Root and Butt Rots of Forest Trees loth International Conference on Root and Butt Rots Proceedings of the IUFRO Working Party 7.02.01 Québec City, Canada, September 16-22, 2001 G. laflamme, J.A. Bérubé, G. Bussières (éditeurs/editors) Centre de foresterie des Laurentides - Laurentian Forestry Centre Rapport d'information - Information Report LAU-X-126 .+. Ressources naturelles Canada Service canadien des forêts Natural Resources Canada Canadian Forest Service Canada Root and Butt Rots of Forest Trees Proceedings of the IUFRO Working Party 7.02.01 Quebec City, Canada, September 16-22, 2001 G. Laflamme, J.A. Bérubé, G. Bussières (éditeurs/editors) Ressources naturelles Canada - Natural Resources Canada Service canadien des forêts - Canadian Forest Service Centre de foresterie des Laurentides - Laurentian Forestry Centre Rapport d'information - Information Report LAU-X-126 DONNÉES DE CATALOGAGE AVANT PUBLICATION (CANADA) / NATIONAL LIBRARY OF CANADA CATALOGUING IN PUBLICATION DATA International Union of Forestry Research Organizations. Working Party 7.02.01 (Root and Butt Rots of Forest Trees) (lOlh: 2001: Quebec, Canada) Root and butt rots offorest trees : proceedings of the IUFRO Working Party 7.02.01, Québec, Canada, September 16-22, 2001 (Information report; LAU-X-126) Includes prefatory material in French. Includes bibliographic references. ISBN 0-662-33332-2 Cat. no. F046-18/126E 1. Root rots - Congresses. 2. Roots (Botany) - Diseases and pests - Congresses. 3. Trees - Diseases and pests - Congresses. 1. Laflamme, G. II. Bérubé, Jean, 1962­ III. Bussières, Guy, 1953- IV. Laurentian Forestry Centre. V. Information report (Laurentian Forestry Centre); LAU-X-126. Photos de la couverture: (Gauche) Rond de mortalité de pins rouges causé par Heterobasidion annosum (R. Blais, SCF) (Droite) Fructifications de Heterobasidion annosum sur une souche de pin rouge (C. Moffet, SCF) Cover photos: (Left) Circular patch of dead red pines killed by Heterobasidion annosum (R. Blais, CFS) (Right) Heterobasidion annosum fruiting bodies on a red pine stump (C. Moffet, CFS) SB741.R75 157 2003 634.9'63 C2003-9800 19-9 © Sa Majesté la Reine du Chef du Canada 2003 Numéro de catalogue F046-181126E ISBN 0-662-33332-2 ISSN 0835-1589 Il est possible d'obtenir sans frais un nombre restreint d'exemplaires en français de cette publication auprès de : Ressources naturelles Canada Service canadien des forêts Centre de foresterie des Laurentides 1055, rue du P.E.P.S., c.P. 3800 Sainte-Foy (Québec) G1V 4C7 Site Web du CFL : http://www.cfl.scf.mcan.gc.ca Des copies ou des microfiches de cette publication sont en vente chez: Micromédia Ltée 240, rue Catherine, bureau 305 Ottawa (Ontario) K2P 2G8 Tél. : (613) 237-4250 Ligne sans frais: 1-800-567-1914 Téléc.: (613) 237-4251 Les textes apparaissent dans la version fournie par les auteurs, avec l'autorisation de publier. Ces derniers demeurent responsables tant de la forme que du fond de leurs écrits. © Rer Majesty the Queen in Right of Canada 2003 Catalog Number F046-18/126E ISBN 0-662-33332-2 ISSN 0835-1570 Limited additional copies of this publication are available at no charge from: Natural Resources Canada Canadian Forest Service Laurentian Forestry Centre 1055 du P.E.P.S., P.O. Box 3800 Sainte-Foy, Quebec G1V 4C7 LFC Web Site: http://www.cfl.cfs.nrcan.gc.ca Copies or microfiches ofthis publication may be purchased from: Micromedia Ltd. 240 Catherine St., Suite 305 Ottawa, Ontario K2P 2G8 Tel.: (613) 237-4250 Toll Free: 1-800-567-1914 Fax: (613) 237-4251 The texts included in these proceedings are the original versions provided by the authors with authorization to publish and the authors remain responsible for both the form and content oftheir papers. TABLE OF CONTENTS / TABLE DES MATIÈRES FOREWORD AND ACKNOWLEDGEMENTS XI AVANT-PROPOS ET REMERCIEMENTS XII SESSION 1: PHYLOGENY AND TAXONOMY Preliminary characterization of Armillaria isolates from tea (Camellia sinensis) in Kenya A.P. Sierra, W. Otieno and A. Termorshuizen . Phylogenetic relationship among Laetiporus spp. in Japan Y. Ota and T. Hattori 9 Studies in Polyporus subg. Polyporellus: on congruence ofthree biological, morphological and phylogenetic species D. Krüger, K.W. Hughes and R.H. Petersen 14 Identification of Armillaria spp. in north-west Spain using molecular techniques O. Aguin Casai, A. Pérez Sierra, M. Sabaris Roma and J.P. Mansilla Vazquez .. . . . . . . . . . . . . .. 24 Investigations on Heterobasidion in central and eastern Asia K. Korhonen, Y.-c. Dai, J. Hantula and E. Vainio 27 Polymorphism within the 26S rDNA and intergenic spacer (IGS-I) ofwild and artificial genets of Armillaria spp. reveal putative natural hybrids and phylogenic relationships G.!. McDonald, N.B. Klopfenstein and M.-S. Kim 32 Phylogenetic reconstruction of North American Armillaria species and related European taxa based on nuclear ribosomal DNA internaI transcribed spacers M.B. Hughes, A. Weir and S.O. Rogers 32 SESSION Il: ECOLOGY AND BIODIVERSITY Armillaria and Annosl/111 root diseases in a mOllntain pine (Pinus ml/go var. uncinata) stand in the Alps D. Rigling 35 The influence of plant series and plant association groups on the incidence and severity of root diseases in Southwest Oregon forests E.M. Goheen, D.J. Goheen and K. Marshall. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 40 Biodiversity in monocultures: the Sitka spruce stump S. Woodward. . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 47 Bacterial diversity in Sitka sprllce stllmps and their interactions with decay-causing fungi A.C. Murray and S. Woodward 55 111 Swiss-stone pine trees and spruce stumps may represent the primary habitat for Heterobasidion annosum sensu stricto in Western Italian Alps G. Nicolotti, P. Gonthier, M. Garbelotto, G.C. Varese and G.P. Cellerino 63 Relationship between soil factors, root infection by Collybia fusipes and tree health in Quercus robur and Q. rubra C. Camy and B. Marçais 71 Fire and Armillaria: effects on viability and dynamics in eastem Oregon, USA G.M. Filip, S.A. Fitzgerald and L. Yang-Erve 78 Effects ofnutrients on Armillaria root disease in greenhouse-grown lodgepole pine (Pinus contorta) KT. Mallett and D.G. Maynard 85 Investigations on the distribution and ecology of Armillaria species in Albania B.M. Lushaj, M. Intini and E. Gupe 93 Impact of Annillaria rRNA - IGS groups on crown condition of maples in Portneuf County P. DesRochers, M. Dusabenyagasani, J.A. Bérubé and R.C. Hamelin 105 The effect of soil/root microfungi on Armillaria rhizomorph formation H. Kwa.sna 113 Effects of nutrient and water stress on Armillaria disease incidence in Maritime pine B. Lung-Escarmant, M.L. Desprez-Lousteau, D. Loustau, A. Giraud and G. Capron 117 An experimental study of the effects of ozone on tree-Armillaria interactions D. Rigling, P. Lawrenz, H. Blauenstein and U. Heiniger 122 Effects of stump treatments with Phlebiopsis gigantea on mycodiversity J. Hantula, E.J. Vainio, K. Lipponen, A.-M. Hallaksela and K Korhonen 127 Impact of stand and soil factors on the distribution of C. fusipes root rot in oak forests B. Marçais, C. Camy and O. Caël 128 Spatial molecular analysis and monitoring of Inonotus tomentosus R.C. Hamelin, G. Laf1amme, L. Bernier, M.l Bergeron and H. Germain 128 Association of Inonotus tomentosus with spruce beetle attack F.A. Baker 129 IV Incidence of Tomentosus root disease relative to spruce density and slope position in south-central Alaska KJ. Lewis, L. Trummer, R. Shipley and S. Parsons . Red-rot ofNorway spruce (Picea abies) in Austria - Relations with site factors based on a survey by the Austrian forestry inventory C. Tomiczek, R. Buechsenmeister and Th.L. Cech 129 130 The role of soil moisture content in Armillaria root disease K.I. Mallett, D.G. Maynard and c.L. Myrholm 130 SESSION III: CONTROL Integrated control ofArmillaria mellea by Trichoderma harzianum and fosetyl-Al F. Raziq and R.T.V. Fox 133 Improving stump treatment by harvesting machine J.E. Pratt, D.J. Brooks and M.A. Lipscombe 139 Impact ofbiological and chemical treatments against Heterobasidion annosum on non-target micro­ orgalllsms G.c. Varese, P. Gonthier and G. Nicolotti 145 Growth ofinoculated Heterobasidion annosum in roots of Picea abies - Effects ofthinning and stump treatment with Phlebiopsis gigantea M. Pettersson and J. Ronnberg 155 Effect of stump treatment on transfer of Heterobasidion annosum root rot in Norway spruce LM. Thomsen 160 Operational stump treatment against Heterobasidion annosum in European forestry - CUITent situation M. Thor.................................. . . . . . . 170 Stump treatment experiments against Heterobasidion in the Halian Alps N. La Porta, R. Grillo, P. Ambrosi and K. Korhonen 176 Microbes inhabiting Picea wounds and their antagonism to Haematostereum sanguinolentum M.T. Dumas and J.A. McLaughlin 181 Costs and effects of biological control of root rot in Poland Z.H. Sierota 194 Stump inoculation with Pleurotus ostreatus (Jacq.: Fr.) P. Kummer A. Zolciak 197 Colonisation and degradation of Sitka spruce sapwood by the Rotstop strain of Phlebiopsis gigantea P.J. Bailey, S. Woodward and J.E. Pratt 200 Simulated stump treatment experiments for monitoring the efficacy of Phlebiopsis gigantea against Heterobasidion K. Korhonen 206 v Preliminary results using biological control against Heterobasidion annosum on Silver tir in southem Italy G. Sicoli, L. Trigona, N. Luisi and F. Mannerucci 211 Testing of Rotstop on Sitka spruce, Douglas-fir and larch LM. Thomsen and J.B. Jacobsen. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 216 Potential for biological control of Heterobasidion annosum in the UK using Rotstop® J. Webber and K. Thorpe 221 Results of Heterobasidion annosum eradication performed in 1993-94 in two red pine plantations G. Laflamme, R. Blais and G. Bussières 226 Biological control of Armillaria spp. with Basidiomycetes P. Lakomy 226 Variation in Phlebiopsis gigantea and monitoring the effects of release J. Fatehi, C. Wood and J. Stenlid 227 Intimate mixtures of susceptible species and the spread of Armillaria root disease B.J. van der Kamp 227 SESSION IV: GENETICS AND POPULATION DYNAMICS Molecular markers reveal genetic isolation and phylogeography in the S- and F-intersterility groups of Heterobasidion annosum H. Johannesson and J. Stenlid 231 Studies on the ecology and genetics of hybridization in Heterobasidion M.M. Garbelotto, W.J. Otrosina, LH. Chapela and P. Gonthier 238 Air-borne inoculum composition, patterns of inter-group gene flow, of Heterobasidion annosum coll. species in pure and mixed natural forests in the Alps P. Gonthier, G. Nicolotti, M. Garbelotto, G.c. Varese and G.P. Cellerino 245 Isolation and molecular structure of a laccase gene from the root rot fungus Heterobasidion annosum (S-type) F.O. Asiegbu 253 Analyses of selected Expressed Sequence Tags (EST) from Heterobasidion annosum (P-type) - Pinus sylvestris pathosystem F.O. Asiegbu, J. Nahalkova, W. Choi, J. Stenlid and R.A. Dean 260 Evolution ofArmillaria genets over eight years (1992-2000) J.-J. Guillaumin and Ph. Legrand 267 VI Population structure and mating system of Climacocystis borealis M.A. Büttner, T.N. Sieber, and O. Holdenrieder 276 The mating behavior of Stereum sanguinolentum M. Calderoni, T.N. Sieber and O. Holdenrieder 280 Development of Simple Sequence Repeat (SSR) markers in Armillaria ostoyae S.R.H. Langrell, B. Lung-Escarmant, A. Giraud and S. Decroocq 282 Sequence polymorphism in laccase genes of S, P & F types of Heterobasidion annosum M.S. Abu, F.O. Asiegbu, H. Johannesson and J. Stenlid 287 Genetic variation in Heterobasidion abietinum (H annosum F group) population P. Capretti, S. Tegli, P. Lakomy and L. Zamponi 293 Application of genetic markers for biological studies of Armillaria M.-S. Kim, N.B. Klopfenstein and G.!. McDonald 296 Identification ofpathogenicity genes in Heterobasidion annosum using Expressed Sequence Tags (ESTs) M. Karlsson, A OIson and J. Stenlid 296 Population structure of Armillaria spp. and Megacollybia platyphylla in Quercus rubra stumps over a 17-year period M.B. Hughes, P.M. Wargo, J.J. Worrall, S.O. Rogers, and A. Weir 297 Presence of dsRNA in Heterobasidion annosum 1. Ihrmark, J. Zheng, E. Stenstrôm and J. Stenlid 297 Population structure oftwo Armillaria species coexisting in managed mountainous Norway spruce forests S. Prospero, D. Rigling and O. Holdenrieder 298 SESSION V: PATHOGENICITY, RESISTANCE AND ETIOLOGY Genetic variation in susceptibility to Heterobasidion annosum infection in spore-inoculated fresh stumps of Picea abies clones G. Swedjemark and B. Karlsson 301 Pathogenicity ofP-, S-, and F-intersterility groups of Heterobasidion annosum to Scots pine, Norway spruce and common fir in inoculation experiments A. Werner and P. Lakomy 310 Burt rot of old growth of Chamaecyparis pisifera caused by Serpula himantioides y. Abe, T. Hattori and M. Kawai 318 Vll VlIl Characterization of fungal isolates from Newtonia buchananii trees in sub-montane rain forest in Tanzania F.A. Mrema, F.O. Asiegbu, A. Rosling and K. Wahlstrôm 323 The study of chitin-binding lectin from Pinus nigra seeds during the interaction of conifer seedlings with the necrotrophs Heterobasidion annosum and Fusarium avenaceum J. Nahalkovâ, F. Asiegbu, G. Daniel, J. Hrib, B. Vookova and P. Gemeiner 333 Extent of decay in Parashorea malaanonan developing from logging injuries in Sabah, Malaysia M. Sudin, M.A. Pinard, S. Woodward and S.S. Lee. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 343 Phenological change and distribution ofbasidiocarps of Phaeolus schweinitzii in a severely infected larch stand T. Yamaguchi 348 Heterobasidion root rot - A threat to the forests in Estonia M. Hanso and S. Hanso 351 Comparative study between Norway spruce and silver fir tree health status infected by Heterobasidion annosum using electrical resistance and chemical coloration V. Vujanovic and D. Karadzic 356 Preliminary evaluation of Scots pine plantations "resistant" to Heterobasidion annosum Bref. (Fr.) V. Lygis, R. Vasiliauskas, J. Stenlid, A. Vasiliauskas 362 Inoculation test of Armillaria mellea on Hinoki cypress under controlled temperature and soil water conditions E. Hasegawa 366 Virulence of Armillaria cepistipes and Armillaria ostoyae isolates on Norway spruce seedlings S. Prospero, O. Holdenrieder and D. Rigling 370 RESROBS: Resistance of spruce to root and butt rot disease, an EU-funded research program S. Woodward, J. Stenlid, M. Michelozzi, H. Solheim, B. Karlsson and P. Tsopelas 375 Virulence of Heterobasidion annosum S-P hybrids is determined by mitochondria A. Oison and J. Stenlid 378 Assessment of loblolly pine decline in central Alabama N.J. Hess, W.J. Otrosina, E.A. Carter, J. Steinman, J.O. Jones, L.G. Eckhardt, A.M. Weber and C.H. Walkinshaw 378 Root and butt rot of Chamaecyparis obtusa caused by Perenniporia subacida M. Tabata, T. Kato, M. Ohkubo, Y. Abe and S. Yoshinaga 379 A novel method of collecting and storing Heterobasidion annosum basidiospores for use in stump inoculation trials G. MacAskill and H. Steele .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 379 Pathogenicity of Rotstop to Sitka spruce in Britain J. Pratt and LM. Thomsen 380 SESSION VI: INCIDENCE AND EPIDEMIOLOGY Monitoring root rots in young Scots pine plantations (up to 20 years) M. Manka and W. Szewczyk 383 Incidence ofbutt rot in consecutive rotations of Picea abies in south-western Sweden J. R6nnberg, U. Johansson and M. Pettersson 388 Characterization and confirmation of enzyme activity in an endochitinase protein expressed in Phellinus weirii-infected Douglas-fir A. Zamani, R.N. Sturrock and AK.M. Ekramoddoullah 394 Risk of spread of Heterobasidion abietinum on Abies alba stands in the Mediterranean region P. Capretti and A. Santini 396 Infection and distribution of Heterobasidion species in stumps of Douglas-fir C. Delatour, A Soutrenon, J.L. Flot and G. Sylvestre-Guinot 400 Incidence ofroot diseases in the fir forest of Mount Parnis National Park, Greece P. Tsopelas and A. Angelopoulos 408 Root and butt rots in semi-mature, pre-commercially thinned stands ofbalsam fir in Newfoundland G. Warren and B. English 413 Early events of infection of roots of Pinus sylvestris seedlings with Heterobasidion annosum strains of P-, S-, and F-intersterility groups - Scanning electron microscopy A. Werner, P. Lakomy and K. Idzikowska 419 MOHIEF: Modelling ofHeterobasidion in European forests, a EU-funded research program S. Woodward, J.E. Pratt, T. Pukkala, K.A. Spanos, G. Nicolotti, C. Tomiczek, J. Stenlid, B. Marçais and P. Lakomy .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 423 Infectious cycle ofArmillaria ostoyae on maritime pine stands of different ages B. Lung-Escarmant, F. Maugard, A Giraud, M.A. Escrivant, F. Molinier, F. Meril1eau, G. Vida 428 Early development of Heterobasidion root rot in young Norway spruce stands T. Piri and K. Korhonen 432 Preliminary study on the survival and spread of Armillaria mellea in mulches in gardens A. Pérez Sierra 436 Schade Lake root disease survey K. Knowles 439 IX Growth reduction of Douglas-fir due to non-Iethal infection by Armillaria ostoyae M.G. Cruickshank 441 Leptographium species and their vectors as components of loblolly pine decline L. Eckhardt, J. Jones, N. Hess, E. Carter and J. Stienman 442 Characterisation of phenylalanine ammonia lyase production following challenge ofSitka spruce with Heterobasidion annosum M.-T. Hsu and S. Woodward 442 Distribution of the Heterobasidion annosum intersterility groups in Poland P. Lakomy and A. Werner 443 APPENDIXI List of participants 445 x FOREWORD IUFRO Working Party 7.02.01 usually convenes every four years. The 10th International Conference on Root and Butt Rot was he1d in Quebec City, Canada, from September 16-21, 2001. The event attracted 67 participants from 13 countries, although 29 members were unable to attend due to the tragic events of September Il, 2001. Gaston Laflamme, the conference organizer, and Ariane Plourde, Research Director at the Laurentian Forestry Centre, Canadian Forest Service, welcomed the delegates. Working Party co-ordinator Claude Delatour chaired the business portion of the meeting, in which it was decided that Europe, possibly Poland, wou1d be the site of the next meeting. After the scientific communication sessions, participants travelled to the Montreal and Ottawa regions for field trips. The delegates visited forests infested by Inonotus tomentosus, Heterobasidion annosum and Armillara sp. The proceedings contain the texts or abstracts of papers submitted to the organizing committee and of the 55 scientific posters presented at the meeting. They also include texts by those who were registered for the conference but who could not attend. Authors are responsib1e for content. Texts are presented under the following six headings: "Phylogeny and Taxonomy", "Eco10gy and Biodiversity", "Control", "Genetics and Population Dynamics", "Pathogenicity, Resistance and Etiology" and "Incidence and Epiderniology". Gaston Laflamme, Conference organizer Co-ordinator, IUFRO WP 7.02.01 Jean Bérubé and Guy Bussières, Conference co-organizers Conference Website: www.cfl.scf.rncan.gc.caliufro-rbr200 1 ACKNOWLEDGEMENTS We sincerely thank our official host, the Laurentian Forestry Centre, Canadian Forest Service in Quebec City, as well as two other CFS units, the Great Lakes Forestry Centre in Sault Ste. Marie and the Science Branch in Ottawa. Thanks are a1so extended to our other sponsors: the Canadian Phytopathological Society, the ministère des Ressources naturelles du Québec, Tembec Forest Products Group, Régent Instruments Inc., Alliance Forest Products Inc., Kruger Inc., Conseil de la recherche forestière du Québec and Forintek. In addition, we wou1d like to thank ail those who helped organize the meeting at the Château Frontenac. We would also 1ike to express our appreciation to the field trip organizers, particularly Robert Blais and Julie Dubé (CFS-LFC), Louise Innes and Solange Simard (MRNQ), Mike Dumas (CFS-GLFC) and John McLaughlin (OMNR). In addition, we are grateful for the assistance provided by the staff of Communications Services, CFS­ LFC. We thank Charles-Paul Cou10mbe and Benoit Arsenault, who created and updated the conference's Web site with the help of the LFC's Informatics Services team, and we thank the editing team made up of Isabelle Lamarre, Diane Paquet and Pamela Cheers, who took on the arduous chore of producing this document and without whom the publication ofthese proceedings wou1d have been very difficult indeed. G. Laflamme, J.A. Bérubé and G. Bussières XI AVANT-PROPOS Le groupe de travail 7.02.01 de l'IUFRO se réunit habituellement à tous les quatre ans. C'est dans la ville de Québec, Canada que s'est tenu la lOc Conférence sur la pourridiés des arbres forestiers, du 16 au 21 septembre 2001. L'événement a regroupé 67 participants provenant de 13 pays, bien que la tragédie du Il septembre 2001 nous ait privés de la participation de 29 membres. Les délégués ont été accueillis par Gaston Laflamme, organisateur du congrès, et par madame Ariane Plourde, directrice de la recherche au Centre de foresterie des Laurentides du Service canadien des forêts. Une session d'affaires du groupe de travail s'est tenue sous la présidence de M. Claude Delatour, coordonnateur du groupe de travail; parmi les différents sujets de discussion, il a été décidé de privilégier l'Europe comme site de la prochaine réunion, qui serait possiblement tenue en Pologne. Après les sessions des communications scientifiques, les congressistes se sont rendus dans les régions de Montréal et d'Ottawa pour participer à des excursions en forêt. Ils ont eu l'occasion de visiter principalement des stations affectées par Inonotus tomentosus, Heterobasidion annosum et Armillaria sp. Le compte rendu regroupe les textes ou les résumés des communications soumis au comité organisateur et ceux des 55 affiches scientifiques présentées pour la réunion de ce groupe de travail. Les textes de ceux et celles qui étaient inscrits au congrès mais qui n'ont pu se rendre sur les lieux de la conférence sont intégrés au document. Il est à noter que les auteurs sont responsables du contenu de leurs textes. Les textes sont regroupés sous les six thèmes suivants: « Phylogénie et taxonomie », « Écologie et biodiversité », « Contrôle », « Génétique et dynamique des populations », « Pathogénicité, résistance et étiologie », ainsi que « Incidence et épidémiologie ». Gaston Laflamme, organisateur de la conférence Coordonateur du IUFRO WP 7.02.01 Jean Bérubé et Guy Bussières, co-organisateurs de la conférence Site Web de la conférence: www.cfl.scf.mcan.gc.caliufro-rbr2001 REMERCIEMENTS Nous tenons à remercier vivement le Centre de foresterie des Laurentides, du Service canadien des forêts à Québec, qui était 1'hôte officiel de cette rencontre, ainsi que deux autres composantes du Service canadien des forêts, soit le Centre de foresterie des Grands Lacs de Sault Ste. Marie et la Direction générale des sciences forestières à Ottawa. Des remerciements sont également adressés à tous nos autres commanditaires : la Société canadienne de phytopathologie, le ministère des Ressources naturelles du Québec, Tembec - Groupe des produits forestiers, Régent Instruments Inc, les Produits forestiers Alliance, Kruger Inc, le Conseil de la recherche forestière du Québec et Forintek. Un grand merci à tous ceux et celles qui ont participé à l'organisation de la rencontre au Château Frontenac. Il nous faut souligner de façon particulière tous ceux et celles qui ont pris part à l'organisation des visites de terrain, en particulier Robert Blais et Julie Dubé, SCF-CFL, Louise Innes et Solange Simard, MRN-Québec, Mike Dumas, SCF-CFGL, et John McLaughlin, MRN-Ontario. Enfin, nous avons grandement apprécié l'aide du personnel des Services des communications du SCF­ CFL. Nos remerciements vont à Charles-Paul Coulombe et Benoit Arsenault qui, avec l'équipe des Services informatiques du CFL, ont créé et mis à jour le site Web du congrès, ainsi qu'à l'équipe d'édition constituée d'Isabelle Lamarre, Diane Paquet et Pamela Cheers, qui ont su mener à terme ce long travail d'édition et sans qui ce compte rendu aurait été difficilement réalisable. G. Laflamme, J.A. Bérubé et G. Bussières xii Phylogeny and Taxonomy PRELIMINARY CHARACTERIZATION OF ARMILLARIA ISOLATES FROM TEA (CAMELLIA SINENSIS) IN KENYA A.P. Sierra', W. Otien02 , and A. Termorshuizen3 'Royal Horticultural Society, Wisley, Woking, Surrey GU23 6QB, UK 2Tea Research Foundation of Kenya, P.O. Box 820, Kericho, Kenya 3Biological Fanning Systems, Wageningen University, Marijkeweg 22,6709 PG, The Netherlands SUMMARY The taxonomy of Armillaria in several African countries remains unresolved, but A. heimii and A. mellea are the two main species described in Kenya. A survey covering the main tea growing districts in Kenya was carried out in 1997 for the presence of Armillaria spp. and 47 isolates were collected from infected tea plants. Cultural morphology, somatic incompatibility reactions, PCR-RFLPs ofITS and lGS and DNA sequencing of the lGS were perfonned for the characterization of the collected isolates. For comparison purposes, Kenyan isolates of A. mellea (K5 and K8) and Kenyan isolates K10 and K12 (belonging to a yet unnamed biological species) were selected as reference isolates. The isolates were separated into two groups by morphology and somatic incompatibility. The restriction pattern of the ITS using Alu I, Hinfl, and Nde II and the restriction pattern of the lGS using Alu l confirmed the morphological grouping. Group l may represent A. heimii and group II could be a new species. The latter was supported by the sequencing of the lGS. Comparison of the sequences with those published in the Genbank database showed that they were different from A. mellea and identical to KlO and K12. No A. mellea was found during the survey. Keywords: Armillaria, lGS, lTS, phylogeny, Kenya INTRODUCTION Tea (Camellia sinensis) is a crop of great significance in Kenya and Armillaria is a primary pathogen causing root rot. It is one of the most important diseases of tea and the losses in small tea farms can be as high as 50% (Onsando et al. 1997). The losses in larger fanns are lower due to disease control programs. There are two Armillaria species recorded in this country, A. mellea (Vahl) Kummer and A. heimii Pegler. Ota et al. (2000) confinned by isozyme and RAPD analysis that isolates of A. mellea in Africa were identical to isolates of A. mellea from Japan, they derived from the same origin and migration may have occurred from Asian countries to Africa. The identification of African Armillaria species is often limited by the absence or scarcity of basidiomata in tropical regions (Gibson 1960). The lack of basidiomata and haploid testers has led to the identification of the Armillaria species based on vegetative mycelia, and in the field the disease is mainly confinned by the white mycelium found underneath the bark of infected roots. Different methods have been used for the identification of Armillaria in Africa based on somatic incompatibility (Abomo-Ndongo and Guillaumin 1997), isozyme electrophoresis (Agustian et al. 1994; Mwangi et al. 1989; Mwenje and Ride 1996, 1997), molecular markers such as DNA restriction fragment polymorphisms (Anderson et al. 1987; Chillalli et al. 1997; Smith and Anderson 1989) and DNA sequence analysis (Anderson and Stasovski 1992). Mohammed (1994) used RAPD markers to distinguish African isolates with diverse origins. RFLP and nucleotide sequence data of the intergenic spacer region of the ribosomal DNA operon were recently used to distinguish between Southern Africa isolates of Armillaria (Coetzee et al. 1997, 2000). The authors Phylogeny and Taxonomy showed that both nuclear and organelle DNA-based molecular markers provide an alternative to mating tests and basidiomata morphology that can aid systematics of Armillaria in Africa. Identification of Armillaria species in sorne African countries remains unresolved and this was highlighted by Coetzee et al. (2000) who studied Armillaria in pine plantations in South Africa where A. mellea and A. heimii were also described. They concluded in their research that isolates identified as A. heimii were in fact Armillaria sp. or A. fuscipes (Petch 1909). The objective of this study was to characterize the Armillaria species affecting tea in Kenya using different methods based on morphology, somatic incompatibility, PCR-RFLPs of the ITS and IGS region and sequence of the IGS region. MATERIALS AND METHODS Isolates In 1997, 14 districts were surveyed in the main tea growing areas in Kenya. Forty-seven isolates of Armillaria were collected mainly from infected roots oftea (Camellia sinensis), but also Dombeya sp., Dioscorea sp., Coffea arabica, Musa acuminata and Eucalyptus sp. African isolates of A. mellea (K5, K8 and STl) and isolates KlO and K12 from Kenya were donated by Dr J.-J. Guillaumin (INRA Clermont-Ferrand, France) for comparison purposes. Basidiomata and cultural morphology Morphological features of basidiomata were recorded when these were present. Production of basidiomata was attempted in vitro in 1 litre flasks. The medium consisted of 50 g of milled beech wood sawdust, 20 g of whole blended orange, 60 g of whole grain rice, lOg of peptone and 5 g of agar (Tirro 1991). The mixture was topped up with 400 ml of water and autoclaved for 1 hour at 121°C. The flasks were inoculated and incubated at 25°C for 4 weeks in the dark. After this period the temperature was adjusted at 20°C with a 12 hour photoperiod. The isolates were cultured on 2% MEA and 3% MEA with added peptone (0.06%). The plates were incubated in the dark at 22°C over four weeks and cultural morphology was described. Somatic incompatibility Isolates were paired as 3 mm plugs on 2% MEA overlaid with cellophane using the method described by Hopkin et al. (1989). The plates were incubated in the dark for three weeks at 20°C. After this time, a 2 x 2 cm square of cellophane was cut out containing the paired isolates and immersed in freshly prepared L-Dopa solution (0.05%). These were incubated at 37°C for 1 hour before being examined for the presence of the black line between the thalli (Mallet et al. 1986). DNA extraction The isolates were grown in liquid culture (1 % malt extract, 0.5% yeast extract and 1% glucose) and incubated in the dark for three weeks at 20°e. The flasks were not shaken during this time. The mycelium was harvested, rinsed with distilled water, frozen in liquid nitrogen and stored at -80 0e. The DNA was extracted from the frozen mycelium using a DneaslM Plant Mini Kit (Quiagen). 2 Phylogeny and Taxonomy PCR-RFLPs The internaI transcribed spacer (ITS) was amplified by PCR with the universal primers ITS 1 and ITS4 (White et al. 1990). The intergenic spacer (IGS) region between the 26S and 5S was amplified with two different sets ofprimers. The first set included LR12R, 5' CTG AAC GCC TCT AAG TCA GAA 3' (Veldman et al. 1981) and 0-1,5' AGT CCT ATG GCC GTG GAT 3' (Duchesne and Anderson 1990) recommended by Anderson and Stasovski (1992). The second set ofprimers included: P-l, 5' TTG CAG ACG ACT TGA ATG G 3' and 5S-2B, 5' CAC CGC ATC CCG TCT GAT CTG CG 3' recommended by Coetzee et al. (1997). Ready-To-Go PCR beads (Amersham Pharmacia Biotech) were used for the PCR amplification. Individual reactions were brought to a final volume of 25 Jll. Each reaction contained 1.5 units of Taq DNA polymerase, 10 mM Tris-HCl, 50 mM KCI, stabilizers including BSA, 1.5 mM MgClz, 200 JlM of each dNTP, 0.1 JlM of each primer and purified water (Sigma Chemical Co.). The PCR amplification program to amplify the ITS was as described by Chillali et al. (1997). The PCR program to amplify the IGS region consisted of 1 cycle of 95°C for 95 sec, followed by 35 cycles of 60°C for 60 sec, noc for 120 sec and 95°C for 60 sec and a fmal extension at noc for 10 min. The amplifications were performed on a Progene (Techne, UK) thermocycler. The ITS and IGS amplified products were purified with a QIAquick lM Purification Kit (Quiagen). The ITS was digested separately with 5 units of the restriction enzyme HinfI, Alu land Nde II and the IGS was digested with 5 units of Alu 1. The restriction patterns were visualized in 3% agarose gels stained with ethidium bromide. Sequencing of the IGS was carried out by MWG Biotech. Lasergene (DNASTAR 2000) software for Macintosh was used for editing and aligning the sequence files. Additional sequences from the GenBank databases available through the National Center for Biotechnology Information (NCBI, Bethesda, MD) were obtained. The alignments were made with Megalin and the indels coded using MacClade (Maddison and Maddison 1992). Phylogenetic analyses were performed using PAUP version 4.0b (Swofford 1998) RESULTS Basidiomata and cultural morphology Basidiomata were only present in one location (Kericho) at >2180 m altitude during the rainy season. They appeared in clusters fused at the base. The pilei were 8.5-16.5 mm in diameter, convex, applanate to umbonate, with a non-striate margin, light ochraceous but dark-brown at the disk centre. The stipe was creamy­ white in colour, 45-50 x 3-6 mm with a whitish, fugacious annulus attached to the upper quarter of the stipe. Basidia were 30-35 x 6-7 Jlm, elongate clavate with four sterigmata. Lamellae were white in colour. The basidiospores were sub-globose to ovoid, 4.5-7.5 x 4-6.5 Ilm. Clamp connections were absent. Only very immature basidiomata were obtained in the in vitro attempts. The isolates were separated into two groups based on morphology in culture. Group l consisted of isolates whose colonies were mainly rhizomorphic and only mycelium was observed at the centres. Group II consisted of isolates which had raised mycelial colonies with submerged rhizomorphs. Somatic incompatibility A reaction was considered incompatible when a pigmented line was observed at the interfaces and compatible when the colonies merged. Two different groups were found which corresponded with the two morphological groups. Isolates from group l were compatible among themselves and incompatible with group II. Isolates from group II were incompatible both with isolates from group land with each other. Isolates from group II when paired against each other showed a clear and distinct pigmented line consisting of melanized hyphae. Phylogeny and Taxonomy 3 PCR-RFLPs The TTS region of isolates from group 1 by cultural morphology was amplified and a PCR product of about 700 bp was obtained. A PCR product of about 900 bp was amplified for isolates in group II. The IGS region of isolates in group 1was amplified with the primers P-l and 5S-2B and gave a PCR product of over 1000 bp. The IGS region of isolates in group Il was amplified with the primers LR12R and 0-1 and gave a PCR product of about 800 bp. A similar band was obtained for the isolates of A. mellea K5, K8 and sn and isolates KlO and K12. The digestion of the TTS of group 1 with the restriction enzyme HinfI gave for group 1 a pattern of about 220, 190, 170 and 72 bp and for group II a pattern of about 360, 230, 150, 100 bp (Figure 1a). Figure 1 (a, b). Digestion ofITS region with HinfI (fig la) and with Nde Il (fig lb) on 3% agarose gel stained with ethidium bromide. Lanes 2-6: isolates from group II. Lanes 7-17: isolates from group 1. A 100bp ladder was used as a size marker in lane 1 and lane 18. 4 ----------------------------- Phylogeny and Taxonomy The digestion of the ITS with restriction enzyme Alu 1 gave for group 1 a pattern of about 480, 160, 85 bp and for group II a pattern of about 510, 225, 95 bp. The digestion of the ITS with restriction enzyme Nde II for group 1 gave a pattern of about 390, 250 bp and for group II a pattern of about 590, 270 bp (Figure 1b). The digestion of the IGS with the restriction enzyme Alu 1 gave three different patterns, for group 1 a pattern of about 380, 245, 135 bp was obtained, for group II a pattern of 310, 220, 135 bp was obtained and for A. mellea a pattern of about 310, 170, bp was obtained. Iso1ates K10 and K12 had similar restriction patterns that group II (Figure 2). Figure 2. Digestion of IGS region with Alu I. on 3% agarose gel stained with ethidium bromide. Lanes 2-4: isolates K5, K8 and sn. Lanes 5-9: isolates from group II. Lanes 10-11: isolates KlO and K12. Lanes 12-16: isolates from group I. A 100 bp ladder was used as a size marker in lane 1 and lane 17. The phylogenetic studies of the IGS region were performed only for group II. The results showed that aIl the isolates from group II and isolates KlO and K12 formed one clade (100% Jackknife support) that was separated from the group of A. mellea also supported by a 100% Jackknife value. The isolate K5 (A. mellea from Kenya) was grouped with isolates of A. mellea from Japan and South Korea supporting Ota et al. (2000) theory. The clade support for these clusters had 100% Jackknife value. DISCUSSION Cultural morphology and somatic incompatibility separated the 47 isolates into two groups. Group 1 with rhizomorphic colonies and group II with mycelial colonies and submerged rhizomorphs. Molecular data based on the US and IGS regions separated the isolates into the same two groups. Basidiomata were only found once and they corresponded to group I. The description of the basidiomata conforms to that of A. heimii (Heim 1963; Pegler 1977) except for the stipe size which was slightly larger compared to the original description (2.5-4.5 x 2-3 mm). The cultural morphology of the isolates in this group also resembles A. heimii (Mwangi, pers. comm.). The amplification of the US region by PCR gave a product of about 700 bp that was similar to the size described by Chillali et al. (1997) for A. heimii and the restriction pattern with the enzyme Nde II was similar, but the restriction patterns with the enzymes Alu 1 and Hinf 1 were completely different from the ones obtained for A. heimii. The IGS region of group 1 was amplified and a band of over 1000 bp was obtained. This PCR product was different in size from A. heimii from other African countries. The digestion with the restriction enzyme Alu 1 gave a different pattern to the A. heimii but was similar to the one obtained by Coetzee et al. (2000) and identified as Armillaria sp. or A. fuscipes. Even though the morphological Phylogeny and Taxonomy 5 data point to A. heimii, molecular data suggested that there are sub-groups within A. heimii or they are a complex of several species. No basidiomata were found in nature for group II. This group was different from group 1 and different from A. mellea. This was supported by the PCR-RFLP results obtained for the ITS and IGS regions. Isolates from this group were identical to isolates KlO and K12 previously described as potential new Armillaria species (Chillali et al. 1997). The phylogenetic analysis showed that the IGS region between the 26S and 5S of the isolates from group II was identical to the IGS region of isolates KI0 and K12 and different from A. mellea and other Armillaria species. The IGS sequence of group II was different from any other Armillaria sequence published in GenBank. The data suggest that group II could be a new species but basidiomata are essential for the complete description and naming of the species. So far only very immature basidiomata have been produced in the in vitro attempts. Somatic incompatibility is one of the methods that has been used for the identification of genotypes and the incompatible reaction is characterized by the presence of a black line along the demarcation zone. The isolates from group II showed this reaction when paired against each other. The black line is usually absent in pairings between two genotypes of the same species (Guillaumin et al. 1991). This phenomenon has been reported for African Armillaria by Abomo-Ndongo (1997). A similar phenomenon has been observed for Ganoderma in oil palms where most isolates, even when taken from the same plant, were somatically incompatible with one another (Miller et al. 1999). Care should therefore be taken in interpreting somatic compatibility tests aimed at delimiting species in Armillaria. It can be concluded from this study that two different Armillaria species were found affecting tea plantations. One is suspected to be A. heimii and the other a possibly new Armillaria species. It was surprising that no isolates conforming to A. mellea were found and this may be an indication that its importance in Africa has been overestimated. More research is needed to resolve the taxonomy of Armillaria species in Kenya. ACKNOWLEGEMENTS We thank Dr J.-J. Guillaumin and Mr P. Desray of INRA Clermont-Ferrand, France, for their generous donation of isolates and for their helpful advice and support and Dr M. Coetzee from the Forestry and Agricultural Biotechnology Institute (FABI), South Africa, for his generous help. We also thank Dr C. Prior, Dr B. Henricot, Dr C. Gorton of the Royal Horticultural Society, UK for their advice and Dr J. Nicklin and Prof. Bridge ofBirkbeck College (London, UK) for their helpful comments. REFERENCES Abomo-Ndongo, S.; Guillaumin, J-J. 1997. Somatic incompatibility among African isolates. European Journal of Forest Pathology 27: 201-206. Agustian, A.; Mohammed, c.; Guillaumin, J-J.; Botton, B. 1994. Discrimination of sorne African Armillaria species by isozyme electrophoresis analysis. New Phytologist 128: 135-143. Anderson, J.B.; Petsche, D.M.; Smith, M.L. 1987. Restriction fragment polymorphisms in biological species of Armillaria. Mycologia 79: 69-76 Anderson, J.B; Stasovski, E. 1992. Molecular phylogeny of northern hemisphere species of Armillaria. Mycologia 84: 505-516. Chillali, M.; Idder-Ighili, H.; Guillaumin, J-J.; Mohammed, c.; Botton, B. 1997. Species delimitation in the African Armillaria complex by analysis of the ribosomal DNA spacers. Journal of General Applied Microbiology 43: 23-29. Coetzee, P.A.; Coutinho, T.A.; Wingfield, M.J. 2000. Identification of the causal agent of Armillaria root rot of Pinus species in South Africa. Mycologia 92: 777-785. 6 ------------------------------ Phylogeny and Taxonomy Cotzee, M.P.A.; Wingfield, B.D.; Wingfield, M.J.; Coutinho, T.A. 1997. Identification of the causal agent of Armillaria root rot in South African forest plantations. In Proceedings of the Ninth International Conference on Root and Butt Rots (C. Delatour, J-J. Guillaumin, B. Lung-Escannant and B. Marçais, eds): 49-61, INRA, Paris. Duchesne, L.C.; Anderson, J. B. 1990. Location and direction of transcription of the 5S rRNA gene in Armillaria. Mycological Research 94: 266-269. Gibson, I.A. S. 1960. Armillaria root rot in Kenya pine plantations. Empire Forestry Review 39: 94-99. Guillaumin, J-J.; Anderson, J.B.; Korhonen, K 1991. Life cycle, infertility and biological species. In Armillaria root disease. Agriculture Handbook No. 691: 10-20. United States Department of Agriculture. Washington D.C.. Harrington, T.c.; Wingfield, B.D. 1995. A PCR-based identification method for species of Armillaria. Mycologia 87: 280-288. Heim, R. 1963. L'Armillariella elegans Heim. Revue de Mycologie XXVIII: 89-94. Hopkin, A.A.; Mallet, KI.; Blenis, P.V. 1989. The use of L-DOPA to enhance visualization of the 'black line' between species of the Armillaria mellea complex. Canadian Journal of Botany 67: 15-17. Korhonen, K 1978. Interfertility and clonai size in the Armillaria mellea complex. Karstenia 18: 31-42. Maddison, W.P., Madisson, D.R. 1992. MacClade: analysis of phylogeny and character evolution. Version 3.0. Sinauer Associates, Saunderland, Massachusetts. Mallet, K.I.; Hiratsuka, Y. 1986. Nature of the 'black line' produced between different biological species of the Armillaria mellea complex. Canadian Journal of Botany 64: 2588-2590. Miller, R.N.G.; Holderness, M.; Bridge, P.D.; Chung, G.F.; Zakaria, M.H. 1999. Genetic diversity of Ganoderma in oil palm plantings. Plant Pathology 48: 595-603. Mohammed, C. 1994. The detection and species identification of African Armillaria. In Modern Assays for Plant Pathogenic Fungi; Identification, Detection and Quantification (A. Schot, F.M. Dewey and R.P. Oliver, eds): 141-147. CABI International, Willingford. Mohammed, C. 1994. Ecology and pathogenicity of Armillaria in Kenya, Zimbabwe, and the Congo. In Proceedings of the Eighth International Conference on Root and Butt Rots (M. Johansson and J. Stenlid, eds): 34-44. IUFRO. Uppsala, Sweden. Mohammed, C.; Guillaumin, J-J. 1994. Armillaria in tropical Africa. In Aspects of Tropical Mycology (S. Isaac, J. C. Frankland and R. Watling, eds): 207-217. Cambridge University Press, Cambridge, UK. Mohammed, C.; Guillaumin, J-J.; Botton, B.; Intini, M. 1994. Species of Armillaria in tropical Africa. In Proceedings of the Eighth International Conference on Root and Butt Rots. (M. Johansson and J. Stenlid, eds): 402-410. IUFRO, Uppsala, Sweden. Mwangi, L.M.; Lin, D.; Hubbes, M. 1989. Identification of Kenyan Armillaria isolates by cultural morphology, intersterility tests and analysis ofisozyme profiles. European Journal of Forest Pathology 19: 399-406. Mwenje, E.; Ride, J.P. 1996. Morphological and biochemical characterization of Armillaria isolates from Zimbabwe. Plant Pathology 45: 1031-1051. Mwenje, E.; Ride, J.P. 1997. The use ofpectic enzymes in the characterization of Armillaria isolates from Africa. Plant Pathology 46: 341-354. Onsando, J.M.; Wargo, P.; Waudo, S.W. 1997. Distribution, severity, and spread of Armillaria root disease in Kenya tea plantations. Plant Disease 81: 133-137. Ota, Y., Intini, M., Hattori, T. 2000. Genetic characterization ofheterothallic and non-heterothallic Armillaria mellea sensu stricto. Mycological Research 104: 1046-1054. Pegler, D.N. 1977. A preliminary agaric flora of East Africa. Kew Bulletin Additional Series VI: 91-95. Pegler, D.N. 1986. Agaric Flora of Sri Lanka. Kew Bulletin Additional Series XII: 82-86. Petch, T. 1909. New Ceylon fungi. Annals of the Royal Botanic Gardens, Peradeniya IV: 39. Shaw, c.G.; Kile, G.A. 1991. Armillaria root disease. Agriculture Handbook No. 691. United States Department of Agriculture. Washington D.C.. Smith, M.L.; Anderson, J.B. 1989. Restriction fragment length polymorphism in mitochondrial DNAs of Armillaria: identification of North American biological species. Mycological Research 93: 247-256. Swofford, D.L. 1998. Phylogenetic analysis using parsimony 4.0b2 version. Sinauer Associates, Saunderland, Massachusetts. Tirro, A. 1991. Technica per la produzione in vitro dei carpofori di Armillaria. Micologia Italiana 3:73-77. Phylogeny and Taxonomy 7 VeIdman, G.M.; KIootwijk, J.; de Regt, V.C.H.F.; Rudi, R.J. 1981. The primary and secondary structure ofyeast 26S rRNA. Nucleic Acids Research 9, 6935-6952. White, T.J.; Bruns, T.; Lee, S.; Taylor, J. 1990. Amplification and direct sequencing of fungal ribosomaI RNA genes for phyIogenetics. In PCR protocoIs. A guide to methods and applications (M. A. Innis, D. H. GeIfand, J. J. Sninsky, and T. J. White, eds): 315-322. Academie Press, San Diego. 