw versus the brownish red colour of Hw; andthe more massive and compact structure of Hw (Table 3.3).The non—woody horizons consisted of 3 types. A matted Fm45Table 3.3: Physical Descriptions of Organic Soil Horizons onCH and HA Sites.Horizon Composition Colour Structure RootinciFm >60% plant 1OR 3-4/4-6 compact, abundant<20% amorphous 2.5YR 3/4—6 matted fine to>20% fungi 5YR 2.5-3 coarseFw >90% wood 1OR 3/4 woody few2.5YR 3/4-6 structure5YR 3/4—6 holdsHrw <80% wood 1OR 2.5—3 woody few>20% amorphous 2.5YR 1—2 structure5YR 1—2 failsHw <20% wood 1OR 2.5—3 crumbly plentiful>80% amorphous 2.5YR 1-2 greasy to5YR 1-2 abundantHh no wood 1OR 2.5-3 massive, plentiful>80% amorphous 2.5YR 1-2 blocky to5YR 1-2 greasy abundantHhi >95% amorphous 5YR 2.5-5 massive, very few7.5 YR 0-1 very greasyblocky to finegranular46(mycogenous) horizon, containing abundant fungal hyphae and plantroots; a well decomposed Hh (humified) horizon, which was greaterthan 80% amorphous, with a massive structure, greasy texture anda dark colour; and a Hhi, a very massive, very greasy, blackhorizon, greater than 95% amorphous, containing intermixedmineral particles (17—35% organic carbon mass), found immediatelyabove the mineral soil.3.2.2 HUmUS profilesHumus profiles along the two trenches from the CH and HAsites were classified according to Green.j,. (1991) and areillustrated in Figures 3.1 for CH and 3.2 for HA. Estimating theproportions of various humus forms was done by simply dividingthe length of humus type along each trench by the total length oftrench; results are shown in Table 3.4.The HA trench consisted of 37% Hemimors (specificallyVelohemimors), 22% Huntimors (8% Orthihumimors and 14%Melahumimors), and 41% Lignomors (26% Hemilignomors and 14%Orthilignomors). The presence of Hemimors, in which the Fmhorizon comprises greater than 50% of the combined thickness ofthe F and H horizons, occurs over recently windthrown mixtures oforganic and mineral soil. The Lignomors consist of greater than35% by volume of decaying wood, and are indicative of a windthrowevent. The Humimors have a well developed H horizon reflectingmaturity and a relative lack of disturbance.The CH trench has a greater diversity of humus,MELAH[.ORTHILIGNOMORLitterLEGENDWoodyFwCHTrenchProfile:O-6mNon-woodyFmHh IHrwHw4811109876IIUGNOHUMIMORMELAHUMIM0PLIGNOHUMIMORMELAHUMIMORBf BfgLitterE I’)] If) aLEGEND5043Figure3.2a.HAtrenchprofilefrom0to4m.E to 0LEGENDWoodyNon-woodyFwFm2VELOHEMIMORORTHIHUMIMORHEMILIGNOMORLitter0SBhfu_______Bfh2________Bfh3_________Bfh4_______Htib AhebBf—51987654E I’) 0HEMILIGNOMORVELOHEMIMORAh-S-S-S-S-S-S5-LitterBfBhfuLEGENDFigure3.2b.HAtrenchprofilefrom4to9in.WoodyNon-woodyFwFmHrwHh5214131211Figure3.2c.HAtrenchprofilefrom9to14in.10Hw9ORTHILIGNOMORMELAHUMIMORHEMILIGNOMORLitterBfE LO dLEGENDWoodyNon-woodyFwFmHrw53Table 3.4: Proportion of humus forms on CH and HA trenchprofiles.ORDER SITESubgroup CH HAHEMIMORVelohemimor 0 37.4Melahemimor 2.3 0HUMIMOROrthihumimor 18.8 7.6Melahumimor 9.2 14.2Lignohumimor 28.7 0LIGNOMOROrthilignomor 17.3 14.4Hemilignomor 0 26.4HYDROMOROrthihydromor 7.6 0Hemihydromor 4.5 0Lignohydromor 11.7 054including 2% Hemimors (Melahemimors), 57% Humimors (19%Orthihumimors, 9% Melahumimors, and 29% Lignohumimors), 17%Lignomors (Orthilignomors), and 24% Hydromors (5% Hemihydromors,7% Orthihydromors and 12% Lignohydromors). In comparison to theHA trench, the CH has had few recent disturbances, as indicatedby the relatively large proportion of Humimors and smallproportion of Hemimors and Lignomors. The Hydromors, which werenot found in the HA trench, develop under the influence offluctuating,stagnant water that is generally less than 50 cmbelow the ground surface, in this case over Gleysolic soils. Acloser examination indicated that the gleyed soil occurred over ahighly cemented mineral horizon, which seemed to impede drainage.3.2.3 Soil variabilityTo determine the variability between CH and HA sites, theten plots from each of the five CH or HA stands were combined togive a total of 50 plots for each of the CH and HA sites.Characteristics were averaged based on occurrence.The soils on the HA sites were found to be drier and lesscompacted than on the CH. For example, 64% of CH plots hadstanding water occurring at an average depth of 23 cm, while forthe HA sites, only 16% of the plots had standing water, occurringat an average depth of 24 cm (Table 3.5). Root restrictinglayers, such as cemented or compacted horizons or standing water,occurred in 98% of the CH plots, but in only 70% of the HA plots.The mean rooting depth in the CII was thus significantly shallower55Table 3.5: Site property values including mean, standarddeviation and occurrence (% of plots) on the CH andHA sites.Site TypeProperty CH HASeepage Depth (cm) 22.6 24.0(15.1) (4.4)64 16Rooting Depth (cm) 21.4 28.8(19.6) (18.0)100 100Root Restricting 21.0 21.3Layer Depth (cm) (16.5) (14.7)98 70% Coarse Fragment 19 17Content (0—10cm) (14) (17)% Coarse Fragment 51 32Content (10—20cm) (25) (25)Total Forest Floor 23.0 28.8Depth (cm) (13.6) (16.8)100 100Follisol Depth 48.3 56.5(9.4) (10.9)7 1156(21 cm) than the HA (28 cm). Also, the coarse fragment contentof the CH was significantly higher in the 10 to 20 cm depth thanthe HA.The overall depth of humus was greater for the HA (29 cm) thanthe CH (23 cm), and this was because the HA had a greaterquantity of decaying wood than CH (Table 3.6). This contradictsthe findings of Lewis (1982). The least decomposed woodyhorizons (Fw and Hrw), occurred in 68% of the HA plots, with amean depth of 20 cm, while they occurred in 38% of the CII plotswith a mean depth of 19 cm. The well decomposed 11w was found in56% of the HA plots (16 cm depth), but only 38% of the CH plots(15 cm depth). The rionwoody Fm and Hh were found in virtuallyall of the CH and HA plots, but the depth on the HA (6 cm) wassignificantly less than on the CH (9 cm). The well humified Hhioccurred on 32% and 28% of the CH and HA plots respectively, butthe depth on the CH was significantly greater than the HA (5.2 cmand 2.5 cm respectively). Folisols (forest floor depths greaterthan 40 cm) occurred on 22% of the HA plots (57 cm depth), and on14% of the CH plots (48 cm depth). Overall, the HA tended tohave deeper forest floors with significantly more woody huiuasthan the CH, which had significantly more non-woody humus. Thehigh occurrence of rotting wood on the HA reflects the relativelyrecent catastrophic windthrow event of 1908. The woody horizonsfound on the CH reflect single tree blowdowns.The mineral soil horizons were examined superficially andhorizons designated by experience using colour and texture; no57Table 3.6: Forest floor horizon depths including mean, standarddeviation and occurrence (%) by site.Horizon CH HAFm—Hh 8.7 6.3(4.5) (3.3)98 92Hhi 5.2 2.5(3.1) (1.3)32 28Fw—Hrw 18.7 19.9(15.2) (14.2)38 68Hw 15.0 15.6(8.8) (9.6)38 5658chemical analyses were done. Several interesting observationswere made (Table 3.7). The CH plots tended to have deeper andmore consistent occurrence of A horizons (including Ah, Ae andAhe), probably from lack of disturbance. The HA plots had agreater occurrence of organic enrichment in the B horizons,probably burial of organic horizons during windthrow. The CH hada much higher occurrence of gleyed horizons than the HA. The HAtended to have much more friable mineral soils, while the CH wasmore compact.These general observations along with the more empiricalinformation seems to indicate that the windthrow process is veryimportant to the rejuvenation and higher productivity of HAsites. The windthrow process not only breaks up pans and othercemented horizons, but also mixes mineral soil with humus andwoody materials which aerates the soil, increases the friability,and encourages deeper rooting by enriching the mineral soilfertility. In contrast, the CH site, which have had very littledisturbance, have almost continuous root restricting layers and ahigh occurrence of standing water over some portion of the year.The formation of Hydromors, occurring under periodic anaerobicconditions, is associated with accumulation of organic compoundsduring the hydromorphic period. When the profile dries out,these organic acids migrate down and behave as active agents ofpodzolisation and complex formation (Bloomfield, 1975). Thus,without a major disturbance event, such as windthrow, CH siteswould tend to degrade further with a greater occurrence and59Table 3.7: Mineral soil horizon depth (cm), standarddeviation, and occurrence (% of plots) by sitetype. (+ indicates greater than).Horizon Site TypeCH HAAh 4.5 2.9(2.7) (1.3)17 19Ahe 6.7 4.2(3.8) (2.3)37 36Ae 2.6 0(1.5)9Bh 9.5 7.8(8.1) (4.8)4 9Bfh 14.3 12.1(13.5) (11.0)28 36Bf 11+ 15+18 20Bfgh 14.1 14.0(8.4) (2.8)11 2Bfg 14+ 11+11 2Bg 15+ 14+8 860build—up of Hhi humus horizons, cemented horizons, and standingwater.3.3 Nutrient Concentrations of Humus Horizons3.3.1 Nutrient concentrations of woody horizonsWoody horizons develop from logs decomposing in place on theforest floor, with a gradual build—up of forest floor over them.There is virtually no mixing of woody material with non-woodymaterial until they approach a more decomposed stage, when theybegin to turn dark and lose the characteristic red colour. TheFw and Hrw therefore have very low ash contents and very lowquantities of total and available nutrients (Table 3.8). Asdecomposition proceeds, there tends to be an accumulation ofnutrients, presumably from decomposer organisms, and mixing withnon—woody and inorganic materials, so that the ash contentincreases significantly, as does total N, total S, mineralizableN, and available N. As expected, C/N ratios decreasedsignificantly with decomposition within the woody humus types,decreasing by almost half from Fw to Hrw, and again from Hrw to11w. This occurs because of a reduction in total C and asimultaneous accumulation of total N, which almost doubles fromFw to Hrw, and again from Hrw to Hw. This decrease in C/N ratiowith decomposition is well documented. According to Stevenson(1981), the decay of organic residues by soil organisms leads tothe incorporation of part of the C into microbial tissue, withthe remainder being liberated as CO2. At the same time, organic61Table 3.8: Mean nutrient concentrations and standard deviationby horizon on oven—dry and ash-free basis. Similarletters indicate no significant difference betweenall horizons.Woody Non-WoodyFw Hrw Hw Fm Hh HhipH 3.67ab 3.61a 3.56a 3.92ab 3.59a 4.16b(H20) (.07) (0.11) (0.11) (0.51) (0.15) (0.42)ash 0.9a 1.3a 2.9b 4.8b 5.7b 34.Oc(%) (0.3) (0.7) (1.0) (3.0) (4.1) (13.2)C* 56.5c 56.3bc 54.8bc 50.8bc 53.2bc 60.3a(%) (2.2) (2.4) (0.7) (0.9) (1.1) (19.2)N 0.286a 0.479b 0.876c 1.115c 1.lOlc 1.752d(%) (0.130) (0.171) (0.217) (0.202) (0.142)(0.324)S* 665a 920b 1585c 1649c 1766c 3132d(ppm) (157) (180) (254) (256) (140) (910)C/N 226d 127c 66b 47ab 49b 35a(77) (32) (17) (10) (8) (9)Avail. N 23a 29a 34a 61c 61c 61c(ppm) (5) (12) (11) (13) (31) (19)Avail. P 8a 9a 9a 33c 18b 9ab(ppm) (3) (2) (4) (15) (9) (2)Avail. S 23a 25a 27a 64b Sob 44b(ppm) (10) (7) (5) (26) (16) (23)Exch. Ca* 6a 8a 8a 13a 13a ha(me/bog) (2) (6) (5) (9) (7) (11)Exch. Mg* 3a 6ab 7ab 4a 9b 6ab(me/bOg) (1.2) (3.3) (3.5) (0.7) (2.9) (5.5)Exch. K* O.4a 0.6a 0.6a 2.3c 1.2b O.7ab(me/bOg) (0.2) (0.4) (0.2) (0.9) (0.4) (0.3)Extr. Mn 15a 22a 4a 207b 21a 14a(ppm) (11.7) (43.8) (4.9) (201) (17.0) (18.9)* Variances equal after log transformation.62N is converted to available N (NH3 and NO3-) which soil organismsutilize for the synthesis of new cells. Also, available N can befixed into humic substances. The result is a gradualtransformation of plant material into stable organic matter witha fairly consistent C/N ratio.Similarly, total S accumulates significantly withdecomposition, from an average of 665 ppm in the least decomposed(Fw) to 1585 ppm in the most decomposed (Hw). Total N and Sconcentration are not significantly different between the Hwhorizon and the Hh horizon, indicating that N and Sconcentrations in woody materials accumulate as decompositionproceeds, towards that of well decomposed non—woody materials.The available nutrients N, P and 5, were not significantlydifferent between woody horizons. Similarly, the inorganicnutrients, Ca, Mg and K were not significantly different, butthere appeared to be a trend towards increasing concentrationwith decompostion. Manganese was extremely variable.3.3.2 Nutrient concentration of non-woody horizonsThe non—woody horizons are formed from the input of litterto the forest floor, including leaves, roots and fruits of theplants growing in the immediate area. Nutrients are moreconcentrated in litter material than in wood, and the horizonsderived from litter therefore, have significantly greater totalN, total S, and available N, P and S than the woody horizons.The relationships between the non—woody horizons are morecomplicated than between woody horizons. For example, the ashcontent of the Hhi is very much higher than that of either the Fm63or the Hd, presumably because of the long residence time and theclose proximity of the Hhi horizon to the mineral soil.Therefore, on an oven—dry ash—free basis, the concentrations ofC, N and S appear to be significantly greater than that of the Fmand Hd (Table 3.8), whereas, on an oven—dry basis alone, theconcentrations of C are significantly lower, and that of N and Snot significantly different (not shown). Similarly, theavailable nutrients N, P,and S are not significantly different onan ash—free basis, but are significantly lower on an ash basis(not shown).There were no significant differences between the Fm and Hdhorizons for total C, N, S, and available N and S, but availableP was significantly lower for the more decomposed Hd. Althoughnot significant, there appeared to be a trend towards decreasingconcentrations of available N, P and S from Fm to Hd to Hhi. Alarger number of samples would likely improve the significance,consistent with natural trends in decomposition towards moreresistant, less available nutrient forms.For the inorganic nutrients, the only significantdifferences in the non—woody horizons were that Mg wassignificantly higher in the Hd, K significantly lower for theHhi, and Mn significantly higher for the Fm.3.3.3 Nutrient concentration differences betweensitesThe nutrient concentrations for each horizon by site areshown in Table 3.9 and Figures 3.3 to 3.8, and the Analysis ofVariance probabilities for horizon, site and horizon X siteinteraction are shown in Table 3.10. As previously described,64significant differences between horizons were found for all thenutrients except Ca. Generally, the non—woody horizons weresignificantly higher in nutrients than the woody horizons.Significant interactions between site and horizon occurredfor total N, total S and C/N ratio (Table 3.10). The 11w horizonhad significantly greater total N and S and significantly lowerC/N ratio on the HA site than the CH, but there was nosignificant difference between sites for the other horizons.The site effect and the horizon X site interactions weresignificant for the available nutrients N, P and S. The Hh andthe 11w horizons from the HA site had significantly greateravailable N than the CH sites, and the Hrw and the Hhi horizonshad greater available N, although not significantly so. Neitherof the F horizons were significantly different in available Nbetween sites, indicating that mineralization tends to be lowerin more humified horizons on the CH site than on the HA.The HA site had significantly greater available P in theHrw, Hw, and Hhi horizons than in similar horizons on the CHsite. There were no significant site differences for the woodyor non—woody F horizons, or the Hh. Again, this seems toindicate that P is less available in more humified horizons ofthe CH than of the HA. Turnover of organic P is a biologicalprocess, and mineralization and immobilization are stronglyinfluenced by the physical and chemical properties of the soil.Differences therefore, must be a function of site.Finally, the CH site has significantly greater Savailability in the Fm and Hrw horizons; greater, but notsignificantly greater, available S on the Fw and Hhi horizons;65.—.Cl)C)—.Z—.c)E——9)O\O\009)thO(flZ—rIr1(D05—(DIImc-Iw(5)(5)C)C)C)C)C)0-.)Ists)01—..H.1.--. 001—..001—010’.—’0101—.01—.01.--.01%01—.W01—sWpHO01—)—JO0’O•H’—J)01.001000000’0.001OWI-J’.D•)•W•.OW•.•.•01•01—OHH.H.01—J‘--‘H-.3D0-sJ009)9)01001—9)9)HW01i-’•b019)9)‘--‘‘-9)9)oHitflt—P)‘--‘9)9)9)—H01—.H.—.I--01—.04---s0t’.J—01—-01)-..W.i—.0.t’.—H)_)tIJC)I0rI’•PI1hHi)0s).P.WH0.••HUlW-.J01WHO•0.0.0.II00100100IH•)01•.0101aIM.0•-J.01•a0014’‘-‘00M—)MM0-..)‘-“M’..O—.301001HOlmP’9)9)-.101010H‘9)9)010‘.DH9).-9)9)—9)‘—Os•_9)•-‘9)9)HiUI(DOH01—.01-.3--.—.301—. H—.0(..).—. 0101—01W—sW01.—..W.—.M01.MW”sW01.-.Wmi.•toH4’-H.t-H010’0.0.t’00101.01M0.H•0’0•01W001•0-.1.-.)•.•.01WMM.0.0••01(D.bø0101‘—9)‘—HWHOl00—5--01MMD0—)OW015--’00101M.09)b—10100‘-‘-mtiioop’01—. H01—..1’-O.—..01—sH—-He—.0101—’-0101—-.M01—’.W—.W01—.01.—‘.W01—-.ib.II(-I0(DI-’•I-’HOl.D01H00.O•HH00W01‘—JH0.W•O•0.H•H01•O•.WW01.0•0•H.01•Mo9)‘—‘9)MCIM-.3MO—J’-J—1010100.-.1mow001H(..)009)b01—.3(DPI9)0H‘-9)—i))‘-9)—‘b‘--‘9)—ba“‘-(DW.P---’.HM—s.‘—-.01M—-’.H.----.H01—.01i.-’.0101--sW01--s—I01--sw..—.wto I’-I-P1H—)WOl.DH0.0•01..)HI.)MHHOl-.1•M•O•O•0‘H•0••H—.)00)‘-.1•01•.P..019)9)00)HalW0I.O—0)0101HO001HOlI—JOOIH09)‘—‘9)HHHI.)9)00‘—‘9)9)‘—‘9)‘-‘9)‘--“9)‘--‘9)0—9)—‘9)001—’.I.)W—-’.W01—.WW—. H01—..H01.--.01-.1.—..0101.--. H—.1.--..W01—.I.)-.1—-.I.)01.—.1..)-4-’..DI-’•(DHW01WHOlO•O•HWMiW.0Ht-I-HWO•O•O1.D•CI•0)••)01M0101.•—1•eI-’01019)‘--‘9)HOlWO0)0)•C’.)—.1•‘.0•01H’—JWO01‘.0--.)1.0’.OH.9)‘—‘ba—.3W01H•00‘--‘9)WI.)‘--‘9)“-‘9)..n....OVP0II-•‘-m1W1<00—‘.%Ci)CDCDCDCDI-’CDI-’—o00(1)1‘dlZIoo0UC):BC)BC)C)C)C)C)U1.—..l-.—.W01—%W01—SO01—OU1—.M01.—-J.-..M01..—.M2I---I—l-0•O•MMOl-GiWMI-OOWlI-I-“W0)0)•W•—0))M-‘Ci)‘—Ci)i-.—0)‘—0)00J-.JCi)-)))DM‘-0).—p)..0 øi01..WU101.•-J01.O.-0.-..01U1E\).-W(flI01-.W-sMU1..O01MM01O•0•MI-IOiOMOOMI-OOWl01Ci)Ci)•W•Ci)—)-0M-Ci)—‘Q)i—.oi-a’Ci)Ci)Ci)‘—0)00Ci)‘—a)-b01—.O-W(7-.01‘)D010sO01O—O01MM01.---3—.J01‘.JsM(fl-M0•0•MW’.D0I0IMMWI-01W-JM-..1—Q)-Ci)‘—Ci)P3••Ci)Ci)—0)0Ci)0)0)OWOz-9)‘—Ci).—Oi.—.W01—O01—.l-0—sWUI—.--)O.-sl-sOPs.sM-se.-sOi-sOi01.-.I-I--JtJ.S010.•.0)-Dl-O’iMMI-O5i0‘.Ol-l-MM30O‘—P)—Ci)•Ui010•Ci)000..J•-_P)0WUl‘.DO’iO‘-P3‘-‘—Ci)—0)0)‘—0)‘o—01 —Ci)Ci)•O•.MOl-Ml-OU1OI—JO01l-WOW0))•‘..O•M—iOI-0I-’l—‘—Ci‘-Ci)M’.D.—3Ci)Q)—Ci)‘—Ci)Ci)—0)0)“—‘UiWCi)—‘Ci)“—“Ci)Ci)W.’-UiU1.”-sMW..s4sU1sOW-001OW.’-sOU1-s)-W.sMU1s-’.UlW-s(JI.-sOZW...O’iU1s01WMMOl——30•0.l-GMips)—‘MMOOMO.PSOICi)WCi)Ci)•01•-.30)WWCi)0‘.0‘—Ci)I-’--iWU1“—‘a)“—‘P3‘—“Ci)Ci)‘—“0)MWMC”i‘-‘Ci)•-“CiI-’-.-I-J.I-I.0) p)zr1(Dr1‘lCD(D’l00(DObrtmmctrtCD(DO ZH0‘1N01<‘-I.rtCD*0r1CDC’)ICICCoI0ICIC0I000-‘CDC) 0CDI.IjI-’.C‘1CD00)0)00)0)VC,C,0C) CDD-‘0ci)o(110oo0--z0•TJHCDO\ 00.•rt0I-’-C)00)Ob0)(DO‘10CDI-i.-.-f?nH-ctH•00)CD1J0)H•ctHH.H)OCD(DCDCDr1(DO(DH-NCflOH-ZctCDCDCDr1CDIC3CCoI0‘.1I-‘ICICCoI0II_L01001000IC)z0—‘iicX O\Oa)I____• .2 .EowF c a)‘—,. 0 I.. t1J4-)4-.a)o C)o owo Li.1wC,) Zq-io 0 0 0 0 0 0 0o 0 0 0 0 0 0U) 0 U) 0 U) 0 U)C’) C’) C’l C4 ‘ ‘0N4J •d_________ _ _ _ __O’4-4a)0o E.!Zo I‘ i• D 4j.C I I 0-U-‘‘ 0IE-r4_IC I(Hz I I I I-4’-C%.I1----v-0OOO0,‘I:i: CIC0,I0II•TjI-I.CD(n.I-..<Q,-I-s.I-J.0I-’)-b’.QP)CDbW’CD(-t-wzPrt.IIH)bH)CDCDCD<P)C)C)I0CDI-’ZbOCDCDI-’ZrtCDrCDP) CDct0-(1)0I-’.‘ct0CDtIO-.-(1)C0(1)CDbQrJ)I-i-a.CD(I)o•I-I.I-’C•) —I-’-.