Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The role of microorganisms in the phenomenon of hemlock brownstain Kreber, Bernhard 1995

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

Item Metadata


831-ubc_1995-983019.pdf [ 18.18MB ]
JSON: 831-1.0088852.json
JSON-LD: 831-1.0088852-ld.json
RDF/XML (Pretty): 831-1.0088852-rdf.xml
RDF/JSON: 831-1.0088852-rdf.json
Turtle: 831-1.0088852-turtle.txt
N-Triples: 831-1.0088852-rdf-ntriples.txt
Original Record: 831-1.0088852-source.json
Full Text

Full Text

THE ROLE OF MICROORGANISMS IN THE PHENOMENON OF HEMLOCK BROWNSTAIN by BERNHARD KREBER B.Sc.(Wood Science), University of Hamburg, Germany M.Sc.(Forest Products), Oregon State University, U.S.A. A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Forestry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA March, 1995 ©BERNHARD KREBER In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. (Signature) ILJft K N Department of I ft M A ' The University of British Columbia Vancouver, Canada Date Y£_ LLLAS DE-6 (2/88) Abstract ii Hemlock brownstain, a discoloration which varies in type and intensity and occurs in amabilis fir (Abies amabilis (Dougl.) Forbes) and western hemlock (Tsuga heterophylla (Raf) Sarg.) was investigated because it is a serious problem in the high-value lumber markets. The objective of this study was to understand the causes of hemlock brownstain, more specifically with emphasis on the role of microorganisms, and to suggest means for its control. While hemlock brownstain can vary macroscopically, a similar microscopic distribution of the brown coloration was demonstrated to be mainly associated with parenchyma cells and to a lesser degree with longitudinal tracheids. The brown deposits which were frequently associated with hyphae and bacteria, contained catechin as shown histochemically. Inoculation of western hemlock sap and wood with fungi and bacteria produced brownstain in vitro. The brown colorations which can develop in western hemlock during seasoning, were then investigated in field studies on logs and lumber. Extensive log storage time was demonstrated to promote brownstain as was salt water storage of logs, the latter producing more brownstain than logs stored on land. Fungi were isolated from freshly felled logs and from sawn lumber and they were believed to represent an endemic wood microflora. A link was suggested between fungi and brownstain. Low solubility phenols were associated with iii brownstained regions when compared to non-stained areas and migration of phenols to the wood surface was indicated. Infection of western hemlock lumber with Ophiostoma piceae produced brown deposits but subsequent immunolabeling with a monoclonal antibody proved unsuccessful in linking the brownstain to the fungus. However, laboratory experiments demonstrated that sapstaining fungi can produce browning in western hemlock sap as they shifted the pH from 5 to near 7, which caused ionization and oxidation of phenols. Browning induced by fungi was inhibited when the sap was buffered in the acidic range. While browning did occur at neutral pH, it did not occur in the absence of oxygen. Browning was also demonstrated in 15 /xm sections when infected with 0. piceae, suggesting that browning can occur in the presence of very small amounts of colour precursors. Furthermore, light microscopy demonstrated a lack of pigmentation of the 0. piceae hyphae when grown on western hemlock which contrasted with the formation of pigmented hyphae (sapstain) when grown on lodgepole pine. This observation suggested that physiological factors associated with the presence of the fungus triggered brownstain in western hemlock and that hemlock brownstain was unrelated to pigment formation by the fungus itself. iv Catechin was demonstrated to play a major role in brownstain of western hemlock but involvement of other, unknown, sap constituents was also indicated. Information gained in this study suggested that a faster processing of western hemlock logs into sawn lumber may lessen the extent of brownstain problems. In addition, biocides supplemented with reducing agents and/or chelating agents and the use of a buffer to stabilize the pH of the wood surface, should be investigated. As a spin-off of this research, the physiological factors which greatly reduced pigmentation of the sapstaining fungus 0. piceae in western hemlock, should be further investigated. V TABLE OF CONTENTS Abstract ii Table of Contents v List of Figures x List of Tables xii Acknowledgement xiv Preface xvi 1.0 INTRODUCTION 1 2.0 DISCOLORATIONS OF HEM-FIR WOOD: A REVIEW OF THE MECHANISMS 3 2.1 Background of Non-microbial Stains 3 2.2 Economic Significance of Discolorations in Western Hemlock and Amabilis-Fir 5 2.3 Investigations of Hem-Fir Discolorations 6 2.4 Extractive Chemistry of Western Hemlock 13 2.5 Factors Possibly Contributing to Discolorations in Hem-Fir 18 2.5.1 Factors Inherent in Living Trees 19 2.5.2 Felling Season 21 2.5.3 Post-Mortem Factors 22 2.5.4 Log Age 23 2.5.5 Log Storage 25 2.5.6 Storage of Lumber 2 6 2.6 Summary 28 3.0 ADVANCES IN THE UNDERSTANDING OF HEMLOCK BROWNSTAIN 2 9 3.1 Objectives 29 3.2 Material and Methods 2 9 3.2.1 Microscopic Examinations 29 VI 3.2.2 Histochemical Examinations 30 3.2.3 Sap Experiments 3 0 3.2.4 Solid Wood Experiment 32 3 Results 34 3.3.1 Microscopic observations on discoloured hem-fir 34 3.3.2 Histochemical observations 40 3.3.3 Colour changes in inoculated sap 45 3.3.4 Microscopic production of brownstain in inoculated western hemlock 47 4 Discussion 51 5 Conclusions 60 0 MONITORING PRODUCTION OF BROWNSTAIN IN WESTERN HEMLOCK LOGS AND LUMBER DURING STORAGE 62 1 Objectives 62 2 Materials and Methods 62 4.2.1 Field work at Chamiss Bay 62 Logging site description 62 Selection of western hemlock trees 63 Initial sampling of western hemlock logs 63 Second sampling of western hemlock logs 64 Inspection of western hemlock logs after 9 months storage 65 Sawing of western hemlock logs 65 Inspection of sawn lumber after storage 66 4.2.2 Laboratory work using Chamiss Bays samples 66 Assessment of susceptibility to brownstain in disks 66 Selection of logs susceptible to brownstain 66 Sample preparation from selected disks 67 Initial isolation of fungi 69 Harvesting of sap 70 pH measurement 70 HPLC analysis of pressate 70 Total soluble phenol determination 70 Sample preparation from disks after 2 months log storage 71 Isolation of fungi and quantification of bacterial forming colonies 72 Sap analysis 73 V l l Results and Discussion 74 4.3.1 Production of brownstain in fresh and 2 month-old logs 74 4.3.2 Inspection of cross-cut ends after 9 month storage 7 8 4.3.3 Inspection of lumber after 2 months of outdoor storage 84 4.3.4 Isolation of microorganisms 88 4.3.5 Sap analysis 95 Conclusions 103 0 MICROFLORA AND TOTAL SOLUBLE PHENOLS ASSOCIATED WITH BROWNSTAIN IN WESTERN HEMLOCK LUMBER 105 1 Objective 105 2 Materials and Methods 105 5.2.1 Sampling at CIPA sawmill 105 5.2.2 Sample preparation and isolation of fungi 106 5.2.3 Determination of TSP content 108 5.2.4 Assessment of antisapstain treatment 108 3 Results and Discussion 109 5.3.1 Observations on brownstained lumber 109 5.3.2 Isolation of fungi 110 5.3.3 Sap analysis 113 4 Conclusions 117 0 IN VITRO PRODUCTION OF HEMLOCK BROWNSTAIN 118 1 Immunogold-silver staining of O. piceae when grown in western hemlock 118 6.1.1 Objective 118 6.1.2 Materials and Methods 118 Sample preparation 118 Preparation of fungal inoculum 120 Infection of western hemlock 120 Visual and microscopic examination of wood samples 121 Immunolabeling of 0. piceae # 3871 in western hemlock 121 Sap analysis 123 Vlll 6.1.3 Results and Discussion 124 6.1.4 Conclusions 13 0 2 Production of brownstain in western hemlock sap 131 6.2.1 Objective 131 6.2.2 Materials and methods 131 Sample preparation 131 Infection of sap 132 Sap analysis 132 6.2.3 Results and Discussion 135 6.2.4 Conclusions 147 3 Infection of wood sections with 0. piceae on a glass-slide 148 6.3.1 Objective 148 6.3.2 Materials and Methods 148 Slide preparation 148 Wood sections used 14 8 Infection of wood sections 149 Microscopic examination of sections 149 6.3.3 Results and Discussion 150 6.3.4 Conclusions 154 4 Infection of western hemlock and lodgepole pine with 0. piceae 155 6.4.1 Objective 155 6.4.2 Materials and Methods 155 Wood used 155 Preparation of inoculum and infection of wood 155 6.4.3 Results and Discussion 157 6.4.5 Conclusion 161 ix 7.0 ELUCIDATION OF THE MECHANISMS OF SAP BROWNING 162 7.1 Objective 162 7.2 Materials and Methods 162 7.2.1 Effect of pH on sap browning 162 7.2.2 Effect of oxygen on sap browning 162 7.2.3 Effect of heat on sap browning 163 7.2.4 Production of sap browning by heat and by pH alteration 164 7.2.5 Amendment of water with phenols 164 7.2.6 Amendments of sap with known phenols 165 7.2.7 Effect of buffer on sap browning 165 7.3 Results and Discussion 167 7.3.1 Effect of pH on browning 167 7.3.2 Effect of oxygen on browning 169 7.3.3 Effect of heat on browning 172 7.3.4 Production of sap browning by heat and by pH alteration 174 7.3.5 Amendment of water and sap with known phenols 177 7.3.5 Buffer experiment 180 7.4 Conclusions 183 8.0 Summary and Recommendations 184 9.0 Literature Cited 190 X List of Figures Figure la: Representative specimen showing grey stain on the wood surface 35 Figure lb: Representative specimen showing brownstain on the board end and on the wood surface 36 Figure lc: Representative specimen showing zebra stain on the wood surface 37 Figure 2: Abies amabilis. Radial section (63x), ray parenchyma cells with brown deposits 3 8 Figure 3: Tsuga heterophylla. Transverse section (lOOx), isolated tracheids with a filled lumen among stain-free tracheids 39 Figure 4: Tsuga heterophylla. Radial section (200x), hyphae enclosed in coloured deposits of a tracheid 41 Figure 5: Tsuga heterophylla. Radial section (lOOx), bacterial infection of tracheids 42 Figure 6: Tsuga heterophylla. Radial section (80x), red DMB reaction products observed in ray parenchyma cells 43 Figure 7: Tsuga heterophylla. Radial section (80x), red DMB reaction products observed in axial tracheids 44 Figure 8: Tsuga heterophylla. Radial section (320x), colourless globules observed in clean ray parenchyma cells 4 8 Figure 9: Tsuga heterophylla. Radial section (160x), presence of brown deposits in 0. piceae #3871 inoculated wood after 6 weeks 50 Figure 10: Tsuga heterophylla. Transverse section (250x), brown deposits in a pit connecting discoloured tracheids 53 Figure 11: Tsuga heterophylla. Transverse section (400x), hyphae surrounded by a brown sheath in an otherwise unstained tracheid 54 Figure 12: Flowchart showing sample regime of log #1 from brownstained and non-stained regions in fresh and two month old disks 68 XI Figure 13: Representative, water-stored log showing dark colorations after 9 months of storage 81 Figure 14: Representative, land-stored log showing brownstain in sapwood after 9 months of storage 82 Figure 15: Production of severe brownstain on the boards edges and faces of lumber sawn from log #4 86 Figure 16: Establishment of a standard calibration using gallic acid 98 Figure 17: Total soluble phenols measured in brownstained regions of test logs after 0 (A) and 2 (B) months of storage 99 Figure 18: Total soluble phenols measured in two non-stained regions of control logs after 0 (A) and 2 (B) months of storage 101 Figure 19: Total soluble phenols measured in four different regions in test and control logs after months of storage 102 Figure 20: Flowchart showing sample regime of 5 cm x 10 cm western hemlock lumber 107 Figure 21: Frequency of fungi isolated from western hemlock lumber 111 Figure 22: Total soluble phenols measured in three different regions within a board 114 Figure 23: Total soluble phenols measured in three different regions within a board 115 Figure 24: Flowchart showing experimental design 119 Figure 25: Colour changes in western hemlock sap (8A; 5A) incubated with different fungi for 10 days at room temperature 13 6 Figure 26: Tsuqa heterophylla. Radial section (63x), Brown deposits associated with hyphae of O. piceae in a 15 /xm section 151 Figure 27: Production of sap browning under oxygen 170 Xll List of Tables Table 1: Changes in pH of sap after three weeks of incubation 46 Table 2: Classification of selected western hemlock logs 75 Table 3: Production of brownstain on log ends after 9 month of outdoors storage 79 Table 4: Fungi isolated from western hemlock logs selected at Chamiss Bay 89 Table 5: Moisture content (MC %) and colony-forming units (CFU) in brownstained (BS) and non-stained (NS) logs 94 Table 6: Moisture content (MC %) and pH in brownstained (BS) and non-stained (NS) sample regions 96 Table 7: Total soluble phenol content (jug/mL) measured in sap from western hemlock incubated with 0. piceae for 6 weeks 127 Table 8: Fungi evaluated for their potential to cause browning in western hemlock sap 13 3 Table 9: Changes in western hemlock sap (8A) incubated with different microorganisms for 12 days at room temperature 137 Table 10: Changes in western hemlock sap (5A) incubated with different microorganisms for 12 days at room temperature 13 8 Table 11: Changes in western hemlock sap (4A) incubated with different microorganisms for 12 days at room temperature 14 0 Table 12: Changes in western hemlock sap (7D) incubated with different microorganisms for 12 days at room temperature 14 0 Table 13: Repeated assessment of changes in western hemlock sap (8A) incubated with different microorganisms for 12 days at room temperature 141 Table 14: Changes in pH 7 adjusted and non-adjusted western hemlock sap (8A) following incubation for 12 days at room temperature 168 xiii Table 15: Effect of oxygen on colour and TSP (/xg/mL) changes in western hemlock sap 171 Table 16: Changes in heated sap-8A after incubation with different microorganisms for 12 days at room temperature 173 Table 17: Changes in heated sap and in pH modified sap 175 Table 18: Colour and TSP in pH adjusted sap-5A amended with phenols after 12 days of incubation at room temperature 178 Table 19: Changes in buffered and non-buffered western hemlock sap (9A) incubated with different fungi for 12 days at room temperature 181 xiv Acknowledgements Foremost, I wish to thank my research supervisor Dr Roger Smith for his advice and guidance during the course of my graduate study. Apart from his invaluable experience in wood products research Roger passed on two things which I greatly appreciated: first he taught me to always thoroughly question my results before drawing any conclusions; second he made me understand that as a scientist you may have one or two highlights (advancements) in your whole career and therefore science means taking little steps at a time and often without making headway. Thanks Roger! A special thanks to Dr Colette Breuil for guiding me through the bureaucracies of my PhD program and also for her advice concerning my research. I also wish to extend my thanks to Drs Bruce Bohm, Simon Ellis and Bart van der Kamp for their interest in hemlock brownstain which resulted in many useful discussions. I also owe a lot of gratitude to friends and research colleagues in the Treated Wood Department at Forintek Canada Corp. I sincerely thank Tony Byrne for his advice and guidance throughout my study and for his great editorial skills. Tony reviewed numerous papers of mine and his last words after reviewing my thesis were "Deo gratias" . Tony, there will be more to come! I also thank Paul Morris for his scientific advice and also for many enjoyable hours while hiking or skiing. Bob Daniels assistance with the HPLC XV analysis is greatly appreciated. Futong Cui, Glenn Weigel, Maria Chan, Jean Clark and Rob Scott are also thanked for their advice and help during my studies. I also acknowledge the help of K.A. Seifert who identified some of the fungi isolated in this study. I must thank Rob Scheel and Bill Gilpin from Interfor Ltd. and the Interfor crew at Chamiss Bay. Without their support the field study on hemlock brownstain would have been impossible and I would not have been introduced to the adventurous world of lumberjacks. I also acknowledge Mr. Lome Holman of CIPA Lumber Co. Ltd. who provided lumber whenever needed. My studies at U.B.C. were supported by a VanDusen Fellowship and an University Graduate fellowship (MacMillan Bloedel) made available through the Department of Wood Science. Last but not least Forintek Canada Corp. and the Canadian Forest Service financially assisted my studies. xv i Preface Some of the material contained in the thesis has appeared previously in publications produced during the course of the research: 1. Kreber, B. and A. Byrne. 1994. Discolorations of hem-fir wood: a review of the mechanisms. Forest Products Journal 44 (5) :35-42. Permission was granted from Forest Products Society (Madison, WI) to reproduce the above article in whole or in part in this thesis. 2. B. Kreber. 1993/94. Advances in the understanding of hemlock brownstain. Material and Organismen 28 Bd. , Heft 1:17-37. Permission was granted from Material and Organismen (Berlin) to reproduce the above article in whole or in part in this thesis. 1.0 INTRODUCTION Variation in colour is one of the most distinctive, natural properties of wood. While the colour of wood is a very important characteristic in terms of its market value colour descriptions such as dark brown or "whitewood" are relative and subjective terms. The colour of woods is due mainly to extraneous compounds rather than structural cell wall components. Coloured extractives, found as deposits in the cell lumina or within the cell wall, give a characteristic colour to a species (Kuo and Arganbright, 1980; Kai and Swan, 1990; Kucera and Katuscak, 1992). Generally, wood colours can range from various shades of white to yellow, reddish to brown and gray to black. Some wood species display splendid and much-prized colorations, for instance mahogany (Swietenia macrophylla King) , black walnut (Juglans nigra L. ) , cherry (Prunus serotina Ehrh.), ebony (Diospyros ebenum Koenig), eastern redcedar (Juniperus virginiana L.) and they have been used for furniture and fancy goods since ancient times. The coloration varies not only among different wood species but also within a species and often in the same piece of wood. The latter colour variation can often be caused by sapwood and heartwood content, knots, or by grain orientation. For instance an interlocked grain, as is common in many tropical timbers, influences the colour intensity. Furthermore wood colours can change with time, for instance true mahogany has a pinkish cast when freshly cut and turns into a rich reddish brown with age and 2 exposure to light (Panshin and de Zeeuw, 1980). In fact most wood species change their colour with exposure to light, heat and other environmental factors, becoming either lighter or darker depending on the factors involved (Fengel and Wegener, 1984) . When changes in the coloration of woods cause an uneven or unwanted appearance, thereby decreasing their decorative market value, the problem is commonly defined as discoloration. Generally wood discolorations affecting light-coloured wood species are very detrimental because the natural light colour of these woods is readily disfigured. In this context it is unfortunate that current market trends favour the natural appearance of light-coloured woods for decorative purpose. In contrast dark-coloured woods including artificially darkened woods, which were highly regarded a few decades ago, are less in demand for decorative purpose. Western hemlock (Tsuga heterophylla (Raf) Sarg.) and amabilis fir (Abies amabilis (Dougl.) Forbes) are both light-coloured woods. They are highly regarded for their decorative appearance but they are also very susceptible to abnormal colorations. The discolorations affecting these wood species can vary from shades of grey and brown to black. In the current study brown discolorations were investigated in western hemlock with emphasis on the microbial involvement in the staining phenomenon. Understanding the causes of brownstains in western hemlock may enable some means for their control to be devised. 3 2.0 DISCOLORATIONS OF HEM-FIR WOOD: A REVIEW OF THE MECHANISMS 2.1 Background of Non-microbial Stains Wood discoloration problems have been known to lumber producers and customers worldwide for many years and have caused large economic losses to the wood industry (Hubert, 1926; Scheffer and Lindgren, 1940; Scheffer, 1973). Discolorations can be divided roughly into microbial ("biological") and non-microbial ("chemical") types. The most obvious form of wood discoloration is sapstain caused by fungi. Numerous investigations have been conducted on microbial stains since Robert Hartig's first description in 1878, resulting in an understanding of biological discolorations and their prevention (Munch, 1907; Lagerberg et al. 1927; Findlay, 1959; Liese and Schmid, 1961,1964; Schmid and Liese, 1965; Zink and Fengel, 1988, 1989, 1990). Non-microbial wood discolorations are not understood as well despite their common occurrence in both hardwoods (e.g., alder, maple, oak) and softwoods (e.g., pine, western hemlock). The lack of knowledge about chemical staining reflects the fact that these wood discolorations have long been considered to be less important because of their generally superficial nature. Thus colour disfigurations were commonly planed off, which normally removes the problem and restores the natural wood appearance. 4 However, advanced sawing technology such as thin-kerf sawing, which produces a dimension rather than oversized product, has renewed interest in non-microbial discolorations. Furthermore, overseas buyers are increasing their demands for kiln-dried lumber and prefer light-coloured wood. These trends have led to increasing concern about non-microbial wood discoloration. Unfortunately, non-microbial wood discolorations are generally not understood and literature on the subject is limited. This type of stain commonly develops as wood dries and involves the formation of coloured polymers, the chemical structures of which are not clearly understood. Chemical constituents of the particular wood species and the role of uncertain factors (e.g., temperature, humidity, microbial infestation) have to be understood to determine the cause(s) of non-microbial discolorations. The complexity of non-microbial staining is reflected in confusing terminology; it is also termed chemical or oxidative stain. A classification was recently proposed for microbial and non-microbial wood discoloration (Bauch, 1986). While most fungal staining can be easily recognized, a full understanding of the causes of other discolorations is required to use Bauch's classification. Furthermore, it is not a trivial task to be certain that microorganisms or their enzymes are not involved in a chemical stain. A thorough knowledge of the discoloration is essential to develop protective means to maintain the natural appearance of the species. 5 2.2 Economic Significance of Discolorations in Western Hemlock and Amabilis Fir Western hemlock and amabilis fir have only relatively recently been recognized as high value products of the British Columbia wood industry (COFI, 1983). Historically, these species have had minor end uses, for instance the production of tannin from hemlock bark for the booming leather industry in the 19th century (Hergert, 1989). The hemlock logs were often left in the woods (until 1900-1910) because there was a lack of technology to convert them into lumber. Competition from other wood species, for instance Chestnut (Castanea sp.) or Quebracho (Schinopsis sp.), led to a decline in hemlock tannin production. Today commodity lumber production in British Columbia (B.C.) is economically very important, worth approximately $6.5 billion annually, of which about $5 billion is exported (Goudie, 1992). Western hemlock and amabilis fir are coastal whitewoods and sold as "hem-fir". About 0.5 billion board feet of shop and better grade hem-fir, worth approximately $800 million, are marketed annually from British Columbia (Byrne and Smith, 1991). Hem-fir lumber is successfully marketed in Europe and Japan where it is widely used for windows, doors, mouldings and other millwork. Hem-fir lumber is highly regarded for its wood quality, for instance fine grain, 6 strength, and ease of working (COFI, 1983). Both species (western hemlock and amabilis fir) present in Canadian hem-fir display an overall similar, whitish or tan, appearance, with only a slight difference between heartwood and sapwood. Wood of the two species can only be separated with certainty by microscopic examinations. Unfortunately, the bright, whitish colour of hem-fir is particularly prone to darker (brown, black) disfigurations and downgrade from shop and better to lower grades can result in an approximate 30% loss in value (Byrne and Smith, 1991) . While sapstain fungi can be controlled in hem-fir lumber (Byrne and Smith, 1991) brown discolorations, for which there are no known controls, are cyclically a market issue. In 1990 a major problem in the market for Canadian hem-fir in Europe led to industrial requests for additional information (Byrne, 1992). 2.3 Investigations of Hem-Fir Discolorations Brown discolorations in clear-grade hemlock were first reported by Eades (1932). Concerns were highlighted during the World War II years as a result of the search for wood suitable for airplane manufacture (Eades, 1943; Englerth and Hansborough, 1945). Western hemlock was investigated for this application but discolorations found in the wood led to speculation that incipient decay or mechanical weakness was present. However, it was concluded that discolorations did not indicate wood degradation or sapstain fungi and no significant reduction in strength properties resulted 7 (Englerth and Hansborough, 1945) . More extensive scientific work on brownstain in hemlock was initiated in 1960 when mill operators from the Pacific coast of North America, from Oregon to B.C., reported a high incidence of brown discolorations (Evans and Halvorson, 1962). A light or dark brownstain was found to develop during both air and kiln drying but it affected the wood surface only, penetrating to a depth of no more than 0.5 mm. In addition discoloration was noticed most often in lumber from the sapwood-heartwood boundary region and was most apparent on the end or edge grain of affected lumber. Bacterial colonization in stained hemlock sapwood suggested a possible relationship between bacteria and brownstain development. Further bacteriological studies revealed that discoloured hemlock contained sap constituents which could be oxidized to coloured polymers. Evans and Halvorson (1962) speculated that the oxidation involved a bacterial phenoloxidase, which triggered polymerization of leucoanthocyanidins carried to the wood surface under favourable drying conditions. Chemical screening trials demonstrated that thiourea, an antioxidant, prevented surface browning and it was also assumed that brownstain was caused by atmospheric oxidation. About the same time intensive research was conducted at the Western Forest Products Laboratory in Vancouver to elucidate the chemistry and biology of the brownstain (Barton, 1962; Whittaker, 1962a). Bacterial infection was found in some apparently healthy hemlock 8 trees and in felled logs but bacteria were frequently encountered in brownstained logs. However, experiments to link the presence of bacteria to the production of the brown discoloration were inconclusive (Whittaker, 1962a). Barton (1962) did much work on the extractive chemistry of amabilis fir and western hemlock. Among other phenolic compounds the presence of water-soluble phenolics such as catechin, epicatechin and leucoanthocyanidin, which are known to polymerize under certain circumstances (Haslam, 1989), were identified in sapwood extracts. Barton, reproducing a brownstain in expressed hemlock juice, concluded from chromatographic examination that leucoanthocyanidin is probably not involved in brownstain formation. Attempts were made to chemically inhibit brownstain formation with pH modifiers, chelating or reducing agents, and it was reported that pH reducing components showed potential to moderate discoloration (Barton, 1962). Cross-sectional distributions and seasonal changes of catechin, epicatechin and leucoanthocyanidin were demonstrated in hemlock sapwood extractives (Barton, 1963; Barton and Gardner, 1966). Subsequent work employing synthetic catechin produced a brownstain on a papergram when reacted with juice expressed from hemlock sapwood but only a slight discoloration developed with synthetic leucoanthocyanidin. The authors suggested interaction of an enzyme system with catechin, both constituents of sound hemlock sapwood, and subsequent oxidation, as wood dries, produces a brown polymer at the wood surface. However, contribution of additional 9 compounds, for instance epicatechin or leucocanthocyanidin, to the final colour was not excluded. Inhibition of brown stain formation by chemical means failed in field trials although chemicals evaluated had shown promising results under laboratory conditions. Several serious cases of an intense, black-brown surface discoloration found in kiln dried amabilis fir initiated chemical and anatomical research (Barton and Smith, 1971). Microscopic work on discoloured specimens demonstrated the presence of high concentrations of dark-brown extractives in parenchyma cells and bacterial infection in longitudinal tracheids. While cell wall degradation of parenchyma cells and longitudinal tracheids was not significant, pit breakdown was observed. Sections of discoloured wood further revealed a change in fluorescence, when compared with sections of wood which was not discoloured, suggesting lignin degradation. Chemical analysis indicated that 3,3'-dimethoxy-4,4'-dihydroxystilbene (DDS) was responsible for the deep brown colour of kiln-stain in amabilis fir. It was further hypothesized that bacterial degradation of lignin might have produced DDS. Chemical treatments, with 8 different compounds, did not prevent brown discolorations on stored hem-fir (Swan, 1984a). However, it was verified that brownstain developed in amabilis fir as well as in western hemlock. The author concluded that basic research on the biological and chemical aspects of brownstain was needed before a solution to this problem could be found. 