8 PHYLOGENETIC RELATIONSHIP AMONG LAETIPORUS SPP. IN JAPAN y. Ota and T. Rattori Forestry and Forest Products Research Institute, Ibaraki, 305-8687, Japan SUMMARY Two species and one variety of Laetiporus have been hitherto reported from Japan. Laetiporus sulphureus var. sulphureus auct Japan has a lemon yellow pore layer and yellow pileus surface, though it is distinct from the European form by its non-imbricated pilei and its southern distribution. Laetiporus sulphureus var. miniatus auct Japan has a white or a lemon yellow pore layer, pinkish orange pileus surface, and imbricated pilei. Laetiporus versisporus has semi-globose basidiocarps with abundant chlamydospores in the context when matured and usually does not produce hymenophores. In order to define the intra-generic taxa of Japanese and European Laetiporus spp., 38 Japanese isolates of Laetiporus spp. and five European isolates of L. sulphureus were analyzed for variation in the internai transcribed spacer (ITS) region of the nuclear ribosomal DNA. Phylogenic analysis of the ITS region sequences resulted in five groups: Group A) the Japanese isolates of L. sulphureus var. miniatus associated with hardwoods, which has a white pore layer, Group B) the European L. sulphureus isolates, Group C) the Japanese isolates of L. sulphureus var. miniatus associated with conifers, which has a lemon yellow pore layer, and Groups D) and E) the Japanese isolates of both L. sulphureus var. sulphureus and L. versisporus. These results suggest that 1) all the Japanese Laetiporus spp. are distinct from European L. sulphureus, 2) Japanese L. sulphureus var. miniatus in association with different host types (hardwoods vs. conifers) belong to distinct taxa, and 3) Japanese L. sulphureus var. sulphureus and L. versisporus belong to the same species with different morphology. Keywords: Internai transcribed spacer INTRODUCTION Laetiporus spp. occur worldwide from boreal to tropical zones and cause red-brown cubical heart-rot in the wood of many deciduous and coniferous trees. Recently, Laetiporus sulphureus (Fr.) Murr. sensu lato in North America has been shown to be a complex of at least four taxa based on incompatibility tests, restriction fragment length polymorphisms in the ITS region of the nuclear ribosomal DNA, macro morphological characteristics, geographical distribution, and host range (Banik and Burdsall 1999, 2000; Banik et al. 1998). In Europe, phylogenetic analyses based on sequences of ITS region of the nuclear ribosomal DNA indicate that L. sulphureus may be separated into at least two taxa in association with diferent host types (Rogers et al. 1999). In Japan, two species and one variety of Laetiporus spp. have been hitherto reported. Laetiporus sulphureus var. sulphureus auct Japan has a lemon yellow pore layer and yellow pileus surface, though it is distinct from the European form by its non-imbricated pilei and its southern distribution. Laetiporus sulphureus var. miniatus auct Japan has a white or a lemon yellow pore layer, pinkish orange pileus surface, and imbricated pilei. This fungus is mainly distributed in cool temperate to boreal areas of Japan. Laetiporus versisporus has semi-globose basidiocarps with abundant chlamydospores in the context when matured and usually does not produce hymenophores. Its basidiocarps are at first lemon yellow then turn white to brown. The objective ofthis study was to define the intra-generic taxa of Laetiporus spp. from Japan and Europe and based on sequence ofITS regions of the nuc1ear ribosomal DNA. Phylogeny and Taxonomy 9 MATERIALS AND METHODS Thirty-eight isolates of Laetiporus spp. collected throughout Japan, five isolates of European L. sulpureus, and two isolates of L. porutentosus were used in this study (Table 1). dT bilLa e . aettporus ISO ates use III t IS stu ly. Isolate Species Host Origin Source no. 10 L. sulphureus var. sulphureus Kouchi, lapan 16 L. sulphureus var. sulphureus Yamaguchi, lapan 20 L. sulphureus var. sulphureus Ehime, lapan 95 L. sulphureus var. sulphureus Quercus sp. Kumamoto, lapan 107 L. sulphureus var. sulphureus Hardwood Miyazaki, lapan 108 L. sulphureus var. sulphureus Miyazaki, lapan III L. sulphureus var. sulphureus Hardwood Miyazaki, lapan 118 L. sulphureus var. sulphureus Castanopsis cuspidata Kyoto, lapan 3 L. versisporus Castanopsis cuspidata var. Sieboldii Tokyo,lapan 4 L. versisporus Castanopsis cuspidata var. Sieboldii Chiba, lapan 13 L. versisporus Kagoshima, lapan 19 L. versisporus Mt. Baishanzu, China 65 L. versisporus Nata, lapan 67 L. versisporus Osaka,lapan 70 L. versisporus Castanopsis cuspidata Kyoto, Japan 72 L. versisporus Castanopsis cuspidata Kyoto, lapan 109 L. versisporus Hardwood Miyazaki, lapan 24 L. sulphureus var. miniatus Hokkaido, lapan 26 L. sulphureus var. miniatus Yamanashi, lapan 34 L. sulphureus var. miniatus Prunus Mume Hokkaido, lapan 1. Yamaguchi 36 L. sulphureus var. miniatus Prunus Sargentii Hokkaido, lapan 1. Yamaguchi 38 L. sulphureus var. miniatus Quercus mongolica Hokkaido, lapan 1. Yamaguchi 40 L. sulphureus var. miniatus Prunus salicina Hokkaido, lapan 1. Yamaguchi 84 L. sulphureus var. miniatus Quercus mongolica Shizuoka, lapan 94 L. sulphureus var. miniatus Quercus sp. Miyazaki, lapan 99 L. sulphureus var. miniatus Ehime, lapan 104 L. sulphureus var. miniatus Kumamoto, lapan 8 L. sulphureus var. miniatus Abiesfirma Chiba, lapan 15 L. sulphureus var. miniatus Yamanashi, lapan 27 L. sulphureus var. miniatus Yamanashi, lapan 28 L. sulphureus var. miniatus Aomori, lapan 31 L. sulphureus var. miniatus Picea Glehnii Hokkaido, lapan T. Yamaguchi 33 L. sulphureus var. miniatus Picea Glehnii Hokkaido, lapan T. Yamaguchi 35 L. sulphureus var. miniatus Picea Glehnii Hokkaido, lapan 1. Yamaguchi 41 L. sulphureus var. miniatus Abies sachalinensisi Hokkaido, lapan 1. Yamaguchi 73 L. sulphureus var. miniatus Abies sp. Yamanashi, lapan 87 L. sulphureus var. miniatus Yamanashi, lapan 44 L. sulphureus Taxus baccata Belgium C. Decock 45 L. sulphureus Prunus sp. Belgium C. Decock 47 L. sulphureus Malus sylvestris Belgium C. Decock 48 L. sulphureus Hardwood Germany C. Decock 50 L. sulphureus Salix sp. Belgium C. Decock 52 L. portentosus Nothofagus pumilio Argentina M. Rajchenberg 56 L. portentosus Eucaryptus macrororrhyncha Argentina M. Rajchenberg 10 Phylogeny and Taxonomy DNA was extracted using a CTAB (cetyltrimetylarnmorium bromide) procedure. The ITS regions of nuclear rDNA were amplified with PCR using primer ITS4 and ITS5 (White et al. 1990). PCR amplification was performed using Perkin-Elmer GeneAmp PCR Reagent Kit with Taq DNA Polymerase (TaKaRa, Japan). The thermal program was an initial denaturing at noc for 5 min, followed by 40 cycles at 55°C for 1 min (annealing), noc for 2 min (elongation) and noc for 1 min (denaturing). A final elongation was allowed for 5 min at noc. Sequences were determined in both directions with the same primers using a Perkin Elmer Applied Biosystems (Foster City, CA, U.S.A.) PRISM Ready Reaction DyeDeoxy Termitator Cycle Squencing kit and a Perkin Elmer Applied Biosystems Automated DNA Sequencer 310. Sequences were aligned using Clustal X (Jeanmougin et al. 1998) and adjusted manually. Phylogenetic ana1ysis of the aligned sequences was performed using both distance and parsimony methods. For distance analysis, the neighbor-joining method generated from HKY85 distances using PAUP 4.0b (Swofford 2001) was performed. The strength of the internai branches of the resulting trees was statistically tested by bootstrap analysis (Felsenstein 1985) from 1000 bootstrap replications. For parsimony analysis, the maximum-parsimony analysis was performed using PAUP 4.0b. 100 replicate heuristic searches were performed, each with a random taxon addion sequence, and TBR blanch swapping. These trees were rooted with the sequence of ITS of L. portentosus as out-group. RESULTS The ITSl, ITS2 and 5.8S gene were completely sequenced in both directions. There was sequence variation in the ITS sequences of Laetiporus isolates. By contrast, there was little variation in the sequences of the 5.8S gene. The nucleotide sequence data set obtained from the isolates in Table 1 gave a 591-nucleotide a1igned sequence, including sorne indels due to the variable nucleotide sequence of the ITS region. After excluding the inde1s, 513 aligned sites remained, of which 152 sites were variable. Fig. 1 shows a neighbor-joining tree. Laetiporus isolates used in this study were divided into five distinct groups; Group A) the Japanese iso1ates of L. sulphureus var. miniatus associated with hardwoods, which has a white pore layer, Group B) the European L. sulphureus isolates, Group C) the Japanese isolates of L. sulphureus var. miniatus associated with conifers, which has a lemon yellow pore layer, and Groups D) and E) the Japanese isolates of both L. sulphureus var. sulphureus and L. versisporus. Groups A, B, D and E were strongly supported by the bootstrap analysis (almost 100%), the bootstrap value ofthe group C was 78%. A similar tree topology was obtained by the maximum-parsimony method using PAUP4.0b (data not shown). Phylogeny and Taxonomy 11 1 Group B European L. sulphureus Group 0 Japanese "L. sulphureus var.sulphureus" and L. versisporus Group E Japanese "L. sulphureus var. sulphureus" and L. versisporus ;~ L. portentosus Group C Yellow pored form of "L. sulphureus var. miniatus" /00 61 48 44 98 45 50 47 15 8 27 28 41 73 87 31 33 35 III 108 109 20 19 16 13 10 95 107 65 67 70 72 118 104 40 26 84 Group A 99 ~: White pored form of 24 43 "L. sulphureus var. miniatus" 38 34 - 0.0/ substitutions/site Fig. 1. Neighbor-joining tree on distances derived from sequences of the ITS 1, ITS2 and 5.8S rRNA gene of Laetiporus spp. Distances were determined by HKY85 methods. Bootstrap values based on 1000 replications are placed by the nodes. DISCUSSION Phylograms for the ITS region separated Laetiporus isolates into five distinct groups. European form of L. sulphureus (Group B) was separated from all other Japanese Laetiporus spp. Therefore, Japanese Laetiporus spp. are distinguished from European L. sulphureus. Laetiporus sulphureus var. miniatus auct Japan was phylogenetically divided into two groups (A and C) in this study. Distinctive differences in morphological and ecological characteristics occurred between groups A and C. Group A consists of those with white pores occurring on living and dead trunks or logs and stumps of hardwoods. Group C includes the yellow pored forrn, which is mainly distributed in cool temperate to boreal areas in association with conifers. This group was considered to be identical with one of North American Laeriporus group (LIG III), which occur on conifers, fruiting on stumps and living and dead trunks and with a yellow pore layer (Banik and Burdsall 2000). Groups D and E include both L. sulphureus var. sulphureus auct Japan, and L. versisporus. Laetiporus versisporus is different from normal Laetiporus spp. by their abundant chlamydospores in the context when matured, and lack of hymenophores. However, the intermediate form of L. versisporus and L. sulphureus var. sulphureus auct. Japan has been col1ected from mainly southem part of Japan. Laetiporus versisporus is considered to be the ptychogasteric stage of L. sulphureus var. sulphureus auct Japan. There appears to be significant geographical differentiation between Groups D and E. The isolates belonging to Group E were 12 Phylogeny and Taxonomy collected from central part of Japan. The isolates belonging to Group D were collected from southem part of Japan. Morphological study and pairing tests to deterrnine the compatibility of these Laetiporus spp. in Japan are currently underway. REFERENCES Banik, M.T.; Burdsall. RH. Jr. 1999 Incompatibility between Laetiporus cincinnatus and L. sulphureus in culture. Mycotaxon 70: 461-469. Banik, M.T.; Burdsall. RH. Jr. 2000. Incompatibility groups among North American populations of Laetiporus sulphureus sensu lato. Mycologia 92: 649-655. Banik, M.T.; Burdsall, RR Jr.; Volk, T.J. 1998. Identification of groups within Laetiporus sulpureus in the United States based on RFLP analysis of the nuclear ribosomal DNA. Folia Cryptog. Estonica 33: 9-14 . Fe1senstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39: 783-791. Jeanmougin, F.; Tompson, J.D.; Gouy, M.; Higgins, D.G.; Gibson, T.J. 1998. Multiple sequence a1ignment with Clusta1 X. Trends biochem. Sci. 23: 403-405. Rogers, S.O.; Hordenrieder, O.; Sieber, T.N. 1999. Intraspecific comparisons of Laetiporus sulphureus iso1ates from broad1eafand coniferous trees in Europe. Mycol. Res. 103: 1245-1251 . Swofford, D.L. 2001. PAUP 4.0 Phylogenetic ana1ysis using parsimony beta version 8: Sinauer Associates, Inc. Sunderland, Massachusetts. White, T.J.; Bruns, T.; Lee, S.; Taylor, J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phy10genetics. In PCR Protocols: A guide to Methods and Application. Ed. M. A. Innis, D. H. Gelfand, J. J. Sninsky and T. J. White, pp. 315-322. Academie Press Inc. New York, NY. Phylogeny and Taxonomy 13 STUDIES IN POLYPORUS SUBG. POL YPORELL US: ON CONGRUENCE OF THREE BIOLOGICAL, MORPHOLOGICAL AND PHYLOGENETIC SPECIES D. Krüger *, K.W. Hughes, and R.H. Petersen University of Tennessee, Botany Departrnent, Knoxville, TN 37996-1100, USA. * email: dkrueger@utk.edu Dedicated to the victims of the brutal attack on the civilized world, Sept. 11, 2001. This paper appears in place of an invited oral presentation cancelled as a result of limitations on international air traveI. SUMMARY The white-rotting polypore genus Polyporus comprises several infrageneric groups or subgenera, including Polyporellus, which in tum contain several similar morphological species. Crosses using monokaryotic single-spore isolates of Polyporus arcularius, P. brumalis, and P. ciliatus obtained from different geographic locations failed to uncover cryptic biological species. ITS rDNA sequences provided additional information on the phylogeny ofthese closely related members of Polyporellus. Keywords: Basidiomycotina, compatibility, molecular phylogeny, ribosomal DNA INTRODUCTION The group Polyporellus contains, among a few other species, three morphotaxa widespread in the Northem Hemisphere: P. arcularius, P. brumalis, and P. ciliatus (Nuiiez and Ryvarden 1995). P. arcularius is nearly cosmopolitan, but does not extend into the boreal realm. For example, in Germany there is a northem limit reported by Conrad et al. (1995). P. brumalis is circurnhemispherical in Eurasia and North America, and P. ciliatus has been reported from temperate Eurasia. All three species have also been reported from South America (Popoff and Wright 1998 for P. arcularius and P. ciliatus, Nufiez and Ryvarden 1995: by virtue of accepting synonyms to P. brumalis and P. ciliatus from Spegazzini). Closely related to P. ciZiatus appears to be P. trich%ma, included in our phylogenetic analyses for outgroup purposes. Spanish and Costa Rican collections of P. arcularius have previously been shown to belong to the same intercompatibility group (Nufiez 1993). Likewise, Hoffmann (1978) reported collections of P. ciZiatus from Europe and North America to belong to one interfertile group and collections of P. brumaZis from Europe, North America, and Tndia to form another interfertile group. There has been nomenclatural confusion in P. arcularius, P. brumalis, and P. ciliatus, and names have been misapplied (see Jahn 1969; Kreisel 1963). Based on mating study results, Hoffmann (1978) assigned two German and one Canadian collection obtained as P. brumaZis to P. ciliatus. For Hoffmann (1978) all three taxa (with one strain of P. arcularius included) were mutually incompatible. A particular question is whether Polyporus ciliatus occurs in North America, as claimed by Hoffmann (1978) and herbarium labels at DAüM, or in South America, as indicated by Popoff and Wright (1998). Nunez and Ryvarden (1995) regarded P. ciZiatus as unknown from North America. We are also interested in addressing the phylogeny of Polyporellus, and whether different morphospecies correspond to single biological and phylogenetic species entities. 14 Phylogeny and Taxonomy Here we report cUITent progress of our research in Polyporellus as we collect the initial framework of data to answer the above problems, including an assessment of the suitability of the use ofnuclear ITS rDNA data for the questions of interest. MATERIALS AND METHODS Specimens, establishment and maintenance of cultures Our own collections were assigned field book numbers, annotated, dried for preservation, and accessioned into TENN (Holmgren et al. 1981). Phase contrast microscopie observations were undertaken after squash mounting of herbarium specimens in 3% w/v KOH and phloxin dye, at 400 x magnification. Identification based on morphological characters was accomplished with the aid ofkeys by Gilbertson and Ryvarden (1986 - 1987), Ryvarden and Gilbertson (1993 - 1994), and Nunez and Ryvarden (1995). In sorne cases, colleagues fumished spore prints from which cultures were obtained. Techniques for establishing monokaryotic single-basidiospore isolates (SBIs) were described by Gordon and Petersen (1991). Dikaryon cultures were established for a number of collections as described by Petersen and Hughes (1997). Monokaryon and dikaryon cultures were stored on agar disks of malt extract agar (MEA: 1.5% w/v Difco malt extract, 2% w/v Difco Bacto-agar, Nobles 1965) in sterile water in microvials (Burdsall AND Dorworth 1994). Occasional bacterial contamination was overcome by passage through MEA plates supplemented with 1.54 mM chloramphenicol (Calbiochem 220551) and 10.67 JlM streptomycin sulfate (Sigma S0890). In order to obtain lacking herbarium specimens and/or monokaryon cultures, fruiting of dikaryon cultures on Fagus, Acer, Juglans, or Betula wood chips followed the procedure of Psurtseva and Mnoukhina (1998). Additional dikaryotic cultures were obtained from culture collections, and specimens or DNA extractions from other sources below. The following section, ordered alphabetically by country of origin, is read as: COUNTRY. Infracountry. Notes. Collector and date. Field book number or other number obtained elsewhere / TENN number or other collection number if availab1e (GenBank number). Other acronyms: CBS = Centraalbureau voor Schimmelcultures, Utrecht, Netherlands. DAOM = "Department of Agriculture, Ottawa, Mycology" (National Mycological Herbarium, Ottawa, Canada). DSH = D. S. Hibbett collection number. DSMZ-H = cultures used by Hoffmann (1978) and kept at Deutsche Sammlung von Mikroorganismen und Zellkulturen GmbH, Braunschweig, Germany. a = University of Oslo herbarium, Oslo, Norway. SBUG-M = "Sektion Biologie Univ. Greifswald - Myze1pilze" (Univ. Greifswald, Germany, fungal culture collection). VT = "Virginia Tech". P. arcularius Batsch: Fr. AUSTRIA. Niederosterreich: Kaltenleutgeben. On Fagus. H Voglmayr Apr 04 1999. 10299/TENN58370 (SBI2: AB070865, SBI4: AB070866). / CANADA. Ontario: Rondeau Provo Park. On Tilia. R. G. Thorn May 22 1983. CBS 223.911 RGT830522/01 (AB070858). / COSTA RICA. Cartago: km 66 of Interamerican Highway. R. Petersen Jul 01 1998. 9473/TENN56447. / CHINA. Guizhou. R. Petersen Aug 31 1991. 4124ITENN50834 (SBI1: AB070863, SBI2: AB070864). / GERMANY. Meck1enburg-Vorpommem: Usedom. On Fagus. R. Bütow Jun 1997. SBUG-M1244 (fruited for obtaining specimen/spores) /TENN58529, 58569, 58588 (AB070861). / SOUTH AFRICA. Kruisfountein. On Olea. P. Talbot Nov 1955. DAOM 94067 (from PRE52) (AB070859). 1 USA. Florida: Welaka. On wood chips. E. Lickey Mar 19 2001. 10975/TENN58890. 1 USA. Louisiana: Baton Rouge. D. Sime May 21 1997. 9076/TENN54876. 1 USA. Louisiana: Lafayette. R. Petersen May 23 1997. 9101lTENN54925. / USA. Louisiana: West Feliciana. E. Lickey May 26 1997. 9214/TENN55948. 1 USA. Tennessee: Tremont. H. Voglmayr Apr 02 2000. 10929/TENN58412 (SBIl: AB070867, SBI2: AB070868). 1 USA. Tennessee: Knoxville. H. Voglmayr Apr 04 2000. 10930/TENN58438. 1USA. Texas. On Pinus. J L. Mata Jun 10 2000. 10477/TENN58540. / USA. Texas. A. Methven Jun 10 2000. 106931TENN58779. / Extraction from D. S. Hibbett (Hibbett and Donoghue 1995). DSH92.144 (AB070860). 1Extraction from D. S. Hibbett (Hibbett and Vilgalys 1993). VT959 (AB070862). Phylogeny and Taxonomy 15 P. brumalis Pers.: Fr. AUSTRlA. I. Krisai-Grei/huber Sep 28 1996. 80371TENN55596. 1AUSTRlA. I. Krisai-Greilhuber Jun 10 1996. 8038/TENN55597. 1 AUSTRlA. Niederosterreich: Muckendorf. I. Krisai-Greilhuber and H. Vog/mayr Oct 04 1998. 10123/TENN57347. 1 AUSTRIA. Oberosterreich: St. Willibald. On Betu/a. H. Vog/mayr Oct 23 1999. 10666/TENN58382. 1 AUSTRlA. Oberosterreich: Kirchschlag. On Sorbus. H. Vog/mayr Nov 01 1999. 10667/TENN58383. 1 CANADA. Quebec: Kingsmere. J W Groves Oct 31 1955. DAOM 31983 (AB070869). 1 CANADA. Quebec: Kingsmere. J W Groves Oct 31 1955. DSMZ-HI7 (from DAOM 31983) (AB070870). 1 DENMARK. Roskilde Amt: Lellinge. On Fagus. H. Knudsen May 18 1999. 10169ITENN57700 (SBIl: AB070872, SBI3: AB070873). 1 DENMARK. Storstrems Amt: Fakse. On Fraxinus. R. Petersen May 19 1999. 10178/TENN57708. 1 GERMANY. Mecklenburg-Vorpommern: Malchow. On Fagus. D. Krüger May 09 1999. 10147/TENN57678. 1 GERMANY. Mecklenburg-Vorpommern: Neustrelitz. On Fagus. D. Krüger Dec 28 1999. 10908/TENN58391 (SBI4: AB070876, SBI5: AB070877). 1 NORWAy. Oslo. On Betu/a. D. Krüger Mar 25 2000. 10917/TENN58400. 1NORWAY. Telemark: Grasdalen. On Sorbus. A.-E. Torke/sen May 23 1972. 092301 (AB070871). 1RUSSIA. On A/nus. R. Petersen Sep 21 1996. 89711TENN55631. 1 USA. Alaska: Anchorage. On A/nus. R. Petersen Sep 09 1995. 7992/TENN53984 (was identified as P. ciliatus). 1USA. Alaska: Anchorage. On A/nus. R. Petersen Sep 11 1995. 8122/TENN53936 (was identified as P. ciliatus). 1USA. Tennessee. On Fagus. K. McFar/and Nov 07 1999. 10665/TENN58381 (SBII: AB070874, SBI2: AB070875). 1USA. West Virginia. A. Kova/enko Sep 30 2000. 10964/TENN58828. 1 USA. West Virginia. A. Kova/enko Sep 30 2000. 10965/TENN58827. P. ciliatus Fr. AUSTRlA. Niederosterreich: Hainburg. On Popu/us. I. Krisai-Greilhuber and H. Vog/mayr Apr 29 1999. 10300/TENN58371. 1 DENMARK. Ribe Amt: Billund. J Vesterholl May 14 1999. 10507/TENN57737. 1 DENMARK. Roskilde Amt: Lellinge. R. Petersen, H. Knudsen and D. Krüger May 18 1999. 101651TENN57696. 1DENMARK. Roskilde Amt: Lellinge. On Acer. H. Knudsen and R. Petersen May 18 1999. 10167ITENN57698 (SBI9: AB070882, SBIIO: AB070883). 1 DENMARK. Roskilde Amt: Lellinge. On Fagus. H. Knudsen and D. Krüger May 18 1999. 10168/TENN57699. 1 GERMANY. Baden-Württemberg: Tübingen. On Betu/a. D. Krüger May 31 1999. 10521ITENN57751. 1GERMANY. Mecklenburg-Vorpommern: Malchow. On Quercus. D. Krüger May 05 1999. 10149/TENN57680. 1 GERMANY. Mecklenburg-Vorpommern: Zislow. On buried Betu/a wood. D. Krüger May 05 1999. 101511TENN57682. 1GERMANY. Mecklenburg-Vorpommern: Neustrelitz. On Fagus. D. Krüger May 13 1999. 10156/TENN57687. 1GERMANY. Mecklenburg-Vorpommern: Greifswald. On Fagus. M. Scholler and D. Krüger May 22 1999. 101811TENN57711. 1 GERMANY. Mecklenburg-Vorpommern: Carpin, Nature Reserve Serrahn. R. Petersen May 25 1999. 10318/TENN57966. 1 GERMANY. Mecklenburg­ Vorpommern: Zwenzow, Krummer See. D. Krüger May 26 1999. 10320/TENN57968. 1 GERMANY. Mecklenburg-Vorpommern: Carpin, Nature Reserve Serrahn. On Fagus. D. Krüger May 25 1999. 10508/TENN57738. 1FINLAND. EteHi-Hiime: Evo. R. Petersen Sep 13 1994. 7480 (fruited for obtaining spores to make up for lost monokaryons) ITENN53639, 58441, 58823 (SBI2: AB070880, SBI3: AB070881). 1 SWEDEN. Uppland: Tarnby Lund. On Betu/a. R. Petersen Sep 04 1994. 7257/TENN53619 (SBI2: AB070878, SBI7: AB070879). P. tricholoma Mont. (included in phylogeny) COSTA RICA. Heredia: Chilimate. On buried wood. R. Petersen Mar 13 1999. 10240/TENN57563 (AB070885). 1 COSTA RICA. Heredia: Chilimate. R. Petersen Mar 13 1999. 10241ITEN 57564 (AB070888). 1 MEXICO. Chiapas. Small dead angiosperm branchlets. R. Petersen Oct 18 1997. 3870/TENN55844 (AB070884). 1 USA. Puerto Rico: Palmer. Hardwood log. R. Petersen Jun 09 1998. 9568/TENN56481 (AB070887). 1 USA. Puerto Rico: Palmer. Hardwood log. R. Petersen Jun 12 1998. 95911TENN56503 (AB070886). P. a/veo/aris (De.: Fr.) Bondartsev and Singer (Favo/us group; outgroup in phylogeny) Extraction from D. S. Hibbett (Hibbett and Vilgalys 1993). DSH 90.36 (AB070828). 16 Phylogeny and Taxonomy Mating experiments Self-cross pairings for deterrnination of mating types and tester strains were made among 12 randomly selected monokaryotic SEls of at least one collection per species. Tester strains are SEls with known mating type assigned for later testing of new arrivaIs. Subtester strains are auxiliary tester monokaryons selected later from collections of a different geographic origin, without prior knowledge of the actual mating type of the SB!. The technique for self-crosses was described by Petersen (1992). For intercollection pairings, four SBIs (randomly selected, or testers/subtesters when assigned) were paired with four SBIs (randomly selected, or testers/subtesters when assigned) from other collections in this study. DNA extraction DNA was extracted using a CTAB method (either modified from Carlson et al. 1991, see Hughes et al. 1999; or modified from Doyle and Doyle 1987; Zolan and Pukkila 1986; see Krüger et al. 2001), SDS-based method (Lee and Taylor 1990), or more recently, with a xanthogenate/SDS miniprep protocol modified from Tillett and Neilan (2000). AlI methods gave approximately equivalent results and were not always effective. In the CTAB methods, lOto 50 mg of herbarium material, or less of hyphal material scraped off MEA plates, was placed in a 1.5 ml reaction tube with 0.5 ml prewarrned (65°C) CTAB extraction buffer (O.lM Tris, 0.2 M NazEDTA, 1.5 M NaCl, 55 mM CTAB = hexadecyltrimethylammonium bromide, Sigma H5882). The material was incubated for 30 to 60 min at 65°C in a 1.5 ml tube and shaken occasionally before being ground with a sterilized plastic mini pestle (Kontes Pellet Pestle®, Kontes 749520). For tough basidiocarps of herbarium specimens, grinding was supported with sterile sand, and the material repeatedly frozen and heated (three cycles of 10 min at -80°C / 10 min at 65°C (Vralstad et al. 2000). Here, a 3% w/v SDS (sodium dodecyl sulfate, Sigma L4509), 1% w/v mercaptoethanol extraction buffer (500 Ill, 65°C) was used instead of CTAB extraction buffer. With either extraction procedure, approximately the same volume of 24: 1 chloroforrn:isoamyl alcohol was added, the mixture vortexed briefly, and spun at 12,000 rpm for 4 min. The upper, clear phase was transferred to a new 1.5 ml reaction tube and mixed with an equal volume of isopropyl alcohol (4°C). The reaction tube contents were mixed, refrigerated (4°C) for one hour, and centrifuged for 4 min at 13,000 rpm. The resulting pellet was rinsed twice in 70% v/v ethyl alcohol, air-dried, and resuspended in 100 l.tI TE buffer (10 mM TrisHCl, 1 mM NazEDTA; pH 8.0). DNAs were also extracted from fungal tissue using a modified xanthogenate protocol (Tillett and Neilan 2000). Fresh material was soaked for several weeks at 4°C in CTAB / sodium azide preparations (after Rogstad 1992: 6 M NaCI, 3 mM NaN}, 41.1 mM CTAB). Alternatively, fresh or herbarium material was stored in SDS buffer (50 mM Tris/HCl, 50 mM NazEDTA, 10% w/v SDS, pH 7.2). Ten to 50 mg of such material was ground in 50 III TE buffer with a small amount of sterile sand in a 1.5 ml microfuge tube as described above. For material grown 2 - 8 weeks at room temperature in malt extract (ME) broth (ca. 10 ml 1.5% w/v Difco malt extract, in baby food jar), 250 - 500 mg of material was filtered, blotted dry, then ground. After addition of 50 III TE extraction buffer, a minipestle mounted on a drill was used to grind the material. Following grinding, 750 III of potassium ethyl xanthogenate buffer (100 mM Tris/HCl pH 7.2, 20 mM NazEDTA pH 8.0, 1% w/v SDS, 800 mM ammonium acetate, 1% w/v C}HsKOSz = potassium ethyl xanthogenate: Fluka 60045) was added. The tube contents were vortexed, and incubated at 70°C for 60 min, with occasional vortexing. After a final, vigorous vortex for 10 sec, samples were placed on ice for 30 - 60 min, and then centrifuged at 14,000 rpm for 10 min. The supematant was recovered, and DNA precipitated with 750 III isopropyl alcohol (80% v/v, 4°C), and spun at 10,000 rpm for 10 min. The alcohol was aspirated, and the pellet was washed with 250 III 95% v/v cold ethyl alcohol, spun again at 10,000 rpm for 10 min. Remaining alcohol was then pipetted off, removed with a piece of paper towel, and evaporated 2 min at 70 oC on a heating block. The pellet was then resuspended in 50 III TE buffer supplemented with 2 III RNAse Plus (5 Prime - 3 Prime, Inc., now Eppendorf-5 Prime, Ine.), and ineubated 10 min at 50°C. Phylogeny and Taxonomy 17 PCR and sequencing DNA selected for sequencing was primarily monokaryotic in origin. For tester and subtester isolates we attempted to sequence two monokaryons for each collection. The nuclear ribosomal ITS 1 - 5.8S - ITS2 region was amplified with primers ITS 1-F and ITS 4-B (Gardes and Bruns 1993). A 50 ~1 reaction contained 1 X buffer supplied by manufacturer (includes MgCh in case of QBioGene polymerase kit, or separate 3 mM MgCh), 0.8 mM each dNTP, 0.2 ngl~l of bovine serum albumin (Sigma A7906), 0.2 mM primer each, 1.1 units Taq Po1ymerase (QBioGene EPTQA023; for difficult reactions: TaKaRa ExTaq ™ kit used after manufacturer's instructions, PanVera). Parameters were as follows: initial denaturation 94°C / 3 min, fOllowed by 35 cycles of denaturation 94°C / 1 min, annealing 52°C / 1 min, extension noc / 1 min. The final extension was 72°C / 3 min, followed by an indefinite storage step at 4oc. PCR products were electrophoresed in a 1.5% w/v agarose/TBE gel. In spite of secondary bands, amplified DNA product was not excised, but used after cleaning with a Microcon­ PCR device kit (Millipore), according to manufacturer's instructions. This seemed appropriate because disappearance after treatment with restriction enzymes Bfil (MEl Fermentas) and Neil (New England Biolabs) in Polyporus tricholoma proved those bands to be due to secondary structure. InternaI primers ITS 5 and ITS 3 (White et al. 1990) were used as sequencing primers. Cycle-sequencing was performed with the ABI Prism Dye Terminator Cycle Sequencing kit (Perkin-Elmer), following manufacturer's instructions and using approximately 200 ng DNA template. Sequence data Sequences were corrected using Chromas v. 1.45 (Conor McCarthy; Griffith University, Southport, Australia) for viewing and manipulating ABI electropherograms and Prograrnmer's File Editor v. 0.07.002 (Alan Phillips, Lancaster University, UK). Sequence alignment was done with ClustalX 1.64b (Thompson et al. 1997), followed by visual confirmation and manual optimization. The neighbor-joining (Saitou and Nei 1987) algorithm as implemented in ClustalX was used for initial phylogenetic analysis. A NEXUS file (Maddison et al. 1997) was manuallY generated for use in PAUP*4.0b8 (Swofford 2001). Gaps were coded as missing after trying alternative treatment modes. Heuristic searches were performed both in 20% deletion jackknife (Efron and Gong 1983) and bootstrap (Felsenstein 1985) resampling analyses (TBR swapping, MAXTREES set to 1,000, 100 random taxon addition repeats per resampling pseudoreplicate). TreeVIEW 1.5 (Page 1996) was used to view and manipulate trees. Sequences were submitted to EMBL/GenBanklDDBJ databases using the DDBJ Sakura submission system. RESULTS Mating studies and culture morphology We confirmed a tetrapolar mating system for P. arcularius (Vandendries 1936a, Hoffmann 1978), P. brumalis (Vandendries 1936b, Hoffmann 1978) and P. eiliatus (Hoffmann 1978, Petersen et al. 1997). We selected the following tester strains for later research: P. arcularius field book number 10299: AlBI (SBIl), AIB2 (SBl4), A2BI (SBI3), A2B2 (SBI2); P. brumalis field book number 10908: AlBI (SBI1), AIB2 (SBI4), A2BI (SBI8), A2B2 (SBI5); and P. eiliatus field book number 10167: AlBI (SBIll), AIB2 (SBI9), A2BI (SBIlO), A2B2 (SBIl 2). Results of the intercollection pairing experiments are shown in Fig. 1B (lower triangle). 18 Phylogeny and Taxonomy 000 '"- o C) 0 O'l >t § § § § § § § § §! N § § § § § § § S'lI ~ § § >t § § § § § § § III § §! ,-J r-- 8 8........ '0 § § ro 88 .... d .... x ::l :§lQ§§§§§§iCl§ § ~ .. J ~§ X § X § >t '" >t >: >t >t X ~ ~>< § :< § X § X X >: :< ....: :.< :..: ~ ,---1 ro § § li' li' § § §I " §iQ,Q§§] r' 1 ,~ ~ 8 § § §I i-J O.05); species in other phyla showed no significant changes with increasing stump age. Most fungi isolated were identified at least to genus level (Table 1). Many Ascomycota and mitosporic fungi were found at only one sampling time. Heterobasidion annosum was the first hymenomycete found, initially isolated from stumps 7 days after cutting. Bjerkandera adusta and Melanotus proteus were established in stumps within 28 days, and other hymenomycetes colonised between this time and 16 months after cutting. 600 oc ~ ~ 500 o u ~ ·Ci 400 l: ::J.... Ô 300 ~ Z 200 E ::J l: iij 100-o 1- o o Basidiomycotina III Ascomyotina o Deuteromycotina o Mastigomycotina III Zygomycotina o Sterile mycelial spp. IIIVeasts o Myxomycotina o 7 days 28 days 12m 16 m 48/53 m Time since felling Figure 1. Total numbers of fungi in different phyla isolated from Sitka spruce stumps over 53 months from felling. Numbers ofyeast-like organisms and sterile mycelia are presented as separate groups. Table 1. Times of isolation of identified fungi from stumps of Sitka spruce. Isolate identification Basidiomycota Bjerkandera adusta Gleophyllum spp. Heterobasidion annosum Melanotus oroteus Ecology and Biodiversity o 7 days Period of isolation 28 days 12 months 16 months 48/53 months 49 Period of isolation Isolate 0 7 days 28 days 12 months 16 months 48/53 months identification Peniophora pithya Polyporus abietinus Resinicium bicolor Sistotrema brinkmannii Stereum sp. Ascomycota Ascocoryne sp. Nectria fuckeliana Nectria inventa Mariannaea elegans Ophiostoma piceae Pseudoeurotium zonatum Mitosporic fungi Acremonium butyri Arthroderma sp. Aspergillus sp. Botrytis cinerea Cephalosporium (2 sp.) Cladosporium (4 spp.) Cryptosporiopsis sp. Cylindrocarpon sp. Dictyopolyschema pirozynskii Epicoccum sp. Graphium sp. Leptographium lunderbergii Paecilomyces elegans Paecilomyces farinosus Penicillium (3 spp.) Phoma (2 spp.) Trichoderma (3 spp.) Truncatella sp. Ulocladium chartarium Verticillium chlamydosporium Verticillium lecanii Verticillium nigrescens Verticillium nubilum Bacterial diversity: Bacteria were most abundant in the surface layers of stumps (15 mm) than deeper, although numbers found deeper within stumps gradually increased with stump age (Fig. 2). Numbers of CFUs isolated 50 Ecology and Biodiversity within stumps significantly reduced (P>O.OO 1) with stump age after year 3 (Fig. 3), irrespective of the position of isolation. The Shannon-Weaver index indicated that diversity differed significantly between samples taken from nearest the stump surface and those taken from deeper in the stumps (2-sample t-tests; p 5 mm) were determined using well­ separated 7 days old colonies. Degradative abilities tested were cellulase (Barlows 1992), 'ligninase' (Gold et al. 1988), amylase, pectinase and chitinase (Page et al. 1982). Catalase activity was tested using 3% H20 2 and observing for evolution of O2 bubbles. Oxidase (as cytochrome C) was detected using Kovacs' reagent. Based on the different characteristics determined, CFUs were clustered using the Euc1idean distance measure and the group average linking method. Gram positive and Gram negative CFUs were analysed separately. Groups were identified from clusters when 75% of the information remained. Interactions between bacteria and Heterobasidion annosum Interactions on Agar-based media: Petri dish cultures of H annosum were prepared on both sporulation agar (SA) and yeast-peptone-dextrose agar (YDPA; Benko and Highley 1992), co-inoculated with a bacterial isolate (100 tested in total) from stumps on, 4, 6 and 10 years old, and incubated in the dark at 20±2°C. Interactions in Liquid Culture: Fifteen bacteria inhibiting H annosum growth on SA were sub-cultured to 15 ml R2A broth and incubated in the dark at 20±2°C for 72 br. Following mixing, 5 ml of broth was transferred to a 50 ml jar containing 15 ml Norkrans' liquid medium (NLM: Norkrans 1963). Jars were inoculated with actively growing H annosum mycelium on 2% malt extract agar, sealed and incubated at 20±2°C for 21 days. Cultures were filtered through 2 layers of muslin cloth, washed with 50 ml distilled water and dried to constant weight at 105°C. Interactions in P. sitchensis Wood Cubes: Cubes (20 mm3 ) cut from green P. sitchensis timber, were weighed, placed in a 50 ml glass jar and 3 ml tap water added. Jars were sealed and autoclaved at 105 kPa for 1 hr. Ten wood cubes were re-weighed, dried at 105°C to constant weight and re-weighed to obtain an initial dry weight. Fifteen bacterial isolates showing antagonism against H annosum on SA were sub-cultured into 15 ml R2A broth. H annosum was sub-cultured from MEA to 15 ml NLM, incubated at 25±2°C for 10 days and cultures fragmented using a sterile Ultra Turrax homogeniser set at half-speed. Autoclaved wood cubes were inoculated with bacteria and fungi simultaneously, or 10 days apart, in all possible combinations, by placing 1 ml of each culture onto the wood. Controls cubes were not inoculated or, 56 Ecology and Biodiversity inoculated with bacteria or with H. annosum only. Jars were sealed using Parafilm M and incubated at 20±2°C for 140 days after inoculation with H. annosum. Wood cubes were then removed from the jars, rinsed in distilled water and dried to constant weight at 105°C. Percentage weight loss due to fungal or bacterial activity was calculated. Statistical Analysis: Results were analysed using the student's t-test, analysis of variance (ANGYA), Duncan's multiple range test and the X2 test. RESULTS Bacterial isolations from stumps Types of bacteria, based on characteristics: Proportions of CFUs able to hydrolyse cellulose increased significantly (P0.05; Fig. 3). H. annosum alone caused a 2.04% weight loss from cubes, a significant change (t-test; P<0.05) compared with controls. Inoculation of wood cubes with bacteria 10 days after inoculation with H annosum resulted in a significant increase (P0.05) occurred compared with cubes inoculated with H. annosum alone. DISCUSSION This work indicates the potential importance of bacteria in microbial community development in Sitka spruce stumps. Enumeration suggested that bacteria are present in large numbers during the first two years in the upper layers of the stump (Murray 1998; Woodward in press), and that the culturable population decreases markedly after this time, possibly due to the establishment of, and antibiotic production by, hymenomycete decay fungi in the substrate by this time. 58 Ecology and Biodiversity The abilities of the bacteria to degrade different substrates did not correlate strongly with stump age. Lignolytic activity, based on the use of the lignin-like substrate Remazol brilliant blue R, was absent in the isolates tested, although certain bacteria are capable of degrading this substrate (Singh and Butcher 1991). Numbers of isolates with cellulase activity, however, did increase with stump age, and was also correlated with wood moisture content (Murray 1998). Whether this pattern resulted from selection for cellulolytic bacteria under higher moisture conditions, or the higher moisture conditions followed the degradation of cellulose was not determined. Pectinase activity was detected in 3.2% of bacterial isolates from the stumps, although the specific substrate present can be of great importance in inducing this group of enzymes (Dunleavy et al. 1973). 9 7 -1 -3 a T~ .L Bacterial lsolate c ~~~~~~%~%~~%%%~~~ ~ ~ ~ 6 C' ~ ~ 3 6 (5 ~ ~ ~ ~ ~ ~o/· Bacterial isolate b Bacterial lsolate ct ~~~~~~%~%~~~%%~~~ ~ ~ ~ 6 C' ~ ~ 3 6 (5 ~ ~ ~ ~ ~ ~o/· Bacteri a1lsolate Figure 3. Weight loss of Sitka spruce wood cubes 140 days after inoculation with Heterobasidion annosum in combination with 15 bacterial isolates showing inhibition of H annosum on SA. Blocks were inoculated with (a) H annosum followed 10 days later by bacteria; (b) H. annosum and bacteria simultaneously; (c) bacteria followed 10 days later by H. annosum; (d) bacteria alone. Bars for controls (no inoculation) and H. annosum inoculation alone (Ha) are included on each graph. Values represent means ofthree replicates ± SEM. Amylase production may also be important for stump bacteria as starch is the major storage polysaccharide found in wood parenchyma cells (Zabel and Morrel 1992). Approximately 8% of bacteria isolated from stumps were amylase positive, and proportions varied with sturnp age, with high numbers in stumps of 1 and 9 years old. Chitin hydrolysis has a role in the biological control of fungi by bacteria, acting on the fungal cell wall (Whipps 1997). A significant positive correlation was found between the proportion of stump bacteria with chitinase activity and stump age, but the actual numbers of individual isolates found to produce chitinase, 2.1 % in total, was smal!. Under the inducing conditions found in the stump, very different from those in axenic cultures, the production of requisite degradative enzymes may occur. However, it is likely that the use of specific substrates as Ecology and Biodiversity 59 sole carbon sources in in vitro tests will induce the production of the enzymes required to degrade that substrate, if the organism is competent. Siderophores were produced by 29.3% of bacteria tested, but decreased in bacteria from older stumps. There were also significant correlations with position in the stump, and with heartwood vs sapwood isolates (Murray 1998). Production of chelating agents by bacteria commonly occurs under iron deficient conditions, and has been implicated in competition between bacteria and fine root pathogens (Whipps 1997). In contrast, germination and growth of sorne fungi, including H annosum, can be markedly enhanced in the presence of siderophores (Blakeman 1982). The chelating ability ofbacteria found in young stumps may reflect a requirement to sequester nutrients present in low concentrations at this early stage in the degradation process. Four groups of bacteria were identified using c1uster analysis based on 13 characteristics. This method was suitable for a general discrimination between the bacteria, and population changes in the stumps. Greater numbers of characteristics are required to give accurate taxonomic discrimination (Goor et al. 1990). Proportions of the three groups of Gram negative bacteria varied significantly with stump age, whereas the proportions of the single group of Gram positive isolates remained similar in ail stump ages. Few studies have been published on bacterial successions in decaying woody debris in the forest. An appreciation of the numerous interactions that occur between microbial populations in a natural community is essential to an understanding of the functioning of that community (McInemey 1986; Rayner and Boddy 1988). Although there are difficulties inherent in predicting in vivo interactions based on in vitro assays, such tests provide rapid, simple and inexpensive methods for screening interactions between micro-organisms. There is a tendency for in vitro tests to favour species producing of siderophores or antifungal metabolites (Nicolotti and Varese 1996; Dumas 1992). Variations in the proportions of bacteria inhibiting H annosum with medium used to test the interaction c1early illustrate the difficulties in finding suitable media for in vitro studies to mimic natural environments. Liquid media enable immediate and intimate contact between interacting micro-organisms, and differing responses were found with the 15 bactèria which inhibited H annosum on SA, when tested in NLM. Greater access to available nutrients from throughout the medium may increase the importance of competition as a mechanism of antagonism in this environment. Three isolates active against H annosum in SA had Iittle effect on the fungus in NLM, however, possibly resulting from their relative abilities to compete with H annosum in an environment where diffusion of metabolites is relatively easy compared with solidified substrates (Begon et al. 1990). In such a situation, the species with the greater initial inoculum size may inhibit the growth of the competitor. Weight loss in wood cubes inoculated with H annosum was only 2% over the 140 days of incubation, and the effects of co-inoculation with bacteria varied, particularly in relation to the relative timings of the fungal and bacterial inoculations. Similar effects have been noted previously (Hulme and Shields 1972). However, both Henningsson (1967) and Greaves (1970) found bacteria inhibited decay when inoculated into wood between 2 and 21 days before inoculation with fungi. No attempt was made in the current work to re-isolate the bacteria from the wood cubes after incubation, and the lack of any inhibitory effect may have resulted from a failure of the bacteria to establish. Inoculation with bacteria 10 days after H annosum, however, resulted in weight losses over 200% greater than in control cubes. Similar enhancement of decay has been reported previously in tests using bacteria and wood decay fungi (Henningsson 1967; Blanchette and Shaw 1978), and may result in part from an increased availability of certain nutrients produced by the bacteria. Many decay fungi require thiamine and biotin in the external environment, and bacteria may provide these vitamins (Alexander 1977). Fungal cellulase activity also may be enhanced in the presence of bacteria, as the prokaryotes utilise carbohydrate breakdown products which could repress enzyme production (Henningsson 1967; Greaves 1970; Shortle et al. 1978). Given the recent reports citing the detection using molecular methods of high numbers of species of unculturable prokaryotic organisms in many environments (Amann and Luwig 2000), it is likely that tree stumps will also contain far more bacteria than recorded in the work described here. Tt is also probable that the 60 Ecology and Biodiversity unculturable prokaryotes present within the niche alter their irnrnediate micro-environment through the production of enzymes and different secondary metabolites. The results presented here and elsewhere in this volume (Woodward in press), however, emphasize the potential for bacteria to exert considerable influence on the development of the decay commumty in spruce stumps. REFERENCES Aho, P.E., Seidler, R.J., Evans, H.J. and Raju, P.N. 1974. Distribution, enumeration, and identification of nitrogen-fixing bacteria associated with decay in living white fir trees. Phytopathology 64: 1413-1420. Alexander, M. 1977. Introduction to Soil Microbiology. 2nd Edition. John Wiley and Sons, London. Amann, R. & Luwig, W. (2000). Ribosomal RNA-targeted nucleic acid probes for studies in microbial ecology. FEMS Microbiology Reviews 24:555-565. Barlows, A. 1992. The Prokaryotes, A Handbook on the Biology of Bacteria: Ecophysiology, Isolation, Identification, Applications. Vol. 1. 2nd edition. Springer-Verlag, New York. 1028 pp. Begon, M, Harper, J.L. and Townsend, C.R. 1990. Ecology: Individuals, Populations and Communities. 2nd Edition. Blackwell Scientific, Oxford. 945 pp. Benko, R. and Highley, T.L. 1992. Selection of bacteria for screening interaction of wood-attacking fungi and antagonistic bacteria. 1. Interaction on agar. Material und Organismen 25: 161-171. Blakeman, J.P. 1982. Phylloplane interactions. pp. 307-333 in Mount, M.S. and Lacy, G.H. (eds): Phytopathogenic Prokaryotes. Volume 1. Academic Press, London. Blanchette, R.A and Shaw, c.G. 1978. Associations among bacteria, yeasts, and basidiomycetes during wood decay. Phytopathology 68:631-637. Dumas, M.T. 1992. Inhibition of Armillaria by bacteria isolated from soils of the boreal mixedwood forest of Ontario. European Journal of Forest Pathology 22: 11-18. Dunleavy, J.A; Moroney, J.P.; Rossell, S.E. 1973. The association of bacteria with increased permeability of water-stored spruce wood. Record of the British Wood Preserving Association Annual Convention 127­ 148. Gold, M.G.; Glenn, J.K.; Alic, M. 1988. Use of polymeric dyes in lignin biodegradation assays. Methods in Enzymology 161 :74-78. Goor, M.; Kersters, K.; Mergaert, J; Ryckaert, c.; Swings, J.; Vantomme, R.; Van den Mooter, M.; Verdonck, L. 1990. Numerical analysis of phenotypic features. In: Klement, Z.; Rudolph, K.; Sands, D.C. (eds), Methods in Phytobacteriology. Akademiai Kiodo, Budapest. 145-153. Greaves, H. 1970. The effect of selected bacteria and actinomycetes on the decay capacity of sorne wood-rotting fungi. Material und Organismen 5:265-279. Hallaksela, A-M. 1977. Microflora isolated from Norway spruce stumps. Acta Forestalia Fennica 158: 50 pp. Hallaksela, A-M. 1984. Bacteria and their effect on the microflora in wounds of living Norway spruce (Picea abies). Communicationes Instituti Forestalis Fenniae 121: 25 pp. Hallaksela, A-M. and Salkinoja-Salonen, M. 1992. Bacteria inhabiting artificially inoculated xylem of Picea abies. Scandinavian Journal of Forest Research 7: 165-175. Henningsson, B. 1967. Interactions between micro-organisms found in birch and aspen pulpwood. Studia Forestalia Suecica 53: 1-31. Holdenrieder, O. and Greig, RJ.W. 1998. Biological Control. pp. 235-258 in: Woodward, S., Stenlid, J., Katjalainen, R. and Hüttermann, A (eds): Heterobasidion annosum: Biology, Ecology, Impact and Control. CAB International, Wallingford, New York. Hulme, M.A. and Shields, J.K. 1972. Interaction between fungi in wood blocks. Canadian Journal of Botany 50:1421-1426. Kallio, T. 1974. Bacteria isolated from injuries to growing spruce trees. Acta Forestalia Fennica 137: Il pp. McInerney, M.J. 1986. Transient and persistent associations among prokaryotes. pp. 293-338 in: Poindexter, J.S. & Leadbetter, E.R. (eds.) Bacteria in Nature V. 2. Methods and Special Applications in Bacterial Ecology. Plenum Press, New York. Murray, AC. 1998. The bacterial ecology of Sitka spruce stumps. PhD Thesis, University of Aberdeen. Ecology and Biodiversity 61 Nicolotti, G. and Varese, G.C. 1996. Screening of antagomstlc fungi against air-borne infection by Heterobasidion annosum on Norway spruce. Forest Ecology and Management 88:249-257. Norkrans, B. 1963. Influence of sorne cultural conditions on fungal cellulase production. Physiologia Plantarum 16:11-19. Page, A.L.; Millar, R.H.; Keeney, D.R. 1982. Methods of Soil Analysis. 2. Chemical and Microbiological Properties. 20d edition. American Society of Agronomy, Madison. 1159 pp. Rayner, A.D.M.; Boddy, L. 1988. Fungal Decomposition of Wood - its Biology and Ecology. John Wiley, Chichester. Reasoner, DJ.; Geldreich, E.E. 1985. A new medium for the enumeration and subculture ofbacteria from potable water. Applied and Environmental Microbiology 49: 1-7. Schwyn, B.; Neilands, J.B. 1987. Universa1 chemical assay for the detection and determination of siderophores. Analytical Biochemistry 160:47-56. Shortle, W.c.; Menge, J.A.; Cowling, E.B. 1978. Interaction of bacteria, decay fungi and live sapwood in discoloration and decay of living trees. European Journal of Forest Path010gy 8:293-300. Singh, A.P.; Butcher, I.A. 1991. Bacterial degradation of wood cell walls: a review of degradation patterns. Journal of the Institute of Wood Science 12: 143-157. Whipps, I.M. 1997. Developments in the biological control of soil-borne plant pathogens. Advances in Botanical Research 26: 1-135. Woodward, S. (in press). Biodiversity in Monocultures: The Sitka Spruce Stump. Proceedings of the lOlh IUFRO Conference on Root and Butt Rots. Quebec, September 2001. Zabel, R.A.; Morrel, J.J. 1992. Wood Microbiology: Decay and its Prevention. Academie Press, London. 476 pp. 62 Ecology and Biodiversity SWISS-STONE PINE TREES AND SPRUCE STUMPS MAY REPRESENT THE PRIMARY HABITAT FOR HETEROBASIDION ANNOSUMSENSU STRICTO IN WESTERN ITALlAN ALPS G. Nicolotti 1 , P. Gonthier1 , M. Garbelott02 , G.c. Varese3 , and G.P. Cellerino1 1University of Torino, Departrnent for the Exploitation and Protection of Agricultural and Forestry Resources (Di.Va.P.R.A.) -Plant Pathology, Via 1. da Vinci 44,1-10095, Grugliasco, ltaly 2 University ofCalifornia-Berkeley, Departrnent of Environrnental Science, Policy and Management - Ecosystem Sciences Division (ESPM-ES), Berkeley, CA 94720, U.S.A. 3 University of Torino, Departrnent of Plant Biology, Viale Mattioli 25, 1-10125, Torino, Italy SUMMARY In the Western Italian Alps, all three species of the forest pathogen Heterobasidion annosum coll. are present, and may even coexist in the same stand. While the presence of H. parviporum and H. abietinum can be easily correlated to the presence of their respective primary hosts, spruce and fir, the host/niche occupied by H. annosum sensu stricto (s.s.) still remains unclear. Although Scots pine, a major host for this fungal species in other parts of Europe, is abundant in the region, little or no evidence of disease caused by H. annosum is visible in this tree species. An analysis of H. annosum coll. species was perfonned in two natural mixed conifer forests using traditional isolation techniques and a novel direct molecular diagnosis from wood. In site 1 (Cogne, Aosta Valley), a subalpine stand of mixed spruce, larch, and Swiss-stone pine (Pinus cembra 1.), 16 naturally infected spruces and larches only yielded H. parviporum isolates, while a Swiss-stone pine was extensively colonized by both H. parviporum and H. annosum s.s. In site 2 (Charvensod, Aosta Valley), a stand very similar to the above one, an analysis of 15 spruce stumps yielded both H. parviporum and H. annosum s.s. isolates. At both sites Scots pine was absent. These results suggest that 1) Swiss-stone pines and larches may be infected by the locally predominant H. annosum species (i.e. H. parviporum); 2) Swiss-stone pine may be a primary host for H. annosum s.s.; and 3) spruce stumps may be colonized by both Heterobasidion species and may thus offer a suitable habitat for the survival of H. annosum s.s. in the region. To our knowledge, this is the first report of Heterobasidion annosum on native European Pinus cembra and of an adult pine tree to be contemporary colonized by isolates belonging to different species of H. annosum. Keywords: Heterobasidion annosum, Pinus cembra, Larix decidua, ISGs, ecology INTRODUCTION Subalpine forests are, despite their low tree species diversity, structurally and spatially highly heterogeneous. These forests consist of a mosaic of stands, tree groups, single trees and glades. From the boundaries of subalpine ecotones up to the timberline, trees are more and more influenced and controlled by a range of environrnental factors; the temperature is usually identified as the main one (Holtmeier 1993). In the Western Italian Alps (WIA), fir (Abies alba Miller) and especially spruce forests are severely affected by the root and butt rot agent Heterobasidion annosum (Fr.) Bref. coll. (Anselmi and Minerbi 1989, Capretti 1998, Cellerino et al. 1998). While in other parts of the world the damages caused by Heterobasidion tend to be smaller at higher altitudes (Korhonen and Stenlid 1998), in WIA they are locally relevant at high elevations (1800-2300 m a.s.l) as well (Gonthier 2001). Unpublished data point out that sorne subalpine forests in Aosta Valley (WIA) are infected up to 95% of the trees. No more information are available about the epidemiology of H. annosum coll. in subalpine forests. Norway spruce (Picea abies (L.) Karsten), European larch (Larix decidua Miller) and Swiss Stone pine (Pinus cembra L.) are the main components of the subalpine forests in WIA. Among these, larch and especially Ecology and Biodiversity 63 spruce are weIl documented H annosum coll. hosts (Korhonen and Sten1id 1998). Spruce is the main host for H parviporum in Europe (Korhonen et al. 1998) and strong host specificity of this fungus vs spruce has been recently confirmed in pure and mixed forests growing at lower elevations in WIA (Gonthier et al. 2001; Gonthier 2001). Besides, both H annosum sensu stricto (s.s.) and H abietinum have been occasionally found on spruce, mostly in Northem and Southem Europe respectively (Korhonen and Piri 1994, Vasiliauskas and Stenlid, 1998, Capretti et al. 1994). Heterobasidion annosum s.s. and H abietinum are known to infect, in the same areas respectively, L. decidua too (Stenlid 1987, Capretti et al. 1994, Vollbrecht et al. 1995), even if that coniferous species shouldn't be considered as a primary host for H annosum coll. (Korhonen and Stenlid 1998). Little is known about the incidence of the disease and the behavior of Heterobasidion on the Swiss Stone pine, since this tree species is rarely felled in WIA. The fungus has been reported on P. cembra only once, right in the WIA (Nicolotti et al. 1999), and it was typed as H parviporum. The finding of H annosum on P. cembra is an obvious cause for concem because 1) it is generally believed that this tree species is resistant to most decay agents; 2) the pathogen was causing an extensive butt rot in the infected tree, whereas it is normally reported as root rot agent vs other pine species; and 3) H parviporum is generally considered a pathogen of spruce trees and not pines. Recent completed studies on the epidemiology of H annosum coll. in WIA show that in the region the scenario is complex (Gonthier et al. 200 1, Gonthier 2001). Synthetically 1) aIl the three species of H annosum are present; 2) they may coexist in the same stand; 3) strong host specificity is showed by H parviporum and H abietinum vs their preferential hosts. While the presence of H parviporum and H abietinum can be easily correlated to the presence of their respective preferential hosts, the host/niche occupied by H annosum s.s. in that area still remain unclear. This fungus has been found in forests either with or without Scots pine (Pinus sylvestris L.), one of the major host for this fungus in other parts of Europe (Korhonen et al. 1998). In WIA, even in stands where both H annosum s.s and its putative preferential host are presents, no evidence of disease is visible in this tree species. Two different, but not mutually exclusive, hypotheses can explain the presence of H annosum s.s. in the WIA. 1) Scots pines are infected but largely asymptomatic; and/or 2) H annosum s.s. has adapted to different hosts and/or niches. The aims of our study in mixed spruce, larch and Swiss stone pine forests were 1) to describe the symptoms caused by H annosum on P. cembra; 2) to identi:fy the Heterobasidion species present in the infected pine and in adjacent trees, iii)- to investigate the patterns of colonization of individual fungal genotypes. Besides, an adjunctive objective was iv) to evaluate the raie of other hosts/niches in the survival of H annosum s.s. in the regIOn. MATERIALS AND METHODS Sites and stand descriptions Two experimental plots were located in two naturally regenerated subalpine forests, which lay one another in parallel mountain slopes, in Aosta Valley (WIA). Both forests face North in valleys with an alpine sublitoral climate (mean annual rainfall 750 mm). The sites are classified as Picetum subalpinum and the soils as Ochrepts/Umbrepts. The two forests are very similar, both as regards the tree species composition and the structural aspects, and they comprise several large even-aged groups of trees, and as a whole is uneven-aged. Spruce is the dominant species (about 60% of the canopy); larch and Swiss stone pine represent about 20% of the total number of trees. Sorne silvicultural characteristics of the investigated forests are shown in Table 1. Regeneration of P. cembra is present in groups, but the adult trees have been found to be coetaneous, ranging between 85 and 90 years of age. The dominant larches were about twice as old, while spruce trees belonged to different age classes, indicating a continuous regeneration of this species through time. 64 Ecology and Biodiversity Table 1. Characteristics of the forests investigated. Forests Elevation Spruce % Stone pine % Larch % Basal area Density Mean DH3 (m. a.s.l.) [butt rot [butt rot [butt rot (m2 ha-!) (trees DBH2 (m) incidence%] , incidence%] 1 incidence%] , ha-') (cm) Cogne 1,800- 60 15 25 25.48 512 25 26 (site 1) 2,050 [90] [?] [30] Charvensod 1,800- 60 20 20 26.31 400 29 24 (site 2) 1,900 [50] [?] [10] , This estimate was based on the frequency of stumps displaying the typical laminated white rot caused by H annosum coll. (following thinning in 1997-1999) 2 Diameter at breast height 3 Dominant height In site 1 we studied the symptoms caused by H annosum coll. on P. cembra and the patterns of colonization of fungal isolates present in the pine. At this site we also investigated the frequency of the disease in the surrounding trees, and the spatial distribution of individual H annosum coll. genotypes in spruce and larch trees - or just felled secondarily infected stumps - growing within 30 m from the infected pine. In site 2 we studied the frequency of the different species of H annosum coll. in spruce stumps. That epidemiological information, together with data obtained in the same stand typing spores collected by the wood disk exposure method (Gonthier et al. 2001), could be useful to understand sorne ecological behavior of H annosum s.s. in WIA. Site 1: investigations and sample collection from the infected Swiss stone pine tree and from the surrounding stumps and trees The infected Swiss stone pine tree, growing at 1900 m a.s.l,. was cut in May 1998 and it showed an extensive butt rot in the stem section. The tree, growing closed to an old larch, was 14 m tall and its DBH was 34 cm. Annual rings which were visible (not decayed) were counted and measured both upstream and downstream. To assess the probable age of the tree and if the fungal infection would have cause a reduction in tree growth, two cores were extracted by an increment borer upstream and downstream at 40 cm aboveground from each of 10 healthy P. cembra trees growing in the stand. Annual ring values of the infected tree and the healthy ones were then compared. The symptoms were carefully described and, to assess the extent of the decay inside the tree, the trunk was consecutively cut into 50 cm logs. A longitudinally cut was performed in the last log in which discoloration was noticed. From each stem log, a 2 cm thick disk was taken from the major section for isolations. The complete root system was excavated to check for the occurrence of root grafts between the infected pine and the surrounding trees and stumps. AlI the woody roots, down to a diameter of 0.5 cm, were measured (length) and consecutively numbered; root contacts were marked as weil. Roots were sampled by excising transversal disks. Where root contacts were observed, sampies were also taken from roots of surrounding trees/stumps. The P. cembra stump, including about 1.5 m of all its main roots, was dug out, transported to the laboratory, and consecutively cut by a circular saw into 3-4 cm thick disks. AlI trees and stumps (left in 1997 after a thinning) with diameter over 4 cm, growing in a range of 30 m from the infected pine, were carefully mapped. The DBH was measured for trees and estimate for stumps. Each tree was sampled by extracting 2-3 cores from the collar zone and each stump by cutting off its top portion to obtain a 6-7 cm-thick disk near the root collar. Basidiocarps were also collected and stored at 5°C. Ecology and Biodiversity 65 Disks and cores were sprayed with a benomyl solution (0.010 g benomyl, 500 III methanol, 1 1 sterile water) and incubated, according to their size, in plastic bags or in Petri dishes (15 cm diam.), containing damped filter paper, for about 10 days at room temperature. Colonies of H. annosum were recognized by their Spiniger stage. In the Swiss stone pine, any distinct area bordered by interaction zones on the disks was identified and numbered. Corresponding areas on the following disks were numbered according to previous disks. Isolations were made from any area on PCNB-based selective medium (Kuhlman and Hendrix, 1962). Isolates from basidiocarps were obtained from the context. The larch tree, growing closed to the pine (about 0.4 m to each other), was felled down. The stem section showed a typical Heterobasidion discoloration, and a disk was taken. As it was impossible to isolate the fungus, a wood block (1.5 x 1.5 x 1.5 cm), including the discolored area, was cut and frozen at -20°C before DNA extraction. Site 2: sampling of spruce stumps for species composition assessment Fifteen randomly selected healthy spruce stumps were sampled in July 1998 (Gonthier 2001). The trees were cut 1 yr before. A wood disk (2-3 cm thick) was taken about 15 cm below the top surface of each stump. Stumps showing visible wood discoloration at the collar, where discarded, to avoid isolation of colonies spreading from roots. The treatment of samples, the incubation period and the isolation techniques were as above. Mitochondrial and nuclear typing for the assessment of H. annosum species and somatic incompatibility tests. Ali isolates were checked for the presence/absence of clamp connections, that would indicate either heterokaryon or homokaryon, respectively. The following methods were employed to classify the isolates. First, a taxon-specifie competitive-priming (TSCP-PCR) (Garbelotto et al. 1996) in the mt LrRNA gene, as described by Gonthier et al. (2001). To type all three European species of H. annosum by a single PCR amplification and gel, the method was modified as follows. A mix of four primers (MLS, MLF, Mito 5 and Mito 7) was used for DNA amplifications, and PCR products were electrophorezed in 2.5% Metaphor agarose gels in 1 X TBE at 3 V cm- I for 3 hrs. Second, a PCR RFLP on ITS to distinguish between H. parviporum and H. annosum s.s., as published by Gonthier et al. (2001), and to verify the presence of hybrid heterokaryons. Results of molecular typing, both from mitochondrial and nuclear markers, were compared to check for the occurrence of nuclear-mitochondrial chimeras among species. Third, sexual compatibility tests with homokaryon testers of the three species (T4, T5, T6 - kindly provided by Dr. Capretti - , A2r, A27r and A66r) as described by Stenlid and Karlsson (1991). The Buller phenomenon was used to identify heterokayon isolates, and in that case the occurrence of clamp connections was verified in the tester thallus, at least 3 cm from the interaction zone. Direct DNA extraction from the wood of the larch was performed as follows. Sawdust from collected wood was obtained by drilling in the collected wood. The sawdust was placed in a 1.5 ml Eppendorf tube and DNA was extracted from the sawdust by adding an equal volume of 24:24: 1 phenol:chloroform:isoamyl alcohol and vortexing vigorously for 2 minutes. After 15 min centrifugation at 13000 rpm, the supernatant was collected and the extraction was completed using the GeneClean II Kit (QBiogene), following the manufacturer's instructions. DNA was diluted 1: 100 in PCR water and a nested PCR approach was utilized to obtain DNA amplification of Heterobasidion. It should be noted that the pathogen was no longer culturable and expected to be present in very small titer. A first round of PCR (using parameters described by Garbelotto et al. 1996) was performed with primers ITS-IF and ITS4 (Gardes and Bruns 1993; White et al. 1990). The PCR product was then diluted 1: 100 in PCR water and a second PCR amplification was performed using two internai primers (ITS 1 and ITS2) (White et al. 1990). Sequences of the complimentary strands of the PCR product were obtained using an ABI 377 automatic sequencer (Applied Biosystems, Foster City, CA). Sequence alignments were obtained by using Sequencher software (GeneCodes) and optimized by manual alignment. Using the consensus sequences, a BLAST search of the GenBank database (NCBI) was performed. 