THE - Bodleian · PDF file(Vice-Residents sit acr observers on the Governing Council ... NEWS OF MEMBERS ARTICLES Terminalia ... I visited the Forest Research Institute of Malaysia

THE COMMONWEALTH FORESTRY ASSOCIATION

PATRON HER W E S T Y THE QUEEN

Past and Present Officers

&ddcnt 1979 - HIS GRACE THE DUKE OF BUCCLEUCH AND

QUEENSBERRY, KT.

Pavt Res&nts 1921 - 35 H.R.H. THE PRINCE OF WALES, KG, KT, KP, GMMG. 1936 - 42 H.R.H. THE DUKE OF KENT, PC, KG, KT, GCMG, GCVO. 1943 - 56 THE EARL OF ATHLONE, PC, KG, GCMG, GCB, GCVO. 1957 - 72 THEDUKE OFBUCCLEUCHANDQUEENSBERRY, PC, KT, GCVOO, 1973 - 78 SIR SS. MALLINSON, CBE, DSO, MC, DL.

R.T. BRADLEY, MA. J.D. BRAZIER. ISO. DSc.

Chairman: P.J. WOOD, MA. V i e Chairman: J.S. MAINI, PhD. Hon. Treasurer: P.G. ADLARD, MA

Past Chairmen: 1921 - 22 The Rt. Hon. VISCOUNT NOVAR, KT, GCM, PC. 1923 - 24 Colonel The Rt. Hon. LORD COURTHOPE, MC, PC. 1924 - 27 Sir J. STIRLING MAXWELL, Bt, KT. 1927 - 32 Sir P.H. CLUTTERBUCK, CIE, CBE, M. 1933 - 34 Sir J. STIRLING MAXWELL, Bt, KT. 1934 - 37 The Rt. Hon. VISCOUNT STONEHAVEN, GCMG, DSO, PC. 1938 -3 The Rt. Hon the EARL OF CLARENDON, KG, GCMG, GCVO, PC. 1954 - 59 The Rt. Hon. LORD MILVERTON, GCMG. 1959 - 61 SIR H. CHAMPION, CIE. 1961 - 64 E.B. LATHAM, CBE, MM. 1964 - 72 Sir A. GOSLING, KBE, CB. 1972 - 74 J.A. DICKSON, CB. 1975 - 77 C.G. LATHAM, MA, CA. 1977 - 79 J. WYAlT-SMITH, CBE. 1979 - 82 D.R. JOHNSTON, MA. 1982 - 85 J.D. BRAZIER, ISO, PhD. 1985 - 88 M.E.D. POORE, MA, PhD. M1 Biol. 1988 - 90 R.T. BRADLEY, M.A. 1990 - 93 R.L. NEWMAN, B&.

Regional Vice Chairmen D. GWAZE (Africa) J.S. MAINI (Americas) P.J. WOOD (Europe and Mediterranean) S. SHYAM SUNDER (Indian Subcontinent) R.L. NEWMAN (Asia and the Pacific)

Erccutiu Committee PJ. WOOD (Chairman) J.S. MAIM (Vie-Chairman) P.G. ADLARD (Tnosurer) R.D. BARNES R.T. BRADLEY J.D. BRAZIER BUCCLEUCH, The Duke of J. BURLEY M.L. CLARK J.M. DOBBY P.D. HARDCASTLE G.D. HOLMES R.H. KEMP (2.0. L A T W J.S. McBRIDE D.L. McNEIL F.R. MILLER M.E.D. POORE

' on the Standbrg Committee on C o m n w e a l t h Fashy 3?izEzLEY

P.G. AD-D (UK) G.L. AINSCOUGH (Canada) J. ANGWENYI (Kenya S.-m Bu(ahs2) I. AWANG jhlalavsia, R.D. B&S (lk) ' P.J. BERG (New Zealand) A G J . BROWN (Australia) BUCCLEUCH, The Duke of J. BURLEY (UK) M.L. CLARK (Canada) AJ. COMBEN (UK) T.R. CUTLER (UK) J.M. DOBBY (UK) R.G. FLORENCE (Australia) D. GWAZE (Zimbabwe) P.D. HARDCASTLE (UK) G.D. HOLMES (UK) W.J. HOWARD (UK) R.H. KEMP (UK) P.R.O. KIO (Nigeria) L.S. KIWANUKA (Uganda) F.B. MIGHT (Swaziland) J.P. LANLY FAO) J.S. MAWI (Lad.) J.S. McBRIDE IUK) D.L. McNEIL (uK)' C.H. MURRAY (FAO) PJ. MWITWA (Zambia) E . k OCHIENG (Kenya) F L C RE D (Canada) so imc&msoN (New w a n d ) R.W. ROBERTS (Canada) L. ROCHE (UK) G.P. SAKANDA (Malawi) S. SHYAM SUNDER (India) B.S. SIYAMBANGO (Namibia) J. SPEARS (USA) W.R.J. SUITON (New Zealand) A. TOLFTS (Vanuatu) P.A. WARDLE (FAO) D. WIJE-WARDANA (New Zealand) PJ. WOOD (UK) (Vice-Residents sit acr observers on the Governing Council

Editorial Commiitee J.D. BRAZIER (Chairman) P.G. ADLARD R.D. BARNES A.J. GRAYSON R.M. LAWTON J.M. PIPER P.J. WOOD

Desk Ediror A.M.D. OKOLE

m a n t Seuetaty C.L. MACPHERSON

Auditor KS. WRIGHT, FCA, AT11

h & u of the Sehlich Memoricrl Tiuvt Fund CHAIRhiAN VICE-CHAIRMAN DIRECTOR, OXFORD FORESTRY INSTITUTE

Queen's A w d WfiyrsrJ 1988 Dr. J.W. TURNBULL (Australia)

Dr. D, GMiOUR (Australia) runner up

Commonwealth Forestry Review Volume 73(1). M& I994 i

Contents

CFA CHAIRMAN

QUEEN'S AWARD FOR FORESTRY 1989 S.N. RA1

REPORTS OF MEETINGS

FORTHCOMING INTERNATIONAL EVENTS

OBITUARIES

Richard Smeathers Y.S. Rao L.K. Danso Brian and Jenifer Baylis

NEWS OF MEMBERS

ARTICLES Terminalia amazonia (Gmel.) Exell: development of a native species for reforestation and agroforestry 9 D. NICHOLS

Estimating tree volume using a new form factor 14 J.A.A. DA SILVA, B.E. BORDERS and G.H. BRISTER

Bole volume equations for high forest timber species in Ghana 18 J.L.G. WONG and H.L. BLACKETT

Establishment of clone banks from plus trees selected in the OF1 international pine trial network 23 A. POTTINGER

Germplasm health and disease indexing with particular reference to forest trees in tropical countries 28

M.H. IVORY and P.B. TOMPSETT

A four-phase sampling method for assessing standing volume using Landsat-TM-data, aerial photography and field assessments 35 M. KOHL and S.P.S. KUSHWAHA

Variation of cypress aphid Cinara cupressi (Buckton) attacks on the family Cupressaceae 43 J.A.F. OBIRI

Provenance variation in Pinus maximinoi: a promising species for commercial afforestation in Zimbabwe 47 B.I. NYOKA

RESEARCH NOTE

Comparison of the growth and yield response of Pinus patula between natural stands in Mexico and South African plantations 54 C. AGUIRRE-BRAVO and S.A. WINTER

BOOK REVIEWS

Adams, W.T., Strauss, S.H., Copes, D.L., Griffin, A.R. (Eds.) Population genetics of forest trees 56

D.A. ROOK

Aiken, S.R. and Leigh, C.H. Vanishing rain forests: the ecological transition in Malaysia 56 J. VANCLAY

Banuri, T. and Apffel Marglin, F. (Eds.) Who will save the forests?

Colchester, M. and Lohman, L. (Eds.) The struggle for land and the fate of the forests P.J. KANOWSKI 57

Bawa, K.S. and Hadley, M. (Eds.) Reproductive ecology of tropical forest plants P. GIBBS 58

Bonga, J.M. and von Aderkas, P. In vitro culture of trees D.A. ROOK 5 8

Bradley, P.N. Woodfuel, women and woodlots (Vol.1) Bradley, P.N. and Huby, M. (Eds.) Woodfuel, women and woodlots (Vo1.2) J.E.M. ARNOLD 59

Calder, I.R., Hall, R.L. and Adlard, P.G. (Eds.) Growth and water use of forest plantations P. FREER-SMITH 60

Callister, D.J. Illegal tropical timber trade: Asia-Pacific G.M. PLEYDELL 60

Cubbage, F.W., O'Laughlin, J. and Bullock, C.S. Forest resource policy A. WHITEMAN 6 1

Dargavel, J. and Tucker, R. (Eds.) Changing Pacific forests: historical perspectives on the forest economy of the Pacific Basin S.D. RICHARDSON

ii Commonwealth Forestry Review Volume 73(1 J, March 1994

Grayson, A.J. Private forestry policy in Western Europe E.G. RICHARDS 62

Penka, M., Vyskot, M., Klimo, E. and Vasicek, F. Floodplain forest ecosystem I!: after water management measures G. PATTERSON 63

Prinsley, R.T. (W.) The role of trees in sustainable agriculture

F.L. SINCLAIR 63

Richardson, B.A. Wood preservation R. A. PLUMPTRE

Rodgers, A.D. Bernhard Eduard Fernow: a story of American forestry J. BURLEY 64

Rohrig, E. and Ulrich, B. (Eds.) Temperate deciduous forests (Ecosystems of the world 7 ) P.S. SAVILL 65

Schreuder, H.T., Gregoire, T.G. and Wood, G.B. Sampling methods for multiresource forest inventory H.L. WRIGHT 65

~chultes , R.E. and Rattauf, R.F. The healing forest: medicinal and toxic plants of the Northwest Amazonia D.J. MABBERLEY 66

Subba Rao, N.S. and Rodriguez-Barrueco, C. (eds.) Symbioses in nitrogen-fixing trees R.B. PEARCE 66

NOTICES

Brazier, J.D. A revlew of tropical hardwood consumption 67

Canadian Forestry Association Old growth forests 67

Jain, N.C. (Ed.) Tree improvement and provenance research 68

Longman, A. Rooting cuttings of tropical forest trees 68

Swedish Pulp and Paper Association Annual publication 1993 68

Watkins, C. (Ed.) Ecological effects of afforestation: studies in the history and ecology of afforestation in western Europe 68

PUBLICATIONS RECEIVED 68

SUMMARIES

French Spanish

INDEX TO ADVERTISERS

Aberdeen University Australian National University ICRAF Meyer International Nelspot Norbord Stuewe Sylvametrics

inside back 74

5 7 3 13 46 22 34

Tbe Association does not hold itself responsible for any statementsor views expressed by the authors of papers, reviews or other contributions. It holds the copyright of all original material.

l

2 CamaronweaIth Forestry Review Volume 73(1), March 1994

also unique in the Indian subcontinent, is that some of the irnpwtant National Parks and watershed areas are being rigidly protected with the help of the armed forces.