%I-..I-i.I-,.C)P)Ct)CDI-’(1)1-’‘-ICD.L0010010010CDz10CD101001*-nIICICCoI0-DC•) 0C) CD10DC) 10-I3CO*Table 3.10: ANOVA for Horizon and Site (Probability of NullHypothesis) and Squared Multiple R.Horizon Site Horizon*Site R2N .000 .048 0.890S .000 .081 0.900C/N .000 .035 0.884AvN .000 .000 .005 0.791AvP .000 .073 .044 0.765AvS .000 .002 .041 0.714Ca — —— 0.246Mg .002 0.440K .000 .000 .080 0.835Mn .000 —— 0.684TPSS .000 .036 0.642CELL .000 .053 0.45272and no difference in available S on the Hw and Hh horizons(Figure 3.7). According to Stevenson (1982), the amount of Smineralized does not appear to be directly related to soil type,total concentrations of C, N, or S, or to C/SI N/S or C/N ratios,soil pH, or mineralizable N, but is probably caused by thechemical nature of the decomposing fraction and by those factorswhich influence the growth of microorganisms, such astemperature, moisture, pH, and availability of food supply.For inorganic nutrients, there were no significant effectsof horizon, site or horizon X site interactions for Ca (Table3.10 and Figure 3.8). Table 3.9 does show that Ca concentrationin the Fm horizon of the CH site was much higher than for the HAsite, and this is presumably because the litter of both westernred cedar (Minore, 1983) and salal (Klinka, 1976) tends to becalcium—rich, but the large sample variation probably masks anysignificance.Similarly, the CH tended to have higher concentrations of Mgand Mn (Figure 3.9), but the differences were not significant,again probably because of the high sample variation. Oneexception is that the Mn concentration in the CH was extremelyand significantly higher for the Fm of the CH. Carter1 (1991,pers.comm.) found that Mn tends to be higher on nutrient—poorsites, and Klinka (1976) found the mean Mn concentration in salalleaves to be over 2000 ppm. It is unclear whether salal1 R. Carter. Research Associate, University of BritishColumbia, Vancouver, B.C.731rJI-..IICDI-’.I-’.rtb(J)CDwI-JI1.IIIi000ZC) ct(DIICDI-•(DZ0‘(DOfrt,ctI-..cttflCDI-•tflrt•CD0CDI-11<0I-I.*ICIC(0I0CD0C-‘IIIIzjI-,.IICDI-J.0H,I-bi(DNX 000iCDbQ‘<CDb)(Dflb ç1I-’-I-rfCDCDCDCD0‘—.O’OI_I(noo I-I.<I-I.rtCDuII.<p) _—0)(JiCDCDCD00I-’.D‘-JO)CD0)*CD‘-I.0.000) O)U)cttnCDI-•0)I.J.‘.0I-’•H‘-I.Drtu-...hh-hC) 0)oo-‘,c)‘1zICCCoI0IIC,00CD10) -I.0CD00Ci)0I**H0—H)ti(DNX‘Ion CDU)9)OCDbQ0)ct-‘•I-’rt(DCDCDCD5.—.0)tnoz-<CDct(DU)CDSrni.1-10;(1)-10)CDU)O(D ‘-I.I-in•P)S.—‘CD*0)I-’.I-’00) 0)5rtQ(DO)SCDCDU)55I-..I-hI-,.0OH)0)‘1C) 0z0CD-‘0)00-‘0(00)3Co(0C’)0mC)CDC)CDC)CDCI)3mC)CDC)CDC) DC) DCDC’)CDTJI-i.‘ICD‘.3‘.0--‘hDO.•000000‘.1I-IICC’)I0I-Fs.)1%)Cli0Cli0(3100000000•0(310 000I-‘ICICCOI0*Iaccumulates Mn and therefore leaves a high concentration indecomposing litter, or if Mn is a symptom of poor siteconditions.The one inorganic nutrient which was significantly differentby horizon and site was K. Concentrations of K weresignificantly higher in Fm and Hrw horizons from the CH site thanfrom the HA, and higher, but not significantly so, for all otherhorizons of the CH (Figure 3.8). This high concentration mayreflect the presence of salal litter which has been shown to haverelatively high concentrations (6600 ppm) of K (Klinka, 1976).3.4 Organic Composition of Humus Horizons3.4.1 LipidsLipids come from two basic sources, from undecoinposed plantmaterials and from the .bodies of dead and living microfaunalorganisms. Plant materials normally contain waxes which protectthe surface of leaves, trunks, flowers and fruits. These areparticularly resistant to decomposition, and tend to accumulateas other readily digestible carbon sources are utilized. Notsurprisingly then, wood, which is basically lignin and cellulose,has a lower amount of lipids than non—woody horizons which arecomposed of many of these lipid-rich plant structures (Table 3.11and Figure 3.10). Microorganisms also synthesize lipids;bacterial cells contain from 5 to 10% lipids, fungi usuallycontain from 10 to 25% (Stevenson, 1982). As decompositionproceeds, the bodies of dead organisms accumulate, so that the77Table 3.11: Organic composition of humus types including lipids,total polysaccharides, labile polysaccharides andcellulose by site-type based on 4 replicates (unlessotherwise noted).Woody Non-woodyFw Hrw Hw Fm Hh HhiLipids (% oven-dry, ash—free soil)CH 2.36 4.25 7.56 5.24 5.68 5.41(1.47) (1.21) (1.29) (0.42) (1.14) (2.11)HA 1.93 2.22 5.04 4.86 4.82 2.62(0.13) (0.64) (0.49) (0.40) (1.03) (1.58)Mean 2.15 3.23 6.30 5.05 5.39 4.21(1.07) (1.38) (1.61) (0.35) (1.09) (2.30)Total Polysaccharides (% oven—dry, ash—free soil)CH 19.7 13.3 16.0 32.1 24.3 18.5(14.9) (1.9) (3.5) (2.2) (4.8) (8.1)HA 11.1 12.4 14.8 26.6 21.6 15.8(3.0) (3.9) (2.3) (2.8) (1.1) (3.3)Mean 15.8 12.5 15.1 28.1 22.1 17.2(11.5) (3.0) (2.9) (3.5) (4.2) (5.1)Labile Polysaccharides (% oven-dry, ash-free soil)CH 12.9 10.4 13.2 26.2 20.3 15.3(9.2) (1.3) (3.0) (2.9) (3.2) (5.4)HA 7.9 9.5 12.6 21.4 17.3 14.3(1.7) (2.0) (1.5) (1.3) (0.3) (3.1)Mean 10.6 9.6 12.6 22.9 18.3 14.8(7.0) (1.6) (2.3) (3.3) (3.2) (4.8)Cellulose (% oven-dry, ash-free soil)CH 6.8 2.9 2.9 5.9 4.0 3.3(5.8) (0.7) (0.7) (2.0) (1.6) (2.8)HA 3.2 3.0 2.2 5.2 4.3 1.5(1.4) (2.0) (1.1) (1.5) (0.7) (0.8)Mean 5.2 2.9 2.5 5.3 3.9 2.6(4.6) (1.5) (0.9) (1.6) (1.3) (1.8)78I-&.I-’.C) 0C) CDzriIIrt00(00 0N0(0(0I-s.r1CD0CDz1<(0TCDCD(00I-I.I—.(0ITJI-’.IICD0‘1IVc.Ci)ICIIIxIcontent of lipids in more decomposed materials is higher.The lipid concentration of the humus types ranges between 1and 9% which is in the reported range for soils (Stevenson,l982);means are shown in Table 3.11. There were no significantdifferences between humus types. However, two trends are clear.First, the non-woody materials tend to have a higher lipidcontent than woody materials. Second, there tends to be anaccumulation of lipids from the least decomposed to the mosthighly decomposed for both woody and non-woody types. Both thesetrends are expected.Differences between sites for each humus type are shown inTable 3.11 and Figure 3.10. Again, differences are notsignificant, but CH sites have consistently higher lipid contentsand greater variability than HA sites. There are severalpossible reasons why lipid concentrations are higher on the CH:the concentration of lipids originating from salal or western redcedar litter may be higher than that of litter found on the HA;the wetter site conditions of the CH may inhibit the breakdown oflipids by microorganisms leading to accumulation; or to a largerquantity of lipids synthesized by microorganisms on the CH. Moreintensive sampling might provide clues. In any case, the higherconcentration of lipids in the CH would suggest thatdecomposition on these sites is slower, or less complete than onthe HA. The higher concentrations of lipids on the CII suggests arole for lipid oxidation products as allelochemicals, but thiswas not investigated.803.4.2 Polysaccharides and cellulosePolysaccharides are high molecular weight sugar compounds.They originate from plants, mainly cellulose and hemicellulose,and as products of microbial metabolism. The stability ofpolysaccharides in soils is caused by a combination of severalfactors which makes them resistant or inaccessible to microbialattack, including structural complexity, adsorption on clayminerals or oxide surfaces, formation of insoluble salts orchelate complexes with polyvalent cations, and tanning by humicsubstances (Stevenson, 1982).Total polysaccharides in woody horizons were notsignificantly different from each other, but were significantlylower than the non—woody Fm and Hh, which presumably containabundant leaf and root residues (Table 3.11 and Figure 3.11).The well humified Hhi had significantly lower totalpolysaccharide concentration than the Hw. The labilepolysaccharides did not have similar variances so could not bestatistically analyzed, but the trends appear to be similar tothat of the total polysaccharides (Table 3.11 and Figure 3.11).Cellulose, which was calculated as the difference betweentotal and labile polysaccharides, accounted for only 2 to 5% ofthe total polysaccharides (Table 3.11 and Figure 3.12). Therewere no significant differences in cellulose between the two Fhorizons, Fm and Fw, and there was no difference between the twomost decomposed horizons, Hhi and Hw, but the difference betweenthe most and least decomposed horizons were significant. Horizon81I•TjI-I.CDL)I uOC3CCoI.zINO0O-lN0zI-’.IIrti-•CD*O<0CDI.-.. LsJ(OCDI-hOIII.-,(D’< CDU20)(no00I-’.1:r.—I-a.CD000CDr100I.t,I0)CD-U-C’) 0)) o-0=CD0.CD—oC)D•—-‘...-0CD D•1hh1%.)1%)Cs)Cs)0Cli001ppioCli•00000000--r.)C,)p!•Pç.npnp0000000C.)wOz00______•14-a)0______C0C)a)LL.a,G)‘4-I(0>1a)0*a)U)(I)0N•d0(0co‘4-I004.)x 4.)C a)UEN 0IJL0 UII a)(0 (00.I-I--..II rIr-Ii1 rlwoI..(Ni-1Cla)z•r-Iand site effects were significant (Table 3.10), but there was nosignificant interaction between horizon and site. When total andlabile polysaccharides were compared by site within horizon, theCH was significantly higher for the Fm, and higher but notsignificantly so for all other horizons. Similarly, cellulosewas higher, but not significantly so, for all horizons on the CH.A greater sampling intensity to reduce overall variability wouldincrease the significance. The trends are in agreement withearlier findings of lower nutrient availability and higher lipidcontents on the CH, and that decomposition seems to be slower orless complete on the CH than on the HA.3.4.3 Bound phenolic acidsThe phenolic acids, p—hydroxybenzoic, protocatechuic,vanillic, and syringic acids (hydroxybenzoic acid derivatives),and p—coumaric and ferulic acids (cinnamic acid derivatives), arecommonly found in living plants and in soils. The kind andquantities of these phenolic acids in soil depends to a largedegree on the origins of organic material. Gymnosperm ligninconsists predominantly of coniferyl units with low amounts of p—coumaryl units and low or absent syringyl units; angiospermdicotyledon lignin contains an equal amount of coniferyl andsyringyl units with very low p—coumaryl units; and monocotyledonscontain approximately equal amounts of all three units (Crawford,1981). The p—coumaryl units can also be derived from suberin,and to a lesser degree, from cutin (Kolattukudy, 1980).84Protocatechuic acid is a common intermediate product formedduring catabolism of lignin (Kirk, 1984). Thus in this study,woody horizons can be expected to be high in vanillic acid, whilethe non—woody horizons, with a greater angiosperm contribution,would have higher syringic acid content. Similarly, the litter-derived Fm horizon would have higher coumaryl derived units thanwoody Fw horizon.As expected, a comparison of humus horizons (Table 3.12)indicates that woody horizons tended to be higher in vanillicacid while the non-woody horizons tended to be higher inprotocatechuic, p—coumaric and ferulic acids. The Fm horizon,with the greatest proportion of angiosperm inputs, had thehighest concentrations of protocatechuic, syringic, p—coumaric,and ferulic acids; the Fw, which is pure coniferous wood, had thehighest concentration of vanillic acid; the Hhi, was found to behighest in p-hydroxybenzoic acid. Interestingly, the Fm, whichis the most active horizons, was found to have the highestoverall concentration of phenolic acids, while the Hhi, whichwould be expected to accumulate the mobile organic acids fromoverlying horizons, was found to have the second highest overallconcentrations of phenolic acids.Changes with decomposition of woody humus were examined withno significant differences, but the following trends were found(Table 3.12). Vanillic acid decreased with decomposition, whichwas consistent with the reduction in total lignin content content85Table 3.12: Concentration of bound phenolic acids (standarddeviation) by humus horizon (micrograms per gram ofoven—dried, ash—free soil).ACID HORIZONWoody Non-woodyFw Hrw Hw Fm Hh HhiProto— 114a 120a 121a 392b 342b 141acatechuicp—Hyroxy— 9Oab 52a 116b 108b 123b 248bbenzoicVanillic 361b 278b 279b 135a 161a 294bSyringic lBab lOa l7ab 35b l6ab 26abp-Coumaric 3Oab ha l7ab 125c 83bc 75bcFeruhic 23ab h2a l6ab 75c 52bc 71c86as discussed inprevious sections. A similar trend was found byKogel (1986). Syringic was found to initially decrease from Fw toHrw, but then increased from Hrw to Hw. This may occur becauserooting density is greater in the more decomposed Hw, and couldtherefore be directly affecting the concentration of syringicacid.A comparison was made of phenolic acid concentrationsbetween CII and HA (Table 3.13). The Fw horizon, which had thehighest vanillic acid concentration was not significantlydifferent between the two sites. For the Fm horizon, the CH sitewas significantly lower in vanillic acid than the HA site, butsignificantly higher in syringic acid, which is consistent withthe fact that the CH has higher angiosperm litter inputs. Therewere no significant differences between CH and HA sites in any ofthe humus horizons for protocatechuic, ferulic or p—coumaricacids.A correlation analysis was performed to examine therelationship between the concentration of bound phenolic acidsand a) whether or not the horizon was woody, b) the proportion ofthe plot covered by salal, or a) the proportion of the plotcovered by any other herbaceous or shrubby vegetation includingVaccinium spp., Cornus canadensis (bunchberry) and Blechnumspicant (deer fern). Vegetation data was taken from Table 3.2 inthe preliminary study which examined vegetational differencesbetween CH and HA sites. The results are shown in Table 3.14.87Table 3.13: Concentration of bound phenolic acids (standarddeviation) by humus horizon and site (micrograms pergram of oven—dried, ash—free soil).ACID SITE HORIZONFin Fw Hrw HwProto— CH 483 113 151 108catechuic (250) (19) (54) (39)HA 302 116 90 135(51) (11) (30) (49)p-Hyroxy- CH 111 72 83 136benzoic (44) (28) (24) (44)HA 105 109 22 96(13) (64) (9) (15)Vanillic CH 107 357 275 245(13) (45) (48) (80)HA 162 365 281 312(33) (54) (39) (48)Syringic CH 46 15 12 17(14) (4) (3) (6)HA 24 20 7 17(7) (8) (3) (4)p—Coumaric CH 129 19 16 24(96) (19) (10) (16)HA 121 40 6 10(12) (23) (6) (4)Ferulic CH 89 29 17 19(21) (18) (13) (5)HA 62 17 7 14(8) (5) (4) (4)88Table 3.14: Pearson correlation matrix between bound phenolicacid concentration and abundance of salal orother shrubs, presence of wood, and site type (CHor HA).Phenolic Acid Salal Other Wood Site_______________Shrubs Presence TypeProtocatechuic .278 .036 —.623 —.193p—Hydroxybenzoic .423 .486 —.392 —.299Vanillic —.017 .125 .557 .178Syringic .342 .354 —.498 —.194p—Cournaric .290 .138 —.704 —.114Ferulic .470 .275 —.772 —.31689The presence of salal was most highly correlated with thepresence of p—hydroxybenzoic, ferulic and syringic acids. Thepresence of all other herbaceous and shrubby vegetation was mosthighly correlated with p-hydroxybenzoic and syringic acids. Thewoody horizons were highly positively correlated with vanillicacid only, and highly negatively correlated with all other acids.Site type was not highly correlated with phenolic acids, but thenegative values indicate that HA tended to have lowerconcentrations of phenolics than CH site.The results indicate that vegetation type has a largeinfluence on the concentrations of bound phenolic acids present.Salal, and possibly other shrubs, such as Vaccinium (alsoericaceous), have waxy, thick cuticles that provide a source ofcoumaric and ferulic acids. The angiosperms also contributesyringic acids. Wood, on the other hand, has very few phenolicacids other than vanillic acid, a lignin derivative.4.0 SUMMARYSignificant differences were found in physical and chemicalproperties between CH and HA sites. Although CH and HA sitestend to occur side by side, stand structures, humus profiles,soil properties and moisture regimes are different. HA sitestend to occur on ridgetops, with CH sites adjacent on lowerslopes. The topographic position is correlated with moisturegradients, being driest at the top and wetter towards the bottom.The CH stands therefore occupy a wetter site.90The higher topographic position also makes the site lessprotected from strong winds, making the HA stands moresusceptible to windthrow. And once windthrown, the standstructure of the resulting stand would tend to further its’susceptibility to windthrow. The undisturbed CII stands arerelatively open and uneven—aged, which encourages windfirmness;in contrast, the HA stands are dense and even—aged, whichencourages windthrow (Oliver and Larsen, 1991). Windthrowdisturbances are very important in causing site differences.In terms of differences in understorey vegetation, therelatively open stand structure occurring in the CH phase,supports abundant and dense growth of salal. Other vegetationincluding Vaccinium spp., Menziesia ferruginea, Blechnum spicantand Cornus canadensis, are much less abundant. The densestructure of the HA stands supports only sparse shrub cover,including Vaccinium spp., and Blechnum spicant. Mosses such asHylocomium splendens, Kindbergia oregana and Rhytidiadeiphusloreus occur on both phases.The CH phase, which is thought to be undisturbed for over1000 years, was found to have almost continuous root restrictinglayers such as compacted or cemented horizons, a high occurrenceof standing water over some portion of the year and gleyedmineral soil horizons. The HA had more visible windthrowactivity, including a greater proportion of woody humus andcorrespondingly deeper forest floors, more friable mineral soils,and deeper rooting zones than the HA. This suggests that91windthrow disturbance is important in physically rejuvenating thesite by breaking-up hardpans, mixing organic matter with mineralsoils which increases the soil friability and encourages deeperrooting by enriching the mineral soil fertility. Thus, withoutmajor disturbance events, such as windthrow, CII sites would tendto degrade further with a greater occurrence and build—up of Hhihumus horizons, cemented horizons, and standing water.Humus profiles of the CH and HA phases were found to havesix predominant humus horizons. Three of these horizons were ofwoody origin, ranging from the least decomposed Fw horizon, tothe residuic Hrw, to the well decomposed Hw. The remaining threehorizons were of non—woody origin, ranging from the leastdecomposed Fm horizon, to the well decomposed Hh, and an Hhiwhich was an organic horizon containing intermixed mineralparticles found immediately above the mineral soil. The majorityof fine roots were found in the Fm, Hh and Hw horizons.The windthrow process was responsible for the deposition ofa large amount of woody material on HA sites. Poorly decomposedwoody horizons (Fw and Hrw) were found to occur on 68% of HAplots, but only 38% of CII plots. Well decomposed Hw horizonswere found on 56% and 38% of HA and CH plots respectively. Wherethey did occur, there was not significant difference in depth ofwoody horizons between CH and HA. The non—woody horizons Fm, Hhand IIhi were deeper and occurred more often on the CII plots.Humus profiles also reflected differences due to windthrow.The HA was found to have 38% Hemimors (consisting of thin F and H92horizons occurring over a windthrown mixture of organic andmineral soil), and 41% Lignoxnors (consisting of greater than 35%decaying wood), but only 22% Humimors (reflecting relativelyundisturbed organic horizons). In contrast, the CR had morediverse humus profiles including 57% Huinimors, 17% Lignomors, 24%Hydromors (which develop under fluctuating, stagnant water) and2% Hemimors.Nutrient concentrations were found to vary between horizons.Not surprisingly, the woody humus was found to be significantlylower than non—woody humus for pH, total N and S, available N, Sand P, and total inorganic K and Mn. Nutrient concentrationstended to increase with increasing decomposition of woodyhorizons. The 11w horizon was found to have abundant fine roots,and was thus the most biologically important of the woodyhorizons. Nutrient concentrations within the 11w horizon werefound to differ between CII and HA sites; the pH, moisturecontent and C/N ratio of those from the CH were higher than thosefrom the HA, while total N and S, and available N and P weresignificantly higher than those of the HA compared to the CR.This indicates that the CR