10 A recently reported discoloration on exported hemlock lumber appeared as grey streaks along the wood surface (Smith and Spence, 1987) . Although macroscopically the discoloration did not resemble the traditionalbrown stain in hem-fir lumber, microscopic examination showed brown globular deposits within the ray parenchyma cells in the grey-streaked wood. Additionally the authors saw fungal hyphae within the rays. Isolation of fungi from discoloured wood, which interestingly showed non-pigmented hyphae, yielded Ophiostoma piceae (Munch) H. and P. Syd. and Sporothrix sp. , a possible asexual stage of Ophiostoma sp. Subsequent laboratory inoculations of wood with these fungi resulted in a brown discoloration. The authors suggested that oxidative activity of fungi growing within the rays caused polymerization of catechin, abundantly present in hemlock sapwood, and thus produced a grey or brown disfiguration in the lumber. Other hem-fir samples sent to Forintek from European customers also displayed a grey or brown disfiguration as previously described by Smith and Spence (1987) and brown, globular deposits were microscopically demonstrated in ray parenchyma and to a lesser extent in tracheids (Byrne and Smith, 1991). Unfortunately, studies on the presence of fungi were not conducted but preliminary bacterial investigations showed the presence of bacteria in the pits of some samples (Byrne, 1992). These grey discolorations of hem-fir microscopically resembled grey stain found in oak (Clark, 1957; Forsyth and Amburgey, 1991), but in trials sodium bisulfite, 11 which can control grey oak stain (Forsyth, 1988), did not prevent colour formations in hem-fir (Byrne and Smith, 1991). Further samples returned from a French customer showed a different discoloration presumably developing when hem-fir is kiln dried (Byrne and Smith, 1991) . Preliminary studies indicated a light red/brown coloration, occurring just under the wood surface. This discoloration was concentrated in earlywood giving a striped wood appearance which led to the term "zebra stain". During investigation of very severe cases of almost black discoloured samples, iron and/or manganese were detected using X-ray spectroscopy (Byrne and Smith, 1991). The intensity of this black colour increased with the amount of these elements present. Dilute phosphoric acid removed the black component and the metal ions, leaving behind the brown component of the discoloration. The origin or source of these metals in stained samples remained unknown. Microbial investigations were not conducted, although both iron and manganese play a role in the metabolism of some microorganisms. Biogeochemical studies have also indicated that western hemlock trees can accumulate high contents of copper and zinc (Warren and Howatson, 1947) and iron and manganese (Warren et al, 1952). Furthermore these elements varied between different parts of the same tree and appeared related to the age of the tree as well as to soil type and climate. High accumulation of manganese was also recently demonstrated in the xylem of western hemlock needles (Ballard, 1992). 12 Numerous agents have been evaluated in other attempts to find suitable chemical treatments to prevent brownstain in hem-fir (Byrne and Smith, 1991). The chemicals tested included pH reducers, chelating/sequestering agents, and reducing agents, but only a quaternary ammonium compound (didecyldimethylammonium chloride - DDAC) controlled brownstain in small specimens while drying under ambient laboratory (2 0° C) conditions. The mode of action of this chemical was not understood. However, field tests using DDAC on commodity hem-fir lumber gave disappointing results and it was hypothesized that the quantity of extractives moving (longitudinally), as wood dries, through the end grain to the surface, might have exceeded the potential of the DDAC to inhibit brown stain formation. Recently, a sample of amabilis fir with brown stain resulting from "kiln-burn" was investigated for causal agents (Sutcliffe and Miller, 1991) . A conversion of DDS to a highly coloured compound was demonstrated but it was hypothesized that other polyphenols could contribute to the discolorations. In addition high amounts of calcium indicated by energy dispersive X-ray analysis (EDXA) were observed on the surface of the stained specimen but manganese or iron were not present. The effects of some kiln drying variables were investigated on the 13 development of brownstain in hem-fir lumber (Avramidis et al. , 1993) . The authors concluded that low drying temperatures and a more gradual drying reduced the incidence of brownstain during drying. Furthermore, Avramidis et al. (1993) reported that the presence of less oxygen resulted in less stain development. 2.4 Extractive Chemistry of Western Hemlock Undoubtedly a thorough knowledge of the composition of wood extractives is required to fully understand the cause of brown discoloration in hem-fir products. Many extractive compounds of western hemlock were isolated and identified in the 1960s. The significance of these extractives as potential colour precursors is discussed in this section, providing examples from pulp and paper research. In the U.S. Pacific Northwest large quantities of sawmill residues used in producing pulp are subject to discoloration during storage (Springer, 1983) . Thus large losses in wood brightness occur and in some instances pulp cannot be bleached to satisfactory levels (Springer, 1983). Considerable research has therefore attempted to elucidate compounds responsible for loss in wood brightness of chips or pulp during storage, as well as searching for ways of prevention. 14 In western hemlock, which is widely used for production of pulp, a variety of phenolic compounds give rise to undesirable chromophores (Barton, 1973a). Extracts from western hemlock developed a red-brown colour upon treatment with mineral acids in the presence of alcohol (Pigman et al. , 1953) . Flavan-3, 4-diols, which belong to the proanthocyanidins, appeared to be colour precursors. These compounds could be either water-soluble or water-insoluble in sapwood but heartwood contained largely the insoluble type. The richest source of water-soluble flavan-3, 4-diols, however, was the inner bark (including cambium). Purified phlobaphenes (red-coloured, water-insoluble reaction products of tannin extract treated with mineral acid) were also suggested to be rich in flavan-3, 4-diols. Research on the chemical composition of tannins and polyphenols from conifer woods and bark has demonstrated that monomeric catechins and leucoanthocyanidins were commonly present in sapwood and cambium of all species studied (including western hemlock) but absent in heartwood (Hergert, 1960). In place of these compounds polymeric proanthocyanidins (flavan-3,4-diol type) were present in the heartwood. The author further noted that species with intensely coloured heartwood (e.g., Douglas-fir) also exhibited high amounts of flavan-3,4-diols in the sapwood but species with light coloured heartwood displayed low flavan-3,4-diol contents. Polymeric tannins in hemlock, however, were suggested to be built from compounds such as catechin, epicatechin, gallocatechin, 15 epigallocatechin and leucoanthocyanidin (Hergert, 1960). Cambium and sapwood of western hemlock were examined for the presence of low molecular-weight constituents as potential intermediates in lignin formation (Goldschmid and Hergert, 1961). Cambium and sapwood contained alicyclic acids, lignans and glycosides of lignan-like compounds, sugars and catechin, epicatechin and leucoanthocyanidin but depsides of cinnamic acid derivatives were present in cambium only. Phenolic extracts of western hemlock, possibly involved in brown discoloration of hemlock lumber and low brightness of hemlock pulp, were reviewed by Barton (1968) . Flavonoid-type phenolic extractives, such as catechin and leucoanthocyanidin, known precursors of highly coloured tannins and polymeric phenol, were strongly suggested as producing brown stain. Lignans such as coniferin, hydroxymatairesinol or matairesinol were thought not to contribute to colour formation. Lignans, colourless or pale lemon-coloured substances, lack vicinal hydroxyl groups and their stable ring system was believed to preclude coloured oxidation by-products. Guaiacylglycerol, a simple phenolic compound, was suggested to form colour precursors under acid or alkaline conditions but the relatively small amount present precluded its significant participation. A glycoside of a recently detected lignin dimer (phenyl coumaran), in which the sugar moiety is attached through an alcoholic linkage, was also suggested as 16 contributing to coloured products in pulp and lumber. Another lignan, liovil, was identified after its isolation from western hemlock sapwood (Barton, 1970). In addition, the structure of a recently detected new lignin dimer in western hemlock was determined. The significance of these compounds in colour disfiguration was described (Barton, 1968). The chemical composition of the extractives and their effect on optical properties of western hemlock pulp was investigated by performing three treatments: a sequestering (ethylenediaminetetracetic acid - EDTA) treatment, ethanol-benzene extraction and acetone-water extraction (Polcin et al., 1969). Brightness of pulp was improved by EDTA treatment and removal of copper and manganese cations, previously reported to be present in relatively high amounts in western hemlock groundwood (Wayman et al. , 1968), was suggested as one possible explanation for this observation. Ethanol-benzene extraction did not improve the colour of pulp. However, extraction with acetone-water removed polyphenols and low molecular weight phenolic compounds and improved pulp brightness. A subsequent study examined the heat stability (105°C for 18 hours) of three (EDTA, ethanol-benzene, acetone-water) extractions from western hemlock and also model compounds absorbed onto sheets of pure bleached cotton (Polcin and Rapson, 1971). Acetone-water 17 extracts of western hemlock sapwood developed a yellow-brown colour, presumably by air oxidation, soon after evaporation of the solvent. Investigating the contribution of model compounds to the discoloration indicated the importance of the flavan-3ols such as catechin. Heat treatment experiments with d-catechin produced a brown colour but addition of unsaturated fat to d-catechin substantially intensified the discoloration. The discoloration was even more intense when d-catechin was heated in the presence of water. The authors concluded that oxidation of d-catechin occurred at elevated temperatures, forming simple and complex quinoid structures. It was further suggested that peroxi-radicals were involved in the oxidation of d-catechin in the presence of unsaturated fats and water producing an increase in discoloration. While formation of free radicals is known to increase in the presence of some metallic (e.g., copper, manganese) cations it was concluded that removal of these cations by EDTA treatment prevented heat discolorations of groundwood pulp. Changes in western hemlock wood extractives during refining and reduction of chromophore production during and after refining by chemical applications have been studied (Barton, 1973b). During refining, high temperatures (exceeding 8 0° C) were recommended to modify flavonoid and lignan materials as well as lignin precursors and lignin. Several chemical additives such as acids, reducing agents and oxidising agents improved brightness of western hemlock mechanical pulp. 18 Chemical agents for maintaining the brightness of stored western hemlock wood chips have been evaluated in a laboratory trial (Springer, 1983). Dilute (2%) aqueous solutions of sodium bisulfite enhanced and maintained the brightness of chips during 12 weeks storage. Bisulfite ions, effective biocides and enzyme inhibitors, were demonstrated to oxidize in solution to bisulphate ions causing a large decrease in pH; liquors squeezed from western hemlock sapwood dropped from 5.4 to 2.6. Loss of brightness in mechanical pulp as a function of storage time of western hemlock chips was correlated with a decline in concentration of d-catechin monomers (Hrutfiord et al., 1985). It was hypothesized that oxidation of d-catechin produced a brown polymer, which was retained in the wood fibres of freshly chipped western hemlock. 2.5 Factors Which May Contribute to Discolorations in Hem-Fir Although this section has summarized the most significant research on discolorations in hem-fir it is useful to speculate on factors which may play a role in these colour formations. Examples will be given from research on stains found in other wood species. 19 2.5.1. Factors Inherent in Living Trees The occurrence of discolorations in living trees has been recognized for many years (Sachs et al. , 1966; Bauch and Baas, 1984; Shortle, 1984). It is possible that conditions leading to discoloration of felled logs and wood products may have already been initiated in the living tree. The following examples illustrate links between the living tree and discolorations in wood products where predisposition of the wood substrate to discoloration may have originated in the tree. Discolorations in living trees have been shown to arise from wounding. For instance, broken branches, pruning, severe logging damage (Shigo and Hillis, 1973; Phelps and McGinnes, 1984), and/or dying branches (Aufsees, 1984) have been shown to cause wood discolorations. Minor wounds may produce slight colour changes due to the formation of chemical protection barriers. However, severe wounding can lead to interactions between microorganisms (sapstainers, decay fungi) and living cells, producing intense colour disfigurations in the living tree (Shigo and Hillis, 1973). Discolorations have also been found in living trees attacked by insects, for instance bark maggots (Cheilosia alaskensis Hunter) causing black streaks in western hemlock (Englerth and Hansborough, 1945; Moeck, 1968) or by the symbiotic fungi of ambrosia beetles which cause localized sapstain (Funk, 1965) . In a recent study 20 discolorations induced by larvae of Sermanotus japonicus Lacordaire were demonstrated in living trees of Cryptomeria japonica D. Don. (Yamada et al., 1987). In this study cation (K, Mg) accumulation appeared to explain an increase in pH over a five year time period. Although microorganisms (bacteria and non-hymenomycetes) were found in the discoloured areas, cation concentration, but not composition of the microflora, has been proposed to explain the advance in discoloration. Thus a shift in pH may have triggered polyphenolic reactions of accessory compounds. Interestingly, accumulation of manganese (Warren et al., 1952; Ballard, 1992) or copper (Warren and Howatson, 1947; Wayman et al. , 1968) has been detected in western hemlock trees. The role of bacteria in standing trees as a factor in discoloration also requires consideration. Bacteria have been shown in apparently healthy western hemlock trees (Whittaker, 1962a) and have also been associated with wetwood present in softwoods, for instances in western hemlock (Bauch et al. , 1975; Ward and Zeikus, 1980) and white fir (Abies concolor (Gord. & Glendl.) Lendl.) (Wilcox and Oldham, 1972). Bacteria have also been associated with wetwood in hardwoods such as Populus sp. (Knutson, 1973; Sachs et al.,1974; van der Kamp et al. , 1979; Scott, 1984; van der Kamp, 1992) and Fagus sylvatica L. (Walter, 1993) . In addition to participation in wetwood formation bacteria have been linked to the pH of the wood and to polymerization of phenolic compounds, producing discolorations (Schink and Ward, 1984; Schmidt, 1986; 21 Schmidt and Mehringer, 1989). Bacterial infestation of western hemlock trees has been specifically suggested as a causal factor in discoloration of lumber produced from such trees (Whittaker, 1962a). 2.5.2 Felling Season In temperate zones trees undergo seasonal changes which are most obvious in deciduous wood species because of defoliation in fall. Metabolism of trees is reduced during the winter season and this is reflected in a decrease in water uptake and sap flow. These seasonal changes also influence the composition of certain compounds in living trees. Extractives possibly involved in discolorations of hem-fir lumber were also shown to undergo seasonal changes in living western hemlock trees (Barton and Gardner, 1966) . For instance maximum catechin levels were detected in April, May and June, declining to lower but still significant levels in July - October. In kiri wood (Paulownia tomentosa Steud.) peroxidase activity triggering discoloration was demonstrated to occur in September and October at levels 12 fold those of June or November (Ota et al. , 1991) . The authors suggest that harvesting Kiri trees should be avoided during this time. 22 2.5.3 Post-Mortem Factors Post-mortem changes causing discolourations in the sap of trees upon exposure to air have been reported in hardwoods and softwoods (Bailey, 1911). For instance freshly cut Alnus sp. or Liquidambar sp. can develop discolorations within hours after sawing under favourable conditions of temperature and humidity. Bailey (1911) demonstrated that heating of the woods in boiling water controlled discolorations thus indicating a plant enzyme as the cause of the problem. An orange coloured polymer, oregonin has been identified developing in red alder (Alnus rubra Bong.) after sawing (Karchesy, 1992). Hrutfiord and Luthi (1981) demonstrated that phenoloxidases reacting with accessory organic compounds after oxygen penetrated the wood tissue produced the observed colour formations in red alder. A similar reaction causing discolorations in freshly sawn Kiri wood has been shown (Ota et al. , 1991). Disruption, or spatial separation of enzymes (peroxidases) and wood extractives (Ota and Taneda, 1989), caused by the sawing action, was suggested as producing the colour reaction. Peroxidase activity on phenolic extractives has been demonstrated to cause brownstain in sugar pine (Pinus lambertiana Dougl.), in 23 eastern white pine (P. strobus L.) and in western white pine (P. monticola Dougl.) (Stutz, 1959; Stutz et al. , 1961). Unlike brownstain in white pines which was shown to also develop under air seasoning conditions, kiln drying was required to produce discolorations in sugar pine. The control of brownstain in white pines and sugar pine has been shown in laboratory and field experiments (Stutz, 1959; Stutz et al. , 1961; Arganbright, 1972; Shields et al., 1973; Oldham and Wilcox, 1981). 2.5.4 Log Age The time elapsed from stump to saw has been suspected as one important factor affecting the formation of hemlock brownstain (Evans and Halvorson, 1962) . The significance of log age was also underscored in a survey on brown discolorations in sugar pine developing during kiln-drying (Herman, 1937) . Subsequent research on sugar pine indicated that nine month old sugar pine logs discoloured three times as much as fresh ones (Rasmussen, 1940). It was speculated that higher oxygen tensions or higher levels of insect damage could cause the "log age effect" observed in pine species (Stutz, 1959). Brown discolourations in maple (Acer pseudoplatanus L.) logs which intensified with time have been reported (Koltzenburg, 1974). These discolorations can appear without microbial interaction at a given ratio of oxygen, temperature and wood moisture. However, 24 bacteria were demonstrated to produce a similar brown discoloration in the same wood species (Zimmermann, 1974). In a detailed study (Starck et al. , 1984; Bauch et al. , 1985; Yazaki et al. , 1985) a reddish-brown discoloration found on the end grain of freshly cut logs or on the surface of freshly cut lumber of Ilomba (Pycnanthus angolensis Excell) was traced to the presence of bacteria. The authors suggested that some bacterial strains, for instance Pseudomonas fragi Hussong et al. , can alter the pH of the wood from about 5.5 to 7.5 thereby triggering chemical reactions of accessory compounds. Stabilization of the pH of the wood surface with formic acid was reported to inhibit discolorations (Bauch, 1986). In addition, (+)-catechin and (-)-epicatechin were suggested as possible contributors to colour formation in Ilomba (Yazaki et al.,1985). It should be emphasized that changes within the logs, for instance in moisture content and wood temperature, and chemical changes of accessory compounds, are more likely to be responsible for colour formation than log age per se. The role of microbial populations interacting with certain wood extractives must also be considered in producing discolorations. In this context, however, it is interesting to mention that an aging process has traditionally been used on Kiri logs to prevent a discoloration, which develops when lumber is immediately sawn from fresh-cut logs (Ota et al., 1991) . Thus Kiri logs are exposed to the weathering action for 6-9 months 25 followed by an outdoor exposure of the sawn boards for another two years to enhance the appearance. 2.5.5 Log Storage Delay between felling of logs and processing into kiln-dried lumber probably increases the risk of potential discolorations (including sapstaining) in most wood species. Unfortunately, it is not possible to totally avoid delays and thus western hemlock and amabilis fir logs are commonly stored from several months to 1.5 years before sawing (Kreber and Byrne, 1993). In most cases short term storage of logs is in dry decks but for longer periods water storage is employed. Microbial infestation can be controlled for long time periods by keeping wood in water saturated condition (Liese and Peek, 1984). However, logs stored in ponds or sprinkled are known to develop high bacterial populations (Smith, 1975). Bacteria may cause changes in wood permeability (Unligil, 1972; Johnson, 1979) due to decomposition of pit structures (Liese and Karnop, 1968; Greaves, 1969). Several species of Bacillus (Ellwood and Ecklund, 1959) and Pseudomonas (Grosu et al., 1973) have been reported in ponded logs. Discolorations of lumber sawn from sprinkled hardwood and softwood logs have been attributed to the presence of bacteria (Stout, 1959; Stutz, 1961; Lane and Scheffer, 1969; Hedley and Meder, 1992) . 26 Water storage/ponding of logs was also reported to influence discolorations in beech (Hoster, 1974) and to promote and accelerate the development of gray discolorations in red oak (Forsyth and Amburgey, 1992). The effectiveness of sodium bisulfite in preventing discoloration of lumber sawn from freshly-cut oak, decreased dramatically with increasing storage time of the sawlogs (Forsyth and Amburgey, 1992) . Water storage or floating of logs has been suggested as a means to redistribute accessory compounds (flavonoids) from the phloem of unpeeled spruce logs into the xylem (Adler, 1951). In B.C. hem-fir logs are often floated to mill yards. Interestingly it has been claimed, that western hemlock trees felled and immediately salt water floated produced less discolorations than fresh water floated logs (Barton, 1962). 2.5.6 Storage of Lumber Storage of unseasoned lumber in close piles, the common practice for exported lumber, decreases air drying rates significantly. This may influence colour formation in hem-fir. Storage of lumber in bulk piles was demonstrated to influence brown discolorations in sugar pine lumber (Stout, 1950) . The author showed that pine lumber stacked in bulk piles between sawing and stickering for more than 5 days developed severe discolorations. 27 These observations were also confirmed for close-piled, white pine lumber (Cech, 1966) . In a recent study (Hansen, 1988) green and red discolorations developing on the surface of freshly sawn Samba (Triplochiton scleroxylon K. Schum.) boards were investigated. The colour formation was most noticeable in the centre of stacks. The presence of a bacterium Pseudomonas aerucrinosa Migula was suggested to have produced the discolorations. The author proposed faster drying of the wood and treatment with a bactericide to prevent discolorations in this white wood species. A bright yellow discoloration has been reported in heartwood of green oak lumber when stored with insufficient ventilation, particularly when thin stickers did not permit drying (Bauch et al. , 1991) . The mould Paecilomyces variotii Bain, was suggested as causing the colour formation from hydrolyzable tannins. The same (yellow) discoloration was also found in kiln dried oak. P. variottii, however, was shown to tolerate acidic substrates as well as high (50° C) temperatures, conditions which are prevalent during the first stage of kiln drying. The authors concluded that growth of P. variotii was probably not inhibited during the initial stage of kiln drying thus explaining yellow discolorations found during drying. 2.6 Summary 28 Discolorations of hem-fir, other than those caused by sapstain fungi, have become an economically important problem. With the move towards increased kiln-drying of the wood species mixture and to more added-value products, such discolorations are less tolerable. Although discoloration of hem-fir lumber has been a puzzle for many years, knowledge of the cause(s) remains rudimentary. Most research into hem-fir discolorations has been conducted on only a few wood samples. While polymerization of wood extractives has been proposed as the probable cause, involvement by bacteria and fungi has also been suggested in the literature. Factors involved in discolorations of other wood species, such as factors inherent in the living tree, season of tree felling, post mortem changes, and log age and storage, may also be involved. A thorough understanding of the cause(s) of hem-fir discolorations is necessary before recommendations or preventive treatments can be devised to maintain the natural colour in hem-fir products. 29 3.0 ADVANCES IN THE UNDERSTANDING OF HEMLOCK BROWNSTAIN 3.1 Objectives The objectives of this study were to: a) describe the location of the stain in the wood tissue; b) elucidate the composition of brownstain deposits histochemically; c) evaluate the potential of three Ophiostoma piceae (Munch) H. & P. Syd. strains and a mixed bacterial culture to produce brownstain in sap of western hemlock, amabilis fir and lodgepole pine and in western hemlock wood. 3.2 Material and Methods 3.2.1 Microscopic Examinations Representative samples showing various types and intensities of colorations, were selected from Forintek's collection of brownstained specimens. The specimens came from edge-grain stock and had been sent from Europe to illustrate the market problem. Transverse and radial sections (15-20 /xm) were prepared from twenty five discoloured samples of western hemlock and amabilis fir using a sliding microtome. Sections were dehydrated through a series of ethanol solutions, and finally passed through xylene and mounted using Permount resin (Fisher Scientific, Nepean, Ontario). Light microscopy and phase contrast microscopy were employed to examine specimens using a Zeiss photomicroscope. 30 3.2.2 Histochemical Examinations Radial sections were also sampled from freshly sawn hem-fir boards and from small (66 x 18 x 6 mm) sapwood beams of western hemlock as outlined (see section 3.2.1). Additional hem-fir sections showing microscopical brown deposits, were extracted either in 10 mL of methanol or in 10 mL of acetone/water (7:3) using an ultrasonic bath for two hours at room temperature. Every section (non extracted or extracted) was placed in 1% 2,4 dimethoxybenzaldehyde (DMB) reagent (Aldrich Chemical Co., Milwaukee, WI) prepared as described by Mace and Howell (1974) mounted on a glass slide in a drop of DMB and then stored at room temperature to yield a red coloured product when reacting with catechin and its derived tannins (Halloin, 1982) . Control sections were prepared according to the procedure outlined but DMB was omitted from the reagent. Wood species and presence of brownstain deposits was microscopically verified on additional sections mounted in lactophenol. Microscopic examination of treated samples was performed using light microscopy. 