The work of ecosystem restoration in the middle hill slopes wmveryirnpressive. Someofthe degradedmhaveregenerated very well with protection. The level of detailed microplanning and the extehsion work for protection of the forest areas was commendable. However, it came to light that techniques for regeneration in the moist deciduous forests were not well developed. Through a presentation and in discussions, I shared my experiences in Karnataka, where polyculture gapregeneration, using native species, is being successfully achieved in moist deciduous forests practically devoid of regeneration.

In our afforestation, native species are mixed with Acacia auriculiformis andCusuarinaequisetifolia, timber species which act as a nurse and provide an intermediate, mainly fuelwood crop. The foothills of the Himalayas that form part of Nepal have conditions similar to those in parts of India with Imperata grasslands which need to be afforested. I shared my experience of these matters with local colleagues and exchanged notes on prevalent afforestation practices and soil conservation. Forestry- reIated matters were discussed at the King Mahendra Trust with its Member Secretary andat the International Centre for Mountain Development. I met the Director General of National Parks and Wildlife and learnt about the excellent work they are doing. A captive breeding programme of a rare deer was being contemplated. /

August 1992 in Malaysia. During August 1992 I participated in arneeting at KualaLumpuron SilviculturalResearchin Tropical RainForestsandlatervisitedresearchsites in Peninsular Malaysia. The ongoing work on tending operations, in which saplings are freed of competition, is very valuable. Rate of growth was approximately three times higher under near-open conditions than was obtained under full overhead shade. Growth of the saplings which were freed fromcornpetition was better, indicating that cleaning operations in recently worked areas are a beneficial activity. I visited atropical pine plantation where under planting with rattan was carried out. The growth of rattan is impressive and it has good scope for practice. I visited some of the afforestation sites, where the three main species being planted are Acacia mangium, Pinus caribaea var. hondurensis, and Eucalyptus deglupta. It was considered there is scope to introduce some of the tropical hardwood species which are capable of coming upon heavy soils. Specieslike Lagerstroemia lanceolata, Terminalia paniculata, T. arjuna, Pterocarpus msup ium and Dalbergia latifolia can be successfully introduced in most of the afforestation sites or in the gap regeneration areas. Introduction of teakin these areas was attempted, but the growth looked poor,apparently due to heavy soils and impededdrainage.

I visited the Forest Research Institute of Malaysia and the atboreturn, and shared my research and professional experience with several scientists.

There are some excellent species of tropical fruits that can be introduced from Malaysia into Vopical India, including durian, rambhutan, Artmarpm champaka and rnangosteen. Serious &srts to inkrodwe them do oot seem to have been made in the P@-

August 1993 in SriLanka. In Sri Lanka I had the opportunity to visit Sinharaja National Park which has been declared a World Heritage site. Some 70% of its species are endemic. There is a profuse growth of ground orchids; alamanda vines grow wild and plenty of kitul palm (Caryota urens) which is tapped for toddy and jaggery (locally made unrefined sugar). Jack fruit, bread fruit and coconut palm are abundant. The Kandian home gardens, with nutmeg, cinnamon and clove among other spices, are a virtual treasure of plant collections.

I participated in the Fifteenth Asia Pacific Forestry Commission meeting atcolombo, and visited the dry forest area. This has an interesting combination of species: several of Diospyros including ebony (D. ebenum), satinwood (Chloroxylon swierenia), and a species of Garcinia. In wet evergreen forest, Garcinia spp. and mostDiospyros spp.-are understorey trees but where, with an abrupt transition from wet evergreen to relatively dry conditions, they occur with deciduous vegetation, they appear capable of excellent growth to reach a large size, as is found with satinwood and ebony in Sri Lanka.

All around Sinharaja a 3km wide strip has been planted with tropical pines as a buffer zone, with some interplanting with rattans and other species to provide non-wood forest products to the surrounding villages.

I also visited the hill ranges called Knuckles, due to their appearance. These ranges have a montane climate with cloud and pygmy forests in the top reaches. In the past the land was extensively cultivated with teaand as such there areconsiderable stretches of poorly managed tea plantations and grassy blanks. Eucalyptus grandis and Pinus caribaea var. hondurensis have been doing well as afforestation species.

At an afforestation site near Anuradhapura, planted mainly with Eucalyptus camaldulensis, Holoptelia integrifolia, Azadirachta indica and Acacia auriculiformis, I discussed choice of species and other technical matters with local officers. At the research station and nursery at Kumalpulla there is a successful programme of research on mangrove areas, with several species being regenerated without difficulty. I saw plantation sites in Kaltharu district.

I gave a talk to the foresters of Sri Lanka on 'Ecosystem restoration and afforestation activities in Kamataka', based primarily on work which I had done as Conservator of Forests, Kanara Circle, Dhanvad.

One of the remarkable things noticed in Sri Lanka was Feronia elephantum growing in the moist zone as a fruit tree, whereas in India it is considered as a species of the tropical dry zone. It has a well developed root system and, when used as a root stock, grafts are more successful than those of citrus, as it is hardier in dry areas.

There is an ongoing biodiversity survey, of both flora and fauna, to provide benchmark data and several other excellent studies in the forest of Sri Lanka.

September 1993 in Malaysia. My attendance at the 14th Commonwealth Forestry Conference at Kuala Lumpur provided an opportunity to meet and interact with foresters and other scientists from different parts of the world. A view which emerged was that people's participation, particularly where thereis apressure of population, is essential for sustainable forest

Commonwealth Forestry Review Volume 73(1), March 19% 3

management and the Indian experience is very much under focus. It also emerged that with a long history of forest management and recent strides in forest policy changes and afforestation, India is in the forefront of tropical forestry. However, the experiences need to be shared more widely for the benefit of mankind.

There was a pre-conference meeting of the Commonwealth Forestry Association attended by members of a core group in order to restate and emphasize the role the Association can play by being a bridge for North-South relationships. There are many common problems and an active and expanding Association can help meet and resolve these.

A visit to the forest areas for the planting ceremony of the Commemorative Commonwealth Forestry Park was instructive. The pioneer bamboo Gigantochloa sp., which comes up in the openings in the tropical rain forest in these areas, can be an excellent bamboo for cultivation as, free from branches for most of its length, it is easy to harvest.

REPORTS OF MEETINGS

International Dalbergia Workshop, 31 May - 4 June 1993

An International Workshop on the genus Dalbergia was held in Hetauda, Nepal from 3 1 May to 4 June. Sixty-oneDalbergia researchers and practitioners from 14 countries attended the workshop, sharing information on the diversity of the genus; management and use of the different species; and the potential for genetic improvement of Dalbergia sissoo. During the week. participants divided into working groups to assess current knowledge of various Dalbergia species. These working groups produced: 1) recommendations for further research on Dalbergia species, 2) a draft manual on the use and production of D. sissoo, and 3) draft copies of Nitrogen Fixing Tree Highlights on D. latifnlia and D, melanoxylon. The manual, highlights and workshop proceedings will be available in 9 to 12 months.

The major species of discussion at the workshop wasDalhergia sissoo. A trulymultipurpose tree, sissoo produces nitrogen-rich fodder and green manure, high-quality fuelwood and charcoal, strong durable poles and timber, and beautiful dark brown wood for furniture and panelling. It is also used in agroforestry systems to protect the soil, improve crop production, and provide long- term financial security. These characteristics make Dalbergia sissoo a popular species for industrial, community, and farm- forestry planting. It is a valuable resource for national forestry programmes, commercial enterprises, and private farmers.

Given the importance of Dalbergia sissoo at all levels of forestry, participants at the Dalbergia Workshop recommended the establishment of a RegionaWalbergia sissoo Improvement Programme for South Asia and Southeast Asia. Objectives of the programme would be to: - interlink all regional organizations and agencies currently

active in Dalbergia sissoo improvement work; - assemble distinct provenances of Dalbergia sissoo from

the region and identify their physiological and wood quality characteristics;

- select, test, multiply and distribute genetically superior seed and clones of Dalbergia sissoo;

- develop and distribute diseaseresistant Dalbergia sissczo varieties;

- establish in situ and ex situ gene conservation facilities; - train Dalbergia sissoo research and extension workers. For the present, the Nitrogen Fixing Tree Association (NFTA) will function as a communication link for individuals, organizations and agencies interested in pursuing the establishment of the Dalbergia sissoo Improvement Programme. When established, the programme could have one or more offices associated with different forestry research organizations in the region.

For publications from the International Dalbergia Workshop or the proposed Dalbergia sissoo Improvement Programme, please contact James Roshetko, NFTA, 1010 Holomua Road, Paia, Hawaii 96778, USA.

Growth and Yield Estimation from Successive Forest Insentoiies, 13-1 7 June 1993

Thirty-three papers were given during three days of presentations at this conference (IUFRO S4.01/S4.02/S4.1 l), held at the Royal Veterinary and Agricultural University, Copenhagen. Papers focused on four major topics: the nature of successive samples, the statistical analysis of remeasured samples, the forest management interface, and general growth and yield studies. During a field trip, participants visited three long-term thiiingexperiments inoak, beech andNorway spruce.

Proceedings are available from: Danish Forest and Landscape Research Institute, Skovbrynet 16, DK-2800 Lyngby, Denmark. Price Dkk 150+ Dkk 30postage. Selected papers will be published shortly in a special issue of Forest Ecology and Management.