66 Ecology and Biodiversity To study the presence and the patterns of colonization of H annosum genotypes, aIl isolates from Cogne, and belonging to either of H annosum species, were paired at least twice in aIl possible combinations according to the method described by Stenlid and Karlsson (1991). RESULTS The infected Swiss stone pine tree was apparently in good health conditions. The color of the foliage and the crown density were similar to those of uninfected P. cembra growing in the stand. Seventy-five annual rings were visible on the pine stump, but it is likely that the tree was between 88 and 90 yrs old (age of sUITounding P. cembra trees). A great reduction in ring size was observed in the last 25 yrs, both vs rings of previous yrs in the same tree (0.72 vs 2.34 mmlyr) and vs the last 25 annual rings from sUITounding P. cembra trees (1.47 ± 0.02 mm1yr). Stem decay at breast height occupied about 29% of the stem section and it was confined to the heartwood. The butt rot was fibrous, soft, pale brown to dark brown and appears quite dry. Peripheral areas of the decay were stained either brown or lilac and were bordered bluish or grey. A hollow was noticed from the collar zone up to 2 m of height, while the fungus was isolated from disks taken up to 4 m above the collar. Decay was also present in the central cylinder of all the seven tree's main roots, down to a diameter of 7 cm. Wood rot and staining appeared similar to those found in the stem. The fungus has also been isolated from 30% of fine roots (0.5-7 cm), that did not show any visible symptom of decay. In that case H annosum was present on the whole section of the sampled disks. No H annosum mycelium was observed between the bark and the wood. AlI sampled disks yielded heterokaryon isolates. Isolates from the stem and from one root were typed as H parviporum and they ail belonged to the same genotype, whereas the other six main roots were colonized by a single H annosum s.s. genet. In the stump, a distinct dark line of about 2 mm width was present between wood occupied by either of the two fungal individuals. Root contacts were detected between infected fine roots of the pine and infected roots of two closed spruce stumps. Both stumps yielded H parviporum isolates, and one of these was from the same genotype of the pine, suggesting a direct tree-to-tree spread of the fungus. Of the 35 sampled sUITounding trees/stumps (27 spruce, 4 larch, 4 Swiss stone pine), the fungus was isolated from 17 (49%). No isolates were obtained from other surrounding Swiss stone pine trees. The fungus was present in 17 out of 27 sampled spruce tree/stumps (63% incidence) and it was detected also in the larch close to the infected pine. A single Heterobasidion basidiocarp was found in a decay pocket of a spruce stump. AlI isolates possessed clamp connections, ail were typed as H parviporum, and each tree/stump, except for the P. cembra tree, was colonized by a single Heterobasidion individual. The BLAST search resulted in a perfect match between the isolate amplified from larch wood and H parviporum. Among the H parviporum isolates, 14 genets were identified, and these occupied 18 trees (17 spruces and the infected pine) (average of 1.29 tree/stump per genet). The largest genet occupied three trees within a 7 m distance, while other two smaller genets occupied two trees each. Eleven genets (61.1 %) were detected in only a single tree. A single H annosum s.s. genotype was present, and it was confined to the Swiss stone pine tree. Twenty-five H annosum coll. isolates were collected from spruce stumps in Charvensod, and ail ofthem came from the sapwood: Il (46%) were typed as H parviporum and 13 (54%) as H annosum s.s. Three sturnps were colonized by isolates of both species of the fungus, and homokaryotic strains were more frequent than heterokaryotic ones. Nuclear PCR-RFLP typing indicated the absence ofboth heterokaryon hybrids and nuclear-mitochondrial chimeras. Ecology and Biodiversity 67 DISCUSSION Although H. annosum coll. has already been reported on pine species of the Section Cembra (Debazac 1977) before (Kolomiets and Bogdanova 1992, Arefev 1991,Darozhkin and Fedarau 1976, Negrutskii 1963), all such reports were from Eastern Europe and Siberia. These areas are well outside the natural range of P. cembra L., but correspond to the natural range of the closely related P. sibirica Du Tour (Tutin et al. 1993). While we cannot exclude that the pine species in those reports may have been P. cembra that had been planted outside its natural range, it is more likely that those reports referred to P. sibirica (Korhonen, personal communication). To our knowledge this is the first report of H. annosum on P. cembra L., at least in the natural range of this pine speCles. Heterobasidion spp. are generallY reported to kill pine trees, by infecting the cambium and then spreading into the root sapwood (Korhonen and Stenlid 1998). The infected Swiss stone pine examined in this study was apparently symptomless and displayed almost exclusively an internai decay. This unusual characteristics for a pine has already been reported about Heterobasidion infections right on trees of the closely related P. sibirica, suggesting that these two pine species are mostly susceptible of butt rot instead of root rot. Internai decay was also observed in the root cylinder, often just lirnited to the heartwood portion of the root. Internai decay of the roots and stem results in a significant loss of structural integrity, and may also, when a large section of the functional sapwood is colonized, lead to a physiologicalloss of vigor. Both H. parviporum and H. annosum s.s. were present in the Swiss stone pine tree, causing a similar heart rot whether in the roots or in the stem. Isolates collected from P. sibirica in Southern Finland by Prof. Laine were typed as H. parviporum (Korhonen, personal communication). In white fir stands in the United States, multiple H. annosum genets were commonly found in living trees (Garbelotto et al. 1999), but the simultaneous presence in single trees of strains belonging to different species of Heterobasidion have been reported, as far as we know, only twice, and those reports referred to P. abies (Vasiliauskas et al. 1998, C. Delatour, unpublished, in Delatour 1998). Therefore, this is the first report of both pathogen species in a single live pine tree. The presence of both fungal species may indicate that host specificity for these pathogens is not strict, but may be regulated by ecological conditions and site history. In our stand, H. parviporum was by far the predominant species, as exemplified by the observation that about 60% of spruces were infected by this fungal pathogen. Tt has already been shown that the predominance of a pathogen species in a host may result in high percentages of that species in adjacent hosts that may not be generally considered as common hosts for that pathogen (Capretti et al. 1994). This phenomena was verified while studying the Swiss stone pine, as we assessed a secondary vegetative spread of aH. parviporum isolate from a spruce to the pine (or vice versa). The same could be also true for the infected larch, as reported by Capretti et al. (1994) about infections ofthis host in areas characterized by high H. abietinum inoculum density. To our knowledge, this is the first report of aL. decidua tree to be infected by H. parviporum. Heterobasidion infections on larches were previously reported as caused either by H. abietinum or by H. annosum s.s., mostly in Southern and Northern Europe, respectively (Stenlid 1987, Capretti et al. 1994, Vollbrecht et al. 1995). In the stand at Cogne, the most common host for H. annosum s.s., Scots pine, is absent. In general, Pinus sylvestris and P. cembra do not coexist in the same sites in the Aosta valley and adjoining areas. Pinus sylvestris, in fact, is virtually absent at higher elevations, where P. cembra grows. In that stand, we found a single H. annosum s.s. genet, and it was exclusively associated to the Swiss stone pine tree, while all surrounding sampled trees were infected by H parviporum. This finding suggests that Swiss stone pine is virtually a primary host for H. annosum s.s. Both H. parviporum and H. annosum s.s. isolates have been found in the forest of Charvensod, and they may even coexist in the same stump. Data on the aiborne inoculum composition of Heterobasidion species in that forest have been recently published (Gonthier et al. 2001): 99% of spores were from H. parviporum and only 1% of spores were from H annosum s.s. Although further samplings are in need to correctly assess the relative abundance of the two species both in spruce trees and in spruce stumps, this study show a quantitative equilibrium between H. parviporum and H. annosum s.s. inside primarily infected spruce stumps (46% vs 54%). Resu1ts 68 Ecology and Biodiversity indicate that H parviporum is almost the only species present in the air and thus able to infect stump surfaces. We suppose that large fruit body production and the subsequent massive release of air spora normally occur after colonization of large volumes of substrate, and thus, as found in Cogne, spruce trees are mostly infected by H parviporum. The significant presence of H. annosum s.s. inside spruce stumps may be explained by a particular ecological fitness of this pathogen vs this substrate. H annosum s.s. has already been reported to display a saprotrophic degradating ability greater than H parviporum in spruce wood (Daniel et al. 1998). Thus, spruce stumps may represent, in areas were P. sylvestris is absent or resistant, an alternate and suitable habitat for the survival ofH annosum s.s. Heterobasidion parviporum and H annosum s.s. co-exist both in spruce stands (Gonthier et al. 2001) and in mixed forests growing at high elevations in the WIA. We show in this study that at least two niches are simultaneously shared by both species in subalpine forests. This finding may have sorne important implications in the population biology of this fungus, because of the essential role played by true niche overlaps in enhancing the potential for hybridization. Nevertheless, neither heterokaryon hybrids nor nuclear-mitochondrial chimeras between Heterobasidion parviporum and H. annosum s.s. were found in this study. Absence of inter-group gene flow between the two species has already been reported in the WIA (Gonthier et al. 2001). This finding supports the hypothesis that strong genetic barriers reinforced by negative selection on hybrids are the main factors involved in the lack of hybridization between the two populations in the region. ACKNOWLEDGEMENTS We are grateful to Dr. Kari Korhonen for his helpful information about the epidemiology of H. annosum on P. sibirica. We also thank the lumberjacks of the "Région Autonome Vallée d'Aoste", with particular regards to Mr. Cesare Bionaz. REFERENCES Anselmi, N.; Minerbi, S. 1989. Root rots involved in forest decline in Italy. In Proceedings of the 7th IUFRO Conference on Root and Butt rots. Canada, August 1988. Forestry Canada, British Columbia, Canada. Edited by D.l Morrison. pp. 503-512. Arerev, c.P. 1991. Xylotrophic fungi - the causal agent of Siberian pine (Pinus sibirica Du Tour) rot in the central taiga Irtysh River Basin. Mikol. Fitopatol. 25: 419-425 (in Russian). Capretti, P. 1998. Italy. In Heterobasidion annosum, Biology, Ecology, Impact and Control. Edited by S. Woodward, J. Stenlid, R. Karjalainen, and A. Hüttermann. CAB International. pp. 377-385. Capretti, P.; Goggioli, V.; Mugnai, L. 1994. Intersterility groups of Heterobasidion annosum in Italy: distribution, hosts and pathogenicity tests. In Proceedings of the 8th IUFRO Conference on Root and Butt Rots, Wik, Sweden and Haikko, Finland, August 9-16, 1993. Edited by M. Johansson and J. Stenlid. Swedish University of Agricultural Sciences, Uppsala, Sweden. pp. 218-226. Cel1erino, G.P.; Gonthier, P.; Nicolotti, G. 1998. Diffusione di Heterobasidion annosum su abete rossa in Valle d'Aosta ed interventi di lotta biologica e integrata. Secondo Congresso Nazionale di Selvicoltura per il miglioramento e la conservazione dei Boschi Italiani. Atti Convegno Interregionale Lombardia, Piemonte, Valle d'Aosta. Vercelli, 28 febbraio 1998. pp. 201-204. Daniel, G.; Asiegbu, F.O.; Johansson, M. 1998. The saprotrophic wood degradating abilities of Heterobasidion annosum intersterility groups P and S. Mycol. Res. 102: 991-997. Darozhkin, M.A.; Fedarau, V.M. 1976. Fungus diseases of conifers introduced in the Central Botanical Gardens of the Byelorussian Academy of Sciences. Biiyalagichnykh Navuk. 3: 47-50. Debazac, E.F. 1977. Manuel des conifères. Ecole Nat. des Eaux et Forêts. Nancy. Delatour, C. 1998. France. In Heterobasidion annosum, Biology, Ecology, Impact and Control. Edited by S. Woodward, l Stenlid, R. Karjalainen, and A. Hütterrnann. CAB International. pp. 369-376. Ecology and Biodiversity 69 Garbelotto, M.; Ratcliff, A.; Bruns, T.D.; Cobb, F.W.; Otrosina, W. 1996. Use of Taxon-Specifie Competitive­ Priming PCR to study host specificity, hybridization, and intergroup gene flow in intersterility groups of Heterobasidion annosum. Phytopathology 86: 543-551. Garbelotto, M.; Cobb, F.W.; bruns, T.D.; Otrosina, W.J.; Popenuck, T.; Slaughter, G. 1999. Genetic structure of Heterobasidion annosum in white fir mortality centers in California. Phytopathology 89: 546-554. Gardes, M.; Bruns, T.D. 1993. ITS primers with enhanced specifity for fungi and Basidiomycetes: application to the identification of mycorrhizae and rusts. Mol. Ecol. 2: 113-118. Gonthier, P. 2001. Studi sull'epidemiologia di Heterobasidion annosum nelle Alpi Nord-occidentali e indagini di lotta biologica e chimica. Ph D Thesis, University ofTorino. Gonthier, P.; Garbelotto, M.; Varese, G.c.; Nicolotti, G. 2001. Relative abundance and potential dispersal range ofintersterility groups ofHeterobasidion annosum in pure and mixed forests. Can. J. Bot. 79: 1057-1065. Holtmeier, F.K. 1993. The upper timberline: ecological and geographical aspects. In Ecologia delle foreste di alta quota, Proc. XXX Corso di Cultura in Ecologia, University of Padova. Edited by T. Anfodillo and C. Urbinati. pp. 1-26. Kolomiets, N.G.; Bogdanova, D.A. 1992. Diseases and pests of the forest stands ofNovosibirsk Scientific Centre of the Siberian Branch of the Russian Academy of Sciences. Sibirskii Biologicheskii Zhumal 4: 53-55 (in Russian). Korhonen, K; Piri, T. 1994. The main hosts and distribution of the Sand P groups of Heterobasidion annosum in Finland. In Proceedings of the 8th illFRO Conference on Root and Butt Rots, Wik, Sweden and Haikko, Finland, August 9-16, 1993. Edited by M. Johansson and J. Stenlid. Swedish University of Agricultural Sciences, Uppsala, Sweden. pp. 260-267. Korhonen, K.; Stenlid, l 1998. Biology of Heterobasidion annosum. In Heterobasidion annosum, Biology, Ecology, Impact and Control. Edited by S. Woodward, J. Stenlid, R. Karjalainen, and A. Hüttermann. CAB International. pp. 43-70. Korhonen, K.; Capretti, P.; Karjalainen, R.; Stenlid, J. 1998. Distribution of Heterobasidion annosum intersterility groups in Europe. In Heterobasidion annosum, Biology, Ecology, Impact and Control. Edited by S. Woodward, J. Stenlid, R. Karjalainen, and A. Hüttermann. CAB International. pp. 93-104. Kuhlman, E.G.; Hendrix F.F. Jr. 1962. A selective medium for the isolation of Fomes annosus. Phytopathology 52: 1310-1312. Negrutskii, S.F. 1963. Sorne features of the infection of Pinus sibirica by Fomes annosus. Lesnoi Zhurnal 2: 22­ 26 (in Russian). Nicolotti, G.; Gonthier, P.; Varese, G.c. 1999. First report of Heterobasidion annosum on native European Pinus cembra. Plant Disease 83: 4, 398 (Disease note). Stenlid, J. 1987. Controlling and predicting the spread of Heterobasidion annosum from infected stumps and trees of Picea abies. Scand. J. For. Res. 2: 187- 198. Stenlid, J.; Karlsson, lO. 1991. Partial intersterility in Heterobasidion annosum. Mycol. Res. 95,1153-1159. Swedjemark, G.; Stenlid, J. 1993. Population dynamics of the root rot fungus Heterobasidion annosum following thinning ofPicea abies. Oikos 66: 247-254. Tutin, T.G.; Burges, N.A.; Chater, A.O.; Edmondson, J.R.; Heywood, V.H.; Moore, D.M.; Valentine, D.H.; Walters, S.M.; Webb, D.A.1993. Flora Europaea (2nd edition), Vol. 1: Psilotaceae to Platanaceae. Cambridge University Press, Cambridge. Vasiliauskas, R.; Stenlid, l 1998. Spread of Sand P group isolates of Heterobasidion annosum within and among Picea abies trees in central Lithuania. Can. J. For. Res. 28: 961-966. Vollbrecht, G.; Johansson, u.; Eriksson, H.; Stenlid, J. 1995. Butt rot incidence, yield and growth pattern in a tree species experiment in southwestern Sweden. Forest Ecology and Management 76: 87-93. White, T.J.; Bruns, T.; Lee, S; Taylor, J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. Chapter 38. In PCR Protocols: a Guide to Methods and Applications. Edited by M. Innis, D. Gelfand, J. Sninsky, and T. White. Academie Press, Orlando, Florida. 70 Ecology and Biodiversity RELATIONSillP BETWEEN SOIL FACTORS, ROOT INFECTION BY COLLYBIA FUSIPES AND TREE HEALTH IN QUERCUS ROBUR AND Q. RUBRA C. Camy and B. Marçais Laboratoire de Pathologie Forestière, INRA, Centre de Nancy F54280, Champenoux, France SUMMARY A method of rating of C. fusipes root infection severity was used in 2000 in two mature pedonculate oak stands to investigate the relationship between C. fusipes infection severity, degree of oak decline and soil factors. This method was also used in young and mature red oak stands, successively in 1995 and in 2001, to investigate the effect of soil factors, such as water logging, and the effects of tree vigour, such as concurrence between trees, on the infection induction and evolution. Thus, four plots in France were selected with a minimum of 51 studied trees on each site. The results clearly showed that pedonculate oaks were more infected by C. fusipes when roots were less submitted to water logging and that their proportion of declining increased when they were more infected by the parasite. The level of decline was also influenced by the depth of a gravel layer able to reduce the vertical growth of roots, but C. fusipes as been shown to play the major role in the decline. Conceming the work on red oaks, water logging was also shown to negatively influence the infection induction and its evolution. The evolution of infections did not appear to depend on the age of trees or on their initial degree of infection. The period required for an healthy red oak to become heavily infected by C. fusipes was computed to be of about 30 years. It was also observed that a moderate or heavy degree of infection could be related to a reduction of tree radial growth. C. fusipes appeared to be at the origin of growth loss. Keywords: Collybiafusipes, root rot, oak decline, soil factors INTRODUCTION The European oak forest has undergone several diebacks in the last century (Delatour 1983). Around 1980 field studies have mentioned that the primary parasite C. fusipes was involved in the decline of pedonculate oak (Q. robur), in particular in soils not submitted to water logging (Delatour and Guillaumin 1984; Guillaumin et al. 1985). Presence of C. fusipes was also reported in red oak (Q. rubra) stands in association with problems of growth. Artificial inoculations have confirmed that C. fusipes is a primary parasite and that vigorous seedlings of different oak species are susceptible to C. fusipes. Moreover, the parasite was more aggressive on red oak than on pedonculate oak (Marçais and Caël 2000). C. fusipes appeared to be a slow parasite, as two years after seedlings had been inoculated, no mortality could be observed as would have occurred with Armillaria mellea or A. ostoyae on young pine seedlings (Rishbeth 1982). However, the period required for a mature oak to become heavily infected remained unknown. C. fusipes is not considered as a weakness parasite, as artificial inoculations evidenced that seedlings defoliated for two years did not show an increased susceptibility to the parasite (Marçais and Delatour 1996). Furthermore, it was observed in a field study that C. fusipes did not develop preferentially on trees not favoured by selective logging (Piou et al. 2001). This parasite induces typical orange lesions on large roots and can cause a drastic destruction of the root system. Infected pedonculate oaks, located in sandy soils with presence of a gravellayer, are more declining. Such soils are known to be unfavourable to Q. robur and we do not know whether C. fusipes was the cause of decline or whether soil conditions are involved in decline. A method allowing a good estimation of root infection severity of a mature tree was established by Marçais et al. (1999). Using this method, we were able to investigate the relationship between tree health, the severity of C. fusipes infection and soil characteristics, in two pedonculate oak stands. We were also able to investigate the induction and evolution of disease in two naturally infected red oak stands during a 6 years period and to relate this evolution to soil and tree characteristics. Ecology and Biodiversity 71 MATE~SANDMETHODS Study sites Trees were sampled in four sites where C. fusipes showed a scattered distribution (Table 1). Two out of four sites, with Q. robur, were located in alluvial forests. The soit consisted in a sandy loam coarse textured layer of heterogeneous thickness covering a layer of grave\. Most roots over 1 cm in diameter did not extend into this gravel layer. In the two other sites, with Q. rubra, soil consisted in a layer with a coarse texture covering clay where vertical root growth was not limited (Table 1). Table 1 . Description of the study sites. Site Oak species No. Studied No. declining Soil texture Parental Sail pH (Area in France) trees oaks material Ainvelle Q. robur 60 20 Clayey sandy silt Gravel 4.5 (Haute-Saône) Mersuay Q. robur 60 22 Sandy clayey silt Gravel 5.2 (Haute-Saône) Les Barres Q. rubra 93 0 Sandy loam Clay 4.0 (Loiret) Agre Q. rubra 51 0 Clayey silt Clay (Tarn-et-Garonne) Tree sampling Relationship between decline, C. fusipes infection severity and soit parameters were investigated at Ainvelle and Mersuay. Trees were selected on the basis oftheir crown status, rated on a 0 to 3 scale adapted from Nagelesein (1995): (0) crown healthy and opaque with dense secondary ramifications; (1) crown moderately healthy, with dead twigs present and/or gaps present in the canopy; (2) crown moderately declining, with the gaps in the canopy coalescing at the periphery of the crown and fonning openings towards the outside in the upper part of the crown. The skeleton of large limbs is fully visible; (3) crown severely declining, with large dead limbs in the upper part of the crown and/or loss of more than half of fine branches. The 60 oak trees were selected to obtain dominant trees 20 being healthy oaks rated as zero or one 20 being moderately declining oaks rated as 2 and 20 being declining oaks rated as 3. The age oftrees was about of80-150 years. Evolution and induction of the disease in a 6 years period was monitored at Les Barres and Agre. Oak trees (Q. rubra) were selected on the basis of their social status. Only healthy dominant and codominant oaks were selected. The age oftrees varied between 40 and 70 years old at Les Barres and was of20 years old at Agre. Site and tree investigation The diameter at breast height (dbh) was investigated at Les Barres in 1996 and in 2001 and at Agre in 1995 and in 2001. Oak height was measured at Agre only on 30 trees in 1995, and on every trees in 1996 at Les Barres. The number of neighbouring trees within 4 m was evaluated in 1995 only at Agre. In this stand, a silviculture trial was set up in the past and result in very heterogeneous tree density. In the 4 stands, a core of soit was excavated at the base of each tree. Soils were described especially for their variability in depth of water logging traces (traces of iron deposition or flight) and in depth of layer able to limit vertical growth of roots such as a gravel layer. 72 Ecology and Biodiversity Root damage assessment Root systems were studied for C. fusipes infection severity as described by Marçais et al. (1999). Briefly, the root collar and major roots were partially excavated to a depth of about 20-30 cm and a distance of about 80­ 100 cm from the trunk base. Lesions caused by C. fusipes are very characteristics and are easily detected as patches of dead bark that are orange in colour with small white fans of mycelium scattered in the necrotic tissues. On pedonculate oak, an hypertrophy of the bark is usually observed as infected bark is thickened up to 3-4 cm. In contrast, on Q. rnbra, thickening of bark tissue is rarely observed (Marçais et al. 1999). Previous work showed that C. fusipes is consistently isolated from such lesions on oak roots (Guillaumin et al. 1985; Marçais et al. 1999). The infection status of each major root was assessed within the following four classes: 0) no necrosis detected; (1) superficial necrosis present, but covering less than half of the root circumference (penetration of C. fusipes in the bark of less than 1-2 mm); (2) necrosis covering one side of the root entirely (penetration usually 2­ 5 mm for Q. robur); (3) C. fusipes infection over the entire root circumference but root still alive (penetration usually more than 4-5 mm for Q. robur); (4) root dead with decayed wood. Diameter of each root was measured at about 10 cm from the trunk base. The root infection index of a tree was computed as: L (root diameter X root rating) / L (root diameter). This index therefore takes value from 0 to 4. Trees with an index value of 0-0.3 were rated as undamaged, those with an index value of 0.3-2 and 2-4 were rated as respectively lightly and heavily damaged trees. C. fusipes infection severity was rated in 2000 in Mersuay and Ainvelle, and successively in 1995 and 2001 on the same trees in Les Barres and Agre. Statistical analysis Root infection index and infection evolution in a 6 years period were subjected to an analysis of variance using a GLM procedure (SAS Inc. 1989). The frequency of oaks uninfected in 1995 that became infected during the 6 years period and the frequency of declining oaks were analysed by generalised linear modelling with the SAS procedure GENMOD. A binomial distribution was assumed and the logit link function was used. RESULTS Relationship between C. fusipes infection severity, pedonculate oak health and soU characteristics In both sites, Ainvelle and Mersuay, oak decline and infection severity were significantly correlated (Table 2). The majority of oak trees rated as severely damaged by C. fusipes were rated as moderately or heavily declining trees (Ainvelle: 12 of 12 trees severely damaged, Mersuay: 14 of 16 trees). Table 2. Correlation analysis between oak decline and infection severity. Plots Ainvelle Mersuay Spearman correlation coefficients 0.497 0.428 Probability 0.0001 0.0006 At Ainvelle and Mersuay, the root infection index increased significantly with the depth ofwater logging in the soil (Fig l, respectively: F = 5.14, df= 5, P < 0.001; F = 8.09, df= 4, P < 0.001). Root infection index decreased when the gravel layer was close to the soil surface in Mersuay (F = 3.23, df= 3, P = 0.029) and when the gravellayer appeared deeper in the soil in Ainvelle (F = 3.44, df= 4, P = 0.014). Ecology and Biodiversity 73 3. ~.:: ~ 2. ID en c o U ~ c 15 T ••1 35 45 55 75 100 15 25 40 60 75 ---Ainvelle-- --Mersuay-- Depth ofwater logging (cm) Figure 1. Mean root infection index for different depth of water logging, measured as the first traces of iron deposition, in Ainvelle and Mersuay. The bars represent the confidential interval of the mean. The number oftrees in the categories is given above the bar. In Mersuay, the percentage of declining oaks increased significantly when the water logging depth was deeper in the soil (likelihood X2 = 19.6, df = 4, P < 0.001) and when the gravel layer appeared closer to the soil surface (likelihood X2 = 10.4, df= 3, p = 0.014). However, the relation between tree crown status and depth of gravel layer was due to the fact that severely infected trees are more frequent when the gravel layer appears close to the soil surface (Fig 2). There was no significant relationship between the proportion of declining oaks and the depth of water logging or the depth of gravel layer in Ainvelle. Collybia infection severity : o none .. moderate _ heavy 5s 3 O....l....-_~~ 40-60 60-S0 >=so Depth of gravellayer (cm) en ~ CIl o Cl c c Ü ID o (%) 100 80 60 40 20 Figure 2. Proportion of declining oaks (rated as 3) in Mersuay according to their level of infection by C. fusipes, for different depth of gravellayer. The number oftrees in the categories is given above the bars. Effects of soil and tree characteristics on infection induction and evolution over a 6 years period, on red oaks. Age of oaks did not appear ta influence the evolution of infection at Les Barres. At Agre, in a situation very different for stand, soil and climate, disease evolution over the same period, on younger trees, was on the same range. In these two plots, evolution of infection between 1995 and 2001 did not depend on the initial level of infection in 1995. Thus, it is possible to make the hypothesis that the speed of the disease is about constant through the time. This allowed us to estimate the period required for an healthy oak (rated as 0) to become severely infected (rated as 2). This period was computed as the ratio: [(2 / infection evolution between 1995 and 2001) x 6]. The period appeared to be very long bath at Les Barres and at Agre (Table 3). 74 Ecology and Biodiversity Table 3. Number of years required for an uninfected red oak (rated as 0) to become severely infected (rated in 2) by C. fusipes. Les Barres Agre First quartile (25% of trees before) 17 17 Median (50% oftrees in both sides) 35 27 Third quartile (25% of trees after) 80 70 At Les Barres, the frequency of oaks not infected in 1995 that became infected between 1995 and 2001 increased significantly with the depth of water logging (Fig. 3 ; likelihood X2 = 6.68, df = 1, P = 0.02). This effect did not appear at Agre (Fig. 3). However, in this plot, evolution of infection on trees already infected in 1995, significantly depended on the depth of water logging (Fig. 4, F = 9.76, df= 1, P = 0.02). The infection evolution on already infected tree was not linked with water logging at Les Barres (Fig. 4). 6 ...... co "- 0 inN 7{i .~ 0-0 -0 Q) Q) ..... ..... ü ü Q) Q)­ _ c c .­ c Q) :::J E_ cc o ü >,Q) ü.o c ..... Q) cc :::J~ eJ ..... ~ LL 0.8 0.6 10 ::.:.1IiI IiIII 0.0 -t..===----===--===--__-==:..........:==--===---.J 0-40 40-70 70-110 10-30 30-60 > 60 -- Les Barres -- --- Agre --- Depth of water logging (cm) Figure 3. Frequency ofuninfected oaks that became infected between 1995 and 2001 for different depth ofwater 10gging at Les Barres and Agre. The number oftrees in the categories is given above the bar. 9 16 20 11 o 0-40 40-70 70-110 0-30 30-60 60-90 -- Les Barres-- --- Agre Depth of water logging (cm) 0.0 1......::==--===---==:.- ----' 0.6 1.0 0.4 0.2 0.8 ~ .;:: LO Q)O) >0) ~ ...... c "~ .Q-o ..... Q) ü ..... ~ ü c~.- c o·;;; c.::.:. o cc :.;:::; 0 :::J L.. - 0 ~-w Figure 4. Evolution of root infection severity during 6 years from oaks infected in 1995, for different level of water logging at Les Barres and Agre. The bars represent the confidential interval of the mean. The number of trees in the categories is given above the bar. Ecology and Biodiversity 75 At Agre, evolution of the root infection severity and frequency of uninfected oaks in 1995 that became infected in 2001 were not related to the degree of competition with neighbouring trees in 1995, evaluated either as the density of neighbouring trees within 4 m or as the ratio height/dbh. In contrast, at Les Barres, the frequency of oaks uninfected in 1995 that became infected in 2001 decreased with the degree of concurrence undergone between trees in 1995, measured as the ratio height/dbh (likelihood X2 = 5.54, df= 1, P = 0.019). The mean radial growth between 1995 or 1996 and 2001 decrease in the two plots when trees were moderately and severely infected by C.fusipes in 1995 (Table 4, Les Barres: F= 9.5, p = 0.003; Agre: F = 25.97, P < 0.001). No significant relationship between growth and the induction or evolution of infections could be evidenced (results not shown). Table 4. Mean radial growth (mm / year) ofred oaks between 1995 or 1996 and 2001, at Agre and Les Barres, for different degree of C. fusipes infection in 1995. Agre Les Barres Infection level Nb oftrees Mean Std Error Nb oftrees Mean Std Error None 26 5.30 0.28 28 6.95 0.31 Moderate-Iow 17 4.76 0.31 26 7.30 0.32 Moderate-high 8 3.54 0.33 18 6.76 0.38 Severe 0 21 4.74 0.52 DISCUSSION Water logging is the major factor explaining distribution of C. fusipes infection in the four studied stands. Indeed, the parasite seems to be favoured by a reduced water logging level that might be related to oxygen availability. In vitro, the colonisation of little hazel stem segments by C. fusipes is strongly influenced by a good aeration of the culture. Furthermore, during an experiment of artificial inoculations, C. fusipes had not survived in the majority of inocula located in pots where fortuitous water logging had occurred (Marçais and Caël 2000). In a field study, Piou et al. (2001) observed that C. fusipes developed well on final crop trees submitted to poor water logged conditions. In the Netherlands, Nanta and Vellinga (1995) reported preferential distribution of C. fusipes on dry soils. Root rot fungus are often mentioned in literature as favoured by dry soil conditions. Indeed, in British Columbia, the incidence of disease induced by Inonotus tomentosus was higher on soils with a dry moisture regime as influenced by slope position and texture (Bernier and Lewis 1998). Whitney (1984) showed that black spruce located on dry moisture regimes are more heavily infected by A. mellea, than those on moist or wet sites. Kuhlman (1980) observed that root segments of loblolly pine were more decayed by Heterobasidion annosum when they were buried in dry soils. The low degree of competition between trees was Iinked with an higher frequency of tree that became infected in the 6 years study period at Les Barres. In fact, trees subjected to a poor level of competition were located in different soil situation than those subjected to a poor competition level, with water logging appearing deep in the soil. Then, they were located in favourable conditions for C. fusipes. Our study also showed that the proportion of declining pedonculate oaks increased with their severity of infection by C. fusipes. This observation confirms those made by Marçais et al. (2000) through other sites in France. We confirmed that in this case C. fusipes was the major cause of decline. Soil conditions, though they were unfavourable for Q. robur, appears to play a secondary role in decline. The radial growth of red oaks of Les Barres and Agre during the 6 years period was lower on those that were the more heavily infected by C. fusipes at the beginning of the period. C. fusipes was the major factor linked with reduced growth. In those stands, Marçais and Caël (2001) have already observed, through a dendrochronological study, that trees severely infected by C. fusipes exhibited poor growth for many years. However, the time at which trees became infected could not be determined in this study and, as a consequence, it 76 Ecology and Biodiversity could not be concluded whether the root rot was the cause of the poor growth or whether oaks with poor growth had higher chance to become infected and then to develop high level of infection. Observation of the disease evolution on single trees during 6 years gives a part of the answer. Indeed, previously infected trees have a reduced growth but conversely there is no effect of tree growth on infection induction or evolution. Thus, as the parasite does not appear to be influenced by growth, that could allow us to conclude that C. fusipes is probably the origin of growth reduction. We were also able to estimate the period required for the parasite to induce severe damage in field conditions. Because of the slow evolution of C. fusipes infection, this estimation could never have been done by inoculation experiments. REFERENCES Bernier, D.; Lewis, K. J. 1999. Site and soil characteristics related to the incidence of!nonotus tomentosus. Forest Ecology and Management. 120: 131-142. Delatour C. 1983. Les dépérissements des chênes en Europe. Revue Forestière Française. 25 (4):265-282. Delatour C.; Guillaumin l-J. 1984. Un pourridié méconnu: Collybiafusipes (Bull. ex Fr.) Quel. C.R. Acad. Agri. de France; séance du 18 Janvier 1984. 70(1):123-126. Guillaumin ll; Bernard Ch.; Delatour c.; Belgrand M. 1985. Contribution à l'étude du dépérissement du chêne: pathologie racinaire en forêt de Tronçais. Ann. Sei. For. 42 (1): 1-22. Kuhlman E. G. 1980. Influence of moisture on rate of decay of loblolly pine root wood by Heterobasidion annosum. Cano 1 Bot. 58:36-39. Marçais B.; Delatour C. 1996. Inoculation of oak (Quercus robur and Q. rubra) with Collybia fusipes. Plant. Dis. 80: 1391-1394. Marçais B.; Caël O.; Delatour C.1999. Measuring the impact of Collybiafusipes on the root system of oak trees. Ann. Sc. For. 56:227-236 Marçais B.; Caël O.; Delatour C.2000. Relationship between presence of basidiomes, above-ground syrnptoms and root infection by Collybiafusipes in Oaks. Eur. J. For. Path. 30:7-17. Marçais B.; Caël O. 2000. Comparison of the susceptibility of Quercus petraea, Q. robur and Q. rubra to Collybiafusipes. European Journal of Plant Pathology. 00:1-6. Marçais B.; Caël O. 2001. Relation between Collybiafusipes root rot and growth ofpedonculate oak. Cano J. For. Res .. 1 For. Res. 31:757-764. Nageleisen L.M. 1995. Méthode d'évaluation de l'aspect du houppier (protocole DEPEFEU). Paris: Département Santé des Forêts- Echelon technique Nord-Est. Bulletin technique. Il pp. Nanta M.; Vellinga E.C. 1995. Atlas van Nederlandse Paddesstoelen. A. A. Balkema, Roterdam, Brookfield. p. 352. Piou D.; Delatour C.; Marçais B. 2001. Hosts and distribution of Collybia fusipes in France. Eur. J. For. Path. (accepted) Rishbeth J. 1982. Species ofArmillaria in southern England. Plant Pathology. 31 :9-17. Whitney R.D. 1984. Site variation of Armillaria mellea in three Ontario conifers. In : Kile, G. A., ed. Proceedings of the 6th international conference of root and butt rot of forest trees; 1983 August 25-31; Melbourne, Victoria, Gympie, Queensland, Australia. Melbourne: International of Forestry Research Organizations: 122-130. Ecology and Biodiversity 77 FIRE AND ARMILLARIA: EFFECTS ON VIABILITY AND DYNAMICS IN EASTERN OREGON, USA G.M. Filip, S.A. Fitzgerald, and L. Yang-Erve Forest Science Department, 321 Richardson Hall, Oregon State University, Corvallis, OR 97331 USA and OSU Extension Service, 1421 S Hwy 97, Redmond, OR 97444 USA SUMMARY This study detennined the effects of prescribed burning, soil depth, antagonistic fungi (Trichoderma harzianum Rifai), and time since burning on the viability of Armillaria ostoyae (Romagnesi) Herink in wood pieces buried in the soil of a mixed-conifer forest in northeastern Oregon, USA. Red aider (A/nus rubra Bong) stem segments colonized with A. ostoyae were buried at two soil depths in plots that were burned and not burned. Half of the Armillaria segments were buried with segments of T. harzianum. Prescribed burning in the fall significantly reduced the recovery of A. ostoyae one day after the burn at a soil depth of 8 cm but not at a soil depth of 30 cm. Adding T. harzianum inoculum to the soil did not appear to reduce A. ostoyae recovery immediately after the fire, but A. ostoyae recovery appeared to decrease after several months. Differences in A. ostoyae recovery may also be due to the season (fall or spring) of the prescribed burns. Keywords: Armillaria ostoyae, Trichoderma, inoculum, prescribed fire, mixed-conifers INTRODUCTION Fire has shaped the forests of western North America for at least 350 million years (Agee 1993). Fire return intervals have historically ranged between 40-150 years in the upper elevation, mixed-conifer forests of northeastern Oregon (Mutch et al. 1993, Fitzgerald et al. 2000). Fire exclusion and selective harvesting of pine and larch that began at the turn of the 1900s have resulted in an unprecedented abundance of Abies grandis (Dougl. ex D. Don) Lindl. and Pseudotsuga menziesii (Mirb.) Franco in many areas in the interior West (Filip 1994). Mortality from root disease and other pests is much greater in stands with more Abies spp. (Filip and Goheen 1984, Filip et al. 1996). As a result, these stands are overstocked with diseased, suppressed, or dead A. grandis, P. menziesii, and Pinus ponderosa Dougl. ex Laws. Subsequently, wildfire frequency in the western USA and even throughout the country appears to have increased with 84 690 wildfires on 43 057 ha in the USA in 2000 (National Fire News 2000). Prescribed burning is increasingly used throughout western North America in an attempt to return fire to its natural role in the ecosystem. Only a few studies, however, have directly linked fire in the Pacific Northwest to the changes in the incidence or severity of root pathogens (Thies 1990). In the mixed-conifer forests of northeastern Oregon, Armillaria ostoyae is common and widely distributed (Schmitt et al. 1991). Schmitt found that A. grandis had the highest incidence of damage from several species of root pathogens, that P. ponderosa was damaged more by A. ostoyae than other root pathogens, and that Pseudotsuga menziesii was damaged by root pathogens in most sample strata and plant associations. Armillaria ostoyae clone sizes of 1134 ha and ages of approximately 2400 years have been reported in a study area adjacent to ours (Ferguson et al. 1999). Biological control of Armillaria is difficult because effective populations of antagonistic fungi are not sustainable under nonnal field conditions. Filip and Yang-Erve (1997) provide a summary of the interactions between Armillaria spp. and Trichoderma spp. Because prescribed burning can influence antagonistic soil microorganisms (Margaris 1977, Panneter 1977), controlled burns may indirectly help minimize Annillaria root disease. In central Oregon, Reaves et al. (1984) found that ash leachates from prescribed fire in ponderosa pine forests reduced the growth of A. ostoyae in culture. Isolates of Trichoderma spp. obtained from burned soils were 78 Ecology and Biodiversity more antagonistic to A. ostoyae in culture than isolates from unburned soils (Reaves et al. 1990). Rood and Sandberg (1989) showed a significant reduction in viable rhizomorph recovery for A. limonea (Stevenson) Boesewinkel and A. novae-zelandiae (Stevenson) Herink after a prescribed fire in New Zealand. The objective of our study was to determine whether prescribed fire reduces the viability of A. ostoyae in wood segments artificially buried in soil in a mixed-conifer forest. We were specifically interested in four issues: (1) whether there is a significant difference in the viability of A. ostoyae inoculum segments between burned and unburned treatments; (2) whether the treatment difference is associated with the depth of buried segments (8 cm vs. 30 cm); (3) whether the treatment difference is associated with the presence of Trichoderma harzianum; and (4) whether the treatment difference is associated with time since the prescribed burning treatment (over a year). MATERIALS AND METRODS Research site The study was conducted on land managed by the USDA Forest Service, Prairie City Ranger District, Malheur National Forest of Oregon (Filip and Yang-Erve 1997, Fitzgerald et al. 2000). The study site is at an elevation of 1,370 m with a west to south-facing slope. Four plant associations occur in the research area: Abies grandis/Calamagrostis rnbescens, A. grandis/Carex geyeri, A. grandis/Vaccinium membranaceum, and A. grandis/V. scoparium (Johnson and Hall 1990). Stands have a dominant overstory of ponderosa pine with scattered large-diameter western larch (Larix occidentalis Nutt.), Douglas-fir (Pseudotsuga menziesii var. glauca (Beissn.) Franco), and grand fir. Large diameter trees are common in the study area, and stands contain dying and dead grand fir and Douglas-fir due to an outbreak of western spruce budworm (Choristoneura occidentalis Freeman) from 1980 to 1992. Armillaria ostoyae, Armillaria NABS X, and Heterobasidion annosum sensu lato occur naturally in the study area. The study area is adjacent to an Armillaria clone study (Ferguson et al. 2000) and a tree-density reduction study (Fitzgerald et al. 2000). Three units in the study area were selected because oftheir prescribed fire treatment priority. Units A and B were selectively harvested in order to reduce excess stand density and favor ponderosa pine and western larch. Unit H was clearcut. Prescribed fire was planned in aIl three units in order to reduce fuel loads and create planting spots. Because of the uneven distribution of Armillaria root disease in the study area, an isolate of Armillaria ostoyae from northeastern Oregon was used to artificially inoculate the treatment blocks. Trichoderma spp. were isolated from the soil at the study sites. A preliminary laboratory test was performed to screen for the Trichoderma isolate that was the most antagonistic to A. ostoyae; an isolate of T. harzianum was used as the inoculurn. Red aIder (Alnus rnbra Bong.) stem segments were used as substrates for A. ostoyae and Trichoderma harzianum, and inoculum preparation has been described previously (Filip and Yang-Erve 1997) Experimental design and plot establishment Two blocks each were randomly established in units A, B, and H in the study area: blocks 1 and 2 in unit A, blocks 3 and 4 in unit B, and blocks 5 and 6 in unit H. Each block was designed with a split-plot configuration. The use of prescribed burning was the whole-p10t factor, and soil depth and presence of Trichoderma were the sub-plots factors. Each 12 x 24 m sub-block (burned or unburned) contained four plots (Figure 1). Treatments were randomly assigned to the plots. Fifteen red aIder stem segments colonized with Armillaria were buried in each plot before burning; five segments were buried for each ofthree sampling times (1 day, 1 month, and 2 months after burning). A Trichoderma-inocu1um segment was buried adjacent to and touching an Armillaria-inoculum segment where indicated (AT). An iron stake was placed at each corner and at the center of each plot. The position of each inoculum segment was measured from the closest iron stake and recorded on a map for future sampling. Ecology and Biodiversity 79 The configuration was a 23 factorial design; the treatments were bumed or unbumed, 8 or 30 cm soil depth, and presence or absence of Trichoderma inoculum. Ali segments were randomly assigned to each treatment and buried between August and September 1993. Before the prescribed fire treatment, woody debris was evenly scattered over each plot so that buming was uniform. Prescribed burning Unfortunately, local weather and the availability of personnel at the Prairie City Ranger District affected our ability to conduct prescribed buming. Instead of buming aU six blocks in the fall of 1993 as planned, the fire treatment occurred at three different times over two years. Blocks 5 and 6 in H unit were bumed for an hour on 27 October 1993. The woody debris and litter layer on both blocks were barely consumed; soils were moist to the touch. On 31 October 1993, the bumed plot of block 6 was rebumed for four hours because of the poor original bum; the first sampling of segments occurred the next day, the second sampling occurred in June 1994, and the third in August 1994. Block 3 in unit Band block 5 in unit H were bumed on 8 June 1994, nine months after inoculation. The fire lasted one hour, was extinguished with water, and completely consumed the woody debris and Iitter layer. We began sampling the next day, sampled a second time in July, and a third time in August. Blocks l, 2, and 4 in units A and B were bumed on 27 October 1994, one year after inoculation. The fire lasted one hoUT and completely consumed the woody debris and litter layer. Water was used on ail plots to extinguish the fire. Sampling for Blocks l, 2, and 4 was done only on the day after the bum. These three blocks received the same treatment, so data could be statisticaUy analyzed. Inoculum recovery Following each prescribed fire treatrnent, five segments from each treatment were randomly selected in order to test the viability of A. ostoyae at each sampling time. Segments were excavated individually, bagged in separate paper bags for each treatment, and stored at 1-2°C until they were processed in the laboratory. We attempted to isolate the fungi by splitting each segment asepticaUy and plating five wood chips (1 x 1 x 1 mm) onto a selective medium for Armillaria (2% malt agar with 1.5 ppm each of prochloraz, benomyl, and thiabendazole; 30 ppm rose bengal; and 100 ppm streptomycin sulfate; Goldfarb et al. 1989). The recovery rate of A. ostoyae was scored from 0-5 based on the number of isolates recovered from that segment. The number of A. ostoyae isolates recovered was converted to 0-100% scale. Treatrnent means were the average of the 15 segments in that treatrnent. Statistical analysis Percent recovery of A. ostoyae was tabulated by buming treatrnent, soil depth, and presence of Trichoderma. Data from blocks l, 2, and 4 were examined by using analysis of variance for a 23 factorial with an F test (P<0.05; SAS Tnstitute, Tnc. 1994). We explored the main and interactive effects of prescribed buming, depth of inoculum, and the presence of Trichoderma on the percentage recovery of A. ostoyae. Comparisons were made between ail combinations of treatments. RESULTS There were no significant differences among blocks l, 2, and 4 (P = 0.94) in the percentage recovery of Armillaria immediately after the prescribed bum in October 1994. The effect of prescribed buming on Armillaria viability was significant (P = 0.0138; Table 1). The interaction between the prescribed bum and soil depth on Armillaria viability was also significant (P = 0.0138). The 3-way interaction of prescribed buming, soil depth, and the presence of Trichoderma (P = 0.3003), the interaction between prescribed buming and the presence of 80 Ecology and Biodiversity Trichoderma (P = 0.4313), and the interaction between soil depth and the presence of Trichoderma (P = 0.9193) did not significant1y affect the viability ofArmillaria. Table 1. Results of an analysis of variance of the effects of prescribed burning (Burn), soil depth, and presence of Trichoderma (Trich) on the viability ofA. ostoyae inoculum one day after fall treatment. Source NDF I DDF I Type III FI Burn 1 14 7.92 Soil depth 1 14 0.06 Trich 1 14 0.14 Burn x Depth 1 14 7.92 Burn x Trich 1 14 0.66 Dept x Trich 1 14 0.01 Burn x Depth x Trich 14 1.16 0.3003 0.0138 0.8059 0.7181 0.0138 0.4313 0.9193 INDF = numerator degrees of freedom; DDF = denominator degrees of freedom; Type III F = Type III F-value; and Pr> F = probability greater than F-value. Bolded values are significant. Prescribed burning and soil depth The recovery of Armillaria decreased significant1y from 36 to 5% at soil depths of 8 cm in burned plots but remained unchanged at soil depths of 30 cm (Table 2). The recovery of Armillaria was lowest in the burned and highest in the unburned treatments at a soil depth of 8 cm. The mean recovery of Armillaria at the 8 cm depth with prescribed burning was 5% which is not statistically different than 0% recovery (P = 0.43); thus Armillaria at a soil depth of 8 cm had virtually no viability after the prescribed burning. The mean recovery of Armillaria in unburned plots was 36% at 8 cm and 19% at 30 cm; however, depth itself did not have a significant effect on the percentage recovery ofArmillaria in unburned plots (P = 0.8059; Table 1). Table 2. The percentage recovery of Armillaria ostoyae, at two soil depths and at burned and unburned sites one day after fall treatrnent in northeastern Oregon. Treatment Soil depth (cm) Mean l SE Burned 8 5 (0.23)a2 0.29 Burned 30 19 (0.95)ab 0.29 Unburned 8 36 (l.83)b 0.29 Unburned 30 19 (0.95)ab 0.29 1 Means are the percentage and actual number (in parentheses) of viable isolations of A. ostoyae out of a total of five attempts from each of 15 segments per treatment. 