3.2.3 Sap Experiments Sampling and sterilization of sap: Two freshly sawn boards of western hemlock (hereafter referred to as Hem-1, Hem-2), one of amabilis fir and one of lodgepole pine (Pinus contorta (Dougl. ex. Loud.)) were used as sources of sap. Visual examination of the 31 boards indicated no signs of brownstain or sapstain but ambrosia beetle infestation was noticed in some regions of Hem-2. Microscopic examinations verified the presence of brown deposits in the amabilis fir and western hemlock. Sap was pressed from each board using a hydraulic press and was filtered twice through Whatman # 1 filter paper followed by filter-sterilization using a 0.2 pirn cellulose acetate membrane (Nalge Company, Rochester, NY). The sterilized sap was stored in a refrigerator (4°C) overnight. Microorganisms used: Ophiostoma piceae strains WFPL # 3871, WFPL # 387K, WFPL # 387T were cultured on 1.5% malt agar plates for approximately 5 weeks at 25°C. A suspension containing spores and mycelial fragments was prepared from each culture by gently scraping the culture surface. Mycelial fragments and spores were then washed from the Petri plates using sterile, distilled water. The crude fungal suspensions were individually blended in a Waring blender for 15 seconds and then drained into sterile 16 oz glass jars. Sterile, distilled water was added to give approximately 300 mL of each fungal suspension. A mixed bacterial culture (Mix-B) isolated from the sap of a board of western hemlock which showed brownstain, was cultured in a 250 mL Erlenmeyer flask containing 50 mL of nutrient broth (Difco Laboratories, Detroit, MI) for approximately 24 hours. A 32 suspension was prepared by adding about 5 mL of the Mix-B nutrient broth culture to approximately 300 mL of distilled, sterile water. Inoculation of filter-sterilized sap: Approximately 20 mL of filter-sterilized sap was added to a series of autoclave-sterilized (121°C, 30 minutes) 125 mL Erlenmeyer flasks and inoculated with 0.5 mL of either bacterial or fungal suspension using a sterile pipette tip. One replicate was set up for each treatment and for each control containing no inoculum. Flasks were stored in a laminar flow hood at room temperature (22°C) for three weeks. Evaluation of changes in inoculated sap: Colour changes in inoculated sap was visually recorded over time. In addition, the pH of filtered saps was measured prior to inoculation and again after a three week incubation period. Readings were taken while stirring using a pH-meter. Qualitative HPLC analysis, as described by Kreber and Daniels (1993), was conducted on inoculated Hem-2 sap after incubation and was compared to HPLC analysis of the uninoculated control and freshly pressed Hem-2 sap. 3.2.4 Solid Wood Experiment Inoculation of solid wood: Small (66 x 18 x 6 mm) sapwood beams of western hemlock, showing no brownstain (verified microscopically) , were used in this study. The beams had been cut from a freshly-felled tree within 24 hours of felling and they had been kept in a 33 freezer for 3 years. Four beams were employed for each treatment, placed on a polypropylene mesh in an aluminum tray containing 3 sheets of cotton. To keep the chamber moist, distilled water (90 mL) was added to each tray and trays were then autoclaved (30 min, 121°C) prior to inoculation. Each wood sample was dipped in inoculum (fungal or bacterial suspension prepared as described above) for three seconds prior to being placed in the sterile tray. Each tray was sealed in a polyethylene bag to maintain a moist environment and stored at room temperature for six weeks. In addition, four control beams were dipped in sterile, distilled water only. Assessment of brownstain in inoculated wood beams: Two wood beams selected at random were removed from each treatment after 4 and 6 weeks. Radial sections (15 /xm) were prepared as described above, mounted in lactophenol and the extent of brownstaining was determined with a photomicroscope. Sections from treated wood specimens were compared to untreated controls kept at room temperature or in a freezer. 3.3 Results 34 3.3.1 Microscopic observations on discoloured hem-fir Generally, microscopic examination indicated similar distribution of brown deposits for all discoloured samples investigated. This observation was rather surprising since specimens were sampled from boards showing different macroscopic patterns of discoloration (Figure la-c). Brown deposits were most noticeable in rays but they varied in colour intensity (pale yellow to dark brown), in shape (e.g., spherical, pillow-like) and size (Figure 2). Furthermore, an apparently random distribution of brown deposits was observed across and between growth rings. In severely stained samples the lumina of ray parenchyma cells were seen to be filled with chestnut brown deposits. Ray tracheids in western hemlock were never found to contain such deposits. Although brown chromophores entered the half-bordered pit from parenchyma cells there was no penetration through the margo into the ray tracheids. Axial tracheids with deposit-filled lumina were rare in sections examined. Tracheids with a filled lumen were often isolated among stain-free tracheids (Figure 3) . In most cases the lumina of discoloured tracheids were lined with a thin deposit ranging from yellow to brown in colour. 35 JiJ Figure la: Representative specimen showing grey stain on the wood surface. 36 Figure lb: Representative specimen showing brownstain on the board end and on the wood surface. 37 I h r \ n Figure lc: Representative specimen showing zebra stain on the wood surface. 38 1* i . i l l Figure 2: Abies amabilis. Radial section (63x), ray parenchyma cells with brown deposits. 9& .j. 2 ct f # Figure 3: Tsuaa heterophylla. Transverse section (lOOx), tracheids with a filled lumen among stain-free tracheids. 40 Fungal hyphae were frequently seen in ray parenchyma and tracheid cells and they were most often hyaline. Some hyphae within the rays and within the tracheids were seen enclosed in coloured deposits or surrounded by a brown sheath (Figure 4). Hyphae were also detected in some stain-free cells. Bacteria were frequently observed in bordered pits in otherwise unstained tracheids but bacteria were also seen enclosed in coloured deposits in the lumena of tracheids (Figure 5). 3.3.2 Histochemical observations Sections prepared from sapwood beams of clean, fresh western hemlock showed no signs of brown deposits in either parenchyma cells or axial tracheids, as verified microscopically. However, sections mounted in DMB developed a red reaction product in parenchyma cells and axial tracheids within six hours suggesting presence of catechin/epicatechin (Figure 6, 7). The intensity of DMB reaction varied between and among both parenchyma and tracheid cells but the colour change was not observed when specimens were mounted in the absence of DMB. Every section prepared from each of the hem-fir boards showed brown deposits. The presence and spatial distribution of brown deposits within sections confirmed microscopic observations as described above. DMB also demonstrated a red reaction product in brown 41 Figure 4: Tsuqa heterophylla. Radial section (200x), hyphae enclosed in coloured deposits of a tracheid. 42 f» Figure 5: Tsucra heterophylla. Radial section (lOOx) , bacterial infection of tracheids. 43 Figure 6: Tsuqa heterophylla. Radial section (80x), red DMB reaction products observed in ray parenchyma cells. 44 J " I L * m J w 1 Figure 7: Tsuaa heterophylla. Radial section (80x), red DMB reaction products in axial tracheids. 45 deposits within 6 hours in the sections including additional sections which had been extracted in methanol or acetone/water prior to treating with DMB. Sections mounted in the absence of DMB produced no colour change. 3.3.3 Colour changes in inoculated sap Microbial inoculation gradually produced a distinct colour change only in sap of Hem-2 when compared to the control. After three weeks a brown colour was recorded in all samples of inoculated Hem-2 sap irrespective of the inoculum used. Unlike uninoculated, uncontaminated sap, the pH changed in inoculated sap of Hem-2 from pH 5 to above pH 8 over the three weeks; interestingly a similar pH shift was recorded in each of the treated Hem-2 sap samples independent of the inoculum (Table 1). Qualitative HPLC analysis (data not shown) of inoculated compared to uninoculated sap of Hem-2, indicated that higher molecular weight compounds were no longer present, while the proportion of lower molecular weight compounds considerably increased. Also many fewer compounds were detected in inoculated sap of Hem-2 after incubation using HPLC analysis. These observations were recorded for all treatments and independent of inoculum type. Although the control sample (uninoculated) of Hem-2 was not discoloured, a slight turbulence (possible contamination) was noted. This may 46 TABLE 1: CHANGES IN pH OF SAP AFTER THREE WEEKS OF INCUBATION. Treatment Control/frozen Control/22°C 0.p_.3 3871 0.£. 387K 0.p_. 387T Mix-B4 WOOD SPECIES Amabilis fir 5.7 8.41 8.7 8.6 8.7 8.7 Western hemlock (1) 5.9 8.62 8.4 8.1 8.2 7.5 Western hemlock (2) 5.2 4.72 8.2 8.5 8.4 8.5 Lodgepole pine 7.0 7.0 8.7 8.3 8.4 8.7 1 = sap indicated heavy contamination 2 = sap indicated slight contamination 3 = Ophiostoma piceae (Munch) H. & P. Syd. 4 = mixed bacterial culture isolated from a western hemlock board showing brownstain 47 have caused the minor changes observed in the phenolic composition. Treated sap of amabilis fir, which already showed a light brown colour at the time it was pressed, demonstrated some darkening over time but controls (uninoculated) became contaminated and developed brownstain. Therefore the effect of inoculation on the production of colour changes in sap of amabilis fir was inconclusive. However, a brown precipitate was observed in amabilis fir sap independent of treatment. Measurements of pH indicated shifts in pH similar to those recorded in sap samples of Hem-2 (Table 1). Neither the sap of Hem-1 nor of lodgepole pine incubated with microorganisms changed colour. Interestingly an increase in the pH of Hem-1 and lodgepole pine sap was also recorded after incubation and again this pH shift was independent of treatment (Table 1). 3.3.4 Production of brownstain in inoculated western hemlock Microscopic examination of frozen, clean western hemlock sapwood beams showed no brown deposits. However, ray parenchyma cells contained oval and rounded, colourless globules (Figure 8) instead of the brown deposits previously described. These colourless structures were accumulated in particular ray parenchyma cells but they were not necessarily present in adjacent cells. Ray tracheids or axial tracheids did not contain colourless globules. 48 Figure 8: Tsuga heterophylla. Radial section (320x), colourless globules observed in clean ray parenchyma cells. 49 Examination of control (autoclaved, non inoculated) specimens demonstrated a few small, yellow to light brown deposits in ray parenchyma cells but these had increased in number after 4 and 6 weeks. Light-coloured deposits were recorded in a few ray parenchyma cells following autoclaving (121°C, 30 minutes). Thus heating alone produced some coloured globules, however, atmospheric oxidation was shown to form additional internal discolorations. Specimens inoculated with # 3871 strain of 0. piceae showed hyphae in the tracheids and in ray parenchyma cells after 4 and 6 weeks of incubation. Brown deposits were seen in ray parenchyma cells containing hyphae and also in cells apparently free of hyphae but they were not seen in ray tracheids or in axial tracheids (Figure 9) . Several axial parenchyma cells also demonstrated brown discolorations. A considerable increase in brown deposits and numerous small brownish particles was seen in ray parenchyma cells when compared to controls after 4 and 6 weeks. In some cases light brown deposits completely filled the lumen of ray parenchyma cells. Hyphae and brown deposits were less frequent in wood inoculated with # 387K or # 387T when compared to specimens inoculated with # 3871. Wood inoculated with Mix-B demonstrated a few small, light brown deposits in ray parenchyma cells but smaller, light brown particles were more frequent than in controls. Surprisingly, bacteria were 50 I m • HPNI i l l tfmMLm**tt Figure 9: Tsuga heterophylla. Radial section (160x), presence of brown deposits in O. piceae #3871 inoculated wood after 6 weeks. 51 rarely observed in either ray parenchyma or tracheid cells of the inoculated wood. 3.4 Discussion The pronounced occurrence of brown deposits in ray parenchyma cells indicates their central role in the process of hemlock brownstain. Biosynthesis of phenolic substances occurs in ray parenchyma cells as they undergo cytological changes upon aging (Fengel, 1970; Parameswaran and Bauch, 1975). It is also known that parenchyma cells throughout the sapwood are engaged in storage and distribution of nutrients but are also involved in heartwood formation (Panshin and de Zeeuw, 1980) . Proanthocyanidins (PAs), condensed tannin precursors (Hemingway, 1989) are colourless but they undergo chemical changes, coincident with disintegration of cytoplasm as wood matures (Stafford, 1988) . In some cases PAs may come into contact with enzymes such as polyphenoloxidases or peroxidases, forming condensed tannins (Stafford, 1988). The localized nature of the distribution of brown deposits is quite striking. Possible explanations are the variable distribution of PAs (Stafford, 1988) and the high degree of physiological specialisation of ray cells even within an apparently homogeneous ray (Sauter and van Cleve, 1989). 52 The occurrence of discoloured tracheids surrounded by stain-free tracheids was seen in this study, and agreed with a recent report (Ellis and Avramidis, 1993) . These authors suggested that physical-chemical differences between tracheids may create critical conditions which are necessary for formation of stain. They also reported that the lumina of some tracheids were lined with thin deposits. However, in the current study brown chromophores were also seen in bordered pits (Figure 10) and in intercellular spaces. Thus for the first time this microscopic study has provided strong evidence that precursors of hemlock brownstain are mobile. The presence of microorganisms in wood specimens examined is of particular interest. Bacteria were frequently present in bordered pits but bacterial degradation of cell wall substance was rarely seen, unlike the observations recorded for kiln burned amabilis fir (Barton and Smith, 1971) . Bacteria have been demonstrated to induce discolorations in other wood species (Bauch et al. , 1985; Hedley and Meder, 1992). In this study fungi, rather than bacteria were frequently seen in discoloured hem-fir samples, the hyphae being observed in both rays and tracheids. Hyphae were usually hyaline and were frequently enclosed in a coloured deposit. Of particular interest was an observation showing hyphae entering the lumen of a stain-free tracheid via a bordered pit but the hyphae were enclosed in brown sheaths (Figure 11). 53 Figure 10: Tsuga heterophylla. Transverse section (250x), brown deposits in a pit connecting discoloured tracheids. 54 Figure 11: Tsuga heterophylla. Transverse section (400x), hyphae surrounded by a brown sheath in an otherwise unstained tracheid. 55 This observation agreed with a recent study in which the presence of non-pigmented 0. piceae or Sporothrix sp. was strongly correlated with brown deposits in hem-fir lumber (Smith and Spence, 1987). Yellow discolorations recorded in green oak (Quercus sp.) heartwood were explained as being caused by a mold fungus altering the cell contents (Bauch et al., 1991). In the current study the fungus appeared to modify the cell contents and produce a brown polymer locally around the hyphae. In this study DMB reacted to catechin and epicatechin when spotted on filter paper at low (50/xg/mL) concentrations. The DMB reagent also showed that catechin and/or epicatechin were present in microscopic sections free of brown deposits. The localized distribution of catechin and/or epicatechin was therefore histochemically demonstrated for the first time in wood. The specific sensitivity of DMB to catechin type units has been demonstrated on condensed tannin precursors (catechin) in sections of one week old roots of cotton seedlings (Mace and Howell, 1974) . To the best of my knowledge, however, the use of DMB on wood has not previously been reported. In addition the DMB reagent provided strong evidence that catechin/epicatechin were not only present in brown chromophores but also in brown deposits of methanol or acetone/water (7:3) extracted hem-fir sections (15ptm) . The latter observation indicates the polymeric nature of brown chromophores agreeing with 56 a similar study (Halloin, 1982) in which DMB was used to locate catechin and their derived condensed tannins in cottonseeds. Catechin is thought to be the main tannin precursor in cotton plants and cottonseeds (Halloin, 1982). In the current histochemical study the brown deposits in hem-fir lumber were demonstrated to contain units of catechin/epicatechin. Inoculation of sap used in this study showed that bacteria and fungi were capable of growing in sap without additional nutrients. This observation agrees with a recent study (Schmidt, 1986; Schmidt and Mehringer, 1989) . Furthermore the increasing pH of inoculated sap indicated that proteins were utilized independent both of microorganisms and the source of sap evaluated. Schmidt (1986) reported that sap with a high glucose content promoted acidity while a low glucose containing sap encouraged alkalinity upon inoculation with bacteria. In the current study glucose or arabinose were not indicated in sap of Hem-2 (uninoculated and inoculated) but they may have been present below the detection levels used (Sutcliffe, 1992). Since proteins are known to bind to phenols and PAs (Hagerman, 1989) a release of phenolics upon microbial utilization of proteins may have contributed to observed colour changes in Hem-2 sap. Solubility and reactivity of phenols increase at higher pH encouraging rearrangements and oxidation of these compounds (Daniels, 1993a). In the current study qualitative changes in 57 lower molecular weight compounds in inoculated Hem-2 sap and the decrease of detectable (HPLC) compounds suggested that phenolics were oxidized. Thus oxidized compounds may have become insoluble in inoculated sap and did not elute from the column used for HPLC analysis. Bacteria have been shown to discolour solutions of polyphenols by a peroxidase system (Shortle et al. , 1978) . Also sapwood sawdust, in the presence of hydrogen peroxide could be discoloured by bacteria (Shortle et al. , 1978) . Evans and Halvorson (1962) have speculated that bacterial phenol oxidase caused hemlock brownstain. Brown colorations were produced in the cambial region of freshly debarked western hemlock when mushroom peroxidase and hydrogen peroxide were added (Hrutfiord et al. , 1985) . In the present study no attempt was made to verify phenol oxidizing enzymes in inoculated sap. The pH change associated with colour change of inoculated sap was not continuously measured over time and thus the precise pH at which the colour change started remains unknown. In a recent study (Schmidt and Mehringer, 1989) in vitro brown discolorations were recorded in sap of beech (Fagus sylvatica L.) inoculated with different bacteria at pH 7.3. The authors, however, prevented in vitro discoloration of beech sap when glucose and fructose were added thus keeping the pH below 7. In another study stabilization of the pH of the wood surface of Ilomba (Pycnanthus angolensis Exell.) wood with formic acid inhibited discoloration (Bauch, 1986) . This was linked to a pH shift caused by bacteria (Starck et 58 al., 1984; Bauch et al. , 1985; Yazaki et al. , 1985). In this context acidic chemicals have been shown to inhibit hemlock brownstain in laboratory trials (Barton and Gardner, 1966) while alkaline treatments have been thought to intensify hemlock brownstain (Swan, 1984b). The reasons for colour changes observed in sap of Hem-2 and in amabilis fir but not in Hem-1 are not yet understood. Ambrosia beetle infestation, indicated by pinholes and the accompanying staining fungus in some regions of the Hem-2 board, may have predisposed this wood substrate to discolorations. The presence of fungi and bacteria in wood of Hem-2 was not verified but Whittaker (1962a) has reported bacteria in insect-damaged trees of western hemlock. Changes in the total phenol content may have occurred in Hem-2 as has been reported in cricket bat willow wood infected with a bacterium (Wong and Preece, 1978), in wounded red maple (Shevenell and Shortle, 1986) and in wetwood of western hemlock (Schroeder and Kozlik, 1972). 0. piceae isolates produced in vitro brownstain in western hemlock sapwood but the discoloration seemed to be less intense than found in a similar study (Smith and Spence, 1987). These authors demonstrated a dark brown discoloration 6 weeks after inoculation with 0. piceae isolates. Smith and Spence (1987) employed 5 cm x 10 cm boards which had produced brown endstain prior to inoculation. Thus the authors believed that their material was 59 highly prone to brownstain. However, heating (103°C) commonly used to identify hem-fir lumber susceptible to brownstain, produced no macroscopic brownstain and indicated that the substrate was not highly susceptible to brownstain. It is possible that insufficient precursors to cause hemlock brownstain are present in small size samples (Ellis, 1993) or surface drying is too fast to enable mass flow of sap to the surface for oxidation reactions to occur. The Mix-B inoculated wood provided inconclusive results. Brownstain has been transmitted from bacterial infected, discoloured western hemlock slabs to sound wood when placed in a pond but variable results were observed (Evans and Halvorson, 1962; Whittaker, 1962a). It is possible that moisture content of autoclaved wood specimens used in the current study was too low to allow prolific growth of bacteria. Autoclaving of wood prior to inoculation may have altered composition of wood extractives and thereby reduced the effect of microbial infection on in vitro production of brownstain. Heating of sap of western hemlock has been indicated to cause darkening from light yellow to brown (Evans and Halvorson, 1962) and a slight discoloration was also microscopically recorded in the present study. Wood used by Smith and Spence (1987) was not autoclaved prior to inoculation. 60 In the present study it was not possible to determine whether enzymes present in living cells were able to discolour parenchyma cells since autoclaving would almost certainly have inactivated them. Enzymes of living cells have been reported to react with accessory compounds causing wood discolorations (Bailey, 1911; Stutz, 1959; Hrutfiord and Luthi, 1981; Ota et al. , 1991). However, atmospheric oxidation has also been demonstrated to produce less discoloration than microbial oxidation for inoculated beech wood (Schmidt and Mehringer, 1989) . 3.5 Conclusions Microscopically observed, the spatial distribution of brown deposits in hem-fir lumber was irregular in ray parenchyma cells and even more so in tracheids. This is possibly a result of a localized microbial infections. For the first time in situ histochemical examinations provided evidence that brownstain chromophores in hem-fir lumber are at least partially composed of catechin or epicatechin. Catechin/epicatechin were also demonstrated in wood free of brownstain. The localized nature of the distribution of catechin/epicatechin was further demonstrated and this observation was found to resemble the variable distribution of brown deposits seen in discoloured wood. 61 In vitro production of brownstain was demonstrated in both inoculated wood and inoculated sap of western hemlock. A range of microflora were found to shift the pH of sap from slightly acidic to slightly alkaline. Alkaline conditions promoted brown discoloration in sap but other, unknown factors (e.g., composition of extractives) were also implicated. 62 4.0 MONITORING PRODUCTION OF BROWNSTAIN IN WESTERN HEMLOCK LOGS AND LUMBER DURING STORAGE. 4.1 Objectives The objectives of this study were to: a) evaluate the effect of storage time and storage condition on the production of brownstain in freshly felled trees and in lumber sawn from the trees; b) to determine the microflora associated with brownstain; c) to determine changes in the gross phenolic composition associated with brownstain. 4.2 Materials and Methods 4.2.1 Field work at Chamiss Bay Logging site description A logging site belonging to International Forest Products LTD (Interfor), was selected at Chamiss Bay on Vancouver Island. According to Interfor's preharvest silvicultural prescription the site (Forest classification HB: 951, hemlock-"balsam" over 260 years old, Site Class I) is on East to South-East aspect, 350-370 metre elevation above sea level and was located within the coastal western hemlock wet biogeoclimatic subzone. The soil depth was estimated to be 55 cm and to be well drained. Deer fern (Blechnum 63 spicant (L.) Roth), sword fern (Polystichum muniturn (Kaulf.) Underw. (Sward F.)) and huckleberry (Vaccinium parvifolium Smith) were the main indicator species associated with the mature western hemlock and amabilis fir trees. Medium western hemlock dwarf mistletoe (Arceuthobium camplyopodum Engelm. forma tsucrensis Gill) infection was scattered through the site. Selection of western hemlock trees Twenty eight western hemlock trees were selected in Interfor's logging site 211A on August 5, 1993. The trees were classified by two Interfor foresters for basal diameter and stem degrade (e.g. frost scars, dead branches, mistletoe, sweep) and they were felled the following morning (August 6, 1993). On August 7, 1993, thirteen logs were transported from the logging site to the sorting yard at Chamiss Bay. The other 15 logs were left at the site because the grapple yarder was unable to reach them. Initial sampling of western hemlock logs At the sorting yard the logs were bucked (10-15 cm) at the butt end prior to cutting off two consecutive, 10-15 cm thick disks (hereafter referred to as outer disk and inner disk) from the butt end. The disks were labelled and individually wrapped in polyethylene and they were then stored outdoors overnight. Additional western hemlock logs felled on August 4, 1993 and other 64 western hemlock logs which had been logged approximately 3-4 weeks prior to this date at site 211A, and which had been on the ground since then, were also sampled as described previously. The logs were cut to 3 metre long sections following the sampling of disks and they were then oriented on dry-land in a North-South direction. The disks were transported back to Forintek Canada Corp., Vancouver, B.C. on August 9, 1993. Second sampling of western hemlock logs Logs stored at Chamiss Bay were visited again after two months. The ends of the logs were examined visually for brown endstain. Two disks (10-15 cm) were then sawed from the same butt end of the logs, labelled and wrapped as described above (see . Logs were then selected at random to be put in ocean water storage (#'s 2, 3, 4, 19, 20, 25, 27, 33, 36, 40, 43, 46, 47, 48, 50; 51, 54, 55) or to be placed on dry land (#'s 1, 5, 6, 16, 17, 24, 26, 32, 35, 37, 41, 42, 44, 45, 49, 52, 53, 56). The dry-land logs were stacked in one pile whereas all water-stored logs were bundled together prior to placing in the water. Disks sawn from 17 logs (#'s 1, 3, 4, 6, 16, 17, 24, 25, 26, 32, 33, 40, 41, 42, 43, 44, 48) were transported back to the laboratory while the others were left behind due to a restricted load capacity of Forintek's vehicle. 65 Inspection of western hemlock logs after 9 months storage The western hemlock logs were evaluated for production of brownstain at the log endcuts following approximately 9 months of storage. The logs stored in ocean water had been taken out after 5-6 months and subsequently stored exclusively on land. Logs were visually examined on both cross-cut ends. Wood chips were then taken at random from discoloured and non-discoloured regions and they were analysed quantitatively for calcium, manganese and iron using X-ray spectroscopy. Ten logs (five stored in water and five stored on land) were then chosen to be sawn up into lumber. The 10 logs included the six logs (#'s 1, 16, 40, 42, 44, 48) evaluated for microorganisms and wood extractives plus 4 additional logs (#'s 4, 6, 35, 43) which were selected at random. Sawing of western hemlock logs The ten selected logs were sawn up into lumber 9-10 months after felling. A bandsaw was used to produce 5 cm thick boards at Chamiss Bay. Each log was rotated after the first opening cuts to produce edge grain material. Board surfaces were cleared of sawdust and a fungicide, F2 concentrate (11.4% didecyldimethyl ammonium chloride + 16.8% disodium octaborate tetrahydrate) diluted 12:1, water:F2 concentrate, was brush-applied to the wood surface to control mould fungi and sapstaining fungi. Photographs were taken of representative boards from all logs. Boards produced from 66 each log were closed-piled following normal industrial practice. Inspection of sawn lumber after storage Visual examination of boards was undertaken approximately two months after sawing and about 12 months after felling of the trees. The boards were evaluated for surface brownstain and photographs were taken of representative examples from each log. 4.2.2 Laboratory work using Chamiss Bay samples Assessment of susceptibility to brownstain in disks Susceptibility to brownstain was assessed in the outer disks collected at the initial sampling. To determine brownstain samples were kiln-dried (48 hours at 60°C followed by 48 hours at 80°C) using a laboratory kiln immediately following the arrival at the Forintek laboratory. The samples were then visually examined and brownstained regions were delineated. The inner disks were stored in a cold room (3-5°C) until use. Selection of logs susceptible to brownstain After the initial brownstain assessment six disks from six different logs were selected for further laboratory studies. Three 67 disks were from logs # 1, 40 and 44 which produced brownstain and three disks, from logs # 16, 42, and 48 which remained stain-free, were selected as controls. Sample preparation from selected disks For logs which developed brownstain on the outer disks during kiln-drying the outlines of six stained regions were transferred onto the inner disks (Figure 12). The first area was selected around the centre of the brown discoloration, the second from a non-stained region, the third from the edge of a discoloured region and the fourth from a non-stained area. These four regions were coded (e.g. 1-1; 1-2; 1-3; 1-4) and they were then sawn out using a bandsaw. Each sample, approximately 4 cm x 4 cm x 15 cm, was then divided into three equal sized (4 cm x 4 cm x 5cm) pieces, a centre piece and two outer samples. Each centre piece was immediately sealed in a plastic bag and was stored in a refrigerator (4°C) for 24 hours prior to undertaking fungal isolations. The outer pieces prepared from each sample region were used to determine moisture content (MC). Two additional samples were prepared from a brownstained and a stain-free region, coded (e.g., 1-BS; 1-NS) and placed in a freezer prior to collection of sap (Figure 12). Sample preparation from non-stained disks (# 16, 42, 48) was undertaken at random from similar cross-sectional regions of the inner disks to correspond with the sampling from stained disks. 