site seems to be decomposing either ata slower rate than the HA, or that decomposition is lesscomplete.For non—woody horizons, the Fm horizon was found to besignificantly higher than other non-woody horizons in available Pand exchangeable K and Mn. There were no significant differencesin nutrient concentrations between Fm and Rh horizons. The Hhi93horizon, occurring close to the mineral soil interface, hadsignificantly higher ash contents, and total C, N and S. For theFm horizon, samples from the CH sites were found to besignificantly higher than from the HA for pH, moisture content,available S, and exchangeable Ca, K, and Mn; this most likelyreflects differences in litter input between the sites, and tothe wetter conditions of the CH. For the Hhi horizon, samplesfrom the CH sites were found to be significantly higher than forthe HA for pH and water content, but available P was higher forHA than for CE.Lipids concentrations were found to increase with increasingdecomposition, while total and labile polysaccharides andcellulose were found to decrease with increasing decomposition.Lipids, total and labile polysaccharides and cellulose were allfound to be higher for CH horizons than for HA. This suggest, aswith the nutrient concentration differences between sites, thatthe CH site seems to be decomposing either at a slower rate thanthe HA, or that decomposition is less complete.A comparison of bound phenolic acid concentrations betweenhumus horizons found that woody horizons tended to be higher invanillic acid while the non-woody horizons tended to be higher inprotocatechuic, p—coumaric and ferulic acids. The Fm horizon,with the greatest proportion of angiosperm inputs, had thehighest concentrations of protocatechuic acid, syringic acid (adegradation product of angiosperm lignin), and p—coumaric andferulic acids (thought to originate from the cutin of leaf and94other plant tissues and from the suberin of roots); the Fw, whichis pure coniferous wood, had the highest concentration ofvanillic acid (a degradation product of coniferous lignin); theHhi, was found to be highest in p-hydroxybenzoic acid.Interestingly, the Fm, which is the most biologically activehorizon, was found to have the highest overall concentration ofphenolic acids, while the Hhi, which would be expected toaccumulate the mobile organic acids from overlying horizons, wasfound to have the second highest overall concentrations ofphenolic acids.When the CH site, with characteristically abundant salal wascompared with HA sites, with relatively sparse understoreyvegetation, the CH site was found to be significantly higher insyringic, p—coumaric and ferulic acids. The concentration of p—hydroxybenzoic acid, originating from suberin of roots, was notsignificantly different between sites, but tended to be higher onCH sites than on HA.5.0 CONCLUSION5.1 Soil ClassificationHypothesis: That distinct and recognizable humus horizonsoccur commonly on both the CH and HA sites, but that therelative abundance varies between the sites.Conclusion: Accept.955.2 Chemical Differences in Humus Horizons Between CH and HAHypothesis: That similar horizons from the CH and HA sitesare different with respect to chemical composition includingtotal and/or available nutrient concentrations, lipid andpolysaccharide contents, and phenolic acid content.Conclusion: Accept that some available nutrients, lipidsand polysaccharides and phenolic acid contents are differentbetween some horizons from CH and HA sites.96Chapter IVChemical Characterization of Humus Horizons Using 13Carbon NuclearMagnetic Resonance’1.0 INTRODUCTIONNuclear magnetic resonance (NMR) spectroscopy is a techniqueoriginally developed for organic structure determination inchemistry. However, since 1970, developments in NNR havefacilitated its application to many areas of applied science,including soil science. Solid-state 13C NMR has recently beenused to characterize forest litter and humus layers (Kogel1988; Kogel-Knabner J., 1988), litter decomposition andhumification (Hempfling j., 1987; Hammond g 1985).Other applications of ‘3C NMR in soil science and forestry arelisted in Preston and Rusk (1990). A review of the NNR techniqueis beyond the scope of this thesis but is well described byWilson (1987)This study used ‘3C NNR spectroscopy to examine chemical1 This chapter is based on the publication by de Montigny(1992).97differences in the humus types between the ecosystem phases whichcould explain the differences in forest productivity afterclearcutting.2.0 METHODS2.1. Sample PreparationSamples representative of the 6 humus types from the 2 sitephases (as described previously) were collected, air dried andground to pass a 20 mesh sieve. In addition, coniferous litterand salal litter were collected from each of the 5 CII and/or HAsites, air—dried, bulked and ground to pass a 20 mesh sieve.Replication was not possible due to the high costs of 13C NNR.However, the excellent agreement between samples of similar humusbetween sites indicates that choosing representative humus typesfrom the CH and HA sites provided adequate sampling for thisstudy.2.2 NMR SpectroscopyThe NNR spectroscopy was done by Dr. Patrick G. Hatcher ofPennsylvania State University. Dry samples were packed into abullet-type rotor that was placed in the probe of a ChemagneticsInc. M-100 NMR spectrometer operating at an 1H field of 100 MHz.The sample, which was spun at a rate of approximately 3.5 kHz atthe “magic—angle” of 54.7 ° to the magnetic field, was analyzedby the technique of cross polarization with magic-angle spinningas described by Hatcher (1987). Dipolar—dephased (DD) CPMAS98spectra were generated by inserting a delay period of 40-100 jswithout 1H decoupling between the cross—polarization andacquisition portions of the CPMAS pulse sequence (Opella andFrey, 1979). Chemical shifts are reported relative totetramethylsilane (TMS) at 0 ppm.2.3 Spectral AnalysisSpectra were divided into chemical shift regions accordingto chemical types of C as follows: A, aliphatic 0-50 ppm; B,methoxyl 50-60 ppm; C, 0-alkyl 60-96 ppm; D, di-O-alkyl andaromatic 96—141 ppm; E, phenolic 141-159 ppm; F, carboxyl 159-185ppm; and G, aldehyde and ketone 185-210 ppm. In the context ofthis paper, the term aromatic C is used to designate,specifically, the nonoxygenated aromatic C occurring at 96—141ppm, and the term phenolic is used to designate the oxygen-substituted carbons at 141-159 ppm.Areas of the chemical shift regions were measured by cuttingand weighing the spectra and are expressed as percentages oftotal area (relative intensity). The proportions of lignin andcarbohydrate C were then determined using the procedure describedby Preston al. (1990). Briefly, these include:Total Lignin CarbonCL = 4.5 E + BTotal Carbohydrate Carbon= 1.2 (C — 1.5E)99Ratio of Carbohydrate to Lignin MonomersCm = 1.2 (C— l.5E)Lm 3EAromatic LigninAr* = D — 0.2 (C — 1.5E)Ratio of Carbohydrate to Aromatic Lignin Monomers= 1.2 (C — 1.5E)Lm E+Ar*The problems and uncertainties inherent in analyzing thespectra this way have been described by Preston . (1990).The most notable is that there are problems with peak overlap andlack of completely specific chemical shift regions, for which theuse of vertical divisions and correction factors cannot fullycompensate. Thus the analysis of lignin signals is based only onthe guaiacyl structural unit, which is a major component ofconiferous lignin. This is satisfactory for the woody horizons,Fw and Hrw, which are almost exclusively the carbohydrate—depleted residue of coniferous wood with little alteration oflignin structures. The results are less meaningful for horizonsin which decomposition has proceeded further, resulting inalteration of lignin structures and in non—woody horizons derivedfrom salal and needle litter. Therefore, these calculations werenot done for the Hhi horizons.3.0 RESULTS AD DISCUSSION3.1 Carbon-13 CPMAS NMR Characterization of Humus Horizons1003.1.1 Woody horizonsTo aid in the interpretation of the NNR spectra, thestructural units of cellulose, guaiacyl lignin and condensedtannins are shown in Figure 4.1. Spectra of woody and non—woodyhorizons from CH and HA sites are shown in Figures 4.2 and 4.3,respectively, and the relative proportion of C in the chemical-shift regions in Table 4.1. The general features of the spectra,and the values in Table 4.1, indicate that there is little todistinguish samples taken from the CH and HA sites. There arealso only small differences between Fw and Hrw spectra, but alarger difference between this pair and the Hw spectra.The spectra of the Fw and Hrw horizons are similar to thosereported previously for well—decomposed gymnosperm wood (Preston1990; Hatcher, 1987). These spectra (Figures 4.2 a—d)are dominated by signals typical of the guaiacyl lignin unit(Figure which predominates in softwoods; these includemethoxyl C at 56 ppm, and aromatic and phenolic C at 110 to 160ppm. The phenolic region (141-159 ppm) includes the guaiacyl C3at .148 ppm, free C4-OH at 146 and C4 participating at C-O-C4 etherlinkages at 153 ppm CLeary j., 1986). The aromatic region(96—141 ppm) arises from guaiacyl C1, C2, C5 and C6. 0—alkyl C(including carbohydrate) is depleted in these spectra, duelargely to the three—C side chain of lignin.In fresh wood, the 0-alkyl region (and in fact, the wholespectrum) is dominated by signals from carbohydrates, mainlycellulose (Hemmingson and Newman, 1985; Newman and Hemmingson,101(A) cellulose repeaLing unit (B) lignin repeating unitecjH2O HH—_o_?\gH/H—H OH H— COHR2 R1(OH0 —. C= OCH3, R2 = H gualacyl= R2 = OCH3 syringytR1 = R2 = H phenyipropane(C) condensed tannin repealing unitOHR = H procyanidin unitR = OH prodeiphinidin unitFigure 4.1. Structural units of a) cellulose, b) lignin and c)condensed tannins.102Table4.1.:Relativepercentagesofcarboninchemicalshiftregions(ppm)ofhumustypesbysite.AliphaticMethoxylo-AlkylAromaticPhenolicCarboxylCarbonyl(0—50)(50—60)(60—96)(96—141)(141—159)(159—185)(185—210)Non—WoodyHorizonsLitterCH2653319961HA2744414641FmCH20532241063HA21529241074HhCH19428271185HA18427271094HhiCH3852018884HA255202411105WoodyHorizonsFwCH81027341533HA111123351532HrwCH14924331443HA101025341443HwCH23724271163HA17524281385103Table 4.2: Ratios of calculated total lignin (Liii), carbohydrate(Cm) and aromatic (Ar) carbon and associated ratiosfor humus types by site. Cm/Lm* is calculated usingAr.Lm Cm Ar Cm/Lm CmILrn*Non—Woody HorizonsLitterCH 43.6 24.5 15.1 1.1 1.0HA 30.9 41.5 7.5 2.3 3.1FmCH 48.1 21.1 25.7 0.7 0.6HA 48.7 17.2 18.2 0.6 0.6HhCH 52.8 13.3 21.0 0.4 0.4HA 50.0 14.4 21.5 0.5 0.5HhiCH 39.4 10.1 14.0 0.5 0.4HA 54.8 4.4 20.0 0.1 0.1Woody HorizonsFwCH 78.7 4.7 28.6 0.1 0.1HA 80.2 0 30.1 0 0HrwCH 70.0 4.6 28.3 0.1 0.1HA 74.5 4.2 28.8 0.1 0.1HwCH 54.5 9.2 22.3 0.3 0.3HA 62.6 5.7 23.1 0.2 0.2104Figure4.2.IIIIIIIIPPM240180120600-60PPM240180120600-60Carbon-13CPMASNNRspectraofwoodyhorizons(Fw,Hrw,Hw)fromCHandHAsites.CIISite(a)FwHASite(b)Fw(c)[trw(d)Hrw1051990; Preston 1990). These include the crystalline (65ppm) and non-crystalline (84 ppm) components of C4, and theanomeric C. Total carbohydrate C was calculated to be about 4%for the less-decomposed Fw and Hrw, and 7.5% for the Hw (Table4.2). The ratio of carbohydrate to lignin moieties (Cm/Lm) wasin the range of 0.1 to 0.3. These are similar to values foundfor highly—decomposed logs of Douglas—fir, western red cedar andwestern hemlock. By contrast, ratios of 2—3 are found for freshwood (Hemmingson and Newman, 1985; Preston j., 1990).Also in contrast to fresh wood, there is a broad region ofintensity in the aliphatic region (0—50 ppm). This increasesfrom about 10% in the Fw , to 12% in the Hrw, and to 20% in theHw (Table 4.1). There is also an increase in resolution; for theFw spectra, the aliphatic region is a broad shoulder most likelydue to the selective preservation of waxes and resins in theoriginal wood. The Hrw spectra begin to show a peak at 30 ppm,which is better defined for the CH site and for the Hw spectra;this region has a strong, well-defined peak at 30 ppm. The peakat 30 ppm is characteristic of aliphatic -CH2units in longchains, such as fatty acids, and is typically seen to increasewith increasing decomposition in Folisols and Histosols (Hammond1985; Preston 1987; 1989). It is thought to belargely a byproduct of production of microbial biomass coupledwith incomplete decomposition. The increase in 0-alkyl,aliphatic and carboxyl C may be the result of microbial or fungalactivity (Baldock 1990; Preston and Ripmeester, 1983).106There is a greater contrast between the Fw and Hrw spectra,vs. the Hw. The Hw have more intensity characteristic ofcarbohydrates, at 62, 73 and 101 ppm. The carboxyl and carbonylintensity also increase from Fw to Hrw to Hw; this could be dueboth to oxidation of lignin and to the production of fatty acidsin microbial biomass. Decomposition of lignin, or production ofother aromatic structural units, is also consistent with arelative decrease in methoxyl, aromatic and phenolic C, and achange in the aromatic region, as the intensity at 130 ppmincreases relative to that from 115-125 ppm. Total lignin C iscalculated to be about 80% for the Fw, 72% for the Hrw, and 58%for the Hw (Table 4.2).It is not easy to ascertain the origin of the 0—alkylintensity in the Hw spectra; it may arise largely from theoriginal sugars in woody and non-woody litter input, or there maybe greater contribution from microbial activity. However, theusual pattern in organic soils is for plant—derived carbohydratesto decrease, while microbial aliphatic C increases (Hammondal., 1985; Hempfling j., 1987; Kogel—Knabner 1988;Preston et al., 1987; 1989; Zech j., 1990). This, and thegood resolution in the spectra, suggests that the 0-alkyl C mostlikely derived from the original inputs.3.1.2 Non—woody horizonsThe spectra of the non-woody Fm and Hh horizons (Figure 4.3)are similar to those reported for forest litter layers under107CIISiteHASitePPM240180120600-60PPM240180120600-60Figure4.3.Carbon-13CPMASNMRspectraofnon-woodyhorizons(Fm,Hh,Hhi)fromCHandHAsites.(a)Fm(b)Fm(c)Hh(d)Hh(e)HhI(I)HhiIIIIII108conifers (Kogel 1988; Krosshaven 1990; Preston1992; Zech ., 1987). The dominant signal is that from0-alkyl carbon (Table 4.1), indicating that polysaccharides arequantitatively the most significant compounds. 0-alkyl Caccounts for 33 to 44% of the total C in the litter, 30% in theFm horizon, 27% in the Hh horizon and 20% in the Hhi horizon(Table 4.2). The aromatics are the next most significantcompounds accounting for 14 to 19% of the total C in litter, 24%in the Fm horizon, 27% in the Hh horizon and 18 to 24% of thetotal C in the Hhi horizons. Aliphatics, which would include thecutin, suberin, and highly aliphatic polymers of plant cuticles,account for about 26% of the total C in the litter, 18 to 22% inthe Fm and Hh, and 25 to 35% in the Hhi. Compared to the woodysamples, the methoxy C and phenolic C content is lower, 4 to 5%and 6 to 11%, respectively.From the estimates in Table 4.2 for these non—woodyhorizons, total lignin increases with decomposition from Fm toHh, while total carbohydrate C decreases. The resulting ratio ofCm/Lm decreases from a high of 2.3 in HA litter to approximately0.5 for the Hh (values were not calculated for the Hhi horizon).The overall effects of decomposition of the non—woody horizonsare a decrease in easily—decomposable 0—alkyl C, which wouldmainly be due to carbohydrates of plant origin, and an increasein aliphatics and carboxyls, most likely derived from residualplant material or microbial biomass.There is also some evidence for tannins in the spectra of109the non—woody horizons. Tannins are present in low quantities inmost litter materials (Kogel-Knabner p1., 1991), but aredifficult to identify in NMR spectra because the peaks occur inthe same regions of those of lignin and carbohydrate (Czochanska1980; Morgan and Newman, 1986; Preston and Sayer, 1992).However, the peak due to C3’ and C5’ of condensed tannins(procyanidins and prodeiphinidins, Figure occurs at 144ppm, in a region that is usually clear in wood and litterspectra. As discussed above, the phenolic region of the Fwhorizon from both CH and HA (Figure 4.2) shows a sharp, intensepeak at 149 ppm, with a light shoulder at 153 ppm, typical ofguaiacyl lignin C3 and C4. The Fm and Hh horizons from the CHsite (Figure 4.3), which would have both coniferous and salallitter inputs, show a broad signal combining the tannin peak at145, the guaiacyl peak at 148, and the syringyl peak at 153. Thepeak at 144 ppm is less prominent in the HA than in the CH sites.It also decreases from Hh to Hhi for both sites, possibly fromdecomposition or complexation of tannins as they are leachedthrough the soil profile.3.2 Dipolar-dephased Carbon-13 CPMAS NMR Characterization ofHUmUS HorizonsThis technique can be used to examine features that may bemasked in the normal CPMAS spectra. During a delay period beforea signal acquisition, intensity is lost more quickly from carbonsthat have strong dipolar interactions with protons; i.e., carbonswith directly bonded hydrogens. The dipolar interaction is110weakened in two cases: for nonprotonated C which have a greaterseparation from hydrogen nuclei, and where there is molecularmotion in the solid state. This occurs for methyl C due mainlyto methyl group rotation, while long-chain aliphatics can alsodisplay sufficient backbone vibrations to weaken the dipolarcoupling in the solid state (Opella and Frey, 1979; Wilson,1987)Dipolar-dephased spectra have proven useful in detectingtannins, as the non—protonated C4 and C8 at 108 ppm can beobserved in DD spectra in a region that is otherwise masked byaromatic and anomeric CH (Wilson and Hatcher, 1988). Thisprovides a useful test for tannins in complex matrices, for whichthe only interference is the ketal C of carbohydrate, which wouldnot likely be a problem for litter and soils.For the woody horizons, the DD spectra for the Fw and Hrwhorizons in Figure 4.4 a-d show typical lignin peaks for methoxylat 56 ppm, phenolic at 148 ppm with a shoulder at 153 ppm, andnonprotonated aromatic C (guaiacyl C1) at 132 ppm, as well asweaker signals due to carboxyl (172 ppm) and carbonyl (195 ppm) C(Hatcher, 1987; Preston J., 1990). Tannin signals are veryweak or absent. For the Hw horizons (Figure 4.4e and f), thereis some intensity at 108 ppm, but without the othercharacteristic tannin peak at 144 ppm. There is alsoconsiderable residual aliphatic intensity for both CH2 and Cl!3,consistent with the presence of long—chain hydrocarbon moietieswith considerable molecular motion in the solid state.111CHSite-DOHASite-DUIIIPPM240180120IIIIII600-60PPM240180120600-60Figure4.4.Dipolar-dephasedCarbon-13CPMASNMRspectraofwoodyhorizons(Fw,Hrw,Hw)fromCHandHAsites.(a)Fw(b)Fw(d)Hrw(c)Hrw(1)11w112CHSite-DOHASite-DO(h)FmIIIPPM240180120600-60IIIIIPPM240180120600-60Figure4.5.Dipolar-dephasedHhi)fromCHandCarbon-13HAsites.CPMASNMRspectraofnon-woodyhorizons(Fm,Hh,(a)Fm(c)Hh(d)Hh(e)HM(f)Hhi113For the non—woody horizons, the DD spectra of the Fm and Hhhorizons from the HA site (Figures 4.5 a-d) lack a clearly-defined peak at 145 ppm, although there is some intensity at 108ppm. The tannin signal appears to be stronger for the CHsamples; however, the differences are small. Quantitativeanalysis would require a series of DD spectra with differentdipolar dephasing times (eg, Hatcher, 1987; Wilson, 1987), aswell as analysis of replicate samples. Dipolar—dephased spectraof the most decomposed Hhi samples indicate much lower tannincontents, consistent with the trend shown in the non—dephasedspectra (Figure 4.3e and f). As was seen for the DD Hw spectra,those from the non—woody horizons also show considerable residualintensity from more mobile long-chain aliphatic C.3.3 Carbon-13 CPMAS NMR Characterization of Litter InputsTo trace the source of the tannins in the organic horizons,some litter inputs and other salal plant parts were examined.The CH and HA litter (Figures 4.6b and c) are similar toconiferous litter reported elsewhere (Norden and Berg, 1990;Preston 1992; Zech 1987), as well as to the Fmhorizons except that the resolution is better as there has notbeen any decomposition. Tannins, if present are only indicatedby the breadth of the phenolic signal with poorly resolvedintensity at 145 ppm. The salal litter (Figure 4.6 a) isdifferent, as it clearly shows a tannin peak at 144 ppm. Thespectra of salal flowers, leaves and roots, shown in114PPM 200 100 0Figure 4.6. Carbon-13 CPMAS NMR spectra of litter materialsfrom: a) CH site salal, b) CH site coniferouslitter, and c) HA site coniferous litter.