Institure of Foresters ofAustralia Bien~ia l Conference, 19-22 September 1993

The 15th Biennial Conference of the IFA was held at Alexandra Headland, Queensland. Peter Wood attended as an observer and reports as follows:

'It was a particular privilege to be able to attend this meeting, which was of a very high standard. Dr Jeff Sayer of CIFOR, in his inaugural address, put development as the first goal of CIFOR, noting that aid todeveloping countries should concentrate on developing valid policies and self-help capability. He expressed a preference for multilateral aid over bilateral and suggested that all aid should move away from support for the public sector, where its effect was tocreate market distortions. He summarized the recent UN initiatives succinctly and highlighted the differing aspirations of the developed and devloping worlds at UNCED, which had notreally beenresolved. Thus global ecology, climate change, third world population increase, and concern about women and forest dwellers (the concerns of the no*), were set against national sovereignty, the

4 C81~1~nwenlth Forestry Review Volume 73(1), March 1994

need for technology tram*, aid and poveay alleviation, which wexp: central to the concerns ofthe south. He ended on an upbeat note outlining same of the priorities in forestry research for m R . Many ofthe papets at the conference discussed the pressures

on politicians over forestry poticy and practice. Discussions indicated that there was a widespread feeling that the politicians se4danresponded to their electorate, but only to pressure groups. A major task for land use managers was therefore to enable the silent, sensible majority to inform their elected representatives. This aspact was particularly evident at the Conondales forest ami, which 1 visited on a field excursion and in which conflict resolution in an indigenous forest area had been particularly well done.

In the margins I was able to speak to many Australian foresters concerned with tropical forestry. Much of their work has concentrated on the use of fast growing Australian species, but they also have a good record in rural development forestry, notably but not exclusively in Nepal. The Australian Forest Tree Seed Centre in Canberra has unique access to special germplasm and is a most important collaborator in any scheme using planted trees.'

Modelling StandResponse toSilvicultura1 Practices,27September - l October 1993

Eightyfour participants from 13 countries attended the IUFRO S401 Conference at Blacksburg, Virginia. The purpose of the oonference was todiscuss philosophy andapproaches tomodelling stand responses tosilviculturalpractices such as genetic selection, control of competing vegetation, application of fertilizers, and thinning.

Sessions onmodelling stand response to spacing and thinning (7 papers), fertilizer applications (5 presentations), genetic improvement (3 papers), vegetation control ( l presentation), and combinations of treatments (2 papers), followed the presentations on general stand modelling considerations. A field excursion viewed research studies on response of loblolly pine stands to intensive silvicultural practices.

Proceedings are available for $20 per copy from: Harold E. Burkhart, Department of Forestry, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24061-0324, USA.

FORTHCOMING INTERNATIONAL EVENTS

Full detaib of events already noticed in previous issues have h e n omitted.

15/05/!34-19/OJ/94 USA. Spatial Accuracy of Natural R e ~ o u w Data Bases (IUFRO S4.02).

al Forestry Experiences F c m q Commission

Association. Themes include: social forestry extension approaches; woodland management; agroforestry; planning, monitoring and evaluation of social forestry projects; research in social forestry; role of women in social forestry development. Contact Mr Peter Gondo, ForestryCommission, Forest Research Centre, PO Box HG 595, Highlands, Harare, Zimbabwe. Tel. 263 4 6878: Fax 263 4 7066.

16/05/94-2015194 USA. IUFRO International Symposium on Spatial Accuracy of Natural Resource Data Bases, Williamsburg, Virginia.

19/05/94-25/05/94 CANADA. Future of Northern Forests. Contact Mary Richardson, Box 3009, Athabasca, Alberta, Canada TOG OBO.

01/06/94 FRANCE. FOREXPO (Biennial Exhibition in the Gascogne Forest), Audenge.

20-24/06194 COSTA RICA. International Conference on Ecology and Environment (Foundation for Primary Forest Protection). Topics include: models for development and planning; pacific uses of nuclear energy and environment; quality and environment; sustainable development of forests; forest goods and services; community and ecology; ecology and education; native cultures and ecology; ecological experiences. Contact Celso Vargas, Department of Computer Science, Costa RicaInstitute of Technology, PO Box 159, Cartago, CostaRica. Fax 506 51 53 48. email: [email protected](bitnet)

05/07/94-10/07/94 GERMANY. INTERFORST 94. 7th International Trade Fair for forestry and log timber technology.

06/07/94-08/07/94 UK. First International Conference on Wood Development, Missenden Abbey Management Centre, Great Missenden, Buckinghamshire. Hosted by the Buckinghamshire College of Brunel University in association with the University of Forestry and Wood Science, Sopron, Hungary. Contact Lorna Davis, Conference Coordinator, Buckinghamshire College, Queen Alexandra Road, High Wycombe, Bucks HP1 1 252, UK. Tel: 0494 522141.

1 1/07/94-13/07/94 USA. Stand Inventory Technologies for Forest Ecosystem Management, Portland, Oregon.

24/07/94-27/07/94 CHINA, Harbin. International IUFRO/ NEF'U/FAO Seminar on Forest Operations under Mountainous Conditions, with special attention to Ergonomics, Accessibility and Environmental Protection. Pre-conference cuItural tour 19-23 July; postconference excursion 27-31 July. Contact Prof. LiGuangda, Dept. of ForestOperations, Northeast Forestry University, 8 Hexing Road, 150040 Harbin, People's Republic of China, Fax 86 451 240-146; or Prof. John Sessions Dept. of F m t Engineering, Oregon State University, Corvallis, OR 97331-5706, USA, F a 503 737-4316.

Commonwealth Forestry Review Volwne Bfl), Mmck 1994 5

28/08/94-03/09/94 THAILAND. Measuring and Monitoring Biodiversity in Tropical and Temperate Forests.

30/08/94-3 1/08/94 USA. Third Annual Workshop on Wood Machining and Cutting Tools.

05/09/9408/09/94AUSTRIA. AltemativeMethods forPuposes of Management. (IUFRO S 4.04) Topics include forest stand inventory, inventory by temporary and permanent sample plots, forest mapping based on aerial photographs, satellite scenes and GIS, managerial planning and controlling, comparative assessment andeconomic valuation. Contact Dr Wolfgang Sagl, Universitat fiir Bodenkultur, Institut fiir forstliche Betriebswirtschaft, Gregor Mendel-strasse 33, A- l 180 Vienna, Austria.

18/09/94-21/09/94 USA. Inventory, Socio-economics and Management of the Boreal Forests.

21/W/94-23/09/94 UK. APF International Forest Machinery Exhibition 1994. Thoresby Estate, Nottinghamshire. c o n t i t Show Secretary, Ian Millward, 10 Warren Road, Reigate, Surrey RH2 OBN. Tel: 0737 245 104: Fax: 0737 225619.

25/09/94-28/09/94 USA. 4th International inorganic Bonded Wood and Fiber Composite Materials Conference.

26109194-30/09/94 BRAZIL. International Symposium on Resource and Environmental Monitoring (International Society for Photogrammetry and Remote Sensing). Contact INPE, do, CRI. AV. dos Astronautas 1758, 122227-010 Sao Jose dos Campos, SP-Brazil. Fax 55 123 21 8543.

27109194-0 1/10/94 JAPAN. IUFRO International Symposium on Growth and Yield of Tropical Forests. Sponsored by IUFRO IV, Japan Society of Forest Planning, and International Society of Tropical Foresters. Purposes: 1) to bring together researches on growth, yield, management and environment of tropical- subtropical forests including natural and manmade; 2) toevaluate the state of knowledge achieved; 3) to exchange information on methodological tools such as programs, databases, models; 4) to define research needs and to urge further development; and 5) to provide growth and yield information to the Database of Tropical Forests which willbedevelopedby the IUFRO special programme fordeveloping countries. Contact: Dr Yukichi Konohira,Deputy Coordinator, Division 4, Tokyo University of Agriculture and Technology, 3-5-8 Saiwaicho Fuchu, Tokyo 183, Japan. Fax 8 1 423 64 7812.

03/10/94-06/10#4 CZECH REPUBLIC. Growth Models for Policy Making. IUFRO S4.01-04 Symposium in Prague. The meeting will focus on the use of models for management decisions on different organizational levels. The sessions will span three days; a one day fieid excursion to view permanent research plots andtor air pollution damage to forests is also planned. Contact: Ivo Kupka, Forest and Game Management Research Institute, Jiloviste-Sanady, 156-04 Fraha 516, C x c b Republic. Tel: 422 591612; Fax: 422 591413.

1 l/lO/94- 131 lO.94 USA. Integrating Soolal and Perspectives to Sustain Forest Health.

13/11/94-19/11194PHILIPPINES. IntemationaiWohhopin Albizia and Paraserianthes Species, Nitmga Fixing Tree Association, M i o . Individuals intemsted in attenrting the workshop should submit an abstract of the meazch to be presented by 30 June, 1994. A limited amount of financial assistance will be available for participants. Those who wish to receive financial assistance must send an abstract as well as a letter of explanation and examples of published work rehted to Albizia andParaserianthes. Full papers will be due 30 September 1994. Contact: James Roshetko, Nitrogen Fixing Tree Association, 1010 Holomua Road, Paia, Hawaii %779, USA. Tel: 808 579 9568; Fax: 808 579 8516.

20/03/95-24/03/95 NEW ZEALAND. StandEstablishment and Inter-rotation Management: Second International Conference on Forest Vegetation and Management, Rotorua.

23/04/95 - 26104195 NETHERLANDS. Conference on Ungulates in Temperate Forest Ecosystems.

07/08/95- 12/08/95 FINLAND. IUFRO XX World Congress: Caring for the Forest - Research in achanging World. Contact: Prof. Risto Seppala, Finnish Forest Research Institute, Unioninkatu 40A, SF-00 170 Helsinki, Finland. Tel: 358 0 857 05 1 : Fax: 358 0 625 308.

September 1996. GERMANY. Conference on Effects of Environmental Factors on Tree and Stand Growth, Dresden. Contact: Prof. G. Wenk, Technische UniversitatDresden,Institut fur Waldwachstum und Forstliche Informatik, Wildsdmffer Strasse 18,O-8223 Tharandt, Germany.