68 1-NS * ISOLATION OF FUNGI * BACTERIAL COLONIES * MOISTURE CONTENT PRESSATE *pH *TSP *HPLC Figure 12: Flowchart showing sampling regime of log #1 from brownstained and non-stained regions in fresh and two month-old disks. 69 Initial isolation of fungi Fungal isolation attempts were performed by splitting each centrepiece (4 cm x 4 cm x 5 cm) in half using a flame-sterilized chisel. Eight small (approximately 8 mm x 2 mm) chips were removed from the freshly exposed wood surface using a flame-sterilized chisel and forceps. Four flame-sterilized wood chips were then individually placed in the centre of a Petri dish containing tetracycline malt agar (2% malt extract, 1% bacto-agar plus 100 ppm tetracycline to control bacterial growth) (MA-T) while the other four flame-sterilized chips were plated on a starch-casein-nitrate agar (SCN - containing starch 10g; casein 0.3g; potassium nitrate 2.0g; sodium chloride 2.0g; potassium monophosphate 2.0g; magnesium sulfate 0.05g; calcium carbonate 0.03g; ferrous sulfate O.Olg; rose bengal 0.35g; 1% bacto agar, distilled water 1 litre). The dishes were incubated in the dark at 25°C for 8 weeks. Petri dishes were inspected after 1, 2, 3, 4 and 8 weeks and colonies growing out of the wood into the media were sub-cultured into petri dishes containing either MA-T or SCN. The purified cultures were then incubated on malt agar (2% malt extract, 1% bacto agar) prior to culturing on slant agar tubes. The purified cultures were stored at 4°C. Harvesting of sap 70 The two samples prepared from each disk (e.g., 1-BS; 1-NS) for collection of sap were individually placed in plastic bags and then put into a hydraulic press. The pressate, decanted into 2 0 mL vials with teflon seal caps, was frozen until use. pH measurement A pH meter was used to determine hydrogen ion concentration in the stirred sap following thawing. HPLC analysis of pressate Thawed sap was injected onto the column of a HPLC and the effluent was qualitatively monitored to determine the gross phenolic composition at a wavelength of 280 nm as described by Daniels (1993b). In addition, the amount of catechin and epicatechin was quantified in the sap using standards purchased from Sigma (St. Louis, MO). Total soluble phenol determination Western hemlock sap was analysed for total soluble phenol (TSP) content using the Folin-Ciocalteau reagent (Sigma, St. Louis, MO). Sap was diluted 25 times by adding 0.02 mL of the sap to 0.480 mL 71 of distilled water followed by 2.5 mL of Folin-Ciocalteau reagent, diluted 10 times with distilled water, and then 2 mL of sodium carbonate solution (8.25 g sodium carbonate monohydrate in 100 mL of distilled water). The solution was first placed in a water bath at 50°C for 5 minutes and then in a cool water bath (10- 15°C) for approximately 5 minutes. Samples were transferred into polystyrene cuvettes (1 cm path length) and absorbance was read at 76 0 nm against a blank (distilled water). TSP content was calculated using the average absorbance of three readings for each sample. Gallic acid (Sigma, St. Louis, MO) was used to establish a standard curve (5-80 /zg/mL) . The measurements were performed with a Shimadzu spectrophotometer (Shimadzu Scientific Instruments, Inc., Columbia, MD). Sample preparation from disks after 2 months log storage The outer disks sampled after two months of outdoor log storage, were heated to induce brownstain as outlined in The disks that were used for these further studies, came from the same logs as previously sampled, log 1, 40, 44 with brownstain and #16, 42, 48 without brownstain. Six regions were mapped on each inner disk and these areas were then sampled according to the procedure described in 72 Isolation of fungi and quantification of bacterial forming colonies Microorganisms were isolated from the centre of the three pieces sampled from each region according to the procedures described in, except that all wood chips were plated aseptically on MA-T and SCN, respectively. The plates were incubated in the dark (25°, 70% RH) , monitored and sub-cultured as previously described in In addition bacterial colony-forming units (CFU) were counted using one of the two outer pieces per sample region (e.g., 1-1) which had been sealed in a plastic bag and then stored in a refrigerator for approximately three weeks. It was assumed that each CFU was formed from one viable cell. For each sampling one outer piece from each region (e.g., 1-1) was split in half using a chisel. Both pieces were flame-sterilized in a laminar flow hood and then individually heat-sealed in a polyethylene bag. The sealed wood sample was then positioned at the upper edge of the bag and placed in a hydraulic press. As pressure was applied sap was collected in the lower, hanging part of the bag. The bag was removed from the press and re-sealed below the wood sample to separate the sap from the wood. The sap was then removed to a laminar flow hood and a serial dilution plating was performed using 1.5% nutrient agar (NA) (Difco Laboratories, Detroit, MI) as growth medium. Two plates were employed per serial dilution and for control (sterile, distilled 73 water). All plates were incubated at room temperature for 48 hour prior to counting CFU. Prior to using this specially developed technique a number of checks were run. Additional NA Petri dishes were seeded with sterile, distilled water which was transferred in a plastic bag prior to plating. This experiment which was undertaken to verify the methodology employed, showed that the polyethylene bags were free of contamination. Other NA plates were incubated to check whether the pressing procedure would assure aseptic handling of the sap. For this purpose distilled, sterile water and a bone-dry, flame-sterilized wood sample were sealed in a similar plastic bag, and handled according to the procedures described previously. All wood samples were weighed prior to pressing sap and following drying (103°C, 12 hours) to determine moisture content. Sap analysis The pH, qualitative HPLC analysis of the gross phenolic composition, quantitative catechin and epicatechin analyses and TSP were determined (see 4.3 Results and Discussions 74 4.3.1 Production of brownstain in fresh and 2 month-old logs. The cross-cut ends of the freshly-felled western hemlock trees looked brownstain and defect-free. Wet zones were noticed in the inner heartwood and/or in patches within the sapwood. The absence of hemlock brownstain on the fresh log ends suggested that this type of coloration is not produced in the living tree and thus is different from such discolorations (Basham and Taylor, 1965; Sachs et al. 1966; 1974; Shigo and Sharon, 1968; Cosenza et al. , 1970; Shigo et al. 1971; Wilcox and Oldham, 1972; Shigo and Hillis, 1973; Mackay, 1975; Blanchette et al. , 1981; Bauch et al. , 1982; Djuesiefken et al., 1984; Walter, 1993)). Furthermore, hemlock brownstain, requiring storage to occur, differs from that observed for example on log ends of red alder, which develop within hours after the felling of the tree (Hrutfiord and Luthi, 1981; Terazawa et al., 1984; Kreber et al., 1994) . Brownstain was produced in inner heartwood regions and in outer regions of some disks following kiln-drying (Table 2) . In the outer regions brownstain seemed to be associated with the heartwood/sapwood interface. In addition, brownstain was observed in regions where decay was conspicuous. For instance, disk # 44 showed a decayed zone across the section and this zone was associated with brownstain. While wet zones, perhaps "wetwood" 75 TABLE 2 : CLASSIFICATION OF SELECTED WESTERN HEMLOCK LOGS. LOG # l2 (70)3 2 (85) 3 (40) 4 (45) 5 (30) 6 (40) 16 (75) 17 (80) 19 (30) 20 (30) 24 (45) 25 (40) 26 (30) 27 (45) 327 (60) 34 (60) 35 (70) 37 (50) 408 (45) 41 (50) 42 (45) 43 (40) 44 (50) 45 (30) 46 (45) 48 (50) TREE DEGRADE1 Frost crack; clear 7 m; light mistletoe infection frost crack grown over; large dead branch @ 5 m; moderate mistletoe infection frost crack @ 7m; large dead branch @ 4m; heavy mistletoe on top clear; heavy mistletoe infection; dead branches; heavy mistletoe infection; dead branches; scars; sweep; moderate mistletoe; dead branches; frost scars; heavy mistletoe; scars; moderate mistletoe; scars; ND6 mistletoe; dead branches; frost scars; ND ND ND ND ND ND ND ND ND ND ND ND ND STAIN H4 No No No No No No H;H/S5 No No No No NO No No H No No H No No No H; H/S No No No 76 49 (45) 50 (30) 51 (45) 52 (40) 53 (30) 54 (30) 55 (45) 56 (30) ND ND ND ND ND ND ND ND No No No No No No No No 1 = refers to degrade assessed on the standing tree by Interfor. 2 = trees 1-27 were selected at site 211A. 3 = numbers in brackets refer to diameter (in cm) at the butt. 4 = brownstain observed in heartwood after kiln-drying. 5 = brownstain observed in heartwood/sapwood interface after kiln drying. 6 = no degrade was determined 7 = logs 32-37 were selected at the sorting yard at Chamiss Bay. 8 = logs 40-56 were selected at the sorting yard at Chamiss Bay and they had been harvested from site 211A 3-4 weeks prior to selection. 77 (Ward and Zeikus, 198 0) were frequently observed in the fresh western hemlock disks these areas showed no particular susceptibility to brownstain following kiln-drying. After two months of land-storage hemlock brownstain was not found on the original cross-end cuts of western hemlock logs. However, a visit to the logging site 211A to inspect some of the tree stumps sampled two months before showed pronounced brownstain. For instance brownstain was found in the heartwood of stump #16 whereas disk segments removed from this tree remained stain-free both at the initial brownstain assessment and after two months. It is likely that sap flow from the root system continued following felling of the tree which resulted in an accumulation of phenolic compounds producing brownstain at the stump surface. Heating of disks from the logs stored for two months on land provided similar results to those recorded in the fresh disks. Accordingly brownstain was found in the same regions within the disks of logs # 1, 40 and 44 whereas disks #16, 42 and 48 remained stain-free. Thus heating of western hemlock disks produced consistent results in determining those logs with potential for brownstaining. The fact that the control logs (# 16, 42, 48) remained stain-free upon heating also suggested that the two month-old logs were not yet predisposed to brownstain. 78 4.3.2 Inspection of cross-cut ends after 9 months storage Inspection of log ends demonstrated that most logs had produced some endstain after 9 months of outdoor storage (Table 3). Two distinct differences were noticed between water-stored and land-stored logs. Firstly, the water-stored logs showed a much darker colour ranging from dark brown to black (Figure 13). These dark colorations were not found on the cross-end cuts of logs stored on land. In the water-stored logs dark colorations were noticed in the heartwood and/or sapwood regions where they were associated with groups of adjacent growth rings. When wood chips were removed from these regions the brown-black colour in the water-stored logs was found to be superficial. X-ray spectroscopy (data not shown) for calcium, manganese and iron in black discoloured wood chips showed no qualitative difference compared to wood chips with lighter discolorations (Ingram, 1993). Secondly, water-stored logs produced dark brown to black stain in heartwood and sapwood (Figure 13) while land-stored logs showed brownstain only in the sapwood and/or in the sapwood/heartwood interface (Figure 14). These observations demonstrated that log storage time, independent of storage conditions, promoted development of brownstain because it was not observed on the log ends of the two month-old logs. Log age has been suspected to influence both production of hemlock brownstain (Evans and Halvorson, 1962; Kreber and Byrne, 1993; Scheel, 1993) and discolorations in other softwoods and hardwoods 79 TABLE 3: PRODUCTION OF BROWNSTAIN ON LOG ENDS AFTER 9 MONTHS OF OUTDOOR STORAGE. LOG# 1 2 3 4 5 6 16 17 19 20 24 25 26 27 32 33 35 37 40 41 42 43 44 45 46 STORAGE land water water water land land land land water water land water land water land water land land water land land water land land water APPEARANCE OF LOG ENDS stain-free HW1 darker stained than SW2; black-brown stain associated with certain growth rings SW light brown in places; HW dark-brown w/ black patches; HW/SW3 interface brownstained dark and wet; black stain in some growth rings; inner HW enclosed by a black band SW light brown; SW w/ some brownstain; HW/SW interface w/ narrow brownstained band SW greyish; dark-brown zone around HW SW greyish w/ brown patches wet and dark, possibly dirt log was missing SW w/ brownstain inner HW w/ dark brown stain stain-free stain-free SW w/ brownstain (outer 10 cm) log was missing SW w/ brownstain (outer 5 cm) SW w/ brownstain HW w/ patches of black and brownstain SW w/ brownstain (outer 5 cm) stain-free HW dark brown surrounded by a black band HW w/ dark brown patches SW w/ brownstain HW grey-green; SW w/ brownstain 80 47 48 49 50 51 52 53 54 55 56 water water land water water land land water water land HW grey surrounded by a 2-3 mm brown band SW w/ black patches SW w/brownstain SW w/ brownstain (outer 1-2 cm) HW light brown enclosed by a black band SW w/ light brownstain SW w/ brownstain HW grey w/ brownstain in some growth rings; HW enclosed by a 1-2 cm black band SW w/ light brownstain stain-free 1 = HW refers to heartwood 2 = SW refers to sapwood 3 = HW/SW refers to transition zone between heartwood and sapwood 81 .** \ * Figure 13: Representative, water-stored log showing dark colorations after 9 months of storage. 82 maBm •tffjk ' Figure 14: Representative, land-stored log showing brownstain in sapwood after 9 months of storage. 83 (Herman, 1931; Rasmusseen, 1940; Stout, 1950; Koltzenburg, 1975; Forsyth, 1988). In the current study salt water storage also demonstrated a striking effect on the intensity of the coloration developing on the log ends. Several studies have shown that fresh water storage of logs can promote discolorations on the log ends or on the lumber sawn from the logs (Hoster, 1974; Braun and Lewark, 1992; Forsyth and Amburgey, 1992; Hedley and Meder, 1992 Lubbers and GroiS, 1992; Rathke, 1991). Some explanations have been given for this water-storage phenomenon, for example: infection of water-stored logs by bacteria which appeared to modify glycosidic flavonoids (Hedley and Meder, 1992) ; death of parenchyma cells causing dissolving of phenols (Hoster, 1974); and redistribution of bark phenols into the wood (Adler, 1951; Lubbers and GroS, 1992). Interestingly, Lubbers and GroS (1992) showed that debarking of spruce (Picea sp.) and true fir logs prior to storing in water decreased brownstain on the logs ends and on the lumber surface. Similar observations were made when investigating brownstain in radiata pine lumber produced from logs debarked prior to water storage (Hedley and Meder, 1992). In the present study redistribution of bark phenols may have contributed to hemlock brownstain associated with sapwood regions but it is unlikely to have influenced the dark colorations found deeper in the wood, for instance in the heartwood/sapwood interface or even in the heartwood. However, it is possible that salt water 84 with a pH of above 7 and with trace amounts of metal ions promoted the dark colorations on the log ends of the water-stored western hemlock. Extensive quantitative metal ion analysis of wood would be required to determine whether these ions are associated with, or play a role, in the dark colorations occurring in stored logs in ocean water. 4.3.3 Inspection of lumber after two months of outdoor storage Freshly cut western hemlock boards produced from logs which had been stored for 9.5 months, showed no brownstain on the wood surface. This observation agreed with recent studies (Lubbers and GroS, 1992; Rathke, 1991) wherein brown discolorations were also absent in freshly sawn spruce and true fir lumber but developed with time. Observations made in this study suggests that hemlock brownstain, rather than being formed in the living tree, as for example with the black colorations associated with the heartwood of yellow cedar (Smith, 1970) or the discolorations associated with tree wounds (Shigo and Sharon, 1968), involved atmospheric oxidations. However, brownstain was observed on the lumber surface when inspecting it after 2 months of storage; this was the first inspection, so it is not known how quickly (before two months) the stain first appeared. Staining soon after cutting into lumber has been frequently reported by mill managers (Blake, 1992). Lubbers 85 and GroS (1992) and Rathke (1991) have also demonstrated that brown surface colorations formed a few days after sawing spruce and true fir logs. The boards sawn from log #4 (water-stored) developed the most severe brownstain, darkly disfiguring the board edges (Figure 15). In addition the faces of # 4 boards were also discoloured, mostly concentrated at the end of the board and in the outer regions of the board (Figure 15) . Boards from logs # 48 and 43 (both logs stored in water) showed brownstain at the wood surface in a striped pattern. The coloration was associated with the outer regions of the boards and rarely affected the whole length of the board. Lumber from logs # 1 , 6, 16 and 35 showed a little brownstain, most noticeably on the edges whereas the faces were less affected. Lumber from logs # 4 0 , 42 and 44 looked very clean, showing only minor brownstain on the edges of some boards whereas the faces were stain-free. In this study water-stored logs produced considerably more brownstain in sawn lumber than did logs stored on land. However, water-storage did not always result in brownstained wood: the lumber sawn from the water-stored logs #4 0 and 44 remained stain-free. While both logs (40 and 44) developed brownstain upon heating and thus were thought of as being susceptible to brownstain this observation indicated that brownstain produced in kiln-dried western hemlock differed from those colorations developing during 86 jfe Figure 15: Production of severe brownstain on the boards edges faces of lumber sawn from log #4. 87 seasoning. However, the factors which prevented surface brownstain on sawn lumber of logs # 40 and 44 are not well understood, wood moisture content may have been involved. It is possible that logs #40 and 44 were not submerged during water-storage as some logs in a bundle always float above the water surface. Thus drying could have occurred which then yielded lumber with a lower moisture content. Lubbers and GroS (1992) reported that spruce and true fir logs with a higher moisture content produced more brownstain in the sawn lumber. In the current study another interesting observation was made, namely that brownstain did not develop on a closed-piled board or at the sticker location, but brownstain was evident on the wood surface that was exposed to air, where boards were unevenly stacked. This observation indicated that reduced photooxidation, inhibition of moisture movement or reduced oxygen content, could result in reduced brownstain in the sawn lumber. Inhibition of moisture movement and reduced oxygen content have been proposed to explain the occurrence of less discoloration underneath a sticker in western hemlock (Evans and Halvorson, 1962) as well as in other softwoods (Anderson et al. , 1960; Miller et al. , 1983) and hardwoods (Bauch et al. , 1991) . However, photooxidation has not been related to hemlock brownstain, but photo-labile constituents of western hemlock may well be involved as indicated in the current study. Many wood species have been known to change colour when 88 exposed to light and for example, woods with stilbenes have been reported very photo-labile (Morgan and Orsler, 1968). Light-induced colour changes of catechin, epicatechin and dihydroquercetin have recently been demonstrated in a study on chemical brownstain in Douglas-fir (Arvey, 1993) . 4.3.4 Isolation of microorganisms Of 192 isolation attempts made from the freshly, felled western hemlock, 25 fungal cultures resulted. From this initial sampling the fungi cultured were categorized as: green moulds (3) , pigmented fungi (16) and basidiomycetes (6) (Table 4). After two months of log storage some similar 27 cultures were isolated yielding 22 pigmented fungi and 5 yeasts but no decay fungi and green moulds. At both sampling times log # 44 yielded the most fungi and log # 42 the second most isolates. Logs # 16 and 40 each yielded one isolate after 2 months but 5 fungi (log # 40) and 4 fungi (log # 16) were isolated at the initial sampling. Logs # 1 and 48 yielded no fungi except one isolate cultured from log # 48 at the initial sampling. Clearly, it can not be assumed that all fungi were isolated and that the relative frequency of their occurrence represents somewhat a measure of ease of isolation. Fungi producing red pigmentation on malt agar plates were the most frequent isolates from western hemlock at both sampling times. Similar growth characteristics of these isolates indicated the same 89 TABLE 4: FUNGI1 ISOLATED FROM WESTERN HEMLOCK LOGS SAMPLED AT CHAMISS Bay. # l-l*4 1-2 1-3* 1-4 16-1 16-1 16-1 16-2 16-3 16-4 40-1* 40-1* 40-2 40-3* 40-3* 40-3* 40-4 42-1 42-2 42-3 42-4 42-4 42-4 42-4 42-4 44-1* 44-1* 44-1* 44-1* 44-1* 44-1* 44-1* FIRST SAMPLING2 Code A5B5 A6B5 A7B5 A2M AIM A7S A2S A5S A5aS A10S A11S A8S A12S A9S A1S A2B6 A13S A6S A4B6 Fungus ----basidiomycete basidiomycete basidiomycete green/brown mould --Penicillium sp. Phialocephala virens Ophiostoma piceae Trichocladium canadense Ascocoryne sarcoides -A. sarcoides A. sarcoides A. sarcoides A. sarcoides A. sarcoides ---Leptodentium elatius basidiomycete Phialophora sp. Phialophora melinii basidiomycete --SECOND SAMPLING3 Code BY B21S B7S BIB B8S B9S BIOS B U S B20S BIS B2S B3S B4S B13S B14S B15S Fungus -Yeast -P. virens _ _ A. sarcoides A. sarcoides -A. sarcoides A. sarcoides A. sarcoides A. sarcoides L. elatius L. elatius L. elatius L. elatius L. elatius Phialophora sp Phialophora sp 90 44-1* 44-2 44-3* 44-3* 44-3* 44-3* 44-3* 44-3* 44-4 44-4 48-1 48-2 48-3 48-4 A3B A4Sa A4Sb A14S A3M A3S A1B -A. sarcoides A. sarcoides black pigmented T. canadense _ -_ green mould P. melinii -_ basidiomycete -B16S B12S B5S B6S B17S B18S B19S Leptographium sp. A. sarcoides A. sarcoides dark pigm. dark pigm. T. canadense A. sarcoides -_ ----1 = species were tentatively identified by K.A. Seifert, Agriculture, Canada. 2 = logs were sampled latest 3-4 weeks after felling the trees. 3 = logs were sampled following a 2 month storage period (August-October, 93) at the Chamiss bay sorting yard. 4 = sample regions in bold with asterisk refer to regions associated with brownstain 5 = fungi showed a similar growth morphology on malt agar plates. 6 = fungi showed a similar growth morphology on malt agar plates. 91 fungus tentatively identified as Ascocoryne sarcoides (Jacq. ex. S.F. Gray) Groves & Wilson by Dr. Keith A. Seifert, Agriculture Canada. A. sarcoides has been reported to be a common heartwood inhabitant of living spruce, true fir, pine, Douglas-fir (Pseudotsucra menziesii (Mirb.) Franco), hemlock, and larch (Larix sp.) in Canada (Etheridge, 1970) and in Norway spruce (P. abies (L.) Karst) (Roll-Hansen and Roll-Hansen, 1979). In western hemlock this fungus has been demonstrated to be confined to the central heartwood column in the lower one-third of the stem and in the roots (Etheridge, 1970). Etheridge also showed that the fungus, a primary non-decay invader preceded wood-decay fungi. The fact that A. sarcoides was isolated from disks of log # 42 at both sampling times but wood-decaying fungi were not cultured, supports Etheridge's observations (1970). A connection between A. sarcoides and discolorations has been reported by Whittaker (1962b), who associated it with redheart in lodgepole pine. A. sarcoides has now been cultured from brownstained western hemlock in this study specifically regions 40-3 and 44-3. Dark-pigmented fungi were isolated from western hemlock both at the initial sampling and after two months storage: Phialophora sp., Leptodent ium elatius (Mangenot) de Hoog var. elatius de Hoog, Leptocrraphium sp. and Trichocladium canadense Hughes all tentatively identified by Dr. Seifert. The isolates of dark pigmented fungi were cultured from the disks of log #44 which also showed signs of decay. Dark pigmented fungi may have infected this 92 tree as secondary invaders following the establishment of decay organisms. Phialophora sp. and L. elatius have also been reported in hem-fir lumber by Chung and Smith (1986) and Seifert and Grylls (1991) but Leptocrraphium sp. and T. canadense were not isolated in either previous investigation. In the current study one 0. piceae isolate was purified from a freshly felled western hemlock tree. This fungus is not a common invader of standing western hemlock trees yet 0. piceae has been reported to be the most frequent fungus from western hemlock lumber in B.C. (Seifert and Grylls, 1991). In the current study dark-pigmented fungi were commonly associated with brownstain regions but a larger sample population would be required to show a definite relationship. Miller et al. (1983) reported that Graphium sp. and Leptographium sp. were the most common fungi in brownstained Douglas-fir lumber. T. canadense was the fungus most frequently isolated from red heart in white birch (Betula papyrifera Marsh.) (Basham and Taylor, 1965) yet its capacity to produce red heart in vitro has not been demonstrated (Fritz, 1931; Siegle, 1967). In the current study strong evidence is provided that the fungi cultured from freshly felled western hemlock trees represent the resident microflora. Based on available literature it appears highly unlikely that fungi would have invaded and penetrated the log ends to a depth of approximately 3 0-40 cm from the butt end 93 within 3-4 weeks after felling the trees. For instance a radial growth rate of 4.22 mm/day has been reported for Ophiostoma sp. in pine (Breuil et al.,1988) and 4.5 mm/day for 0. piliferum in pine (Gibbs, 1993) under laboratory conditions. In the cooler conditions of forest storage lower growth rates would be anticipated. Fungi isolated in this study did not match these reported growth rates even on malt agar media at 25°C where growth should be more rapid than in wood. Although more pigmented fungi were isolated after 2 months of log storage than at the initial sampling, this was attributed to an improved isolation technique and not to fungal invasion following felling of trees. Therefore the evidence indicates that all fungi isolated after 2 months of outdoor storage in this study represented a resident microflora in mature western hemlock trees at the time of felling. Quantification of bacterial colony-forming units showed no distinct difference between discoloured and stain-free wood following dry-land log storage for two months (Table 5) . Several studies have presented evidence for involvement of bacteria in discolorations in standing trees (Cosenza et al. , 1970, Shigo, 1971), in freshly cut logs (Zimmermannn, 1974; Bauch et al., 1985) and in discoloured lumber cut from water-stored logs (Hedley and Meder, 1992). However, wood samples were stored in a refrigerator for 3-4 weeks prior to undertaking this experiment and this may have decreased viability of the colony-forming units of bacteria. Also, obligate anaerobic bacteria would not be detected using the 94 TABLE 5: MOISTURE CONTENT (MC %) AND COLONY-FORMING UNITS (CFU) IN BROWNSTAINED (BS) AND NON-STAINED (NS) LOGS. LOG # 1-1 1-2 1-3 1-4 16-1 16-2 16-3 16-4 40-1 40-2 40-3 40-4 42-1 42-2 42-3 42-4 44-1 44-2 44-3 44-4 48-1 48-2 48-3 48-4 SAMPLE REGION BS NS BS NS NS NS NS NS BS NS BS NS NS NS NS NS BS NS BS NS NS NS NS NS MC % 104 95 88 85 55 60 137 119 118 144 85 85 43 65 41 61 48 73 58 54 71 87 81 63 CFU1 52 4 2 4-5 4 3-4 4 3 3 4-5 3 3 n.d.3 5 n.d. 2 24 l4 34 4 2 3 5 1-2 1 = a dilution 0.1 mL in 1 mL was performed and 0.1 mL was plated. 