2 Means followed by a different letter are significantly (P<0.05) different according to ANOYA (SAS Institute Inc. 1994). Season of burning Block 6 in Unit H was burned in the fall 1993 and sampled three times: fall 1993, spring 1994, and faH 1994. Because this was the only block effectively treated in faH 1993, there was no statistical analysis. In plots treated with prescribed burning, the recovery of Armillaria appeared to remain the same at the 30 cm soil depth regardless of the time of sampling; however, recovery decreased at the 8 cm depth at the second sampling. In unburned plots, the recovery of Armillaria at a depth of 8 cm was higher than at 30 cm at the second and third samplings. Following the burning treatment, the recovery of Armillaria appeared to decrease with time in the presence of Trichoderma but increased in the absence of Trichoderma. Without the burning treatment, recovery Ecology and Biodiversity 81 ofArmillaria increased at the second sampling, then decreased at the third sampling, especially in the presence of Trichoderma. Block 3 in Unit B was burned in spring 1994 and sampled three times: June, July, and August 1994. Block 5 in Unit H was also bumed in the spring, but the unbumed plot was accidentally destroyed. The spring bum seemed to have no effect on the recovery of Armillaria during the sampling period. At the second and third sampling, Armillaria recovery in bumed plots at the 8 cm depth was higher than recovery at 30 cm. The presence or absence of Trichoderma appeared to have little affect on Armillaria recovery. DISCUSSION AND CONCLUSION There was a significant reduction in Armillaria recovery immediately following a fall bum. The recovery of Armillaria was lowest (5%) in bumed plots and highest (36%) in unbumed plots in segments at a soil depth of 8 cm. The results of this field study suggest that the heat from the fire was able to kill a high proportion of Armillaria at a soil depth of 8 cm but not at 30 cm. The fact that most segments were scorched by the fire in ail three blocks further supported the conclusion that the reduction in Armillaria recovery was caused by heat. As mentioned previously, Hood and Sandberg (1989) reported that there was a highly significant reduction in the yield from isolates of rhizomorphs, although no significant change in rhizomorph frequency, after felling and buming of forest vegetation in New Zealand. Trichoderma harzianum can kill Armillaria mycelium in vitro. The shift in the microbial balance from ash leachates in forest soils (Reaves et al. 1990) that favors the antagonist Trichoderma following a prescribed fire is a rather slow process compared to lab tests. We did not find a decrease in Armillaria with Trichoderma in samples taken immediately after the fire, although 8-10 months after the prescribed buming we found a decrease in Armillaria recovery. Additional research is needed in order to evaluate whether the reduction in Armillaria recovery is short-tenn. AIso, there are large natural populations of Trichoderma spp. in forest soils that we did not monitor. There was no apparent difference in the recovery of Armillaria between bumed and unbumed plots after the spring buming treatment. This differs from the results after the fall buming; we found a significant difference between bumed and unbumed plots at the 8 cm soil depth. More studies are needed to establish the timing and intensity offire that are necessary to effectively reduce Armillaria recovery. Woody roots that are naturally infected with Armillaria are normally present at depths much greater than were inoculum segments in our study. Obviously, an operational bum with conditions similar to our study would only destroy a small portion of the total Armillaria inoculum. Unless the bum was hotter or more penetrating, a situation that could lead to soil damage and significant nutrient loss, the direct effects of prescribed fire on Armillaria populations appear to be negligible. The long-term effect of buming as related to changes in Trichoderma populations and subsequent indirect effects on Armillaria, however, may be more important. Long­ tenn studies that explore the effects of prescribed buming on naturally infected roots in soil need to be conducted. Advanced A. ostoyae clone sizes and ages in northeastem Oregon (Ferguson et al. 1999) testifY to the population stability and resilience through centuries offire and forest structural changes. 82 Ecology and Biodiversity Studyarea Sub-block Burned Black Unburned A A A A A A A A A A A A A A A AT AT AT AT AT AT AT AT AT AT AT AT AT AT AT Figure 1. Diagram of the plot configurations and randomized treatment applications within each block. Red aider segments, colonized by Armillaria (A) were buried 8 or 30 cm under the soil surface (soil depth). In sorne cases, a segment colonized by Armillaria and a segment colonized by Trichoderma were buried together (AT). The plots in which these segments were buried were subsequently subjected to prescribed burning treatments. REFERENCES Agee, J.K. 1993. Fire ecology ofPacific Northwest forests. Island Press, Washington, D.C. 493 p. Ferguson, B.; Dreisbach, T.; Parks, c.; Filip, G.; Schmitt, C. 1999. Armillaria root disease species and clone diversity across a mixed-conifer landscape in the Blue Mountains of eastern Oregon. In Proceedings of the Fifth Joint Meeting of the Western International Forest Disease Work Conference and Western Forest Ecology and Biodiversity 83 Insect Work Conference, Goheen, E.M. (ed.), USDA Forest Service, Southwest Oregon Forest Insect and Disease Service Center, Central Point, Oregon. Pp. 146-151. Filip, G.M. 1994. Forest health decline in central Oregon: a 13-year case study. Northwest Science 68:233-240. Filip, G.M.; Goheen, D.J. 1984. Root diseases cause severe mortality in white and grand fir stands of the Pacifie Northwest. Forest Science 30: 138-142. Filip, G.M.; Yang-Erve, L. 1997. Effects ofprescribed burning on the viability of Armillaria ostoyae in mixed­ conifer forest soils in the Blue Mountains of Oregon. Northwest Science 71 (2): 137-144. Filip, G.M.; Torgersen, T.R.; Parks, C.A.; Mason, R.R.; Wickman, RE. 1996. Insect and disease factors in the Blue Mountains. In Search for a Solution: Sustaining the Land, People, and Economy of the Blue Mountains. Jaindl, R.G.; Quigley, T.M. (eds.). American Forests, Pub!., Washington, D.C. Pp. 169-202. Fitzgerald, S.A.; Emmingham, W.H.; Filip, G.M.; Oester, P.T. 2000. Exploring methods for maintaining old­ growth structure in forests with a frequent-fire history: a case study. In Fire and forest ecology: innovative silviculture and vegetation management. TaU Timbers Fire Ecology Conference Proceedings, No. 21, Moser, W.K.; Moser, C.F. (eds.). Tall Timbers Research Station, Tallahassee, Florida. Pp. 199-206. Goldfarb, B.; Nelson, E.E.; Hansen, E.M. 1989. Trichoderma spp.: growth rates and antagonism to Phellinus weirii in vitro. Mycologia 81:375-381. Hood, 1 A; Sandberg, C.l. 1989. Changes in soil populations of Armillaria species following felling and burning of indigenous forests in the Bay of Plenty, New Zealand. ln Proc. 7th Int. Conf. on Root and Burt Rots, IUFRO Working Party S2.06.01. Morrison, D J.(ed.). Pacific Forestry Research Centre, Victoria, British Columbia, Canada. Pp. 288-296. Johnson, c.G., Jr.; Hall, F.C. 1990. Plant associations of the Blue Mountains. USDA Forest Service R6-ECOL AREA 3. Pacific Northwest Region, Portland, Oregon. 116 p. Margaris, N.S. 1977. Decomposers and the fire cycle in Mediterranean-type ecosystems. ln Proc. Symposium on the Environmental Consequences of Fire and Fuel Management in Mediterranean Ecosystems. Mooney, RA.; Conrad, C.E. (eds.) USDA Forest Service General Technical Report WO-3. Washington, D.C. Pp. 37-45. Mutch, R.W.; Arno, S.F.; Brown, J.K.; Carlson, C.E.; Ottmar, R.D.; Peterson, J.L. 1993. Forest health in the Blue Mountains: a management strategy for fire-adapted ecosystems. USDA Forest Service General Technical Report PNW-GTR-31O, Portland, Oregon. 14 p. National Fires News. 2000. Wildfire season overview, January through October 2000. Website http:/www.nifc.gov/fireinfo/nfnlO-17Summ Parmeter, J. R. 1977. Effects of fire on pathogens. In Proc. Symposium on the Environmental Consequences of Fire and Fuel Management in Mediterranean Ecosystems. Mooney, RA.; Conrad, C.E. (eds.). USDA Forest Service General Technical Report WO-3. Washington, D.C. Pp. 58-74. Reaves, J.L.; Shaw, c.G. III; Martin, R.E.; Mayfield, J.E. 1984. Effects of ash leachates on growth and development of Armillaria mellea in culture. USDA Forest Service Research Note PNW-418. Pacific Northwest Forest and Range Experiment Station, Portland, Oregon. 11 p. ___. 1990. The effects of Trichoderma spp. isolated from burned and non-bumed forest soils on the growth and development of Armillaria ostoyae in culture. Northwest Science 64:39-44. SAS Institute, Inc. 1994. SAS User's Guide to Statistics. SAS Institute, Inc., Cary, North Carolina. Schmitt, c.L.; Goheen, D.J.; Gregg, T.F.; Hessburg, P.F. 1991. Effects of management activities and stand type on pest-caused losses in mixed conifer stands on the Wallowa-Whitman National Forest. USDA Forest Service BMPMZ-01-91. Wallowa-Whitman National Forest, La Grande, Oregon. 78 p. Thies, W.G. 1990. Effects of prescribed fire on diseases of conifers. ln Natural and Prescribed Fire in Pacific Northwest Forests. Walstad, J.D.; Radosevich, S.R.; Sandberg, D.V. (eds.). Oregon State University Press, Corvall is, Oregon. Pp. 117-121. 84 Ecology and Biodiversity EFFECTS OF NUTRIENTS ON ARMILLARIA ROOT DISEASE IN GREENHOUSE­ GROWN LODGEPOLE PINE (PINUS CONTORTA) K.I. Mallett and D.G. Maynard Natural Resources Canada, Canadian Forest Service, Northern Forestry Centre, 5320 - 122 St. Edmonton, AB. T6H 3S5. Telephone: (403) 435-7314 FAX: (403) 435-7359 Email: KmaIlett@NRCAN.GC.Ca SUMMARY Armillaria root disease is thought to occur in nutrient stressed trees; however, sorne studies have found a greater incidence of Armillaria root disease on high versus low productivity sites and in fertilized versus non­ fertilized plantations. A greenhouse experiment was initiated to test whether young lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.) grown under low nutrient conditions were more susceptible to Armillaria root disease than trees grown under optimal nutrient conditions. Lodgepole pine trees, grown in the greenhouse and fertilized with either fuIl- or quarter-strength Hoagland's solution were inoculated with Armillaria ostoyae (Romagn.) Herink. Trees from each treatment were examined 3, 6, 9, and 12 months after inoculation for evidence of infection. Disease incidence was greater in trees grown with the full-fertilizer treatment than those grown with the quarter-strength fertilizer treatrnent; however, inoculurn survival was better in the quarter treatment. Infection was not related to size of the tree. The increased incidence of Armillaria root disease in the full-strength fertilizer treatment may be related to changes in the chernical constituents of the tree root or the fungus's ability to use the food base in creating rhizomorphs or mycelia. Keywords: Armillaria root disease, Armillaria ostoyae, stress, nutrient INTRODUCTION Armillaria root disease is one of the most important diseases of forest trees and occurs in aIl forested regions of the globe on a variety of host species (Wargo and Shaw 1985). In Canada, Armillaria root disease is found in aU forest regions and can cause significant losses in coniferous and deciduous tree stands of aIl ages (Hiratsuka 1987, Myren 1994). Until recent1y, Armillaria root disease was considered to be a disease of severely stressed trees (Wargo and Harrington 1991) and the incitant, Armillaria mellea (sensu lato), was thought to be a secondary pathogen. With the discovery that there are many different species of Armillaria, it became apparent that sorne species such as Armillaria ostoyae (Romagn.) Herink were prirnary pathogens (Gregory et al. 1991). Armillaria ostoyae is the principal species of Armillaria found in conifers in western North America (Wargo and Shaw 1985, Mallett 1992). Little is known about the environmental conditions that must be present for these primary pathogenic species of Armillaria to damage trees. A higher incidence of Armillaria TOot disease is thought to occur in trees growing on nutrient deficient sites (Wargo and Harrington 1991). The disease in conifers has been associated with low soit nitrogen (Singh 1983, Entry et al. 1986, 1991). Very few studies have examined the effect of edaphic factors on A. ostoyae and Armillaria TOot disease (Shields and Hobbs 1970, Entry et al. 1986, 1991, Blenis et al. 1989). In a greenhouse study, soit type significant1y affected the inoculum viability, rhizomorph production, and amount of Armillaria root disease in lodgepole pine (Pinus contorta Dougl. var. latifolia Engelm.) caused by A. ostoyae (Blenis et al. 1989). Seedlings grown in a soit from a highly productive site (site productivity was measured by periodic annual increment) had the least incidence of disease whereas those grown in soil from a low productivity site had the greatest incidence of disease. In contrast, a life table study of 10dgepole pine in west central Alberta showed that there was a greater incidence of mortality due to Armillaria TOot disease on high versus low productivity sites (Ives and Rentz 1993). Ecology and Biodiversity 85 In this study we report on the effect of soit nutrients on Armillaria root disease caused by A. ostoyae, in greenhouse-grown lodgepole pine seedlings. Specifically we tested whether lodgepole pine seedling grown under low nutrient conditions were more susceptible to Armillaria root disease than seedlings grown with optimal nutrients. Seedling heights, diameters, shoot and root dry weights as weil as elemental analysis of seedling tissues were measured to determine the response to nutrient stress. MATERIALS AND METHOnS Isolate identification Two isolates of Armillaria ostoyae (NoF 1076 and NOF 898) isolated from basidiocarps were used in the experiments. Armillaria species identifications were made by haploid pairing tests using basidiospore isolates from basidiocarps (Anderson and Ullrich 1979). Diploid isolates used were isolated from the stipe tissue of basidiocarps. Ali isolates were grown on carrot agar (Mallett and Colotelo 1984). Inoculum preparation Inoculum was prepared from branch segments of Populus tremuloides Michx. (approximately 10 x 2 cm). Branch segments were autoclaved in metal surgical instrument trays (46 x 12.5 x 6 cm) containing 300 ml of water (60 branch segments per tray) for 60 min. Three hundred milliliters of malt extract-dextrose-peptone broth (MPDB) (3% w:v malt extract 2% dextrose, 0.5% peptone, in distilled water) was then added to each tray before autoclaving for an additional 20 min. Isolate cultures grown on carrot agar were comminuted in 100 ml of sterile distilled water in a sterile blender and then added aseptically to the trays containing the branch segments and MDPB. Inoculated branch segments were kept at 25°C in the dark for 3 months. Fertilizer experiment Lodgepole pine seeds, collected near Hinton, Alberta, were seeded in 3 litre plastic pots containing limed peat moss (pH 5), three seeds per pot. The exposed surface of the peat was covered with horticultural grit (approximately 0.5 cm in depth) to control bryophyte growth. After the seedlings had emerged, seedlings were thinned to one seedling per pot. Then an inverted 2 x 25 cm test tube was inserted into the soil parallel to each seedling's stem. Seedlings were grown in a greenhouse compartment with artificial light (high pressure sodium vapour lamps, 400 W, with an intensity of 363 !lmol m-2s- l ) with a photoperiod of 18 h and day and night temperatures of25°C and 20°C, respectively. The seedlings were watered twice weekly and fertilized once a week with full-strength Hoagland's No. 2 solution (l ml of 1 M KH2P04, 5 ml of 1 M KN03, 5 ml of 1 M Ca(N03)2 2 ml of 1 M MgS04, 0.5% Fe tartrate 1 ml microelement stock [0.286% H3B03, 0.181 % MnCIz4H20, 0.008% ZnS04"7H20, H2Mo04H 20] 1 L distilled water) (Hoagland and Amon 1950) for 4 months. Four months after seeding, half of the 384 seedlings were fertilized with full-strength Hoagland's and the remaining seedlings were fertilized with quarter-strength Hoagland's solution. Three hundred milliliters of fertilizer solution were dispensed to each pot. Seedlings in each fertilizer treatrnent were further subdivided into those that received isolate NOF-I076, NOF-898, or a sterile inoculum piece. The seedlings were inoculated when they were 6-months old by removing the test tubes and replacing them with inoculum pieces. The seedlings in the experiment were assigned to a treatrnent in one of four blocks (greenhouse benches) in a randomized complete block design. The heights and diameters (measured at the soilline) of the seedlings were recorded at the time of inoculation. Sixteen seedlings from each treatrnent were measured and examined 3, 6, 9, and 12 months after inoculation. Heights and diameters of the seedlings were recorded before the roots were examined for symptoms and signs of Armillaria root disease. Those seedlings with resinous lesions and an attached rhizomorph or a characteristic mycelial fan of Armillaria were considered diseased. Seedlings that died during the experiment were examined for fans of mycelium beneath the bark. Cultures were made from mycelial fans to confirm that it 86 Ecology and Biodiversity was A. ostoyae. Species identification was made using the method of (Hopkin et al. 1989). Root and shoot dry weights were obtained by separating roots, stems, and foliage and then drying them in ovens at 60°C to a constant weight. Inoculum pieces were retrieved and examined. If rhizomorphs, yellow stringy rot, and/or a mycelial fan were observed, the inoculum was considered viable. Elemental analysis Total nitrogen (N), phosphorus (P), sulfur (S), calcium (Ca), magnesium (Mg), and potassium (K) were detennined for the needles, stems, and roots of all seedlings. Total nutrient analyses were performed on oven­ dried samples that had been ground in a Wiley mill and passed through a 0.25 mm sieve. The samples were digested with HN03-H20 2-HCl in a microwave oyen (Kalra et al. 1989) and analysed for S, P, Ca, Mg, and K. Total N was determined using a modified Kjeldahl digestion technique with a Tecator Kjeltec 1030 automated system (Kalra and Maynard 1991). Samples were digested in an aluminum block digester using an H2S04 and K2S04-CUS04 catalyst mixture (Kjeltab). Statistical analysis Annillaria root disease incidence and inoculum survival data were analysed using a linear model in the Categorical data modeling procedure (CATMOD) of the Statistical Analysis System (SAS) Institute, Inc. (Statistical Analysis System Institute, Inc. 1990. SAS/STAT user's guide. Vol 1,2. Ver. 6. 4th ed. SAS Institute, Cary, North Carolina). The Chi-square statistic was used to analyze the disease incidence and inoculum survival data because it does not assume any underlying distribution of the data. The foliar, stem, and root nutrient and the growth data were analyzed by analysis of variance (ANOVA) using the generallinear model (GLM) procedure of SAS. Least square means (LSMEANS), using Fisher's protected least significant difference, was utilized for mean comparisons where significant effects were detected by the ANOVA. RESULTS Different concentrations of Hoagland's solution had an effect on Armillaria root disease in lodgepole pine. There was a greater incidence of disease in the full-strength fertilizer treatrnent compared to the quarter­ strength treatment (Table 1, Chi-square = 4.03, P = 0.04). Only seedlings that were exposed to viable inoculum were considered in the analysis. A greater incidence of disease occurred in the seedlings that received the full­ strength fertilizer treatrnent at each sampling date. There was no difference in the incidence of Annillaria root disease (infected + dead seedlings) caused by the two isolates (Chi-square = 1.97, P = 0.16); however, isolate NOF-I076 caused mortality, NOF-898 did not. There was no significant difference in the amount of mortality between the full- and quarter-strength treatments (7.3 and 6.7%, respectively) or when it occurred. Inoculum in the quarter-strength treatrnent had slightly greater survival than the inoculum in the fuIl­ strength treatrnent except at the second sampling date (Table 2, Chi-square = 3.62, P = 0.06). There was a significant difference in inoculum survival between isolates regardless oftreatrnent; inoculum pieces ofNOF-898 did not as survive as weil as NOF-I076 (Chi-square = 23.3, P = 0.00). Selected growth characteristics of non-inoculated seedlings in each of the treatrnents after one year were compared (Fig.1). There were no differences in dry root weights, height and diameters (at the beginning of the experiment) between the two treatrnents; however, there were differences in the dry shoot weights and the heights and diameters of the seedlings after 12 months growth. Seedlings in the fuil-strength treatment were larger than seedlings in the quarter-strength treatment. There was no difference between infected and non-inoculated seedlings in dry root weight (P = 0.14), dry shoot weight (P = 0.71), height (P = 0.61) or diameter (P = 0.97). The concentrations of N, P, and S in the needles, stems and roots of non-inoculated seedlings fertilized with the fuil-strength Hoagland's solution were higher than in the needles, stems and roots of seedlings fertilized Ecology and Biodiversity 87 with quarter-strength Hoagland's solution (Table 3). There were no differences in the concentrations of Ca, Mg, and K in any plant part between nutrient treatments (data not shown). DISCUSSION Armillaria root disease was greatest and occurred earlier in the full-strength treatment compared to the quarter-strength treatment. Seedlings in the quarter strength treatment were nutrient stressed as shown by the significant difference in seedling height, diameter, and nutrient (N, P, S) content of the seedlings as compared to the full strength seedlings. Increased Armillaria root disease has been observed in forest stands fertilized with N (Rykowski 1981 a, b, Entry et al. 1991 b). Rykowski (1981 a, b) found that the mortality of seedlings was higher in two ofthree Scots pine plantations fertilized with N, P, K, Mg and Ca. Entry et al (1991b) observed the highest rate of infection by A. ostoyae in a Douglas-fir stand thinned and fertilized with 360 kg N ha- I compared to unthinned, unfertilized and thinned, unfertilized stands. In addition, a lifetable study of young lodgepole pine in west-central Alberta found a greater incidence of Armillaria root disease on high productivity sites versus medium and 10w productivity sites (Ives and Rentz 1993). In contrast, severe nitrogen deficiency has been associated with a higher incidence of Armillaria root disease (Entry et al. 1986, 1991 a, Singh 1983). Under the severe N deficient conditions created in these studies, seedlings may be more susceptible to Armillaria root disease. It is unlikely, however, such severe conditions would exist in most forests. We provided sufficient N to maintain the seedlings, although growth in the quarter­ strength treatment was limited and the N concentration of the needles was in the range considered deficient (Morrison 1974, Maynard and Fairbams 1994). The contrasting results could also be due to other factors, such as virulence of the A. ostoyae isolates involved, in addition to nutrient availability and the severity of the nutrient stress. Added N may increase Armillaria root disease by either effecting the chemical constituents of the tree root as suggested by Entry et al. (199Ia, b) or effecting the fungus directly. Entry et al (199Ia, b) suggested nutrient imbalances (either too little or too much) may have decreased the concentration of phenolic and lignin compounds in root bark tissue while increasing the sugar content making the tree more susceptible to A. ostoyae. Conversely, the effect of the increased nutrients in the full-strength treatment may have been on the fungus's ability to use the food base in creating rhizomorphs or mycelia (in the case ofroot to inoculum piece contact) with greater inoculum potential. There was a trend for 10wer inoculum survival in the full-strength treatment than in the quarter-strength treatment. The decrease in inoculurn survival in the full-strength treatment may have been associated with an increase in the populations of other soil microorganisms due to a higher concentration of soil nutrients (Alexander 1977, Cook and Baker 1983). These organisms may have been antagonistic to A. ostoyae. Edaphic factors such as nutrients may not only predispose seedlings to Armillaria root disease but may affect inoculum quality. Abundant soil nutrients may increase the ability of A. ostoyae to utilize its food base quickly, possibly by decreasing the carbon to nitrogen ratio, creating a greater inoculum potential. Therefore, seedlings growing on high productivity sites, with respect to nutrients, may be at greater risk of contracting Armillaria root disease than seedlings growing on less productive sites. This will require further investigation, however, before appropriate management prescriptions can be made for specific sites. 88 Ecology and Biodiversity Table 1. The effect of full- and quarter-strength Hoagland's solution on Armillaria root disease in lodgepole pine seedlings caused by two isolates of Armillaria ostoyae (NOf-\076 and NOf-898) three to twelve months after inoculation. Isolate Fertilizer Health Months after Total Percentage treatment Status inoculation (%) 3 6 9 12 NOf-1076 Full strength infected 1 0 4 7 12 21.8 dead 0 0 3 1 4 7.3 healthy 12 14 9 4 39 70.9 Total 13 14 16 12 55 100.0 NOf-l076 Quarter strength infected 0 1 4 2 7 11.7 dead 0 0 2 2 4 6.6 healthy 15 13 9 12 49 81.7 Total 15 14 15 16 60 100.0 NOF-898 Full strength infected 0 5 0 3 8 22.2 dead 0 0 0 0 0 0.0 healthy 9 5 8 6 28 77.8 Total 9 10 8 9 36 100.0 NOF-898 Quarter strength infected 0 1 0 3 4 9.5 dead 0 0 0 0 0 0.0 healthy 10 6 13 9 38 90.5 Total 10 7 13 12 42 100.0 • Count Data Ecology and Biodiversity 89 Table 2. The effect of full- and quarter-strength Hoagland's solution on inoculum survival of two isolates of Armillaria ostoyae (NOF-1076 and NOF-898). Isolate Fertilizer Status Months after Total Percentage treatment inoculation (%) 3 6 9 12 NOF-I076 Full strength inviable 3 2 0 4 9 14.1 viable 13 14 16 12 55 85.9 Total 16 16 16 16 64 100.0 NOF-1076 Quarter strength inviable 2 1 0 4 6.2 viable 15 14 15 16 60 93.8 Total 16 16 16 16 64 100.0 NOF-898 Full strength inviable 7 6 8 7 28 43.7 viable 9 10 8 9 36 56.3 Total 16 16 16 16 64 100.0 NOF-898 Quarter strength inviable 4 9 3 4 20 32.3 viable 10 7 13 12 42 67.7 Total 14 16 16 16 62 100.0 • Count Data Table 3. Total Nitrogen (N), Phosphorus (P), and Sulfur (S) concentrations (g kg-') in needles, stems, and roots of 16-month-old lodgepole pine seedlings treated with quarter- and full-strength Hoagland's solution. Values are means ± standard error. Nutrient Needles Stems Roots Full Quarter Full Quarter Full Quarter strength strength strength strength strength strength N 13.58 ± 0.43 7.38 ± 0.40 12.28 ± 0.42 5.72 ± 0.39 13.38 ± 0.50 5.13±0.47 P=O.OOOI P=O.OOOI P=O.OOOI P 1.53 ± 0.04 0.87 ± 0.04 1.98 ± 0.11 1.23 ± 0.10 1.95 ± 0.09 1.13 ± 0.08 P=O.OOOI P=O.OOOI P=O.OOOI S 2.07 ± 0.12 1.51 ± 0.11 1.70 ± 0.08 1.25 ±O.O8 2.48 ± 0.11 1.47±0.10 P=0.001 P=0.0002 P=O.OOOI • P value for the comparison between full- and quarter-strength treatrnents 90 Ecology and Biodiversity 80 -Q)-­.... ..c 60 Q) "Ci> ~....oo ~ ~ o 40.... o o ..c en 20 a oc:x:xJ full =:::::J quarter T T T ~ 80 -E E--~ 60 Q).... Q) E ct! "'C ~ 0-40 E ü--.... ..c Q) Q) ..c 20 a sht. wt rt. wt. hgt. Growth parameter dia. Figure 1. Means of growth parameters of non-inoculated greenhouse-grown lodgepole pine seedlings treated with full- and quarter-strength Hoagland's solutions (sht. wt.= shoot weight, rt. wt. = root weight, hgt. = height, dia. = diameter). Different letters for each growth parameter indicate a statistically significant difference (P = 0.05). Vertical bars = ± 1 S.E. ACKNOWLEDGMENTS We thank C. Myrholm, F. Radford, S. Graham, D. Williams, and Y. Kalra for excellent technical assistance and Drs. R. Blanchette and Y. Hiratsuka for reviewing the manuscript. Financial support from Canada­ Alberta Partnership Agreement in Forestry and the Green Plan, Integrated Forest Pest Management Initiative is gratefully acknowledged. REFERE CES Alexander M. 1977. Soil microbiology. Second edition. John Wiley and Sons. New York. pp. 467. Anderson, J. B., and UlIrich, R.C. 1979. Biological species of Armillaria mellea in North America. Mycologia, 78:837-839. Blenis, P.V., Mugala, M.S., and Hiratsuka, Y. 1989. Soil affects Armillaria root rot of lodgepole pine. Cano J. For. Res. 19: 1638-1641. Cook, R.J., and Baker, K.F. 1983. The nature and practice of biological control of plant pathogens. American Phytopathological Society. St. Paul. pp. 539. Ecology and Biodiversity 91 Entry, J.A., Martin, N.E., Cromack Jr., K., and Stafford, S. 1986. Light and nutrient limitation in Pinus monticola: Seedling susceptibility to Annillaria infection. For. Ecol. Manage. 17: 189-198. Entry, J.A., Cromack Jr., K, Hansen, E. and Waring, R. 1991a. Response of western coniferous seedlings to infection by Armillaria ostoyae under limited light and nitrogen. Phytopathology 81: 89-94. Entry, J.A., Cromack Jr., K., Kelsey, R.G. and Martin, N.E. 1991b. Response of Douglas-fir to infection by Armillaria ostoyae after thinning thinning plus ferti1ization. Phytopathology 81: 682-689. J.A., Martin, N.E., Cromack Jr., K, and Stafford, S.G. 1986. Light and nutrient limitation in Pinus monticola: Seedling susceptibility to Armillaria infection. For. Ecol. Manage. 17: 189-198. Gregory, S.c., Rishbeth, J., and Shaw, III., c.G. 1991 Pathogenicity and virulence. In: Annil1aria root disease. Ed. c.G. Shaw III and G.A. Kile. USDA, For. Serv., Agricu1tura1 Handbook No. 691. Washington, D.C. pp. 76-87. Hiratsuka, Y. 1987. Forest tree diseases of the prairie provinces. Cano For. Serv., North. For. Cent., Edmonton, Alberta. Inf. Rep. NOR-X-286. Hoagland, D.R., and Amon, D.I. 1950. The water culture method of growing plants without soil. USDA, For. Serv., Calif. For. Range. Exp. Stn., Berkeley, California., Circ. 347. Hopkin, A.A., Mal1ett, KI., and Bienis, P.V. 1989. The use of L-DOPA to enhance visualization of the "black line" between species of the Armillaria mellea complex. Cano J. Bot. 67: 15-17. Ives, W.G.H., and Rentz, c.L. 1993. Factors affecting the survival of immature lodgepole pine in the foothills of west-central Alberta. For. Can., Northwest Reg., North. For. Cent., Edmonton, Alberta. Inf. Rep. NOR-X­ 330. Kalra, Y.P., and Maynard, D.G. 1991. Methods manual for forest soil and plant analysis. For. Can., Northwest Reg., North. For. Cent., Edmonton, Alberta. Inf. Rep. NOR-X-311. Kalra, Y.P., Maynard, D.G., and Radford, F.G. 1989. Microwave digestion of tree foliage for multi-element analysis. Cano J. For. Res. 19: 981-985. Mal1ett, KI. 1992. Armillaria root rot in the Canadian prairie provinces. For. Cano Northwest Reg., North. For. Cent., Edmonton, Alberta. Inf. Rep. NOR-X-329. Mallett, K.I. and Colotelo, N. 1984. Rhizomorph exudate of Armillaria mellea. Cano J. Micro. 30: 1247-1252. Mallett, KI., and Hiratsuka, Y. 1988. Inoculation studies of lodgepole pine with Alberta isolates of the Armillaria mellea complex. Cano J. For. Res. 18: 292-296. Maynard, D.G., and Fairbarns, M.D. 1994. Boreal ecosystem dynamics of ARNEWS plots: baseline studies in the prairie provinces. Nat. Res. Can., Cano For. Serv., Northwest Reg., North. For. Cent., Edmonton, Alberta. Inf. Rep. NOR-X-327. Morrison, I.K 1974. Mineral nutrition of conifers with special reference to nutrient status interpretation: a review ofliterature. Environ. Can., Cano For. Serv., Ottawa, Ontario. Publ. 1343. Myren, D.T. 1994. Tree diseases of eastern Canada. Nat. Res. Can., Cano For. Serv., Sei. Sustainable Dev. Dir., Ottawa, Ontario. Rykowski, K. 1981 a. The influence offertilizers on the occurrence of Armillaria mellea in Scots pine plantations. 1. Evaluation of the health of fertilized and non-fertilized plantations and the variabil ity of A. mellea in the areas investigated. Eur. J. For. Pathol. Il: 108-119. Rykowski, K. 1981 b. The influence of ferti1izers on the occurrence of Armillaria mellea in Scotch pine plantations. II. The influence of Armillaria mellea on chemical changes in needles and wood of roots under minerai fertilization. Eur. J. For. Pathol. II: 178-186. Shields, Jr., W.J., and Hobbs, S.D. 1979. Soil nutrient levels and pH associated with Armillariella mellea on conifers in northern Idaho. Cano 1. For. Res. 9: 45-48 Singh, P. 1983. Annillaria root rot: Influence of soil nutrients and pH on the susceptibility of conifer species to the disease. Eur. J. For. Pathol. 13: 92-101 Wargo, P.M., and Harrington, T.C. 1991. Host stress and susceptibility. In: Armillaria root disease. Ed. c.G. Shaw, III and G.A. Kile. USDA, For. Serv., Agricultural Handbook No. 691. Washington, D.C. pp. 88­ 101. Wargo, P.M., and Shaw, III, c.G. 1985. Armillaria root rot: The puzzle is being solved. Plant Dis. 69:826-832. Whitney, R. D. 1984. Site variation of Armillaria mellea in three Ontario conifers. In: Proceedings of the sixth international conference on raot and butt rots of forest trees. Ed. G.A. Kile. Melbourne, Victoria, and Gympie, Queensland, Australia. Aug. 25-31 1983.CISRO Melbourne. 92 Ecology and Biodiversity INVESTIGATIONS ON THE DISTRIBUTION AND ECOLOGY OF ARMILLARIA SPECIES IN ALBANIA B.M. Lushaj*, M. Intini**, and E. Gupe* * Instituti i Kerkimeve dhe Manaxhimit te Mjedisit, Pyjeve, Kullotave dhe Shfrytezimit Pyjore, Rruga "Kongresi i Lushanjes" 33/1/5, Kutia Postale nr. 74, Tirana, Albania. E-mail: bmlushaj@hotmail.com ** Istituto per la Patologia degli Alberi Forestali dei CNR, - Piazzale delle Cascine 28 -Firenze, ltalia ABSTRACT Five Armillaria species were identified in a nation-wide survey in Albania: Armillaria mellea sensu stricto was found on several conifers and broad-leaved trees in most of the areas examined, except the high altitudes (above 1100-1200 m) in northern Albania. It was found to cause sorne damage to fir, oak, beech, chestnut, poplar, and hop hornbeam, and significant damage to fruit trees and grapevme. Armillaria gallica was common in conifer and broad-leafed-tree forests at altitudes from 600 m to 1600 m, less common at lower altitudes on oaks. The fungus is a weak parasite or a saprophyte of forest trees and was only occasionally found on cultivated plants. Armillaria ostoyae was rare in central and southern Albania, but is widely distributed in the forests of the northern Albania, causing significant damage to several conifers at altitudes from 600 to 1800 m's', occasionally also at lover altitudes. It was not found in fruit orchards and vineyards. Armillaria cepistipes was recorded at high altitudes from 800 to 1800 m's', mostly as a saprophyte in conifer and broad-Ieafed-tree forests, predominating in beech and Silver fir forests. Armillaria tabescens was found mostly in oak forests at altitudes from 0 to 900 m. In fruit orchards, it was occasionally found to cause disease on almond trees and pear trees (Prunus spp.). Keywords: Agaricales, Armillaria mellea, A. gallica, A. tabescens A. ostoyae, A. cepistipes, compatibility test, distribution, ecology, host preference INTRODUCTION Seven Armillaria species have been described in Europe. Six of them are wood-decay fungi with a wide distribution and they are of great ecological and economical importance (Kile et al. 1991; Guillaumin et al. 1993). The distribution and ecological characteristics of Armillaria species are fairly weil known throughout western and northern Europe, but limited information is available from eastern Europe and from the Balkan region, including Albania. Six Armillaria species have been reported from Slovenia, the northemmost Balkan country (Munda 1997), and five species have been found in the southemmost country, Greece (Tsopelas 1999). The rare European species, A. ectypa (Fr.) Lamoure was not found in either ofthese countries, and A. borealis Marxm. & Korhonen was also absent in the material investigated from Greece. Ecology and Biodiversity 93 Five Armillaria species have been reported from Albania: A. mellea sensu stricto (Vahl: Fr.) Kummer, A. gallica Marxm. & Romagn., A. tabescens (Scop.) Emel, A. ostoyae (Romagn.) Herink and A. cepistipes Velen. Sorne aspects of their distribution and ecology have been published in preliminary reports in Albanian language (Lushaj 1992; Lushaj and Shyti 1995; Intini and Lushaj 1998). The work reported here gives more detailed information on the distribution, ecology and host range of Armillaria species in Albania. Furthermore, it focuses on the damage caused by these fungi in different types of forests, fruit orchards and vineyards. MATERIALS AND METHODS Study sites and hosts The presence of Armillaria species was investigated from 1990 to 2000, both in forest and agricultural areas. Permanent experimental plots were established in many parts of Albania in different types of forests, fruit orchards and vineyards at altitudes from sea level to 2100 m, at different levels of surveys. In addition, surveys were made in aIl main districts of the country, from the Tropoje and Shkoder districts in the north to Kosove borders in the east and to the Gjirokaster and Korçe districts in the south, to Greece borders. The surveyed forests, orchards and vineyards were selected and permanent sample plots were established, for the first period, during the years 1992-1995, in a non-systematic manner in sorne areas, in the most attacked trees (Lushaj 1992; Lushaj et al. 1995), and in a systematic manner on permanent sample monitoring plots by a network gird 10 x 10 km of the National Assessment and Monitoring System in Albania, for the second and third periods, during the years 1996-2000. The investigations concemed largely natural forests, while a limited number of reforested areas were examined. Efforts were made to inc1ude many different forest types from several locations. Altogether, 42 different forest areas were investigated, occurring in 40 mountainous forest areas (data available from the author) and on two non-mountainous forest areas. The monitoring plots (50 x 50 m = 2500 m2 ) were laid out on a grid 10 x 10 km, based on a statistically representative method. On each monitoring plot, 30 forest trees were examined for the presence of Armillaria sp. (Lushaj et al. 1996, 1997; Lushaj 1997, 2000). Altogether, forest trees were investigated in 15 different areas of Albania, belonging to 15 districts. Fruit orchards and vineyards were also investigated in 15 different areas of Albania, belonging to 15 districts. Field observations and sampie collection The majority of collections were made at middle and high altitude coniferous forests consisting of Silver fir (Abies alba L.), Greek fir (A. borisi-regis Malf.), Austrian black pine (Pinus nigra Am.), Scots pine (P. sylvestris L.), Norway spruce (Picea abies (L.) Karst.), etc. Deciduous forests, consisting mainly of several oak species, were examined at low and lower altitudes, while beech (Fagus sylvatica L.) was examined at higher altitudes. Hop hombeam (Ostria carpinifolia Scop.), Aleppo pine (Pinus halepensis Mill.), poplar species (Populus spp.), maritime pine (Pinus pinaster L.) and eucalyptus species (Eucalyptus spp.) were also growing in these forests. At the same time, the majority of collections were made at low and middle altitudes in fruit orchards and vineyards. Collections were made from a variety of hosts in these forests as weil as in fruit orchards and vineyards. Dead or unhealthy trees, wind-broken trees and cut trees were excavated, and the root collar and a number of roots were inspected for the presence of the characteristic mycelial mats of Armillaria under the bark and/or rhizomorphs on the roots. Samples of decayed wood tree bark and rhizomorphs were collected. The presence of basidiocarps was noted and it was recorded whether they were associated with dead or living trees, stumps or wood debris in the ground. The location and altitude were recorded and notes were taken on the condition of the 94 Ecology and Biodiversity host, with reference to apparent pathogenicity of the fungus and presence of obvious stress factors (Blodgett and Worrall 1992). During the study years, over 2500 living or dead trees and 600 stumps were investigated in forests, fruit orchards and vineyards of Albania. Culture media and incubation conditions Isolations from basidiospores, wood samples and rhizomorphs were made on potato dextrose agar (PDA) amended with fungicides and antibiotics, on malt extract agar 3% (MEA) and on carrot agar (CA) in Petri plates (Intini and Gabucci 1987; Guillaumin et a1.1989). Morphological studies of diploid cultures were performed on all these culture media. Monosporous isolations were made on 1-2% MEA (Difco, Detroit MI). All the cultures were incubated in darkness at 24°C. ln order to produce basidiocarps in vitro, two types of substrates were used in 500 ml Erlenmeyer flasks: (a) about 100 g of small pieces of oak branches (1.5-2.5 cm in diameter and 2-3 cm long) in 150 ml of deionised water; and (b) a complex medium, consisting of 25 g whole grain rise, 15 g of beech or oak sawdust and 230 ml of 0.5% peptone solution in deionised water (Intini 1991; Tirra 1991). The substrates were autoclaved two times at 121°C for 1 h with 24-h interval. After inoculation with the vegetative mycelia of Armillaria species the flasks were incubated in the darkness at 24°C for about 1.5 month. The temperature was then lowered to 15°C and the cultures were exposed to light with a photoperiod of 11-h (Guillaumin et al. 1989; Tirra and Intini 1989; Tirra 1991; Intini 1993). Isolations Small chips from newly exposed wood, mycelial mats or basidiocarps were plated directly on the surface of PDA, CA or MEA medium. Rhizomorphs were surface sterilized in a solution of NaOCI and 20% ethanol for 2-3 min., then rinsed with sterile water and plated on PDA, CA and MEA (Blodgett and Worrall 1992). Monosporous cultures were isolated both from basidiocarps collected in the field and from those developed in vitro. Sporulating caps were placed on the lid of a plastic Petri plate (9 cm), in which an opening had been made with a hot surgical blade. Basidiospores were allowed to settle on the surface of the MEA through this opening. After incubation for 24 h, single germinated spores were picked up under a microscope and transferred to a fresh MEA medium with a modified Pasteur pipette (Korhonen and Hintikka 1980). Pairings Haploid and diploid isolates were identified in compatibility tests with known haploid tester strains of the European Armillaria species (Korhonen 1978; Guillaumin et aI.1991). In total, about 40 haploid testers were used in this study; most of them originated from Italy and other European countries, sorne were from Albania. Each unknown isolate was paired with at least three different tester strains of each Armillaria species. Pairings were performed by placing two inocula 1-2 mm apart on the surface of the medium (two pairings per plate). Inocula were taken from the margin of a growing culture with the aid of a modified Pasteur pipette (Korhonen and Hintikka 1980). These inocula were cylinders of agar medium approximately 1 mm in diameter and 2 mm in length. They contained mainly submerged mycelium without crust or rhizomorphs and little aerial mycelium (in initial experiments where the diploid inocula with the dense crustose mycelium were used the mating reaction was not always clear). ln haploid-haploid pairings, 3-4 weeks of incubation were sufficient for the observation of the mating reaction, while in the diploid-haploid pairings the time of incubation was prolonged sometimes up to 6 weeks or more (Guillaumin et al. 1991). Before mating tests, diploid cultures of many isolates were preliminarily identified on the basis of culture morphology, so the number of tester strains in diploid-haploid pairings could be reduced to two or three Armillaria species. Ecology and Biodiversity 95 In case the results were not clear in diploid-haploid confrontations, the pairings were repeated using new tester strains (Guillaumin et al. 1991). AIso, in sorne of these pairings, two to three plugs were taken from the side of the haploid testers (3-5 mm past the confrontation line) and they were subcultured in a plate together with the haploid tester. The morphology of these two subcultures, in relation to the reduction in the amount of aerial mycelium, was compared after 2 weeks incubation (Rizzo and Harrington 1992). RESULTS Field observations The surveys carried out during the study years revealed that the average frequency of Armillaria disease was 4% for forests and 6% for orchards. In forests of fir, beech, chestnut, oaks, pine and poplar, the infection frequency was up to 15%, and the incidence of damage (decline) was ranked to the first class (Lushaj et al. 1996, 1997,1998; Lushaj 1997,2000, 2001a, b). Fructification in vitro Ali the five Albanian Armillaria species, A. mellea, A.gallica, A. ostoyae, A.cepistipes and A. tabescens, produced basidiocarps and sorne of them only primordia, in vitro. Basidiocarps were produced on both types of media, but fruiting was more frequent in the complex medium that contained rice, sawdust and peptone. Ali isolates of A. tabescens and A. ostoyae fruited on both types of media, but only 31 % of A. cepistipes, 26% of A. ga/lica and 18% of A. mellea isolates fruited, and the fruiting took place only on the complex medium. Basidiocarps developed usually 1-2 months after the cultures were exposed to light, but certain cultures fruited only after 4-7 months incubation in light conditions. Species identification A number of collections were identified in the field on the basis ofbasidiocarp morphology. Basidiocarps were usually observed in October, November and early December. However, the level of fruiting of Armillaria was low during most years of survey. An abundance of basidiocarps was observed in certain areas of Albania during October-November in 1996 and 1997, as a result ofearly rains. The majorities of Armillaria cultures were isolated from vegetative organs and were considered diploid. Therefore, the species identification was mainly based on diploid-haploid pairings. When monosporous isolates were available, either from natural or cultivated fruit bodies, they were identified in haploid - haploid pairings. In most cases, these pairings verified the species identification that was based on the morphology of the basidiocarps. Identification of most Armillaria isolates was also possible on the basis of morphological characteristics of diploid isolates on MEA, CA and PDA culture media (Fig. 1). In total, 576 Armillaria isolates were identified from 43 different host species (Tables 1 and 2). Of them, 397 were diploid isolates, which were identified, in diploid - haploid pairings. Twenty-eight of these diploid isolates produced basidiocarps in vitro and their identification was verified from basidiocarp morphology and haploid-haploid pairings. Monosporous cultures were isolated and identified from 88 basidiocarp collections found in the field. Of ail the isolates identified, 263 belonged to A. mellea, 135 to A. gallica , 69 to A. ostoyae, 49 to A. cepistipes and 60 to A. tabescens (Tables 1 and 2). 