ICRAF Training Announcement

The International Centre for Research in Agroforestry (ICRAF) will hold its 'Agroforestry Research for Development' Course, 24 October - 1 1 November 1994, in Nairobi, Kenya. The course covers concepts of agroforestry, characterization and diagnosis of landuse system problems and constraints, and design of agroforestry techniques.

Cost (including accommodation, meals transport within Kenya and health insurance): US$5,900.

For further information contact: The Training Coordinator, October 1994 Course, ICRAF Train- ing Programme, PO Box 30677, Nairobi, Kenya. Fax: 254 2 521 001; Telex 22048; E h 1 157:CG1236.

Commonwealth Forestry Review Volume 73(1), March 1994 7

behalf of FAO. This programme was financed by the Asian Development Bank. It appeared that God really carved out this job for him before he snatched him away from us. In a period of one y w , he was instrumental in providing funds to 27 'hands- on' research projects involving 2 1 re-h institutes and more than 30 scientists, held several seminars and workshops which provided forums for generating new ideas in forestry research, prepared adatabase for forest researchers int brought out about 2 1 monographs on various forestry topics. As he 1eftDehraDunon 8thMarch 1993, whereweall wenttoattend a FORSPA meeting, he was considering how to start a satellite project for training researchers from small countries such as Bhutan, Cambodia, Nepal and Laos.

Sudhakar's energy and vitality were irresistible. He was always approachable; invariably available to the upcoming and younger foresters to discuss, advise or enjoy repartee. He was particularly keen to provide opportunities for the younger forest researchers, scientists and professionals of the region, in whom he saw the future. He wouldcompare life with forest management - older trees being removed to create light and space for the younger trees. He loved epigrams and had one for every- occasion.

Only the chronicle remains to be recorded now: Sudhakar was born on 7th April 1933 and took a diploma in journalism (after his M.Sc.!) from the Osmania University. Perhaps this explains his strength as an extrovert, articulate communicator. He went to the Indian Forest College in 1957 and came back to the Andhra Pradesh Cadre to work till 1967, when he went over to the Re-Investment Survey in Delhi. He never looked back from then: with a Commonwealth Scholarship he took a Ph.D. in Forest Economics at Bangor, Wales; as Project Officer with the State Industrial Development Corporation, Hyderabad, he projectized five brand-new forest-based units which are today's leaders in the paper, pulp and particleboard industries in India; he went back to the Government of India in 1975, from where he eventually landed on the very top of an otherwise low-profile profession. The rest is already recounted above.

Sudhakar leaves behind a fine family of wife Uma, two sons, Gautam and Vikram, married and settled in the USA, and brothers and sisters every one of whom is as warm-hearted and affectionate as Sudhakar himself. But the light that lit their lives and those of hundreds of his admirers worldwide is gone forever.

May his soul rest in peace! S. SHYAM SUNDER

L.K. DANSO

Larry Danso, who died in.March 1993, was a graduate in Forestry of the University College of North Wales, Bangor. In 1958 hejoinedtheGhanaForest Service as Assistant Conservator. After several district assignments he want to Sunyani as Principla of the Forestry Training School from 1 %6- 1 973 and then on to the position of Assistant Chief Conservator, responsible for administration of the Ashanti region.

In 1980 he joined the FAO and took up the challenge of Community Forestry on a project called 'Diversified Forest Rehabilitation' in Thailand This was pioneering work in

communityforests. Larrychosetolivewithtriswifeinthe of the Khao Phoo1uang f m s t reserve, necesParily learnirrg the localThai dialectand workingwiththelocalpeujileencoUraging agrofmtry,treeplantingaFwndthehousehoIdamldevekrpment of household enterprises such as bee-keeping. His success in the project wasrec0gNze.d whenhe waspresented with theB.R Sen award at the 1987 FAO Conference. Later he did consultancies in Community Forestry in Kenya, Malawi and Sierra Leone atxi worked in Moshi, Tanzania in the Hai Rural Energy and Afforestation scheme. The women of Hai maintain his memory in their self-help programme in construction and distribution of improved earthenware stoves.

P.A. WARDLE

BRIAN and JENIFER BAYLIS

Many of our members will have heard by now of the tragic death in October, 1992 of Brim Baylis and his wife Jenifer, in a hght aircraft accident during an aerial assessment of drought damage in the Usutu PulpCompany's plantations in Swaziland Both are sadly missed by their many friends and colleagues in the international community of foresters.

Brim was born in Oxford in 1951 and was educated at Cheltenham College. His first degree was in Botany (with Honours) which hecompleted at theuniversity of Natalin South Africa in 1975. He followed this with the Graduate Diploma in Forestry at the University College of North Wales, Bangor.

Brim's first assignment after qualifying in forestry was as Scientific Officer in Khao Yai National Park which was managed by the Royal Forestry Department in Thailand. His brief was to survey the Asian elephant population and act as information officer and guide to foreign visitors. After spending a year abroad, he returned to Britainin l978 and was appointed forester with the Gardens and Woodland Department of Portmehion Ltd., Penrhyndeudraeth, Gwynedd, where he worked on the rehabilitation and rejuvenation of estate woodlands through the preparation and execution of management plans to place silvicultural practice and maintenance on a sound footing.

While at Bangor, Brim met Jenifer Bamtt, who was born in Highgate in 1957 and educated at Skinner's Grammar School for Girls. She also took her first degree at Bangor and followed it with a PhD. in entomology. Brim and Jenifer were married in August, 1978.

In 1984, Brim and Jenifer went to Uganda on a year's Voluntary Service Overseas. Their base was the Kamanje School of Theory in Mityana where Brim was Coordinator1 Instructor. They re-opened a d agricul-hnical school, ran the administration as well as teaching five subjects and developed the school to 0-level standard. In 1986, Brim was appointed as Head of the Forestry Section of Icour Ltd. in ACC& Ghana. Here his remit was to set up and run community forestry schemes in a semi-arid region of Ghana. a s involved the establishment and administration of the section, opening up of nurseries, all aspects of tree production, extension work, consultation with farmers, species testing, plantation establishment and training.

8 Commom~calth Forestry Review Volume 73(1 J, March 1994

After returning to Britain again in 1988, Brim was appointed to a pdst as Research Officer on a six-month ODA-funded research project at the OxfordForestry Institute. The objective of the project was to make a provenance collection from the populations of Pinus caribaea in the Bahamas Islands. Brim completed this mission very successfully and there are many provenance trials already established with the material he collected. His next project, also at the Oxford Forestry Institute, involved travelling to many parts of the world to evaluate the performance of earlier international provenance trials of other tropical pines.

In May, 1990, Brim and Jenifer and their baby son, Menigan, set off again for Africa for Brian to take up the post of Tree Breeder with the Usutu Pulp Company in Swaziland. Brian was the first tree breeder that the Company had ever employed and he was faced with the challenge of establishing a programme that would put in place the seed orchards to supply Usutu with its Pinuspatula seed requirements as soon as possible. This work was considered to be a key part of the Company's continued success and future development in its contribution to increased yield and improved quality of its products. Brian was instrumental in the rapid establishment of enough seed orchards to provide that seed by the end of the decade. At the same time, Jenifer had increasing opportunity to use her entomological skills and she had recently taken up a teaching position at the University of Swaziland. Among other things, she had developed a system to combat the severe

' O b l e m s the forest was facing with defoliating caterpillars. Brian and Jenifer quickly became friends with many

people in Swaziland and they made valuable contributions to the community both in their profession and in their private life. In 20 years from now, all the Pinus patula growing at Usutu will be the product of the breeding and seed production that Brim started. Their contribution to this fine forest will be a fitting memorial to Brian and Jenifer's skill and enthusiasm.

R.D. BARNES

NEWS OF MEMBERS

E.F. Bruenig, Professor Emeritus in Hamburg University's Chair of World Forestry since 1991, is now based mainly in Sarawak where he is active in the ITTO's Year 2000 programme. He can best be reached at the Forest Department, Wisma Sumber Alam, Petra Jaya, 93660 Kuching, Sarawak, Malaysia. Tel. 82 442180 X 414; Fax 82 4413771445640.

Neill Cooper has retired as Chief Protection Officer for the New Zeatand Ministry of Forestry, and is now working as a freelance Forest and Rural Fire Consultant for Temperate and Tropical Forest. He has recently been recognized and registered as a specialist consultant by the New Zealand Institute of Faestry Council.

Dr Jon Heuch has joined LTS Intemational Ltd. as Forestry consultent, Follewing experience in Nepal and with Fiji Pine Ltd., $on will provide consultancy services in forest

management, silviculture and nursery techniques, research management and the utilization and propagation of bamboo species. He will be based at the LTS Intemational Head Office at 10 Woodhall Millbrae, Edinburgh EH14 5BJ (tel. 031 453 6272; fax 03 1 453 1816).

Cordon Pickles was until June 1992 Director of Operations at ZAFFICO, Zambia. He is now a forestry consultant at Stenka Forestry, mainly occupied with tropical forest management and conservation projects. He assisted with the privatization study at ZAFFICO and Zambezi Sawmills in 1992, and is engaged on a strategical study of wood industries in Sarawak from October-December 1993.

Dr M.N. Salleh, Director-General of the Forest Research Institute of Malaysia, has had further recognition of his contribution to the cause of advancing sustainable forestry and forestry research, in the form of the Malaysian National Science Award which he won jointly with Professor Syed Jalaluddin Syed Salim, Deputy Vice-Chancellor of the Universiti Pertanian Malaysia, at the end of last year.

WOODMARK

The Forestry Industry Committee of Great Britain, representing all sectors of the private forestry industry in Britain, has sponsored a new British WOODMARK which for the first time will allow consumers to identify British grown timber.

In line with the Government's commitment to ensuring the sustainable management of Britain's forests, made at the 1992 Earth Summit, the Forestry Authority has over 300 inspecting staff whose job it is to see that private woodlands are managed in accordance with good forestry practice. Trees may only be harvested with their approval and subject to a replanting programme.

Thus customers buying WOODMARK wood can be confident that they are contributing to sound management practices.