2 = higher number represents more bacteria in sap e.g. 5 = more than 105 CFU/mL. 3 = MC of sample was too low to collect sufficient sap. 4 = a dilution 0.1 mL in 2 mL was performed but 0.1 mL was accidently plated. This could have underestimated CFU counts. procedure employed in this study. 95 4.3.5 Sap analysis Generally sap pressed from discoloured regions was distinctively brown, while it was clear or pale yellow when collected from non-stained areas. It was also observed that sap from brownstained regions foamed following pressing perhaps indicating gaseous compounds. Foaming was also noticed in sap pressed from brownstained areas of logs after two months. This corresponded to previous observations (Kreber, 1993) suggesting products of microbial origin, possibly associated with micro-anaerobic conditions. Generally, the pH of sap did not appear to influence brownstain production. The pH of discoloured sap was slightly lower than that of non-stained sap (Table 6). However, the difference in pH may have reflected a variation within the disks. For instance, sap sample 1-BS was sampled from inner heartwood while 1-NS was collected nearby in another inner heartwood region. After two months of log storage a slightly lower pH in stained as compared to non-stained sap was again observed but pH was higher in some disks. Post-harvest changes, such as the production of acetic acid, have been documented and can occur during log storage as a result of biodeterioration (Packman, 1960) . 96 TABLE 6: MOISTURE CONTENT (MC %) AND pH IN BROWNSTAINED AND NON-STAINED SAMPLES. SAMPLE REGION 1-BS1 1-NS2 16-NS 16-NS 40-BS 40-NS 42-NS 4 2-NS 44-BS 44-NS 4 8-NS 4 8-NS INITIAL MC % 94 92 104 114 118 116 48 39 78 76 80 61 INITIAL pH 5.0 5.5 5.3 5.4 5.5 5.7 6.0 5.7 5.3 5.6 5.2 5.9 pH AFTER 2 MONTHS 4.9 5.1 5.4 5.5 5.3 5.6 5.8 ND3 5.5 5.4 5.1 5.3 1 = refers to regions with brownstain 2 = refers to regions without brownstain. 3 = pH was not determined 97 Qualitative HPLC analysis (data not shown) of the gross phenolic composition showed differences between discoloured and non-discoloured sap. Discoloured sap contained more low molecular weight compounds and more oxidized compounds than non-discoloured sap. This observation confirmed previous results seen when producing in vitro brownstain in inoculated western hemlock sap (see chapter 3) and may explain the tendency to foaming noted in expressing the discoloured sap. Quantification of catechin and epicatechin by HPLC showed small amounts (0-10ppm) of these compounds in both discoloured and non-stained wood. Catechin and epicatechin were not detected in all the sap samples and there was no clear presence/absence pattern in discoloured and non-discoloured sap relative to hemlock brownstain. This suggests that either other phenols were also involved in the production of brownstain or that catechin and epicatechin can cause colorations at very low concentrations. HPLC analysis of sap pressed after two months of log storage produced similar results. Total soluble phenol (TSP) content was determined after establishing a standard with gallic acid which produced a linear relationship between 5ppm and 80 ppm (Figure 16) . In western hemlock sap diluted 25 times, the TSP content differed between disks and within stained and non-stained regions of disks. Generally, TSP content was lower in sap collected from brownstained regions of both fresh disks and two month-old disks (Figure 17). 98 Figure 16: Determination of a standard calibration using gallic acid. 99 Figure 17: Total soluble phenols measured in brownstained regions of test logs after 0 (A) and 2 (B) months of storage. 100 The lower mean TSP content associated with brownstained areas was significantly different (95% confidence level) from the higher mean TSP content associated with non-stained regions. There was no significant difference at the 95 % level when comparing the mean TSP content of the two non-stained regions associated with control logs (Figure 18) nor when comparing the mean TSP content of the non-stained regions associated with control logs to the non-stained regions associated with the test (brownstained) logs. Similar observations were made when determining the total soluble phenols content in the 4 regions of disks (e.g. 1-1; 1-2; 1-3; 1-4) prepared for microbial isolation attempts after 2 months of log storage (Figure 19). The lower TSP in sap from brownstained regions corresponded with the qualitative HPLC analysis of the gross phenolic composition which also demonstrated less phenolics in these regions. Thus, qualitative HPLC and TSP analysis supported observations made previously (see chapter 3) that phenolic compounds are more oxidized in regions associated with brownstain. It may well be that soluble phenolic compounds oxidize to quinones when producing brownstain. Attempts to reduce possible quinones with sodium borohydrate failed because the reducing agent interfered with the Folin-Ciolteau reagent. 101Figure 18: Total soluble phenols measure in two non-stained regionsof control logs after 0 (A) and two (B) months of storage.70•REGION-i REGION-2605040-JCDD 3020100102 Figure 19: Total soluble phenols measured in four different regions in test and control logs after 2 months of storage (the 4 regions were sampled as shown in Figure 12). 4.4 Conclusions 103 Generally, brownstain was not observed at the butt cut of the freshly felled trees nor was brownstain noticed at the cross-cuts of the same trees beyond two months of outdoor storage. Most log endcuts showed brownstain after 9 months of log storage demonstrating that storage of logs beyond two months promoted brownstain. Salt water-stored logs were more discoloured than the land-stored logs. Thus storage conditions affect production of brownstain. In sawn lumber brownstain disfiguring the wood surface was related to the brownstain regions discolouring the log end-cuts. Sapwood and/or the sapwood-heartwood transition zone were more prone to brownstain than heartwood. Brownstain was more prevalent in regions associated with the butt ends. Other factors which may promote the production of brownstain are inhibition of moisture movement, the presence of oxygen and photooxidation. Fungi isolated from western hemlock logs represented the resident microflora from the living tree. A. sarcoides was the most frequent isolate from both freshly-felled western hemlock trees and hemlock logs following two months of log storage. 104 Dark-pigmented fungi were the predominant type isolated from brownstained regions whereas only one dark-pigmented fungus was isolated from a non-discoloured area. The specially developed method used in this study to quantify aerobic bacteria in wood produced consistent and repeatable results. Finally total soluble phenols were lower in brownstain regions suggesting the presence of more oxidized compounds. 105 5.0 MICROFLORA AND TOTAL SOLUBLE PHENOLS ASSOCIATED WITH BROWNSTAIN IN WESTERN HEMLOCK LUMBER. 5.1 Objective To understand the relationship between fungi and the total soluble phenols in brownstain. 5.2 Materials and Methods 5.2.1 Sampling at CIPA sawmill On February 22, 94 CIPA's (CIPA Lumber Co. LTD.) sawmill in Nanaimo, B.C. was visited to examine the hemlock brownstain situation in stored wood awaiting offshore shipment and to sample western hemlock 5 cm x 10 cm material showing hemlock brownstain. Fifty hem-fir boards were selected from one package and trimmed to 1 metre length for laboratory studies. According to CIPA's records the lumber was produced and dip-treated on December 1, 1993 with the fungicide NP-1 (containing didecyldimethylammonium chloride + 3-iodo-2-propynyl-butylcarbamate) to control mould fungi and staining fungi. The 50 (1 metre long) boards arrived at the Forintek laboratory on February 25, 1994. The lumber was stored inside overnight prior to sorting by wood species (amabilis fir vs western hemlock) microscopically. 106 5.2.2 Sample preparation and isolation of fungi A 2.5 cm sample was trimmed off the brownstained end of all boards, followed by a 10 cm sample, a 2.5 cm sample, a 20 cm sample and another 2.5 cm sample (Figure 20) . The 2.5 cm samples coded A (end piece), B (centre piece), C (inner piece), were placed in plastic bags and refrigerated for up to 48 hours. To isolate fungi small wood chips (10-15 mm long x 2-5 mm wide x 1-3 mm thick) were prepared from all three positions, A, B and C within a board and also from three different positions across the width of a sample (Figure 20). Wood chips were then placed on glass rods in petri dishes overlaying cotton sheet wetted to maintain a high humidity for 2-3 hours. Five chips were produced from each position, a total of fifteen from each board. The wood chips were then surface flame-sterilized and aseptically placed onto MA-T (2% malt extract and 1.5% agar plus 100 ppm tetracycline). Petri dishes were sealed in plastic bags and incubated in a chamber at 25°C in the dark. Incubated wood chips were frequently checked to enable subculturing of fungi growing out of the wood into the media. Subcultures were purified on malt agar prior to transferring them to slant malt agar tubes which were stored at 4°C in the dark. 107Figure 20: Flowchart showing sampling regime of 5 cm x 10 cmwestern hemlock lumber (region A represents the end of a board)20cm 2.5cm 10cmX\\\TOTAL SOLUBLEPHENOLS IN SAP ISOLATION OF FUNGI108 5.2.3 Determination of TSP content Following preparation of wood chips the remainder of each 2.5 cm sample used for fungal isolation was individually sealed in a plastic bag and placed in a refrigerator prior to pressing of sap. Sap was collected using a laboratory press as previously described (see and the pressate was immediately placed in a freezer until analysis. Total soluble phenols was determined using the Folin-Ciocalteau method as described (see 5.2.4 Assessment of antisapstain treatment To verify that the western hemlock boards had been treated with NP-1 four boards (# 16, 28, 46, 49) were selected at random. Five wood chips with a 6.25 cm2 surface area were removed from different positions on these boards. The chips were analyzed for the major fungicidal active ingredient, didecyldimethylammonium chloride (DDAC) according to the HPLC method developed by Daniels (1992) and the surface retention levels were determined. 5.3 Results and Discussions 109 5.3.1 Observations on brownstained lumber Examination of hem-fir lumber at CIPA demonstrated that many packages contained boards with brown or black discolorations at the cross-cut ends. The amount and intensity of the discoloration was variable within and between packages. The boards sampled were sawn on December 1, 1993 and visibly showed the presence of brownstain. This observation demonstrated that hemlock brownstain can develop in the winter season and does not depend on summer conditions. Brown discolorations have been observed to develop in spruce and fir lumber both in the cool season of November and the hotter months of July and August (Rathke, 1991) . In another study Miller et al. (1983) have reported on brown discoloration in Douglas-fir occurring during the winter of 1980/81. It is also notable that a severe incidence of hemlock brownstain was reported in February 1993 from Mayo, Nanaimo, B.C., a sawmill mill close to CIPA (Kreber, 1993). Upon careful examination of the cross-cut ends green mould fungi, particularly Penicillium sp., were often seen in discoloured regions. In some cases dark green to black pigmented fungi were also observed. In contrast, fungi were not visible when the cross-cut ends were free of brownstain. This observation, which was 110 consistently seen in different hem-fir packages, indicated a relationship between brownstain and fungi on the cross-cut ends. 5.3.2 Isolation of fungi A total of 570 isolation attempts were performed on 38 western hemlock boards. The 12 true fir boards were not sampled. A total of 290 isolates were successfully cultured from 36 different boards. Two boards, 36 and 47, yielded no fungi. Fungi producing a red-pigmentation on malt agar media were the most frequently isolated (total of 104) followed by 70 yeasts and 53 green moulds (Figure 21). Assuming it was the same fungus, the red pigmented fungus, tentatively identified as Ascocorvne sarcoides by Dr. K. Seifert, was isolated from 18 boards and it was present in all three positions (A, B and C) in 10 boards. Interestingly, boards 16, 25, 28, which showed no signs of brownstain at the endcuts, yielded A. sarcoides exclusively from all three regions within the board. The high isolation frequency of A. sarcoides from western hemlock lumber was surprising. While this fungus has been reported to be the most frequent non-decay fungus in living western hemlock trees (Etheridge, 1970) it was not isolated from western hemlock lumber in a recent mill survey in B.C. (Seifert and Grylls, 1991). Also Chung and Smith (1986) have only infrequently isolated Coryne sp. from western hemlock lumber, the anamorph genus of the teleomorph 111OthersDark p1gm. funpiWhite myceIiu7”Green mouldsA. sarcoidesYeastsFigure 21: Frequency of fungi isolated from western hemlock lumber.Ascocoryne sp.. 112 Green moulds, mainly Penicillium sp., were also cultured from 18 boards yet in 3 boards green moulds were the sole fungi isolated from all three positions. This observation demonstrated that moulds deeply penetrated western hemlock lumber and thus agreed with another study (Spradling, 1936) where Trichoderma liqnorum (Tode) Harz has been shown to rapidly grow throughout the sapwood of unseasoned southern pine (P. taeda L.). Furthermore, the high frequency of green moulds in western hemlock lumber observed in this study corresponded with the findings of Seifert and Grylls (1991). In the current study, only 14 isolates of dark pigmented fungi, possibly Ophiostoma sp. , were found. This differs from the investigations of Seifert and Grylls (1991) and Chung and Smith (1986) in which 0. piceae was the most frequent isolate from western hemlock lumber. However, Chung and Smith (1986), surveying for sapstaining fungi on hem-fir lumber in transit, isolated from the discoloured wood surface only. Also Seifert and Grylls (1991) did not clarify in their report whether the high incidence of 0. piceae was drawn from the prolific presence of asexual and sexual fruiting structures, which they have observed at the wood surface, or whether this fungus was isolated from the subsurface of western hemlock lumber. In the current study, there was no evidence that dark pigmented fungi, for example 0. piceae, predominated in 113 microflora. However, it is well possible that some of the whitish fungi isolated from western hemlock were anamorphs of Ophiostoma sp. , for example Sporothrix sp. as it has been reported from western hemlock (Chung and Smith, 1986). The DDAC retention levels of the NP-1 fungicide on the boards sampled were at the lower end of the range of the target retention level of 90 jig DDAC/cm2. Nevertheless it would seem unlikely that such an abundant fungal invasion had occurred in the cold winter season thereby infecting the boards during the 10 weeks since sawing. Therefore, the majority of fungi isolated from the western hemlock were believed to represent a microflora which were already established at the time of sawing the lumber. This conclusion confirmed results obtained in this study showing an endemic microflora in freshly felled western hemlock logs (see section 4.4). Also in another study (Clark, 1994), prolific fungal infection was demonstrated in second-growth western hemlock logs after 5 months of forest storage. The fact that few fungi were isolated from the freshly cut logs (Clark, 1994) provides further support that fungal infections can occur during log storage. 5.3.3 Sap analysis TSP content was variable both within and between boards (Figure 22; 23). In general, boards with brown endstain contained less TSP's while boards without brown endstain (16, 25, 28, 29) demonstrated 114 140 120 100 80 CD ^ 60 40 20 0 • A ; ' B I C I—i—h • • - • II iillli.illllli n lllllll Illi,, I 1 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 19 21 23 24 BOARD # Figure 22: Total soluble phenols measured in three different regions within a board (board # 16 showed no brownstain). 115 120 100 80 (D 60 40 20 0 • ---• ---1+1 • A O B B c 1 + [ l l | l l | l l | l l | l l | M M ' i •III r IJI 25 26 28 29 31 32 33 34 36 37 38 40 41 43 46 48 49 50 BOARD # Figure 23 : Total soluble phenols measured in three different regions within a board (boards # 25, 28 and 2 9 showed no brownstain). 116 a higher TSP content (Figure 22; 23). The average TSP content of the boards with brownstain was significantly (p-value = 0.006) different from the average TSP content of the non-stained boards when performing a Mann-Whitney test. In cases (e.g., #23) where boards with brown endstain produced high amounts of TSP the discoloration was very light and the discoloured region was small. In another instance a very high TSP content in board #3, which showed brown endstain, was related to bark residue on the sample. Inner bark of western hemlock is particularly rich in phenols (Hergert, 1960) . It was also noticed that the TSP content decreased from the cross-cut ends to the inner sample region in 21 boards. This suggested movement of soluble phenols along the grain to the cross-cut ends, which would be expected in close-piled lumber drying from the cut ends. Anderson et al. (196 0) have demonstrated migration of water-soluble extractives, tannins and polyphenols, from the inner core to the outside of redwood (Sequoia sempervirens (D. Don) Endl.) boards during kiln-drying. The authors have also stated that these water-soluble tannins and polyphenols were largely responsible for the occurrence of discolorations found in redwood. 5.4 Conclusions 117 Except for two boards all western hemlock boards yielded fungi with A. sarcoides being the most frequently isolated fungus. The uniformity of fungal isolates from western hemlock lumber which had received anti-sapstain treatment, suggested an endemic microflora which had infected the wood prior to sawing the lumber. A spatial relationship between the presence of fungi and brownstain was found on the cross-end cuts of western hemlock lumber. Total soluble phenol contents were lower in boards with brownstain suggesting that phenol oxidation had occurred. 118 6.0 IN VITRO PRODUCTION OF HEMLOCK BROWNSTAIN 6.1 Immunogold-silver staining of 0. piceae when grown in western hemlock 6.1.1 Objective To determine whether there is a link between the presence of 0. piceae and the production of brownstain in western hemlock. 6.1.2 Materials and Methods Sample preparation A package of 10 cm x 10 cm hem-fir lumber was provided by CIPA, Nanaimo, BC. Boards with brown endstain were selected and ripped into two halves. The faces were planed to remove previous antisapstain treatments. A 30 cm segment was trimmed off the brownstained end of each board and kept for later reference. A 2.5 cm sample was then cut off the same cross-cut end, followed by a 10 cm, a 2.5 cm and a 12.5 cm piece (Figure 24). The 2.5 cm pieces were stored in a freezer until pressate was collected. A total of 3 5 (12.5 and 10 cm) pieces were prepared. Fourteen 10 cm pieces were selected at random and frozen to serve as controls. The freezing of control samples was undertaken to control post-mortem changes and growth of resident microflora naturally present in 119 1 2" X 4" WESTERN HEMLOCK I 30cm I 12.5 cm J. O.PICEAE#387I INFECTED 1 ,T • 2.5 1 1 cm | ' i 1 2. CE 5 a 10 cm i 2.5 cm W T NON-INFECTED - • X MICROSCOPY LABELING PRESSATE PRESSATE * TSP * pH * HPLC 1 1 1 2.5 1 | c m | Jf' 1 Figure 24: Flowchart showing experimental design. 120 wood. The remaining 10 cm samples and 12.5 cm samples were used as described below. Preparation of fungal inoculum O. piceae isolate # 3871 was cultured on five malt agar plates (2% malt extract, 2% agar) for three weeks. A suspension of spores and mycelial fragments was prepared by slightly scraping the fungus off the malt agar media plates and blending the scrapings in 250 mL of sterile, distilled water. The inoculum suspension was then added, while agitating, to 2 litres of sterile, distilled water. Infection of western hemlock The 12.5 cm small pieces previously described (see, were individually immersed in the 0. piceae # 3871 suspension for about five seconds. The inoculated samples were placed in a closed container on a 2 mm polypropylene mesh underlain with water-wetted filter paper to maintain a high humidity during incubation. Between each layer of infected samples a similar 2 mm mesh was placed as spacing. Each container was sealed with a lid and incubated at 20°C and RH 68% for six weeks in the dark. Controls (10 cm samples) were immersed in distilled water and then incubated in a similar container at 4°C and 88% RH as described above. 121 Visual and microscopic examination of wood samples After 6 weeks of incubation visual observations were made on both infected and control pieces to determine fungal colonization at the wood surface. A 2.5 cm sample was then cut from the centre of all infected 12.5 cm samples and 10 cm control samples (incubated at 4°C and stored in a freezer) and used as detailed below. Half of the remaining pieces from all 10 and 12.5 cm samples were oven-dried (103°C, 12 hours) to encourage production of brownstain while the other half were stored at room temperature. The heated samples were visually examined for brownstain on the cross-cut ends and faces. Small wood cubes were cut from the centre of the 2.5 cm pieces and 15 jLtm thick, radial sections were prepared for light microscopy. The remaining portion of the 2.5 cm pieces were used to press sap. Both the collected sap and the small wood cubes were frozen until use. Immunolabeling of 0. piceae # 3 871 in western hemlock A monoclonal antibody (1F3) raised against O. piceae isolate # 3871 (Banerjee et al, 1994) was provided by Dr. D.L. Brown, University of Ottawa, Canada. An enzyme-linked immunosorbent assay (ELISA) was performed in artificial media (data not reported) to determine the reactivity of 1F3 against the 0. piceae #3871 strain used in 122 this study following the procedure described by Banerjee et al. (1994) . Radial sections (15/xm) were cut from infected and uninfected wood cubes and stored overnight in distilled water in a refrigerator. To fix the wood all sections were then incubated for 3 0 minutes at room temperature with gluteraldehyde and Triton X-100 (0.5% gluteraldehyde and 0.2% Triton X-100) in lOOmM phosphate buffer (PB) in a well of a 24-well plate, using 1 mL/well. Sections were then washed 2x5 minutes in 0.5 mL of PB, followed by a 20 minute incubation with 1 mL/well of Triton X-100 (0.5%). Sections were washed 3x3 minutes in PB and 1x5 minute in phosphate buffer saline (PBS) before incubation in glycine (50 mM in PBS) for 3 0 minutes using 1 mL/well. After washing sections in PBS for 2x5 minutes, blocking buffer (washing buffer with 5% goat serum) 0.6 mL/well was applied to the sections for 10 minutes, followed by washing buffer (0.8% BSA, 0.1% gelatin, 2mM NaN3 in PBS) for 5 minutes (1 mL/well). The primary antibody (1F3, diluted 500x in incubation buffer) was applied to sections for 2 hours using 0.5 mL/well, except for the controls which were treated in incubation buffer (washing buffer with 1% goat serum) only. The sections were then passed through washing buffer (3x10 minutes) followed by incubation with biotinyl GAM-IgG diluted 200x in 0.35mL/well of incubation buffer for 60 minutes. Sections were then treated (3x15 minutes) in washing buffer prior to incubation with AuroProbe-1-streptavidin (0.3 mL/well) diluted 50x overnight at 4°C. Washing buffer (3x15 minutes) was then applied to the sections followed by 3x5 minutes treatment with PBS and a 2x5 minutes rinse in filter-123 sterilized (0.2 /xm) , distilled water. Silver amplification with IntenseSE™ M was then performed on wood sections for 8 minutes in the dark using 100 /xL/well while agitating at 200 RPM. Washes (3x5 minutes) were employed using filter-sterilized water prior to mounting sections in 66% Aquamount. The sections were then examined with a phase-contrast microscope. Sap analysis TSP content was determined in sap collected prior to infection of western hemlock and in sap pressed from either 0. piceae infected, or non-infected, western hemlock following 6 weeks of incubation. The Folin-Ciocalteu method was used as described previously (see section . Qualitative HPLC analysis was also performed to determine the gross phenolic composition according to the procedure described by Daniels (1993b). 124 6.1.3 Results and Discussion Western hemlock inoculated with 0. piceae showed mycelial growth typical of Ophiostoma sp. and/or their anamorphs on the cross-cut ends after 6 weeks of incubation. Mycelium and black reproductive structures were also noticed but were less prolific on the faces of the pieces. Non-infected controls stored at 4°C, also showed similar fungal colonization with reproductive structures typical of Ophiostoma sp. and their anamorphs. Fungal isolation attempts from 3 different controls yielded Ophiostoma-like (unidentified) fungi. It seemed likely that these fungi were resident in the western hemlock lumber used for this study. Heating infected samples produced dark colorations at the ends whereas non-infected samples showed the chestnut-brown coloration commonly found when brownstain is produced by heating the wood. The dark coloration was either observed across the whole cross section of infected samples or was associated with certain groups of growth rings. Stereo microscopy of the cross-cut ends of infected samples showed hyaline hyphae and black reproductive structures on the surface thus accounting for the dark coloration on that surface. A dark-grey coloration penetrated the wood 5 mm axially. Pigmented hyphae were not found in these discoloured areas but macroscopically visible brownstain was microscopically associated with non-pigmented hyphae. 125 Observation under a microscope revealed a few brown globules within ray parenchyma cells of western hemlock at the time of starting this experiment. After incubation a second type of brownstain was noticed in parenchyma cells and more rarely in tracheids. This brownstain pattern showed small rounded particles ranging from yellow to brown in colour which appeared to condense in advanced stages of brownstain to form larger brown deposits (globules) within the cells. In advanced stages the whole cell lumen was seen to be occupied with brownstained deposits. However, macroscopic brownstain was limited to the edges of the boards but, interestingly, was almost exclusively observed when hyaline fungi were present microscopically. Similar observations were previously made in the current study (see Chapter 3.0) and also in another investigation (Smith and Spence, 1987). Why 0. piceae, a sapstaining fungus produced a black pigmentation on malt agar medium, but failed to produce pigment in western hemlock wood is unknown; nonetheless the fungus appeared to promote brownstain. Even diluted x 104 the monoclonal antibody 1F3 detected 0. piceae in an ELISA, demonstrating high sensitivity of 1F3 to this fungus. However, in western hemlock 1F3 immunolabeling of 0. piceae was less satisfactory. Effective labelling was still not accomplished when extending the time for silver amplification from 8 to 20 minutes as this was initially thought to be the factor affecting this outcome. 126 Two factors which may have caused the failure to labell are: the degree of infection was too low (providing insufficient accessible hyphae) or interference may have occurred in the western hemlock samples due to an inherent microflora as observed by Breuil (1994) in non-sterile wood. In the current study fungi were isolated from non-infected controls suggesting an inherent microflora in the western hemlock used. Another explanation might be that different antigens were expressed on the cell wall of the non-pigmented 0. piceae observed in western hemlock, which the monoclonal antibody was unable to recognize. Interestingly, Banerjee et al. (1994) labelled 0. piceae with 1F3 in gamma irradiated jack pine (P. banksiana Lamb.) sapwood suggesting that the fungus expressed different anigens in the cell wall when growing in jack pine. However, in the current study immunolabeling was also not satisfactory in ultra-thin sections prepared from western hemlock which was thoroughly colonized with 0. piceae (Ghariban, 1994). Thus evidence provided in the current study gave strong indications that the monoclonal antibody, 1F3, lacked specificity against the strain of 0. piceae grown in western hemlock. However, an ELISA might have produced a different outcome on infected ground western hemlock as 1F3 was highly sensitive in the ELISA undertaken to verify its specificity against 0. piceae. TSP analysis of sap showed differences between and within boards (Table 7) . Variability between boards was expected since phenolics differ qualitatively and quantitatively across the cross-section 127 TABLE 7: TOTAL SOLUBLE PHENOL CONTENT (/ig/mL) MEASURED IN SAP FROM WESTERN HEMLOCK INCUBATED WITH 0. PICEAE FOR 6 WEEKS. SAMPLE # 1C 2C 5C 3D 4D ID 5D 3B 5B IB 4B A1 1248 513 479 324 340 111 68 99 145 1622 1017 B2 1735 1099 370 675 271 181 62 120 110 1733 678 Infected3 1855 998 249 172 768 120 147 247 261 1884 890 C4 1581 1038 281 540 240 234 448 120 132 1921 1069 1 = total soluble phenol content was measured in 12.5 cm sample prior to infection and incubation for 6 weeks. 