96 Ecology and Biodiversity Geographical and latitudinal distribution Armillaria mellea sensu stricto was present in most of the areas surveyed, but majority of the records cornes from northem and central Albania (data available from the author). The species occurred in conifer and in particular in broad-Ieaved forests at altitudes ranging from 0 m to 1400 m. It was quite common on the oak species (Quercus spp.) and on fir species (Abies spp.) at altitudes greater than 1100 m. However, in northem Albania the fungus becomes rare at high altitudes; there it was recorded only at altitudes less than 1200 m. A. mellea was also found in fruit orchards and vineyards throughout Albania from north to south, at altitudes up 800 m. A. gallica was less common than A. mellea in the forests of Albania, especially in the southem parts, but it was more frequent in the coniferous and broad-1eaved forests of the northem and central parts of the country. The fungus was most common at altitudes above 450 m, and occurred up to 1800 m. At low altitudes A. gallica was rare, although it was occasionally detected at the elevation close to sea level. A. gallica was found in fruit orchards and vineyards only in two areas, one of them in the south (Korçe district) and the other one in the north (Diber district). A. ostoyae was found everywhere in Albania but is common on1y in the northem part of the country. It occurred mainly in coniferous forests at altitudes higher than 1200 m and was more frequent at elevations above 1200 m. A. ostoyae was not found in fruit orchards and vineyards. A. cepistipes was found only on mountainous forest areas ofnorthem and central Albania, in beech forests at elevations from 800 to 1600 m, in coniferous forests of pine, fir or spruce and in mixed forests at altitudes ranging from 800 m to 1800 m. A. cepistipes was not found in fruit orchards and vineyards. A. tabescens was least commonly recorded of all Armillaria species. It was found predominantly on oak, sometimes on fir, poplar, eucalyptus etc. It occurred in oak forests at altitudes from 400 m to 900 m, in fir forests at altitudes from 1000 m to 1300 m, in poplar stands at altitudes from 100 m to 300 m and on eucalyptus at altitudes from 0 m to 200 m. The fungus also attacked almond orchards in two different localities: in the Diber district, northem Albania, and in Korçe, southem Albania. Ecology and parasitic behavior of Armillaria species Conifers Fir species. AlI five Armillaria species were recorded from fir forests (Table 1). A. mellea and A. ostoyae cause significant losses, killing trees of aIl ages, single trees or smalI groups of 3-1 0 trees. - A. mellea was the predominant species on Abies a/ba in northem Albania (Table 1). In mixed coniferous forests of pine, spruce, and fir at elevations from 600 m to 1000 m, A. mellea was killing fir trees selectively, while mortality of oak or pine was rare. Infection of fir by A. mellea was also quite cornrnon at higher altitudes up to 1400 m. Basidiocarps of this fungus were usually associated with dead trees or stumps, but they were also detected at the base of living trees, sorne of them with no apparent symptoms in the crown. - A. ostoyae was very cornrnon in Abies a/ba forests in northem and central Albania and in Abies borisi-regis in southem Albania. In mixed forests of fir, beech, pine and spruce the fungus was observed to kilI selectively fir trees only; no infections were noticed on beech trees. - A. gallica was present in rnost of the fir forests examined, and it was the predominant species in Abies borisi-regis forests of southem Albania. It was usually associated with srnalI­ suppressed trees in the understory, but was also found on larger dead trees. In sorne cases A. gallica was isolated from firtrees that were infected also by other diseases. - There were nine records of A. cepistipes on Abies a/ba. ­ A. tabescens was recorded only three times on fir species, on trees infected also by other diseases and / or insects. Spruce species. In Picea abies forests of Tropoje district, in the northem Albania, the rnost frequent cause ofmortality was A. ostoyae while A. gallica and A. cepistipes were recorded as saprophytes. Ecology and Biodiversity 97 Pine species. In Pinus nigra forests of northem Albania, A. ostoyae was found to cause significant problems. A. mellea caused mortality in the reforestation's of this tree species, established 25-30 years ago in cleared broad-Ieaved (beech and oak) forests in Puka district in northem Albania and in Pogradec and Korçe districts in the southem of Albania. In northem Albania, Tropoje district, A. ostoyae killed Scots pine trees. A. cepislipes was usually a saprophyte on Scots pine, with two exceptions where the fungus was considered responsible for killing a tree suppressed in the understory. A. mellea caused mortality of Aleppo pine forests at altitudes from 100 m to 350 m. The fungus was cornmon in mixed stands of Aleppo pine and deciduous oaks, but only Aleppo pine trees were infected. A. gallica was found to cause problems in reforestation of Maritime pine established 35-40 years ago at altitudes from 100 to 300 m. Other conifers. Juniperus communis L., growing in understory of other conifers, was occasionally attacked by A. mellea and A. ostoyae. These fungi were recorded also from Cupressus sempervirens in Berat district, southem Albania. Deciduous forests Beech. Many beech forests throughout Albania were investigated in this study. Four Armillaria species were detected (Table 1). A. gallica was the predominant species. In most cases, it was a saprophyte on stumps and dead wood, producing an abundance of rhizomorphs in the soil and forests litter. However, in sorne cases, A. gallica was found to kill small suppressed trees in the understory of coppice forests, where the fungus had infected the tree through the stump. In high altitude forests, A. cepislipes was observed to cause mortality of smaIl suppressed trees, but generally the fungus was a saprophyte. It produced an abundance of rhizomorphs, similar to those of A. gallica. A. mellea was found in most beech forests. A. ostoyae was recorded as a saprophyte in beech forests but in a few cases it was found to cause mortality to young beech trees. A. tabescens was not found in the beech forests. Oak species. Aiso many oak forests were investigated; A. mellea, A. gallica and A. tabescens were found. These fungi killed isolated young trees or were growing as saprophytes in old stumps. A. mellea was the predominant species on aIl species of oak: Quercus cerris, Q. petrea, Q. frainetto, Q. pubescens. Q. aegilops and Q. ilex. In most cases A. gallica was a saprophyte on stumps and dead wood of aIl species of oak. A. tabescens was also found on aIl oak species, most cornmonly on Q. ilex (Table 1). A. ostoyae and A. cepistipes were not found on oak. Other deciduous trees. A. mellea, A. gallica and A. tabescens were found in pure poplar stands (Populus spp.). A. tabescens was found on Eucalyptus species. In the pure deciduous and mixed deciduous-conifers forests of Albania, A. mellea infected Castanea saliva, Ostria carpinifolia and other species, including occasionally Fraxinus excelsior, Carpinus betulus, Platanus orientalis and Corylus avellana. Armillaria gallica was recorded on Castanea sativa, and A. mellea, A. ostoyae and A. cepistipes were found on Betula pendula and other species too. Fruit trees and garden plants Three Armillaria species were found to cause infections of Albania: A. mellea, A. gallica and A. tabescens (Table 2). A. mellea was the predominant Armillaria species on the fruit orchards, vineyards and private gardens, infecting a variety of omamental plants and trees in different areas. It was found to cause significant damage in the vineyards of every district, especially in areas c1eared from oak forests, often together with other diseases (e.g. Rosellinia spp.) or insect pests. A. mellea was recorded also from lemon tree (Citrus limon), ficus (Ficus carica), walnut (Jug/ands regia), nerium (Nerium o/eander), apple trees (Malus spp. and on Malus sylvestris), mulberry tree (M?rus nigra L. and Morus alba L), olive species (Olea spp. and Olea europaea L.), pear trees (Prunus spp.) and thuJa (Thuja orienta/is). 98 Ecology and Biodiversity Infections of A. gallica on cultivated plants were recorded in two different areas: in Korçe, southem Albania, and in Diber, northem Albania. The fungus caused mortality on cherry, apple and pear. However, the fungus other was not detected in other areas of southem Albania, and it does not seem to play an important function as a pathogen. A. tabescens was found to cause significant damage in almond tree orchards in two different localities of Albania, in Korçe and in Diber. Two orchards had been established on former forestland cleared from broad- leaved trees. According to species, the number of collections was A. mellea 277, A. gallica 135, A. ostoyae 69, A.tabescent 62 and A. cepistipes 49 (Tables 1 and 2). Table 1. Armillaria records on different host species and altitudes in the forests of Albania. No. Host species Working circles (l) Altitude in m A. A. A. A. A. North mellea gallica ostoyae cepistipes tabescens Centre South Conifers Silver tir (Abies a/ba 07,13,24,34,41,42 600-1800 10 14 14 9 2 Mill.) 700-1000 700-1300 2 Greek tir (Abies 41,42 2 4 2 borissi-regis Matt.) 700-900 3 Junipers (Juniperus 37,38,39 3 3 communis L.) 100-800 4 Norway spruces (Picea 07 900-1800 3 3 3 abies (L.) Krast.) 5 Serbian spruce (Picea 07 900-1800 3 3 3 omorica L.) 6 Austrian black pine 12,13,28,31,34 400-1200 2 3 28 (Pinus nigra Arn.), 800-1000 800-1200 7 Aleppo pines (Pinus 20,22 4 ha/epansis Mill.) 100- 300 8 Maritime pine (Pinups 20,22 2 pi/aster Sol.) 100- 300 9 Scots pine (Pinus 07,13 900-1800 4 2 sy/vestris L.) 10 Cypresses (Cupressus 37,38,39 3 3 sempervirens L.) 100-800 II Broad-Ieaved Il Birch (Betu/a pendu/a 31 6 6 6 Roth.) 800-1200 12 Hop hombeam (Ostria 01 400- 900 3 3 carpinifolia Scop.) 13 Hombeam (Carpinus II 600- 800 Ecology and Biodiversity 99 betu/us L.) 14 Sweet chestnut (Castanea saliva Mill.) 15 Hazelnut (Cory/us avellana L.) 16 Cornmon beech (Fagus sy/vatica L.) 17 Ash (Fraxinus excelsior L.) 18 Oriental planetree (Platanus orientalis L.) 19 Poplar sp. (Populus spp.) 20 Turkey oak (Quercus cerris L.) 21 Hungarian oak (Quercus frainetto Ten.) 22 Sessile oak (Quercus petrea (Matt) Liebl.) 23 Pubescent (Downy) oak (Quercus pubescens Willd.), 24 Aegilops oak (Quercus aegilops L.) 25 Holm oak (Quercus ilex L.) 26 Eucalyptus sp. (Eucaliptus spp.) Total 02,06,27,29 09 03,09,10,14,19, 21,23,31 17 22,26 37 01,04,05,08,11, 12,15,16,17,18, 25,38 II,12,15,16,17, 18,38, 01,04,05,12,15, 16,17,18,25,30 01,04,05,12,15, 16,17,18,25,28, 30,39 35 32,33,35 36,40 42 350-1100 700-900 700-900 800-1600 800-1600 900-1100 800-1200 600-800 100-400 400-600 100-300 250-900 300-900 300-900 250-900 300-900 300-900 250-900 300-900 300-900 250-900 300-900 300-900 0-200 0-600 0-200 14 19 3 26 16 37 31 8 6 198 4 24 6 24 Il Il 6 6 6 127 3 69 23 49 6 6 6 6 3 6 9 9 53 Numbers indicate the working circles (forest economy): l, Shkoder (Rosek-Drisht); 2, Shkoder (Shllak 1); 3, Shkoder (Shllak 2); 4, Has (Tregtan); 5, Has (Perollaj); 6, Tropoje (Geshtenjat e Tropojes); 7, Tropoje (Valbone-Dragobi); 8,Tropoje (Bytyç); 9, Tropoje (Nikaj-Mertur); 10, Tropoje (Koigecaj); Il, Puke (Dedaj-Bahot); 12, Puke (Puke-Fushe Ares); 13 Puke (Tuç); 14, Puke (Munelle); 15, Mirdite (Kulmet e Dervenit); 16, Mirdite (Shtane); 17, Mirdite (Lugina e Fanit te Madh); 18, Mat (Derjan 1); 19, Kruje (Qafe Shtame); 20, Kruje (Kraste); 21, Tirane (Bize); 22, Tirane (Parku i Tiranes); 23, Librazhd (Lepushe); 24, Librazhd (Qarrishte); 25, Librazhd (Drogostuje); 26, Librazhd (Librazhd); 27, Pogradec (Geshtenjat e Pogradecit); 28, Pogradec (Verdove); 29, Pogradec (Stropske); 30, Korçe (Gorice); 31, Korçe (Dardhe); 32, Vlore (Maja e Bratit); 33, Vlore (Gumenice); 34, Vlore (Llagara);35, Vlore (Himare); 36, Vlore (Kume-Shashice); 37, Berat (Hija e Roshnikut); 38, Berat (Mali i Bardhe); 39, Berat (Mali Partizan); 40, Fier (Lumi Seman-Rosk); 41, Gjirokaster (Bredhi i Sotires 1) and 42, Gjirokaster (Bredhi i Sotires 2). 100 Ecology and Biodiversity Table 2. Armillaria records on different cultivated hast species and altitudes in the fruit trees of Albania. No. Host species Altitude in m Districts (2) A. A. A. A. A. North mellea gallica ostoyae cepistipes tabescens Centre South Lemon (Citrus e,k 2 limonia Osbeck.) 0-200 2 Fig-tree (Ficus b,d,m,n,i,g,h,o 6 carica L.) 0-500 0-500 3 Walnut- tree 350-1100 a,f 2 (Juglands regia L.) 4 Wild apple-tree 350-800 a,f,m,1 4 (Malus si/vestris (L.) 350-800 Mill.) 350-800 5 Apple- tree sp. 350-800 a,f,l,m,n, 5 2 (Malus spp.) 350-800 350-800 6 White Mulberry- tree 100-600 a,f,m,n,l, 5 (Morus alba L.) 100-600 100-800 7 Black Mulberry-tree 100-600 a,f,m,n,l, 5 (Morus nigra L.) 100-600 100-800 8 Oleander (Nerium d oleander L.) 0-200 9 Olive-tree (Olea c,ç,e,n.g,h,i,k 9 europaea L.) 0-400 0-400 10 Olive-tree sp. (Olea c,ç,e n.g,h,i,k 9 spp.) 0-400 0-400 II Plum-tree (Prunus 350-900 f,1 2 4 avium 1.) 350.900 350-900 12 Plum-tree (Prunus 350-900 f,1 2 domestica L.) 350.900 350-900 13 Plum-tree sp. f,1 2 (Prunus spp.) 14 Plum-tree (Prunus 350-900 f,1 2 6 dulcis L.) 350.900 350-900 15 Plum-tree (Prunus 350-900 f,l 2 persica L.) 350.900 350-900 16 Pear-tree sp. 350-900 f,l,m 3 (Pyrus spp. 350.900 350-900 17 Wild pear -tree 350-900 f,l,m 3 (Pyrus communis L.) 350.900 350-900 18 Wall- Creeper d (Thuja orientalis L.) 0-200 19 Vine (Vitis 350-800 f,l,m,n 4 Ecology and Biodiversity 101 vinifèra L.) 100-800 0-800 20 Vineyard (vine spp.) 350-800 f,l,m,n 4 100-800 0-800 21 Peach- tree (Persica 350-900 f,l,m 3 vulgaris Mill.) 350-900 350-900 Total 15 79 8 9 (2) Letters indicate districts in which collection were made from agricultural plants and vineyards: A, Tropoje; b, Lezhe; c, Lushnje; ç, Fier; d, Tirane; e, Vlore; f Diber; g, Mallakaster; h, DUITes; k, Sarande; i, Elbasan; l, Korçe; m, Shkoder; n, Berat; 0, Kruje. DISCUSSION Albania is a mountainous country and in spite of its small size the landscapes and climatical conditions are variable (Lushaj et al. 1996, 1997). The c1imate on the coastal area is of MediteITanean type with hot and dry summers and mild and rainy winters. In inland there is a transition from the maritime c1imate to a continental and mountainous c1imate, with more abundant rainfall. Covering 28 749 km2 and located at the same latitude as central and southem Italy (Lushaj 2001 a), Albania presents a great variety of plant and fungal flora. Five Armillaria species were identified in a nation-wide survey in Albania. They inc1ude ail the European Armillaria except A. borealis and A. ectypa. Armillaria mellea sensu stricto is the most common species in Albania and the most significant pathogen in forest and cultivated plants. It was recorded from several conifers and broad-leaved trees in most areas examined, except the high altitudes above 1100-1200 m of the forest of the north of Albania. The fungus was mostly a weak parasite or a saprophyte of forest trees, but it also causes significant damage on Abies spp., Quercus spp., Fagus sylvatica, Castanea sativa, Populus spp., Ostria carpinifolia etc. Moreover, A. mellea was commonly found on cultivated plants as well in fruit orchards, causing damage especially on apple and pear species and in vineyards. The fungus is considered therrnophilic, and in most areas of central Europe it is restricted to low altitudes. In France, A. mellea has not been reported at altitudes greater than 1000 m (Guillaumin et al. 1993), but in southern Italy, it has been found at altitudes up to 1400 m (Intini 1991; Grillo et al. 1996). In Greece it was present in most areas examined, except high altitudes (above 1100 m) of northern Greece (Tsopelas 1999). The situation was about the same in Slovenia (Munda 1997) and in Hungary (Szanto 1998). In most areas investigated in Europe and North America, A. mellea sensu stricto occurs mainly on broad­ leaved forests and is rare in coniferous forests (Kile et al. 1991; Legrand and Guillaumin 1993). In Albania, however, A. mellea is the predominating Armillaria species in Abies alba forests, causing considerable damage. Armillaria gallica is a weak parasite or a saprophyte of mostly deciduous trees, occurring in central and southern Europe. Also, in Albania, it is rare on conifers but common on broad-Ieaved trees. At higher altitudes (ca. 600-1600 m) it is found commonly on beech, chestnut, etc. At lower altitudes (0-600 m), the main host trees are several species of Quercus. Occasionally it was found on cultivated plants as apple and pear species at altitudes from 350 to 900 m. A. ostoyae is rare in central and southern Albania, but it is common in the forests of the northern Albania and causes significant damage on Pinus nigra, P. sylvestris, Picea abies, P. omorica, Abies alba, etc., at altitudes 102 Ecology and Biodiversity from 600 to 1800 m. At lower altitudes (ca. 100-800 m), A. ostoyae was recorded on Pinus pinaster, Cupressus sempervirens and Juniperus communis. A. ostoyae was not found in fruit orchards and vineyards. In coniferous forests of Europe and North America, Armillaria ostoyae is the most important pathogen among the Armillaria species (Kile et al. 1991). In areas of central Europe with a continental type of climates, the fungus occurs in the forests independently of altitude, while in zones of Mediterranean climate is found only at high altitudes (Guillaumin et al. 1993). Albania is on the southemmost limit of this fungus in the Balkan region but in Italy A. ostoyae has been found as far south as in Calabria (Intini 1996; Grillo et al. 1996). Armillaria cepistipes was recorded in northem Albania at high altitudes from 800 to 1800 m's'and in southem Albania at altitudes from 800 -1200 m, mostly as a saprophyte on conifers and broad-leaved trees, predominating in beech and silver tir forests. It was not found in fruit orchards and vineyards. The species has a northem distribution but along the mountain ranges its distribution in southem Europe extends to Calabria, southemmost Italy (Intini 1996; Grillo et al. 1996). Armillaria tabescens was found commonly on several species of oak, Populus spp., Eucalyptus spp., etc. Among cultivated plants, it was found to cause disease in almond trees as weil as on pear trees (Prunus spp.). At the moment, it is difticult to talk about the economical importance of Armillaria in Albania, as weil as the need and possibilities to control them, because we do not have the data. It is an obligation for the near future. Ecology and Biodiversity 103 Figure 1. Morphological characteristics of the diploid mycelia of the five Armillaria species present in Albania. The cultures were grown on 3% malt extract agar for 3 weeks. A) A. mellea: white cottonous colony, fiat rhizomorphs with irregular branches, scarce pigmentation around the rhizomorphs. B) A. gallica: rhizomorphs circular in cross section, slight pigmentation around the rhizomorphs. C) A. cepistipes: brown colony with scarce rhizomorphs, circular in cross section, considerable pigmentation around the rhizomorphs. D) A. ostoyae: crustose colony with scarce rhizomorphs, brown pigmentation around the colony. E) A. tabescens: scarce mycelium without rhizomorphs and pigmentation. 104 Ecology and Biodiversity IMPACT OF ARMILLARIA rRNA - IGS GROUPS ON CROWN CONDITION OF MAPLES IN PORTNEUF COUNTY P. DesRochers\ M. Dusabenyagasani2 , J.A. Bérubé', and R.e. Hamelin' 1 Natural Resources Canada, Canadian Forest Service, Laurentian Forestry Centre, Sainte-Foy, Québec, Canada 2 Centre de Recherche en Biologie Forestière, Faculté de Foresterie et de Géomatique, Université Laval, Sainte-Foy, Québec, Canada ABSTRACT Portneuf County, located along the main pollutant transport line from the American Midwest and the Great Lakes, is associated with industrial development. Eighteen plots, located in the southem part of the county, were assessed from 1992 to 1997 for sugar maple health, including Armillaria root rot, and for pollutants in throughfall. One hundred and forty-five rhizomorphs were collected on plot trees and genotyped using rRNA­ IGS. Fifty reliably identified voucher collections were genotyped as controls. Rhizomorphs were identified as follows: A. calvescens (29 samples), A. sinapina (47 samples), a third group sharing A. calvescens - A. gallica NA and A. gallica Eur. patterns (33 samples). Ninety-one samples remained undetermined. Contingency tables were analysed for the effect of Armillaria groups on sugar maple condition, controlling for the pollution class and canker/wound status, using Mantel-Haenszel's mean score statistic. Trees without Armillaria rhizomorphs had 2.3 times more chances of being healthy than those with rhizomorphs. A. sinapina was associated with a higher damage level than A. calvescens. Keywords: Armillaria calvescens, Armillaria sinapina, IGS genotypes, throughfall, sugar maple crown condition INTRODUCTION Sugar maple decline in the early 1980's triggered many research projects. Numerous causes have been suggested, including acid rain, air pollutants, drought, late frost, insect damage and defoliation and Armillaria root rot (Allen et al. 1992; Bauce and Allen 1992; Bernier et al. 1989; Brodeur and Maufette 1989; Gagnon 1986; Lachance 1985, 1989; Roy et al. 1985). The extent ofdecline in the PortneufCounty area reached 60000 ha in 1986 (Gagnon 1987; Gagnon and Bordeleau 1990). Portneuf County is located west of Quebec City, on the north shore of the St. Lawrence River, along the main pollutant transport line from the American Midwest and the Great Lakes (Boulet 1990). It has six industrial parks and 87 smaller industrial areas, covering 2 080 ha (J. Landry, MRC de Portneuf, personal communication). Following the implementation of the Canadian government's Green Plan (Environnement Canada 1990), a study on sugar maple health related to local and long distance transported pollutants was implemented in Portneuf County. The aim of this paper is to report on the impact of Armillaria species on the health of maple stands in Portneuf County in relation to air pollution and other woody tissue damage. MATERIALS AND METHODS Network design The Portneuf County maple study was initiated in 1991 and 1992. Initially, the network included 20 plots distributed along three transect lines (Figure 1). The project ended in 1996, but crown condition was rated in 1997 for 18 plots out of 19. Ecology and Biodiversity 105 • SAINT-RAYMOND @ SAiNTE-CATl-lERINE-DE­ LA-JACQUES-CARTIER QUÉBEC Sl>JNT-AUGUSTlN •. 12 18~ ® ® NElJ\1LLE ,~ "• PONT-ROUGE@) ITJ roRTNEUF • SAINTE-CHRISTINE @ ® @ Sl>JNT--ALBAN. SAiNT-CASIMIR. Figure 1. The Portneuf County Green Plan study on sugar maple health related to local and long distance transported pollutants: location ofplots established in 1991 (square) or 1992 (circle). Transect lines were established as follows: the south line along a quarter of the distance between the St. Lawrence River shore and the Boreal Shield foothills; the middle line along three quarters of the distance between the St. Lawrence River shore and the Boreal Shield foothills; the northern line in the middle of the slope of the first Boreal Shield mountains. Study plots were established at an approximate distance of 5 km along the southem transect and 10 km on each of the other two lines. Plots had to be 1 km away from any major industry or highway. As no sugar maple stands were found between plot 13 and plot 15, plot 14 was alternately located in the middle of the county. One plot was lost in 1993 (plot 17, Saint-Alban). Each plot consisted of three circular subplots randomly located in the selected maple stands. Subplot radius varied from lOto 21 m in order to include 20 sugar maples (Acer saccharum Marsh.) in each subplot. Health rating Tree health was assessed according to the ARNEWS methodology (Magasi 1988; D'Eon et al. 1993). Measurements (ARNEWS form 4) included crown condition; cUITent, insect and disease defoliations; abiotic symptoms; woody tissue damage, including open wounds, cankers and root rots. Only crown condition was rated in 1997 on 18 plots out of 19. Crown condition ratings for live hardwoods ranged from 10 (perfect tree) to 70 (moribund tree), according to the percentage of dead branches and/or bare twigs. Sugar maple was assessed from 1991-92 through 1997. 106 Ecology and Biodiversity An additional rating of open wounds and cankers was perfonned in 1993-94. Canker and open wounds, including maple borer galleries, were assessed on each tree. Non-destructive assessment of Arnillaria infection was perfonned in 1993-94 and reassessed in 1996. Mosses, litter fall and exfoliating dead bark layers were delicately pulled away from superficial roots and root flare without damaging the trees. Rhizomorphs penetrating between bark cracks or layers were recorded (Roy et al. 1985) and were sampled whenever possible. Sixty sugar maples had Annillaria rhizomorphs but could not be sampled. DNA analysis for Armillaria spp. genotyping Genomic DNA from rhizomorphs from 145 sugar maples in the Portneuf network and 50 Armillaria spp. voucher collections, consisting of fruiting bodies, was PCR-amplified with specifie primers of the intergenic spacer of the nuclear DNA coding for the ribosomal gene (rRNA-IGS) according to Harrington and Wingfield (1995). Forty-eight voucher collections, which were used as controls, had been previously identified by crossing with tester strains by Bérubé (Bérubé and Dessureault 1988, 1989) and Morrison (Canadian Forest Service, Pacifie Forestry Centre, pers. comm.). Two other specimens from plot 2 were identified while fresh by their distinctive morphological characters. PCR products were digested with A/uI and analysed by electrophoresis in agarose gels in order to genotype Annillaria samples by comparison with already reported digestion patterns (Harrington and Wingfield 1995). Since A/uI produced a single restriction pattern for A. ostoyae, A. gemina, additional restriction assays were perfonned with Ndel, BsmI, or Hindi!. Throughfall sampling and analysis Throughfall water was sampled in all 19 plots at 12 sampling periods from 1992 to 1996: two collections were perfonned in 1992, three each in 1993, 1994 and 1996 and one in 1995. Four samples plof] period-1 were collected in 1992 and 1993 randomly distributed in subplot 1 and six samples plof' period- 1 were collected in 1994-96. Two bottles were randomly distributed in each subplot in 1994-96. Mean sampling duration was 14.5 days (min. 10 days; max. 19 days). Samples were collected between mid-June and the end of September in 1 L opaque bottles with a 10 cm diameter funnel slightly plugged with mesh cloth to exclude solid material. Sample volumes were measured and sampies were filtered (0.45 Jlm) and kept at 4-6°C for a short period before analysis. Sample acidity was measured with a pHmeter (radiometer PHM82). Samples were analysed for K+ and Na+ by atomic emission spectrophotometry; for Ca2+ and Mg2+atomic absorption spectrophotometry; for NH/ by colorimetry; for cr, F, NO)- and SO/- concentrations by ionic chromatography (Dionex), in 1992-94 (Boutin and Robitaille 1995). Analyses were perfonned by ICP in 1995 and 1996. Fluorine was not analysed in collections 6 and 7 (1994). W concentrations were calculated from pH measurements. Total bottle content for one ion was obtained by multiplying the collection volume by the ion concentration. Mean sample content was then calculated for H+, F, NO)- and SO/- for each plot. Sorne samples were excluded from the means calculation: samples with volumes exceeding 1000 mL, with very low volumes (10-20 mL), with bottle or funnel tipped or with signs of contamination by droppings, such as abnonnally high K+ or~+ concentrations. The minimal and maximal numbers of sampies per mean were 50 and 60 respectively (37 and 48 for F+). Each plot was characterised as "low" (below or equal to the median content) or "high" (above the median content) for each of the four following pollutants: H+, F+, NO)-, SO/-. Plots with three or four high scores were rated "high" in the overall pollution class; others were rated as "low" (Table 1). Ecology and Biodiversity 107 Table 1. Fluorine (F), nitrate (NO)-), sulphate (S04 2 -), acidity (H) and global pollution index by plot in the Portneuf network. Plot 1 2 3 4 5 6 7 8 9 10 Il 12 13 14 15 16 18 20 Statistical analysis F+ Law Law High Law High High Law Law Law Law Law High High Law Law High High High Law Law High High High Law High Law Law High Law High High High Law Low Low Low Low High High High High High High Law Low Law Low Low High Low Low High High Low Low Law Low High High High Low Low Law Low Low High High High Low High High High Global index Low Low High High High High Low Low Low Law Low High High Low Low High High Low Trees were given an overall tree condition rating following the crown condition assessments from 1993 to 1997 (Table 2). Contingency tables were built from the overall pollution class (low - high) by the canker and/or wound class (absent - present) by the Armillaria group (5 groups) by the overall tree condition (3 levels, Table 2). Table 2. Overall tree condition from 5-year observation of annual crown conditions (ARNEWS methodology) in the Portneuf network. Tree condition 2 3 Description Healthy Affected Declining or dead Related annual crown condition codes 4 or 5 years with crown condition code 35 or less and no code 50 or more 2 or 3 years with codes 40 or 45 and no code 50 or more At least one year with code 50 or more or 4-5 years with code 40 or 45 or tree dead in 1997 Armillaria specimens from Portneuf were grouped according ta their genotype identification into five different groups. Group 0 was for absence of Armillaria; group 1, for Armillaria caIvescens Bérubé & Dessureault; group 2 for a group with a mixed pattern from A. calvescens and Armillaria gallica Maxrnüller & Romagnesi (both European (Eur.) and North American (NA)); group 3 for Armillaria sinapina Bérubé & Dessureault and group 4 was for unidentified or unsampled Armillaria spp. or genotypes with only one sample. The 2 x 2 x 5 x 3 contingency table was analysed for effect of Armillaria group on tree condition, controlling for the pollution class and canker/wound status, using the mean score statistic from the general Mantel-Haenszel methodology with SAS FREQ procedure (Stokes et al. 1995). The mean score for each group has been calculated from the tree condition value (table 2), which is ordinal. Two additional 2 x 2 x 2 x 2 tables were built as follows: pollution class (high - low) by canker/wound class by Armillaria (presence or absence) by tree condition (1: healthy versus others; 2: affected versus very affected, declining or dead). These tables were used to calculate the odds of a tree being healthy (affected in the second additional table) without any Armillaria 108 Ecology and Biodiversity infection, as compared with Armillaria infection, controlling for pollution and canker/wound status (Stokes et al. 1995), using the SAS FREQ procedure. RESULTS AND DISCUSSION Armillaria genotypes The results of rRNA-IGS identification of voucher specimens are presented in Table 3. Ail Armillaria gemina Bérubé & Dessureault were identified as such by rRNA-IGS genotyping. Most Armillaria ostoyae (Romagnesi) Herink (28/31) were identi fied as such by their IGS pattern. Five out of six A. calvescens specimens displayed the common A. calvescens - A. gallica NA pattern (Harrington and Wingfield 1995), the sixth one showing A. gallica Eur. pattern. The only Armillaria mellea (Vahl:Fr.) Kummer specimen was genotyped as A. ostoyae. The inadequate rRNA-IGS genotyping of these two voucher specimens shows the limits of this method. Three eastem A. sinapina, from conifers or unidentified wood debris (QFB 7726, 7754 and 7755), were genotyped as A. sinapina. The A. sinapina fruiting bodies from hardwoods (QFB 7756, 7757, 7758, 16744, 16747) showed A. gallica Eur. pattern. As A. gallica Eur. does not exist in North America, rhizomorphs bearing A. gallica Eur. patterns will be from now on considered as A. sinapina. Western collections of A. sinapina exhibited the Armillaria nabsnona Volk & Burdsall rRNA-IGS pattern. Table 3. Armillaria spp. voucher collections: identification by tester strain or distinctive morphological characters compared with rRNA-IGS genomic identification. Tester strains rRNA-IGS Identification by 1 5 2 8 2 2 1 28 TotalQFB1 No. 7729, 7736 7726, 7754 to 7758, 167445,167475 7733, 7734 7753 7748, 7749, 7750, 7751, 7752 7753, 7745 7727 7706 to 7722, 7729, 7731, 7732, 7735, 7737 to 7742, 7931 7713AA. calvescens / A. gallica NA No identity A. sinapina A. nabsnona A. gemina A. mellea A.ostoyae2 A. calvescens A. gallica Eur. A. calvescens A. calvescens / A. gallica NA A. gemina A.ostoyae A.ostoyae A.ostoyae A.ostoyae A. sinapina (ease) A. sinapina (west4 ) 1: From the René-Pomerleau Herbarium, Natural Resources Canada, Canadian Forest Service, Sainte­ Foy, Québec; 2: includes three specimens from British Columbia, from D. Morrison; 3: specimens from eastern Canada; 4: specimens from British Columbia; 5: specimens identified from distinctive morphological characters, while fresh. From the 145 Portneuf maple rhizomorphs analysed, 29 displayed the A. calvescens - A. gallica NA pattern. A. calvescens and A. gallica NA are known to be related (Miller et al. 1994, Smith and Anderson 1989) and give the same Alul pattern (Harrington and Wingfield 1995). The rRNA-IGS pattern obtained from A. calvescens voucher specimens, the specificity of A. gallica for oak in North America (Blodgett and Worrall 1992a, b) and the preference of A. calvescens for maple trees and stands (Blodgett and Worrall 1992a; Sabourin et al. 1990) strongly suggest that rhizomorphs showing A. calvescens - A. gallica NA pattern are from A. calvescens. Ecology and Biodiversity 109 Thirty-three samples had a mixed pattern between A. calvescens - A. gallica NA and A. gallica Eur. patterns. The status of rhizomorphs exhibiting such a mixed pattern remains unclear. Forty-two samples displayed the A. sinapina pattern; five samples displayed a mixed pattern between A. sinapina and Armillaria cepistipes Velenovsky, but, as A. cepistipes has never been found in eastern North America, these are considered to be A. sinapina. One sample showed the A. ostoyae pattern and 95 could not be sampled or genotyped. Sugar roaple health status The impact of Armillaria species on the overall tree condition is summarised in Table 4. Proportions of healthy trees are higher when no Armillaria has been detected (35.7%) and decrease with the presence of A. calvescens (24.2%) and even more with A. sinapina (12.8%), while the number of very affected, declining or dead trees increases from 24.9% without Armillaria to 44.8% A. sinapina infection. The Armillaria group mean scores differ significantly (P- -0 20 0 Dec97 dec98 Dec99 Figure 3. Evolution oftree growth in each treatrnent throughout the experiment o 80 20 Experirrent 2 (june 1999-june 2000) 100 è tii GO 1: o E 40 8 m) yielded genetic distance values not significantly different from zero. The average genet size was an average of approximately 4 m. The distribution of the genetic polymorphisms is consistent with infections from basidiospores, followed by secondary vegetative spread. 128 Ecology and Biodiversity ASSOCIATION OF INONOTUS TOMENTOSUS WITH SPRUCE BEETLE ATTACK F.A. Baker Department of Forest Resources, Utah State University, Logan, UT 84322-5215 USA In 1996, spruce beetles (Dendroctonus rufipennis) infested the T.W. Daniel Experimental Forest in northern Utah. AlI trees with spruce beetles were 10cated, and the proportion of infected roots was determined by pit sampling near each stump and at two adjacent points 100 m away. At each sampling point, eight pits approximately 20 cm x 20 cm x 40 cm deep were excavated with a pulaski. AlI roots >0.5 cm in diameter were examined for root disease symptoms. At seven of the nine spruce beetle infestations, the proportion of roots with symptoms was greater in the area of the infestation than it was 100 m away. There were more roots per sample at the adjacent sampling points than immediately around the beet1e attacked trees. Inonotus tomentosus was the most abundant pathogen. In this stand spruce beetles attacked first, and continue their attacks in areas of the stand with a greater proportion of symptomatic roots than adjacent, unattacked areas. This evidence suggests that root disease may predispose Engelmann spruce to attack by spruce beetles. INCIDENCE OF TOMENTOSUS ROOT DISEASE RELATIVE TO SPRUCE DENSITY AND SLOPE POSITION IN SOUTH-CENTRAL ALASKA K.J. Lewis*, L. Trummer**, R. Shipley*, and S. Parsons* *University ofNorthern British Columbia, 3333 University Way, Prince George, B.e. Canada V2N 4Z9 **USDA Forest Service, 3301 C St., Suite 522, Anchorage, Alaska 99503-3956 Jnonotus tomentosus causes Tomentosus root disease in sub-boreal forests of south-central Alaska. Site variables are related to the incidence of1. tomentosus. At three different locations, a total of 371 plots, 50m2 each, were examined to measure the incidence of J. tomentosus in conjunction with a variety of site variables. These included spruce density, species composition, and slope position (depression, low, mid, upper, flat). The incidence of I. tomentosus in plots ranged from 0­ 100%, and 54% of the plots had no infected trees. No relationship was found between incidence and spruce density (stemslha or % spruce). There was a significant difference in disease frequency by slope position (P=0.0013) with upper slopes having a higher incidence of infected trees. Slope position may be used by forest managers to indicate areas more likely to have a higher incidence of root disease, and can therefore reduce costs of disease surveys. Ecology and Biodiversity 129 RED-ROT OF NORWAY SPRUCE (PICEA ABlES) IN AUSTRIA - RELATIONS WITH SITE FACTORS BASED ON A SURVEY BY THE AUSTRIAN FORESTRY INVENTORY C. Tomiczek, R. Buechsenmeister, and Th.L. Cech Federal Forest Research Centre, SeckendorffGudent-Weg 8, A-1131 Vienna, Austria Norway spruce (Picea abies), the main tree species of Austrian forests (67%), is suffering from root and butt rots more than from any other biotic damaging agent. Up to 100% of the trees may be devaluated by rot, which can be attributed mainly to Heterobasidion annosum. Within the scope of two surveys (one from 1986 to 1990 and the other from 1992 to 1996), the amount of trees with red-rot has been recorded aIl over Austria. On the basis of a comparison between the two surveys, tendencies are shown. Relations to numerous site and stand factors such as sea-level, slope, exposition, soil type, stand composition, age c1ass, water balance, etc. are discussed. THE ROLE OF SOIL MOISTURE CONTENT IN ARMILLARIA ROOT DISEASE K.I. MaIlett*, D.G. Maynard**, and C.L. Myrholm* *Natural Resources Canada, Canadian Forest Service, Northern Forestry Centre, 5320 - 122 St. Edmonton, AB, T6H 3S5, Canada ** Natural Resources Canada, Canadian Forest Service, Pacific Forestry Centre, 506 West Burnside Rd. Victoria, B.C., V8Z IM5, Canada Lodgepole pine and white spruce were grown in peat soil that was kept saturated or at 75%, or 25% of field capacity and inoculated with either A. ostoyae or Armillaria sinapina~ After 12 months, trees were examined for infection, and foliage, stem and root tissues were analysed for nutrient content. Soil moisture had little effect on disease caused by A. ostoyae in lodgepole pine and white spruce. Armillaria ostoyae inoculum survival was poorest in the saturated soil. Likewise, with A. sinapina, disease in white spruce was not affected by soil moisture; however, it was in lodgepole pine. There was progressively more disease in the 75% and saturated treatments. Armillaria sinapina inoculum survival was poorest in the 25% treatment and best in the saturated treatment. Potassium concentrations were greater in the foliage and stems of infected trees compared to those of healthy trees. 130 Ecology and Biodiversity Control INTEGRATED CONTROL OF ARMILLARIA MELLEA BY TRICHDDERMA HARZIANUM AND FOSETYL-AL F. Raziq* and R.T.V. Fox" *Department of Plant Pathology, NWFP Agricultural University, Peshawar, Pakistan and **Department of Horticulture and Landscape, School of Plant Sciences, The University of Reading 2 Earley Gate, Reading RG6 6AU, u.K. SUMMARY Laboratory, glasshouse, and field experiments have been carried out to integrate a fungal antagonist with a fungicide, fosetyl-Al (Aliette) or fenpropidin (Patrol). An antagonistic isolate of Trichoderma harzianum that suppressed Armillaria mellea in vitro interacted significantly when application to inoculated strawberry plants was integrated with the systemic fungicides, fosetyl-Al and fenpropidin. Survival was better after treatment with fosetyl-Al than fenpropidin, which was phytotoxic and inhibited the T. harzianum inoculant. The antagonist was not adversely affected by fosetyl-Al. Keywords: fosetyl-Al (Aliette) fungicide, fenpropidin (Patrol), Trichoderma harzianum, fungal antagonist, integrated control of Armillaria mellea INTRODUCTION Apart from soil fumigation after a susceptible crop with carbon disulphide (Bliss 1951), methyl bromide (Vanachter 1979) or chloropicrin (Hagle and Shaw 1991), there is a lack of experimental evidence to support the use of chemical treatments for control of Armillaria species (Shaw and Roth 1978), even those like Armillatox, which have been marketed (Redfem 1971). Another similar mixture of cresylic acids (Bray's Emulsion) was also marketed, but again control was not sustained, and there were phytotoxicity problems with both products (West 1994). A major reason for the failure of both chemical and bi010gical methods is that it is difficult to reach Armillaria deep inside wood. Campbell (1934) found that after becoming established in roots, the vegetative mycelium ofArmillaria develops a protective layer of thick-walled mycelium, the pseudosclerotial envelope, both within the body of the infected wood and the white mycelial fans in the cambium region, similar to the protective layer that also covers the rhizomorphs. Any successful chemical or biological eradicant of the fungus should penetrate into both these resistant structures, yet still maintain its effectiveness. Using the most active of the newer chemical fungicides evaluated by Turner and Fox (1988), West (1994) found that none could fully eradicate rhizomorphs in soi l, even with concentrations as high as 10,000 mg/l, and even a thin layer of bark prevented any of the chemicals from eradicating the underlying mycelium. Despite fuis, fenpropidin, which exhibited good protectant activity in vitro, slowed the rate of mortality of inoculated strawberry, privet and willow, even though it was fungicidal only above 5000 mg/l and fungistatic at lower concentrations. Fosetyl-Al (Aliette) was as active in vivo but not in vitro. Phosphonic acid, the active ingredient in fosetyl-Al, has been reported by Navarro et al. (1990) to control Armillaria. Although many fungi have also been found to be antagonistic towards Armillaria (Leach 1937; Cusson and LaChance 1974; Federov and Bobko 1989; Pearce and Malajczuk 1990; Hagle and Shaw 1991; Riffle 1973; Eghbaltalab et al. 1975), perhaps the most thoroughly studied antagonists are Trichoderma species (Hagle and Shaw 1991). In the interactions between Armillaria and Trichoderma, one critical factor is the tolerance of Trichoderma to stress factors that inhibit the growth and metabolism ofArmillaria. Control 133 Several examples of integrated control of Armillaria involving Trichoderma viride have been reported after fumigation with carbon disulphide (Bliss 1951) and methyl bromide (Munnecke et al. 1973). Munnecke et al. (1981) reported that Trichoderma spp. were 1.9 to 3.2 times more resistant to methyl bromide than A. mellea, and also suggested that fumigation of soil weakens the defence mechanisms ofArmillaria. MATERIALS AND METHODS The fungicides fosetyl-AI and fenpropidin were applied 40 days before or after the antagonists. Each of the fungicides was applied at 2000 mg/l. Each of the potted strawberry plants was drenched with 500 ml of the fungicide suspensions using a watering can fitted with a sprinkler. The fungicide suspensions were allowed to leach freely from the bottom of the pots. The T. harzianum antagonists were grown on 30 g sterile mushroom compost in 500 ml glass jars for 2 weeks before application. Each of the treatments was replicated four times in RCB design. The experiment started on 6 September 1995, and terminated on 25 March 1997. Along with the above factorial experiment, two other experiments were carried out simultaneously. One of the experiments tested the efficacy of the antagonists alone. Control plants in this experiment were inoculated with A. mellea and 30 g sterile mushroom compost, without any antagonist, added around the roots. The other experiment was done to investigate the effect of the fungicides alone on suppression of Armillaria root rot. In addition to the dose of the fungicides used in integration with the antagonists (2000 mg/l), two higher doses, 4000 and 8000 mgll, were tested. Strawberry plants were chosen as a very susceptible host rapidly infected by A. mellea (Fox 2000). The strawberry plants were drenched with 500 ml suspension ofeach dose of the fungicides. Each of the treatments in these experiments was replicated four times in RCB design. The plants were watered regularly and were sprayed once, on 4 April 1996, with Pentac at 1 mlll to control red spider mites. RESULTS Although no significant differences were found among the antagonists, fungicides or time sequences, the three factors interacted significantly (P<0.05). Seventy five percent of the plants treated first with fenpropidin and then, after 40 days, with T. harzianum, survived until the end of the experiment which lasted 566 days. None of the plants survived that long when the antagonist and the fungicide were applied in the reverse sequence. T. harzianum was more effective when applied before fosetyl-AI and 50% of the plants survived until the end of the experiment when treated in this sequence compared to no plant surviving when the application sequence was reversed. This antagonist interacted differently with fenpropidin as 50% of the plants survived when the fungicide was applied first, compared to 25% when the antagonist was applied first. The antagonists and fungicides did not differ significantly in terrns of mean number of living leaves, but plants treated with the fungicides first had more leaves than those treated with the antagonists first. For the subsequent counts, however, the differences were not significant. The interaction of antagonists with fungicides or time sequences was not signiftcant. The fungicides and time sequences interacted signiftcantly for the number of leaves counted 202 236, and 298 days after inoculation. Plants treated with fenpropidin ftrst (and antagonists later) had signiftcantly (P<0.05) more leaves than those treated with antagonists first (and fenpropidin later), while differences in the number of leaves on plants treated with fosetyl-AI in either sequence were not significant. The antagonists, fungicides, and time sequences interacted signiftcantly (P<0.05) only 202 days after inoculation. Plants treated with fenpropidin ftrst and T. harzianum 40 days later had significantly more leaves than those treated with T. harzianum first and fenpropidin 40 days later, while those treated with T. harzianum and fosetyl­ Al in either sequence had non-signiftcant differences. The strawberry plants treated with the antagonists alone or left untreated had no significant differences in the number of leaves. Plants treated with the fungicides alone showed signiftcant differences in the number of living leaves counted 236 days after inoculation with A. mellea. The number of leaves was significantly higher on plants 134 Control treated with fosetyl-Al than those treated with fenpropidin. The doses did not differ significantly, and the interaction ofthe fungicides with the doses was also non-significant. Neither the effects of the antagonists, fungicides nor timing of the sequences of treatment differed significantly in terms of mean health scores of the strawberry plants. The only significant interaction was between the fungicides and time sequences 202 and 236 days after inoculation. On these dates, the plants treated with fenpropidin first (and antagonists later) had significantly (P<0.05) higher health scores than those treated with the antagonists first (and fenpropidin later), while those treated with fosetyl-Al and the antagonists in either sequence did not differ significantly in health scores. The strawberry plants treated with the antagonists alone or left untreated did not have significant differences in health scores. Plants treated with the fungicides alone showed significant differences only in health scores recorded 236 days after inoculation. Plants treated with fosetyl-Al had significantly higher health scores than those treated with fenpropidin. The doses did not show significant differences and the interaction of fungicides with doses was also non-significant. DISCUSSION There was a significant interaction between the antagonists, fungicides, and the time sequence of application. Seventy five percent of the strawberry plants treated first with fenpropidin and then, 40 days later, with T. harzianum survived until the end of the experiment lasting 566 days, while none of the plants survived that long when treated with the antagonist first and the fungicide 40 days later. This result shows the sensitivity of the antagonist to the direct application of the fungicide. The antagonist's survival and growth was probably adversely affected when subjected to the direct drenches of the fungicide suspension resulting in greatly reduced activity against the pathogen. When the fungicide was applied 40 days before the antagonist, the residual effect was not large enough to hinder the growth of the antagonist. The frequent watering of the potted plants is not likely to have leached fenpropidin significantly as the fungicide and its metabolites have little or no tendency to leach (Tomlin 1994). West (1994) also found little effects of watering on the leaching of fenpropidin. Fenpropidin binds to soil because of its high affinity for organic matter, a characteristic feature of lipophilic chemicals. The compost used in the pots was rich in organic matter. Tt was concluded by West (1994) that when fenpropidin had been in the soil for one week or more, it was unavailable for activity against inoculum of Armillaria, because of binding ta sail. Armillaria is generally considered ta be more sensitive ta chemicals than Trichoderma (Bliss 1951; Munnecke et al. 1981). The residual effect of fenpropidin applied 40 days before Trichoderma would, therefore, be negligible. In addition to strong adsorption, fenpropidin is known ta be rapidly and extensively degraded in sail (Tomlin 1994) and this process over the 40 day period must also have made it ineffective. Early application of the fungicide also possibly helped in the establishment of T. harzianum by reducing the microbial population in the compost that would compete with the antagonist. The survival of the strawberry plants is likely to have resulted from the effect of fenpropidin on Armillaria during the first few days after its application leading to successful control of the weakened pathogen by T. harzianum that was applied 40 days later. No plant survived until the end of the experiment when treated with T. harzianum alone and only 25% of them survived when treated with fenpropidin alone at the dose rate integrated with the antagonist. The mechanism by which fenpropidin application, followed by the introduction of T. harzianum, suppressed Armillaria seems to be similar to the one reported for T. harzianum by Bliss (1951) and Munnecke et al. (1981). They suggested that fumigation of soil with carbon disulphide or methyl bromide prevents Armillaria from producing antibiotics which makes it susceptible to attack by Trichoderma that survives the fumigation and dominates the treated sail in the absence of other competitive organisms as they are killed by the fumigation. Fosetyl-Al did not work in the same way probably because its impact on Armillaria, at the dose rate applied, was not sufficient to prevent it from producing antibiotics and did not eliminate the competing microorganisms to help T. harzianum. Findings of the in vitro experiments prior to this study support this conclusion. Control ------------------------------------ 135 Integration of T. harzianum resulted in the survival of 50% of the strawberry plants when applied before fosetyl-AI or after fenpropidin. Fosetyl-AI suppressed the disease significantly more than fenpropidin. Fifty percent of the strawberry plants treated with fosetyl-AI survived compared to 16.7% of those treated with fenpropidin. Plants treated with the highest dose of fenpropidin (8000 mgll) showed symptoms of severe phytotoxicity. Though these plants apparently recovered, the treatment was not effective as no plant survived until the end of the experiment. This result indicates that even this high dose failed to eradicate the pathogen inoculum. As discussed earlier, this was probably because of the binding of the fungicide to the soil particles and its rapid and extensive degradation. Fosetyl-AI was found safe and the highest dose of 8000 mg/l did not cause any visible symptoms of phytotoxicity. Instead, this dose was found the most effective, resulting in the survival of 75% ofthe treated plants. The fungicides were applied only once in the long duration of the experiment (566 days). Both fenpropidin and fosetyl-Al are reported to be fungistatic rather than fungicidal (West 1994). This would mean that for persistent protection against the disease, the fungicides would have to be applied repeatedly at proper time intervals. The better performance of fosetyl-Al, despite this theoretical limitation, was probably because this chemical can be taken up by the plant roots and stored for prolonged activity. d'Arcy-Lameta and Bompeix (1991) reported that fosetyl-AI was rapidly taken up by tomato roots, and it was postulated that phosphonate, the active compound of this fungicide, can be fixed in the lignified tissues of woody plants and other perennials, allowing its storage and slow release. It seems that for this to happen, the fungicide would have to be applied in heavy concentrations. This should be possible as the fungicide is not phytotoxic. Previous studies on the integrated control of Armillaria mellea have involved affects of phytotoxic fumigants on resident populations of Trichoderma in the absence of the living host plant. Integrated control is desirable especially when a living host plant is involved. Most of the fungicides that have been tested previously for the control of Armillaria are phytotoxic if applied in high enough concentrations to attempt its eradication and sorne of them stimulate the growth of the pathogen when applied in low concentrations. To avoid these problems, it was imperative to integrate the fungicides with antagonists at sublethal non-phytotoxic doses. As the antagonist is likely to be adversely affected by the fungicide, the fungicide should be applied first. The antagonist could then be applied after the toxic effects of the fungicide have diminished. This may also help in the establishment of the antagonist, if the soil flora has been killed by the fungicide, thus reducing competition to the introduced antagonist. But if the antagonist is not likely to be adversely affected by the fungicide, then it could be applied before, or simultaneously with, the fungicide. This could have a combined effect on the pathogen resulting, possibly, in its death. Table 1. Effect of integration of fungicides with antagonists on survival of Armillaria-inoculated strawberry plants on median Survival Time (Days) A F One Two One Two One Two Antag Th2 188.5 378.5 221.0 335.0 207.0 188.5 222.5 234.5 Th23 216.0 238.0 240.5 201.5 174.0 198.5 216.0 Tv3 203.0 171.5 169.0 204.5 165.0 213.5 169.0 180.5 180.5 Thaml 198.0 176.0 176.0 187.0 201.5 198.0 147.0 185.0 185.0 Co 192.0 218.5 245.0 194.5 253.5 180.5 240.0 218.5 199.5 SP 194.5 192.0 180.5 199.5 203.5 194.5 180.5 223.5 192.0 199.5 196.0 191.0 202.0 209.0 188.5 180.5 239.0 136 Control *Table la. Effect of integration of fungicides with antagonists on survival of Armillaria-inoculated strawberry plants on median Survival Time (Days): Fosetyl-AI (A); Fenpropidin (F) Fungicide Sequence Fungicide x Sequence A F One Two One Two One Two Antag Th2 188.5 378.5 221.0 335.0 207.0 188.5 222.5 234.5 Th23 216.0 238.0 240.5 201.5 174.0 198.5 216.0 Tv3 203.0 171.5 169.0 204.5 165.0 213.5 169.0 180.5 180.5 Thaml 198.0 176.0 176.0 187.0 201.5 198.0 147.0 185.0 185.0 Co 192.0 218.5 245.0 194.5 253.5 180.5 240.0 218.5 199.5 SP 194.5 192.0 180.5 199.5 203.5 194.5 180.5 223.5 192.0 199.5 196.0 191.0 Table lb. Effeet of integration of fungieides with antagonists on survival of Armillaria-inoeulated strawberry plants on median Survival Time (% Survival after 566 days): Fosetyl-AI (A); Fenpropidin (F) Fungicide Sequence Fungicide x Sequence Antag A F One Two One Two One Two Antag Th2 12.5 37.5 12.5 37.5 25.0 0.0 0.0 75.0 25.0 Th23 25.0 37.5 37.5 25.0 50.0 0.0 25.0 50.0 31.3 Tv3 0.0 25.0 12.5 12.5 0.0 0.0 25.0 25.0 12.5 Thaml 12.5 0.0 0.0 12.5 0.0 25.0 0.0 0.0 6.3 Co 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 SP 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 8.3 16.7 10.4 14.6 12.5 4.2 8.3 25.0 Table 2. Effeet offungal antagonists on survival of Armillaria-inoeulated strawberry plants (n=4). Treatment Median Survival Time (Days) Control 165.0 Th2 206.5 Th23 212.5 Tv3 154.0 Thaml 192.5 Co 217.5 SP 174.0 Log-Rank Test: X2-Value (6 d.f.) = 12.11 (p>X2 = 0.0596) *Table 3. Effeet of fungieides on survival ofArmillaria-inoeulated strawberry plants. 