(a) SalalLitter(b) CHLitter(c) [IALitterI I I —— I I115PPM 200F---100 0Figure 4.7. Carbon-13 CPMAS NNR spectra of salal: a) flowers,b) leaves and c) roots.(a) SalalFlowers(b) SalalLeaves(c) SalalRoots116Figure 4.7, all have a strong tannin signal at 144 ppm.Interestingly, the tannin content in the leaves is less than inthe flowers and roots. Furthermore, the same peak is clearlyseen in the litter (Figure 4.6a), indicating as expected, thatthe tannins do not readily decompose.A sample of salal leaf tannin was run using solution NMR(Preston2, 1991, pers. comm.). The spectra was similar to thatof proanthocyanidins reported by Czochanska (1980). Morespecifically, the spectra appears to be that of a polymer ofprocyanidin and prodelphinidin units (Figure lc).3.4 Site DifferencesThe organic components of similar horizons from the two sitetypes did not show any unusual features that would explaindramatic differences in seedling performance between the sites.That is, the woody horizons were found to be similar to largegymnosperm logs decomposing on the forest floor as seen in otherstudies. Similarly, the non—woody horizons are similar to thosewidely reported for Histosols and forest organic horizons.However, one major difference is that tannin signals appear to bestronger in CH samples than in HA. This may be due to salal,which was found to have strong tannin signals relative toconiferous litter.The presence of tannins on the CH site may explain some of2 Dr. C. Preston, Pacific Forestry Centre, Forestry Canada,Victoria, B.C. Canada117the differences in forest productivity following clearcutting.Tannins have been found to reduce the biodegradability andhumification of organic matter by three processes; the productionof protein—tannin complexes which are much more resistant tomicrobial decomposition than unaltered proteins; the permeationand coating of nonproteins such as cellulose and hemicellulose bythe protein—tannin complexes, giving them considerable resistanceto microbial attack; and by the inactivation of enzymes importantin the process of decomposition (Benoit and Starkey, 1968a,b).Other factors which may be affecting decomposition is the wettersite conditions of the CH (as previously discussed), and thehigher concentration of aliphatics, particularly of lipids (aspreviously discussed).Small differences seen in the NMR spectra between the CH andHA may reflect differences in decomposition between the CII and HAphases. Litter samples from the HA site have a much highercarbohydrate content than that from the CH sites. Yet, in the Fmhorizon, the HA has less carbohydrate than the CH site. In fact,total carbohydrate C, in both the woody and non—woody horizonstend to be higher on the CII, while total lignin C tends to belower (Table 4.2). The ratio of total carbohydrate to lignin C,also tends to be higher on the CII. All of this seems to indicatethat the carbohydrates on the CII site tend to be more resistantto decomposition than on the HA sites. Furthermore, the loweravailability of nitrogen and phosphorus on the CH site, relativeto the HA site, as found in Chapter 3, may be due to differences118in decomposition. Tannins from salal litter, or aliphaticspeculiar to salal or western red cedar litter, or the wetter siteconditions may be factors important in causing this resistance todecomposition.4.0 SUMMARYHumus horizons identified in the study had an excellentcorrespondence with spectra of ‘3C NNR. Woody horizons weredominated by signals from lignin, but with increasingdecomposition, the relative proportion of lignin decreased, whilealiphatics and carbohydrates increased, presumably from microbialand fungal sources. The non—woody humus types were typical offorest litter layers, which are dominated by signals in thecarbohydrate region. Increasing decomposition resulted indecreasing carbohydrates and increasing aliphatics and carboxyls.There were no apparent differences between the CH and HA sites,with the exception of the presence of tannins. Spectra fromsalal leaves and litter were found to contain tannin compounds.These tannins were also found in the Fm horizons on both CH andHA sites, but intensity was greater on samples from the salal—dominated CH sites. The ratio of total carbohydrate to lignin Ctended to be higher for the CH humus horizons, indicating thatthe carbohydrates in these horizons may be more resistant todecomposition, possibly as a result of the tannins, or ofaliphatic compounds peculiar to vegetation on the CH site, or ofthe wetter soil conditions on the CH. These factors could be119important to the overall reduction of forest productivity seen onsalal-dominated CH sites.5.0 CONCLUSION1) Hypothesis: That the 13C CPMAS NMR spectra obtained from thedifferent humus horizons are distinct and recognizable.Conclusion: Accept that the different humus horizonsidentified are distinct and recognizable.2) Hypothesis: That similar horizons from the CH and HA sitesare different with respect to phenolics and tannins.Conclusion: There is evidence to suggest that the litterand F horizons on CH sites may have a higher tannin content,but more work is needed to determine actual concentrationsbefore the hypothesis can be accepted.120Chapter VAllelopathic Potential of Salal1.0 INTRODUCTIONEricaceous plants have long been known to contain relativelyhigh concentrations of phenolic acids (Harborne and Williams,1973), many of which are known to be allelopathic (Rice, 1984).Many studies suggest that phenolic acids produced by ericaceousspecies could adversely affect the growth of competing vegetationincluding grasses, shrubs and trees. As discussed in Chapter 2,these studies include that of Arctostaphyllos glandulosa inCalifornia (Muller et al., 1968), Erica species in NorthwestSpain (Ballester .]., 1977; Carballeira, 1980; and Carballeiraand Cuervo, 1980) Calluna vulgaris in the United Kingdom(Handley, 1963; Jalal al., 1982; Jalal and Read, 1983a and b),Kalmia angustifolia in Eastern Canada (Mallik, 1987) and Empetrumhermaphroditum in Sweden (Zackrisson and Nilsson, 1989).Salal may also have an allelopathic effect. Towers121(1966) found salal to contain the phenolic acids p—hydroxybenzoic, o—pyrocatechuic, gentisic, protocatechuic,vanillic, syringic, p—coumaric, caffeic, ferulic and sinapicacids (Towers pJ., 1966). The tannins (polyphenolics) foundto be present in the leaves, roots and litter of salal using ‘3CNMR, as described in the previous section, could also bepotential allelochemicals. The fatty acid concentration inericaceous plants is extremely high, and some breakdown productssuch as nonanoic acids are known fungitoxins (Garrett andRobinson, 1969), and octanoic and decanoic acids are knownphytotoxins (Overbeek and Blondeau, 1954). Del Moral and Cates(1971) found some evidence for allelopathy by salal in fieldbioassays using litter extracts on germination of barley (Hordeumvulgaris) and an annual grass (Bromus tectorum).Phenolic acids as allelochemicals were the focus of thisstudy, primarily because of ease of access to a High PerformanceLiquid Chromatograph used exclusively for phenolic acids. Thepresence of free phenolic acids in soil solution was felt tobetter represent natural conditions, since alkali extraction, asis commonly used, may release phenolic acids by degradation oforganic matter (Whitehead J., 1981; Kaminsky and Muller,1978). Whether the phenolic acids found in soil solution, or thepolyphenolics in salal leachates occur in sufficientconcentrations in the soil solution to affect conifer growthremains a critical question.This study was initiated to determine: 1) the seasonal122concentration of the free phenolic acids in soil solution undervigorously growing salal, including p-hydroxybenzoic,protocatechuic, vanillic, syringic, p—coumaric and ferulic acids;and 2) the effects of the phenolic acids at field concentrationsand of leachates from the flowers and fruits of salal plants onthe germination, growth and uptake of radioactively-labelledphosphorus in three species of conifers: Sitka spruce, westernhemlock and western red cedar.2.0 METHODS2.1 Seasonal Phenolic Acid Concentration2.1.1 Soil samplingThe plantations selected for sampling of free phenolic acidswere located in Western Forests Products T.F.L. 42 near PortMcNeill at Misty 100 and 400 (aged 4 years since logging,slashburning and planting), and Rupert 200 and 418 (aged 8 yearssince logging, slashburning and planting). Soil samples weretaken from beneath vigorous young salal plants in February, May,July, September and November, 1990. In the February sampleperiod, four samples were collected from each plantation, two ofwoody origin (including Fw, Hrw and Hw) and two of non-woodyorigin (Fm). Because of high variability in phenolicconcentrations, it was decided to double the sample number, so insubsequent sampling periods eight samples were collected fromeach plantation, four of woody origin, and four of non—woodyorigin. Samples were left overnight in cool shade, and123transported to the laboratory at the Pacific Forestry Centre thenext day. They were kept in cold storage at 4°C until processed(within 1 week). Immediately before centrifuging, soils weresieved through a 10 mm sieve to remove large roots and woodypieces, then through a 4 mm sieve to homogenize the sample. Asmall sample, approximately 20 g, was taken to determine watercontent at 70°C.2.1.2 Sample preparationFresh, sieved soils were centrifuged at 10,000 r.p.m. for 20minutes to extract the soil solution. This method used 30 ccsyringes with the handle sawed of f and an 8 cm piece saweddirectly beneath to be used as a “syringe support”, so that thesyringe support and shaft fit perfectly into the centrifuge tube(see Figure 5.1). A small circle of filter paper was cut to fitat the base of the syringe shaft, and this was covered with asmall amount of polyester wool (used for fish tank filters). Thetube was then weighed and filled with soil. Approximately 40 gof wet soil could be centrifuged per tube, with 8 tubescentrifuged per run, and 2 runs per sample, for a total of about300 g of wet soil centrifuged for each sample. The soil solutionwas poured out of each of the centrifuge tubes into plasticscrew—cap cups, and the bulked sample was then kept frozen at —30°C until shipped to the University of British Columbia SoilChemistry Laboratory for phenolic acid analysis.On thawing, the solution was measured for pH, acidified to12430ccSyringeFigure5.1:Designofsoilcentrifugetubesmodifiedfroma30ccsyringe.Handle_____SupportCentrifugetubeSyringeFilterpapert125pH 1.0 with 6 N HC1, centrifuged at 6000 rpm for 15 minutes, thenfiltered on Whatman #1 filter paper. In a 500 ml separatoryfunnel the solution was extracted three times with 75 mis ofglass distilled di-ethyl ether. Anhydrous Na2SO4 was added tothe collected fractions in a 250 ml glass—stoppered flask, andleft for 30 minutes. This was then filtered (Whatman #1) into a250 ml beaker and dried overnight. The residue was taken up in 2ml of methanol for analysis by high performance liquidchromatography (HPLC).2.1.3 HPLC analysisPrior to injection into the high performance liquidchromatograph, samples were filtered through a 0.2 micrometerPrep—Disc Filter (Biorad). Chromatographic conditions used anacetonitrile gradient as follows:Time % (l% Acetic Acid % Acetonitrile0 92 811 92 816 86 1430 86 1433 40 6035 40 6040 92 843 92 8Flow = 1.5 mi/mmWavelength = 280 nmTemperature = Room temperatureAttenuation = 4 (free phenolics)Chart speed = 0.5 cm/mmDetector Sensitivity = 0.01Standard = 2 ppm (free phenolics)Peak Threshold = 22Peak Width = 6Column = ODS Spheri-5 25 cm (Brownlee)126Guard Pak = Bondapak C18 Inserts (Waters)Solvents = Acetic acid Aristar (BDH)Acetonitrile Omnisolv (BDH)Standards = Protocatechuic, p—Hydroxybenzoic, Vanillic,Syringic, Trans—p—coumaric, Ferulic (Sigma)2.1.4 Statistical analysisData obtained by HPLC was transformed from ppm (equal to mgper litre) to nanograms per gram of oven dry soil (ng/g o.d.soil) as follows:mg/litre X .002 litre X hg o.d.soil X 1000000 ng/mgwhere .002 litre is the volume of injected sample, g o.d. soil isthe oven dry weight of soil centrifuged per sample, and 1000000is the conversion factor from milligrams to nanograms.Results were analyzed using the statistics computingpackages of SYSTAT (Wilkinson, 1990). All data were checked fornormality and homogeneity of variances. Log transformations weredone if variances proved unequal using Bartlett’s test, thenretested to ensure homogeneity. Data for individual species wereanalyzed separately.The analysis involved a one—way analysis of variance forindividual phenolic acid concentration by date of sampling withTukey HSD post hoc tests of pairwise mean differences and MGLHtwo way analysis of variance of phenolic acid concentration bydate and humus horizon. A correlation analysis was also done todetermine correlation between phenolic acid concentration andsoil pH and moisture content.1272.2 Effects of Phenolics on Conifers2.2.1 Seed germinationGermination procedure was done according to InternationalSeed Testing Procedures (Anon., 1985). Seeds of Sitka spruce,western hemlock and western red cedar were obtained from PacificForestry Centre, Victoria, B.C. where they had been stored frozenat -18°C. The Sitka spruce seed was collected in 1988 at SalmonRiver, the western red cedar in 1982 at Qualicum, and the westernhemlock in 1979 at Salmon River.Seeds were given either no pretreatment, or a stratificationtreatment for 21 days at 0°C using distilled water. Fourgermination treatments were tested using the following solutions:1) a control using distilled water (pH 6);2) a phenolic acid solution consisting of protocatechuic,hydroxybenzoic, vanillic, syringic, p—coumaric and ferulicacids (BDH) at concentrations equivalent to maximumconcentrations found in natural soils under salal asfollows (pH 5):Phenolic Acid nanomoles/1 nanograms/lprotocatechuic 0.135 21hydroxybenzoic 0.510 70vanillic 1.15 193syringic 1.58 313coumaric 0.31 51ferulic 0.10 193) an extract of salal flower and berries (pH 4) using 50 gair-dried material per litre of distilled water, shaken24 hours and filtered (#4 Whatman), and1284) a soil solution (pH 4) obtained by centrifuging Flu horizon(taken from under vigorously growing salal) at 11,000 rpmfor 20 minutes.Each treatment consisted of 4 replicates of 100 seeds each,germinated in 4 X 4 plastic germination boxes. The seeds weregerminated on filter paper overlying layers of tissue papermoistened with 40 mls of the appropriate solution. Sitka spruceand western red cedar were grown for 21 days at 30°C days and20°C nights. Western hemlock was germinated at a constant 20°Cfor 35 days. Germination was defined as the radicle reachingfour times the length of the seed coat.Data was analyzed using three indices of germination fornormal germinants:1) R50 value, the number of days to reach 50% of totalgermination;2) Germination capacity (GC), the germination percentage after21 days for Sitka spruce and western red cedar, and 35 daysfor western hemlock;3) Germination value (GV), calculated by the formula GV = MDGX PV. MDG (mean daily germination) is the quotientobtained by dividing the accumulated total number ofgerminants by the number of days of the test, and the PV(peak value of germination) is the maximum quotientobtained by dividing daily the accumulated number ofgerminants by the corresponding number of days.1292.2.2 seedling growing conditionsThirty—two germinants each of Sitka spruce, western hemlockand western red cedar germinated as described above weretransplanted to flats of small pots. Each pot contained 2germinants, with rows of each species interspersed randomlythroughout the flat. Each flat consisted of one treatment. Thesoil media was a standard forestry mix consisting of 3:1peat:verrniculite, with 2 kg lime/rn3 of growing medium added.Germinants were grown in a growth chamber for 12 weeks with aregime of 16 hours of full incandescent and fluorescent light,temperatures of 21°C days and 18°C nights, and relative humidityof 60%.Seedlings were watered twice a week, once with distilledwater, and then with the 4 solutions described for thegermination experiment. About 20 mls of solution were used towater each pot. Seedlings were misted daily until seed coatswere shed. When 80% of seed coats were shed (8 weeks), seedlingswere fertilized with Green Valley Soluble 20-20-20 Plant Food(20% nitrogen, 20% phosphoric acid, and 20% soluble potash, andtrace Fe, Cu, Mn, Zn, B, and Mb). Two grams of fertilizer wereused in 8 litres of water, an equivalent of 100 ppm N.The seedlings were harvested by carefully removing them fromthe growing medium. The roots were excised at the root collar,gently washed under running water and subsequently used for theseedling root bioassay as described in the next section.Following the root bioassay treatments, both roots and stems were130oven—dried at 70°C overnight and weighed to the nearestmilligram.2.2.3 Seedling root bioassayThe seedling root bioassay was carried out as described byMcDonald J., (1991). Seedlings were harvested, the topsclipped of f at the root collars, and the roots washed carefullyunder running tap water and kept under moist paper towels. Theexcised roots from 20 seedlings for each species were placed intocheesecloth bags. One bag of each species was then immersed intoa solution containing 5 X l0 M calcium sulphate, 5 X l06 Mpotassium dihydrogen phosphate and about 1 MBq 32P per litre asorthophosphate. Roots were left for 15 minutes, then immediatelywashed for five minutes in running tap water to remove unabsorbed32p from the root surfaces. The washed roots were removed fromthe cheesecloth, placed on aluminum trays, and dried at 70°Covernight.When dry, the roots were weighed then digested in aTechnicon Block Digester using 5 mls of concentrated sulphuricacid and 1 ml of 30% hydrogen peroxide, for 3 to 6 hours untilcompletely digested. Samples were allowed to cool before adding10 mls of distilled water and pouring into scintillation vials.The 32P content was counted as Cerenkov radiation in a Packardliquid scintillation spectrometer.The data converted from counts per minute (CPM) to decaysper minute (DPM) by correcting for specific activity of the131uptake solution and background, quench and decay factors asfollows:DPM = (CPM - B) X 0.53 X decaywhere DPM is the disintegration per minute, CPM is the counts perminute measured, B is the average background count of blanks,0.53 is the efficiency of the scintillation counter,and decay isthe radioactive decay of each count based on the reference dateof the solution radioactivity.The specific activity of the uptake solution was thencalculated by multiplying the DPM by the total P concentration (5X l0 N K2P04 and dividing by the concentration of radioactive 32P(220 X 106 DPM). The uptake was expressed as moles of Pi.2.2.4 Mature root bioassayRoot samples were collected from mature western red cedarand western