2 = total soluble phenol content was measured in 10 cm sample before incubation for 6 weeks. 3 = total soluble phenols was determined in infected 12.5 cm samples following 6 weeks of incubation. 4 = total soluble phenol content was determined in 10 cm samples following incubation for 6 weeks at 4°C. 128 region within western hemlock (Barton and Gardner, 1966). Nonetheless the large variations observed in the TSP content within boards were surprising. However, no distinct patterns emerged when comparing total soluble phenols in sap from infected samples with sap from non-infected samples and their role in hemlock brownstain was not evident (Table 7). Qualitative HPLC analysis of sap provided similar variability in the gross phenolic composition both within and between boards. As with the TSP analysis a distinct pattern was not noticed when comparing the gross phenolic composition in sap collected prior to infection with sap analyzed after infection and incubation. However, hydroxymatairesinol (OHMR) was found in concentrations up to 10 times lower in some infected samples than in non-infected samples. According to Barton and Gardner (1966) OHMR was predominately associated with the heartwood and less so with western hemlock sapwood. Therefore, brownstain in sapwood may require phenolics other than OHMR. Furthermore, OHMR produced no coloration when spotted on papergrams (Barton and Gardner, 1966) and the authors explained that the fact that OHMR lacked vicinal hydroxyl groups and a stable ring system precluded oxidation by-products (Barton, 1973). This study failed to link changes in TSP and in the phenolic composition with production of brownstain in infected western hemlock, differing from the previous observations (see chapter 3.0) 129 which showed oxidation of the gross phenolic composition following infection with 0. piceae. This may be because some brownstain was already present in the western hemlock at the start of the experiment. Subsequent production of brownstain during incubation with 0. piceae may have been too little to reveal a difference in phenols because soluble phenols already may have been oxidized. 6.1.4 Conclusions 130 O. piceae prolifically colonized the wood surface and produced abundant, black, reproductive structures. The dark coloration observed in the dried samples was attributed to the presence of the black reproductive structures. 0. piceae produced brownstain mainly in the outer regions of the western hemlock where the fungus showed no hyphal pigmentation. The monoclonal antibody 1F3 detected 0. piceae in an ELISA assay but 1F3 was unable to detect the fungus when growing in western hemlock. The total soluble phenols and the gross phenolic composition were not related to the production of brownstain. In further experiments non-infected and stain-free western hemlock should be used to link production of brownstain produced by 0. piceae to changes in the phenols. Use of a polyclonal antibody is recommended to link 0. piceae to brownstain as they generally are considered easier to work with. 131 6.2 Production of brownstain in western hemlock sap 6.2.1 Objective To characterize chemical changes in western hemlock sap incubated with fungi. 6.2.2 Materials and methods Sample preparation Hem-fir (10 cm x 10 cm) showing brown cross-cut ends was sampled at CIPA, Nanaimo, B.C. on June 20, 1994. According to CIPA's record the lumber had been sawn about a week before sampling. The boards were identified microscopically as hemlock and fir and the true firs were discarded. The ten western hemlock boards were ripped in half and a 30 cm piece and a 5 cm piece were cut off one end of each boards. While the 5 cm samples were oven-dried (103°C) overnight to produce brownstain, sap was pressed from each 3 0 cm piece and frozen according to the procedures described (see The remaining western hemlock boards were wrapped and stored in a freezer until use. Sap from two (5A; 8A) boards which showed pronounced brownstain following heating were chosen for the first experiment. Additional sap assays were then undertaken with sap-8A, and also with sap-4A and sap-7D, which similarly originated from boards showing brown endstain after heating. 132 Infection of sap Sap was thawed and filtered (4x) through Whatman # 1. About 100 mL of each sap was then filter-sterilized (0.2 /xm) using a disposable pre-sterilized filter unit (Nalgene Company, Rochester, NY). Approximately 3-4 mL of the filter-sterilized sap was then decanted into a pre-sterilized slant agar tube under aseptical conditions. Sterility of sap samples was verified by placing a few drops on malt agar plates which were then incubated at 2 5°C in the dark. The cultures evaluated for their potential to cause browning of sap (Table 8) , had been previously isolated from western hemlock logs (see chapter 4) and from western hemlock lumber (see chapter 5) . Additional isolates were chosen from the culture collection of Forintek Canada Corp, Vancouver, B.C. (Table 8). A small plug of fungal inoculum was taken from a slant agar culture tube and transferred to a culture tube containing sap. All infected tubes were incubated at room temperature (22 °C) for 12 days. One replicate was used for each fungus. In addition controls receiving no fungal inoculum were used for each sap and incubated as described above. Additional controls were kept in the freezer. Sap analysis A pH-meter was used to determine the pH of sap. TSP and qualitative HPLC analyses were undertaken as described previously (see . Data from infected sap, non-infected sap and sap 133 TABLE 8: FUNGI EVALUATED FOR THEIR IN WESTERN HEMLOCK SAP. ISOLATE 0. piceae 3871 A. sarcoides 12B P. melinii 270B A2S A6S A7S A8S A13S A14S A1B A2B BIS BIBS B15S B16S AIM A3M B1Y C5A1 C13C1 POTENTIAL TO CAUSE BROWNING SOURCE Western hemlock lumber, FCC1 Lodgepole pine lumber, FCC1 White birch log, Western Western Western Western Western Western Western Western Western Western Western Western Western Western Western Western Western hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock hemlock FCC1 log log log log log log log log log log log log log log log lum] lum] Interior2 Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Interfor, Der, CIPA3, Der, CIPA, , Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Chamiss Nanaimo Nanaimo Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay Bay 1 = Forintek Canada Corp., Vancouver, B.C. 2 = International Forest Products Limited, Vancouver, B.C, 3 = CIPA Lumber Co. Ltd., Nanaimo, B.C. 134 kept in the freezer were compared. 135 6.2.3 Results and Discussions The four sap samples used differed in their susceptibility to develop brown colorations. Sap-8A was highly conducive to browning whereas sap-4A and sap-5A demonstrated little colour change upon incubation with microorganisms. No coloration developed in infected sap-7D. With regard to the initial trial with sap-8A, it was noticed that different fungi coloured sap at different intensities and over different incubation times (Figure 25; Table 9). For instance, the first browning was noticed in sap-8A infected with fungus A1B (a basidiomycete) after only two days whereas A13S, tentatively identified as Phialophora melinii, needed 4 days to colour sap-8A. In sap-8A A6S produced brown precipitates in the solution demonstrating that the fungus had modified water soluble components in this sap. As a result supernatant from some saps containing the brown precipitate were clearer than the non-infected controls. The same fungi also caused colour changes in sap-5A, yet at a much lower intensity and over a longer incubation time (Figure 25; Table 10) . For instance fungus A1B caused the first noticeable colour changes in sap-5A following 3 days of incubation. Infected sap-5A also demonstrated no coloured precipitates indicating that coloured compounds were still water-soluble. The controls showed light yellowing following incubation for 12 days at room temperature. 136 •."nil * • « • * * • • • • - - * • jm^ ' " « - I I K . , , / l „ „ , M | K N 'Ml.., K Ml- IS, U l II !,»,,, | »« • • HI M UN..I VI ,,H " Figure 25: Colour changes in western hemlock sap (8A; 5A) incubated with different fungi for 10 days at room temperature. 137 TABLE 9: CHANGES IN WESTERN HEMLOCK SAP (8A) INCUBATED WITH DIFFERENT MICROORGANISMS FOR 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C 0. piceae 3871 A2S A6S A8S A13S A14S AIM A1B A2B BIBS B1Y B16S SAP COLOUR light yellow yellow light brown brown brown yellow brown yellow light yellow brown light brown light brown light brown brown pH 5.1 5.2 7.2 7.3 7.5 6.0 7.5 4.5 5.5 4.8 5.3 6.8 6.1 7.5 TSP1 /xg/mL 594 353 255 200 167 399 172 424 355 0 372 207 198 206 OHMR2 /xg/mL 60 91 43 19 0 85 0 86 77 0 65 10 72 8 1 = refers to total soluble phenol content 2 = refers to hydroxymatairesinol TABLE 10: CHANGES IN WESTERN HEMLOCK SAP (5A) INCUBATED WITH DIFFERENT MICROORGANISMS FOR 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C 0. piceae 3871 A2S A6S A8S A13S A14S AIM A1B A2B BIBS B1Y B16S SAP COLOUR colourless light yellow light yellow light yellow light yellow colourless light yellow colourless light yellow yellow colourless yellow yellow yellow pH 4.8 4.9 7.5 7.2 7.5 6.2 7.5 5.2 6.6 4.9 4.9 6.4 7.0 7.6 TSP //g/mL 85 45 64 58 0 93 0 119 54 0 105 14 71 18 OHMR /xg/mL 25 34 12 10 0 28 0 35 20 0 34 0 11 0 139 The assays with sap-4A and sap-7D affirmed that saps from different boards differed in their susceptibility to undergo colour changes upon infection with fungi. Sap-4A developed a yellow coloration with some fungi, similar to those colour changes produced by the same fungi noted in sap-5A (Table 11). Colorations did not occur in sap-7D upon incubation with fungi (Table 12). Additional assays with sap-8A, undertaken to verify the initial observations, confirmed that this sap was very susceptible to browning. It also affirmed that fungi varied in their potential to cause sap browning (Table 13). Thus fungi with a high potential, for example A6S, consistently produced brown colorations in sap-8A. Most fungi shifted the pH to neutral or slightly alkaline when cultured in the four different sap samples but only sap-8A underwent browning. The fungi which caused intense brown coloration of sap-8A shifted the pH to alkaline (Table 9, 13) . Interestingly, the fungi which were evaluated for their browning potential, produced similar pH changes in sap-4A, sap-5A and sap-8A but less in sap-7D. For example, the fungus B16S shifted the pH in sap-8A from 5.1 to 7.5, in sap-5A from pH 4.8 to 7.6 and in sap-4A from pH 4.7 to 7.6 (Table 9-11). Two of the three basidiomycetes tested in sap-8A produced no pH shift yet browning occurred with A1B. The third basidiomycete (C5A1) evaluated, decreased the pH to 3.4 which was accompanied by a colour change to yellow/brown. 1 4 0 TABLE 11: CHANGES IN WESTERN HEMLOCK SAP (4A) INCUBATED WITH DIFFERENT MICROORGANISMS FOR 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C 0. oiceae #3871 A2S A6S A13S B16S BIBS A. sarcoides #12B BIS P. melinii # 270B B15S SAP COLOUR colourless light yellow light yellow light yellow yellow yellow yellow yellow colourless colourless colourless yellow pH 4.7 4.9 7.3 7.6 7.7 7.8 7.6 5.1 5.8 7.6 5.0 7.6 TSP /xg/mL 96 47 10 0 0 0 0 0 8 0 49 0 OHMR xxg/mL 6 0 0 0 0 0 0 0 0 0 0 0 TABLE 12: CHANGES IN WESTERN HEMLOCK SAP (7D) INCUBATED WITH DIFFERENT MICROORGANISMS AFTER 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C 0. piceae #3871 A2S A6S A13S B16S B15S P. melinii 270B BIS SAP COLOUR colourless colourless colourless colourless colourless colourless colourless colourless colourless colourless PH 5.3 5.2 6.7 6.4 7.2 6.7 6.8 6.8 6.4 6.6 TSP /xg/mL 309 366 360 340 124 126 267 258 315 297 OHMR /xg/mL 113 239 74 65 0 0 36 36 45 36 1 4 1 TABLE 13: REPEATED ASSESSMENT OF CHANGES IN WESTERN HEMLOCK SAP (8A) INCUBATED WITH DIFFERENT MICROORGANISMS AFTER 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C A3M A7S C5A1 C13C1 0. piceae 3871 A2S A6S A13S B16S BIB A. sarcoides 12B BIS P. melinii 270B l_ B15S SAP COLOUR colourless yellow yellow yellow ye11ow/brown yellow brown brown brown brown brown yellow/brown yellow brown light brown brown pH 5.0 5.0 5.0 6.3 3.4 6.5 7.3 7.4 7.8 ND1 7.3 4.7 5.0 7.0 7.3 7.7 TSP /xg/mL 537 237 253 130 5 121 69 17 0 0 31 68 376 115 38 22 OHMR ptg/mL ND1 18 12 9 0 11 0 0 0 0 0 0 8 0 0 0 1 = not determined. 142 Determination of the TSP content showed that sap-5A and sap-4A contained about 6 times less soluble phenols than sap-8A, and sap-7D contained about one-half the TSP amount of sap-8A. Generally, infected samples demonstrated a pronounced decrease in the TSP content as colorations developed. For example, fungi A6S and A13S reduced the TSP content in sap-8A by more than half (Table 9) . Both of these isolates have been tentatively identified as Phialophora sp. (Table 4). In another trial with sap-8A the same fungi modified all the soluble phenols and they were subsequently not detectable (Table 13) . Both isolates caused pronounced browning in sap-8A. A relationship between coloration and decrease in TSP and pH was thus evident. Similar relationships between coloration, pH and TSP content were also seen in infected sap-4A and sap-5A. For instance A6S and A13S produced an alkaline pH shift and modified all the TSP present in sap-4A and sap-5A and this caused some yellow coloration (Tables 9, 10). A relationship between the pH and the decrease in TSP was also observed in sap-7D when infected with A6S and A13S, but colorations were not noticed (Table 12). However, the other fungi shifted the pH in sap-7D but little changes were recorded in the TSP content. Interestingly, an approximate 50% decrease in TSP content was observed in the incubated controls (non-infected samples) except for sap-7D (Table 9-13) . This suggested that auto-oxidation of soluble phenols had accompanied a yellowing in sap-8A and a light yellowing in sap-4A and sap-5A. 143 A qualitative difference in the gross phenolic composition was recorded (by HPLC) in the four sap samples. For example, catechin was observed exclusively in sap-8A and epicatechin only in sap-5A whereas OHMR was present in each of the four saps. The sap samples also varied quantitatively regarding their individual phenols, for example in the OHMR content (Tables 9-13) . These qualitative and quantitative differences were expected as the sap originated from different boards. Barton and Gardner (1966) have shown that the phenolic composition varied within and between western hemlock trees. Most fungi modified the gross phenolic composition in the sap samples during incubation, but the magnitude of change varied between fungi. For example the fungus A6S produced the largest change in the gross phenolic composition in all four sap samples evaluated and this corresponded with large changes in TSP (Tables 9-13) . This study has confirmed earlier observations that a non-specific microflora can cause browning of sap (see chapter 3). However, this study provided additional information regarding the changes in TSP which decreased as browning occurred. Similar observations have been made in discoloured maple (Tattar and Rich, 1973). In another study (Siegle, 1967) the decrease in phenolics has been correlated with the increase in the discoloration rate when incubating hot water extract of birch meal with a fungus. Changes 144 in TSP yielding discolorations have also been reported in cricket bat willow wood infected with a bacterium (Wong and Preece, 1978) and in red maple (Shevenell and Shortle, 1986) . Several observations correlate with the findings made in the current study. Tattar et al. (1971) have also reported that the pH shifted from 5.5 in clear maple to 6.4 in discoloured, infected maple. An increase in pH has also been demonstrated in discoloured maple sawdust, infected with bacteria and fungi (Zimmermann, 1974) and in Ilomba infected with bacteria (Bauch et al. , 1985) . In another study (Schmidt and Mehringer, 1989) sap browning was linked to a pH shift caused by bacteria. Interestingly, Schmidt (1986) has linked the pH shifts produced in infected beech sap, to the availability of sugar and nitrogenous compounds in the sap. Browning in infected sap has been prevented by addition of glucose which kept the pH below 7 (Schmidt and Mehringer, 1989). In the current study sugar and nitrogen levels were not determined in the sap specimens. However, the fact that some fungi always caused a pH shift while others, for example, mould fungi, a yeast and the A_;_ sarcoides isolates, produced little or no pH changes suggested that the fungi differed metabolically. Clearly, the fungi grew prolifically in the sap samples irrespective of their potential to shift pH, indicating that the sap contained adequate nutrition. 145 The current study also demonstrated that dark pigmented fungi, for instance 0. piceae and Phialophora sp. shifted the pH to neutral or alkaline which promoted an intense browning in sap-8A. This indicated a relationship between the dark pigmented fungi and the browning in sap, possibly due to the secretion of either enzymes or phenols. For example, secretion of fungal phenoloxidases into the environment may have catalyzed browning of phenolic compounds present in the sap. Rosch et al. (1969) have shown the presence of laccase, an enzyme involved in the production of fungal melanin in some sapstaining fungi grown in liquid culture. On the other hand secretion of fungal phenols into the culture medium has also been reported to form heterogenous melanin (dark-brown pigments), triggered by an alkaline pH shift (Bell and Wheeler, 1986). However, the current study was unable to determine if fungal activity (other than shifting the pH) was required for sap browning. While alkaline conditions alone have been shown to cause browning (Hathway and Seakins, 1957) slightly acidic conditions also produced browning in the current study, for example, in sap-8A upon infection with the basidiomycete AlB and C5A1 (Tables 9 and 13, respectively). This observation suggested involvement of a phenol oxidizing enzyme. However, certain compounds seemed to be required in the sap to develop browning. This was underscored as browning occurred in sap-8A but not in sap-4A and sap-5A despite similar pH shifts and changes in TSP and in the gross composition of phenols. Further 146 support was given to the observation that the phenolic composition of the sap is critical to browning when sap-7D, which had a TSP content three times larger than the TSP of both sap-4A and sap-5A, developed no coloration upon inoculation with A6S and A13S although the TSP decreased. Catechin was detected in sap-8A and not in the other sap samples but it was present in very small amounts. However, catechin was not detectable (HPLC) after incubation of sap-8A suggesting its involvement in brownstain. Two recent HPLC studies (Hrutfiord et al. , 1985; Daniels, 1994) have also shown that the concentration of catechin declined rapidly and was related to browning of western hemlock chips and western hemlock sap, respectively. In the current study, the production of brown precipitates in infected sap-8A which contained catechin, was of particular interest, because the development of hemlock brownstain is believed to involve formation of insoluble brown compounds from water-soluble compounds. Hathway and Seakins (1955; 1957) have shown that catechin was very reactive under neutral and alkaline condition producing brown polymers. Song (1987) also recorded the production of brown precipitates when investigating brownstain in Douglas-fir. In the current study catechin was histochemically demonstrated in brownstained western hemlock (see chapter 3) and HPLC analysis provided additional evidence for the involvement of catechin in sap browning in this study. 147 6.2.4 Conclusions Fungi isolated from western hemlock can promote colour changes in western hemlock sap to produce water-insoluble brown precipitates. An upward shift in pH in infected western hemlock sap accompanied colour changes. The phenolic composition of the sap is critical to browning. Chemical extractives, such as catechin, can be involved in the production of hemlock brownstain. 148 6.3 Infection of wood sections with 0. piceae on a glass-slide 6.3.1 Objective To determine whether 0. piceae can produce brownstain in thin sections of western hemlock. 6.3.2 Materials and Methods Slide preparation Glass slides were sterilized in an autoclave prior to coating with water agar (1.5% bacto agar) under aseptic conditions. One coated slide was then placed on a bent 2 mm glass rod in a pre-sterilized petri dish with filter paper and distilled, sterile water on the bottom. Wood sections used Two wood cubes were prepared from one freshly sawn western hemlock board (4A) containing sapwood, one from the outer region and one from the centre of the board. Five 15 /xm sections were then produced aseptically from each cube using a microtome. Each section was immediately placed on a agar-coated glass slide, covered with a sterile cover slip and placed in a Petri dish. 149 Infection of wood sections Wood sections were infected by placing a small plug of 0. piceae isolate #3871 at the edge of the cover slip. Four sections from each wood cube were infected while a fifth section received no inoculum and served as control. All plates were incubated in the dark at 25°C for 4 weeks. Microscopic examination of sections Wood sections were removed from the coated glass slide and mounted in lactophenol for microscopic examination. After the initial examination sections were removed from the glass slides and washed in distilled water to extract all hyphae attached to the wood surface. The degree of hyphal colonization within the section was microscopically examined. Photographs were taken using a Zeiss optics microscope. 150 6.3.3 Results and Discussions O. piceae showed prolific growth on the coated glass-slides after 4 weeks of incubation. However, fungi grew less prolifically in the wood sections. Interestingly, the hyphae were non-pigmented when associated with the thin sections. A few pigmented structures were noticed on the coated slides and in instances where the fungus grew on the filter paper at the bottom of a petri dish. Brownstain developed in numerous ray parenchyma cells which had been brownstain-free and without hyphae prior to infection (Figure 26) . The degree of brownstaining varied within and between infected sections. Brownstain was more pronounced at the side of the sections where the inoculum was placed. Non-infected samples developed some brownstain following incubation yet to a much lesser extent than the infected samples. The degree of staining observed in sections as thin as used in this study was rather surprising considering that the quantity of colour precursors present in the ray cells of the 15 jum sections must be extremely small. In a previous study, Smith and Spence (1987) produced brownstain in 5 cm x 10 cm western hemlock when incubated with 0. piceae and brownstain was also produced in 0.6 cm thick western hemlock wafers following infection with the same fungus in the current study (see chapter 3). 151 " >t Figure 26: Tsuga heterophylla. Radial section (63x). Brown deposits associated with hyphae of 0. piceae in a 15 jxm section. 152 Brownstain did not develop upon heating of a wood sample from the same board (4A) suggesting that both the fungus and the phenols were essential for production of brownstain in the sections. Browning also was not observed when sap from the same board (4A) was inoculated with 0. piceae 3871, or with any of the other fungi previously determined to promote browning of sap. This observation strongly suggested that the phenols responsible for brownstain were not released from the wood (4A) when pressing sap because they were linked to wood components. Phenols can be bound to proteins (Stafford, 1988) and they can also occur as glucosides as demonstrated in western hemlock (Goldschmid and Hergert, 1961; Barton, 1962) . For example, catechin, which is thought to play an important role in hemlock brownstain and which was not detected in sap-4A (see section 6.2) , may have been present in the thin sections as glucosides. However, the exact nature of the browning induced by 0. piceae is not known yet. It is possible that the fungus, as it colonized western hemlock, released phenols from their sugar moieties, which then accompanied by a pH shift produced brownstain similar to the pH promoted browning observed in inoculated sap (see chapter 6.2). On the other hand 0. piceae may also have secreted substances into the wood cells which then caused the coloration, for example extracellular melanin. Bell and Wheeler (1986) reported that production of extracellular fungal melanin can occur either by releasing phenol oxidase into the environment to oxidize phenolic 153 compounds or by releasing phenols into the environment, where they are auto-oxidized. Bell and Wheeler (1986) also showed that fungi devoid of wall-bound melanin could still produce extracellular brown pigments. In the current light microscopy study a distinct wall-bound melanization was not observed in 0. piceae when associated with brownstain. However, further experimentation would be required to determine whether this fungus can produce an extracellular brown pigment. In the current study, dark pigmented hyphae grew out of the thin sections when placed on malt agar media. This observation demonstrated that this fungus had the enzymes, either induced or constitutive, required for melanin production. This observation also suggested that certain chemical constituents in western hemlock greatly reduced pigmentation of 0. piceae when growing in wood as it was not detectable by light microscopy. 6.3.4 Conclusions 154 A technique has been developed which demonstrated fungal participation in the brownstaining process in wood. Brownstain formation can occur in the presence of very small amounts of chemical precursors. Cell wall pigmentation of 0. piceae is greatly reduced or not apparent by light microscopy when the fungus grows within the cells of western hemlock. 155 6.4 Infection of western hemlock and lodgepole pine with 0. piceae 6.4.1 Objective To determine whether 0. piceae produces pigmented hyphae when growing in western hemlock. 6.4.2 Materials and Methods Wood used A freshly sawn western hemlock board (2.5 cm x 5 cm) with a high proportion of sapwood, which had been stored in the freezer for two weeks, was used. Two 3 0 cm pieces and two 25 cm pieces were produced after thawing. In addition a lodgepole pine bolt which had been stored in a freezer since 1990 was sawn to 30 cm length and four 5 cm x 5 cm pieces and four 2.5 cm x 10 cm pieces were produced from the sapwood and heartwood. The wood samples were planed on all faces prior to infection. Preparation of inoculum and infection of wood 0. piceae # 3871 was cultured on malt agar (2.0% malt extract; 1.5% agar) for two weeks at 25°C in the dark. A suspension with fungal fragments and spores was prepared by adding a small amount of sterile, distilled water to a culture plate and lightly scraping 156 the surface mycelium. This procedure was repeated three times. The suspension prepared from three plates was then blended (2x15 seconds) in approximately 250 mL of sterile, distilled water using a Waring blender. The blended suspension was added to approximately 2500 mL of sterile, distilled water. The wood samples were dipped in the fungal propagule suspension for 5-10 seconds and they were then close-piled in a container with moist filter paper and glass rods on the bottom. Four stacks were assembled in the container which was covered with a lid and placed at 20°C and 68% RH in the dark. After four weeks of incubation the infected samples were visually examined and a 1 cm slice was cut from one end of a lodgepole pine sample and also a western hemlock sample and 15 ptm radial sections were prepared from them for microscopic examination. The infected samples were then incubated for another 8 week period. After a total of 12 weeks the four faces of each sample were planed (1 mm) to remove the fungal surface flora and to disclose the degree of surface coloration. A 1 cm section was then cut from one end and from the centre of a lodgepole pine and a western hemlock board to produce 15 /xm sections. A Leitz photomicroscope was used for examination of the 15 fJLrci specimens. 