2000 mg/l 4000 mgll 8000 mgll Fungicide Median Survival Time (Days) Log-Rank Test: Category Fungieides (n=12) Doses (n=8) Fungieides x Doses (n=4) Fungicide Fosetyl-AI Fenpropidin 248.0 199.5 231.0 197.0 439.5 242.0 354.5 x2 4.38 0.57 6.44 197.0 d.f. 1 2 5 % Survival after 566 days 2000 mg/l 4000 mg/I 8000 mg/I Fungicide 25.0 50.0 75.0 50.0 25.0 25.0 0.0 16.7 25.0 37.5 37.5 P 0.0363 0.7512 0.2656 * The editors have experirnented sorne problems with the electronic transfer of Tables la and 3, and could not answer from the authors before printing to solve these problems. Ifyou need clarification, please contact one of the authors. Control 137 REFERENCES Bliss, D.E. 1951. The destruction ofArmillaria mellea in citrus soils. Phytopatho10gy, 41: 665-683. Campbell, AH. 1934. Zone lines in plant tissues. II. The black lines fonned by Armillaria mellea (VaW) Quél. Annals of App1ied Bio10gy 21: 1-22. Cusson, Y., LaChance, D. 1974. Antagonism between Scytalidium lignicola and two root rot fungi. Phytoprotection, 55: 17-28. d'Arcy-Lameta, A., and Bompeix, G. 1991. Systemic transport of tritiated phosphonate in tomato p1antlets (Lycopersicon esculentum Mill). Pesticide Science, 32: 7-14. Eghbaltalab, M., Gay, G. and Bruchet, G. 1975. Antagonisme entre 15 espèces de basidiomycètes et 3 champignons pathogènes de racines d'abres. Bulletin de la Société Linnéenne de Lyon, 44: 203-229. Federov, N.I., and Bobko, LN. 1989. Armillaria root rot in Bye10russian forests. In: Morrison, D.J. (ed.). Proceedings of the 7th International Conference on Root and Burt Rots. 1988 August 9-16; Vernon and Victoria, Be. International Union of Forestry Research Organizations, Victoria, Be. pp. 469-476. Fox R.T.V. 2000. Pathogenicity. In: Fox RTV, ed. Armillaria root rot: biology and control of honey fungus. Andover, UK. 1ntercept 113-136. Hagle, S.K., and Shaw, III e.G. 1991. Avoiding and reducing 10sses from Annillaria root disease. In: Shaw, e.G.III and G.A. Kile (eds.). Armillaria root disease. USDA Forest Service Agriculture Handbook Number 691. pp. 157-173. Leach, R. 1937. Observations on the parasitism and control ofArmillaria mellea. Proceedings of the Royal Society of London, Series B 121: 561-573. Munnecke, D.E., Kolbezen, M.J. and Wilber, W.D. 1973. Effects of methyl bromide or carbon disulfide on Armillaria and Trichoderma growing on agar medium and relation to survival ofArmillaria in soil following fumigation. Phytopathology, 63: 1352-1357. Munnecke, D.E., Kolbezen, M.J. Wilbur, W.D. and Ohr, H.D. 1981. Interactions involved in controlling Armillaria mellea. Plant Disease, 65: 384-389. Navarro, M.e.M., Nunez, AC. and Montesdeoca, M.M. 1990. Root rot (Armillaria mellea (Vahl ex Fr) Kummer) on kiwi (Actinidia dilciosa) on the island of Tenerife. (Sp.). Cuadernos de Fitopato10gia, 7: 70-72. Pearce, M.H., and Malajczuk, N. 1990. Inoculation of Eucalyptus diversicolor thinning stumps with wood decay fungi for control ofArmillaria luteobubalina. Mycological Research, 94: 32-37 Raziq, F. (1998) Biological and integrated control ofArmillaria. Ph.D. Thesis, The University of Reading. Raziq, F. (2000) Biological and integrated control of Armillaria root rot. In: Fox RTV, ed. Armillaria root rot: biology and control ofhoney fungus. Andover, UK. Intercept 183-197. Redfem, D.B. 1971. Chemical control ofhoney fungus (Armillaria mellea). In: Proceedings of the British Insecticide and Fungicide Conference. 6: 469-474. Riffle, J.W. 1973. Effect oftwo mycophagous nematodes on Armillaria mellea root rot ofPinus ponderosa seedlings. Plant Disease Reporter, 57: 355-357. Shaw, C.G.III., and Roth, L.F. 1978. Control of Armillaria root rot in managed coniferous forests. European Journal ofForest Pathology, 8: 163-174. Tomlin, e. (ed.). 1994. The pesticide manual. lOth Edition. The British Crop Protection Council and Royal Society ofChemistry, Farnham, UK. Turner, J.A, and Fox, R.T.V. 1988. Prospects for the chemical control of Armillaria species. In: Proceedings of the Brighton Crop Protection Conference: Pests and Diseases 1. 1988: 235-240. Vanachter, A 1979. Fumigation against fungi. In: Mubler, D. (ed.). Soil infestation. Elsevier Scientific Publishing Co., Amsterdam. pp. 163-183. West, J.S. 1994. Chemical control ofArmillaria. Ph.D. Thesis, The University ofReading. West, J.S. 2000. Chemical control of Armillaria root rot. In: Fox, R.T.V., ed. Armillaria root rot: biology and control ofhoney fungus. Andover, UK. Intercept 183-197. 138 Control IMPROVING STUMP TREATMENT BY HARVESTING MACHINE J.E. Pratt, D.l. Brooks* and M.A. Lipscombe** *Forest Research, Northem Research Station, Roslin, Midlothian Scotland EH25 9SY ** Forest Research, Rydal Rouse, Colton Road, Rugeley, Staffs, UK, WS15 3HF ABSTRACT Equipment that enables measurement of the volume oftreatment fluid applied to stumps by harvesters and of its collateral waste is described. The equipment was used to demonstrate how the current design of chain-saw bars can be modified to improve coverage and reduce waste, resulting in savings estimated to exceed 30%. Keywords: Stump treatment, harvester, application rates, wastage INTRODUCTION Stump treatment is increasingly used in Europe for the prophylactic control of Heterobasidion annosum, and the majority is applied by harvester during felling (Thor and Stenlid 1998; Pratt and Thor 2001). Many systems now use special chain-saw bars for delivery of the protectant, which is pumped on demand from a storage tank to the felling head whence it is discharged onto the stump surface via holes spaced along the length of the chain saw bar. Sophisticated computer control systems may be fitted which determine when the pump is activated, and to an extent this affects the volumes applied. However, treatment inevitably produces waste, either when it is applied in excessive doses, or is sprayed onto the ground surrounding the target stump. The former results in economic loss and is contrary to the spirit of European legislation which seeks to reduce pesticide doses to the minimum (Anon 1994), whilst the latter has environmental implications which may weIl become unacceptable (Westlund and Nohrstedt 2000). There is clearly a need for a system that can be calibrated for the accurate application of protectants to stumps. There would also be benefit in measuring waste. Direct measurement of the volumes applied to the surface of stumps is not possible using existing technology, and indirect techniques are needed. This paper describes such a method, which allows calculation of the volume applied to stumps by comparing the amount oftluid dispensed at the felling head with the volumes sprayed to waste and captured in a measuring tray. MATERIALS AND METHODS The equipment, known as Stump Treatment Calibrator (STC), consists of four instruments which are housed in a portable waterproof (IP64) box, 420 x 305 x 155 mm, gross wt 7 kg, and are powered by an 12v rechargeable battery. The equipment is connected in-line into the delivery pipe of the stump treatment system on a harvester by means of Y4 inch BSP ports and 10 m of standard hydraulic hose. The instruments, mounted in series, consist of a volume/flow meter, pressure meter and two temperature probes in that order. The maximum tluid pressure that the instrument can record is 10 bar. The volume/flow meter displays either the cumulative volume (ml) of liquid discharged, or the current rate offlow (ml/sec) of the liquid in the delivery pipe. The default display shows the cumulative volume and can be reset to zero manually at any time (for example, between stumps). The volume counter continues to update while the flow rate is displayed. The turbine tlow sensor is accurate with tlows 0.4 ­ 6.5 litres/min (7 - 108 ml/sec), and is calibrated for use with water or with fluids of similar viscosity. The system is calibrated by comparing the volume delivered from the uncoupled delivery pipe at the harvester head (measured in a measuring cylinder) with the displayed value. Corrections can be made by recalibrating the flow meter. The pressure meter displays the pressure (in bars) in the delivery pipe, and records the maximum and minimum pressures measured. It is reset manually. The CUITent temperature of fluid in the delivery pipe is displayed, and there is an added facility to display remote temperatures at distance « = 12m) using a thermocouple probe. Control 139 To simulate felling and stump treatment, a log of appropriate diameter, 2 - 4 m long, is grasped in the felling head, the lower end resting on a heavy duty steel tray (collecting tray (Fig 1). A dise circa 5 - 10 mm thiek is eut from the end in a simulated felling cut while the stump-treatment system is aetivated. The dise is prevented from falling into the eolleeting tray by a mesh eovering the tray. The dise aets as a treated proto-stump. The volume of fluid dispensed is read from the display The waste eaught in the tray is poured into a flask for volume measuremenl. The volume applied to the proto-stump is ealculated by subtraetion. The fluid ineorporates a dye which allOWS the coverage of each proto-stump to be estimated and described. The spray distribution (i.e. volumes from different parts of the bar when not performing a felling cut) can be measured in a patternator. This is a flat tray constructed out of 2 mm mild steel in the form of a truncated arc of 120°, divided into ten parallel curved channels each 5 cm wide and 2 cm deep. Like the eolleeting tray, the patternator is plaeed beneath the felling head and the saw and spray system are activated over il. Measuring bottles fixed at the end of each channel collect fluid sprayed over the tray as the bar describes a simulated felling eut above il. It is assumed that the presence of timber when actually felling could affect the distribution pattern found in this method. Figure 1. Measuring the distribution of stump treatment fluid applied through a drilled bar. A log, held vertically in the snedding knives of a harvester head, rests on a colleeting tray. As the saw cuts through the base of the log (as if simulating the felling of a tree), the treatment fluid that is dispersed to waste either side of the felling eut together with that running off the stump surface is collected and measured in a jug. The flow rate, pressure and temperature is monitored in the STe, which is fitted into the treatment fluid delivery system via a 10 m length of standard hydraulic hose. 140 Control RESULTS 1. Standard bars The distribution of fluid sprayed from a commercially-available standard bar with an evenly-spaced array of 32 holes was measured. The results from three separate trials are summarised in Table l, in which volumes caught in each pair of adjacent channels are amalgamated for the sake of brevity. The channels and bar holes were numbered sequentially from the sawbox end of the bar (Channelland 2, bar holes 1 - 7) to the tip (Channel 10, bar hole 32) Table 1. Relative distribution of fluid from a standard bar, as measured in the pattemator channels. Patternator channel no. 1,2 3,4 5,6 7,8 9,10 Bar holes registering on channels 1-7 8-15 16-23 24-31 32 Vol ml (and %) fluid dispensed per 201 (30%) 141 (21 %) 132 (19%) 106 (16%) 95 (14%) channel Area under full sweep of (sq cm) 433 621 811 999 1187 Vol (ml) applied per sq cm 0.46 0.23 0.16 0.11 0.08 As is evident from the table, a third of the fluid was expressed through the 7 holes proximal to the saw box. Because of the configuration of the cutting head, few of these holes would connect with any but the larger stumps, and much of the product would thus be sprayed to waste. In addition, the areas covered by the sweep of the saw bar are much smaller proximal to the sawbox than at the tip, and this provides another cause for uneven distribution of fluid onto a stump along with a higher rate of application than is required (normally 0.1 ml cm-2 ). The volumes sprayed to waste using standard bars were measured using the STC and an Oregon 752LUFI04 VZ 28167 bar with 42 holes (1.2 mm diameter) set 13 mm apart over a length of 53 cm. Ten proto-stumps each of 35 cm, 21 cm or 12 cm diameter were cut and treated to an acceptable standard (90% minimum coverage of the stump) and the waste was measured in the collecting tray as described above. Waste amounted to 30%, 48% and 55% respectively of the volumes discharged, the greater waste being among the smaller stumps. 2. Improved bars Observations made in 1999 showed that the holes in a standard bar often extend beyond the dimensions of the stumps being cut. Complete coverage, could be achieved with fewer holes if these were correctly positioned on either side of the centre of the cutting arc. Standard bars were modified by blocking holes, and thus reducing their number (Fig 2). There is clear evidence (Table 2) that in these modified bars fluid is forced laterally out of treatrnent holes such that adequate coverage can be achieved with significantly reduced doses and less collateral waste. Control 141 Star,chrd bar CenlIe of cu[tmg arc Modifie 0 -, -2 0 0 2 3 , -2 -, 4 ,l 1 5 2 Axis 2 6 7 -, o HF 2 4 3 ~~..... 5 1,5 2.0 7 TH. o 0,5 1,0 -0.5 0 ~S2 -2.0 -1.5 -1.0 ., n. Figure 1. Scatterplot of the correspondence analysis of mycocenoses one year after the treatments (first trial). Figure 2. Scatterplot of the correspondence analysis of mycocenoses one and two years (bold and italics) after the treatments (first trial). From the stumps treated in the second trial, 84 fungal taxa were identified. Of these, 19 were from the first sampling and 27 from the second. The F values of taxa with a F:::: 60% in one or more treatments are listed in Table 3. The number of fungal taxa is generally lower in the treatments than in C, mainly after one year. However, only B treatment showed significant differences. Control ------------------------------------ 149 Table 3. Colonisation frequency (F) of the fungal taxa during the second tria1. C UIO U20 U30 B CU PG 1 yr 2yrs 1 yr 2yrs 1 yr 2yrs 1 yr 2yrs 1 yr 2yrs 1 yr 2yrs 1 yr 2yrs Alternaria alternata 75.0 75.0 30.0 40.0# 50.0 20.0# 10.0 20.0# 100.0* 100.& 80.0 80.0 38.46 15.4# Cladosporium 25.0 58.3 20.0 10.0# 60.0 40.0 10.0 20.0 0* 0# 20.0 53.3 0* 15.4 cladosporioides 70.0# 0# 20.0 0 0#Cylindrocarpon 16.67 25.0 40.0 40.0 0 20.0 90.0* 0 0 magnusianum 0# 0# 15.4#Epicoccum nigrum 83.3 58.3 60.0 60.0 90.0 50.0 20.0* 0* 33.3* 46.7 15.4* Geomyces pannorum 16.7 0 0 0 0 0 0 0 0 7.1 0 0 0 84.5# Mucor hiemalis f. 91.7 91.7 80.0 70.0 90.0 70.0 80.0 60.0 7.1 * 0# 60.0* 0# 84.6 76.9 hiemalis Penicillium 0 0 0 0 0 0 90.0* 0 0 0 0 0 0 0 purpurogenum 30.0# 70.0#Penicillium rugulosum 0 0 0 0 0 20.0 0 0 0 0 7.7 15.4 Penicillium verrucosum 16.7 75.0 0 90.0 0 60.0 0 70.0 0 14.3# 13.3 33.3# 0 69.3 Penicillum viridicatum 0 0 0 0 0 80.0# 0 40.0# 0 0 0 0 0 0 Phialophora 0 0 0 0 0 0 0 0 0 0 66.7* 0 0 0 cinerescens Phlebiopsis gigantea 0 33.3 0 0# 0 0# 0 20.0 0 0# 0 0# 53.8* 61.5 Phoma putaminum. 50.0 83.3 60.0 80.0 80.0* 90.0 30.0 90.0 71.4 42.8# 53.3 20.0# 0* 0# Ulocladium atrum 16.7 16.7 JO.O JO.O 10.0 10.0 0 0 85.7* 85.7# 40.0 40.0 15.4 15.4 TOTAL ISOLATED 31 24 20 21 20 25 15 22 5* 1 1# 23 21 23 25 TAXA included those with F<60%a * and # values significantly different (Kruscal-Wallis test, p::; 0.05) respect to CI and Cafter one and two years Most species were significantly enhanced or decreased respect to control stumps (C) by one or more treatments and sorne species were selectively associated with specifie treatments. It is noteworthy that P. gigantea was isolated, after one year, only from stumps treated with this fungus (F = 50%). After two years, that P. gigantea was also isolated from the control stumps and from stumps treated with urea 30%. Correspondence analysis relative to the results one year after the treatments (Fig 3) showed that B is separated from the others along the first axis. Four groups are recognizable along the second axis: the treatments VI, V2 and control (C), the treatment Cu, the treatment U3 and the treatment PG. The scatterplot relative to the whole data (Fig. 4) showed that after two years B treatment is still segregated from the control along the first axis. After two years the treatments with urea (10, 20 and 30%) and with Cu group together with the control, whereas the treatment with PG diverge from it (along the second axis). 150 Control ·Cu - ~ , '" _ U3 Ul U2 ·.C ~ ~ , '" c~ U3 Ul • ~ ·Cq,U1.. U3 - U2. é·ü;c Pp PG 8 o o0- 2.5 1.5 0.5 _,~5i -2.5 -0.5 -1.0 -1.5 - _2.0-3.5 o - 1 B. c::3> 'S o 1.0 - 0.5 0 AxIs 2 2.0 1.5 -, 1 , _,01 -2 -3 -2.0 -1.0 -1.5 2.0 1.5 1.0 0.5 0 0 -0.5 Axis 2 -, Figure 3. Scatterplot of the correspondence analysis of mycocenoses one year after the treatments (second trial). Figure 4. Scatterplot of the correspondence analysis of mycocenoses one and two years (bold and italics) after the treatments (second trial). DISCUSSION Many of the isolated species are weIl known colonizers of stumps of forest trees (Kaarik and Rennefelt 1957; Meredith 1959, 1960; Rayner 1977a; Domsch et al. 1980 Mugnai and Capretti 1987; Matta 1996; Nicolotti and Varese 1996). The qualitative and quantitative predominance of Deuteromycetes has a1ready been reported on Scots pine and black pine stumps (Meredith 1960) and on silver fir stumps (Mugnai and Capretti 1987), and may be partly due to the isolation method we employed, which favors the development of the anamorph rather than the teleomorph. The presence of Zygomycetes, especially of Mucor hiemalis f. hiemalis is appreciable. Although mucoraceous fungi are not regularly associated with wood, there are several reports of their occurrence in wood at an advance stage of decomposition (Rayner and Boddy 1988). The mycocenoesis associated with the control stumps, after one and two years in both trials, are similar to each others, both from a quanti- and qualitative point of view. This would indicate a relative stability of the natural microfungal communities of P. abies stumps. The number of fungal taxa isolated from the treated stumps after one year was always lower than from the control stumps. However, after two years, the number of fungi isolated from treated stumps was generally similar to the one of control stumps. This result suggests that the effects of the different treatments on the biodiversity of the stump fungi decrease over the time, thus enabling the restoration of the natural mycocenosis. Multivariate analysis illustrated the influence of the biological and chemical treatments on the quali- and quantitative composition of the microfungal community of Picea abies stumps. In the first trial, they also differentiated the two groups of non-treated stumps (Cl and C2).The covered stumps hosted more and differently composed populations, presumably because the disk resulted in greater humidity and acted as a shield against radiation. Thus, the effects of each treatment may be due to both the biological or chemical agent and the presence of the disk. These effects could be either direct or the result of interactions at the community level. Control 151 The TH treatment was the most divergent from the control, both after one and two years. It may be supposed that Trichoderma harzianum's high saprotrophic ability and its ecological adaptation to the low forest temperatures encourages its complete stump colonisation and drasticaHy reduced the fungal diversity. Marked reduction of spruce stump biodiversity had been described by Kallio and Hallaksela (1979) in their successful use of Trichoderma viride against Heterobasidion annosum. T. harzianum, however, had no effect on this pathogen in our study (Nicolotti et al. 1999). The treatments with the three lignivorous Basidiomycetes (HF, PV, VC) showed very similar results; one year after the treatments the related mycocenosis were much more similar to each other than to aH the others, however two years after the treatments the differences minimized. Treatments YB and FVB were the least divergent treatment from the control both after one and two years. Their effects on the fungal community, indeed, were similar to the application of a wood disk only, as in CI. Since they are among the best treatments against Heterobasidion annosum (Nicolotti et al. 1999) they can be regarded as a suitable form ofbiological control. The mycocenosis associated with the chemical treatment with propiconazole (TT) after one year was different from those of the other treatments and similar to that of C2; after two years the differences minimized. Other chemical compounds have already been shown to selectively and significantly reduce or enhance the incidence of sorne fungal species (Meredith 1960; Rayner 1977 a,b; Lipponen 1991). In this study, the propiconazole favoured the establishment of Phoma putaminum and Aureobasidium pullulans, two species whose F values were also high in C2. This finding, referred to a treatment that has been proved to be very effective against Heterobasidion annosum (Nicolotti et al., 1999), emphasises the propiconazole's good compatibility with the environment. The second trial showed that treatments with urea 10% and 20% had the least effects on the stump mycocenosis. These treatments were grouped together with the control both after one and two years. The effects of the treatments with urea 30% and copper were pronounced in the first year however mitigated during the second year, showing that the impact ofthese treatments decrease over the time. The treatment with borate had the worst effects on the stump mycocenosis. In both samplings, the number of fungal species isolated from the stumps treated with borate was significant lower than from control stumps. Besides, from a qualitative point of view sorne species were significantly enhanced and others significantly decreased in their frequencies of colonization. The marked negative impact of this treatment on the microfungal communities seems to persist over the time. This result suggest caution in the evaluation of the ecocompatibility of this treatment since until now borates have shown null or low toxicity for mammalian, plants, fishes and invertebrates (Anonymous 1995, 1996; Pratt 1996). P. gigantea had a marked effect on the stump mycocenosis as weil. This effect is mainly evident on the qualitative composition of the mycoflora. The impact of this treatment on the stump mycocenosis seems to persist over the time and even to increase. It is noteworthy the persistence of this fungus on most of the inoculated stumps and its spread to other treated and untreated ones. The high ability of colonization and persistence of P. gigantea on stumps is very important in forest disease management, but it may also represent an environmental hazard. The fungus we used is an exotic strain of an indigenous organism quite rare in Western Italian Alps (Breitenbach and Krazlin 1986; Bernicchia, personal communication). As underlined by Holdenrieder and Greig (1998), any introduction of exotic living materials present a potential hazard which can lead to an unwanted shift in the biodiversity of the system. In conclusion, this paper provides evidence that the patterns of fungal colonization of P. abies stumps is influenced, sometimes greatly by the treatments against H annosum. Hence, the final choice of a biological or chemical treatment against H annosum should not leave out of consideration the impact of these treatments on non-target organisms inhabiting stumps. The need to prolong the treatments for many years makes particularly 152 Control important a carefu1 eva1uation of the environmental hazard since the possibly negative effects could add up as times goes by. ACKNOWLEDGEMENTS We thank Dr. Giorgio Buffa for its he1p in data analysis. This study was supported by M.U.R.S.T, grant 40% and 60%, and by the "Région Autonome Vallée d'Aoste". REFERENCES Anonymous. 1995a. Reproductive and general toxico10gy of sorne inorganic borates and risks assessment for human beings. Technical report No 63, European Centre for Ecotoxicology and Toxicology of Chemicals. Brussels. 91 p. Anonymous. 1996a. Guidelines for the efficacy eva1uation for plant protection products. Conduct and reporting of efficacy evaluation trials. Bulletin OEPP/EPPO Bulletin 26: 251-271. Breitenbach J.; F. Kriinzlin. 1986. Pi1ze der Schweiz, Nichtbliitterpilze, Band 2. Verlag Mykologia, Luzem. 416 p. Butcher, lA. 1968. The ecology of fungi infecting untreated sapwood of Pinus radiata. Canad. J. Bot. 46: 1577­ 1589. Dix, N.J.; 1. Webster, 1995. Fungal Ecology. Chapman and Hall, London, UK. 549 p. Domsch K.H.; W.Gams; T.-H. Anderson, 1980. Compendium of Soil Fungi. Academic Press, London, UK. 865 p. Etheridge, D.E. 1968. Factors affecting infections ofbalsam fir (Abies balsamea) by Stereum sanguinolentum in Quebec. Canad. J. Bot. 47: 457-479. Holdenrieder O.; RJ.W. Greig. 1998. Biologica1 methods of control. In: Heterobasidion annosum Biology, Ecology, Impact and Control. Woodward S., Stenlid 1., Karjalainen R. and Hütterrnann A. (eds). CAB International, pp.235-258. Kiiiirik , A., and E. Rennerfelt, 1957. Investigations on the fungal flora of spruce and pine stumps. Meddeland. Statens. Skogs-Forskningsinst. 47: 2-87. Kallio, T.; A.M. Hallaksela, 1979. Biological control of Heterobasidion annosum (Fr.) Bref. (Fomes annosum) in Finland. Eur. J. Forest Pathol. 9: 298-308. Kaufmann M.R.; R.T. Graham; D.A. Boyce; W.H. Moir; L. Perry; R.T. Raynolds R.L. Bassett; P. Mehlop; L.B. Edminster; W.M. Block; P.S. Corn. 1994. An Ecological Basis for Ecosystem Management. USDA Forest Service, General Technical Report 264.22 p. Lipponen, K. 1991. Stump infection by Heterobasidion annosum and its control in stands at the first thinning stage. Fo1ia. Forest. 770: 1-12. Matta, A. 1996. Fondamenti di Patologia Vegetale. Pàtron Editore, Bologna, Haly. 494 p. Meredith, D.S. 1959. The infection ofpine stumps by Fomes annosus and otber fungi. Ann. Bot. 23: 455-476. Meredith, D.S. 1960. Further observations on fungi inhabiting pine stumps. Ann. Bot. 24: 455-476. Mugnai, L.; P. Capretti. 1987. Osservazioni sulla microflora fungina di ceppaie di abete bianco. Giorn. Bot. Ital. 121: 305-312. Nicolotti, G.; G.C.Varese. 1996. Scrrening of antagonistic fungi against air-borne infection by Heterobasidion annosum on Norway spruce. Forest Ecol. Managem. 88: 249-257. Nicolotti G.; Gonthier P.; Varese G.C. 1999. Effectiveness of sorne biocontro1 and chemical treatrnents against Heterobasidion annosum on Norway spruce stumps. Eur. J. For. Path. 29: 339-346. Pratt J.E. 1996. Borates for stump protections: a literature review. Forest Commission Technica1 Paper 15. Edinburg. 19 p. Pratt J.E.; J. Johansson; A. Hütterrnann. 1998. Chemica1 control of Heterobasidion annosum. In: Heterobasidion annosum Biology, Ecology, Impact and Control. Woodward S., Stenlid l, Karjalainen R. and Hütterrnann A. (eds). CAB International, pp.259-272. Control 153 Rayner, AD.M. 1977a. Fungal colonization of hardwood stumps from natural sources. 1. Non-Basidiomycetes. Trans. Brit. Mycol. Soc. 69: 291-302. Rayner, A.D.M. 1977b. Fungal colonization of hardwood stumps from natural sources. Il. Basidiomycetes. Trans. Brit. Mycol. Soc. 69: 303-312. Rayner, A.D.M.; L. Boddy. 1988. Fungal Decomposition Of Wood. Its Biology and Ecology. John Wiley and Sons, Chichester, UK. 587 p. Redfern , D.B.; J. Stenlid. 1998. Spore dispersal and infection. In: Heterobasidion annosum Biology, Ecology, Impact and Control. Woodward S., Stenlid J., KaIjalainen R. and Hüttermann A (eds). CAB International, pp.l 05-124. Rishbeth, J. 1959 a. Stump protection against Heterobasidion annosum. 1. Treatment with creosote. Ann. Appl. Biol. 47: 519-528. Rishbeth, J. 1959 b. Stump protection against Heterobasidion annosum. Il. Treatment with creosote. Ann. Appl. Biol. 47: 529-541. Stenlid, J.; D.B. Redfern. Spread with the tree and stand. In: Heterobasidion annosum Biology, Ecology, Impact and Control. Woodward S., Stenlid J., KaIjalainen R. and Hüttermann A. (eds). CAB International, pp.125-142. Syntax-pc 1993. Version 5.0 Computer Programs for Multivariate Data Analysis in Ecology and Systematics. Scientia Publishing, Budapest. Varese G.c.; G. Buffa; AM. Luppi; P. Gonthier; G. Nicolotti; G. Cellerino. 1999. Effects of biological and chemical treatrnents against Heterobasidion annosum on the microfungal communities of Picea abies stumps. Mycologia 91(5): 747-755. 154 Control GROWTH OF INOCULATED HETEROBASIDION ANNOSUM IN ROOTS OF PICEA ABIES­ EFFECTS OF TIllNNING AND STUMP TREATMENT WITH PHLEBIOPSIS GIGANTEA M. Pettersson*, and J. R6nnberg* *Swedish University of Agricultural Sciences, Southem Swedish Forest Research Centre, P.O. Box 49, SE-230 53 Alnarp, Sweden SUMMARY The spread rate and incidence of Heterobasidion annosum (Fr.) Bref. in the roots of Norway spruce (Picea abies (L.) Karst.) was studied in three unthinned first rotation Norway spruce stands on former arable land in southwestem Sweden. A known isolate of H. annosum was inoculated in 135 trees. One year after the inoculation, 2/3 of the trees were thinned and 1/3 of the trees were left standing. Half of the thinning stumps were treated with spores of Phlebiopsis gigantea (Fr.) Jül. and half were left untreated. The length of spread and the incidence of H. annosum was investigated three and five years after the inoculation of H annosum. Three years after the inoculation, H .annosum had spread significantly longer (50.0 cm) in the roots of the untreated stumps than in the roots of the standing trees (28.8 cm). The length of spread in the roots of the treated stumps was 41.0 cm. Five years after inoculation the length of spread could only be measured in the roots of the standing trees (46.6 cm). The incidence of H annosum was compared at different distances from the inoculation point and was significantly lower for the standing trees than the other treatments. The study shows that thinning of P. abies infected by H annosum will increase the spread rate of H. annosum within the root system of decayed trees and the risk for transfer of the disease to adjacent trees. Keywords: Picea abies, H. annosum, stump treatment, Phlebiopsis gigantea, spread rate, incidence INTRODUCTION Root and butt rot causes major economic losses in the Swedish forestry sector. The most important cause of decay is Heterobasidion annosum (Fr) Bref. About 15% of ail mature Norway spruce trees, Picea abies (L.) Karst. in Sweden are infected by root and butt rot (Bengtsson 1975). The main route of infection of H annosum is by airbome basidiospores that establish on freshly cut stumps and wounds, and subsequently spread by growth of mycelia (Rishbeth 195Ia,b; Isomiiki and Kallio 1974). Since spores are spread during the warm part of the year thinning during this period favours the spread of H annosum. Furthermore, Bendz-Hellgren et al. (1999) found that the growth rate of H annosum in the roots of Norway spruce increased after felling of the infected tree. However, the study by Bendz-Hellgren et al. (1999) was not a true comparison between the growth rate of H. annosum in standing trees and stumps since the disease was not introduced in the same way. Consequently forest management without thinning might limit the spread of H annosum in healthy stands as weil as in infected stands. Stump treatment can reduce spore infections on fresh, uninfected spruce stumps by 90-95% (Korhonen et al. 1994, Thor and Stenlid 1997). However, survival of H annosum for 15-60 years in spruce stumps has been reported (Greig and Pratt 1976; Piri 1996). Therefore an infected stump can function as a source of infection for a long time and the effect of stump treatment in infected stands can be questioned. Korhonen et al. (1994) have shown that there is a potential for Phlebiopsis gigantea (Fr.) Jül, competing with H annosum in Norway spruce stumps, to reduce the subsequent transfer of H. annosum to adjacent trees. However, this statement was based on a small sample. Further investigation was recommended before drawing any practical conclusions. Control 155 The aim with this study was to evaluate the growth rate and incidence of H. annosum, inoculated in standing P. abies, in the root systems of standing trees and stumps and the potential of P. gigantea in reducing the growth rate and incidence of H. annosum when applied to stumps already infected with H. annosum. MATERIAL AND METHODS The experiment was established in 1995 in three unthinned first rotation Norway spruce stands on former arable land in southeastem Sweden (56°40' N, 13°10' W). It was designed as a randomised block trial. Each stand consisted of 15 blocks with three trees in each block. The trees were sound and the difference in diameter at breast height between the thickest and the thinnest tree was less than four centimetres. Between each tree in the block at least one growing tree was left. The three trees in each block were inoculated with cutter shavings infected by a known isolate of H. annosum.Inoculation was done 10 cm above ground over the main root c10sest to south, using an increment borer. In 1996, two trees in each block were randomly selected and felled. The standing tree will be referred to as the T-treatment. The surface on one of the two stumps in each block was manually treated with a suspension of P. gigantea oidiospores (Rotstop, Ig/ml) (P-treatment) and the other stump was left untreated (C-treatment). In the autumn of 1998, seven of the 15 blocks in each stand were randomly selected. The remaining standing tree in each block was felled. In the fresh stumps (T-treatment) and the untreated stumps (C-treatment), ail discoloured roots were sampled. The discoloration was measured and discs were taken at 20, 40 and 60 cm from the inoculation point and 10 cm beyond the end of the discoloration. In the stumps treated with P. gigantea (P-treatment) discs were cut at 20, 40 and 60 cm from the inoculation point for aH main roots. Discoloration was measured and a disc was cut 10 cm beyond the end of the discoloration. Ail discs were directly transferred into plastic bags and incubated for 10 days. Examination for H. annosum in its conidial stage was done using a stereomicroscope. The examination of P. gigantea was based on the colour of discoloration and the presence of oidia in mycelia taken from the cut discs. In the autumn of 2000, seven of the remaining eight blocks on each location were randomly selected. The remaining standing tree on each block was felled and the T-stumps were sampled as in 1998. The C and P-stumps were excavated. Due to the general degradation ofthese stumps that made it difficult to follow discoloration in the roots, discs were cut 20, 40 and 60 cm from the inoculation point in aU main roots. Discs were handled and examined as in 1998. For aU stumps in 1998 and 2000, the total number of main roots were counted. The individual of H. annosum was tested by using pure cultures made from conidia taken from the sample discs. The pure cultures were grown on agar media (Hagem Agar) together with the known isolate. If no borderline was formed between the two isolates the samples were regarded as being of the same genotype. The individual from 58 roots in 1998 and from Il roots in 2000 was tested. The length of spread of H. annosum was based on the length of discoloration but adjusted according to the presence of conidia. If the discoloration extended to 60 cm and conidia of H. annosum was found at 40 but not at 60 cm the length of discoloration was used, i.e. 60 cm. If the length of spread was 40 cm and conidia was found at 40 and at 60 cm the length ofspread was set to 60 cm. A mean of the adjusted values per stump and a mean for each treatment and year were calculated. Due to the general degradation, no numbers for the C- and P-treatments for 2000 are presented. The incidence of H. annosum in the roots was calculated as the proportion of main roots with conidia 20, 40 and 60 cm from the inoculation point for each stump. Thereafter the mean of these proportions for each treatment and year was calculated. The SAS general linear models procedure for split-plot designs (SAS Institute Inc. 1998) was used to perform statistical tests. Differences between individual treatments were evaluated with Tukey's significant difference mean separation test. 156 Control RESULTS ln 1998, H. annosum had spread longer in the stumps of the C-treahnent than in the stumps of the T­ treatment (p<0.05, Table 1). The P-treatment did not significant1y differ from any of the other treatments. Maximum length of spread was 200 cm for the C-treatrnent, 140 cm for the P-treatment, and 85 cm for the T­ treatment. The mean length of spread in 2000, only measured in the T-treatment, was 46.62 cm, and the maximum 80 cm. The incidence of H. annosum in 1998 and in 2000, i.e. the presence of conidia in the discs at 20, 40 and 60 cm distance from the inoculation point, was significantly 10wer for the T-treatment than for the other treatments, except at 60 cm in 1998, where no difference was found (Table 2). There was no significant difference between the P- and C-treatments. Both in 1998 and in 2000, 91 % of the isolates were of the same genotype as the inoculated isolate. Phlebiopsis gigantea was found in 19% of the treated stumps in 1998 and in 83% in 2000. In stumps where H. annosum was absent, the presence ofP. gigantea was the same as in stumps where H. annosum was present. Table 1. Mean spread length of H annosum in roots of P. abies based on the length of discoloration in roots adjusted according to the presence of conidia. "c" means untreated stumps, "P" means stumps treated with P. gigantea, and "T" is standing trees. Treatment Length of spread (cm) 1998 2000 C 50.04 al P 40.99 ab T 28.76 b 46.62 1 Figures for length of spread with different letters are significant1y different (p<0.05). Table 2. Incidence of H. annosum in the roots of P. abies, i.e. the proportion of main roots containing H. annosum at different distances from the inoculation point. "c" means untreated stumps, "P" means stumps treated with P. gigantea, and "T" is standing trees. Incidence at different distances from inoculation point 20cm 40cm 60 cm Treatment CPT CPT C P 1998 30.4 al 29.5 a 0.8 b 20.8 a 19.8 a 0.0 b 14.4 a 11. 8 a 2000 56.5 a 46.0 a 16.8 b 65.8 a 45.8 a 14.8 b 48.9 a 42.9 a 1 Figures with different !etters are significantly different (p<0.05). DISCUSSION T 0.0 a Il.3 b In this study, the average growth rate of H annosum significant1y increased after felling of infected trees. This finding is supported by Bendz-Hellgren et al. (1999). The increased growth of H. annosum after felling is probably due to the defence mechanisms of living root wood that is inactivated by felling of the tree. The barrier zones to decay associated with H. annosum and Armillaria mellea (Vahl. Ex Fr.) Quel. in conifer roots has been described by Tippett and Shigo (1980, 1981). Control 157 Corresponding with the study by Bendz-Hellgren et al. (1999), the average growth rate of H. annosum in roots of living trees was about 9.5 cm x yea(l. Contrarily, the average growth rate of H. annosum in stump roots in the study by Bendz-HeIlgren et al. (1999) was 25 cm x yea(1 and about 16.7 cm x year- I for the C-treatment this study. However, in the study by Bendz-HeIlgren (1999), H. annosum was inoculated into stump roots and in this study the fungus was inoculated in trees that were felled one year later. Therefore, these numbers are not comparable. The spread length for the C-treatment in 1998 were significantly higher than the spread length for the T­ treatment. At the same time the incidence of H. annosum identified by its conidial stage was almost zero in the T­ treatment. It could therefore be argued that the tree's natural defence in the roots of the T-trees had killed infections of H. annosum. This is also supported by Bazzigher (1986) who found a dying off of H. annosum in artificially infected Sitka spruce trees over a two-year period. However, in 2000, the spread length had increased and the incidence of H. annosum identified by its conidial stage was much higher. Since the T-treatment in 1998 had no infection by H. annosum at any distance from the inoculation point except at 20 cm, despite the length of discoloration, and not in 2000, the result can be questioned. Tt is possible that H. annosum was present in the wood without being identified by the presence of its conidial stage in 1998. H. annosum had on average spread about 9 cm shorter in the P-treatment than in the C-treatment in 1998. The effect of the P. gigantea treatment on the growth rate of H. annosum was however not significant. Furthermore, the incidence of H. annosum was slightly lower, however not significant, for the P-treatment than for the C-treatment in 2000. As there was no difference in incidence but in length of spread in 1998, it could be discussed that P. gigantea first slows down the spread rate of H. annosum in the root and after that takes over the space initially occupied by H. annosum. If the length of discoloration had been possible ta measure in 2000, this might have supported this idea. However, Korhonen et al. (1994) have reported growth rates for P. gigantea of more than 20 cm in three months. This indicates that P. gigantea may have had the potential to quickly compete with and possibly overgrow H. annosum in the stumps. However, such a rapid colonization could be questioned since only a few main-roots in the stumps of the P-treatment seemed to be colonised by P. gigantea in 1998 and no other significant effects were found. The incidence of H. annosum in roots increased for ail treatments between 1998 and 2000. This shows that a higher number of main roots were occupied by H. annosum in 2000 than in 1998. As the incidence was estimated for the same distances from the incoculation point in aIl roots, similar levels of the incidence H. annosum could have expected for bath sampling years, at least for the C- and T-treatments. However, since the inoculation was carried out on one side of the stump the H. annosum infection had to grow across the fibre direction to be able ta reach the roots on the opposite side. The growth across the fibre direction may slow down the development of the infection (Rennerfelt 1946). Due to problems with extensive contamination by bacteria and viroses on agar-cultures, the number of roots tested for the genotype of H. annosum became much lower for roots sampled in 2000 than in 1998. Due to the small sample in 2000, it can be argued that other individuals of H. annosum than the one first inoculated have been present in the roots. However, as only one of the tested samples was of an unknown genotype in 2000, the influence by other individuals of H. annosum should have been small. CONCLUSION To conclude, due to an increased spread of the H. annosum in roots of infected trees after felling, caution should be shown before taking decision on thinning of Norway spruce stands infected by H. annosum. Furthermore, stump treatment with P. gigantea may reduce the spread of H. annosum in the root systems of stumps already infected by H annosum. However, this was not significant, and therefore additional studies of the influence of P. gigantea on the development of H. annosum infections are suggested before any further conclusions are drawn. 158 Control ACKNOWLEGEMENTS The authors would like ta thank Rolf Overgaard and Anna Eidelin and the staff at Tonnersjoheden Experimental Forest for help with the field work. Urban Nilsson deserves a thanks for support with statistical analyses and Gudmund Vollbrecht for useful comments on the manuscript. REFERENCES Bazzigher, G. 1986. Infection studies with Heterobasidion annosum on young trees of Picea abies. Eur. J. For. Path. 16: 125-128. Bendz-Hellgren, M., Brandtberg, P.