hemlock in late February, 1991 in the Pacific SpiritRegional Park in Vancouver, B.C. (adjacent to the University ofBritish Columbia). The site selected was a 75 to 80 year oldmixed western red cedar and western hemlock stand with a siteindex of 32 m at 50 years and a standing net merchantable volumeof 750 m3 per ha. Root segments, usually mycorrhizal, of 10—20cm axial length and 0.5—2.0 mm diameter were carefully removedfrom the 0—5 cm forest floor (LFH) horizon. These samples weretransported to the laboratory between moist paper towels inplastic bags. The roots were washed carefully under running tapwater, and twenty roots of each species placed into each of 3132cheesecloth bags, which were then kept under moist paper towelsuntil assayed (almost immediately). One bag of each species wasthen immersed into each of the 3 treatment solutions eachcontaining 5 X M calcium sulphate, 5 X 106 M potassiumdihydrogen phosphate and about 1 MBq 32P per litre asorthophosphate. Roots were left for 15 minutes, then immediatelywashed for five minutes in running tap water to remove unabsorbed32p from the root surfaces. The washed roots were removed fromthe cheesecloth, placed on aluminum trays, and dried at 70°Covernight. These were digested, counted and analyzed aspreviously described.2.2.5 Statistical analysesResults were analyzed by analyses of variance using thestatistics computing packages of SYSTAT (Wilkinson, 1990). Alldata were checked for normality and homogeneity of variances.Log transformations were done if variances proved unequal usingBartlett’s test, then retested to ensure homogeneity. Data forindividual species were analyzed separately.Germination data was analyzed by analysis of variance andDuncan’s multiple range test. Seedling growth and root bioassayswere analyzed by one—way analysis of variance by treatment withTukey HSD post hoc tests of pairwise mean differences.1333.0 RESULTS AND DISCUSSION3.1 Seasonal Phenolic Acid ConcentrationThe mean concentration of free phenolic acids in soilsolution over a one year period for both woody and non—woodyhumus in nanograms per gram of oven—dry soil are presented inTable 5.1. The values are within the range reported by Whiteheadet al. (1983). Vanillic acid, hydroxybenzoic acid and syringicacid were the most concentrated, accounting for 39%, 26% and 19%of total measured phenolic acids respectively, followed byprotocatechuic, p—couxnaric and ferulic acids (accounting for 8%,6% and 2% respectively).When yearly means of phenolic acids were compared betweenwoody (Hw) and non—woody (Fm) horizons, the non—woody horizonswere always significantly higher in phenolic acid concentrationsthan woody horizons, with the exception of hydroxybenzoic acid,which was not significantly different (Table 5.1). This isconsistent with the results of the bound phenolic acids inChapter 3.When the phenolic acids were compared over the 5 samplingperiods, significant differences occurred both between datessampled and horizons (Table 5.1 and Figures 5.2 and 5.3). Woodyhorizons had no significant differences in phenolic acidconcentration over the season for hydroxybenzoic, vanillic, p—coumaric or ferulic but differences were significant for syringicacid, which was highest in September and lowest in May, andprotocatechuic acid, which was highest in May and lowest in134Table5.1:SeasonalpH,watercontentandphenolicacidconcentration(nanograms/go.d.soil),andstandarddeviationbymonthandhumus-type(non-woodyFm,andwoodyHw).Similarlettersindicatenosignificantdifferencebetweenmonthswithinhumus-type;*indicatessignificantdifferencebetweenhumus-typewithinmonth.FebJulySeptNovMean4.15ab3.93ab3.69a*3.77ab4.08b3.89*(0.73)(0.36)(0.57)(0.47)(0.27)(0.49)3.91b3.78b3.35a3.60ab4.01b3.71(0.45)(0.29)(0.35)(0.34)(0.36)(0.43)79.4b78.6b78.5b66.4a76.5b75.9(1.3)(3.3)(3.2)(9.5)(2.9)(5.41)78.1b80.1b78.1b71.7a78.7b77.3(1.3)(2.0)(3.2)(7.4)(2.0)(3.3)3.8bc4.9c3.3abc*1.4a1.8ab2.76*(7.3)(3.4)(2.2)(2.4)(2.2)(3.60)1.3ab2.8b1.4ab0.6a1.6ab1.62(1.3)(2.0)(1.8)(1.1)(2.2)(1.9)20.2b12.7b5.0a5.2a7.6ab8.71(23.6)(13.2)(4.2)(3.0)(5.6)(11.0)8.5a9.9a6.3a4.0a5.6a6.61(6.1)(4.9)(6.3)(3.0)(5.6)(5.2)41.0c*18.6bc*13.4abc*6.5a11.3ab*14.7*(66.6)(9.8)(8.4)(8.6)(11.2)(24.8)9.0a9.5a6.0a7.0a6.5a7.4(4.5)(8.8)(6.7)(6.3)(10.5)(7.8)pH watercontentProtocatechuicp-HydroxybenzoicVanillicFm Hw Fm Hw Fm Hw FmHw Fm Hw135Table5.1ContinuedSyringicFm Hw FmcoumaricHwFerulicFm HwFebJulySeptNovMean1.6a4.9ab*11.4b*12.9b1.5a7•7*(1.8)(5.5)(13.6)(29.3)(2.8)(18.3)0.7a2.0a0.9a13.1b0.9a2.8(0.7)(3.8)(1.8)(13.9)(1.4)(6.7)0.6a2.2ab6.0b*1.7ab2.4ab*2.6*(1.0)(2.1)(11.8)(3.1)(2.7)(5.7)0.5a2.1a1.0a1.0a0.8a1.1(0.4)(3.1)(1.7)(3.1)(1.3)(1.9)0.4a*0.6a0.8a0.4a2.1b*0.9*(0.4)(2.1)(11.8)(3.1)(2.3)(1.9)0.1a0.7a000.4a0.3(0.2)(1.1)(0)(0)(1.1)(0.8)136PhenolicAcidsinHwHorizons14 12 10 8 6 4 2Phenolic0 Janacidconcentration(ng/go.d.)JulAugSepOctNovDecMonthProtocatechuicSyringicIp-Hydroxybenzoicp-CoumaricVanillic0FerulicFigure5.2.Seasonalphenolicacidconcentration(ng/go.d.soil)inHwhorizonsundersalalincutovers.FebMarAprMayJun13750 40 30 20 100PhenolicAcidsinFmHorizonsPhenolicacidconcentration(ng/go.d.)JanFebMarAprMayJunJulAugSepOctNovDecMonthProtocatechuicIp-Hydroxybenzoic-*••VanillicFigure5.3.Seasonalphenolicacidconcentration(ng/go.d.soil)inFmhorizonsundersalalincutovers.Syringic<p-CoumaricFerulic138September. Since syringic acid occurs in very small quantitiesin coniferous lignin, the high concentration presumably mustoriginate from either salal or other angiosperms rooted in thewoody layer, or from above—ground leachates of salal.The patterns of phenolic acid concentration in the non—woodyhorizons (Table 5.1 and Figure 5.3) are more complicated than forwoody horizons, differing with season and type of phenolic acid.The phenolic acids vanillic, protocatechuic and ferulic acids,were significantly higher in wet, colder months than in thedriest month (September). For vanillic this was in February, forprotocatechuic acid in February and May, and for ferulic inNovember. For syringic and p—coumaric, the pattern is reversed,with significantly higher concentrations in drier, summer monthsof July and/or September than the colder, wetter month ofFebruary. If the phenolic acids are released as hydrolysatesfrom decomposing organic materials then temperature and moisturecontent would have a great affect on concentration.A short study was done to determine the effects of soilsaturation on the release of phenolic acids. Leaching tubes werefilled with non-woody humus, saturated with distilled water, andthen drained to collect soil solution. One treatment was leftdrained, while the other was again saturated with distilledwater. Both treatments were drained after 1 week, and theexperiment repeated for a period of 4 weeks. It was found thatthe humus in the saturated columns produced much more phenolics139than the drained columns. These results suggest that the wettermonths should produce more phenolic acids, as is the case for thevanillic, protocatechuic and hydroxybenzoic acids. However, thisdoes not explain why the syringic and p—coumaric acids reachtheir peak concentrations in the drier, summer months. Thesedrier months are coincident with production of flowers and fruitsby the salal plant. It is interesting to note that the syringicacid concentration in woody humus is as high in September as thenon—woody humus. This may be evidence that salal is directlyproducing syringic acid which leaches down into lower humuslayers.A seasonal pattern of phenolic acid concentrations was alsofound by Jalal and Read (1983). The concentrations cannot bedirectly compared because they used an alkali extraction whichextracts both H—bonded and free phenolic acids. Concentrationswere found to be highest in May and through the summer months,and were lowest in winter months. The seasonal nature of acidproduction was attributed to the effects of temperature bothdirectly upon microbial activity and indirectly upon soil waterstatus. Increasing aeration and temperatures increased fungalmetabolism and growth which were responsible for the productionof organic acids from the fatty and phenolic acid rich residuesof the healthland raw humus.These results suggest several explanations. First,protocatechuic, hydroxybenzoic and vanillic acids are probably140readily utilized by microbial populations, so that during thecolder, wetter months, microbial metabolism has slowedsufficiently to allow accumulation of these acids. Oncetemperature, aeration and microbial metabolism increases, theconcentration of these phenolic acids decreases.In contrast, the higher concentrations of syringyl andcoumaryl derivatives in the summer months may result from 3factors: a) phenolic acids associated with salal increase inproduction when salal is more physiologically active, b) therelease of these phenolic acids occurs through microbialdecomposition of the phenolic acid rich residues of the salal orother plant litter, or c) these phenolic acids are broken downless easily than the other three and are therefore more readilyavailable and detected. Some or all of these factors may beinvolved.In summary, seasonal concentrations of free phenolic acidsunder salal suggests that salal may be in part responsible forthe production of free phenolic acids, either directly or throughmicrobial metabolism of its phenolic acid rich litter,particularly syringic and coumaric acids. Whether theseconcentrations are sufficient to elicit an allelopathic responseby conifers is now examined.1413.2. Effects of Phenolic Acids and Salal Leachates onConifers3.2.1 Seed germinationThe overall germination performance of the seeds variedgreatly with species. Germination performance was good for theSitka spruce, but low for western red cedar, and very poor forwestern hemlock. This probably occurred because of thedifferences in age of the seeds, the Sitka spruce being theyoungest and the western hemlock the oldest. Comparisons cannevertheless be made within species between treatments andpretreatments.The effect of the solutions varied with species andpretreatment. Germination capacity (GC) was not significantlydifferent by treatment for those seeds which were unstratified,regardless of species, but stratification resulted in treatmentshaving significant effects within species (Table 5.2). For Sitkaspruce, the stratification treatment resulted in significantlygreater GC than the unstratified treatment. This is consistentwith the need for stratification treatment in Sitka spruce. Forseeds given the stratified pretreatment, the phenolic solutionresulted in significantly lower GC than the control and soilsolutions, and the salal treatment was significantly lower thanall other treatments.For western red cedar, the effect of pretreatment gave nosignificant difference in GC between the control and soilsolution treatments, but gave significantly lower GC for the142Table 5.2: Germination indices by species, pretreatment andtreatment. Treatments with similar lower caseletters are not significantly different betweenpretreatments; treatments with similar upper caseletters are not significantly different withinpretreatments.a) R50Species Pretreatment Treatment_____________Phenolic Soil Salal ControlSs Strat 9.3 dB 8.9 deC 13.4 bA 8.7 eCUnstrat 12.8 cB 13.5 bB 15.9 aA 12.9 cBWh Strat 0 0 0 0Unstrat 0 0 0 0Wrc Strat 12.6 cB 9.7 dC 18.1 aA 10.2 dCUnstrat 14.0 bcA 12.6 cB 14.6 bA 12.4 cBb) Germination CapacitySpecies Pretreatment TreatmentPhenolic Soil Salal ControlSs Strat 92.3 aA 95.5 aA 94.3 aA 94.0 aAUnstrat 90.0 abA 84.8 bA 85.0 bA 90.0 abAWh Strat 27.8 abA 33.3 aA 19.8 bB 26.5 abAUnstrat 24.5 abA 26.0 abA 27.0 abA 30.0 abAWrc Strat 65.0 bB 76.5 aA 55.8 cC 77.3 aAUnstrat 74.5 aA 71.3 abA 70.5 abA 75.8 aAc) Germination ValueSpecies Pretreatment Treatment___Phenolic Soil Salal ControlSs Strat 31.9 bB 36.2 aA 24.9 cC 35.9 aAUnstrat 21.6 dA 17.6 eB 16.7 eB 21.0 dAWh Strat 0.7 abAB 1.2 aA 0.3 bB 0.7 abABUnstrat 0.5 abA 0.6 abA 0.7 abA 0.8 abAWrc Strat 13.1 bB 22.1 aA 7.6 cC 21.1 aAUnstrat 14.9 bAB 14.8 bAB 12.8 bB 16.8 bA143stratified pretreatment in the phenolic and salal solutiontreatments. This is presumably the effect of stratification,which actually stresses the seeds. Those seeds that werestressed by stratification, were then further stressed by thephenolic and salal solution, to a point where they could notgerminate.For western hemlock, there was no difference in GC withpretreatment. Those seeds given a stratification treatment, thesalal treatment was the only treatment with a CC significantlylower than the control.The R50 value, which is a measure of the rate of germination,could not be calculated for western hemlock because thegermination capacity was so low that 50% germination was notreached. For Sitka spruce and western red cedar, stratificationresulted in a significantly lower R50 than no pretreatment for allsolution treatments. The R50 was not significantly differentbetween the soil and control treatments regardless ofpretreatment, and the soil treatment was slightly, but notsignificantly lower for the stratified pretreatment. For Sitkaspruce, the phenolic treatment was significantly lower than thesoil and control treatments for stratified, but not forunstratified. The salal solution had significantly lower R50 thanall other treatments regardless of pretreatment. For western redcedar, the phenolic solution had a significantly lower R50 thancontrol and soil solutions, for both pretreatments, and the salal144solution was significantly lower than the phenolic solution forthe stratified treatment only.The GV, is a factor which combines rate of germination withgermination capacity. The GV for the soil solution was higherthan that of the control for the stratified pretreatment,regardless of species, although the differences were not alwayssignificant. This indicates that the soil solution had somebenefit on germination. For western hemlock, there was nosignificant difference between pretreatment, and no significantdifferences between treatment solutions for unstratified seedsonly. However, for the stratified seeds, the GV wassignificantly different between the soil solution and the salalsolutions only. For stratified Sitka spruce and western redcedar, the phenolic solution had a lower GV than the control andsoil solutions, and the salal solution was significantly lowerthan the phenolic solution.In summary, stratification tended to give faster germinationrates for the species tested, but resulted in lower germinationcapacities for western red cedar, higher germination capacitiesfor Sitka spruce, and no difference in western hemlock. Thestratified seeds were more susceptible to the treatment effects.None of the germination indices were significantly differentbetween soil and control treatments; for this reason and becauseof the difficulties involved in obtaining adequate quantities ofsoil solution, the soil solution treatment was not continued in145subsequent bioassays. Western red cedar was always significantlymore affected by the salal solution, than by the phenolicsolution, but the phenolic solution was still significantlydifferent than the control solutions. For Sitka spruce, thissame trend was apparent for the R50 and the GV indices. Forwestern hemlock, the salal treatment gave the lowest GC and GV,although they were not significantly different from the control.3.2.2 seedling growthResults of the seedling biomass study are presented in Table5.3. There were no significant differences in root biomass, butshoot biomass for salal treatment was significantly lower thanthe control treatment for all species. Shoot biomass for thephenolic treatment was significantly lower than for the controltreatment for Sitka spruce only, and was significantly greaterthan the salal treatment for Sitka spruce and western red cedar.Total biomass for Sitka spruce was significantly lower than thecontrol for both the phenolic and salal treatments. For westernhemlock, biomass of seedlings given the salal leachate solutionwas significantly lower than the control, but the phenolic acidtreatment was not significantly different from control or salaltreatments. Biomass of western red cedar given the salaltreatment was significantly lower than both control and phenolictreatments.It is important to consider that the various treatments had146different pH values: control, pH 6; phenolic solution, pH 5; andleachate solution, pH 4. The effect of the solution pH cannot beseparated from the effects of the compounds within the solution.The natural soil solution pH was found to be 3.9 for Fm horizonsand 3.7 for Hw horizons, so that of the leachate solution is mostsimilar to natural pH. This means though that the controlsolution which provides a more favourable pH, may be biasing theresults. However, these differences in pH may reflect naturalconditions; rainfall falling directly onto forest soils would beclose to neutrality; rainfall passing over salal biomass beforereaching the soil, could contain leachates at a much lower pH;and soil solution pH is near 4. However, these effects would beimportant and must be considered when discussing the varioustreatment effects.Also, the treatment effects may be exacerbated because asthe rooting medium dries, the solutions would concentrate,thereby leading to concentrations of potential toxins that aremuch higher than used at the time of watering. A similar effectcould be seen under field conditions as soils dry out in thesummer, but the drying would be much more gradual.3.2.3 seedling root bioassayShort term uptake of inorganic phosphorus (Pi) in distilledwater by roots of seedlings which had been germinated and wateredwith solutions of phenolic acids or salal berries and flowers is147Table 5.3: Seedling shoot, root and total biomass (mg) anduptake of inorganic P (nanoMoles per g oven—dry root)by species and treatment. Treatments followed by thesame letter are not significantly different withinspecies.Species Treatment Biomass Pi UptakeShoot Root Total(mg) (mg) (mg) (nNol/ci)Ss Control 22 c 10 a 31 b 15.93aSalal 14 a 9 a 23 a 19.88aPhenolic 18 b 8 a 25 a 17.77aWh Control 24 b 7 a 31 b 6.21aSalal 11 a 6 a 17 a 6.84aPhenolic 15 ab 7 a 22 ab 7.37aWrc Control 13 b 6 a 19 b 4.44bSalal 8 a 5 a 13 a 5.39bPhenolic 13 b 5 a 18 b 1.50a148shown in Table 5.3. There was no significant difference betweentreatments for Sitka spruce or western hemlock, but the phenolictreatment resulted in significantly lower uptake of Pi forwestern red cedar. The lack of significance for the phenolictreatment on Pi uptake for Sitka spruce and western hemlock isconsistent with studies by Glass and Dunlop (1974) and Harper andBalke (1981) who found that the inhibition of Pi or K uptake byphenolic acids was almost immediately reversed on removal of thephenolic solution.Uptake of Pi was higher (but not significantly so), for thesalal and phenolic treatments than for the controls. This can beexplained by the widely observed phenomenon that plants have thecapacity to increase transport capacity in response to lownutrient availability (Glass, 1989). This suggests that the Piuptake of the seedlings given the phenolic acid or salal leachatesolutions were below the optimum level such that when theseedlings were removed from that environment, uptake increasedimmediately. Uptake may be below the optimum because the phenoland salal treatment are either reducing ion uptake directly, orthey inhibit mobilization of the Pi in the soil, or that the pHdifferences are affecting uptake. This is examined further in thenext section.The significant reduction in Pi uptake for the phenolictreatment in western red cedar suggests that root damage may haveoccurred. Harper and Balke (1981) found that absorption of149salicylic acid at a low pH of 4.5 caused sufficient membranedamage to allow leakage of ions and organic solutes out