157 6.4.3 Results and Discussion Incubated lodgepole pine sapwood samples were covered with a thick mycelial mat after three weeks. In contrast lodgepole pine heartwood was free of mycelial growth. Fungal growth was also well established at the wood surface of western hemlock. After 4 weeks of incubation some mold fungi, for instance Penicillium sp., were also growing on the surface of both lodgepole pine and western hemlock pieces. Visual inspection after 8 and 12 weeks demonstrated an unidentified decay fungus colonizing some of the lodgepole pine samples. Lodgepole pine heartwood also showed localized mould (orange) infections on the surface but no mycelium of 0. piceae. The surface of the western hemlock samples was thoroughly colonized with 0. piceae except for a few localized, green mould fungi. Inspection of the 1 cm piece trimmed off a lodgepole pine sample demonstrated that sapstain was established after 4 weeks. The lodgepole pine sample showed the classical blue stain pattern frequently seen in pines. The discoloration of lodgepole pine sapwood was due to abundant, pigmented hyphae growing in ray parenchyma cells and axial tracheids as seen microscopically. After 12 weeks of incubation bluestain was seen on the faces of the planed lodgepole pine sapwood. However, the degree of bluestain varied within and between lodgepole pine pieces. For example, 158 bluestain was more pronounced on the 1 cm cross-end samples whereas much less coloration was noticed on the cross-section of the centre samples. This observation was also confirmed microscopically as very few pigmented hyphae were found in sections cut from the centre of the lodgepole pines. In contrast visual inspection of the 1 cm slice trimmed off western hemlock demonstrated no sapstain after 4 weeks of incubation. However, microscopic examination showed prolific growth of hyaline hyphae within the wood. Pigmented hyphae were only present in the outer (1-5) cells close to the wood surface. Brownstain also developed in parenchyma cells containing hyphae where there was no brownstain and no visible hyphae existed prior to infection. After 12 weeks of incubation visual examination of the planed western hemlock showed that the board surfaces were stain-free except for a brown coloration disfiguring the edges of the surfaces. Microscopic examinations of these stained regions revealed brownstain in ray parenchyma and tracheids and abundant non-pigmented hyphae. Very few pigmented hyphae and a few coremia were associated with brownstain but they seemed to be closely related to the wood surface. The brown coloration penetrated the wood to a depth of 2-5 mm. The western hemlock slice cut from the centre was free of brownstain and hyphae were not observed microscopically. This experiment reconfirmed that 0. piceae can produce brownstain in western hemlock as observed many times during the course of this 159 study. Macroscopic brownstain was concentrated at the edges of the board surfaces and penetrated about 5 mm. Smith and Spence (1987) also demonstrated that 0. piceae can produce brownstain and a deeper penetration of the brownstain was shown by these authors. However, the presence of non-pigmented hyphae associated with brownstain was consistent in both studies. Both brownstain and hyphae were absent in the centre of the sample suggesting a relationship between brownstain and 0. piceae. The current study has also shown that pigmentation of 0. piceae was promoted in lodgepole pine sapwood whereas it was suppressed in western hemlock sapwood. 0. piceae produced pigmentation on the surface of western hemlock wood but not when deeply penetrating the wood. However, at present the factors suppressing pigmentation of 0. piceae in western hemlock have not been elucidated. Wood moisture content may have influenced the staining intensity. According to Lagerberg et al. (1927) maximum pigmentation occurred in pine and spruce at a wood moisture content of 60-80%. Sapstain has been controlled by keeping Scots pine (P. sylvestris L.) above a wood moisture content of 100-120% (Liese and Peek, 1984). In the present study both, western hemlock and lodgepole pine, showed a wood moisture content of 120%. On the other hand wood species has also been reported to influence the degree of fungal pigmentations (Liese and Schmidt, 1961, Smith, 1994). It is well possible that western hemlock extractives inhibited pigmentation of 0. piceae but not its growth. 160 Therefore, evidence from the current study suggested that this fungus was not capable of causing sapstain in western hemlock. The current study also demonstrated that the association between brownstain and 0. piceae occurring in western hemlock does not occur in lodgepole pine. 6.4.4 Conclusions 161 0. piceae developed pigmented hyphae when grown in lodgepole pine but, except for a small amount of surface growth, the same fungus remained hyaline when colonizing western hemlock. Brownstain developed macroscopically and microscopically in western hemlock infected with 0. piceae. A chemical difference must exist between lodgepole pine and western hemlock to account for the fact that 0. piceae develops pigmentation when growing in the former and not in the latter wood species. 162 7.0 ELUCIDATION OF THE MECHANISMS OF SAP BROWNING 7.1 Objective To determine factors involved in the browning of western hemlock sap. 7.2. Materials and Methods 7.2.1 Effect of pH on sap browning Filter-sterilized (0.2/xm) western hemlock sap (8A) was inoculated with different microorganisms according to the procedures described (see . Additional control sap-8A samples adjusted to pH 7 with NaOH and unchanged sap-8A (pH 5) were set up but they received no inoculum. The sap samples were incubated at room temperature (22°C) for 12 days. The changes in colour, pH, TSP and qualitative changes of the gross phenolic composition in inoculated and control sap-8A were compared according to the procedures outlined previously (see 7.2.2 Effect of oxygen on sap browning Four mL of filter-sterilized (0.2 fxm) sap-8A was transferred into a sterilized (121°C, 15 minutes) glass ampoule. Additional sap-8A was adjusted to pH 7 with NaOH prior to filter-sterilization 163 (0.2 nm) and 4 mL was then delivered into additional glass ampoules. The ampoules were each sealed with a cotton plug and stored in a freezer at -2 0°C overnight. Two ampoules, one with pH 7 adjusted sap and one with unchanged (pH 5) sap were degassed by-drawing a vacuum over the frozen sap to remove the oxygen present in the ampoule. The sap was then thawed to release the dissolved oxygen prior to rapid freezing of the sap with liquid nitrogen. A vacuum was drawn again and the total procedure (thawing-freezing-vacuum) was repeated three times. Nitrogen was then introduced into the ampoules prior to sealing them with a Bunsen burner. The other two ampoules, one with pH 7 adjusted sap and one with unchanged (pH 5) sap, were neither degassed nor flushed with nitrogen and they were then sealed. The coloration, TSP and the gross phenolic composition were determined after 12 days of incubation at room (22°C) temperature as described (see 7.2.3 Effect of heat on sap browning Sap-8A was autoclaved at 121°C for 15 minutes, poured into pre-sterilized test tubes using 4 mL per tube and then inoculated with various fungi. Two additional test tubes containing heated sap without inoculum served as controls. After 12 days of incubation the colour, pH and TSP was determined in heated/inoculated sap and in controls according to procedures described previously (see 164 7.2.4 Investigation of sap browning by heat and by pH alteration Sap from boards # 2B, 5B, 5D, 8B, 9A and 9D, which were sampled at CIPA, Nanaimo, B.C. (see, was heated at 121°C for 15 minutes in an autoclave. Additional sap samples from the same 6 boards were adjusted to about pH 7 with NaOH, filter-sterilized and then incubated at room temperature (22°C) for 12 days. The heated and the pH adjusted sap specimens were compared to controls (non-heated and non-pH adjusted) regarding coloration, changes in pH and TSP as described ( 7.2.5 Amendment of water with phenols Distilled water was adjusted to pH 7 with NaOH prior to making up solutions containing approximately 500 |iig/mL of catechin, epicatechin, hydroxymatairesinol and alfa-conidendrin, respectively. In addition a single solution was prepared mixing together approximately 150 /xg/mL of each compound. The solutions were placed in an ultrasonic bath for 1 hour to promote a thorough dissolving of the compounds. Test tubes, each with 4 mL of the different solutions, were then incubated for 12 days at room temperature (22°C) prior to assessing the changes in colour and TSP. 165 7.2.6 Amendments of sap with known phenols Sap-5a, which had develop no browning upon infection with fungi in this study, was filtered (3 x) through Whatman # 1 and adjusted to pH 7 with NaOH. The sap-5A was then amended with approximately 500 Aig/mL of either catechin, epicatechin, alf a-conidendrin or hydroxymatairesinol. Additional sap-5A was amended with a mixture of approximately 150 Aig/mL of each of the four individual compounds or with about 50 Aig/mL of catechin. Furthermore, sap-5A adjusted to pH 7 and unchanged sap-5A were included without amendments. The sap solutions were placed in an ultra-sonic bath for 1 hour. Approximately 4 mL of each filtered-sterilized (0.45 Aim) sap was placed in individual test tubes and incubated for 12 days at room temperature (22°C) prior to evaluating the colour and the TSP changes as described previously (see 7.2.7 Effect of buffer on sap browning Sap-9A was thawed and filtered through Whatman #1 filter paper. A biological buffer, containing 4.8 g of MES (Sigma, St Louis, MO) and 0.1 g NaOH, was then added to 50 mL of sap prior to filter-sterilization (0.2 Aim). Additional sap which received no buffer was also filter-sterilized. About 3-4 mL of the sterile sap was then decanted into pre-sterilized test tubes and inoculated with a fungus. One replicate was established for each fungus evaluated and also for the controls (with buffer and without buffer). The 166 test tubes were incubated at room temperature (22°C) for 12 days. Changes in coloration, pH and TSP were compared to controls kept at room temperature or in a freezer using the procedures outlined (see 7.3 Results and Discussions 167 7.3.1 Effect of pH on browning Brown colorations developed in sap-8A upon infection with microorganisms. The browning process was accompanied by a pH shift from slightly acidic to slightly alkaline and a decrease in TSP and OHMR (Table 14). However, sap-8A adjusted to pH 7 and incubated without fungal inoculum also underwent browning (Table 14) . A very slight colour change was already noticed at the time the sap was pH altered. The brown colour and the decrease in TSP and OHMR were similar in the pH 7 adjusted sap-8A and in the infected sap-8A. This observation demonstrated that the phenols were highly reactive in sap-8A at pH 7. However, the pH decreased in the adjusted sap-8A from 7 to 6.1 over incubation time, contrasting with the pH increase recorded in the inoculated samples. This observation indicated that carbon dioxide was absorbed from the atmosphere into the pH 7 adjusted sap and possibly reacted with protons released from phenols as oxidation and condensation occurred. Interestingly, brown precipitates formed in the pH altered sap-8A which looked similar to that observed in inoculated sap-8A. This observation demonstrated that neutral pH conditions alone can produce brown reaction products in western hemlock sap. Hathway and Seakins (1955; 1957) have demonstrated that catechin formed quinones followed by oxidative condensation at a neutral and a TABLE 14: CHANGES IN pH 7 ADJUSTED AND NON-ADJUSTED SAP (8A) FOLLOWING INCUBATION FOR 12 DAYS AT ROOM TEMPERATURE. TREATMENT control/frozen control/22°C Sap-8A/pH 7/22°C 0. piceae #3871 A2S A6S A13S A16S BIB BIS B15S SAP COLOUR colourless yellow brown brown brown brown brown brown yellow yellow brown pH 5.1 5.1 6.1 7.6 7.8 7.9 7.9 7.6 5.5 7.4 8.0 TSP /xg/mL 537 237 105 122 114 70 86 119 135 112 78 OHMR /xg/mL 16 13 0 0 0 0 0 0 9 0 0 169 slightly alkaline pH. In the current study catechin might have condensed to brown polymers as it was detected in the sap-8A before, but not after, incubation. In addition, the yellowing observed in the controls of sap-8A perhaps suggested formation of quinones from catechin as its catechin content decreased considerably over incubation time. 7.3.2 Effect of oxygen on browning A brown coloration and a brown precipitate developed in sap-8A, adjusted to pH 7, under atmospheric conditions whereas little coloration formed under nitrogen (Figure 27) . Interestingly, a slight colour change occurred when adjusting sap-8A to pH 7. Unchanged (pH 5) sap-8A developed a yellow coloration when incubated under oxygen but remained colourless under nitrogen (Table 15) . As previously noticed the changes in colorations were accompanied with TSP changes (Table 15). Thus neutral conditions and oxygen were required to produce browning in sap-8A containing catechin while colorations were absent in the unchanged (pH 5) sap-8A when kept under nitrogen. Hathway and Seakins (1955) have also shown that absence of oxygen arrested colorations of catechin at neutral pH. However, the findings made in the current study disagreed with an investigation (Miller et al. , 1983) reporting on browning in Douglas-fir sap incubated with nitrogen after 47 days. In another study (Arvey, 1993) colorations were also observed with phenols extracted from Douglas-fir and incubated in a helium 170 l ' l l , l I l K - A I IMV.1 I . I / . V . l l V»l i i>IE-t*J> HEMLOCK SAP AFTER 4 DAY OF INCUBATION (October 10, 94) OXYGEN Figure 27: Production of sap coloration under oxygen. 1 7 1 TABLE 15: EFFECT OF OXYGEN ON COLOUR AND TSP (jug/mL) CHANGES IN WESTERN HEMLOCK SAP. TREATMENT Nitrogen Nitrogen Oxygen Oxygen SAP COLOUR light brown colourless brown yellow PH 7 5 7 5 TSP /xg/mL 225 425 125 250 172 atmosphere after 32 weeks (Arvey, 1993) . It is possible that Arvey (1993) and Miller et al. (1993) did not completely remove oxygen from their systems as other studies have also reported on oxygen requirements for colorations in redwood (Anderson et al. , 1960), in rosewood (Millettia sp) (Kondo et al. , 1986) and in oak (Wassipaul et al. , 1987) . Also vacuum kiln-drying has controlled brown colorations in oak (Charrier et al., 1992). 7.3.3 Effect of heat on browning Heating of sap-8A (121°C, 15 minutes) produced a brownish coloration without changes in pH and in TSP (Table 16). However, TSP decreased in the heated sap-8A (controls) during incubation even though no changes in colour or pH occurred (Table 16) . In contrast, a shift in pH and a drastic drop in the TSP were recorded in the heated sap-8A following incubation with fungi and the colour of the sap further darkened (Table 16) . The fungi which produced a dark brown coloration also caused the most changes in the TSP and in pH. These observations suggested that resident sap enzymes were not involved in the browning which developed upon infection with fungi. Barton and Gardner (1966) have suggested that enzymes inherent in western hemlock sapwood, were involved in brownstain. In another study (Azim-Musbah, 1993) phenol oxidases were isolated from Douglas-fir sapwood and they were implicated in brownstains in this wood species. Residual tree enzymes or bacterial enzymes have also been suggested as causing brownstain in sugar pine and in white 173 TABLE 16: CHANGES IN HEATED1 SAP-8A AFTER INCUBATION WITH DIFFERENT MICROORGANISMS FOR 12 DAYS AT ROOM TEMPERATURE. TREATMENT control-frozen control-heated-frozen control-heated-22°C A3M A2S A6S B15S P. melinii 270B SAP COLOUR colourless light brown light brown light brown brown dark brown dark brown dark brown PH 5.0 5.0 5.0 5.0 6.4 6.8 6.7 6.8 TSP /xg/mL 494 465 306 305 205 32 122 63 1 = refers to heating of sap in an autoclave at 121°C for 15 minutes. 174 pines (Stutz, 1959) . However, it was reasonable to assume that heating denatured proteins in sap-8A in the current study and thus subsequent sap colorations were induced by the fungi. 7.3.4 Production of sap browning by heat and by pH alteration Heating and pH-modification of six different sap specimens produced brown precipitates in four of the sap samples while both heat and pH caused little coloration in the other two saps (Table 17). Thus heating and pH modification were able to identify sap specimens which were susceptible to browning. Evans and Halvorson (1962) have also shown that heating of western hemlock sap produced browning when collected from boards with brown endstain. Heating itself produced no changes in the pH of the sap samples corresponding with results obtained when heating water extracts of redwood (Anderson et al. , 1960) . These authors have also shown that heating of aqueous redwood extract, adjusted to pH 7, produced greatly intensified colorations compared to heating at pH 3 and pH 5. As expected, incubation of the pH-altered sap samples produced browning and changes in the TSP content in susceptible sap specimens (Table 17). Similar TSP changes were produced in both pH-altered and heated samples of sap-2B and sap-9D. This observation clearly suggested heat-modification of sap phenols. 175 TABLE 17 : CHANGES IN HEATED SAP AND IN pH MODIFIED SAP. SAMPLE # 2B 5B 5D 8B 9A 9D HEATED Sap Colour brown light yellow light yellow brown brown brown TSP /xg/mL 225 (350)2 450 (475) 150 (125) 775 (725) 300 (300) 125 (400) pH-ALTERED1 Sap Colour brown yellow yellow brown brown brown TSP /xg/mL 200 (350) 350 (425) 50 (100) 300 (500) 100 (325) 200 (475) 1 = refers sap adjusted to pH7 and incubated for 12 days at room temperature. 2 = numbers in brackets refer to TSP in sap before treatment. 176 Polcin and Rapson (1971), when studying the heat stability of some western hemlock extractives, have demonstrated that heat treatment produced brown colorations with flavan-3ols (e.g., catechin) whereas lignans were quite stable. However, browning developed in the heated sap-8B and sap-9A without TSP decrease (Table 17) suggesting that sap compounds other than phenols also formed colorations upon heating. Millett (1952) has implicated sugars in brownstaining of kiln-dried sugar pine (P. lambertiana Doul.) and reported that sugars produced an even more intense coloration in the presence of amino acids. In the current study, involvement of sugars in the browning of heated sap was not shown because similar infrared spectra were recorded for brown precipitates formed in heated sap-8B and in pH adjusted sap-8B (Weigel, 1994). Both infrared spectra indicated free phenolic, carbonyl and ether groups. The presence of ether groups was of particular interest because condensed tannins built from catechin monomers, have shown this linkage (Hemingway, 1989; Goldschmidt and Hergert, 1960) . Catechin was recorded only in the samples producing browning upon pH alteration and its concentrations decreased by 50 % during incubation. This observation strongly suggested that catechin played a major role in the browning process. 177 7.3.5 Amendment of water and sap with known phenols A yellow coloration developed in water amended with catechin and epicatechin, whereas the solutions with alfa-conidendrin and hydroxymatairesinol remained colourless after 12 days of incubation. Less coloration developed in water supplemented with a mixture of the four phenols than in the samples with catechin and epicatechin. This demonstrated that flavan-3ols were unstable and were auto-oxidized in water. However, brown precipitates were not formed with catechin or epicatechin, indicating that any oxidation products were still water soluble. It also suggested that other, unknown chromophores contribute to sap browning. Sap-5A, which had developed no brown coloration upon inoculation with microorganisms, nor following heating, produced brown colorations under alkaline conditions when amended with catechin, epicatechin and a mixture containing the four phenols (Table 18). In contrast, neither alfa-conidendrin nor OHMR alone produced browning in amended sap-5A, substantiating the observation made with water. TSP analysis also demonstrated that OHMR changes did not occur in sap-5A even when incubating under neutral conditions. This confirmed Barton's (1968) observation that OHMR's were stable and probably not involved in brownstain. However, alfa-conidendrin seemed more reactive under neutral conditions, as indicated by the large decrease in TSP; the poor water solubility of alfa-conidendrin may have cause this outcome. 178 TABLE 18 : COLOUR AND TSP IN pH ADJUSTED SAP-5A AMENDED WITH PHENOLS AFTER 12 DAYS OF INCUBATION AT ROOM TEMPERATURE. AMENDMENT Control-5A/22°C Control-5A/pH7/22 °C Catechin Epicatechin alfa-conidendrin hydroxymatairesinol Mixture2 Catechin (50 /xg/mL) SAP COLOUR light yellow yellow brown brown yellow yellow yellow brown TSP /xg/mL 98 (195)1 75 (138) 283 (703) 253 (705) 110 (320) 445 (525) 373 (753) not determined 1 = numbers in brackets refer to TSP of sample kept in a freezer 2 = refers to sap-5A containing a mixture of approximately 150 /xg/mL of each individual compound. 179 Browning also occurred in the pH 7 altered sap-5A when amended with 50 /xg/mL of catechin; only yellowing developed in additional sap-5A with 50 /xg/mL of catechin at pH 5. Interestingly, catechin was rapidly oxidized under neutral conditions to the point that it was not detected (HPLC) after 24 hours of incubation. Once again observations underscored that catechin was extremely reactive at a neutral pH and demonstrated that very small amounts of catechin can produce pronounced colour changes in sap. However, other compounds may contribute to the browning process as only a yellow coloration was produced in water amended with catechin and epicatechin. Anderson et al. (1960) showed that a mixture of redwood water extractives had a darker colorations than any of the individuals extracts alone. However, analytical extractive chemistry is required to determine which compounds can contribute to browning in western hemlock sap. One striking observation was the fact that naturally occurring epicatechin, which was detected in controls (unamended sap-5A), produced no browning under neutral conditions. However, browning occurred when sap-5A was amended with additional epicatechin, which resembled the coloration observed in sap-5A when amended with catechin (Table 18). It is possible that the natural epicatechin content of about 10-15 /xg/mL was too little to support browning in sap-5A even under highly reactive (neutral) conditions. 7.3.5 Buffer experiment 180 Browning occurred in the non-buffered sap-9A upon inoculation with different fungi (Table 19). As expected, browning was accompanied by a pH shift towards neutral and a decrease in TSP. In contrast, the same fungi were unable to produce colour changes in the buffered sap-9A and both pH and TSP remained unchanged (Table 19). Because sap-9A contained catechin, this demonstrated that catechin was much more reactive at a near neutral pH than at a slightly acidic pH. Fungi grew well in both the buffered sap and non-buffered sap-9A. Since dark color accompanied the rise in pH, stabilization of the pH of sap-9A effectively controlled browning. Bauch (1986) has reported that pH stabilization of the surface of Ilomba inhibited browning, related to a pH shift by bacteria (Starck et al. , 1984) . Oldham and Wilcox (1981) also controlled surface brownstain in solid-piled sugar pine lumber when keeping the pH of the wood surface low with phosphoric acid. Springer (1983) has prevented browning in western hemlock chips with sodium bisulfite and recorded a decrease in pH of sap from 5.4 to 2.6. Hathway and Seakins (1955) have also arrested browning of catechin with sodium hydrogen sulfite. In the current study, pH seemed critical to promotion of sap browning: for example, the fungus B15S produced browning at pH 6.6 181 TABLE 19: CHANGES IN BUFFERED AND NON-BUFFERED WESTERN HEMLOCK SAP (9A) INCUBATED WITH DIFFERENT FUNGI FOR 12 DAYS AT ROOM TEMPERATURE. Fungus Control2 Control4 Op 3871 A2S A6S A13S B15S BUFFERED1 Colour yellow3 yellow yellow yellow yellow yellow yellow pH 5.2 5.1 5.2 5.2 5.2 5.2 5.2 TSP /xg/mL 300 278 303 325 300 295 318 NON-BUFFERED Colour yellow3 yellow yellow yellow brown brown brown pH 5.2 5.1 6.3 6.3 6.9 6.9 6.6 TSP iiq/mL 300 203 300 305 140 135 157 1 = 50 mL of sap was buffered with 4.8 g of MES and 0.1 g NaOH. 2 = sample kept in the freezer during incubation. 3 = coloration was light yellow. 4 = sap incubated at room temperature. 182 but not at pH 6.3. However, 0. piceae and A2S also shifted pH to 6.3 but caused no browning. A critical pH has been demonstrated for browning in beech sap (Schmidt and Mehringer, 1989). In this study the composition of the sap and the metabolism of the fungi were also suggested to play a role in the pH changes of the non-buffered sap. Clearly, the fungi differed in their potential to induce pH shifts accompanied by colorations. 7.4 Conclusions 183 Alkaline conditions alone can promote browning in western hemlock sap. Oxygen is essential for browning of sap. Browning of inoculated sap following heat treatment was not produced by an enzyme resident to western hemlock sap. Heat and a neutral pH produced browning in sap predisposed to colorations. Catechin played a major role in browning but other sap constituents contributed to the coloration. Stabilization of the sap pH in the acidic range effectively controlled sap browning produced by fungi. 8.0 Summary and Recommendations 184 In this study, conclusive evidence was presented that fungi can produce brown colorations in western hemlock during seasoning. Brown discolourations can disfigure both amabilis-fir and western hemlock, but the research concentrated on the latter species which is economically more important. These discolourations, clearly different from sapstain, can occur in several types and intensities and are a serious problem in high-value markets. Because little is known about their causes, means for their control are still unavailable. Therefore, fundamental research was initiated to elucidate the biology and chemistry of hemlock brownstain and to suggest control measures. As a first approach into the cause of hemlock brownstain microscopic examinations of hem-fir samples were performed. While samples exhibited different macroscopic types of colorations, a similar microscopic distribution of the colouring matter was demonstrated. Subsequent histochemical studies provided first evidence that the brown colouring matter contained catechin. Because fungi and bacteria were frequently associated with brownstain, as seen microscopically, in vitro brownstain experiments were performed on western hemlock sap and wood. These experiments clearly demonstrated that a broad microflora can 185 produce brownstain, which led to the hypothesis of this study that microorganisms are involved in hemlock brownstain. To determine the role of microorganisms in hemlock brownstain, two field studies were performed with emphasis on the brown colorations disfiguring western hemlock logs and lumber during seasoning, the most troublesome discoloration to industry. In the first field study dark-pigmented fungi were isolated predominatly from western hemlock logs showing brownstain and it was shown that the brownstained regions contained a lower quantity of soluble phenols than non-stained areas. While this observation suggested a link between brownstain and the presence of fungi in western hemlock logs, additional factors promoting brownstain became evident when monitoring the western hemlock logs and sawn lumber produced from the logs over time. Prolonged log storage time promoted discoloration. Salt water storage caused severe colorations in logs and lumber much more than in the logs stored on land. However, several limitations of this field trial were recognized; for example the small sample size with logs from one growth site, one age class and harvested in one season of the year. Future research in this area might address the following questions: * Do felling season, log age and growth site influence the extractive compositions with respect to brownstain? * Does death of parenchyma cells influence hemlock brownstain? 