CommonweaIrh Forestry Review Volwne 73(l), M&& IJWIQ 9

Terminalia amazonia (Gmel.) Exell: development of a native species for reforestation and agroforestry

DOLAND NICHOLS Department of Forestry and Natural Resources, Purdue University, West Lafayefte, /IN 47907, USA

SUMMARY

Terminalia amuzonia, (Gmel.) Exell, Combretaceae, is found in the humid tropics in natural forests from Mexico to Brazil. When mature it is a tree up to 50m in height, with adiarneterof up to two metres. Its yellowish wood, usually streaked withred, is suitable formany cons-S and for furniture making. Trees from natural populations have been the source of timber, and the species is now being incorporated into reforestation efforts and agroforestry systems, through the work of farmers and extensionists. It is suggested that the silviculture of this species be developed more thoroughly. There is an urgent need to preserve natural populations of T. ammonia, especially of provenances with low seed production but high potential for use in reforestation projects.

Keywords: Terminalia amazonia, tropics, reforestation, agroforestry. C

INTRODUCTION

It has been estimated that there are some 758 million ha in the tropics whichhave apotential for forestreplenishment (Grainger, 1988). The selection of species for tree planting projects is often not successful and leads to the existence of 'afforestation ruins'. Larnprecht (1989) advises several steps in the process of selecting species:

Analysis of plantations already in an area - considering the failures as well as the apparent successes.

Investigation of indigenous pioneers, their ecology and silviculture. Both Lamprecht (1989) and Evans (1992) suggest that primary rain-forest species are difficult to regenerate and not likely to perform well when planted in brush fields or pastures. More suitable are pioneer gap species from secondary forests, which though capable of invading open areas, could produce fairly large volumes of usable timber on short to moderate length rotations.

Evaluation of relevant publications.

1n the present document &e three listed strategies are employed to evaluate the potential of Terminalia amazonia for use in a variety of management systems, including natural regeneration, agroforestry systems and plantations.

Terminalia amazonia is one of several species of Tenninalia important for timber production, including T. lucida Hoffk., from Costa Rica, and T. ivorensis A. Chev. and T. superbu Engl. & Diels from Africa. Members of this genus frssuently produce branches in horizontal whorls. T. obovata (R. & P.) Steud. is a synonym. In Costa Rica it is known as arnarill6n ormble coral.

Common names, ac~ordingtoLongwood(l%2), includenargusta (US trade name), white olivier (Trinidad and Tobago), cochun (Mexico), guayabo le6n (Colombia), pau-mulato branch0 (Bra- zil), pardillo negro, chichano (Venezuela), and mble (Ecuador).

T. amazonia is native to moist forests from southern Mexico through Central America to Brazil and Peru, and on Trinidad. Both Pennington and Sarukhan (1968) and Longwood (1962) state that the tree is found on a large variety of soils, from sands to clays. In southwestern Costa Rica it ranges from slightly above sea level on the Osa Peninsula to above 1,200 m in Sabalito de Coto Brus. These areas correspond to 'tropical wet forest' and 'premontane wet' and 'premontane moist' transition types according to the Holdridge life zone system (Tosi, 1969). T. amazonia's frequency in primary lowland forests appears to be much lower than in the county of Coto Brus (700- 1 100 m) where it is a major component of many forests, both in old undisturbed stands and in regenerating secondary forest. In the premontane life zone of this county, most forests have been cleared for agricultm, but residual stands are often dominated by T. amazonia where the species can have more than 20 m2 of basal area per hectare.

The species appears to belong to the clas 'long-lived pioneer', with wind-dispersed seed that can produce nearly pure stands after fire and/or clearing (Plate 1).

Individual trees of T. amazonia can be up to 50 m in he@, and grow to two mews in diameter. They fresuently have well- developed buttresses, straight clear boles up to 20 m long, and dark-brown, fimmi, i?d& bark. Small (2 to 4.5 cmin Eensl;th) sunple leaves, with entire mwgins and acute tips, m conoen- trated at the ends of branches F i 1).

beds. Samaras with seed are planted since separating the seed wouldbedifficult and germinationdoes not seemtobe inhibited. They are subsequently covered with an inch or so of soil. Seeds germinate over a period from two to six weeks after sowing.

5. Seedlings germinated in the raised beds discussed above are transplanted, when 5.0 to 7.5 cm tall, either to bags or to raised beds. They are then treated in the same way as in (3) and (4) above.

PLANTATION ESTABLISHMENT AND MANAGEMENT

The usually accepted techniques for plantation forestry in the Zona Sur of Costa Rica are described as follows. An area one metre in diameter is cleared down to mineral soil and maintained for two years after planting. A hole 30 cm deep is made, and 50 grams of 10-30- 10 fertilizer are mixed with the soil in the bottom of the hole at the time of planting. Similar to many trees, T. amazonia has difficulties competing with pasture grasses, and with growing on highly compacted clay soils. It appears to be able to grow adequately on soils where species such as Eucalyp- tus deglupta and Gmelina arborea do not do well. Spacing is usually 3 X 3 m. The lower whorl of branches seems to have a weed suppressing function during the early years of plantation establishment. Foresters have suggested that after two years lower branchesshould be pruned off to improve the value of the first log. The species has a tendency to form multiple stems which need to be cut back to one leader.

GROWTH OF PLANTATIONS

Marshal1 (1939) reported growth of planted trees in Trinidad as 40 feet (12.3 m) in height and a diameter breast height of 1'6" (14.6 cm) at 7 years. Growth rates at seven sites in southwestern Costa Rica are given in Table 1. Life zones are according to the Holdridge system as mapped by Tosi (1969). This dataindicates that T. amazonia may be able to grow at an initial rate of one to two and one half metres per year in height over a wide range of soils, in areas where annual temperatures are approximately 2 1 - 24OC., and rainfall is from 2,000 to 4,000 mm per year.

TABLE 1. Growth ofrerminalia amazonia in southwestern Costa Rica

Location ~ l ~ ~ ~ t i ~ ~ Life Age Mean Mean

(m) zone (months) height diameter (m) (cm)

Salitre de Buenos Aires

Volcan

Veracruz

San Antonio

Uvita de Osa

Uvita de Osa

San Isidro del General

500 Premontane forest, transition

400 Premontane wet forest, transition

S00 Tropical moist forest

700 Tropical moist forest

50 Tropical Forest

S0 Tropical wet forest

670 Premontane wet forest

Ford (1986) observed attacks of between 4 and 45% of the trees in plantations of T. amazonia and other species such as T. oblonga, T. superba, and T. ivorensis. Larvae of the cossid species (genus Cossula. Lepidoptera: Cossidae) bore under- neath the bark and into the xylem and then bore upward 25-40 cm, making galleries 12-14 mm in diameter. Moulaert and Arguedas (1993) found moderate damages done to plantations of T. amazonia by thedefoliatorExophthalmus spp. (Coleoptera: Curculionidae), on a lowland humid site of the Atlantic zone of Costa Rica.

NEEDS FOR FURTHER STUDY PROVENANCES

Given that there are nearly pure forests of T. amazonia at 700- 1100 m as1 in the county of Coto BNS, much of the seed collection for the entire southern Pacific zone of the country takes place there. Nurseries supply seedlings for plantings as low as sea level. This may lead to plantation failures if higher elevation provenances are unable to adapt to conditions at lower elevations.

DISEASES AND INSECTS

Loggers report that often large trees are partially hdlow in the center. Fungal attacks may be a problem, especially if caution is not taken with the season itr which p h n g is done, and if it is not done carefully.

As a rule, farmers and foresters have been advocating the planting of Terrninalia amazonia in areas where it is native. It appears to be able to grow on many soil types, but may not grow wellin areas intheZonaSurwherethereis adroughtof more than three months. For example, in the seasonally dry Potrero Grande area, (elevation 140-700 m asl) Gmelina arborea is planted at lower elevations. In the moister, cooler zones above 700 m, where T. amazonia is a common tree, it is the preferred species for reforestation. Data needs to be collected about the natural range of the species and on its ecology, and limits set on where it can and cannot be planted successfully.

The various propagation techniques discussed in this document need to be compared for their effectiveness and efficiency. Research with known seed sources could l e d to the development of a set of techniques which produce healthy plantations

rapidly. Spacing, pruning, and thinning experiments over a range of sites should be designed to relate these factors to the quantity of wood that can be produced, and to suggest rotation lengths. Inagroforestry, management techniques for T. amazonia with pasture and with coffee can be developed.

CONSERVATION

One of the most urgent needs is to conserve native populations of T. amazonia, especially from low elevation humid forests, where it is not acommon species. Seedcollection in forests from Mexico to Brazil could be used to conserve the species and develop projects which regenerate forests and create productive plantations and agoforestry mixtures.

ACKNOWLEDGEMENTS

The author gratefully acknowledges the assistance of Scott Stanley of the Centro Agronomico Tropical de Investigacih y Ensefianza; Eugenio G o d e z , Chuck Schnell and Dr. Martha Rosemeyer of the Organization for Tropical Studies; Ignacio Retana, Sigifredo Bolaiios, and Marco Robles of the Direcci6n General Forestal of Costa Rica.

REFERENCES

ALLEN, P.H. 1956. The rain forests of Golfo Duke. University of Florida Press, Gainesville, Florida. 417 pp.

CAMACHO, P. 198 l. Ensayos de adaptabi1idad y rendimiento deespecies forestales en Costa Rica. MSc. Thesis. Instituto Tecnol6gico de Costa Rica, Cartago.

CHUDNOFF, M. 1984. Tropical timbers of the world, Agriculture Handbook Number 607. United States Department of Agricul- ture, Forest Service, Washington, DC.

DIRECCI~NGENERALFORESTAL. 1988. ResultadosD&metricos Pacifico Sur, San Jos6, Costa Rica.

EVANS, J. 1992. Plantation forestry in the tropics. Clarendon Press. Oxford.

FORD, L.B. 1986. The Terrninalia borer in Costa Rica. Turrialba 36(2):248-25 1.