of theroot cells of oats. The pH of the greenhouse growing medium inthis study was about 5. That only western red cedar was affectedmay reflect the fact that species differ in response to phenolicacid additions (Harper and Balke, 1981). This suggests thateffects of salal leachates and phenolic acids at maximum soilsolution concentrations found, do not have irreversible effectson the Pi uptake capacity in Sitka spruce or western hemlockroots, but the phenolic acid concentration may impair rootfunctioning in western red cedar.Some problems with the technique of McDonald (1992)used for this bioassay became apparent. Adequate aeration,important for oxidative respiration of roots, was not maintainedduring the immersion of the roots in the solutions, as isnormally done (Harper and Balke, 1981; Glass, 1974).Furthermore, the roots were washed in distilled water rather thanin a similar unlabelled solution, which would tend to cause animmediate change in ion uptake by passive diffusion. Theseproblems are not addressed in the published methods, but shouldbe considered in future studies.3.2.4 Mature Root BioassayUptake of inorganic phosphorus (Pi) by mature roots in thevarious solutions is shown in Table 5.4. Again, it is important150to consider the pH effects of the treatments, as well as thesolutions themselves. For western red cedar the salal leachatesolution, which had the lowest pH, had the greatest effect bycausing a significant reduction in Pi uptake to about 15% ofcontrol levels; the phenolic acid solution caused a significantreduction in uptake to about 36% of control levels. For westernhemlock, the salal treatment caused a significant reduction in Piuptake to about 9% of control level; the phenol treatment causeda reduction in uptake to 69% of control levels, but this was notsignificantly different from the control.The fact that western red cedar is more strongly affected bythe treatments than western hemlock is consistent with theearlier findings for roots of western red cedar seedlings. Thegreater reduction of uptake by the salal leachate solution overthat of the phenolic acid solution may result not only fromcompounds in the solution, but also from the greater acidity (pH4 versus pH 5 of the phenolic solution and pH 6 of the distilledwater). Ideally, the pH should have been kept constant betweenthe solutions. However, the lower acidity is not different fromthat of the pH of soil solution occurring naturally, varyingbetween 3.5 and 4.1. In fact, the effect of the phenolictreatment may be even greater under natural soil pH conditions.Harper and Balke (1981) examined the interactions of the effecton K uptake of phenolic concentration and pH and found that muchlower phenolic concentrations were required to produce the same151Table 5.4: Root biomass (mg) and uptake of inorganic phosphorus(Pi) in nanoMoles per oven—dry g of mature roots.Treatments with the same letters are not significantlydifferent within species.Species Treatment Root Pi RelativeBiomass Uptake Uptake(mci) (nMol Pi/ci) (% Control)Wrc Control 33 a 39.29 c 100.0Salal 25 b 5.82 a 14.8Phenolic 33 a 14.00 b 35.6Wh Control 24 a 30.00 b 100.0Salal 28 ab 2.80 a 9.3Phenolic 31 b 20.55 b 68.5152effect at low pH as that of higher concentrations of phenolics athigher pH.4.0 SUMMARYThe concentration of free phenolic acids under salal werefound to vary with horizon and season. The yearly meanconcentrations of free phenolic acids in soil solution undervigorously growing salal in young conifer plantation, wassignificantly higher in non-woody humus materials than woodyhumus for all phenolic acids tested except p-hydroxybenzoic acidwhich did not differ significantly. Concentrations of vanillic,protocatechuic and p—hydroxybenzoic acids were significantlyhigher in colder, wetter months, while that of syringic and pcoumaric were higher in warmer, drier months. Syringic acidconcentration was found to be as high in the woody humusmaterials as in the non—woody humus in the driest month. Thissuggests that syringyl and coumaryl units originating from salalare at their highest concentrations when the young conifer treesare most drought stressed.The effect of the phenolic acids at soil solutionconcentrations and of salal leachates on conifers was thenexamined. It is important to remember that the solutions werenot buffered, so the overall treatment effects includes both thepH and allelochemical effects. Germination of stratified westernred cedar and Sitka spruce seeds was significantly affected by153both the phenolic solution and the salal leachates, with thesalal treatment having significantly more effect than thephenolic treatment. Western hemlock did not appear to besignificantly affected, but this may reflect poor seed stock morethan actual treatment affects.The effects of phenolic acids at pH 5 and salal leachatesat pH 4 were tested on total biomass of seedlings. After twelveweeks growth, there was no significant difference in root biomassbetween treatments for any of the species tested. However, totalbiomass was significantly lower for the salal leachate treatmentthan for control treatments for all species tested. Totalbiomass was lower for the phenolics but significantly differentfrom the controls only for Sitka spruce. Longer term treatmentscould produce more significant differences.The 32P root bioassay of the mature conifer roots indicatesthat short-term uptake of Pi is affected by the nanomoleconcentration of phenolic acids found in soil solution at pH 5.For western red cedar, uptake was only was 36% of the controlsolution which was highly significant. For western hemlock,uptake was 69% of the control, but this was not highlysignificant. The effect of the salal leachates at pH 4 is evenmore dramatic, with only 15% and 9% of control uptake in westernred cedar and western hemlock respectively.The effect of the treatments over a twelve week period onroot uptake in distilled water were not significantly different154for western hemlock or Sitka spruce. This reversibility of thephenolic effect (as compared with significant short term effectsas tested on the mature roots) is in agreement with Glass andDunlop (1974), who found that the effect of the phenolics is mostlikely at the root membrane, since recovery was almost immediatewhen roots were transferred to phenolic—free solutions. Thesignificantly lower uptake in the phenolic treatment of westernred cedar indicates that irreversible root dysfunctioning mayhave occurred. Root damage by phenolic acids at low pH was alsofound by Harper and Balke (1981). The inorganic phosphorusuptake in the phenolic and salal leachate treatments tended to behigher than the controls, which may indicate that the seedlingswere deficient in Pi relative to the controls.5.0 CONCLUSION5.1 Seasonal Phenolic Acid Concentrations Under SalalHypothesis: That the concentration of free phenolic acidsfound under salal in plantations will vary with season andwill be highest during the summer months when salal is mostphysiologically active.Conclusion: Accept that concentrations vary withseason, and that syringic and p—coumaric acids arehighest during the summer months.1555.2 Effects of Phenolics on ConifersHypothesis: That solutions using the maximum concentrationsof free phenolic acids found in soils under salal and ofleachates from the flowers and berries of salal, will causereduced seed germination, biomass growth, and 32P uptake inroots of Sitka spruce, western hemlock and western redcedar.Conclusion: Accept156CHAPTER VIOverall Discussion1.0 INTRODUCTIONThe causes for growth—check of regenerating conifers on theCH phase, but not the HA, on Vancouver Island involves manyinteracting factors. The differences in productivity begin longbefore the stands are cut, so it is important to understand thebasic differences between CH and HA phases.The terms CH and HA phases were first used by Lewis (1982)to describe site differences within the Thula plicata - Tsugaheterophylla - Gaultheria shallon - Rhytidiadeiphus loreus or“salal — moss” ecosystem of the windward, submontane, wettervariant of the wet subzone (CWHvm) of the Coastal Western Hemlockbiogeoclimatic zone on northern Vancouver Island. This singleecosystem association was subdivided based on what appeared to besuccessional differences. The cedar—hemlock (CH) phase, the157climatic climax community, consisted of old (perhaps 1000 years),somewhat open, western red cedar and western hemlock stands witha minor component of amabalis fir and a dense understorey ofsalal. The hemlock-amabalis fir (HA) phase, a seral stageoccurring on sites with a history of soil disturbance, consistedof young (about 80 years) even—aged, densely stocked, westernhemlock and amabalis fir, with an understorey largely composed ofmosses and sparse shrub and herb cover. The occurrence ofcommunities intermediate between the climatic climax CII phase andthe seral HA phase lead Lewis (1982) to state that the HA phasehad the potential to develop into the CH phase, given asufficiently long period of time without soil disturbance.In this study, the CH and HA phases were found to occur sideby side, but the HA phase occurred on drier ridgetops, while theCH phase were directly adjacent on lower, wetter slopes. Thisdifference alone profoundly affects the site history, standstructure, and corresponding differences in physical and chemicalsoil properties between CII and HA phases. The higher topographicposition of the HA phase, makes the site more susceptible towindthrow. The most recent catastrophic winds occurred in 1908,and most of the HA stands examined in this study originated fromthis single wind event. The dense, even—aged structure rendersthe stands susceptible to subsequent wind—throw events, and theprocess of stand renewal to even—aged conditions is repeated. Inthese stands, salal is rarely seen and other understoreyvegetation is sparse, presumably because the dense structure of158the stand prevents light penetration to the forest floor. Incontrast, the undisturbed CH stands are relatively open anduneven—aged, which encourages windfirmness (Oliver and Larsen,1991).Windthrow disturbances were found to be very important incausing differences in pedogenic processes. Windthrown treestend to create pits and mounds and mix the soil horizons (Bowers,1987). The mixing of the soil by windthrown trees has been foundto contribute to improved soil productivity by favouring organicmatter decomposition, releasing tied—up nutrients, aerating thesoil, breaking hardpans and reversing the process ofpodsolization (Bowers, 1987). The differences in pedogenicprocesses resulting from windthrow are seen in soils from CH andHA phases.The forest floor humus profiles were found to consist of sixmajor humus horizons, three of woody origin and three of non—woody origin, varying in degree of humification. The woodyhorizons (given the suffix w) included the Fw, Hrw and Hw. TheFw, consisted of over 90% poorly decomposed wood in which thewoody structure held when rubbed between the fingers. The Hrw or“residuic” woody H horizon consisted largely of rooting wood, butthe amorphous component was greater than 20%, and the woodystructure failed when rubbed between the fingers. The Hw, orhumified woody horizon, consisted of less than 20% woody and over80% amorphous materials with a crumbly, greasy structure. Thenon-woody horizons included the Fm, Hh and Hhi. The Fm159(mycogenous F horizon) consisted of a mix of plant, fungi andamorphous materials with a compact and matted structure. The Hw(humified H horizon) consisted of greater than 80% amorphousmaterials and has a massive, block and greasy structure. The Hhi(humified intrusive horizon) was very black and greasy consistingof over 95% amorphous materials and tended to occur only at theforest floor—mineral soil interface.The classification of humus horizons in this study hadexcellent correspondence with the spectra of ‘3C NMR. Woodyhorizons were dominated by signals from lignin, but withincreasing decomposition, the relative proportion of lignindecreased, while aliphatics and carbohydrates increased,presumably from microbial. The non—woody humus types weretypical of forest litter layers, which are dominated by signalsin the carbohydrate region. Increasing decomposition resulted indecreasing carbohydrates and increasing aliphatics and carboxyl.Although all six humus horizons were found to occur on bothCH and HA sites, their relative abundance was found to vary.This variation was found to be associated with site history. Thewindthrow process was responsible for the deposition of a largeamount of woody material on HA sites. Poorly decomposed woodyhorizons (Fw and Hrw) were found to occur on 68% of HA plots, butonly 38% of CH plots. Well decomposed 11w horizons were found on56% and 38% of HA and CH plots respectively. Where they didoccur, there was not significant difference in depth of woody160horizons between CH and HA. The non—woody horizons Fm, Hh andHhi were deeper and occurred more often on the CH plots.Soil profiles along a trench through one HA site consistedlargely of lignomors (41%) with greater than 35% woody debris.Within the lignomors was an equal proportion of young (Fw andHrw) and old (Hw) horizons indicating that the windthrow processhas been repetitive. HA phase soils consisted of a largeproportion of hemimors (37%), characterised by thin Fm horizonsover windthrow—disturbed, mixed organic—mineral mounds. Only 22%of the trench consisted of humimors, characteristic of mature,undisturbed profiles. Mineral soils were friable, well—aeratedand well—drained. Pans were found under the humimors, but thesetended to be thin and discontinuous.In contrast, soil profiles along a trench through a CH standconsisted of a relatively large proportion of humimors (57%),with characteristic well developed H horizons reflecting maturityand lack of disturbance. Lignomors comprised 17% of the soilswith a larger proportion of older (Hw) horizons, indicatingrelatively old woody inputs. Hydromors, which develop under theinfluence of excessive moisture on poorly drained soils, werealso common (24%). Hydromors are poorly aerated humus formswhich develop, in part, under hydrolysis and periodic anaerobicfermentation, which results from standing water near the mineralsoil surface. The mineral soils of the CH trench tended to becompact, with continuous, thin pans over which humus—enriched Bflayers were found, reflecting a lack of disturbance.161The inherently different site histories and their associatedpedogenic processes make the CH and HA sites quite different interms of potential productivity. Root restricting layers arisingfrom water, compacted soils or pans tended to be much morefrequent on CH than HA sites (98% vs. 70% respectively) with theresult that the CH sites had shallower mineral soil rootingdepths than HA (21 vs. 29 cm respectively). Also, the presenceof gleyed horizons occurred much more frequently on CH than HA(30% vs. 12% respectively).The complexities of the origin (woody vs. non-woody), stateof decomposition, and depths of forest floor profiles made bulksampling for nutrient analysis unfeasible. Therefore, each ofthe six horizon identified were sampled for chemical analysis.Nutrient concentrations were found to vary between horizons. Notsurprisingly, the woody humus was found to be significantly lowerthan non-woody humus for pH, total N and S, C/N ratio, availableN, S and P, and total inorganic K and Mn. Nutrientconcentrations tended to increase with increasing decompositionof woody horizons from Fw to Hrw to Hw. The Hw horizons werefound to have abundant fine roots, and were thus the mostbiologically important of the woody horizons. Nutrientconcentrations within the Hw horizon were found to differ betweenCH and HA sites; the pH, moisture content and C/N ratio of thosefrom the CH were higher than those from the HA, while total N andS, and available N and P were significantly higher for the HAcompared to the CH. The higher pH of the CH probably reflects162the higher pH of litter associated with CH sites, such as westernred cedar and salal. The higher moisture content is notsurprising, giving the earlier findings that the CH tends tooccur on wetter sites. But the nutrient concentrationdifferences seems to indicate that the CH site is eitherdecomposing at a slower rate than the HA, or that decompositionis less complete. Again, this may be a function of the wetter CHsite, or to differences in litter inputs.For non—woody horizons, the Fm horizon, which is the mostbiologically active horizon in terms of root abundance, was foundto be significantly higher than other non-woody horizons inavailable P and exchangeable K and Mn. There were no significantdifferences in nutrient concentrations between Fm and Hhhorizons. The Hhi horizon had significantly higher ash contents,and total C, N and S. The high ash content reflects the factthat the Hhi is located at the humus-mineral soil interface. Thehigher concentration of total C, N and S probably reflects theadvanced state of humification. The Hhi horizon was felt to bethe least biologically active of all humus horizon.For the Fm horizon, samples from the CH sites were found tobe significantly higher than those from the HA for pH, moisturecontent, available 5, and exchangeable Ca, K, and Mn; this mostlikely reflects differences in litter input between the sites,and to the wetter conditions of the CH. For the Hhi horizon,samples from the CH sites were found to be significantly higherthan for the HA for pH and water content, but available P was163higher for HA than for CH.Lipids, which arise from undecomposed plant material andfrom the bodies of microfaunal organisms, ranged between 1 and9%. Not surprisingly, non-woody horizons tended to have higherlipid concentrations than woody horizons, and concentrationsincreased with increasing decomposition. The CH sites hadconsistently higher and more variable lipid concentrations thanHA sites. Similarly, C’3 NNR indicated that the more decomposedHrw, Hw and Hhi horizons from the CH site had a greater relativepercentage of C in the aliphatic region than did similar horizonsfrom the HA. This may occur for several reasons: theconcentrations of lipids originating from salal or western redcedar litter may be higher than that of litter found on the HA;the wetter site conditions of the CH may inhibit the breakdown oflipids by microorganisms leading to an accumulation; ormicroorganisms synthesize a larger quantity of lipids on CHsites.Total and labile polysaccharides and cellulose were found todecrease with increasing decomposition and were all found to behigher for CH horizons than for HA. The same trends forcellulose was also found using C’3 NMR. This suggest, as with thelipid and nutrient concentration differences between sites, thatthe CH site seems to be decomposing either at a slower rate thanthe HA, or that decomposition is less complete.Tannins were identified in the Fm layers of both CH and HAsites using C’3 NMR, but intensity was greater on samples from the164CH phase. Tannins were also identified in the leaves, roots,flowers and litter of salal using ‘3C NMR. Furthermore, the juiceof salal berries were found to contain large quantities oftannins. Both the flowers and berries may thus be a source ofreadily leachable tannins, particularly in cutovers, while litterinputs from leaves and roots provide a source of tannins throughdecomposition. The tannin from salal leaves was thought to be aeither a procyanidin or a procyanidin and prodeiphinidin mix.Tannins are known to reduce the biodegradability andhumification of organic matter by three processes: the productionof protein—tannin complexes which are much more resistant tomicrobial decomposition than unaltered proteins; the permeationand