186 * Is brownstain formed in lumber even when sawn from freshly, felled logs? * Are metal ions involved in the dark-brown colorations observed in water-stored logs? In a second trial, fungi were clearly associated with brownstain on the endcuts of sawn lumber. The fungi isolated were most likely resident in the wood prior to the sawing of lumber. Longitudinal movement of water-soluble phenols was shown towards the cross-end cut. However, shortcomings of the study, for example that this survey dealt with one sawmill only, at one time of the season and with no information on log source and log storage conditions, could be addressed in a future study To demonstrate microbial involvement in the phenomenon of hemlock brownstain, an immunolabeling technique was applied on infected western hemlock. While in situ production of brownstain was associated with hyphae, the antibody was unable to detect the fungus in wood, possibly due to a lack of specificity. In future research in this area a polyclonal antibody, which is much easier to work with, should be used instead of a monoclonal antibody providing that interference problems, which can arise from an inherent wood microflora, can be eliminated. Fungal involvement in brownstain was then conclusively demonstrated in sap assays. Fungi shifted the sap pH from slightly acidic to 187 near neutral, or above neutral, which ionized and oxidized phenols causing browning of sap. Oxygen and a near neutral pH were essential to produce colorations. However, fungi did not produce colorations in buffered sap and it was not possible to study production of browns tain in the absence of oxygen, due to the aerobic nature of these organisms. Furthermore, 0. piceae infection of 15 /xm wood sections, which must have contained very small amounts of brownstain precursors, and also of infection of wood blocks, confirmed that the fungus can produce brownstain in solid western hemlock. Interestingly, the hyphae of 0. piceae remained hyaline in western hemlock, whereas pigmentation was observed when it grew in lodgepole pine and on a nutrient medium. Thus, physiological factors appeared to trigger production of brownstain in western hemlock when infected with 0. piceae as well as suppress the development of sapstain. Catechin was demonstrated to play a major role in the browning of western hemlock. However, other yet unknown sap constituents probably were also involved in brownstaining. Future research in this area would involve thorough extractive and analytical chemistry. For example, western hemlock extractives could be fractionated based on their ability to produce browning followed by isolation, purification and identification of individual compounds. This approach may clarify which of the compounds, other than catechin, are involved in hemlock brownstain. However, based on 188 information gained from the current study, the following simplified mechanism is proposed for the involvement of fungi in the brownstain developing in western hemlock during seasoning. In addition to any inherent tree microflora, which could be extensive especially in older trees, western hemlock is likely colonized within hours after the felling of the trees. Nutrients and water-soluble wood extractives migrate to the wood surface during storage of logs and lumber which leads to an accumulation of phenols. As wood moisture content decreases, fungal colonization can progress axially and can alter wood extractives within the wood, for example by hydrolysing glycosides. Oxidation, for example of flavan-3ols, can then readily occur upon exposure to air, forming coloured condensation and polymerization products especially when accompanied by an increase in pH. However, it is emphasized that production of brownstain is likely more complex and other mechanisms can not be excluded. Based on the elucidation of some of the factors involved in brownstaining the following means for control are suggested: * Because fungi are implicated, biocides should be reassessed but might need a supplement for example a chelating agent or a reducing agent. * Because phenols are highly reactive at higher pH, stabilization of the pH of the wood substrate with a strong 189 buffer or acidification of the wood substrate should be investigated. * Because oxygen is essential for browning, kiln-drying of lumber in the absence of oxygen should be investigated, for instance the use of a super-heated steam vacuum dryer. * Because high temperatures promote coloration, drying schedules using lower temperatures should be investigated for high-value western hemlock. * Because precursors to brownstain migrate to the surface, pre-steaming western hemlock before kiln-drying should be investigated to remove water-soluble wood extractives. However, the practicality of these recommendations as commercial means of control of hemlock brownstain remains to be determined. Certainly a strategy of minimizing fungal infection and enhancing rapid handling of the tree breakdown into final wood products should help reduce the problem. 190 9.0 Literature Cited Adler, E. 1951. Sulfite pulping properties of spruce wood from unpeeled, floated logs. Part 1. Catechol tannins in the surface layer of the sapwood. Svensk Papperstidning 13:445-449. Anderson, A.B., E.L. Ellwood, E. Zavarin and R.W. Erickson. 1960. Influence of extractives on seasoning stain of redwood lumber. Forest Prod. J. 10 (4) : 212-218. Arvey, S.W. 1993. Chemical brown stain of Douglas-fir wood. PhD thesis. Oregon State University. Corvallis, Oregon. 76p. Aufsess, H. von. 1984. Some examples of wood discolorations related to mechanisms for potential protection of living trees against fungal attack. IAWA Journal 5(2):133-138. Avramidis, S., S. Ellis and J. Liu. 1993. The alleviation of brownstain in hem-fir through manipulation of kiln-drying schedules. Forest Prod. J. 43 (10) : 65-69. Azim-Musbah, D.A. 1993. Extraction and properties of Douglas-fir sapwood polyphenol oxidase. PhD thesis. Oregon State University. Corvallis, Oregon. 83p. Bailey, R.W. 1911. Oxidizing enzymes and their relation to "sap stain" in lumber. Botanical Gazett, vol.50. No. 26:142-147. Ballard, T.M. 1992. Personal communication with B. Kreber. Banerjee, S., J. Little, M. Chan, B.T. Luck, C. Breuil and D.L. Brown. 1994. Production and characterization of antibodies to the sapstaining fungus Ophiostoma piceae. Can. J. Microbiol. 3 5-44. Barton, G.M. 1962. Brown stain on hemlock lumber. Progress Report. No. 1. Canada Dept. of Forestry, Forest Products Laboratory, Vancouver, B.C. Proj. No. V-132-2. 7p. Barton, G.M. 1963. Brown stain on hemlock lumber. Progress Report No. 3. Canada Dept. of Forestry, Forest Products Laboratory. Proj. No. V-132-2. Vancouver, B.C. 26p. Barton, G.M. 1968. Significance of western hemlock extractives in pulping and lumber. Forest Prod. J. 18(5):76-80. Barton, G.M. 1970. New polyoxyphenols from western hemlock sapwood. Wood and Fiber Sci. 2:144-150. Barton, G.M. 1973a. Extractive chemistry of western hemlock. Canada Dept. of the Environment, Western Forest Products Laboratory, Vancouver, B.C. Working Plan Study No. WFP-20-1(VP-015). 191 Barton, G.M. 1973b. The significance of western hemlock phenolic extractives in groundwood pulping. Tappi 56 (5) : 115-118. Barton, G.M. and J.A.F. Gardner. 1966. Brown stain formation and the phenolic extractives of western hemlock Tsuga heterophylla (Raf.) Sarg.). Canada Dept. of Forestry Publ. No. 1147. 20p. Barton, G.M and R.S. Smith. 1971. Brown stain in kiln-dried Abies amabilis lumber. Canada Dept. of Fisheries and Forestry Bi-Monthly Research Notes, Vol.27(3), May-June. 21-23. Basham, J.T. and L.D. Taylor. 1965. The occurrence of fungi and bacteria in normal and discoloured heartwood of second-growth sugar maple in Ontario. Plant Dis. Rep. 49 (9) : 771-774. Bauch, J. 1986. Verfarbungen von Rund- und Schnittholz und Moglichkeiten fur vorbeugende SchutzmaSnahmen. Holz-Zentralblatt 112 (152) =2212-2218. Bauch, J. and P. Baas. 1984. Development and characteristics of discoloured wood. IAWA Journal 5(2):91-98. Bauch, J., W. H611 and R. Endeward. 1975. Some aspects of wetwood formation in fir. Holzforschung 29 (6) :198-205. Bauch, J., G. Seehann and R. Endeward. 1982. Characterization of sapwood and brown heart in Terminalia superba Engl, et Diels.. Holzforschung 36:257-263. Bauch, J., H. v.Hundt, G. WeiSmann, W. Lange and H. Kubel. 1991. On the causes of yellow discolorations of oak heartwood (Quercus Sect. Robur) during drying. Holzforschung 45(2):79-85. Bauch, J., 0. Schmidt, Y. Yazaki and M. Starck. 1985. Significance of Bacteria in the Discolorations of Ilomba wood (Pvcnanthus angolensis Exell). Holzforschung 39:249-252. Bell, A.A. and M.H. Wheeler. 1986. Biosynthesis and functions of fungal melanin. Ann. Rev. Phytopathol. 24:411-451. Blake, D. 1992. Personal communication with B. Kreber. Blanchette, R.A., J.B. Sutherland and D.L. Crawford. 1981. Actinomycetes in discoloured wood of living silver maple. Can. J. Bot. 59:1-7. Braun, M. and S. Lewark. 1992. NaSlagerung von Buchenstammholz aus den Sturmwiirfen (1990) im Saarland (2) . Verfarbungen von Sageholz 1991 nach unterschiedlicher Zeitdauer zwischen Aufarbeitung und Beregnungsbeginn. Holz-Zentralblatt 118:1739;1742 . 192 Breuil, C. 1994. Personal communication with B. Kreber. Breuil, C. , K.A. Seifert, J. Yamada, L. Rossignol and J.N. Saddler. 1988 . Quantitative estimation of fungal colonization of wood using an enzyme-linked immunosorbent assay. Can. J. For. Res. 18:374-377. Byrne, A. 1992. Unpublished observations. Forintek Canada Corp. Vancouver, B.C. Byrne, A. and R.S. Smith. 1991. A review of hemlock brownstain and related discolorations. Paper presented at BIOFOR '91, Sault Ste. Marie, Ont. Forintek Canada Corp. Vancouver, B.C. 7p. Cech, M.Y. 1966. Brown stain in white pine. Forest Prod. J. 16 (11) :23-27. Charrier, B., J.P. Haluk and G. Janin. 1992. Prevention of brown discolorations in european oakwood occurring during kiln drying by a vacuum process: Colorimetric comparative study with a traditional process. Holz als Roh- und Werkstoff 52(11):433-437. Chung, W.C. and R.S. Smith. 1986. Staining fungi in packaged Hem-fir lumber. Forintek Canada Corp. Vancouver, B.C. Proj. No. 02-17-43-108. 21p. Clark, J.W. 1957. A gray non-fungus seasoning discoloration of certain red oak. Southern Lumberman 194:35-3 8. Clark, J.E. 1994. Personal communication with B. Kreber. Cosenza, B.J., M. McCreary, J.D. Buck and A.L. Shigo. 1970. Bacteria associated with discoloured and decayed tissues in beech, birch and maple. Phytopathol. 60 (11) :1547-1551. Council of Forest Industries of British Columbia. 1983. B.C. Hem-Fir, a wood that works well. Vancouver, B.C., Canada. Daniels, C.R. 1992. Determination of didecyldimethylammonium chloride on wood surfaces by HPLC with evaporative scattering detection. J. Chromatogr. Sci. 30:497-499. Daniels, C.R. 1993a. Personal communication wtih B. Kreber. Daniels, C.R. 1993b. Chemical analysis development. Forestry Canada Report. Proj.No. 1712L022. 8p. Daniels, C.R. 1994. Unpublished results. Forintek Canada Corp., Vancouver, B.C. Dujesiefken, D. , W. Liese and J. Bauch. 1984. Discolorations in the heartwood of oak-trees. IAWA Journal 5(2): 128-132. 193 Eades, H.W. 1932. Stains and rots in lumber. Forest Products Laboratories of Canada. Vancouver Lab. Lecture to Pacific Lumber Inspection Bureau, March 24, 1932. Vancouver, B.C. 31p. Eades, H.W. 1943. Investigation of brown streak in western hemlock used for aircraft purposes. Canada Dept. of Mines and Resources, Project No. V-56. Ottawa. 5p Ellis, S. 1993. Personal communication with B. Kreber. Ellis, S. and S. Avramidis. 1993. Brown stain in Pacific coast hemlock. IAWA Journal 14(l):23-28. Ellwood, E.L. and B.A. Ecklund. 1959. Bacterial attack of pine logs in pond storage. Forest Prod. J. 9(9):283-292. Englerth, G.H. and J.R. Hansborough. 1945. The significance of the discolorations in aircraft lumber: noble fir and western hemlock. USDA Forest Pathol. Spec. Release 24, For. Prod. Lab., Madison, WI. lOp. Etheridge, D.E. 1970. Ascocoryne sarcoides (Jacq. ex Gray) Groves and Wilson and its association with the decay of conifers. Fonds. Rech. For. Univ. Laval. Bull. 13:19-26. Evans, R.S. and H.N. Halvorson. 1962. Cause and control of brown stain in western hemlock. Forest Prod. J. 12 (8) :367-373. Fengel, D. Ultrastructural changes during aging of wood cells. Wood Sci. and Techn. Vol. 4:176-188. Fengel, D. and G. Wegener. 1984. Wood. De Gruyter, Berlin, Germany. 613 pp. Findlay, W.P.K. 1959. Sapstain of timber. Forestry Abstracts 20(1,2):1-14. Forsyth, P.G. 1988. Control of non-microbial sapstains in southern red oak, hackberry, and ash lumber during air seasoning. M.S. Thesis, Mississippi State University, 49 p. Forsyth, P.G. andT.L. Amburgey. 1991. Microscopic characterization of non-microbial gray sapstain in southern hardwood lumber. Wood and Fiber Sci.23(3):376-383. Forsyth, P.G. and T.L. Amburgey. 1992. Prevention of non-microbial stains in water-stored oak logs. Forest Prod. J. 42(4):59-61. Fritz, C.W. 1931. Stain and defects in standing white birch. Pulp paper Mag. Can. 31:565-566. 194 Funk, A. 1965. The symbiotic fungi of certain ambrosia beetles in British Columbia. Can. J. Bot. 43:929-932. Gibbs, J.N. 1993. The biology of Ophiostomatoid fungi causing sapstain in trees and freshly cut logs. p. 153-160. IN: M.J. Wingfield, K.A. Seifert and J.F. Webber. Ceratocvstis and Ophiostoma. Taxonomy, ecology and pathogenicity. APS Press. St. Paul, Minnesota, pp. 293. Ghariban, S. 1994. Unpublished results. Forest Products Biotechnology, University of British Columbia, Vancouver, B.C. Goldschmid 0. and H.L. Hergert. 1961. Examination of western hemlock for lignin precursors. Tappi 44 (12) : 858-870. Goudie, D. 1992. Personal communication with B. Kreber. Greaves, H. 1969. Micromorphology of the bacterial attack of wood. Wood Sci. and Techn. 3:150-166. Grosu, R., M. Boiciuc, and R. Vladut. 1973. The influence of bacteria on the quality of wood and wood products. I. The influence of bacteria on wood and wood stored in water. (In Romanian). Ind. Lemnului 24 (4) :143-147. Hagerman, A.E. 1989. Chemistry of tannin-protein complexation. p. 335-341. IN: R.W. Hemingway and J.J. Karchesy. Chemistry and Significance of Condensed Tannins. Plenum Press, New York, N.Y. 553p. Halloin, J.M. 1982. Localization and changes in catechin and tannin during development and ripening of cottonseed. New. Phytol. Vol. 90:651-657. Hansen, K. 1988. Bacterial Staining of Samba (Triplochiton scleroxvlon). Int. Res. Group Wood Preserv. IRG/WP/1362. Stockholm, Sweden. Haslam, E. 1989. Plant polyphenols. Vegetable tannins revisited. University Press, Cambridge, U.K. 23Op. Hathway, D.E. and J.W.T. Seakins. 1955. Autooxidation of catechin. Nature 176:218. Hathway, D.E. and J.W.T. Seakins. 1957. Autooxidations of polyphenols Part III. Autooxidation in neutral aqueous solutions of flavans related to catechin. J. Chem. Soc. 1562-1566. Hedley, M. and R. Meder. 1992. Bacterial brown stain on sawn timber cut from water-stored logs. Int. Res. Group Wood Preserv. IRG/WP/1532. Stockholm, Sweden. 195 Hemingway, R.W. 1989. Structural variations of proanthocyanidins and their derivatives, p. 83-107. IN: R.W. Hemingway and J.J. Karchesy. Chemistry and Significance of Condensed Tannins. Plenum Press, New York, N.Y. 553p. Hergert, H.L. 198 9. Hemlock and Spruce Tannins: An Odyssey, p. 3-20. IN: R.W. Hemingway and J.J. Karchesy (ed.). Chemistry and Significance of Condensed Tannins. Plenum Press, New York, N.Y. 553p. Hergert, H.L. 1960. Chemical composition of tannins and polyphenols. Forest Prod. J. 10 (11) :610-617. Herman, A. 1937. Chemical brown stain in sugar pine. Western Pine Assoc. Res. Laboratory Note No. 24. Hoster, H.R., 1974. Verfarbungen bei Buchenholz nach Wasserlagerung. Holz als Roh und Werkstoff 32:270-277. Hrutfiord, B.F. and R. Luthi. 1981. Chemistry of oregonin. IN: Chemistry and morphology of wood and wood components. Int. Symp. on Wood and Pulping Chemistry, Stockholm. Vol. 1:95-96. Hrutfiord, B.F., R. Luthi and K.F. Hanover. 1985. Colour formation in western hemlock. J. of Wood Chem. and Tech. 5(4):451-460. Hubert, E.E. 1926. The brown stains of lumber. Timberman. 27(7):44-45, 48, 50. Johnson, B.R. 1979. Permeability changes induced in three western conifers by selective bacterial inoculation. Wood and Fibre 11 (1) :10-21. Ingram, J.K. 1993. Unpublished results. Forintek Canada Corp. Vancouver, B.C. Kai, Y. and E.P. Swan. 1990. Chemical constituents contributing to the colour of western redcedar heartwood. Mokuzai Gakkaishi 36 (3) :218-224. Karchesy, J.J. 1992. Personal communication with B. Kreber. Koltzenburg, C. 1975. Zur Entstehung von Verfarbungen in gelagertem Bergahornholz (Acer pseudoplatanus L.) . Holz als Roh und Werkstoff 33:420-426. Kondo, R., T. Mitsunaga and H. Imamura. 1986. The chemistry of the colour of wood II. The effect of environmental factors on the coloration of murasakitagayasan (Millettia sp.) heartwood. Mokuzai Gakkashi 32 (6) :462-466. 196 Knutson, D.M. 1973. The bacteria in sapwood, wetwood and heartwood of trembling aspen (Populus tremuloides) . Can. J. Bot. 51: 498-500. Kreber, B. 1993. Unpublished observations. Forintek Canada Corp. Vancouver, B.C. Kreber, B. and A. Byrne. 1993. Causes of hemlock brownstain: A sawmill survey of hemlock brownstain in 1992. Forintek Canada Corp., Vancouver, B.C. 7p. Kreber, B. and C.R. Daniels. 1993. Phase 4: Monitoring of biological and chemical parameters in hem-fir lumber over time. Forintek Canada Report. Vancouver, B.C. Proj. No. 1712K024. 28p. Kreber, B., E.L. Schmidt and A. Byrne. 1994. Methyl bromide fumigation to control non-microbial discolorations in western hemlock and red alder. Forest Prod. J. 44 (10) : 63-67. Kucera, L.J. and S. Katuscak, 1992. Das Phanomen Holzfarbe. pp 43-52. .In: Holz-Farbe-Gestaltung, 24. Fortbildungskurs der Schweizerischen Arbeitsgemeinschaft fur Holzforschung in Weinsfeld, Lignum Zurich. Kuo, M.L. and D.G. Arganbright. 1980. Cellular distribution of extractives in redwood and incense cedar. Holzforschung 34(2) : 41-47. Lagerberg, T., G. Lundberg and E. Melin. 1927. Biological and practical research into blueing in pine and spruce. Svenska Skogsvardsfor. Tidskri. 25:145-272. Lane, P.H. and T.C. Scheffer. 1960. Water sprays protect hardwood logs. Forest Prod. J. 10 (6) : 277-288. Liese, W. and G. Karnop. 1968. Uber den Befall von Nadelholz durch Bakterien. Holz als Roh- und Werkstoff 26 (6) :202-208. Liese, W. and R. Schmid. 1961. Licht- und elektronenmikroscopische Untersuchungen iiber das Wachstum von Blauepilzen in Kiefern und Fichtenholz. Holz als Roh und Werkstoff 19 (9) : 329-337. Liese, W. and R. Schmid. 1964. Uber das Wachstum von Blauepilzen durch verholzte Holzwande. Phytopathol. Zeitschrift 51(4) : 385-394. Liese, W. and R.D. Peek. 1984. Experience with wet storage of conifer logs. Dansk. Skovforen. Tidsskr. 69:73-91. Lubbers P. and M. GroS. 1992. Entwicklung der Holzqualitat beregneter Fichten/Tannen nach der Auslagerung. Holz-Zentralblatt 118(132):2133;2136;2138. 197 Mace, M.E. andC.R. Howell. 1974. Histochemistry and identification of condensed tannin precursors in roots of cotton seedlings. Can. J. Bot. 52:2423-2426. Mackay, J.F.G. 1975. Properties of northern aspen discolored wood related to drying problems. Wood and Fiber Sci. 6(4):319-326. Miller, D.J., D.M. Knutson and R.D. Tocher. 1983. Chemical brown staining of Douglas-fir sapwood. Forest Prod. J. 33(4) :44-48. Millett, M.A. 1952. Chemical brown stain in sugar pine. Forest Prod. J. 3:232-236. Moeck, H.A. 1968. Surfaced Hemlock-Balsam Lumber Sorting on the Basis of Bark-Maggot Damage. Reprint from British Columbia Lumberman. 3p. Morgan, J.W. and R.J. Orsler. 1968. The chemistry of colour changes in wood. I. Significance of Stilbenes. Holzforschung 22(1):11-16. Munch, E. 1907. Die Blauefaule des Nadelholzes. Naturw. Zeitschr. Forst und Landw. 5:531-573. Oldham, N.D. and W.W. Wilcox. 1981. Control of brown stain in sugar pine with environmentally acceptable chemicals. Wood and Fiber 13 (3) :182-191. Ota, M. and K. Taneda. 1989. The chemistry of colour changes in Kiri wood (Paulownia tomentosa Steud.) 1. - Caffeic acid sugar esters responsible for colour changes. Mokuzai Gakkaishi 35(5):43 8-446. Ota, M., S. Onodera, M. Ohira and K. Taneda. 1991. The chemistry of colour changes in Kiri wood (Paulownia tomentosa Steud.) 2. Radial distribution and seasonal variations of the contents of total phenolics, of caffeic acid sugar esters, and of kiri peroxidase activity in xylem, plus some properties of kiri peroxidase. Mokuzai Gakkaishi 37 (3) :254-260. Packman, D.F. 1960. The acidity of wood. Holzforschung 14(6):178-183. Panshin, A.J. and C. de Zeeuw. Textbook of wood technology. McGraw-Hill Book Company. New York, N.Y. 722p. Parameswaran, N. and J. Bauch. On the origin of phenolic compounds in the wood rays of Abies alba. Wood Sci. and Techn. Vol 9: 165-173. Phelps, J.E. and E.A. McGinnes Jr. 1984. Pruning of Black Walnut -Subsequent discoloration and how to minimise it. IAWA Journal 5 (2) :110. 198 Pigman, W. , E. Anderson, R. Fischer, M.A. Buchanan and B.L. Browning. 1953. Colour precursors in spruce- and western hemlock woods and inner bark. Tappi 36(1):4-12. Polcin J., M. Wayman, C.B. Anderson and W.H. Rapson. 1969. Sapwood and heartwood groundwood of western hemlock and Jack pine. Part 1. Influence of extractives on optical properties. Pulp and Paper Mag. Canada 70 (10) :91-98. Polcin, J. and W.H. Rapson. 1971. Sapwood and heartwood groundwood of western hemlock and Jack pine. Part 2. Heat stability of extractives. Pulp and Paper Mag. Canada 72 (10) : 84-90. Rasmussen, C.A. 1940. Age of logs as related to kiln brown stain of sugar pine. Western Pine Assoc. Res. Laboratory Note No. 28. Rathke, K.H. 1991. Untersuchungen beim Einschnitt von beregnetem Fichtenrundholz aus der Sicht der Sageindustrie. Holz-Zentralblatt 117:2442. Roll-Hansen F. and H. Roll-Hansen. 1979. Microflora of sound-looking wood in Picea abies stems . Eur. J. For. Path. 9(5):308-316. Rosch, R. , W. Liese and H. Berndt. 1969. Untersuchungen iiber die Enzyme von Blauepilze. I. Cellulase-,polygalakturonase-, pektinesterase- und Laccase Aktivitat. Arch. Microbiol. 67:28-50. Sachs, I.B., J.C. Ward and R.E. Kinney. 1974. Scanning electron microscopy of bacterial wetwood, sapwood, and normal heartwood in poplar trees. Proc. 7th Annual Scanning Electron Microsc. Symp., Part 2, IIT Res. Inst., Chicago, p. 453-460. Sachs, I.B., J.C. Ward and E.H. Bulgrin. 1966. Heartstain in red oak. Holz als Roh und Werkstoff 24 (10) :489-497. Sauter, J.J. and B. van Cleve. 1989. IAWA Journal 10(4): 395-403. Scheel, R. 1993. Personal communication with B. Kreber. Scheffer, T.C. and R.M. Lindgren. 1940. Stains of sapwood products and their control. Techn. Bull. 714. U.S. Dept. of Agric. Washington, D.C. Scheffer, T.C. 1973. Microbial degradation and the causal organisms, p. 31-106. In:D.D. Nicholas (ed) . Wood deterioration and its Prevention by Preservative Treatments. Syracuse University Press, Syracuse, N.Y. Schink, B. and J.C. Ward. 1984. Microaerobic and anaerobic bacterial activities involved in formation of wetwood and discoloured wood. IAWA Journal 5(2):105-109. 199 Schmid, R. and W. Liese. 1965. Zur Aussenstruktur der Hyphen von Blauepilzen. Phytopath. Zeitschr. 54 (3) :275-284. Schmidt, 0. 1986. Investigations on the influence of wood-inhibiting bacteria on the pH-value in trees. Eur. J. For. Path. 16:181-189. Schmidt, 0. and H. Mehringer. 1989. Bakterien im Stammholz von Buchen aus Waldschadensgebieten und ihre Bedeutung fur Holzverfarbungen. Holz als Roh und Werkstoff 47:285-290. Schroeder, H.A. and C.J. Kozlik. 1972. The characterization of wetwood in western hemlock. Wood Sci. and Techn. 6:85-94. Scott, E.S. 1984. Populations of bacteria in popular stems. Eur. J. For. Path. 14:103-112. Seifert, K.A. and B.T. Grylls. 1991. A survey of the sapstaining fungi of Canada. Forintek Canada Corp. Ottawa, ON. 32p. Shevenell, B.J. and W.C. Shortle. 1986. An ion profile of wounded red maple. Phytopathol. 76 (2) : 132-135. Shields, J.K., R.L. Desai and M.R. Clarke. 1973. Control of brown stain in kiln-dried eastern white pine. Forest Prod. J. 23(10):28-30. Shigo, A.L. and E.M. Sharon. 1968. Discolorations and decay in hardwoods following inoculations with hymenomycetes. Phytopathol. 58 (11) :1493-1498. Shigo, A.L. andW.E. Hillis. 1973. Heartwood, discoloured wood, and microorganisms in living trees. Annu. Rev. Phytopathol. 11:197-122 Shigo, A.L., J. Stankewich and B.J. Cosenza. 1971. Clostridium sp. associated with discoloured tissue in living trees. Pythopathol. 61(1): 122-123. Shortle, W.C. 1984. Biochemical mechanisms of discoloration, decay, and compartmentalisation of decay in trees. IAWA Journal 5(2) :100-101. Shortle, W . C , J.A. Menge and E.B. Cowling. 1978. Interaction of bacteria, decay fungi and live sapwood in discoloration and decay of trees. Eur. J. For. Path. 8:293-300. Siegle, H. 1967. Microbiological and biochemical aspects of heartwood sta in in Betula papyr if era Marsh. Can. J. Bot. 45(2) :147-154. Smith, R.S. 1970. Black stain in yellow cedar heartwood. Can. J. Bot. 48:1731-1739. 200 Smith, R.S. 1975. Economic aspects of bacteria in wood. In: W. Liese (ed). Biological transformation of wood by micro-organisms. Proceedings of sessions on wood products pathology and 2nd International Congress of Plant Pathology, Sept. 10-12, 1973, Minneapolis, Minn. Springer-Verlag, New York. Smith, R.S. 1994. Personal communications with B. Kreber. Smith, R.S. and R.J. Spence. 1987. Fungal involvement in hem-fir oxidative brown stain. Report. Forintek Canada Corp. Vancouver, B.C. Project No. 1743K207. 51p. Song. H-K. 1987. Chemistry of phenolic stains in Douglas-fir sapwood. PhD thesis. Oregon State University, Corvallis, Oregon. 146p. Spradling, M. 1936. Penetration of Trichoderma lignorum into sapwood of Pinus taeda. Journ. of Agric. Research 52(7) : 541-546. Washington, D.C. Springer, E.L. 1983. Evaluation of chemical treatments for maintaining brightness of stored wood chips. Tappi J. 66(2) : 93-96. Stafford, H.A. 1988. Proanthocyanidins and the lignin connection. Phytochemistry Vol 27(1):l-6. Starck, M. , J. Bauch, and M.H. Simatupang. 1984. Characteristics of normal and discoloured wood of Ilomba (Pvcnanthus angolensis Exell). Wood Sci. and Techn. 18:243-253. Stout, A.W. 1950. Storage factors and the occurrence of chemical brown stain in sugar pine. Western Pine Assoc. Res. Laboratory Note No. 39. Stout, A.W. 1959. Water sprinkling protects decked logs. Western Pine Assoc. Res. Laboratory Note No. 2.114. 6 pp. Stutz, R.E. 1959. Control of brown stain in sugar pine with sodium azide. For. Prod. J. 9 (12) :459-463. Stutz, R.E. 1961. The prevention of sugar pine brown stain. Western Pine Assoc. Res. Laboratory Note No. 4.5114. 4pp. Stutz, R.E. P. Koch and M.L. Oldham. 1961. Control of brown stain in eastern white pine. Forest Prod. J. 11 (6) :258-260. Sutcliffe, R. 1992. Unpublished report prepared for B. Kreber. Forintek Canada Corp. Vancouver, B.C. 2p. Sutcliffe, R. and D. Miller. 1991. Study of sample of amabilis fir suspected of brownstain resulting from kiln-burn. Forintek Canada Corp. Eastern Laboratory, Ottawa. 4p. 201 Swan, E.P. 1984a. A preliminary look at hemlock brown stain. Forintek Canada Corp. Western Laboratory, Vancouver. lOp. Swan, E.P. 1984b. Unpublished results. Forintek Canada Corp, Vancouver, B.C. Tattar, T.A. and A.E. Rich. 1973. Extractable phenols in clear, discoloured woody tissues and bark of sugar maple and red maple. Phytopathol. 63:167-169. Tattar, T.A., W.C. Shortle and A.E. Rich. 1971. Sequence of microorganisms and changes in constituents associated with discoloration and decay of sugar maples infected with Fomes connatus. Phytopathol. 61:556-558. Terazawa, M., M. Miyake and H. Okuyama. 1984. Reddish compounds in living tissue of woods V. Reddish orange staining in keyamahannoki (Alnus hirsuta) and hannoki (A. iaponica) [Betulaceae] caused by interaction of hirsutoside and catechol oxidase after cutting the woods. Mokuzai Gakkaishi 30 (7) : 601-607. Unligil, H.H. 1972. Penetrability and strength of white spruce after ponding. Forest Prod. J. 22(9):92-99. van der Kamp, B.J. 1992. Personal communication with B. Kreber. van der Kamp, B.J., A. A. Gokhale and R.S. Smith. 1979. Decay resistance owing to near-anaerobic conditions in black cottonwood wetwood. Can. J. of Forest Res. 9(l):39-44. Walter, M. 1993. Der pH-Wert und das Vorkommen niedermolekularer Fettsauren im NaiSkern der Buche (Fagus sylvatica L.) . Eur. J. For. Path. 23:1-10. Ward, J.C. and J.G. Zeikus. 1980. Bacteriological, chemical and physical properties of wetwood in living trees, p. 135-165. IN: J. Bauch. Natural variations of wood properties. Mittg. der Bundesforschungsanstalt fur Forst- und Holzwirtschaft 131. Hamburg-Reinbeck, Germany. 2 05p. Warren, H.V. and C.H. Howatson. 1947. Biogeochemical Prospecting for Copper and Zinc. Bull, of the Geological Soc. of America Vol. 58:803-820. Warren, H.V., R.E, Delvault and R.I. Irish. 1952. Preliminary studies on the biogeochemistry of iron and manganese. Economic Geology Vol. 47:131-145. Wassipaul, M. , M. Vanek and J. Fellner. 1987. Verfarbung von Eichenholz bei der kiinstlichen Holztrocknung. Holzf orschung und Holzverwertung 39(1):l-5. 202 Wayman, M., C.B. Anderson and W.H. Rapson. 1968. Bleaching characteristics of hemlock heartwood and sapwood groundwood of jack pine and western hemlock. Pulp and Paper Mag. of Canada 69(9:51-60. Weigel, G. 1994. Unpublished results. Forintek Canada Corp. Vancouver, B.C. Whittaker, E.I. 1962a. Brown stain on hemlock lumber. Progress Rept. No. 2: The significance of bacterial infection. Canada Dept. of Forestry, Forest Products Laboratory. Vancouver, B.C. Proj . Rept. V-132-2 lOp. Whittaker, E.I. 1962b. The interaction of Coryne sarcoides and fungi associated with red heart in lodgepole pine. Can. J. Bot. 40:255-256. Wilcox, W.W. and N.D. Oldham. 1972. Bacterium associated with wetwood in white fir. Phytopathol. 62 (3) : 384-385. Wong, W.C. and T.F. Preece. 1978. Erwinia salicis in cricket bat willows: phenolic constituents in healthy and diseased wood. Phys. Plant Pathol. 12:349-357. Yamada, T., H. Tamura, K. Mineo and K. Suzuki. 1987. Discoloration and cation accumulation in the wood of living sugi trees attacked by the sugi bark borer. J. Jpn. For. Soc. 69 (4) : 121-126. Yazaki, Y., J. Bauch and R. Endeward. 1985. Extractive components responsible for the discolorations of Ilomba wood (Pycnanthus angolensis Exell). Holz als Roh und Werkstoff 43:359-363. Zimmermann, G. 1974. Untersuchungen iiber Art und Ursachen von Verfarbungen an Bergahorn-Stammholz (Acer pseudoplatanus L.). Forstw. Cbl. 93:247-261. Zink, P. and D. Fengel. 1988. Studies on the colouring matter of blue-stain fungi. Part 1. General characterization and the associated compounds. Holzforschung 42 (4) :217-220. Zink, P. and D. Fengel. 198 9. Studies on the colouring matter of blue-stain fungi. Part 2. Electron microscopic observations on the hyphae walls. Holzforschung 43 (6) : 371-374. Zink, P. and D. Fengel. 1990. Studies on the colouring matter of blue stain fungi. Part 3. Spectroscopic studies on fungal and synthetic melanins. Holzforschung 44 (3) : 163-168. 


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