G o W , E., B-, R., SEOLEAU, J., and ~ P I N O Z A , M. (Eds.) 1990. Encuentro regional sobre especies forestales nativas de la zona norte y atldntica (1989, Chilamate, Sarapiqui, Costa Rica). Organizaci6n paraEstudios Tropicales. Direcci6nGeneralForestal Instituto Tecn6iogico de Costa Rica, Cartago, Costa Rica.

GONZALEZ, E. and Qmos, G. (in press). Notas sobre la viabilidad de semillas de Terminaha amazonia. Brenesia.

GRAINGER, A. 1988. Estimating areas of degraded tropical lands requiring replenishment of forest cover. International Tree Crops Journal S (ID), 31-62.

JMIWZ, M.Q. and POVEDA, A.L.J. 1991. m l e s maderables nativos de Costa Rica. Museo Nacional de Costa Rica. San Jod, Costa Rica.

LAMPRECHT, H. 1989. Silviculture in the tropics. Technical Coopera- tion, Federal Republic of Germany, Eschborn. 2% pp.

LONGWOOD, F.R. 1%2. Present andpotential commercial timbers of the Carribean with special reference to the West Indies, the

US Department of Agriculture, Forest Service, Washington, D.C. 167pp.

MARSHALL, R.C. 1939. Silviculture of the trees of Trinidad and Tobago, British West Indies. Oxford University Press, London.

MOULAERT, Q.A. and ARGUEDAS, G.M. 1993. Insectos Mvoros asociados con diez especies fmstales en la Regi6n Huetar Norte de Costa Rica Revista Forestal Centroamericana 2: 24-26.

NICHOLS, D. andGONZALEZ, E. (Eds). 1992. Especiesnativasy exbticas para la reforestacibn en la Zona Sur de CostaRica. Organizacick para Estudios Tropicales. Universidad Estatal a Distancia. San Jod. 73 pp.

P E N N I N ~ N , T.D. ~ ~ ~ S A R U K H A N , J. 1968. Manualparala identijicacibn de losprincipales arboles tropicales &Mexico. Commonwealth Forestry Institute, University of Oxford, UK.

R ~ A N A , I. 1990. Problematica del estudio del amarill6n. (Unpub- lished manuscript). Instituto Tecnol6gicode Costa Rica, Cartago. 6 PP.

TRESEMER, J. 1989. A brief report on reforestationefforts near Uvitade Osa, Costa Rica. (Unpublished manuscript).

Tos~, J. 1%9. Mapa ecol6giw de Costa Rica. Centro Cientifico Tropical. San J& CO& Rica.

WORLD LEADER I N FORESTRY MARKING PAINTS AND PAINT MARKING EQUIPMENT

I Write, phone or fax THE NELSON PAINT COMPANY for information: P.O. B O X 2040 TEL. 906-774-5566 KINGSFORD, MICHIGAN 49802 F M : 906-774-4264 USA.

Guianas, and British Honduras. Agriculture Handbook No. 207,

14 Commonwealth Forestly Review Volume 73(1), 1994

Estimating tree volume using a new form factor

JOSE A. ALEIXO DA SILVA1, BRUCE E. BORDERS2, GRAHAM H. BRISTER2 l Universidade Federal Rural de Pernambuco, Recife-PE, Brazil

University of Georgia, Athens, Georgia, USA

SUMMARY

The use of cylinder form factor in tree volume estimation is discussed together with the use of Girard form class and the normal and absolute form quotients. A new estimator of cy linder form factor is derived which uses two stem diameter measurements at 0.5 feet (15.2cm) and 5.0 feet (1.52m) above ground level. These diameters are conveniently measured by a person standing next to the tree. The estimated form factor is used to calculate the total stem volume of 2509 trees distributed over seven large, independent data collections consisting of three species: loblolly, slash and Caribbean pine. The estimator performed very well ineach case. This estimator could be useful in forest inventories for species which do notexpress significant butt swell, and could obviate the need for specific volume equations.

INTRODUCTION

A cylinder form factor (ff) is defined as the ratio of the volume of a tree to the volumeof acy linderhaving the same diameter and height as diameter at breast height (dbh) and total height of the tree. When the cylinder form factor of a tree is known or can be estimated, the cubic foot volume of the tree is calculated as the product of the basal area of the tree (sq. ft.), tree height (ft.) and cylinder form factor. It is well known that ff will vary from stand tostandas wellasfromtreetotree withinthe stand.Cy1inderform factor is usually estimated indirectly by predicting the volume of the tree using a volume equation or directly by taking section measurements along the stem of the tree and calculating the tree volume using Smalian's or some other cubic volume formula.

According to Gomes (1957), one of the fundamental measures of forest mensuration is individual tree volume. If cylinder form factor could be determined accurately for standing trees. an estimate of tree volume could be obtained without reference to volume equations or other estimation procedures. However, measurement ofcylinder form factor inthe field is adifficult task. Foresters usually avoid the direct use of cylinder form factors in favour of volume equations that require dbh only (local volume tables), or dbh and height (standard non-form class volume tables) as input variables. Of course, this implies that a representative sample of the trees must have been measured previously in order to construct the volume tables.

In the southeastern USA, tree volumes are commonly estimated using form class volume tables that use dbh, height and Girard form class (Gfc) as input. Girard form class is a form quotient defined as the ratio of diameter inside bark (dib) at 17.3 feet (5.27 m) off the ground (the top of the f i s t 16 foot (4.88 m) log allowing for a one foot (30.48 cm) stump and 0.3 feet (9.14 cm) of trim) over dbh over bark (Girard, 1933). Horn (1956) developed an approximation to Gfc by using a ratio of dib at 7 feet (2.13 m) above ground level to dob at 2.25 feet (68.6 cm) above ground level. This approximation of Gfc works well over alarge sample of trees but is apoorestimatoron an individual tree basis. It has found very little application in practice.

Two other form quotients that have been used extensively in Europe are the normal form quotient and the absolute form quotient. Schiffel(l899)' defined the normal form quotient as the ratio between the upper stem diameter at half the total tree height and dbh. Jonson (1910)' defined the absolute form quotient as the ratio of the diameter of the stem halfway between breast height and total height and dbh.

Cylinder form factor is a multiplier applied to the product of basal area and height to calculate stem volume in cubic feet or cubic metres. Form quotients are hold-overs from early days when tree volume was estimated from graphs whose coordinate axes were tree volume (cubic feet, cords, board feet, cubic metres) and some function of dbh or dbh and height. Separate lineson the graph were identifiedaccording to formclassor form quotient values. The advent of statistical methods and digital computers ended the need for graphical methods and tree form is now usually accounted for implicitly in the parameter estimates of algebraically expressed tree volume functions.

Cylinder form factor, however, is still a useful concept in estimating tree volume. If cylinder form factor can be measured or estimated on standing trees, then the need for volume equations and their derivation is obviated. This paper explores the possibilities of estimating cylinder form factor from measurements which are within the reach of a person standing next to the tree.

METHODS AND MATERIALS

The datausedin this study consist ofdetailed stemmeasurements taken on loblolly pine (Pinus taeda L.), slash pine (Pinus elliotti Engelm) and caribbean pine (Pinus caribaea var hondurensis Morelet). Loblolly pine data consist of 695 trees collected in 1977 and 70 trees collected in 1981 from the Coastal Plain of Georgia and North and South Carolina, and 276 trees collected

' Quoted in Husch et al., 1972.

Estimating tree volume using a new form factor 15

in 1982 and 164 trees collected in 1983 from the Piedmont region of Georgia. Slash pine measurements are from 685 trees from the lower Coastal Plain of Georgia and north Florida coliected in 1976, and 270 trees from the Upper Coastal Plain of Georgia collected in 198 1. The ages of the trees ranged from 15 to 30 years old. Caribbean pine data consist of 349 trees from plantations in Sri Lanka collected in 1984, with ages ranging from 7 to 17 years old. The distribution of sample trees by height and dbh classes is presented in Tables 1 though 3. All trees in all areas were felled, leaving a stump of 0.50 feet (15.24 cm). The volume of the first butt section, between 0.5 feet (15.24cm) and5.0feet (1.52 m), was calculatedusing Smalian's or Newton's formulae. Thereafter, the stem was cut into 5.00 feet (1 S 2 m) long sections until the last section, containing the 1.00 inch diameter o.b., was reached. Total height of the tree was calculated by summing section lengths and stump height (0.5 feet or 15.24 cm). The volume of the topmost section was calculated on the assumption that it was conical. The volumes of the remaining sections were calculated using Smalian's formula. Total tree volume, in cubic feet o.b., excluding stump volume, was obtained by summing the volumes of all sectio'ns.

It is known that:

where: f f = cylinder form factor, V, = total tree volume (cu.ft.), V, = volume (cu.ft.) of acylinder with diameter equal

to dbh and height equal to total tree height.

In this study it was assumed that form factor could be written as:

Where:

P =

vs =

D, =

D" =

VC =

TH =

h =

constant of proportionality

volume of the first butt section (cu. ft.), between D, and D,,, calculated by Smalian's formula,

diameter at 0.5 feet (15.2 cm) from ground level, in inches,

diameter at 5.0 feet (1.52 m) from ground level, in inches,

volume of the cylinder with diameter equal to D, and height equal to 4.5 feet (1.37 m), in cu. ft,

total tree height, in feet,

height where the upper diameter DU was measured ( 5 feet or 1.52 m ).

Substituting D, and DU in (1) and considering p=0.5, produces :

where:

f = estimated cylinder form factor.

Estimated cylinder form factor is used to obtain an estimate of the true form factor of the tree. Observed cylinder form factor was found to vary from 0.33 to 0.71 with over 80% of the values less than 0.5. Note that generally only the smallest trees in the data bases had observed cylinder form factors greater than 0.5. The ratio (vJvc) generally varied between 0.8 and 0.9. Thus, the need for the proportionality constant. A grid search of proportionality constants from 0.3 to 0.8 by increment of 0.1 showed that 0.5 explained the greatest amount of variation in observed cylinder form factor. It was necessary to add the term [TH/(TH-h)] to obtain estimates of cylinder form factor greater than 0.5 which were observed for the smaller trees.