coating of nonproteins such as cellulose and hemicellulose bythe protein—tannin complexes, giving them considerable resistanceto microbial attack; and by the inactivation of enzymes importantin the process of decomposition. The ratio of total carbohydrateto lignin C (as found by C’3 NMR) tended to be higher in CH humushorizons, indicating that the carbohydrates may be more resistantto decomposition, possibly as a result of the tannins. Thuslower concentrations of available nutrients, and higherconcentrations of lipids, polysaccharides, and cellulose in CHhumus horizons may be occurring, in part, because of the effectsof tannins arising from the presence of salal on CH, but not HAsites. Other associated factors which could be affecting rate ofdecomposition are the wetter site conditions of the CH phase, and165also the presence of fatty acids, which were not examined in thisstudy.A comparison of bound phenolic acid concentrations betweenhumus horizons indicated that woody horizons tended to be higherin vanillic acid while the non-woody horizons tended to be higherin protocatechuic, p—coumaric and ferulic acids. The Fm horizon,with the greatest proportion of angiosperm inputs, had thehighest concentrations of protocatechuic acid, syringic acid (adegradation product of angiosperm lignin), and p—coumaric andferulic acids (thought to originate from the cutin of leaf andother plant tissues and from the suberin of roots); the Fw, whichis pure coniferous wood, had the highest concentration ofvanillic acid (a degradation product of coniferous lignin); theHhi, was found to be highest in p-hydroxybenzoic acid.Interestingly, the Fm, which is the most biologically activehorizon, was found to have the highest overall concentration ofphenolic acids, while the Hhi, which would be expected toaccumulate the mobile organic acids from overlying horizons, wasfound to have the second highest overall concentrations ofphenolic acids.When the CH site, with characteristically abundant salal wascompared with HA sites, with relatively sparse understoreyvegetation, the CH site was found to be significantly higher insyringic, p—coumaric and ferulic acids. The concentration of p—hydroxybenzoic acid, originating from suberin of roots, was notsignificantly different between sites, but tended to be higher on166CH sites than on HA.The concentration of free phenolic acids in soil solutionunder salal in clearcuts was examined. Average concentrationsover five sampling periods in one year, were found to be in therange of nanograms per gram of oven—dry soil. Woody horizons hadsignificantly lower phenolic acid concentrations than non—woodyhorizons. Concentrations in woody horizons did not varysignificantly over the year, with the exception of syringic acidwhich peaked very sharply in September to concentrations as highas that found in non—woody horizons. Since syringic acid occursin very small quantities in coniferous lignin, the highconcentrations could originate from either salal rooted in thewoody layer, or from above—ground leachates of salal.The concentrations of phenolic acids in the non—woodyhorizons varied significantly over the course of one year. Thewetter winter months had the highest concentrations of vanillic,protocatechuic, and ferulic acids. This probably occurredbecause microbial metabolism had slowed sufficiently to allowaccumulation of these acids. Once temperature, aeration, andmicrobial metabolism increased, concentrations of phenolic acidsdecreased. In contrast, the drier summer months had the highestconcentrations of syringic and p—coumaric acids. This may resultfrom several possible processes: a) these phenolic acids areassociated with salal and increase in production when salal ismost physiologically active as, for example, leaching from theflowers or berries; b) the release of these phenolic acids167through decomposition of salal litter exceeds the utilization bymicrobes in the summer, leading to higher concentrations, and c)these phenolic acids are less transient than the other phenolicacids. It is unclear which process would be more important.The low concentration of these phenolic acids made abioassay important. The effects of the maximum concentrations ofphenolic acids found in the field, and of a 5% solution of salalberry and flower leachates were examined on conifer seedgermination, seedling growth and root phosphorus uptake. It isimportant to remember that the solutions were not buffered, sothe overall treatment effects include both the pH andallelochemical effects.For Sitka spruce and western red cedar, the germinationvalue, which combines germination rate and capacity, wassignificantly lowest for the salal leachate solution, while thephenolic solution was significantly lower than the controlsolution. The results for western hemlock germination value werenot significant, but this may reflect poor seed source, since theseeds had been stored for over ten years. The lower germinationvalues could be significant for naturally regenerating seedlingsin clearcuts.Biomass of seedlings were affected by weekly watering withthe treatment solutions. Root biomass of Sitka spruce andwestern red cedar seedlings given the salal leachate or phenolicacid solution were smaller, but not significantly, than thosegiven the control solutions. For western hemlock seedlings, root168biomass of seedlings given the salal leachate treatment weresmaller, but not significantly, than those given the controlsolution.The salal leachate treatment resulted in shoot biomass ofall seedlings that was significantly smaller than those given thecontrol treatment; the phenolic treatment solution resulted inshoot biomass of Sitka spruce that was significantly smaller thanthose seedlings given the control treatment. The resultsindicate that seedling biomass can be significantly affected bythe salal leachate or soil phenolic acid solution. If fieldconditions were to result in the same trend towards smallerseedling biomass, this could be significant to the overallproductivity of the trees. Whether the effect of the phenolicacids and salal leachate treatment is direct or indirect wasexamined using radioactive phosphorus uptake in roots.When root samples taken from mature trees were placed intothe treatment solutions augmented with a 32P—labelled phosphorussolution (unbuffered), phosphorus uptake was significantly lowerin treatment solutions than for controls. The phenolic acidsolution reduced uptake to 36% and 69% of that of controls inwestern red cedar and western hemlock respectively. The tanninsolution had a even more pronounced effect, reducing phosphorusuptake to 15% and 9% of that of controls in western red cedar andwestern hemlock respectively. These short term effects appear tobe reversible over the long term, at least for Sitka spruce andwestern hemlock. When the roots of seedlings which had been169watered with the treatment solution were placed into alabelled phosphorus solution without the treatments, the short-term uptake was significantly lower for the phenolic acidtreatment in western red cedar only. The total inorganic Puptake in Sitka spruce and western hemlock were actually higherfor the treatments than for the controls (although notsignificantly so). The higher total inorganic uptake in thesalal and phenolic acid treatments can be explained by the widelyobserved phenomenon that plants have the capacity to increasetransport capacity in response to low nutrient’availability.This suggests that the seedlings were deficient in P and, giventhe new conditions of higher P concentrations, uptake increased.The phosphorus deficiency may have resulted from reduced ionuptake either directly at the root by phenolic acids or tanninleachates, (as was found in the mature root bioassay), orindirectly by inhibition of P mineralization in soils containingtannins (as was found in CH humus).The significant reduction in inorganic P uptake for thephenolic treatment in western red cedar suggests that root damagemay have occurred in the long term. This may be because thephenolic concentration would increase as soils dried out betweenwatering.The overall hypothesis of this study, that there aredifferences in chemical characteristics of similar humus horizonsbetween CH and HA sites, and that these differences may be thecause of the poor productivity of conifer regeneration on170clearcut CH sites is therefore accepted.The implications of this study for the management of CHphase cutovers are not straightforward. If the windthrow processis important in rejuvenating the site, then scarification shouldreplicate the process by breaking up hardpans, increasing soilaeration, and improving soil fertility by mixing mineral and soilhorizons. A trial at Port McNeill is presently underway whichexamines this treatment.Another alternative is to remove salal from the site, and toavoid situations in which salal is allowed to reestablish inopen—grown conditions. In practice, however, this is difficultor almost impossible to do. Slashburning, physical sitepreparation, prompt and dense conifer regeneration, and treefertilization to speed up crown closure are methods presentlyused to eliminate salal on CH sites. Some work has been done toeradicate salal using herbicides. Field research trials usingGarlon with a diesel oil delivery has been shown to be reasonablyeffective. It is unlikely that this technique will be widelyused in commercial forestry. Some evidence also exists that asolution of calcium nitrate at a rate of 1.13 micromoles per gramof soil, may reduce the buildup and toxicity of phenoliccompounds in agricultural soils (Farquharson j., 1990). Thismay occur because phenolic acid toxicity is greatest in itsundissociated form at low pHs of 3 to 4, or that calcium can aidin the adsorption of phenolics onto clay or humus particlesrendering them inactive (Rice, 1984). However, in a forestry171situation, rates of fertilization needed to get the necessaryincrease in pH could be excessive.One possible treatment, which has not been tried, would beto use red alder (Alnus rubra), a nitrogen—fixing, indigenoustree, as a nurse crop in plantation following scarification.This would improve the nitrogen balance of the site, add a largeannual source of readily decomposable litter, and contribute torapid shading of salal understorey. No trials have beeninitiated to date.Fertilization with nitrogen and phosphorus is the best knownsilvicultural tool to relieve the growth-check of coniferregeneration, even though salal responds well to fertilizationwith increased above— and below—ground vigour. Any toxic effectsof phenolic compounds are not as great following fertilization(Stowe and Osborne, 1980) and fertilization may also cause apriming effect, resulting in an increase in nitrogenmineralization and decomposition rates of humus. Furthermore,applications of fertilizer that result in an increase in thedensity of the conifer crown canopy cause a decline in the vigourand cover of the salal understorey.172CHAPTER VIISummary and Conclusion1.0 SUMMARY1. Differences between the CH and HA phases begin withdifferences in topographic position. The HA phase,which tends to occur on drier ridgetops, is moresusceptible to windthrow events than the lower, wettertopographic position of the CH phase. The resultingdifferences in stand structure are such that HA standsare dense, young, even—aged, while those of the CH areancient (possibly 1000 years since last disturbance)fairly open stands with abundant understorey vegetationdominated by salal.2. Six distinct humus horizons were found to occurcommonly on the CH and HA sites. They can be173distinguished on the basis of origin (woody vs. non-woody) and on the degree of decomposition (fermented towell—humified). Woody horizons were found to be lowerin nutrients, total and labile polysaccharides andbound phenolic acids than were non—woody horizons.Lipids were also lower in woody horizons, except thatthe Hw horizon had very high concentrations.3. CII sites were found to have a higher proportion of wellhumified woody and non—woody horizons, reflectingecosystem maturity and a lack of disturbance. The HAsites were found to have a greater proportion of so—called residuic” woody horizons, more friable mineralsoils, and less root restricting layers than the CIIsites, indicative of repetetive windthrow events.4. Although not consistently significant, the followingtrends in nutrient concentrations were apparent: CHhumus horizons differed from HA humus horizons in thatthey tended to have higher concentrations of K, Ca, Mnand available S, lipids, and total and labilepolysaccharides, as well as a higher pH. Some of thedifferences can be attributed to the greater inputs oflitter from western red cedar and salal on CII sites.The HA humus horizons were found to be higher inavailable N and P and tended to have a lower C/N ratio174for the more well decomposed horizons. These findingssuggest that for the same humus horizons, those fromthe HA sites are better able to achieve a more advancedstate of decomposition than the CH. The wetter siteconditions of the CH phase, or the presence of tanninsfrom salal are possible reasons why.5. Carbon-l3 Nuclear Magnetic Resonance confirmed themorphological distinction between the six humus typesand that these humus horizons are similar betweensites. Woody horizons were dominated by signals fromlignin, but with increasing decomposition, the relativeproportion of lignin decreased, while aliphatics andcarbohydrates increased, presumably from fungalsources. There were no apparent differences between CHand HA sites for woody horizons. Non—woody horizonswere dominated by signals from the carbohydrates butwith increasing decomposition, the relative proportionof carbohydrates decreased and aliphatics and carboxylincreased. The ratio of carbohydrates to lignin Ctended to be higher for CH horizons, indicating thatthe carbohydrates may be more resistant todecomposition than for the HA horizons.6. Tannin signals were found in the Fm horizons of both CHand HA, but the intensity was greater on the CH175samples. Tannin signals were very strong in salalroot, leaf, flower, litter and berries. The tannin isa proanthocyanidin, and has been tentatively identifiedas a mix of procyanidins and prodeiphinidins.7. Concentrations of free phenolic acids under salal oncutovers were found to vary with season.Concentrations were higher in non—woody horizons thanin woody horizons. Concentrations of vanillic,protocatechuic and p-hydroxybenzoic acids originatingfrom conifers, were significantly higher in the colder,wetter months. Concentrations of syringic and p—coumaric acids were significantly higher in warmer,drier months. In fact, the concentrations of syringicacid was found to be as high in the woody humus as inthe non-woody in the driest month. Syringic acid isnot a degradation product of coniferous lignin, andhigh concentrations are coincident with physiologicalactivity of salal, particularly flower and fruitproduction in the summer months.8. Phenolic acid solutions at field concentrations, and ata 5% salal flower/berry solution (unbuffered)significantly reduced the germination value of Sitkaspruce and western red cedar. Watering seedlings withthe salal leachate solution resulted in total biomass176of Sitka spruce, western red cedar and western hemlockthat was significantly lower than the control after 12weeks. The phenolic acid solution resulted in totalbiomass of all seedlings to be lower than controls butonly the Sitka spruce seedlings were significantlylower. Longer term trials could produce moresignificant results.9. The uptake of 32P by mature roots was significantlyreduced by the phenolic acid and salal leachatesolutions (unbuffered) to 15% and 36% of controlsrespectively for western red cedar and to 69% and 9%respectively for western hemlock.10. The uptake of 32P by excised roots of seedlings whichhad been watered with the phenolic acid and salalleachate solutions (unbuffered) were not significantlydifferent from controls for Sitka spruce and westernhemlock. This indicates that the phenolic effect isreversible. Uptake of inorganic P was actually higherthan the control, indicating the seedlings wereprobably deficient in P. Uptake of inorganic P bywestern red cedar was significantly lower than that forthe controls, indicating possible long term rootdamage.17711. Possible forest management techniques to improve siteproductivity on the CH phase may be to attempt toreplicate the windthrow process using scarification tobreak up hardpans, increase soil aeration, and improvesoil fertility by mixing mineral and soil horizons.The use of a nurse crop such as red alder, couldimprove the nitrogen balance of the site, add a largeannual source of readily decomposable litter, andcontribute to rapid shading of salal understorey.Fertilization with nitrogen and phosphorus is the bestknown silvicultural tool to relieve the growth—check ofconifer regeneration.2.0 CONCLUSIONThis study provides evidence to suggest that the growth-check of conifer regeneration involves a number of interactingfactors. These include: 1) the presence of hardpans andcompacted mineral soils which contribute to annual periods ofanaerobic soil conditions; 2) the presence of large pockets ofnutrient poor woody humus, and 3) the presence of tannins, lipidsand phenolic acids in active humus horizons which may becontributing to decreased decomposition, mineralization andnutrient uptake in trees. Site scarification to break uphardpans and mix mineral soil and humus horizons may improve siteproductivity in much the same way as natural windthrow processes.Fertilization with nitrogen and phosphorus are the best178management tools presently available to relieve the growth—checkof conifer regeneration.179LITERATURE CITEDAnonymous. 1985. International rules for seed testing. mt. SeedTesting Assoc. Seed Sci. Tech. 13: 299—355.Baldock, J.A., J.M. Oades, A.M. Vassallo and M.A. Wilson. 1990.Significance of microbial activity in soils asdemonstrated by solid-state 13C NNR. Env. Sci.Tech.24:527—530.Balke, N.E. 1985. Effects of allelochemicals on mineral uptake andassociated physiological processes. pp. 161-178. M.J.Comstock (ed.). The Chemistry of Allelopathy: BiochemicalInteractions Among Plants. Amer. Chem. Soc., Washington,D.C. 470 p.Ballester, A., J.M. Albo, and E. Vieitez. 1977. The allelopathicpotential of Erica scoparia L. Oecologia (Ben.) 30:55—61.Behre, K.E. 1986. Anthropogenic Indicators in Pollen Diagrams.A.A. Balkema. Boston. 158 p.Benoit, R.E. and R.L. Starkey. l968a. Inhibition of decompositionof cellulose and some other carbohydrates by tannins. SoilSci. 105: 291—296.Benoit, R.E. and R.L. Starkey. 1968b. Enzyme inactivation as afactor in the inhibition of decomposition of organic matter bytannins. Soil Sci. 105: 203—208.Bjorkman, E. 1970. Forest tree mycorrhiza - The conditions for itsformation and the significance for tree growth andafforestation. P1. Soil. 32: 589—610.Blanchette, R.A., T. Nilsson, G. Daniel and A. Abad. 1990.Biological degradation of wood. pp. 141-174 j R.M.Rowell and R.J. Barbour (eds.). Archaelogical Wood:Properties, Chemistry and Preservation. Adv. Chem. Ser.No. 225., Am. Chem. Soc. 286 p.Bowers, F.H. 1987. Effects of windthrow on soil properties andspatial variability in southeast Alaska. Ph.D.Dissertation. Coll. For. Res., U.Wa., Seattle, Wa. 153p.Braathe, P. 1950. Granas veksthemning pa lyngmark. Tidsskr.Skogbruk 58: 42-45.180Bradley, R., A.J. Burt and D.J. Read. 1982. The biology ofmycorrhiza in the Ericaceae. VIII. The role of mycorrhizalinfection in heavy metal resistance. New Phytol. 91: 197-209.Bremner, J.M. and C.S. Muiraney. 