-O., Johansson, M., Swedjemark, G. and Stenlid, J. 1999. Growth rate of H annosum in Picea abies established on forest land and arable land. Scand. J. For. Res. 14: 402-407. Bengtsson, G. 1975. Skador pa skog belysta genom rikstaxen. Skog och virkesskydd. Sveriges skogsvardsforbund, pp. 58-79. (In Swedish). Greig, B. J. and Pratt, J. E. Sorne observations on the 10ngevity of Fomes annosus in conifer stumps. Eur. J. For. Path. 6: 250-253. Isomiiki, A. and Kallio, T. 1974. Consequences of injury caused by timber harvesting machines on the growth and decay ofspruce (Picea abies (L.) Karst.). Acta For. Fenn. 136,25 pp. Korhonen, K., Lipponen, K., Bendz, M., Johansson, M., Ryen, L, Venn, K., Seiskari, P. and Niemi, M. 1994. Control of Heterobasidion annosum by stump treatment with "Rotstop", a new commercial formulation of Phlebiopsis gigantea. Tn Johansson, M. and Stenlid, J. Proceedings of the eigth international conference on root and butt rots, Aug. 9-16, 1993, IUFRO Working Party S2.06.0 l, Uppsala, Sweden., Univ. Agric. Sci. pp. 675-685. ISBN 91-576-4803-4. Piri, T. 1996. The spreading of the S type of Heterobasidion annosum from Norway spruce stumps to the subsequent tree stand. Eur. J. For. Path. 26: 193-204. Rennerfelt, E. 1946. Om rotrôtan (Polyporus annosus Fr.) i Sverige. Dess utbredning och siitt art upptriida. Medd. Statens Skogsforskningsinstitut 35(8): 88p. (In Swedish with German summary.) Rishbeth, J. 1951 a. Observations on the biology of Fomes annosus, with particular reference to East Anglian pine plantations. II. Spore production, stump infection, and saprophytic activity in stumps. Annals of Botany 15: 1-21. Rishbeth, J. 1951b. Observations on the biology of Fomes annosus, with particular reference to East Anglian pine plantations. III. Natural and experimental infection of pines, and sorne factors affecting severity of the disease. Annals of Botany 15: 221-246. Tippet, J. T. and Shigo, A. L. 1980. Barrier zone anatomy in red pine roots invaded by Heterobasidion annosum. Cano J. For. Res. 10: 224-232. Tippet, J. T. and Shigo, A. L. 1981. Barriers to decay in conifer roots. Eur. J. For. Path. Il: 51-59. Thor, M. 1997. Stump treatment against Heterobasidion annosum - techniques and biologica1 effect in practical forestry. Licentiate's dissertation. Control 159 EFFECT OF STUMP TREATMENT ON TRAN8FER OF HETEROBASIDION ANNOSUM ROOT ROT IN NORWAY SPRUCE LM. Thomsen Danish Forest and Landscape Research Institute, Hersholm Kongevej Il, DK-2970 Hersholm, Denmark. E-mail: imt@fs1.dk. SUMMARY A stump treatment experiment with Rotstop and urea at 20% and 30% (w/v) was carried out in a first rotation, unthinned Norway spruce stand in Denmark. Six months after inoculation with H. annosum using S type conidia, the incidence of infection was 15% (Rotstop), 5% (30% urea) and 3% (20% urea) among the treated stumps and 88% for untreated stumps. The Rotstop treated stumps with infection by H. annosum had mostly very small colonies of H. annosum and when resampled after another six months, the infection incidence had dropped to 5%. Five years after inoculation, infection had spread to 38% of trees adjacent to untreated stumps but to oruy 7-14% oftrees next to treated stumps. Infection was proved to originate from stumps by somatic compatibility. In a number of cases, the extent of decay and stains could be classed as severe. Based on whether stumps were found to be infected or uninfected in the first assessment (after six months), 43% of trees adjacent to infected stumps (untreated or failed treatment) had rot at the stump surface when examined five years later. In contrast, only 7% of trees adjacent to treated or untreated stumps, that were classed as uninfected in 1995, had rot at the stump surface. It may be concluded that trees next to untreated stumps or to stumps, where treatment has been inadequate, are at greater risk of infection than trees next to stumps, where infection by H. annosum has been prevented. An additional finding from this trial was made possible, because the parentage of each tree or stump was known. Part of the observed variation in disease expression among standing trees couId thus be accounted for genetically. Keywords: stump treatment, Heterobasidion annosum, rot transfer, Picea abies INTRODUCTION Since the middle of the 1960's, stump treatment experiments against Heterobasidion annosum have been common in Denmark, (Yde-Andersen, 1963, 1966, 1982). Mostly, those experiments were designed to test the efficacy of various protective agents such as creosote, sodium nitrite (NaN03), urea (CO(NH2)2) and, recently, Rotstop® (Phlebiopsis gigantea). Thus, the experiments mainly consisted of applying stump treatment on halfthe selected stumps, leaving half untreated, waiting for natural infection, and then retuming several months later to take off discs for assessments. Efficacy of an applied agent was considered proven, if none of the treated stumps had colonies of H. annosum, identified by the presence of conidiophores after incubation, and the untreated stumps had a significantly higher amount of infection. However, sometimes the experiments were inclusive due to the lack of infection on the untreated stumps. According to the technician in charge of conducting almost all the Danish stump treatment experiments, an infection frequency of 20% on untreated stumps was considered very satisfactory, but seldom achieved when relying on natural infections (M. Egebjerg Petersen, pers. comm.). Most stump treatment experiments have been carried out in first generation, unthinned conifers, mainly Norway spruce (Picea abies) or Scots pine (Pinus sylvestris). This reduces the likelihood that H. annosum is already present in the stumps, which could influence assessment results. Only rarely have experiments been carried out on sites already infected by H. annosum, and the Danish results achieved with urea have not been convincing. Even more rarely have the stump treatment experiments been followed up by later assessments of subsequent transfer into neighbour trees. One very early study (Paludan 1966) illustrated the effect of stump treatment on subsequent spread of H. annosum. Results showed that 23% of stumps in the rows treated with creosote were infected by H. annosum, whereas 68% of untreated stumps were infected. During the 10 years after 160 Control stump exposure, 25% of trees next to untreated rows of stumps died due to attack by H. annosum. By making wood borings, it was estimated that the fungus was present in up to 42% of trees next to untreated rows. In contrast, only 8% of trees next to rows of treated stumps died, although the fungus was proved present in 4 tirnes the number of trees. Stump treatment thus reduced the impact of H. annosum, even though the stump treatment experiment was done with creosote, which in later studies was shown to be even worse than no treatment (VoIlbrecht and J0rgensen 1999). In the last decade, many countries have renewed efforts to find and test stump treatment materials. Sweden and Finland have been especially active in this field. This research led to the introduction of Rotstop®, a formulation of the vegetative spores of Phlebiopsis gigantea, and to the recommendation of 30% urea instead of the 20% traditionally used in Denmark. Due to these new findings, a Danish stump treatment experiment was set up in 1995 to test the efficacy of Rotstop and to compare 30% urea with 20% urea. This paper reports the findings conceming the stump treatments and subsequent transfer to living trees. MATERIALS AND METHOnS The experiment site (F 175A) was a first generation Norway spruce planted in 1978 on former agricultural land north of Copenhagen, Denrnark. Average rainfaIl 613 mm per year. The plant material consisted of 3-year­ old seedlings originating from controlled crosses between selected mature parent trees. The mating design was factorial with three female clones each mated to a set of Il other males. The field design was randomized blocks with single-tree plots and 30 blocks. Adjacent to F175A, there were two other plantings (F175B and F209) with plants of other genetic origin. In addition, reference plants from a bulked stand collection (Rye N0rskov) were mixed into ail three stands. The stands were mapped, so the genetic origin of each tree was known. On the last two sites, no stump treatments were done and aIl stumps were inoculated with either an S or a P type of H. annosum. The stump treatment site consisted of 18 rows with 35 trees in each row. Distance between rows was 2 meters. Around the planting a border of two trees was planted to minimise edge effects. Until 1995, no thinnings had been carried out. The heights and diameters had been measured, and Pilodyne readings for wood density had been done for every tree. In 1995 the 20-year-old trees had a mean DBH of 116 mm, varying from 33 to 183 mm. Stump treatment In the beginning of June 1995, every third row oftrees were cut (row no 2, 5, 8, Il, 14, and 17), leaving 100 cm high stumps. Felled trees were removed from the plot. On June 19, the stump treatments and inoculations were carried out. The treatments were Rotstop, 20% and 30% urea and untreated, and stumps were assigned according to randomised blocks design, with more stumps for the Rotstop treatment and untreated group than the two urea treatments (table 1). Out of a total of 21 0 stumps, five stumps were rejected, either because the tree had died from suppression or was less than 3 cm in diameter at stump height. The 100 cm high stumps were cut down to approximately 50 cm with a chain saw, and the treatment applied immediately after cutting. Rotstop was sprayed on the stump surface from a flask containing a suspension made according to the manufacturer's instruction on the bag (l gr per liter). The urea treatments, 20% (1 kg urea per 4 litres of water) and 30% (l.5 kg urea per 4 litres of water), were applied with a brush. After treatment aU stumps (treated and untreated) were inoculated with conidia from an S type isolate of H. annosum. The spores were suspended in sterile water and sprayed evenly over the whole surface of every stump. The number of spores pr ml and the exact amount used per stump were not recorded, but inoculum levels were undoubtedly much higher than normal, natural spore infection levels. After the experiment was set up, the weather tumed hot (temp. above 25°C) and dry for several weeks. Five months after inoculation, assessment of stumps began. This took two months, with one row of stumps being assessed every two weeks. The top of the stump was removed and a 2 cm thick disc was cut with a chain saw, put Control 161 in a paper bag and taken to the labo Here the discs were wrapped in moist newspaper (outside tne paper bags) and incubated for 10 days at 20°e. Each disc was carefully examined for presence of H. annosum conidiophores with a stereo microscope. Most of the Rotstop treated discs had tumed bright orange in the sapwood, and it was noted when this had not happened on a few discs. Six months after the initial assessment (i.e. one year after inoculation) the treated stumps which had infections, were resampled to check whether H. annosum couId still be found. Transfer of H. annosum In June 2000, five years after treatment and inoculation, the remaining rows oftrees were cut. Seventeen trees were missing or dead and therefore excluded from assessment and calculations. The stumps were examined for presence of staining or incipient rot. The size of any rot was measured and the maximum extent of the rot colurnn was found by sectioning the stem. Three cases where basal wounds seemed more likely to be the source of infection were noted. With a chain saw stump discs were cut from both the newly felled trees and the untreated stumps from 1995, marked to show positions in the stand (row no and position in row, e.g. 10.57), and taken to the lab in plastic bags. Discs from trees were also marked to show the side facing the previously cut row. In the lab the position of the rot areas and the relative sizes compared to stump diameter were drawn on circles representing trees. Those stains not resembling typical H. annosum rot were marked with a "1". Isolations were made from discoloured areas by removing small wood samples with an increment hammer and placing them on PDA in petrie dishes. The discs were then incubated in the plastic bags for 10-12 days at 20°e. Presence of H annosum was determined by appearance of conidiophores on incubated discs or by growth of H. annosum mycelium from the wood samples. When isolations were successful (sorne failed due to contamination), the resulting isolates were tested for somatic compatibility against each other. This was done by placing plugs of mycelia 2 cm apart on PDA and observing the mycelial front where the two isolates made contact. If two isolates formed one continuos mat of mycelium with no barrier zone of thickened mycelium or brownish colour, they were considered to be identical. Ail the F 175A observations from 1995 and 2000 were charted on a map of the whole experiment site (Fig. 1). The map did not illustrate spacing within or between rows or actual size oftrees or rot areas. The purpose was to show position of trees and rot areas in relation to treated and untreated stumps, and to depict the rot size relative to each tree. As nearly ail untreated stumps were infected, the few uninfected stumps were highlighted by a special mark, whereas treated stumps were usually uninfected and therefore marked only, if they were infected (Fig. 1). Calculations From the data collected, several results could be obtained concerning the effects of stump treatment on infection by H annosum and the subsequent transfer into neighbouring trees. The effect of stump treatment against infection of stumps was calculated for both the first assessment after 5-7 months and for Rotstop also 12 months after inoculation. Analysis of variance was used to test the statistical effect and significance of stump treatment in relation to the 1995 results conceming presence ofH. annosum in stumps. Rot amount in trees was cIassified as severe, if the rot height was above 100 cm, or the rot occupied more than half the stump surface. Growth rates of H annosum were calculated by assuming that the distance from inoculation point to stump surface level in neighbour trees was minimum 260 cm. This came from 50 cm stump height in 1995, 200 cm distance between rows, i.e. from stump to nearest tree, and 10 cm stump height in 2000. The average growth rate was calculated by using the actual rot heights of the trees with H. annosum transferred from stumps, added with the distance (250 cm) from inoculation point. FinaIly, the maximum growth rate was calculated from the trees, where the rot height exceeded 100 cm. . Conceming the transfer of H annosum to neighbour trees, rot/stains were excluded if they were consldered as not H. annosum (12 trees) or thought to originate from wounds rather than stumps (3 trees). These 15 trees were then counted as being without rot. Neighbour trees to the 5 stumps missing in 1995 showed no rot in 162 Control four cases, but one tree (7.62) next to a missing stump (8.62) had rot towards another stump (8.61) which was untreated. Therefore the two neighbour trees to 8.62 were counted as neighbours to 8.61, but the other neighbours ta missing stumps were left out of the calculations. The number of neighbour trees with rot was related ta the original treatments of the stumps as weU as to whether the stumps were infected or not. Analysis of variance was used to test the statistical effect and significance of stump treatment in relation to the presence of rot in neighbouring trees. The influence of genetic origin of donor stumps and host trees were tested by analyses of variance of square roots of rot areas on host tree stumps. Square root transformation was applied to counter the skew distribution of this variable. Applying the same statistical method, it was also tested whether transfer of H annosum rot was more likely between stumps and trees with common genetic origin - in this case represented by half-sibs. Figure 1. A section from the top of the assessment map with results from F175A. Row numbers are shown in circles above the plot and position within rows to the left. The combination of row and position within rows is shown for stumps from 1995 and can thus be inferred for trees felled in 2000. Dark shaded circles are untreated stumps, of which ail shown were infected, except stump 8.68 (black dot mark). Light shaded circles are treated stumps, of which aU shown were uninfected, except 11.70 (white square mark). Rot in trees is shown as black stains. The stain with "7" in tree 12.71 was classified as not H annosum, and the H annosum found in 12.76 originated from a wound. The number "1" next to 12.70 and 10.68 shows that rot height was above 100 cm. "~" means that isolates were somatically compatible. The rot in tree 7.70 was towards stump 8.69 rather than 8.70, same for 10.72 towards 11.73. The interpretation could be that rot originated from the two untreated, infected stumps rather than the treated, uninfected stumps. RESULTS In the following section "stumps" refer to inoculated stumps of trees cut in 1995 and "trees" refer to trees cut in 2000. "Infection frequency" is used for H annosum infection of stumps in 1995, and "transfer rot frequency" is used for the rot frequency in trees in 2000, when the rot was originating from the stumps inoculated in 2000. Control 163 Stump treatment Assessment of the stump treatment in 1995 showed that 88% of the untreated stumps were infected by H. annosum (table 1). One stump (17.42) counted as uninfected may have had a small colony of H annosum. For the treated stumps, only 2.6% of the stumps treated with 20% urea and 5% of the stumps treated with 30% urea were infected, but 14.5% of the Rotstop treated stumps had infections of H annosum. In most cases, the infected area was rather small, e.g. 1 cm2 at the edge of the stump or in the heartwood. The rest of the stump was c1early colonized by P. gigantea in the sapwood, but not in the small circles of heartwood. This was evident from the orange staining and the strands of P. gigantea growing from the edge of the stains. On a few stumps the treatment with Rotstop had c1early failed, as few or no orange stains were present, and large areas were colonized by H annosum. The difference between untreated and treated stumps was significant (P95% >5225 Finland 9000 3000 >95% >11 400 France 300 0 Great Britain 60000 9000 80% 55200 Ireland 9950 8500 100% 18450 Norway 400 0 >95% >380 Poland 70000 10 0% 0 Sweden 35000 50 >95% >33298 Sum 189650 21060 >123953 Two countries, Finland and Poland, have reported that there are subsidies available for stump treatment. In Finland, private forest owners are compensated for the cost of material, while in Poland the State forest districts pay for costs of material, application and monitoringlfollow-up. 172 Control The costs for treatment varied between 0.2 and 2 euro per m3 solid under bark (Table 4). Locally, the cost depends on stand conditions, nature of thinning or final felling etc. Sorne of the countries did not separate costs for thinning and final feHing. In Ireland, the harvesting contractor is not paid extra for stump treatment, but the costs which have been reported coyer the supply of pre-dissolved urea solutions to the contractors. In terms of a weighted mean value, stump treatment costs - as expressed by the country representatives - are about 1.2 euro per m3sub in thinning and 0.4 euro in final felling. Table 4. The cost for stump treatment, including the cost for material and application, euro per rn3 solid under bark. Country Cost, euro/m3 sub thinning c1earfell 1.35 1.35 0.75 0.4 Denmark Finland France Great Britain Ireland Norway Poland Sweden Weighted Mean 0.5 0,2 1.35 2 1.3 1.22 0.5 0,2 0.33 0.38 DISCUSSION The country-wise estimates on which this survey relies inevitably include sorne uncertamtIes. The quantities of rnaterials used and the number of hectares treated are seldom accurately reported, but must be estimated from related facts and knowledge such as the volumes of control agents deployed. Nevertheless, the estirnates are sufficiently realistic to allow observations to be made on the approxirnate levels and trends nationally and on a European level. Calculating the area to be treated annually has sorne practical difficulties, and may not necessarily reflect the whole picture. In terms of area treated, thinning accounted for 93% while the thinning proportion in terms of harvested volume should be about 65% (assuming a yield of 50 m3sub per hectare in thinning and 250 m3sub per hectare in final felling). Differences between countries in the materials used depend on several factors, e.g. historical background, efficacy on specific tree species and legislation. Urea has a long history in stump treatment, but has lost ground in many countries. In Great Britain, which is one of the two rernaining significant users of urea, the intention is ta abandon urea because of its poor efficacy on Sitka spruce. In addition, urea is transported in bulk solution, which makes the logistics more difficult and expensive in comparison with P. gigantea and DOT. P. gigantea is a natural coloniser of pine stumps, and it is most frequently used on pines in Great Britain and Poland. In Finland, however, sorne strains of P. gigantea were discovered that are successful colonizers also on spruce sturnps (Korhonen et al. 1994). This strain was formulated into a product, "Rotstop", which is used in Denmark, Finland, Norway and Sweden. In Poland, a biological product is used with sawdust as a substrate ("PG­ IGL"), whereas in Great Britain, the "PG Suspension" is formulated with sucrase suspension packed in srnall sachets. Details about these products are given in Pratt et al. (2000). DOT has been subject to decreased use in Sweden due to its classification as a pesticide, which disqualifies it for use in woodland certified according to Swedish FSC criteria. In France, where DOT is also used, albeit in smaIl quantities, the compound is classified as a fertiliser, and is unregulated for forestry purposes. Until the European Union has harmonized the legislation on pesticides over aIl countries, the rnember states' nationallaw will state what is legal or not. Since there are few approved and suitable control agents available, and Control 173 a low probability for further products to be developed, the forest sector must ensure that those avaüable are used carefully, effectively and responsibly. Stump treatment represents an investment for the future, and must render the expected payback on the investment (he cost-effective). Assumed the extraction volumes given above, the countries in the survey invest at least 13 million euro annually in stump treatment. In Great Britain, it is no longer recommended to carry out stump treatment on sites where the risk from the disease is low. In Finland, Great Britain and Denrnark treatment is recommended and carried out in final felling as weIl as in thinning. In contrast, Sweden enjoys an ongoing discussion about whether to begin stump treatment in final felling or not. The problem is that the longer the rotation period, the harder it is to get a proper payback on the investment from treating stumps in the final feIling, given the net present value of felling assuming a specifie interest rate. Preliminary calculations indicate stump treatment in final felling to be profitable in stands with the relatively short rotation periods cornrnon in southern Sweden, but not on the Norway-spruce sites in the north where the rotation periods are too long. Regardless of these disagreements, there are sufficient areas that are currently not treated, as final-felling stands in southern Sweden and thinning stands in the north, to indicate that an overaIl increase in area treated can be expected. There are examples of incentives for stump treatment. In Finland, the private forest owners are compensated for the cost of protectant, in thinning as weIl as in final feIling. In Poland, the subsidies coyer the cost for proteetant, application and monitoring. In addition to strict economics, there are also sorne aspects concerning the amount of inoculum that can develop in the forests in the long term. To reduce the increase of inoculum, treatment of stumps is necessary. Here, the state forests could play an important part to ensure healthy stands for the future. This has been the case in Denmark, Great Britain and Poland, where stump treatment has been compulsory on state-owned forest land. In spite of the relatively large extent of stump treatment in Europe there are several issues to be addressed further: • The technology and methods for applying stump treatment in a wide range of tree and stand conditions exists, but need to be further improved. There is considerable scope to reduce the consumption of protectant in order to reduce costs and to facilitate the logistics of supply. Using properly designed equipment can help here, since retro-fitted systems can be very inefficient. • Motivation and training is essential in order to accomplish high standards of stump treatment. Without understanding the motives, contractors and forest workers object to apply a treatment that increases their costs and causes them extra work. • Assuring a good quality of work includes foIlow-up routines. Stump treatment should be an integral part of each forest company's follow-up and monitoring programme. • Harmonizing EU directives into national legislation would facilitate the use of the three different control agents that are available. A natural competition with fair conditions for ail manufacturers and suppliers of stump treatment material is essential to provide effective products to reasonable costs. ACKNOWLEDGEMENTS l thank aIl country representatives who contributed to this investigation. Obviously, without your help, this would not have been doable. Special thanks to Jim Pratt for fruitful discussions on the topic as such and for revision ofmy English language. REFERENCES Korhonen, K; Lipponen, K; Bendz, M.; Johansson, M.; Ryen, 1.; Venn, K; Seiskari, P. and Niemi, M. 1994. Control of Heterobasidion annosum by stump treatment with 'Rotstop', a new commercial formulation of 174 Control Phlebiopsis gigantea. Pages 675-685 in Johansson, M. and Stenlid, J. (eds) Proc. 8th Int. Conf. on Root and Butt Rots, Sweden and Finland, August 9-16, 1993, ISBN 91-576-4803-4. Pratt, J.B. and Thor, M. 2001. Improving mechanised stump protection against Fomes root rot in Europe. Quat. J. For. 95 (2): 119-127. Pratt, J.E.; Niemi, M and Sierota, Z.H. 2000. Comparison ofthree products based on Phlebiopsis gigantea for the control of Heterobasidion annosum in Europe. Biocontrol Science and Technology 10:469-479. Rishbeth, J. 1949. Fomes annosus FR on pines in East Anglia. Forestry 22:174-183. Rishbeth, J. 1959. Stump protection against Fomes annosus II. Treatment with substances other than creosote. Ann. Appl. Biol. 47:529-541. Rishbeth, J. 1963. Stump protection against Fomes annosus III. Inoculation with Peniophora gigantea. Ann. Appl. Biol. 52:63-77. Ronnberg, J. and Jorgensen, RB. 2000. Incidence of root and butt rot in consecutive rotations of Picea abies. Scand. J. For. Res. 15:210-217. Thor, M. and Stenlid, J. 1998. Heterobasidion annosum infection fol1owing mechanized first thinning and stump protection in Picea abies. Pages 397-407 in Delatour, C.; Guillaumin, J-J.; Lung-Escarmant, Band Marcais, B. (ed.). Proc. 9th Int. Conf. on Root and Butt Rots, Carcans-Maubuisson (France), Sept 1-7 1997. INRA, France. ISBN 2-7380-0821-6. Woodward, S.; Stenlid, J; Kmjalainen, Rand Hütterman, A. (eds). 1998. Heterobasidion annosum: Biology, Ecology, Impact and Control. CAB International, Wal1ingford ad New York. 589 pp. Control 175 STUMP TREATMENT EXPERIMENTS AGAINST HETEROBASIDIONIN THE ITALIAN ALPS N. La Portal, R. Grill02 , P. Ambrosi' and K. Korhonen3 1 Dept. Natural Resources and Environment, IASMA, Via E. Mach 2, 38010 S. Michele a/Adige (TN), Italy Fax +390461 650956, E-mail: nlaporta@mai1.ismaa.it 2 Dept. ofPhytosanitary Sciences and Technologies, University ofCatania, Via Valdisavoia 5,95123 Catania, Italy 3 Finnish Forest Research Institute, Box 18, FIN-01301 Vantaa, Finland SUMMARY The efficacy of urea, Trichoderma sp. and different strains of Phlebiopsis gigantea in controlling spore infection of Heterobasidion in Norway spruce stumps was investigated in two experiments carried out in a spruce stand heavily infected by Heterobasidion annosum s. str. and H. parviporum. In the first experiment, made in late spring during rainy weather, the protecting agents were applied to freshly-cut stumps. The Rotstop preparation proved most effective (efficacy 70-90%) and an Italian strain of P. gigantea showed sorne efficacy, while urea, Trichoderma and a P. gigantea strain from Germany failed totally. The second experiment was carried out in the auturnn in stem pieces of Norway spruce. The pieces were first incubated in the forest, later in the greenhouse. In this experiment, urea was very effective (90-100%), while Trichoderma and Rotstop were less effective (70­ 80%), and six P. gigantea strains from Italy were unsatisfactory (0-60%). P. gigantea strains grew very weakly in the stem pieces, probably as a result of low initial temperature and later high water content in the wood. Keywords: biological control, stump treatrnent, Heterobasidion annosum, H. parviporum, Phlebiopsis gigantea, Trichoderma sp., Rotstop, Norway spruce INTRODUCTION Chemical and biological control of Heterobasidion in coniferous forests has been focused mainly to preventing the spore infection on stump surfaces since the early establishment of the mycelium is the most susceptible stage to the action of antagonists (Holdenrieder and Greig 1998). Among low-toxicity chemicals, urea and borates have been found to be effective (Pratt 1994, 1996). The efficacy of Phlebiopsis gigantea (Fr.) Jül. on pine stumps was reported by Rishbeth (1963), and later experiments carried out in Finland and other countries have indicated that this fungus can be effective also in Norway spruce stumps (Korhonen et al. 1994, Thor and Stenlid 1998, Pratt et al. 2000, Soutrenon et al. 2000). In Trentino, as in most temperate coniferous forests, P. gigantea is a common inhabitant of pine and spruce stumps, and during warm rainy autumns it fruits quite abundantly on them. A rapid increase in the number of sites infected by Heterobasidion root rot in Italy together with new requirements for "environmentally-safe" pesticides has stimulated research into potential biological control agents to prevent the spread of this pathogen. The aim of this study was to compare the efficiency of the following treatrnents against a heavy natural infection of Norway spruce stumps by Heterobasidion in the Italian Alps: 1) isolates of P. gigantea originating from the Alps, 2) P. gigantea preparation Rotstop® (Kemira Agro OY), which includes a P. gigantea strain isolated from Finland, 3) strains of Trichoderma, and 4) urea. 176 Control MATERIAL AND METHODS Stump treatment experiment. A stump treatment experiment was performed on May 27, 1998, in a pure even-age plantation of Norway spruce about 35 years old, situated at 900 m a.s.l. in Trentino (Eastern Italian Alps), established on a former pasture land. The stand was heavily infected by Heterobasidion. Basidiocarps of the fungus were numerous, and the place looked suitable for a stump treatment experiment, providing a heavy natural basidiospore infection. The following treatments were applied to the stumps: 1) Rotstop, 2) P. gigantea 94135 (Germany, Munich), 3) P. gigantea 97099 (Italy, Trentino), 4) Trichoderma sp. 94268 (isolated from pine wood in Finland), 5) urea solution (30%), 6) control (sterile water). The treatment was applied immediately after cutting a tree. Spore concentration in P. gigantea suspensions was ca. 10 000 spores/ml and in Trichoderma suspension ca. 40 000 spores/ml. The amount of the solution applied to each stump was ca. 10 mlldm 2 , corresponding to a fluid layer of 1 mm on the stump surface. In the case of urea, 20 mlldm 2 was applied. Tt was raining when the treatments were carried out, and also two following days were rainy. The precipitation during these three days was 15.4, 7.0 and 15.8 mm, respectively. The temperature during the following week varied between +4 and +21 oC. The stumps were sampled after two vegetation periods, at the end of September 1999. Two ca. 3 cm thick discs were cut from the stumps. The upper disc was discarded, the lover one was washed in running water, incubated for 7 days in plastic bags, and investigated on both sides for the presence of Heterobasidion conidiophores. Also the area occupied by P. gigantea was approximately determined on the basis of characteristic orange-brown colour. Experiment in stem pieces. This experiment was made relatively late in the autumn, in October 6, 1999, using pieces cut from fresh Norway spruce stems. Four healthy spruce stems with stump diameter ca. 20 cm were selected from a site free from Heterobasidion root rot and carried to the experimental forest The basal part of each stem was cut to pieces, 20 cm long, 27 pieces from each stem. After cutting, the upper surface was divided in two symmetrical halves. One half was sprayed with the protecting agent while the other half was covered. One ta two hours later the whole surface was sprayed with sterile water. The protecting agents were: 1) urea 30%, 2) Trichoderma 94266 (isolated from pine wood in Finland), 3) Rotstop, 4) P. gigantea 97092 (Haly, Trentino), 5) P. gigantea 97097 (Trentino), 6) P. gigantea 97099 (Trentino), 7) P. gigantea 97104 (Trentino), 8) P. gigantea 95051 (Italy, Tuscany), 9) P. gigantea 95042 (Haly, Abruzzo). The spore concentrations were adjusted to ca. 5 000 spores/ml for each P. gigantea strain (including Rotstop) and to ca. 20 000 spores/ml for Trichoderma. The number ofrepetitions was 12 (three repetitions in each spruce stem). After treatment, the stem pieces (spatially randomised) were left standing in a humid place in the forest. The weather during the treatment was sunny and dry, and six following days also were rainless. The stem pieces and the soil around them were sprayed with clean water from time ta time. The temperature during this period varied between +2 and + 13°C. After three weeks, at the end of October, the stem pieces were removed from forest and placed for five weeks in a controlled greenhouse with humidity of 80% and temperature of 25°C. The stem pieces were sampled in December 1999. Two discs were cut; the first disc (3 cm thick) was discarded, the second disc (5 cm thick) was investigated on both sides as described above. The efficacy of the treatment was calculated by comparing the area occupied by Heterobasidion on treated and untreated halves of each disc surface. RESULTS Stump treatment experiment. The experimental spruce stand proved to be more infected by Heterobasidion than expected: the frequency of trees with butt rot was ca. 80%. Two species were present in about equal frequencies: H. annosum (Fr.) Bref) s.str. and H. parviporum NiemeUi & Korhonen. Altogether, 14 Heterobasidion isolates from diseased trees were identified, seven of them belonged to the former and seven to the latter species. Because only sound stumps could be used for the stump treatment experiment, the number of stumps within each treatment remained low. Moreover, a part of the discs had to be discarded because of apparent Control ------------------------------------ 177 Heterobasidion infection coming from the roots. The fmal number of accepted stumps in different treatrnents was: control 19, Rotstop 21, urea 14, P. gigantea (94135) 13, P. gigantea (97099) 12, and Trichoderma (94268) 11. P. gigantea Trichoderma 97099 94268 P. gigantea 94135 Urea • Heterobasidion OP. gigantea LJJ Rotstop 100 90 80 ?ft 70 ë .Q 60rn Ul 'c: 500 Ci u 40 "0 0 0 30 ~ 20 L10 0 Control Figure 1. Wood co1onised by Heterobasidion spp. and P. gigantea in NOIway spruce stumps at the depth of ca. 6 cm from stump surface. The results were calculated as percentage of infected wood of total wood on disc surfaces (all discs counted together). At the depth of 6 cm from the stump surface, Heterobasidion had infected 18.6% of total wood in untreated control stumps (Fig. 1); mean amount of infected wood per stump was 20.6%, and the number of infected stumps 52.6%. Rotstop treatrnent proved relatively effective against Heterobasidion infection but the treatrnents with urea, P. gigantea 94135 and Trichoderma 94268 rather favoured than reduced the infection, a1though the result may not be statistically significant. The efficacy of different treatrnents against Heterobasidion infection was as follows (calculated in three different ways): Table 1. Efficacy of different treatrnents against Heterobasidion measured in percentage of total wood, average of infected wood and number of infected stumps. Treatment Rotstop P. gigantea 97099 Trichoderma 94268 Urea, 30% P. gigantea 94135 Infected wood of total wood Efficacy 90.6% 35.5% 16.3% -173.2% -208.2% Mean infected wood per stump Efficacy 85.4% 38,9% -22.5% -130.6% -162.7% Number of infected stumps Efficacy 70.0% -10.8% -3.6% -35.7% -46.2% Experiment in stem pieces. In this experiment, the spore infection by Heterobasidion was relatively low; only 4-5% of the wood in untreated control halves of the stem pieces was colonised by Heterobasidion (at the depths of 3 and 8 cm from the surface). The colonies were generally small in size. The growth of P. gigantea was even weaker; even in treated halves of the stem pieces the mean colonisation was only 3.5%. The efficacy of different treatrnents against Heterobasidion infection is shown in Fig. 2. In striking contrast to the stump treatrnent experiment, the treatrnent with urea proved to be most effective (efficacy 90-100%). Trichoderma and Rotstop were distinctly weaker (70-80%). P. gigantea isolates from Italy were even weaker (0-65%). 178 Control 100 ~ 900 ë a 80 :0 "iii 70CIl .0e 60 2 - <.l 20CIl <.l lE 10w 0 .3cm 08cm ~ IJ 1 Urea Trichod. Rotstop P. gig. P. gig. P. gig. P. gig. 94266 97092 95051 97097 97104 P. gig. P. gig. 97099 95042 Figure 2. Efficacy of different treatments against Heterobasidion infection in stem pieces of Norway spruce. Two depths, 3 and 8 cm, from the treated surface were investigated. The efficacy of P. gigantea strain 95042 at the depth of 8 cm was slightly negative. DISCUSSION The amount of precipitation and the water content of wood apparently had a major impact on the results obtained in these two experiments. It was raining when the stump treatrnent experiment was made in spring 1998, and the rain continued during the following days. This is probably the main reason why the urea treatrnent failed totally in this experiment. Although urea generally works well in stump treatments, occasiona1 failing of urea has been reported also earlier (e.g. Schonhar 1977, Pagony 1980). Rotstop was quite effective in this experiment, and in stumps treated with Rotstop P. gigantea colonised more than 50% of the wood. The two P. gigantea isolates originating from the Alps as weil as Trichoderma sp. did not work satisfactorily. In the second experiment, the treatments were made to stem pieces of spruce. To obtain a natural spore infection by Heterobasidion, the pieces were first kept in the forest for three weeks, but later the incubation was carried out in the greenhouse with high relative humidity. At the end of the experiment the wood had a higher water content than is usuaI. In these wet pieces of spruce wood, the efficacy of urea against Heterobasidion was very good. The Trichoderma strain also proved to be relatively effective; it was as effective as Rotstop. The Italian isolates of P. gigantea were weaker. The high moisture content of wood apparently inhibited the growth of all the P. gigantea strains used in the experiment. Unexpectedly, the growth of Heterobasidion was not inhibited in the same degree, although in general P. gigantea and the Heterobasidion species seem to have very similar requirements as regards moisture content of spruce wood (K. Korhonen, unpublished results). It is also possible that the relatively low temperature at the beginning of the experiment slowed down the colonisation by P. gigantea. ln conclusion, continuous rain during and after the stump treatrnent may have a great impact on the efficacy in case urea is used as a protecting agent. The rain at the time of application does not seem to affect so much on the efficacy of treatment with P. gigantea, but a long-lasting high water content of wood inhibits the growth of this fungus, and may decrease its efficacy against Heterobasidion. The results also indicate big differences in the capacity of different P. gigantea strains to control spore infection of Heterobasidion in Norway spruce stumps. Control 179 ACKNOWLEDGEMENTS The authors wish to thank the forest owners, family Andreatta, for their collaborative willingness, the Forest Service of the Provincia Autonoma di Trento for useful assistance and logistic help, as weil as Aldo Bianchini, Marco Stefanini and Sanna Kannelsuo for their precious help. REFERENCES Holdenrieder, O.; Greig, B.J.W. 1998. Biological methods of control. In: Woodward, S.; Stenlid, J.; Kmjalainen, R.; Hütterrnann, A. (eds.) Heterobasidion annosum: Bio10gy Ecology Impact and Control. CAB International. Pp. 235-258. Korhonen, K.; Lipponen, K.; Bendz, M.; Johansson, M.; Ryen, 1.; Venn, K.; Seiskari, P.; Niemi, M. 1994. Control of Heterobasidion annosum by stump treatment with "Rotstop", a new commercial formulation of Phlebiopsis gigantea. In: Johansson, M. and Stenlid, J. (eds.) Proc. 8th Int. Conf. Root and Butt Rots. Swedish University of Agricultural Sciences, Uppsala. Pp. 675-685. Pagony, H. 1980. Butt rot: a dangerous pest of Hungarian Scots pine stands [Fomes annosus (Fr.) Cooke]. Erdészeti Kutatasok (Proc. Hungarian For. Res. Inst.) 73(2): 13-23. Pratt, J.E. 1994. Sorne experiments with borates and with urea to control stump infection by H annosum in Britain. In: Johansson, M. and Stenlid, J. (eds.) Proc. 8th Int. Conf. Root and Butt Rots. Swedish University of Agricultural Sciences, Uppsala. Pp. 662-667. Pratt, J.E. 1996. Borates for stump protection: a literature review. Technical Paper - Forestry Commission. Edinburgh, UK: No. 15: 1-19. Pratt, J.E.; Niemi, M.; Sierota, Z.H. 2000. Comparison of three products based on Phlebiopsis gigantea for the control ofHeterobasidion annosum in Europe. Biocontrol Science & Technology 10: 467-477. Rishbeth, J. 1963. Stump protection against Fomes annosus III. Inoculation with Pheniophora gigantea. Ann. Appl. Biol. 52: 63-77. Schônhar, S. 1977. Erprobung von Chemikalien zur Verhütung einer Infektion frischer Fichtenstôcke durch Fomes annosus. Allg. Forst.- u. Jagdzeitung 148: 181-182. Soutrenon, A.; Lévy, A.; Legrand, P.; Lung-Escarrnant, B.; Sylvestre-Guinot, G. 2000. Efficacité de trois traitements de souches contre le Fomes (Heterobasidion annosum) sur pin maritime. Rev. For. Fr. 52: 39­ 48. Thor, M.; Stenlid, J. 1998. Heterobasidion annosum infection following mechanized first thinning and stump treatment in Picea abies. In: Delatour, C. et al. (eds.) Root and Butt Rots of Forest Trees. 9th International Conference on Root and Butt Rots. Institut National de la Recherche Agronomique, Paris. Pp. 397-408. 180 Control MICROBES INHABITING PICEA WOUNDS AND THEIR ANTAGONISM TO HAEMATOSTEREUM SANGUINOLENTUM M.T. Dumas* and J.A. McLaughlin** * Canadian Forest Service, Great Lakes Forestry Centre, 1219 Queen St. E., Sault Ste. Marie, ON, P6A 2E5, Canada ** Ontario Ministry ofNatural Resources, Ontario Forest Research Institute, 1235 Queen St. E., Sault Ste. Marie, ON, P6A 2E5, Canada SUMMARY The Boreal Mixedwood Forest is the most productive forest region in Ontario. Prescriptions to maintain advanced growth in these stands call for partial cutting, which involves the use of machinery in confined spaces. Wounding of residual trees is unavoidable and often results in infection by decay fungi. Wounds of black and white spruce inflicted when the average air temperature was greater than O°C (pre-freeze up) or less than O°C (post-freeze up) were evaluated periodically over 8 months to determine if the ambient temperature at the time of wounding influenced the diversity and establishment of pioneering mycoflora on the wound surface. The size of wounds varied from 0.6 cm2 to 850.8 cm2 with the highest proportions of wounds being on the root, stem and butt respectively. Pre-freeze up wounds had the highest level of biodiversity of microorganisms and Pseudomonas species comprised the greatest proportion of the populations on these wounds. Yeast species (most commonly Crytococcus albius var albius) dominated the surface of wounds inflicted during the post-freeze up period. Isolates of Pseudomonas and Trichoderma species showed the greatest inhibition against the linear growth of H sanguinolentum in in vitro studies whereas none of the yeasts demonstrated antagonistic properties. Over the duration of the study decay fungi were isolated more often from post-freeze up wounds than from pre-freeze up wounds. Poria rivulosa was the most commonly isolated decay organism. INTRODUCTION Partial cutting operations involves the maneuvering of large pieces of equipment in somewhat confined spaces. These operations ultimately result in unavoidable wounds to the residual trees, initiating the complex mechanism often resulting in decay. Wounding of coniferous wood enhances the invasion of bacteria (Kallio 1973, 1974; Roll-Hansen and Roll-Hansen 1980; Hallaksela 1984) and fungi (Isomaki and Kallio 1974; Roll­ Hansen and Roll-Hansen 1980; Vasiliauskas et al. 1996). Affected trees are prone to blowdown and breakage at the wound site and if they survive to rotation age their end values are diminished due to the staining and decay in the wood. The ability of a microbe to function and sustain itself depends on several factors such as suitable substrates, water, oxygen, other gases and temperature (Rayner and Boddy 1988). When trees are wounded a series of events occur that will determine the mycofloral component of the wound face and the succession of conditions that ultimately result in stain and decay. Shigo (1979, 1984) recognized three succession stages. The first includes the physiological processes of the host response to wounding. The second stage of succession occurs when pioneer invaders, primarily bacteria, yeasts and deuteromycetes invade the area and overcome the physiological and chemical barriers of the host. These pioneering microbes are capable of modifying phenolic substances (Shigo and Sharon 1968, 1970; Shortle and Cowling 1978). The final step in the microbial succession is invasion by decay fungi. This is the mode of infection for most decay fungi but sorne, such as Haematostereum sanguinolentum, do not need the wound substrate to be preconditioned by other microbes (Davidson and Etheridge 1963). Also, the season of wounding appears to influence the likelihood of infection by sorne decay fungi. In earlier studies (Kallio and Hallaksela 1979; Beitzen-Heineke and Dimitri 1981; Hallaksela 1984), H sanguinolentum was observed to infect more frequently in the cooler months of the year. Control 181 The purposes of this study were to investigate the diversity of pioneering microbes co\onizing wounds of spruce, the influence of temperature on microbial diversity, and the abi\ities of the various microbes to inhibit H. sanguinolentum. MATERIAL AND METRODS The study areas were located approximately 120 km northeast of Thunder Bay, Ontario, on the management limits of Bowater Inc. within the Black Sturgeon Forest (49°11.4' N, 88°42.5' W). Root, butt (i.e., wounds in the area of 0 to 20 cm above the soil line) and stem wounds (i.e., wounds higher than 20 cm on the stem) on residual black spruce (Picea mariana) and white spruce (Picea glauca) trees were selected for sampling. Manual or feller-buncher cutting and skidding during partial cutting operations caused the wounds. Seventy-six wounded trees were sampled in early October when the mean aerial temperature was higher than O°C and are referred to as pre freeze-up wounds. Twenty-three wounded trees were sampled in late October/early November when the mean aerial temperature was below O°C and are referred to as post freeze-up wounds. In total, 143 pre freeze-up wounds and 48 post freeze-up wounds on 76 and 23 trees respectively were sampled. The size of each wound was measured by photographing it in relation to a 76-mm x 127-mm index cardo The picture was projected onto a screen and traced with a planimetre. The area of the wound was determined through a magnification factor based on the area of the index cardo The first samples were taken 1 week after wounding. The wound surfaces were wiped with sterile cotton swabs moistened with a sterile 20% glycerol solution. The swabs were then placed in sterile 4-mL plastic vials containing 0.5 mL of sterile 20% glycerol and transported to the laboratory in coolers. They were stored at 4°C and processed within 2 days. The two week, 6-7 week, 5 month and 7-8 month samples were thin wood chips approximately 1.5 cm x 1 cm excised from the wound face with a scalpel sterilized with 70% ETOH. The wood chips were placed in plastic vials containing 0.5 mL of sterile 20% glycerol at 4°C. The 18-19 month samples were wood chips approximately 2 cm x 2 cm x 0.5 cm thick eut from the wound face with a chisel cleaned with 70% ETOH. Each sample was placed in a plastic bag for transporting to the laboratory. A selective medium was prepared from both black spruce and white spruce wood chips. Black and white spruce stem sections, free of defects and approximately 1.5 m long and 12 cm in diameter, were debarked and chipped on a woodworking jointer. The chips were collected and kept frozen until needed. The moisture content of the chips was determined by drying 100 g fresh weight of wood chips in an oven at 80°C for 18 br and subsequently placed over CaS04 in a dessicator until a constant weight was achieved. Wood chips, equivalent to a dry weight of 2000 g, were put into a 6-L flask. Four litres of sterile distilled water, brought directly from the autoclave, were poured onto the wood chips and the infusion was left to steep for 16 hours. The infusion was filtered through Whatman 1 filter paper under vacuum and then through 0.45-llm Gelman filter. Eighty grams of sucrose were added to the extract and the volume brought up to 4 L using a volumetric flask and stored at 4°C until needed. Fifteen grams of agar were added to each litre of wood extract-sucrose solution and autoclaved for 15 minutes at 110°C. Twenty mL were dispensed into lOO-mm x 15-mm sterile plastic petri dishes. The other isolation media were 1110 strength TSA (4 g CASO + 15 g agarlL) and PDA (Potato Dextrose Agar). To aid in the isolation of decay fungi, MEA (3% malt extract + 1.5% agar) was supplemented with 5 ppm MBCP (methylbenzamidazole carbamate phosphate). Isolations were made by first adding 1 mL of sterile 0.1 M MgS04 to the contents of each swab sample vial. The vials were then agitated and the suspension streaked onto three defined media and the spruce extract medium, three replicates per medium. The undiluted liquid contents of the vials containing wood chips were also streaked onto li10 TSA to determine if microbes were leached from the wood sample. Colony forming units were assessed only on the swabbed samples. Bacteria and actinomycete isolates were transferred to nutrient agar (NA) and the fungal cultures to PDA. The populations isolated on these media were not tallied in subsequent samples because of their low numbers. Isolation attempts for a broad range of fungi were made after 1 and 2 weeks, 6-7 weeks,5 months, and 7-8 months. The wood chip samples were eut into two pieces. One halfwas placed on MEA 182 Control + 10 ppm MBCP to favour isolation ofbasidiomycetes and the other half on PDA for moulds. Wood chip samples collected 18-19 months after the wounds occurred were plated only on MEA + 10 ppm MBCP. The plates were wTapped with Parafilm7 and incubated in the dark at 25°C. Microbes isolated on synthetic media were then cultured on the spruce extract medium. Only those that were capable of growing on the spruce extract were selected for identification. Bacteria were first identified as either gram-positive or gram-negative according to the method described by Suslow et al. (1982) and their morphological characteristics on 1/10 TSA and nutrient agar. Further identification was accomplished using the MIDr® software system. Fungi capable of growing on the spruce extract medium were transferred to PDA and incubated at 25°C on a 16-hr light/8 hr dark cycle to encourage conidiogenesis. Fungal colonies that were morphologically and microscopically similar were pooled for identification by reference to standard identification keys. Yeast isolates were identified by the MIDI® software system. Basidiomycetes were identified according to Stalpers (1978). The Haematostereum sanguinolentum isolate used in the study was obtained from a spore cast from fruiting bodies on black spruce originating from the study site. Fungi, bacteria and yeasts were paired with H sanguinolentum to detect antagonism. The fungi were tested on 3% malt agar while the bacteria and yeast were tested on the malt extract-yeast extract-peptone medium of Rose et al. (1980). Five-mm-diameter mycelial plugs were cut from the margin of actively growing cultures and plated with H sanguinolentum 55 mm apart on MEA (3%) in 90-mm petri plates. The plates were incubated in the dark at 25°C for 2 weeks. The interactions between the potential antagonists and the H sanguinolentum were observed and recorded after 2 and 4 weeks. Interactions were classified as formation of an unoccupied aversion zone between the pairs (inhibition), meeting at the confluence zone with or without hyphal massing and/or minimal intermingling (collision), and an interaction where either the H sanguinolentum or the other isolate overgrew the other (overgrown). Isolates that inhibited the growth of H sanguinolentum and those that overgrew it were selected for further tests. In this second stage H sanguinolentum was grown for 5 days under conditions described above before the antagonists were introduced to the plates. The pairs were incubated as described above and the ability to overgrow or inhibit the growth of H sanguinolentum was measured after the control colonies reached the edge of the plate. Ali bioassays had three replicates and were repeated three times. RESULTS After one week most of the soil and organic material that had been deposited on the wounds from the skidder tires or logs had been washed by rain from the wound surface. Therefore, the microbes living on the wound surface were presumably obtaining sustenance from the tree and not the soil. The number and ratio of bacteria, actinomycetes, and fungi on one-week-old wounds varied between pre and post freeze-up wounds, wound locations, and media. Random samples of the different classes of microbes isolated from the two spruce species were not significantly different, indicating no relationship between tree species and microbe. Therefore, data sets were combined and are reported as a total population study. The highest numbers of wounds occurred on the stem (94), roots (64) and butt (33), respectively. The largest wounds were found on roots (mean size 137 cm2 , ranging from 7.5 cm2 - 851 cm2 ), fol1owed by stem wounds (mean size 69 cm2 , ranging from 1 cm2 - 785 cm2 ) and butt wounds (mean size 44 cm2 , ranging from 5.3 cm2 - 162 cm2 ). There was no relationship between wound size and the time they were inflicted. More bacteria than fungi were isolated from the pre freeze-up wounds than from the post- freeze up wounds, while fungi were more plentiful than bacteria on the post freeze-up wounds. Bacteria populations were greater on root wounds than butt wounds, which in tum had higher populations than stem wounds (Table 1). Yeasts comprised a large proportion of the fungi from the post freeze-up wounds. Control 183 Identifications using the p\asma1emma fatty acid analysis indicated that there was a comp/ex of bacteria capable of living on the surface of one-week-old wounds. Species of Bacillus and Pseudomonas comprised the highest proportion of the microflora. A much greater diversity of bacterial species was isolated from the pre freeze-up wounds than from the post freeze-up wounds. Bacillus sphaeficus, isolated from the pre freeze-up wounds, demonstrated the highest antagonism to H. sanguinolentum (Table 2). Yeasts numbers and diversity were highest on wounds incurred during the post freeze-up period. Cryptococcus albidus var. albidus was the most frequently isolated yeast (Table 3). The inhibition test demonstrated that ail yeasts were ineffective at inhibiting the growth ofH. sanguinolentum (data not shown). Filamentous fungi were more prevalent than yeasts on the pre freeze-up wounds. Sampling period (i.e., pre or post freeze-up) had no significant effect on the diversity of identified species. However, there were also many fungi that failed to form fruiting structures and as such could not be identified. There were more than twice as many fungal isolations from the pre freeze-up wounds than from the post freeze-up wounds. Trichoderma polysporum was frequently isolated from pre freeze-up wounds (Table 4). Inhibition tests demonstrated that only a few of the fungi isolated were capable of inhibiting H. sanguinolentum. The Penicillium species were the most inhibitory, producing distinct inhibition zones between the cultures whereas the Trichoderma species overgrew the H. sanguinolentum colony and eventually digested the mycelia (Table 5). The highest incidence of isolation of decay fungi was from the samples taken 7 to 8 months following wounding. Poria rivulosa (Berk. & Curt.) Cooke was the most commonly isolated decay fungus. A greater diversity of decay fungi were isolated from wounds incurred during post freeze-up (Il spp.) than during pre freeze-up (8 spp.) (Table 6). DISCUSSION Levy (1975) observed that wood contained appropriate nutrients for primary wood invasion by bacteria. The present study demonstrated that the micro environment at the surface of a wound would determine the initial microflora. AIso, by waiting 1 week after wounding before attempting to isolate microbes we believe that most of the isolated microbes were surviving on nutrients provided solely by the tree. Selection of microbes capable of surviving on the wound surface was further refined through the use of the spruce extract medium. He number of bacteria colonies isolated on synthetic medium were very high but were substantially reduced when grown on the wood extract medium. Kallio's (1973) medium was not used because it was thought that the yeast extract and trypton in that medium might select for bacteria incapable of living only on the nutrients available on the wound surface. Etheridge (1969) found a prevalence of H. sanguinolentum in wounds inflicted at lower temperatures. He suggested that the dominance of H. sanguinolentum of fresh wound surfaces was due to the properties of the substrate, especially at low temperatures, but the reasons why were not thoroughly studied. However, the reason could be that different microbial populations become established on wounds inflicted at different times of year, under different growing conditions, especially temperature. In the pre freeze-up period, bacteria and fungi, which were more active in inhibiting the growth of H. sanguinolentum, dominated the wounds whereas yeasts primarily colonized the post freeze-up wounds. Although yeasts have been shown to possess anti-fungal properties, (Payne et al. 2000; Janisiewicz and Bors 1995; Burr et al. 1995; Walker et al. 1995) the isolates obtained in this study were not antagonistic towards H. sanguinolentum. The diversity of bacteria isolated in this study was much greater than previously reported, probably because previous research concentrated on the bacteria that are capable of inhabiting the inner wood. Nevertheless, in this study Pseudomonas and Bacillus spp. were the bacteria species most commonly isolated from the wound surface. This is similar to the findings of Hallaksela et al. (1991) and Hallaksela and SaLkinoja­ Salonen (1992) who found that these genera were the major types which inhabited the inner portions of the wood 184 Control of Norway spruce. Przybyl and Zlobiilska-Podejma (2000) isolated Pseudomonas species most commonly from discoloured wood of Betula pendula Roth. These species are present in the soil (Hoit et al. 1994) and could enter the inner wood through minute wounds in the mots or be deposited on the wound via splashing from the machinery tires. Sorne of the isolated species may be endophytes that became exposed when the bark was removed. Kallio (1974) isolated bacteria that were antagonistic to H sanguinolentum and other decay fungi but did not identify them. The antagonism of these bacteria to H sanguinolentum is similar to results of tests against Armillaria ostoyae in an earlier study (Dumas 1992). The mode of the antagonistic action is through the production of antifungal compounds and siderophores (Dumas and Strunz 1997), compounds that cou1d possibly be produced on these sites. The diversity of fungal species was high in the initial isolations but as the wounds aged it decreased. The initial large populations and diversity could be similar to the situation with the prokaryotes. The initial abundance and diversity of microbes would include many that can metabolize only simple sugars. As the wound ages there would be a graduaI succession to organisms that can utilize more complex organic compounds for nutrition, similar to soil microbial adaptation as described by Panikov (1999). In the present study deuteromycetes comprised the bulk of the populations in the wounds of black and white spruce at aIl sampling periods. Roll­ Hansen and Roll-Hansen (1980b) found a higher proportion of ascomycetes than deuteromycetes in wounds on P. abies in Norway whereas Michalopoulos-Skarmoutous (1987) observed a greater number of deuteromycetes in the same tree species in Greece. The variations could be due to differences in geographical locations and wound microsites. A large proportion of the fungi isolated in the current study did not inhibit the growth of H sanguinolentum. The Trichoderma species inhibit by acting as a mycoparasite and through the release of anti­ fungal metabolites (Dumas and Strunz 1997). Species of Penicillium were very antagonistic to H. sanguinolentum and the production of a clear inhibition zone indicated the production of antibiotic-like compounds. This is a very common method of antagonism utilized by different species of Penicillium (Land and HuIt 1987). Tt is interesting to note that Beauveria bassiana, an entomopathogen, (Samson et al. 1988) suppressed the growth of H sanguinolentum. Stem decay fungi were isolated much later after wounding occurred and were found more commonly on post freeze-up wounds. Roll-Hansen and Roll-Hansen (1980a) observed a higher recovery of H. sanguinolentum from wounds inflicted in May and September than those caused during July and December, but overall the occurrences of hymenomycetes was not influenced by the wounding period. The discrepancy in results is most likely influenced by the differences between inoculum potential of the decay fungi and tree species. We were unable to continue to monitor the changing diversity of the microbial communities within the wood, first because many of the sampled trees died (primarily due to Armillaria and windthrow) within 2-5 years after the partial cut, and finally because the experimental site was destroyed by wildfire in the spring of 1999. Control 185 00 0\ ~ Q =­.,::.. Table 1. Mean Ilumber of colony lorming units or hacteria, actinomycetcs, and fungi isolated on I! 10 TSA, spmce extract agar and PDA l'rom l-week-old wounds on roots. butls and stem of spmce lI1tlicled pre freeze-up and post freet.e-up. 1/10 TSA Spruce extract agar PDA Wound Bacteria Actinomy.,·cte Fungi Bacteria Actinomycete Fungi Bacteria Actinonlycete Fungi pre l'reeze-up rool (54) a\'g 68.6 13.5 3.3 46.7 4.4 8.6 22.6 JO.7 5.4 b:fratiol ' 24.8: 1 5.9:1 6.2:1 bull (23) avg 27.3 4.6 5.8 [6.5 0.1 6.4 14.5 2.7 6.5 b:f ratio 5.5:1 2.6:1 26.5:1 stcm (66) nvg 12.4 0.2 3.9 S. [ 0.1 4.8 R.5 1.0 4.7 b:f ralio 3.2:1 lI:[ 20.