Now. since

V= k . ~ ~ ~ 2 . t h . f f = k . ~ ~ ~ 2 . ~ ~ . f ,

where: V= cubic foot volume for the tree, k= 0.005454, and remaining terms as defined above, estimated tree

volume is:

V= 0.005454 (DBH)2 (TH) 0.25 + 0.25 2 [ E. 7 l (A] ( 2 )

All tree volumes were estimated using formula (2) and were compared with observed volumes using Fit Index (FI) (similar to R2)(Schlaegel, 198 l ) , average residual and average percent residual. Residuals were calculated as observed minus predicted volume and percent residual was calculated as:

where:

PR, = percent residual of tree i,

Vi = observed volume (cu. ft.) of tree i,

pi = estimated volume (cu. ft.) of tree i using equation (2).

RESULTS AND DISCUSSION

As shown above, the form of the proposed form factor can be written as:

where everything is as previously defined.

16 J.A.A. Da Silva, B.E. Borders, G.H. Brister

TABLE 1. Distribution of the loblolly pine sample trees by dbh (inches) and height classes (ft.)

DBH Height Class Sum Class

22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5 72.5

Sum 84 124 164 194 171 149 101 90 62 36 30 1205

TABLE 2. Distribution uf'the slash pine sample trees h j dhh ( i n d e s ) and herght classes (jt.)

DB H Height Class S u m Class

22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5 72.5

11.5 6 1 1 14 3 1

12.5 I 1 8 10

Sum 32 84 114 140 150 113 113 7 1 56 52 30 955

Statistics of fit for all data sets are presented in Table 4. used in this study, as well as when considering the range of Thew \tatistics clearly show thattheestimatedcylinder form diameter classes. In summary, we found out that good total factor expressed as: volume estimates can be obtained for several species from

different regions using this proposed cylinder form factor to define volumes of individual trees. The procedure only

f.=[0.25 + 0.25 (?l] (&) involves the measurement of three lower-stem diameters and total height of the tree. We recommend using this technique of volume estimation for species of similar form (i.e. those

multiplied by: 0.005454.(DBH)2.(TH) produces very good species without excessive butt swell) in situations where estimates of the observed total tree volume for all data sets existing volume equations are not available.

Estimating tree volume using a new form factor 17

TABLE 3. Distribution of the Caribbean pine sample trees by dbh (inches) and height classes (fr.)

DBH Height Class Sum Class

22.5 27.5 32.5 37.5 42.5 47.5 52.5 57.5 62.5 67.5 72.5

11.5 1 3 4 8

12.5 1 2 2 1 6

Sum 14 13 35 45 59.. 68 64 36 19 5 1 349

TABLE 4. Statistics for all data sets used in the development form quotient

Specie and Number of Fit Index Average of Average of Average of Average of year (data trees (FI) observed predicted residuals per cent collection) volumes volumes (CU. ft.) residuals

(cu. ft.) (cu. ft.)

Loblolly 77 695 0.98 16 6.0277 6.1919 -0.164 1 - 1.6395

Loblolly 8 1 70 0.9784 12.630 1 12.1880 0.442 1 4.0984

Loblolly 82 276 0.9858 3.8221 3.7864 0.0357 2.9252

Loblolly 83 164 0.9860 4.8899 4.7407 0.1492 3.6044

Loblolly All 1205 0.9860 4.8899 4.7407 0 1492 3 6044

Slash 76 685 0.9786 4.8270 4.5259 0.301 1 8.0832

Slash 8 1 270 0.9726 1 1.0706 10.2475 0.823 1 8.7304

Slash All 955 0.9726 6.5923 6.1435 0.4487 8.2662

Caribbean 76 349 0.9453 6.7716 7.1272 -0.3556 -2.4539

REFERENCES

GOMES, A.M. 1957. Medi@o dos arvoredos. Lisboa. Livraria S6 da HUSCH, B., MILLER, C.I. and BEERS, T W. 1972. Forest menswatron. Costa. 413 p. John Wiley & Sons. New York. 410 p.

alftk~, J.W. 1933. V0h-m tables for Mississippi bottomland hard- SCHLAEGEL, B.E. 198 I. Testing, reporting, and using b~omass estima- woods and southern pines. J. Foresr~y 31:34-41. tion models. Proceedings of 1981 Southern Forest Biomass

HORN, A.F. 1956. A simplifd method for estimating form class of Workshop. 95-1 12. loMolly and shortkd pine stands in Mississippi. J. Forestry 3: 185- 187.

18 Commonwealth Forestry Review Volume 73(1), 1994

Bole volume equations for high forest timber species in Ghana JENNIFER L. G. WONG1 and HUGH L. BLACKET Forest Inventory and Management Project, PO Box 1457, Kumasi, Ghana

2do PO BOX 30465, Nairobi, Kenya

SUMMARY

Bole volumes were measured for 3488 trees of 61 species from seven Forest Reserves representing the forest types found in the high forest region of Ghana. Measurements were made using diameter tapes on felled trees and wide angle Spiegel relascope on standing trees with good consistency between the two measurements. From these data two sets of regional volume equations were derived using a single parameter power model based on measurements of diameter at breast height. The model was fitted using a log-normal transformation.

KEYWORDS: Volume equations, Ghana, High forest volume, bole volume

INTRODUCTION

Ghana lies between 4"45'N and l l n l l 'N latitude and between 1°14'Eand3"07'W longitudeandcoversanareaof 238,540km2. The high forest zone covers an area of 81,342 km2 and is found in the south-westem third of the country. By 1980 about 17,180 km' or 21% of the high forest was left (Sayer, Harcourt and Collins 1992), mostly under the jurisidiction of the Forestry or the Game and Wildlife Departments. The Forestry Department presently holds about 15,913 km20f the high forest in constituted Forest Reserves, of which 11,590 km2 is managed for the sustainable production of timber for export using a natural selection system (Ghartey 1989, Nolan 1989). In 1985 anational inventory of thehigh forest estate of Ghana waccommenced with the Forest Inventory Project (HP). The objective of this inventory was to provide an estimate of the national growing stock as timber volume to provide information for the rational development of the timber sector (Fran~ois 1989).

Since it is impossible to measure the volume of every sample tree in an inventory some means of estimating volume from simple measurements of accessible parts of the tree is required. This generally takes the form of predictive regression equations based on analyses of detailed volume measurements of a representative sample of trees. In this case predictiveequations ofbole (=timber) volume against diameter at breast height (d) or above buttress diameter were required. Although there are a few extant volume equations for Ghanaian high forest species (e.g. Anin- Bonsu 1970, Alder 1982) none were suitable forthe FIPbecause of species or geographical restrictions. During the course of the inventory 420 species were enumerated which were classified into three commercial groups. Class 1 contains the 66 species registered as having been exported from Ghana between 1973 and 1988. Class 2 species are those attaining 70 cm d and occurring at a frequency of greater than 1 km2 in the FIP results, all other species were placed into Class 3 (Ghartey 1989).

The latest classification of the Ghanaian high forest, that of HallandSwaine (1981) wasadoptedby the FIP(B1ackett 1989). This used an ordination to classify the high forest into nine types and sub-types ranging from the Wet Evergreen type with a mean

annual rainfallof 1500-2100mmintheextreme south-westtothe Dry Semi-deciduous type with rainfall of 1250-1500 mm on the northern foresttsavanna boundary. For the purposes of the FIP only the most extensive types containing commercial species were surveyed these being the Wet Evergreen (WE), Moist Evergreen (ME), Moist Semi-deciduous south-east sub-type (MSSE), Moist Semi-deciduous north-west sub-type (MSNW) and the Dry Semi-deciduous Fire Zone (DSFZ).

In an attempt to include as many commercial species and to account for possible variation in volume functions with vegetation type, a total of 3488 trees of 61 species from seven Forest Reserves were measuredduringthe course of the FIP. These data were used to generate two sets of volume equations which were used to derivenational volume estimates forGhana(Wong 1989, 1990). The purpose of this paper is to bring these equations to a wider audience.

DATA COLLECTION

There were a number of pragmatic decisions made concerning the selection of sampletrees. Giventhe limited time available for the inventory it was not possible to consider measuring sufiicient trees to produce local volume equations for each of the 66 Class 1 species. Therefore data was collected across all vegetation zones and for as many species as possible in the time available. Only trees of commercial quality were measured. The Inventory used subjective codes ranked from 1 (good) to 5 (poor) for bole form and quality; trees of codes 1 and 2 only were selected as being of timber quality. The two volume sampling teams generally measured any tree of a commercial species and quality that could be easily measured that they encountered. A d l y of the species and sizes of the trees measured was kept and the teams instructed to make up any shortfall the aim being to provide adequate data across all size classes for as many species as possible. Volume sampling was undertaken in seven Forest Reserves chosen to represent the Hall and Swaine (1981) forest zones recognized by the FP. The location of these reserves is shown in Figure 1.

Bole volume equations 19

TREE MEASUREMENTS

FIGURE 1. Location offorest reserves selectedfor volume sampling in Ghana. -- ForestlSavanna boundary. -*- Boundaries of forest types. Italics are the Hall and Swaine (1981) vegetation zones: WE - Wet Evergreen, ME - Moist Evergreen, MSSE - Moist Semi-deciduous South east sub-type, MShW-MoistSermrm-deciduous N o r West s u b type, DSFZ - Dry Semi-deciduous Fire Zone.

Measurements made using a diameter tape on felled trees are the most accurate but it was not feasible for the project to fell sample trees. This was overcome by measuring large trees felled by timber concessionaires and supplementing these with relascope measurements of smaller standing trees. The relascope trees were chosen from those bordering roads and skid trails in order to obtain a clear view of the bole. Since the skid trails were all very recent it was assumed that trees adjacent to them would be representative of the forest. Table 1 gives a breakdown of the volume sampling data by Forest Reserve.