1982. Nitrogen-total. pp. 595-624 Jj, A.L. Page (ed.). Methods of Soil Analysis. Part2. Chemical and Microbiological Properties. Second Ed.Agronomy Monograph. No.9. Soil Sci. Soc. Mn., Madison,Wisc. 411 p.Carballeira, A. 1977. Phenolic inhibitors in Erica australis L.and in associated soil. J. Chem. Ecol. 6: 593—596.Carballeira, A. and A. Cuervo. 1980. Seasonal variation inallelopathic potential of soils from Erica australis L.heathiand. Oecol. Plant. 1(15): 345—353.Cheshire, M.V. 1979. Nature and Origin of Carbohydrates in Soils.Academia. N.Y. 255 p.Chou, C.H. and C.H. Muller. 1972. Allelopathic mechanisms ofArctostaphyllos glandulosa var. zacaensis. Am. Midl.Nat. 88: 324—347.Crawford, R.L. 1981. Lignin Biodegradation and Transformation.Wiley-Interscience, New York. 189 p.Conway, V.M. 1947. Ringinglow Bog near Sheffield. J. Ecol. 34:149—181.Czochanska, Z., L.Y. Foo, R.H. Newman and L.J. Porter. 1980.Polymeric proanthocyanidins. Stereochemistry, structuralunits and molecular weight. J.C.S. Perkin I.: 2278-2286.Damman, A.W.H. 1965. Thin iron pans: their occurrence and theconditions leading to their development. CanadaDep.For.Inform. Rep. N-X-2. 14 p.Darling, F.F. 1947. Natural History in the Highlands and Islands.Collins, London. 303p.Del Moral, R. and R.G. Cates. 1971. Allelopathic potential ofdominant vegetation of Western Washington. Ecology52(6): 1030— Montigny, L.E. 1983. The relation of fire to the assart effect.B.S.F. Thesis, University of B.C. Vancouver, B.C. 108 p.181de Montigny, L.E. and G.F. Weetman. 1990. The effects ofericaceous plants on forest productivity. pp. 83—90 jB.D. Titus, P.F. Newton, W.J. Meades (eds.). The Silvicsand Ecology of Boreal Spruces. 1989 IUFRO Working PartyS1.05-12 Symp. Proc., Newfoundland, 12—17 Aug. 1989. For.Can. Inf. Rep. N-X-271. 189 Montigny, L.E., C.M. Preston, P.G. Hatcher,and I. Kogel-Knabner.1992. Comparison of humus horizons from two ecosystemphases on Northern Vancouver Island using ‘3C CPMAS NNRspectroscopy and CuO Oxidation. Submitted to Can. J.Soil Sci.Dimbleby, G.W. 1962. The development of British heathlands andtheir soils. Oxf. For. Mem. 23: 1—121.Drijber, R.A. and L.E. Lowe. 1991. Bound phenolic acids in humusfractions of Andosols under varying vegetation in theSierra Nevada, Mexico. Geoderma 48: 431—442.Dubois, M., K.A. Gilles, J.K. Hamilton, P.A. Rebess and F. Smith.1956. Colorimetric method for determination of sugarsand related substances. Anal. Chem. 28: 350—356.Duchafour, P. 1977. Pedology-Pedogenesis and Classification.Trans. by T.R. Paton. George Allen and Unwin. London.448 p.Farquharson, B.J., R.P. Voroney, E.G. Beauchamp, and T.J. Vyn.1990. The use of calcium nitrate to reduce phytotoxinaccumulation during crop residue decomposition. Can. J.Soil Sci. 70: 723—726.Frank, W. and A. Schillinger. 1944. Zum Stoffwechsel dersauresfesten Bacterien. Biochem. Z. 316: 313—334.Fox, C.A., R. Trowbridge and C. Tarnocai. 1987. Classification,macromorphology and chemical characteristics of folisolsfrom British Columbia. Can. J. Soil Sci. 67: 765—778.Garrett, M.K. and P.M. Robinson. 1969. A stable inhibitor ofspore germination produced by fungi. Arch. Mikrobiol.67:370—377.Germain, A. 1985. Fertilization of stagnated Sitka spruceplantations on Northern Vancouver Island. MSc. thesis.Univ. B.C., Vancouver, B.C. 185 p.Gimingham, C.H. 1972. Ecology of Heathiands. Chapman and Hall.London. 266 p.182Glass, A.D.M. 1973. Influence of phenolic acids upon ion uptake.I. Inhibition of potassium uptake. Plant Physiol. 51:1037—1041.Glass, A.D.M. 1974. Influence of phenolic acids upon ion uptake.III. Inhibition of potassium absorption. J. Exp. Bot.89: 1104—1113.Glass, A.D.M. 1975. Inhibition of phosphate uptake in barley rootsby hydroxy—benzoic acids. Phytochem 14: 2127-2130.Glass, A.D.M. 1976. The allelopathic potential of phenolic acidsassociated with the rhizosphere of Pteridium aguilinum.Can. J. Bot. 54: 2440—2444.Glass, A.D.M. 1989. Plant Nutrition: An Introduction to CurrentConcepts. Jones and Bartlett. Boston. 234 p.Glass, A.D.M. and J. Dunlop. 1974. Influence of phenolic acidsupon ion uptake. IV. Depolarization of membranepotentials. Plant Physiol. 54: 855—58.Green, R.N., R.L. Trowbridge and K. Klinka. 1991. Towards ataxonomic classification of humus forms. Submitted to For.Sci.Grubb, P.J., H.E. Gren and R.C.J. Merrifield. 1969. The ecologyof chalkheath: its relevance to the caluiole—calcifugeand soil acidification problems. J. Ecol. 57: 175—212.Gustaven, K.H. 1956. The Chemistry of the Tanning Process.Academic Press, New York. 403 p.Haider, K. and J.P. Martin. 1975. Decomposition of specificallycarbon 14 labelled benzoic and cinnamic acid derivativesin soil. Soil Sci. Soc. Am. Proc. 39: 657—662.Hammond, T.E., D.G. Cory, W.M. Ritchey and H. Morita. 1985. Highresolution solid state ‘3C nuclear magnetic resonance.Org. Geochem. 11: 31-39.Handley, W.R.C. 1963. Mycorrhizal associations and Callunaheathland afforestation. For. Comm. Bull. No. 36.Commonwealth For. Inst. Oxford. 70 p.Harborne, J.B. and C.A. Williams. 1973. A chemotaxonomic survey offlavanoids and simple phenols in leaves of theEricaceae. Bot. J. Limn. Soc. 66: 37-54.183Harper, J.R. and N.E. Balke. 1981. Characterization of theinhibition of K absorption in oat roots by salicylic acid.Plant Physiol. 68: 1349—1353.Hatcher, P.G. 1987. Chemical structural studies of natural ligninby dipolar dephasing solid-state ‘3C nuclear magneticresonance. Org. Geochem. 11: 31—39.Hatcher, P.G. and M.A. Wilson. 1991. The effect of samplehydration on ‘3C CPMAS NMR spectra of fulvic acids. Org.Geochem. 17: 293—299.Hemmingson, J.A. and R.H. Newman. 1985. A CP/MAS 13C NMR study ofthe effect of steam explosion processes on woodcomposition and structure. J. Wood Chem. Tech. 5: 159—188.Hempfling, R., F. Ziegler, W. Zech and H. Schulten. 1987. Litterdecomposition and humification in acidic forest soilsstudies by chemical degradation, IR and NMR spectroscopyand pyrolysis field ionization mass spectrometry. Z.Pflanzenernaehr.Bodenk. 150: 179—186.Huang, P.M., T.S.C. Wang, M.K. Wang, M.H. Mu and N.W. Hsu. 1977.Retention of phenolic acids by noncrystallinehydroxyaluminum and iron compounds and clay minerals ofsoils. Soil Sci. 123: 213—219.Ivarson, K.C. and F.J. Sowden. 1962. Methods for analysis ofcarbohydrate material in soil. I. Colorimetricdetermination of uronic acids, hexoses and pentoses.Soil Sci. 94: 245—250.Jalal, M.A.F., D.J. Read and E. Haslem. 1982. Phenolic compositionand its seasonal variation in Calluna vulcfaris.Phytochemistry 21: 1397—1401.Jalal, M.A.F. and D.J. Read. 1983a. The organic acid compositionof Calluna heathiand soil with special reference tophyto- and fungitoxicity. I. Isolation andidentification of organic acids. Plant and Soil 70:257—272.Jalal, M.A.F. and D.J. Read. 1983b. The organic acid compositionof Calluna heathiand soil with special reference tophyto- and fungitoxicity. II. Monthly quantitativedetermination of the organic acid content of Calluna andspruce dominated soils. Plant and Soil 70: 273—286.Jonassen, H. 1950. Recent pollen sedimentation and Jutland heathdiagrams. Dansk. Bot. Ark. 13: 1-168.184Kaland, P.E. 1986. The origin and management of Norwegian coastalheaths as reflected by pollen analysis. pp. 19—36 In K.E.Behre (ed.). Anthropogenic Indicators in Pollen Diagrams.A.A. Balkema. Boston. 154 p.Kirk, T.K. 1984. Degradation of lignin. D.T. Gibson (ed.).Microbial Degradation of Organic Compounds. Dekker, NewYork. pp. 399-437.Klinka, K. 1976. Ecosystem units - their classification, mappingand interpretation in the University of British ColumbiaResearch Forest. Ph.D. thesis. Univ. B.C., Fac.For.,Vancouver, B.C.Kogel, I. 1986. Estimation and decomposition pattern of the lignincomponent in forest humus layers. Soil Biol. Biochem.18(6): 589—594.Kogel, I., R. Hempfling, W. Zech, P. Hatcher and H. Schulten.1988. Chemical composition of the organic matter in forestsoils: 1. Forest Litter. Soil Sci. 146: 124—136.Kogel—Knabner, I., W.Zech, and P. Hatcher. 1988. Chemicalcomposition of the organic matter in the forest soils: thehumus layer. Z.Pflanzenernahr.Bodenk. 151: 331—340.Kogel—Knabner, I., P.G. Hatcher and W. Zech. 1991. Chemicalstructural studies of forest soil humic acids:aromatic carbonfraction. Soil Sci. Soc. Am. J. 55: 241—247.Kolattukudy, P.E. 1980. Biopolyester membranes of plants: cutinand suberin. Science 208: 990—1000.Kowalenko, C.G. and L.E. Lowe. 1972. Observations on the bismuthsulphide colorimetric procedure for sulphate analysis insoil. Commun. Soil Sci. Plant Anal. 3: 79-86.Krosshaven, H., J.O. Bjorgum, J. Krane and E. Steinnes. 1990.Chemical structure of terrestrial humus materials formed fromdifferent vegetation characterized by solid-state 13C NMR withCP-Mass techniques. J. Soil Sci. 41: 371—377.Lavkulich, L. 1982. Methods Manual. Pedology Laboratory. Dep. SoilSci. University of B.C., Vancouver, B.C. 55 p.Leary, G.J., K.R. Morgan and R.H. Newman. 1986. A ‘3C CP/MAS NMRcomparison of wood fractions from spruce. Holzforschung40: 221—224.Lewis, T. 1982. Ecosystems of Port McNeill (Block 4) on Tree FarmLicence 25. Western Forest Products Ltd. Internal Report.185Long, J.N. and J. Turner. 1975. Aboveground biomass of understoreyand overstorey in an age sequence of four Douglas—fir stands.J. Appl. Ecol. 12:179—188.Lowe, L.E. 1974. A sequential extraction procedure for studyingthe distribution of organic fractions in forest humuslayers. Can. J. For. Res. 4: 446-454.Malcolm, D.C. 1975. The influence of heather on silviculturalpractice— an appraisal. Scot. For. 29: 14—24.Malcolm, D.C. 1987. Nitrogen supply for spruce on infertile sites(an ecological problem). Th Leslie L. Schaffer Lectureship inForest Science, Nov. 5, 1987. Faculty of Forestry, Universityof B.C., Vancouver, B.C. 23 p.Mallik, A. 1987. Allelopathic potential of Kalmia ancrustifoliato black spruce (Picea mariana). For. Ecol. Mgmt. 20:43—51.McDonald, M.A., D.C. Malcolm and A.F. Harrison. 1992. The use ofa 32P root bioassay to indicate the phosphorus status of foresttrees: 1. Seasonal variation. Can. J. For. Res. In Print.McKeague, J.A., A.W.H. Damman, and P.K. Herringa. 1968. Iron-manganese and other pans in some soils of Newfoundland.Can. J. Soil Sci. 48: 243—253.Meades, W.J. 1986. Successional status of dwarf shrub heath inEastern Newfoundland. Ph.D. Dissertation. University ofConnecticut.Mehlich, A. 1978. New extractant for soil test evaluation ofphosphorus, potassium, magnesium, calcium, sodium,manganese and zinc. Commun. Soil Sci. Plant Anal. 9:477—492.Messier, C. and J.P. Kimmins. 1990. Factors limiting coniferousseedling growth in recently clearcut sites dominated byGaultheria shallon in the CWHvm subzone. FRDA Rep. ISSN0835 0752; 149. B.C. Ministry of Forests. Victoria,B.C. 30 p.Minore, D. 1983. Western redcedar— a literature review. USDAFor. Serv. Res. Pap PNW-l50. 167 p.Mole, S. and P.G. Waterman. 1987a. A critical analysis oftechniques for measuring tannins in ecological studies.I. Techniques for chemically defining tannins. Oecologia72: 137—147.186Mole, S. and P.G. Waterman. 1987b. A critical analysis oftechniques for measuring tannins in ecological studies.II. Techniques for biochemically defining tannins.Oecologia 72: 148—156.Molisch, H. 1937. Der Einfluss einer Pflannze auf die andereAllelopahtie. Fischer Jena. In A.R. Putnam and C.S.Tang (eds.). 1986. The Science of Allelopathy. WileyInterscience. 317 p.Morgan, K.R. and R.H. Newman. 1986. Estimation of the tannincontent of eucalypts and other hardwoods by carbon—l3 nuclearmagnetic resonance. Appita 40: 450-454.Muller, C.H., R.B. Hanawalt and J.K. McPherson. 1968. Allelopathiccontrol of herb growth in the fire cycle of Californiachaparral. Bull. Torrey Bot. Club. 95: 225—231.Muller, P.E. 1897. De indvunde Erfaringer angaaende reneGrankulturer i gammel vestjydsk Bakkeokede. Beretning omen Skovbrugsekskursion til Jylland, 20-25 Sept., 1897.By V. Gjem and P. Barfoed. pp. 16-21. From: Malcolm,D.C. 1975. The influence of heather on silviculturalpractice— an appraisal. Scot. For. 29: 14-24.Newman, R. H. and J.A. Hemmingson. 1990. Determination of thedegree of cellulose crystallinity in wood by carbon-13nuclear magnetic resonance spectroscopy. Holzforschung44: 351—355.Norden, B. and B. Berg. 1990. A non-destructive method (solidstate ‘3C NNR) for determining organic chemical componentsof decomposing litter. Soil Biol. Biochem. 22: 271—275.Oliver, C.D. and B.C. Larson. 1990. Forest Stand Dynamics. McGrawHill Inc. N.Y. 467 p.Omell, L.G. 1952. Heden. Natur i Halland. Stockholm. 254 p.Opella, S.J. and M.H. Frey. 1979. Selection of nonprotonatedcarbon resonances in solid—state nuclear magneticresonance. J. Am. Chem. Soc. 101: 5854-5856.Overbeek, J. Van, and R. Blondeau. 1954. Mode of action ofphytotoxic oils. Weeds 3:55-65.Pedersen, T. 1970. Effects of fatty acids and methyl octanoate onresting mycelium of Boletus variegatus. Physiol. Plant.23:654—666.187Preston, C.M. and J.A. Ripmeester. 1983. ‘3C labelling for NMRstudies of soils: CPMAS NMR observation of 13C—acetatetransformation in a mineral soil. Can.J. Soil Sci. 63: 495-500.Preston, C.M. and A.C.M. Rusk. 1990. A bibliography of NNRapplication for forestry research. Pac. For. Centre Inf. Rep.BC—X—322. 42 p.Preston, C.M., S.-E. Shipitalo, R.L. Dudley, C.A. Fyfe, S.P. Mathurand M. Levesque. 1987. Comparison of 13C CPMAS NNR andchemical techniques for measuring the degree ofdecomposition in virgin and cultivated peat profiles.Can. J. Soil Sci. 67: 187—198.Preston, C.M., D.E. Axelson, M. Levesque, S.P. Mathur, R.L. Dudleyand H. Dinel. 1989. Carbon-13 CPMAS NMR and chemicalcharacterization of particle—size separates of peatsdiffering in degree of decomposition. Org. Geochem. 14:393—403.Preston, C.M., P. Sollins and B.G. Sayer. 1990. Changes inorganic components for fallen logs by ‘3C nuclear magneticresonance spectroscopy. Can. J. For. Res. 20: 1382—1391.Preston, C.M. and B.G. Sayer. 1992. What’s in a nutshell: aninvestigation of structure by carbon-l3 CPMAS NMRspectroscopy. Submitted to J. Agric. Food Chem.Preston, C.M., M. Schnitzer, R. Hempfling, H.R. Schulten, J.A.Trofymow and D.E. Axelson. 1992. Characterization oforganic matter in the soil profile in a young Douglas-firplantation. To be submitted to Plant and Soil.Putnam, A.R. and C.S. Tang. 1986. The Science of Allelopathy.Wiley—Interscience. 317 p.Read, D.J. 1984. Interactions between ericaceous plants and theircompetitors with special reference to soil toxicity.Asp. of App. Biol. 5: 195—209.Read, D.J. 1987. Recent advances in the study of ericaceousiuycorrhizae. pp. 328-330. D.M. Sylvie, L.L. Hung, and J.H.Graham (eds.). Mycorrhizae in the Next Decade. Inst. FoodAgric. Sci. Gainsville, Fla. 354 p.Rice, E.L. 1984. Allelopathy. 2nd Ed. Academic Press. Orlando,Fla. 422 p.188Robinson, R.K. 1972. The production by roots of Calluna of afactor inhibitory to growth of some mycorrhizal fungi.J. Ecol. 60: 219—224.Salasoo, I. 1981. Alkane distribution in epicuticular wax of someevergreen Ericaceae. Can. J. Bot. 59: 1189—1191.Salasoo, I. 1983a. Effect of leaf age on epicuticular wax alkanesin Rhododendron. Phytochem. 22: 461-463.Salasoo, I. 1983b. Alkane distribution in epicuticular wax ofEpacridaceae. Phytochem. 22: 937-942.Salasoo, I. 1987. Epicuticular wax alkanes of some heath plants incentral Alaska. Biochem. Syst. Ecol. 15: 105-107.Soil Survey Staff. 1975. Soil Taxonomy. U.S. Dept. Agric.Handbook No. 436. U.S. Govt Printing Office, Washington.357 p.Specht, R.L. 1978. Heathlands and related shrublands of the world.pp. 1-18. R.L. Specht (ed.). Heathiands and RelatedShrublands: Descriptive Studies. Elsevier. 214 p.Staneck, W., D. Beddows and D. State. 1979. Fertilization andthinning effects on a Douglas—fir ecosystem at Shawnigan Lakeon Vancouver Island: some observations on salal and brackenfern undergrowth. Can.For. Serv. INf. Rep. BC-R-1. Victoria,B.C. 29 p.Stevenson, F.J. 1982. Humus Chemistry: Genesis, Composition,Reactions. J. Wiley and Sons. New York. 443 p.Stowe, L.G. and A. Osborn. 1980. The influence of nitrogen andphosphorus levels on the phytotoxicity of phenolic compounds.Can J. Bot. 58: 1149—1153.Takijima, Y. 1964. Studies on organic acids in paddy field soilswith reference to their inhibitory effects on the growth ofrice plants. Soil Sci. Plant Nutr. Tokyo. 10:14—21.Tang, C.S. and C.C. Young. 1982. Collection and identification ofallelopathic compounds from the undisturbed roots system ofBigalta Limpograss (Hemarthria altissima). Plant Physiol. 69:155—159.Towers, G.H.N., A. Tse and W.S.G. Maas. 1966. Phenolic acids andphenolic glycosides of Gaultheria species. Phytochem. 5:677—681.189Tschager, A., H. Hilschner, S. Franz, U. Kull, and W. Larcher.1982. Jahreszeitliche Dynamik der Fettspeicherung vonLoiseleuria procumbens und anderen Ericacea der alpinen.Zwergstrauchheide Acta Oecol. 3: 119—134.Ugolini, F.C. and D.H. Mann. 1979. Biopedological origin ofpeatlands in South East Alaska. Nature. 281: 366—368.Vaughan, D., G.B. Sparling, and B.G. Ord. 1983. Amelioration ofthe phytotoxicity of phenolic acids by some microbes.Soil Biol. Biochem. 15: 613—614.Waring, S.A. and J.M. Bremner. 1964. Ammonium production in soilunder waterlogged conditions as an index of nitrogenavailability. Nature 201: 951—952.Weatherall, J. 1953. The checking of forest trees by heather.Forestry 26: 37-40.Webley, D.M., D.J. Eastwood, and C.H. Gimingham. 1952.Development of a soil microflora in relation to plantsuccession on sand—dunes, including the ‘rhizosphere’flora associated with colonizing species. J. Ecol. 40:168—178.Weetman, G.F., R.Fournier, J. Barker, E. Schnorbus—Panozzo, and A.Germain. 1989. Foliar analysis and response offertilized chiorotic Sitka spruce plantations on salal—dominated cedar—hemlock cutovers on Vancouver Island.Can. J. For. Res. 19 (12): 1501—1511.Weetman, G.F., R.Fournier, J. Barker, and E. Schnorbus—Panozzo.1989. Foliar analysis and response of fertilizedchlorotic western hemlock and western red cedarplantations on salal—dominated cedar—hemlock cutovers onVancouver Island. Can. J. For. Res. 19 (12): 1512—1520.Weetman, G.F., R. Fournier, E. Schnorbus—Panozzo, and J. Barker.1990. Post-burn nitrogen and phosphorus availability of deep-humus soils in coastal British Columbia cedar/hemlock forestsand the use of fertilization and salal eradication to restoreproductivity. pp. 451—499. In S.P. Gessel, D.S. Lacate, G.F.Weetman and R.F. Powers (eds.). Sustained Productivity ofForest Soils. Proc. 7th N.Am.For.Soils Conf. Proc. U.B.C.Faculty of Forestry Publication. 205 p.Whitehead, D.C., D. Hazel and R.D. Hartley. 1983. Bound phenoliccompounds in water extracts of soils, plant roots andleaf litter. Soil Biol. Biochem. 15(2): 133—136.Wilkinson, L. 1990. SYSTAT: The System for Statistics. SYSTATInc., Evanston, IL. 677 p.190Wilson, M.A. 1987. Techniques and Applications in Geochemistry andSoil Chemistry. Oxford, Pergamon Press. 353 p.Wilson, M.A. and P.G. Hatcher. 1988. Detection of tannins inmodern and fossil barks and in plant residues by high-resolution solid—state ‘3C nuclear magnetic resonance.Org. Geochem. 12: 539—546.Wyse, 0., J. Ludwig, and R. Joiner. 1945. The fungistatic andfungicidal action of fatty acids and related compounds. Arch.Biochem. 7: 415—425.Zackrisson, 0. and M.C. Nilsson. 1989. Allelopati och dessbetydelse pa svarforyngrade skogsmarker. Skogsfakta.Biologi och Skogsskotsel. No. 59. ISSN 0280—7416. 6 p.Zech, W. M. Johansson, L. Haumaier and R.L. Malcolm. 1987. CPMAS13C NMR and IR spectra of spruce and pine litter and of theKlason lignin fraction at different stages of decomposition.Z.Pflanzenernahr.Bodenk. 150: 262—265.Zech, W.. R. Hempfling, L. Haumaier, H.-R. Schulten and K. Haider.1990. Humification in subalpine Rendzinas: chemicalanalyses, IR and ‘3C NMR spectroscopy and pyrolysis-fieldionization mass spectrometry. Geoderma 47: 123—138.191


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