\ :1 post l'reezc-up root (10) avg 20.4 25.8 22.4 4.5 0.0 94.2 30.5 8.0 74.7 b:fratio 2.1: 1 0.04:1 0.5:1 bull (] 0) avg 7.8 1.0 26.R 1.3 0.0 54.2 5.5 1.2 45.3 b:fratio 0.33:1 0.02:1 0.15: [ stem (28) avg 6.7 0.5 10.1 7.2 0.0 49.8 5.2 0.2 30.8 b:f ratio 0.7:] 0.14:1 0.17: 1 .. Number of\~~~d-;;:- b Ratio ofbacteriallo l'ungal colony l'orming unils. n Q = [ Table 2. Diversity of bacteria isolated fl'om wound surfaces of spnlce at pre n'eeze-up and post freeze-lip and their abilitics lo inhibit the 1inear growth of H. saIIgu illolelltll/II. 00 --l Wound location Root Bult Pre freeze-up Bacteria Arthrohaeter il/ieis (1)"' Bacillus longisporus (2) Baeil/us mycoides subf,'foup A (17) Baeillus pasleurii (3) Baeil/us sphaejicus (3) Baeil/us thuringiensis (8) Cedecca davisae (l) Cellulomol7as 1I11'hata (1) CurlOhacterium flaceumfaciens (\) Klebsiella pneumonie, ozaenae (1) Mierocoecus kristinae (4) Norcardia globerula (1) Noreardia asteroides subgroup A (3) Paellohacillus polymxa (1) Pselldomollas chforoaphis (7) Pseudomasfluorescens Biotype C (Il) Pseudorl/onasjluoreseens Biotype G (1) Pseudofllonas plltida Biotype A (1) Pseudoll7onas putida Biotype 13 (5) Pseudol11onas chloroaphis (7) Pseudomonas marginalis ( 1) PseudOll1onas savastal70f pvfrw;inus (1) Rathayihacter rathayi (l) Rhodococcus/àsicans subgroup 13 (2) R/wdococcus equi subgroup B (1) Serratia phymulhica (1) Stenotrophomonas maltophi/ia (2) Baeil/us eirel/fans (1 ) Bacilfus pastcurii (l) Boeil/us thuringiensis (4) Panoœa aggfotlleralls (1) % inhibition (Range) 45.8 4.1-8.4 36.7-68.5 o 81.0-86.6 58.3-68.3 35.9 2.9 2.5 o 5.5-41.7 65.9 53.3-64.7 2.1 62.2-73.3 55.8-73.6 69.6 67.1 68.5-79.9 62.2-73.3 63.3 68.3 o 62.2-70.8 24.7 o 43.1-49.8 3.3 69.6 63.0-74.5 4.6 Post freeze-up Bacteria Snell/lis mycoides subgroup A (1) Nurcarc!ia restrÎctala (1 ) Pseudol/lonas chlororaphis 0) Pseudoflto/1as/l/lOresc'ens Biotype C (1) Pseudomonasjluorescens Biotype F Cl) Pseudomonas putida Biotype A (l) Pseudomonos putida Biotype B (3) Stenotrophomonas mallOphilia (l) Rahnel/a aquatilis (1) Bi/cil/us l1/ycoides slIbgroup A (2) Psellc!OI1/UIlOS plltida Biotype A (1 ) % inhibition (Range) 42.2 64.8 70.8-71.1 69.6 64.5 67.3 64.5-69.6 36.9 o 39.4-42.8 24.\ 00 00 ("') Q =­.,e.. Pseudoll1onas chlororaphis (1) Pseudoll1onasfl/lorescens Biotype C (8) Pseudoll1onas putida Biotype B (2) Rhodoeoceusfasieans subgroup B (1) Staphylococeus saprophytieus (1 ) Stem Baeillus eereus (3) Baeillus myeoides subgroup A (3) Baeil/us pasteurii (7) Baeillus thuringiensis (4) Paenihacil/us macerans subgroup S (1) Paenibacil/us pabuli (7) Paenibacillus polymxa (1) Pseudomonas chlororaphis (7) Pseudomonas jlllOreseens Biotype C (1) Pseudofllonas putida Biotype B (3) Rabnella aquatilis (1) Unknown (1) 'Number of strains isolated and lested. 49.8 55.7-76.1 69.6-68.4 74.1 o 43.0-64.5 36.7-51.9 43.\.-67.1 62.0-68.3 o 73.9 o 60.7-78.5 50.4 65.9-79.7 o o Baeil/us myeoides subgroup A (1 ) Baeillus pasteurii (1) Pseudomonas ehlororaphsis (1) 53.2 37.8 69.6 Table 3. Frequency of occurrence and diversity ofyeasts isolated from spruce wound surfaces at pre freeze-up and post freeze-up periods. Pre freeze-up Cryptococcus albidus var. albidus (4)3 Candida cacaoi (1) Candida zeylanoides (l) Cryptococcus neoformans subgroup B (2) Rhodotorula rubra (l) Sporobolymces salmonicolor (1 ) No Match (3) a number of occurrences Control Post freeze-up Candida auringiensis (1) Candida blankii (1) Candida cacaoi (3) Candida deserticola (1) Candida guilliermondii (1) Candida hydrocarbofumarica (1) Candida norvegica (3) Candida philyla (1 ) Candida silvicultrix (1 ) Candida sake (1 ) Candida zeylanoides (16) Cryptococcus albidus var. albidus (24) Cryptococcus neoformans subgroup A (2) Cryptococcus neoformans subgroup B (3) Cryptococcus terreus (6) Geotrichum candida (1) Geotrichum candidum (8) Hansenula anomala (1) Rhodotorula minuta subgroup A (2) Rhodotorula minuta subgroup B (1) Rhodotorula rubra (3) Sporobolomyces salmonicolor (4) Trichosporon beigelii subgroup A (1) No Match (36) 189 Table 4. Deuteromycetes isolated from wounds on spruces during pre freeze-up and post freeze-up periods. Time of wounding Pre freeze-up Pre freeze-up Species Root Butt Stem Root Butt Stem Alternaria spp. 6a 4 18 1 2 5 Aspergillus versicolor b 1 Aureobasidium pullans 3 4 8 1 2 3 Beauveria bassiana 1 1 Capitorostrum asteridiellae Cephalosporium sp. Chalara cylindrica Chlaunopycnis alba 1 Cladosporium spp. 13 6 20 2 14 Cladosporium herbarum 1 1 1 Cylindrocarpon spp. 1 4 1 Cytospora spp. 4 2 8 1 6 Epicoccum nigrum 8 7 17 2 3 2 Fusarium spp. 3 Il 2 3 Helicoma dennisii 1 Hormonema type 13 11 24 4 7 33 Lichenoconium spp. 3 Mycelia sterilia 17 20 67 6 Il 39 Oidiodendron griseum 3 1 Paecilomyces farinosus 10 6 9 3 Paecilomyces spp. 10 6 5 Penicillium spp. 42 19 16 9 6 13 Phaeococcus spp. 3 8 2 Phialophora spp. 2 1 1 Phoma spp. 9 4 14 3 5 17 Phoma chrysanthemicola 1 1 Phomopsis spp. 1 7 3 2 Sporotrichum sp. 6 Taeniolina centaurii 1 2 7 2 6 Trichoderma harzianum 3 1 1 1 Trichoderma polysporum 17 4 3 2 3 Trullula sp. 1 Umbelopsis versiformis 1 Valsa friesii 1 2 1 Zythiostroma pinastri 8 5 1 2 Total 172 103 259 43 45 159 a Number of isolations b Not detected 190 Control Table 5. Fungi inhibitory to the growth of Haematostereum sanguinolentum, isolated from wounds on black and white spruce. Wound location Root Butt Stem Pre freeze-up Species Penicillium glabrum Penicillium lividum Trichoderma polysporum Penicillium franisosus Chlaunopncinis alba Penicillium spinolosum Penicillium soppii Penicillium steckii Trichoderma harzianum Beaveria bassiana Unknown 4 Unknown 5 % inhibition 68.5 76.7 100.0 59.3 72.0 71.1 65.8 63.3 100.0 72.1 45.1 71.5 Post freeze-up Species Unknown 3 Beauveria bassiana Trichoderma polysporum Trichoderma harzianum Aspergillus versicolor Penicillium brevicompactum Zythrostroma pinastrii Unknown 2 % inhibition 54.7 61.7 100.0 100.0 66.9 66.3 75.8 61.9 Table 6. Decay fungi isolated from wounds of spruce caused during pre freeze-up and post freeze-up. Decay fungus Cryptoporus volvatus Gloeophyllum protractum Haemostereum sanguinolentum Hymenochaetae tabacina Odontia corrugata Phelbia subserialis Poria rivulosa Poria vaillantii Sistotrema brinkmannii Strecchericium ochraceum UnknownA Unknown B Unknown C Unknown D Pre freeze-up wounds (%) o o o 1.9 1.9 1.9 3.8 0.97 2.9 o o o 0.75 0.75 REFERENCES Post freeze-up wounds (%) 2 2 2 o 6 6 20 4 4 2 2 2 o o Beitzen-Heineke, 1.; Dimitri, L. 1981. Rückeschaden: Entsteühung und die Môglichkeiten ihrer Verhütung. Allg. Forst Zeitschr. 32: 278-280. Burr, T.J.; Matteson, M.C.; Smith, C.A.; Corral-Garcia, M.R.; Huang, T -CO 1996. Effectiveness of bacteria and yeasts from apple orchards as biological control agents of apple scab. Biol. Control 6: 151-157. Davidson, A.G.; Etheridge, D.E. 1963. Infection of balsam fir, Abies balsamea (L.) Mill., by Stereum sanguinolentum (Alb. and Schw. Ex Fr.) Fr. Cano J. Bot. 41: 759-765. Dumas, M.T. 1992. Inhibition of Armillaria by bacteria isolated from soils of the boreal mixedwood forest of Ontario. Eu. J. For. Path. 22: 11-18. Control 191 Dumas, M.T.; Strunz. G.M. 1997. Modes of action of antagonistic microbes to Heterobasidion annosum and Armillaria ostoyae. Proc. In Proc. 9th international Conference on Root and Butt rots, Carcans­ Maubuisson, France: 448. Etheridge, D.E. 1969. Factors affecting infection ofba1sam fir (Abies balsamea ) by Stereum sanguinolentum in Quebec. Cano J. Bot. 47: 457-479. Hallaksela, A-M. 1984. Bacteria and their effect on the microflora in wounds of living Norway spruce (Picea abies). Comm. Inst. For. Feno. 121: 25 p. Hallaksela, A-M.; Vaisanen, O.; Salkinoja-Salonen, M. 1991. Identification of Bacillus species isolated from Picea abies by physiological test, phage typing and fatty acid analysis. Scand. 1. For. Res. 6: 365-377. Hallaksela, A-M.; Salkinoja-Salonen, M. 1992. Bacteria inhabiting artificially inoculated xylem of Picea abies. Scand. 1. For. Res. 7: 165-175. Holt, 1.; Kreig, N.; Sneath, P.; Staley, S.; Williams, S. 1994. Bergey=s Manual of Systematic Bacteriology. The Williams & Wilkins Co., Baltimore, 787 p. Isomaki, A.; Kallio, T. 1974. Consequences of injury caused by timber harvesting machines on the growth and decay of spruce (Picea abies (L.) Karst.) Acta For. Feno. 136: 25 p. Janisiewicz, W.1.; Bors, B. 1995. Development of a microbial community of bacterial and yeast antagonists to control wound-invading postharvest pathogens of fruits. Appl. Environ. Microbiol. 61: 3261-3267. Kallio, T. 1973. Peniophora gigantea (Fr.) Massee and wounded spruce (Picea abies (L.) Karst.). Acta. For. Feno. 133: 28 p. Kallio, T. 1974. Bacteria isolated from injuries to growing spruce trees (Picea abies (L.) Karst.). Acta For. Feno. 137: Il p. Kallio, T.; Hallaksela, A-M. 1979. Biological control of Heterobasidion annosum (Fr.) Bref. In Finland. Eur. J. For. Path. 9: 298-308. Land, c.J.; Huit, K. 1987. Mycotoxin production by sorne wood-associated Penicillium spp. Lett. Appl. Microbiol: 4: 41-44. Levy, J.J. 1975. Bacteria associated with wood in ground contact. p. 64-73. In biological transformation of wood by microorganism. W. Liese Ed., Springer-Verlag, Berlin. Michalopoulos-Skarmoutsos, M. 1987. Occurrence of micro-organisms in the wood of Picea abies Karst. In Greece. Eur. J. For. Path. 17: 305-307. Panikov, N.S. 1999. Understanding and prediction of soil microbial community dynamics under global change. Appl. Soil Ecol. Il: 161-176. Payne, c.; Bruce, A.; Staines, H. 2000. Yeast and bacteria as biological control agents against fungal discolouration of Pinus sylvestris blocks in laboratory-based tests and the role of antifungal volatiles. Holzforschung 54: 563-569. Przybyl, K.; Zlobinska-Podejma, M. 2000. Effects of sorne bacteria (Pseudomonas spp. and Erwinia herbicola) on in vitro growth of Piptoporus betulinus. Forest Pathology 30: 321-328. Rayner, A.D.M.; Boddy L. 1988. Fungal Decomposition of Wood. John Wi1ey & Sons. Chichester, 541 p. Roll-Hansen, F.; Roll-Hansen, H. 1980a. Microorganisms which invade Picea abies in seasonal stem wounds I. General aspects. Hymenomycetes. Eur. J. For. Path. 10: 321-339. Roll-Hansen, F.; Roll-Hansen, H. 1980b. Microorganisms which invade Picea abies in seasonal stem wounds II. Ascomycetes, Fungi imperfecti, and bacteria. General discussion, Hymenomycetes included. Eur. J. For. Path. 10: 396-410. Rose, S.L.;Li, C-Y.; Hutchins, A.S. 1980. A streptomycete antagonist to Phellinus weirii, Fomes annosus and Phytopthora cinnamomi. Cano 1. Micrbiol. 26: 583-587. Samson, R.A.; Evans, RC.; Latgé, J-P. 1988. Atlas ofentomopathogenic fungi. Springer-Verlag, Berlin, 187 p. Shigo, A.L. 1984. Compartrnentalization: a conceptual framework for understanding how trees grow and defend themselves. Anou. Rev. Phytopathology 22: 189-214. Shigo, A.L.; Sharon, E.M. 1968. Discoloration and decay in hardwoods following inoculations with Hymenomycetes. Phytopathology 58: 1493-1498. Shigo, A.L.; Sharon, E.M. 1970. Mapping colurnns of discolored and decayed tissues in sugar maple, Acer saccharum. Phytopathology 60: 232-237. Shortle, W.C.; Cowling, E.B. 1978. Development of discoloration, decay and microorganisms following wounding of sweetgum and yellow-poplar trees. Phytpathology 68: 609-616. 192 Control Stalpers, J.A. 1978. Identification ofwood-inhabiting fungi in pure culture. Studies in Myco1ogy No. 16,248 p. Suslow, T.V.; Schroth, M.N.; Isaka, M. 1982. Application of a rapid method for gram differentiation of plant pathogenic and saprophytic bacteria without staining. Phytopathology 72: 917-918. Vasiliauskas, R.; Stenlid, J.; Johansson, M. 1996. Fungi in bark peeling wounds of Picea abies in central Sweden. Eur. J. For. Path. 26: 285-296. Walker, G.M.; McLeod, A.H.; Hodgson, V.J. 1995. Interactions between killer yeasts and pathogenic fungi. FEMS Micrbiol. Lett. 127: 213-222. Control ----------------------------------- 193 COSTS AND EFFECTS OF BIOLOGICAL CONTROL OF ROOT ROT IN POLAND Z.H. Sierota Forest Research Institute in Warsaw, Poland Bitwy Warszawskiej1920 R. No 3 PL 00-973 Warszawa sierotaz@ibles.waw.pl Root rot caused by Heterobasidion annosum (Fr.) Bref. has been one of the most significant pathological and economic problems in Polish forests for many years. Serious damage has been specifically observed in Scots pine and Norway spruce stands cultivated on soils formerly under agricultural use. After World War II, former agricultural lands, pastures, grasslands and wastelands were continually afforested mostly with Scots pine (Pinus sylvestris) , Norway spruce (Picea abies) , birch (Retula spp.), larch (Larix decidua) and aIder (Alnus spp.) Approximately 1 156 000 ha of former agricultural land and farm abandoned sites were afforested during the years 1947-1997 and 704 000 ha of such forests were state owned and managed by the govemment (State Forests). Afforestation increased forest coyer in Poland from 20.8% in 1946 to 28.0% in 1996. The frrst significant damage (13 700 ha) was recorded in cultures and plantations in the 1950's and 1960's. Within 30 years (by 1976) the progress of the disease was recorded in 122000 ha of stands older than 20 years. Within 50 years (by 2000), damage due to H. annosum in tree crop was found in stands of ail age classes covering an area of201 900 ha ofstate forests (Fig. 1). In 1997, sorne forest divisions reported the high incidence of the disease in affected stands (5-6% of the forested area). In sorne forest districts, H. annosum damage was found in more than 10% of the stands (up to 44.6% of the pine stands area was with the pathogen in roots oftrees) (Sierota 1995). In Poland, losses caused by H. annosum in rniddle-aged Scots pine stands established on former agricultural soils and farm abandoned land were as follows: stand density decreased from 0.9 to 0.6, CUITent annual increment decreased by 50-60%, standing volume decreased by 27.8-69.5 m3lha, depending on intensity of fellings (Rykowski and Sierota 1984, Sierota 1997c). Poland was the second country in the Europe (after UK), where as early as 1970 the Rishbeth's (1959, 1975) ideas to control the root pathogens were introduced and first practical experiments with P. gigantea on semi-economical scale were done (Sierota 1975). This fungus is a known natural component of coniferous forest ecosystems. In the Forest Research Institute in Warsaw, the original method of production and practical application of "Pg-IBL®" preparation with the competitor was developed in 1970s and 1980s and is still being improved (Rykowski and Sierota 1977, Pratt et al. 2000). The preparation "PgIBL®", consisting of viable mycelium of P. gigantea grown on sterilized beech (Fagus) sawdust, is used to prevent the colonization of stump surfaces by H annosum and the development of the pathogen in whole root system. P. gigantea in "Pg-IBL®" decomposes wood of pine roots rapidly - up to 52% of the dry mass of lateral roots in six months (Sierota 1997a, 1998), and in addition reproduces rapidly in the stand. The reduction in H. annosum spread throughout coniferous stands and a significant decrease in tree mortality resulting in an increase of crop production are the main economic benefits of using P. gigantea (Sierota 1998). In State Forest, artificial inoculation of stumps with P. gigantea in PgIBL®-like preparation in first-rotation Scots pine stands, established on former agricultural land, has been obligatory since 1984. By 1992, PgIBL was used on approximately 18.2 million freshly cut stumps; in 1992-1998 the preparation was used satisfactorily in an area of360 200 hectares ofthickets and stands over 20 years old, and in 1999, on 84000 hectares (Fig. 1). The efficiency of spring and fall treatrnents is even 100% (measured next year by presence of fruitbodies and mycelium in stump roots) as long as stumps are sufficiently handled (tapping, covering with litter, etc.). The latter 194 Control increases labour costs by approximately 10%; however, these costs are counterbalanced both by ecological and economical advantages. The success of the biological method of prevention and control of root rot in Scots pine stands in Poland (Sierota 1984, 1997b, 1998) results in: a decrease in primary infection risk from the stump surface side: a decrease in primary infection risk from the root side: greatly reduced production of H. annosum fruitbodies; rapid and effective decay of the root system; enrichment of the developing forest site on former agricultural land with saprotrophic Basidiomycetes (many strains of P. gigantea) which play an important role in the energy balance of the forest ecosystem. Using PgIBL®-like preparation during routine protection treatrnents of pine stands on former agricultural land lessens tree mortality by 43% and increase the production of thickwood volume by 19 m3lha, if compared with stands where protection treatrnents are not carried out. At the scale of PgIBL use amounting in Poland to about 55 000 hectares per year in last decade, avoiding the loss could be estimated at the level of about US$20 million yearly. Forest plantations on former agricultural soils are artificial ecosystems. Diseases and other harmful effects that occur in these plantations reflect natural adaptation. These adaptive changes, however, do not match the goals of traditional forest management and do not introduce advantageous effects from the anthropogenic point of view. The creation of stable and pest-resistant forests on former agricultural land is not possible without properly-directed management efforts at each and every stage of stand development. Biological control of root rot in threatened stands, particularly using P. gigantea and similar acting preparations, according to Rishbeth's ideas, plays still a fundamental role in this concept (Rykowski 1990, Sierota 1995). Area damaged by H. annosum and the PglBL treatment 250 Cl PglBL ea H.a. 200-cu oC o 150o o ~ ><-cu 100 Q)... oC( 50 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Year Figure 1. Area damaged by H. annosum and treated with P. gigantea formula PgIBL. Control 195 REFERENCES Pratt, J.E., Niemi, M., and Sierota, Z.H. 2000. Comparison of three products based on Phlebiopsis gigantea for the control ofHeterobasidion annosum in Europe. Biocontrol Sci. & Technol. 10: 467-477. Rishbeth, J. 1959. Stump protection against Fomes annosus. III. Inoculation with Peniophora gigantea. Ann. Appl. Biol. 52: 63-77. Rishbeth J. 1975. Stump inoculation: a biologicalcontrol of Fomes annosus. In: Biology and Control of Soil-bome Pathogens. G.W. Bruelh, ed. Americ. Phytopath. Soc., St. Paul Minnesota. pp: 158-162. Rykowski, K., 1990: Problemy ochrony lasu na gruntach porolnych / Problems of forest protection in stands on post-agriculturallands - English summary. Sylwan 3-12: 75-88. Rykowski, K., and Sierota, Z. 1977. Badania nad przygotowaniem do produkcji biopreparatu z grzybem Phlebia gigantea (Fr.)Donk. / Investigations on preparation with Phlebia gigantea - English summary. Prace IBL 534: 74-90. Rykowski, K., and Sierota, Z. 1984. Aspekt ekonomiczny wystypowania huby korzeni w drzewostanach sosnowych na gruntach porolnych / Econornical aspect of root rot in pine stands on post agriculturalland - English summary. Sylwan 1: 11-21. Sierota, Z. 1975. Ocena skutecznosci zabiegu sztucznej inokulacji pniak6w sosnowych przy uZyciu grzyba Phlebia gigantea (Fr.) Donk na skaly p61gospodarczR. 600 >. u 50ca .9- 40 liJ 30 20 10 10 20 30 40 50 60 70 80 90 100 Coverage of treatment, % Figure 2. Efficacy of Rotstop against homo- and heterokaryotic isolates of H. parviporum and H. annosum s.str. and their mixtures in pieces of spruce logs. AH the isolates originated from Finland. Results from depths of 3 and 8 cm from the treated surface are presented. The experiment was made in the greenhouse. 208 Control 100 90 ~ >R- 80 ~ 0 ci 70.8 '" 60ë 0::: 50 .3 cm 0 40 >- 08 cm ü 30Ol ~ ~ ,~ 20 rD 10 0 N C'l C'l lO <0 r-- r-- a> a ~ ~ C'l ' C'l a> co 0> <0 0> <0 + a> C'l Cl 0> <0 C'l a 0> + <0 + 0> C'l r-- C'l r-- ' J r-----..-. --- O,\ntagonlsls. [ sUI"\'e~ lIlI ,\ntagoll iSlS. ri ,un '-'y 011. annosull1. 1sut"vcy Ill'lll allnosurn. [[ sut"vey II. annùSUl1l. Il ,un <:y H. annO'Ulll, 1slllvey AntagOl1lsls. JI sUt"vey Figure 1. Mean percentages of antagonists and H annosum isolations from treated stumps, six months (1 survey) and one year (II survey) after inoculation (means referred to each survey and indicated by the same letter do not differ significantly at p=O.OS). ACKNOWLEDGEMENTS We would like to thank Prof. Jan Stenlid and Dr. A.F.S. Taylor, Department of Forest Mycology and Pathology, Swedish University of Agricultural Sciences, Uppsala, Sweden, for kindly providing us with Rostop® and for linguistic suggestions, respectively, and Prof. Ottmar Holdenrieder, Section of Forest Pathology & 214 Control Dendrology, Department of Forest Sciences, Federal Institute of Technology, Zürich, Switzerland, for his kind help in identifying the P. rubescens isolate. This research has been granted by the European Union, the Italian Ministry for University and Scientific and Technological Research and by LN.E.A. within the programme "P.O.M. A 24, Misura 2 -Innovazioni nella difesa dalle malattie di piante agrarie e forestali con mezzi di lotta biologica e integrata - Lotta biologica e integrata ai marciumi radicali delle piante forestali". REFERENCES Capretti, P. 1998. Italy. In: Heterobasidion annosum: Biology, Ecology, Impact and Control (S. Woodward, 1. Stenlid, R. Karjalainen & A. Hüttermann Eds.), CAB International, Oxon UK-New York USA: 283-313. Capretti, P.; Moriondo, F. 1983. Danni in alcuni impianti di conifere associati alla presenza di Heterobasidion annosum (Fomes annosus). Phytopathologia Mediterranea 22: 157-167. Capretti, P.; Mugnai L. 1989. Biological control of Heterobasidion annosum in Silver fir (Abies a/ba Mill.) stands. In: Morrison, D. J. (ed.). Proceedings of the Seventh IUFRO Conference on Root and Butt Rots. Vernon and Victoria, British Columbia, Canada, August 9-16, 1988: 277-287. Capretti, P.; Barzanti, G. P.; Luisi, N.; Puddu, A. 1998. Group dying of Silver fir (Abies a/ba) by Heterobasidion annosum in central and southern Italy. In: Root and Butt Rots of Forest Trees, 9th International Conference (c. Delatour, J.J. Guillaumin, B. Lung-Escarmant, B. Marçais Eds.), Carcans-Maubuisson, France, September 1-7,1997: 440 (Abstract). Guillaumin, J. J.; Anderson, J. B.; Legrand, P.; Ghahari S. 1994. Use ofdifferent methods for mapping the clones of Armillaria spp. in four forest of central France. Proceedings of the Eighth International Conference: Root and Butt Rots. Wik, Sweden and Haikko, Finland, August 9-16, 1993: 437-458. Luisi, N.; Sicoli, G. 1993. Una grave moria dell'Abete bianco associata a Heterobasidion annosum in Basilicata. L'Italia Forestale e Montana 48 (2): 83-92. Nicolotti, G.; Gonthier, P.; Varese, G. C. 1999. Effectiveness ofsome biocontrol and chemical treatments against Heterobasidion annosum on Norway spruce stumps. European Journal of Forest Pathology 29: 339-346. SAS Program for Windows Release 6.12 Copyright © 1989-1996 by SAS lnstitute lne., Cary, NC, USA. Stalpers, J. A. 1978. Identification ofwood-inhabiting fungi in pure culture. Studies in Mycology 16: 248 pp. Woodward, S.; Stenlid, J.; Karjalainen, R.; Hüttermann, E. A. 1998. Heterobasidion annosum: Biology, Ecology, Impact and Control, CAB International, Oxon UK-New York USA: 589 pp. Control 215 TESTING OF ROTSTOP ON SITKA SPRUCE, DOUGLAS-FIR AND LARCH LM. Thomsen1 and J.B. Jacobsen2 1 Danish Forest and Landscape Research Institute, H0rsholm Kongevej Il, DK-2970 H0rsholm, Denmark. E-mail: imt@fsl.dk. 2 The Royal Veterinary and Agricultural University SUMMARY The ability of P. gigantea (Rotstop®) to colonize stern discs of Larch, Douglas-fir and Sitka spruce was tested in the laboratory. P. gigantea was able to grow on fresh discs cut from ail three species, although in sorne cases, growth rates were less than on Norway spruce and Scots pine, which were used as reference trees. P. gigantea was able to prevent infection by H annosum on ail five species. The experiment indicates that Rotstop® may have potential for stump treatrnent on Larch, Douglas-fir and Sitka spruce. However, field trials are necessary to confinn these results. Keywords: sturnp treatrnent, Heterobasidion annosum, Phlebiopsis gigantea INTRODUCTION In Denmark, aU conifers are introduced, and Norway spruce is the dominant soft wood species. However, several other conifers are widely used, mainly Sitka spruce (Picea sitchensis), Douglas-fir (Pseudotsuga menziesii) and Larch (Larix x eurolepis). In addition, various pine species are planted on nutrient poor soils, including Scots pine (Pinus sylvestris). Most of the conifers used are considered susceptible to H annosum rot and butt rot. Stump treatrnent with either urea or Rotstop is therefore recornrnended in thinnings and clear cuts. Rotstop® is a biocide produced by Kemira üy for stump treatrnent against Heterobasidion annosum. It consists of spores of Phlebiopsis gigantea based on an isolate found on Norway spruce (Picea abies) in Finland. In several trials in the Nordic countries, Rotstop® controlled H annosum on Norway spruce (P. abies) and Scots pine (Pinus sylvestris) stumps. Only few trials on other conifer species have been reported. The ability of Rotstop to compete with H annosum on stump discs of Sitka spruce (P. sitchensis), Douglas-fir (P. menziesii) and Larch (L. x eurolepis) was tested in the laboratory as part of a bachelor thesis in Forestry at the Royal Veterinary and Agricultural University (Jacobsen 1998). MATERIALS AND METRODS One tree each of Larch, Sitka spruce, Douglas-fir, Norway spruce and Scots was felled, a 1 m length of stem was surface sterilized (70% ethanol) and cut into 2 cm thick discs on a band saw. Each disc was numbered and packed into a sterile plastic bag. The discs were randomly distributed to treatrnents, and the bottom disc was incubated to make sure that H. annosum was not already present. The average disc diameter was 12 cm. Norway spruce and Scots pine had no visible heartwood. Each combination of tree species and treatrnent was represented by five discs. For inoculation with P. gigantea, an aqueous suspension of working-strength Rotstop® (kindly donated by Kemira Oy) was made and applied by spraying at a rate equivalent to 200-300 spores/cm2 • A P type isolate of H annosum from an infected Douglas-fir was grown on PDA to provide conidia for inoculations at a rate equivalent to 100-400 spores/cm2• 216 Control Spore concentrations were deterrnined by counting chamber (haemocytometer), and the viability of suspensions by plating onto PDA. Treatments: 1. P. gigantea alone. 2. H annosum alone. 3. First P. gigantea and then H annosum on top, both on whole disco 4. Division of disc in six wedgeshaped sections, three ofwhich were treated with P. gigantea, after which the whole disc was inoculated with H annosum. The discs were put in paper bags, wrapped in dampened newspaper and incubated in plastic bags for 10-17 days at 200 e (Table 1). Infection of H annosum was confirmed by presence of conidiophores. Infection by P. gigantea was detected by the presence of orange staining and strands of typical mycelia. In addition, P. gigantea was identified by microscopic examination of hyphae (double clamps and oidia). The areas occupied by each fungus were outlined on the discs and calculated. Re-isolations from the interior ofthe discs were also made. As most discs were totally covered with mycelium of either species, the estimated growth rate based on the abundance of the mycelium provided the means of illustrating the success of infection and thus the suitability of the substrate for supporting superficial fungal growth. The growth rate of aerial mycelium was scored in three categories: 1. Slow (score 1) = mycelium only visible at 40 x magnification. 2. Medium (score 2) = mycelium visible at 6 x magnification. 3. High (score 3) = mycelium visible to the naked eye. Table 1. Number of days since inoculation until the fungi could be identified. Species Scots pine Larch Sitka spruce Douglas-tir Norway spruce H. annosum 10 10 10 10 17 * P. gigantea 10 10 14 17 17 * * The long incubation ofNorway spruce was due to heavy contamination by moulds. After the 10 days of wet incubation, the discs were left 7 days under drier conditions and were then easier to assess. RESULTS Both H annosum and P. gigantea infected discs from the five tree species to the extent that both fungi became widely distributed on discs of all species. However, there were clearly differences in the appearance of the surface mycelium on different species, and this attribute has been used to evaluate the success oftreatment. P. gigantea prevented the growth of H annosum mycelia on all five tree species. On all discs treated with P. gigantea there was no growth of H annosum visible, except for three cases with very small colonies « 1cm2 ). Re-isolation from wood below these colonies yielded only P. gigantea. On discs that were not treated with P.gigantea, mycelium and conidia of H. annosum became visible on all five tree species after 10 days' incubation P. gigantea surface mycelia was markedly less effusive on Douglas­ fir and Sitka spruce compared to Scots pine and Larch. However, P. gigantea managed to colonize entire disc surfaces even where mycelial growth was not abundant and prevent the establishment of H. annosum (Fig. 1). Mycelium of both fungi was more luxuriant on sapwood than on heartwood, and results are therefore shown separately for both these tissues (Fig. 1). On sapwood, P. gigantea grew best on Norway spruce and Larch, almost as well on Scots pine, slower on Douglas-fir, and least on Sitka spruce (Table 2 and Fig. 2). On heartwood there was no significant difference. H annosum in sapwood grew best on Scots pine followed by Norway spruce, Control 217 then Douglas-fir and Larch, and slowest on Sitka spruce. In heartwood, growth on Sitka spruce was better than on Douglas-fir and Larch. On the sectored dises, P. gigantea had overgrown adjacent but untreated sectors on pine after 10 days, but not on larch, Sitka spruce, or Douglas-fir. However, re-isolations from the interior of untreated sectors of dises from these tree species only yielded P. gigantea. After another week, surface mycelium of P. gigantea appeared on large parts of the untreated sections. There were problems with contamination by moulds of dise surfaces, especially on Norway spruce. However, colonies of H. annosum and P. gigantea could still be determined. Table 2. Average mycelial abundance on sapwood for P. gigantea inoculated alone and together with H. annosum, calculated from scores weighted by size of area occupied. Letters indicate 5% significant differences within rows. Tree species Scots Larch Douglas-tir Norway Sitka Growth rate in sap wood pine Spruce Spruce P. gigantea alone 2.6 B 2.9 A 1.7c 3*A 1.5 D P. gigantea + H. annosum 2.3 8 2.8 A 2.4 c 3* 1.6 D * Due to heavy contamination with moulds, results for Norway spruce are not totally comparable with the other species. DISCUSSION Testing the efficiency of stump treatment under laboratory conditions cannot be compared to in situ treatments. However, laboratory tests are, by comparison, faster and easier to carry out and they may provide valuable information. The purpose of the present study was to investigate whether the wood of Sitka spruce, Larch and Douglas-fir could be colonized by the Rotstop isolate of P. gigantea, and whether infection by H. annosum was prevented by this application. Failure under optimal conditions in the laboratory might indicate problems with practical application in the field. Estimating the success of colonisation by assessing the abundance of surface mycelial growth only a few days post treatment may be problematic. However, the three types of mycelial abundance were very distinct, and in addition the size of the area occupied by each growth rate category was calculated carefully. The area weighted score was considered to be the best possible method of quantifying the difference in amount of surface mycelium. How accurately the amount of surface mycelium reflects the suitability of the substrate is of course debatable. In any case, the differences between each tree species were barely significant (at p=O.05), and it is not possible to determine the extent to which these were genuine species responses, or artifacts of a novel system of experimentation. 218 Control Abundance of P. gigantea mycelium wben inoculated togetber witb H. HO' \ \ Medium 1 \ i .0. 1 k' Sap wood Heartwood m l - ----1 Norway spruee Lareh Scots pine Douglas-fir Sitka spruee Figure 1. Rotstop® was able to prevent infection by H. annosum for ail tree species independently of growth rate of the active agent (P. gigantea) as estimated by the amount of mycelium on the dises' surfaces. In Fig. l, each colurnn represents one dise. Norway spruce and Scots pine did not have visible heart wood. The average growth rate of both fungi on sapwood of each tree is shown in Fig 2. Comparison of growth rates in sapwood High Medium Slow l P. gigantea alonc P. gigantea and H. annosum H. annosum alone Norway spruce Larch Scots pine Douglas-fir Sitka spruce Figure 2. Both P. gigantea and H. annosum grew weil on Scots pine, larch, and Norway spruce, less fast on Douglas-fir and slowest on Sitka spruce. As only P. gigantea is present on Rotstop treated dises, the growth rates of H. annosum is just illustrated by the right colurnn. Control 219 CONCLUSION Wood of larch, Douglas-fir and Sitka spruce is suitable as substrates for P. gigantea (Rotstop®). Growth of P. gigantea seemed to be slower and surface mycelium less dense on Sitka spruce and Douglas-fir, compared to Larch and Scots pine. P. gigantea prevented infections of H annosum on Larch, Douglas-fir and Sitka spruce. However, these were prelirninary laboratory tests, and they must be verified by field trials before final conclusions on the suitability of Rotstop for stump treatment oftree species other than Norway spruce and pine. REFERENCES Jacobsen, J.B. 1998 Potentialet for brug af Rotstop® i dansk nâleskovbrug. - En unders0gelse af hvorvidt st0dsi110ringsmidlet Rotstop® kan bruges pa léerk, douglasgran og sitkagran. Bachelorprojekt. Institut for Plantebiologi, KVL, 31 pp. 220 Control POTENTIAL FOR BIOLOGICAL CONTROL OF HETEROBASIDIDN ANNOSUM IN THE UK USING ROTSTOP® J. Webber and K. Thorpe Forest Research, Forestry Commission, Alice Holt Lodge, Farnham, Surrey, UK SUMMARY Currently, Phlebiopsis gigantea is used as a stump treatment to control Heterobasidion annosum in the UK, but is only applied to pine grown in the south east of England. Stumps of other conifers are treated with urea. A possible alternative is Rotstop®, a form of P. gigantea registered for use in Scandinavia, and effective both on Norway spruce (Picea abies) and pine (Pinus spp.). We have examined the genetic diversity of a range of isolates of P. gigantea (including Rotstop), and also compared their ability to colonise the sapwood of Sitka spruce (Picea sitchensis), Norway spruce and pine. On the basis of the our results, we consider the potential of Rotstop for stump treatment on spruce and pine in the UK. Keywords: Rotstop, Phlebiopsis gigantea, Heterobasidion annosum, butt rot, bio-control INTRODUCTION The root and butt rot pathogen Heterobasidion annosum is the most economically significant disease of conifer forests in northern temperate regions. It occurs in virtually all managed coniferous forests of the Northern Hemisphere and causes losses in Europe which exceed €790 million per annum (Woodward et al. 1998). To control the disease, the stumps of freshly felled trees are protected with prophylactic treatments using either chemicals (Pratt 1996, Pratt et al. 1998, Rishbeth 1959) or with the bio-control agent Phlebiopsis gigantea, to prevent invasion by H. annosum (Holdenreider and Greig 1998, Rishbeth 1963). One formulation of P. gigantea available for stump treatment is marketed as PG Suspension (Omex Environmental Ud., Kings Lynn). The isolate it contains came from Scots pine, and PG Suspension is registered for use in the UK, but only on pine. Much earlier work, aimed at establishing whether UK isolates could be effective against H. annosum on species other than pine (Rishbeth 1970, 1963), was not pursued. However, in the UK the major commercial conifer species (c. 28% forest area in Britain) is Sitka spruce, and only chemical stump treatment is available for this species (Pratt 1996). There is an alternative formulation of P. gigantea, Rotstop®, which is registered for use in Scandinavia and effective on both Norway spruce and pine. This could, potentially, provide a non-chemical form of stump treatment for Sitka spruce in the UK. Therefore, two questions were addressed within this study: • Can Rotstop be introduced for use in the UK, or is it genetically distinct from the UK population of P. gigantea? • Will Rotstop be effective on Sitka spruce? The genetic variation within and between Scandinavian and UK isolates of P. gigantea, including the current isolates in PG Suspension and Rotstop®, was examined using molecular markers. The colonising ability of sorne of these isolates was also examined in logs of Scots pine (Pinus sylvestris), Norway spruce (Picea abies) and Sitka spruce (P. sitchensis). Control 221 MATERIALS AND METHOnS Molecular analysis Cultures of P. gigantea (see Table 1) were maintained on 1.5% Oxoid malt agar at 20°C in the dark. DNA was extracted from isolates grown for seven days on MA overlain with cellophane discs. The mycelium was scraped from the cellophane and the DNA extracted by grinding in Tris-HCL, EDTA and SDS using an electric drill (Kim et al. 1999). Molecular markers were generated using RAPD-PCR with Operon primers obtained from VH Bio (Newcastle UK) and also by using RAMS-PCR (randomly amplified microsatellite DNA) as described by Vainio et al. (1996) to characterise the isolates. The banding patterns were visualised on 1.4% agarose gel formulated within Tris-acetate-EDTA (TAE) buffer, incorporating ethidium bromide, and viewed on a GelDoc 1000 system (Bio-Rad, Hemel Hempstead, UK). Colonising ability Two weeks before inoculation into logs, autoclaved wheat grains were soaked in sterile distilled water and colonised with one of four UK or four Scandinavian isolates of P. gigantea. Fourteen 0.5 m logs each of Scots pine, Norway spruce and Sitka spruce were obtained from a site near Alice Holt Lodge in Surrey (Nat. Grid reference SU 807427) in February 2001. The cut ends of alliogs were coated in a bitumen-based sealent (Isoflex, Wetpatch) immediately after felling and then inoculated with the pre-colonised wheat grains the following day. Using a balanced incomplete block design, each log of each species was inoculated with four of the eight isolates. Wheat grains were dropped into holes drilled through the bark with a power drill, at four equally spaced points along a central band, equidistant from the cut ends of each log. The logs were maintained at 18-20°C and after four weeks, the bark was removed from around the inoculation points, and extent of growth assessed by tracing the visibly colonised areas (lesions) on both the sapwood and the inner surface of the bark. The traced lesions were cut out, weighed, and the weight converted to an area measurement. RESULTS Molecular analysis Five of the 22 RAPD primers tested produced clear, consistent banding patterns (OPA02, OPA04, OPAI3, OPA18 and OPK19) and a considerable degree of polymorphism was apparent in the RAPD markers. RAMS markers showed similar levels of polymorphism. This indicated a high level of variation within the P. gigantea UK population. However, Scandinavian isolates (including Rotstop) had many of the molecular markers seen in the UK isolates, and no distinct grouping emerged between isolates from different geopgraphic areas. Typical molecular profiles and a dendrogram are shown in Fig. 1. 222 Control - ~ ~ - - - - - - - J- - - - ~ ~ - - - - - - - 1 -f- - J - - - - - - -f-- ~ r- ~ .l- ~'-- -1-- 1-- - ~ - - - - f- - t- I~ t-- .----L.-.. ~~ -f- - t-- ~ - t-- - - ~ - - -1- r-'- rl 70 100 90 50 - 80 60 N C'"l ~ Cl. > > > > > > .0 :r :r :r :r 1.645, and therefore significant at P=0.05. Finally, thirty individual conidiophores were isolated, and 10 individual conidia from each conidiophore were subcultured. The resulting 300 single-conidium isolates were analyzed for clamps and their ISG was determined by TSCP PCR. Genetics and Population Dynamics 239 Studies on the pathogenicity and virulence of hybrids: inoculation experiments We perfonned two greenhouse experiments and one field inoculation experiment. In greenhouse trials each ISG has shown higher virulence on its corresponding natural hosts (Worrall 1983) (Otrosina et al., in prep.). True firs, sequoias, hemlocks and Douglas-frrs ("S-hosts") were more susceptible to S isolates, while pines ("P­ hosts") were more susceptible to P isolates. Sitka spruce seedlings were always susceptible to both ISGs, and thus this tree species was identified as a "universal" host. In greenhouse experiment 1 (Fig. 1), we compared virulence ofan S, a P, and the natural SP-hybrid isolate. Although these isolates were genetically unrelated, they each infected several trees and stumps, demonstrating their viability in the field. All three isolates were dikaryotic. Isolates were inoculated on seedlings ofwhite fir (S­ host), ponderosa pine (P-host) and Sitka spruce (S and P "universal" host) and virulence cwas mesaured in tenns of seedling mortality. In greenhouse experiment 2, ponderosa pine seedlings were inoculated with one S and one P homokaryon and with an SP dikaryon obtained by mating the Sand P homokaryons in the laboratory, allowing a direct comparison of virulence. Differences in ploidy may be irrelevant in this species, as both haploids and dikaryons of H. annosum can be virulent and are commonly found in nature (Garbelotto 1997). Finally, we perfonned a field inoculation experiment (experiment 3, Fig. 1), in which fungus-colonized wood dowels were inserted into holes drilled in white fir (S-host) tree roots. In this experiment, virulence was expressed as the extent of longitudinal colonization of inoculated roots. We used the same S, P, and SP hybrid isolates as in experiment 1. However, in this case holes were drilled beyond the cambium and the outer layer of the xylem; this bypassed the pathogen-specific host defense responses. It has been shown that sapwood inoculations do not nonnally discriminate between Heterobasidion species (Swedjemark 1999). To further verify this assumption, host reaction was also studied through high pressure layer chromatography (HPLC) analysis of colonized root xylem extracted twice with methanol (Bonello 1993). Stumps as a potential hybridization zone The question arises whether conditions exist in nature that may be conducive to hybridization. Two main requirements exist for successful hybridization: (a) both species must be present in an area, and (b) there must be colonization courts where both species can come into close contact, mate, and their hybrid offspring thrive. Fresh stumps may provide a non-selective colonization court in which the Sand P taxa can mate (Otrosina 1992; Garbelotto 1996). To quantify the impact of stump availability on the composition of Heterobasidion populations, we studied the genetic structure of this fungus in stumps, trees and the air-spora in three Califomia National Forests (NFs) dominated by the P-selective hosts, pine and juniper. Airbome spores provide a measure of the overall population structure in a site. Spores are also essential for the persistence of Heterobasidion in nature (Otrosina 1989). Spores were collected by using the exposed wood-disk method described by James and Cobb (1982). Individual colonies were isolated from the wood-disks and their ISG was detennined by TSCP-PCR. Isolates from wood and stumps were obtained by isolations both from infected wood and from the context of basidiocarps. ISG was detennined by isozyme analysis or by TSCP PCR. Pathogen species are tracking their specifie host(s) While a common habitat is required for physical contact and mating between the two Heterobasidion species, it is also necessary that both species be present in the same geographic location. We analyzed the presence of S isolates in live pine/juniper trees and stumps in relationship to the geographic distance from the two most important hosts for this group: hemlock and true fir (Tsuga and Abies spp.). In this analysis, we included data from three National Forests in which the disease is known to affect both S- and P-hosts (Inyo, Plumas, Modoc), as weil as data from NFs where the disease is known only on S-hosts (Stanislaus, Eldorado) or on P_ hosts (Cleveland, data not shown). Regression analyses were perforrned to verify the presence of a correlation between distance from the specifie host and frequency of retrieval of its adapted pathogen. 240 Genetics and Population Dynamics RESULTS AND DISCUSSION On the genetic stability of the natura) hybrid Ali of the hyphal tip subcultures were c1amped and heterozygous for the P- and S- specific markers (i.e. they were putative hybrids). This result is explainable by hypothesizing a complete lack ofhomokaryotic hyphae in the thallus. This feature would differentiate the hybrid isolate from Sand P heterokaryons, in which homokaryotic hyphae are always present. This characteristic would also imply a significant stability of the hybrid thallus. Both laboratory experiments (Hansen 1994) and field surveys (Garbelotto 1998) have indicated that natural isolates may be mosaics of the two homokaryotic parents and of the resulting heterokaryon. The stability of the hybrid thallus was also confirmed by multiple isolations (over 20) of the same hybrid genotype from at least four different stems (stumps and trees). The analysis of the nuclear condition of the conidia provides us with a further understanding of the ploidy modifications associated with the natural hybrid. The number of uninucleate conidia, as determined by microscopic observation after DAPI-staining, was significantly higher in the hybrid than in two S isolates (Table 1, chi, P). Despite a majority of uninucleate conidia in the hybrid, ail isolates obtained by subculturing of individual mitospores were S-P hybrids as per TSCP PCR determination. These results can be explained by hypothesizing the diploid or polyploid nature of the natural hybrid. Changes in ploidy are commonly reported for hybrids in other groups of organisms as weil as in fungi (Kuldau 1999). While further studies are needed to determine the exact ploidy of hybrids, it should be noted that we could not differentiate the size of DAPI-stained nuclei between the hybrid and the two other isolates. This may indicate that hybrid nuclei may be diploid, rather than polyploid. As ploidy increases over 2, a significant and detectable increase in nuclear size should be evident. Pathogenicity and virulence of hybrids assessed through inoculation trials In experiment 1 and 2 (Fig. 1), the P isolate killed significantly more pines than the S isolate (ANOVA P; a a a ( ;; 0.15 N' ~ 9t isolates and mortality was scored cumulatively 15 0.7": ~. ( ;; ,. < 0.10 ci" in a 12-month period (Worrall, 1983 #29). On ~ 0.6" ,,; 0.05 ::::l the Y axis, mortality is expressed as -./(0.5 0.5-+""1r'"~"r""'1r-t""'+"''r'''''l''''''''~:;~ -""~'r""'l""""""'.'ir="u;;.y~o.Ifo 0 3' +proportional seedling mortality). In s p sp s p sp s p sp s p sp s p sp experiment 3, 60 roots were inoculated in four Expt. no. 1 1 1 2 3 randomized blocks as described in Garbelotto Host sp. pine tir spruce pine tir et al. (Garbelotto 1997). Extent of fungal Seedling 400 colonization in the inoculated roots was or root no. measured after six months (data shown with one SD). ANOVA and Tukey Kramer RSD multiple range tests were performed separately for each species (letters indicate homogeneous groups at alpha=0.05). 242 Genetics and Population Dynamics . 1 Inyo 5 12 Trees lX:2!I Inyo 5 69 Stumps QVOVSXXXXXI Inyo 3 144 Air-spora L~~~ ·~~~·~,:,!·~.:.r~.:.~·~ .:..l'I• .:..I'I• .:..I'~':' .l'~':'.' 'i Plumas 2 29 Trees (2l Plumas 2 38 Stumps IÇXXXXXXXXJ Plumas 2 46 Air-spora L·.':'~·'..:..'·~':' ~;~:. .';1•.:. .';~ ,:,.';'.,:,!;,~,:,~;~,:,~·~4 Modoc 7 105 Trees IX) Medoc 7 79 Stumps Œ:XI Medec 2 28 Air-spora ri;; ·I.i:; i.ir"I·;:i.i;:i.i! j.i!i .il Figure 2. (a) Proportions of Heterobasidion ISGs from trees (T), stumps (S), and air-spora (A). Pooled data from aIl study sites within each NF are presented, however distributions were compared either with t­ tests using individual sites as replicates, or alternatively with X2 tests (Ho= % S ISG in stumps= % S ISG in trees= % S ISG in air-spora). a Nat!. Forest Sites Isolates n n o 20 %SISG 40 60 80 100 (b) Results of regression analyses correlating the proportion of S isolates in stumps (filled squares) and trees (filled circles) and the distance from stands with a co-dominance of true fir or hemlocks (TF/H) for at least two miles. Data were from four TF/H sites (only stumps), three sites in the Inyo NF, two sites in the Plumas NF, four sites in the Modoc NF, and two sites in the Cleveland National Forest (only stumps). Regression analyses showed a statistically significant positive correlation (Y=1.05­ O.OIIX, R2=0.94, P