TABLE 1. Breakdown of Ghana FIP volume sampling data by Forest Reserve

Total measured Forest Vegeta- No. of

Reserve tion species Zone Felled Standing Total

Bia South ME-MS 42 1031 448 1479

Fure River WE 42 307 415 722 Mpameso MSWW 13 218 0 218 R a Anum MSSE 32 78 129 207 Tain DSFZ l 5 103 227 330 Tributaries I1 Tano Anwia ME 32 100 371 471

Tano Suhien MS 12 6 1 0 61

Bole measurements were made in two ways. Large trees of commercially desirable species that had been felled by timber concessionaires were measured using diameter tapes. Trees of lesser known species and those below the prescribed minimum felling diameters were measured using a wide angle Spiegel relascope. For each tree a number of measurements were made. Diameter was recorded at breast height (1.3 m) or above buttress and at approximately 2 m intervals along the bole from the start of the first log which was taken as starting at the point at which the tree was, or could be, felled to the crown point just below the first major branching. The presence of external defects was noted for each section. The buttress length and crown depth were also recorded. On felled trees measurements were also made of stump height as well as the end diameters, lengths and bark thickness of cross-cut logs. All data were entered onto micro-computers for analysis.

Sectional volumes were calculated using Smalian's formula (Philip 1983). Three volumes were calculated for each tree. The bole volume is the sum of all over-bark sectional volumes. Utilizable volume is the over-bark bole volume minus the volume of any sections marked as defective. Commercial volume is the under-bark volume calculated using Smalian's formula from cross-cut log measurements and which equates to the volume used on the Log Measurement Certificates (LMC) issued by the Forest Products Inspection Bureau of Ghana.

COMPARISON OF TAPE AND RELASCOPE MEASUREMENTS

In order to investigate the consistency of the tape and relascope measurements a sample of 180 trees from three Forest Reserves (102 trees from Fure River, 40 trees from Pra Anum and 38 trees from Tain Tributaries 11) were measured using a wide angle Spiegel relascope, felled and re-measured using a diameter tape. The two measurements yielded very similar data with mean discrepancies of -0.8 cm and0.2 m2 for diameter and bole volume respectively. A comparison of volume equation coefficients derived from each dataset showed that they were not significantly different (t test p > 0.05). It was therefore concluded that the tape and relascope data produce statistically identical volume equations and they were combined into a single dataset for all further analyses. This consistency was thought to be the consequence of using only two volume sampling field teams led by specially trained Technical Officers.

DERIVATION OF VOLUME EQUATIONS

Since the FIP was only recording diameter for each tree the choice of volume equations was limited to single parameter models. A number of models (linear, square, second order polynomial and power models) were tested using the F, r2, RMS, t and Furnival index. It was found that the power model was the most satisfactory. The power model has the form;

20 J L G . Wong and H L . Blackeft

where, v is bole volume in m', dr is diameter at the reference height (either 1.3 m or above buttress) in cm and a and b are coefficients. This model has been extensively used in tropical inventories (Philip 1983, Loetsch, Zohrer and Haller 1973) and has previously been used in Ghana by Anin-Bonsu (1970) and Alder (1 982). The model was fitted following the steps given by Baskerville (1 972) which showed that the most appropriate way of fitting the power model was to perform a log transformation and to use a least squares fit of the linear equation;

where In is a natural logarithm.

The first step in the derivation of the volume equations was an investigation into between Forest Reserve (a surrogate for between forest type) variability. All the data for each reserve were used to generate volume equations which were then compared using tests of parallelism and coincidence as recommended by Loetsch et d. (1973). The procedure followed was that given by Zar (1984) and involves an F test (test for differences between h coefficients) followed by a Tukey HSD test to identify those h coefficients which are not significantly different (test of parallelism). A t test was then used to test the a coefficients for those reserves with similar h coefficients (test of coincidence). These tests revealed two groups of reserves with statistically identical volume equations. Reference to the map showed that these could be identified with two distinct zones for the application of volume equations (see Figure 2). Zone 1 can beequated to theMSNW andDSFZforest typesandzone2 with the WE, ME and MSSE types.

Fici~ R I 2. Zonesfor the application o f ~ d u m e equations in Ghana

The split between the MSSE and MSNW types which forms the baundary between volume Zones I and 2 follows the 1500 mm isohyet (Hall and Swaine 1981) and is therefore possibly related to growth potential.

Since 30 is the lower cut off for small sample statistics volume equations could not be formed for species for which less

than this number of trees had been sampled. This meant that volume equations could only be formed for 32 of the 61 species sampled. The sampled species were grouped using the same procedures used for the Forest Reserves. The coefficients for the resultant equations are given in Table 2 with the species in each group listed in Table 3.

TABLE 2. Volume equation coefficients for high forest timber species in Ghana

Zone1 N Diameter range Coefficients Group min max a h

Note: Estimated bole volume (m) = a diameter (cm)h

TABLE 3. Species in Ghanaian volume equutiorl groups

Zone Botanical name Local name Trade name Group

I A Mansonia altissima Triploi~hiton srlero.~ylon

I B Antraris toxir~ariu Entandrc~phru,qnlrr ritile Milicia e\-celso Nesogurdoniu popui~er.ifi,n

I C Af:eliu aji.il~una Pn-rcopsis elato

2A Bumha.r hrevimspr Celtis nrildhreadii

?B Anitlger.iu rohusru Mansonia altissinto Nauclea diderrir hi; Nesogordonia pupaverfera Parkiu hirolor Petersinnthus mamtcarpus Pterygota rnacrocarpa Pynanthus anpiensis Teighemella heckelii Terminalia superha Triplochitun scleroxylon

2C Distemonanthus henthanzianus Entandrophragma angolense Entandrophragrno cvlindricum Entundrophrapna utile Heritipra utdis Khaya ivownsis

Sterculia ohlonga Terminalia ivorensis

2D Guihortia chic Khaya anthotheca Stromhosia glaucescens

OPRONO Mansonia WAWA Oheche. Samba KYEN-KYEN Antiaris. Ako EFOoBRODEDWO Utile, S i p ODUM Iroko, Chlnrophora DANl'A Nesogordonia

PAPAO Atielia. Apa KOKRODUA Afromosia ONYINAKoBEN Bombax, Kondroti ESA Celtis, Ohia SAMFENANlNl Asanfona,Aningeria OPRONO Mansonia KUSIA O P W DANTA Nesogordonia ASOMA Parkia ESIA Combretodendron KYEREYe Koto OTIE Ilomba BAKU Makore OFRAM Frake WAWA Obeche. Samba BONSAMDUA Ayan, Movtngu~ EDINAM Gedunohor PENKWA Sapele EFOoBRODEDWO Utlle, S ~ p o NYANKOM Niangon DUBlNKoKoo African Mahogany,

Khaya OHAA Stercuiia EMIRE Framire ANOKYE-HYEDUA KRUMBEN Anthotheca AFENA

2E Aningeria oltissima SAMFENA AsanfonaAningeria Lophira alata KAKU Miiicra c.rc&a ODUM Imko, Chlowphora Piptadeniastrum africanum DAHOMA

Bole volume equations 2 1

E S ~ M A T I O N O F ERRORS DEFECT ASSESSMENT, UTILIZABLE AND COMMERCIAL VOLUMES

The use of models toestimate the bole volume of trees is subject to many forms of error. Attempts were made to account for as many of these as possible (Wong 1990). One source of bias that can be accounted for is that resulting from the log transformation of the data. It has been shown that if a log distribution is normal (which is the intention of the transformation) then the estimates generated by the use of the fitted equation yield the mean of the log distribution and the median of the arithmetic distribution (Baskerville 1972). There have been a number of correction factors suggested to remove this effect; however, the one most relevant to the Ghana volume equations is that of Sprugel (1983) based on the standard error of the estimate (SEE) of the regression. This correction factor is given as:

y, = actual volume

A y, = predicted volume

To correct for the log-normal bias the volume in arithmetic units should be multiplied by the correction factor (CR) given in Table 4. The effect of the bias was a small (2-5%) underes- timate of the volume which can be easily corrected.

A more significant estimate of error is the 95% confidence interval (CI) for the regressions. The C1 for a log-normal regression is not constant, being at a minimum at the mean of the regression dataset. It is therefore not possible to quote an overall figure for a regression but the C1 at the mean is a minimum error and so is useful as an indication of the magni- tude of the error and also for comparative purposes. Table 4 lists the C1 for the mean of each volume equation as a percent of the mean and shows that they range from 2-20% with the smallest errors associated with the groups containing the most data. There are many other errors that could be considered but most of these cannot be quantified and there is no means of compounding them into an overall estimate of error.

TABLE 4. Correction factors and errors for Gharto volume eyua- U o n c

Zone1 SE of the Correction Predicted 95% Cl m' Group estimate Factor mean

SEE (Sprugel volume Lower Upper 1983) CR m'

During the volume sampling any visible defects on the boles of the trees were sectioned off and a note made of the type of defect. This is subjective and does not attempt to take account of internal rot or shakes but is the assessment that would be made in the field by a felling supervisor, the assumption being that loggers will cross-cut the log to exclude obviously defective sections. Analysis of these data showed that 28% of the sample trees were found to have defects which on average accounted for 17% of the bole volume. Volume losses due to defects, averaged over all trees, came to about 5%. Net utilizable volume can therefore be estimated by multiplying bulked volumes by 0.95. This figure is therefore an assessment of the volume perceived, prior to felling, to be of usable quality. Note that only those trees of suitable quality for timber were included in the volume sampling so this is not an indication of overall tree quality.

For the felled trees the dimensions of cross-cut logs were recorded to provide an estimate of the volume as recorded on the LMC. The under-bark con~iwlcial log volumes were calculated from these and were found to be about 10% less that the bole volumes. If an estimate {)F [.MC volumes is required the bole volume qhould bc nultiplird by 0.90.

DISCUSSION

The fact that Forest Reserves as far apart as Bia South, Pra Anum and Fure River have the same volume function suggests uniform- ity of growing conditions across several forest types in Ghana. However, analyses of the variability of bole length showed that bole lengths are significantly longer (mean of 21.6 m) in the Moist Evergreen and Moist Semi-deciduous types than in the Wet Evergreen (mean 19.8 m) and the Dry Semi-deciduous (mean 16.4 m) forest types. This pattern is the same as that found by Hall and Swaine (1981) who attributed it to water stress in the dry areas and nutrient limitations in the wettest areas. Since the split into volume equation areas did not follow the same pattern it may be that bole form varies with forest type. A possible hypothevis could be that trees in the WE zone have more cylindrical boles, perhaps because they are s~ower~rowingthan those in the a and MS types, giving rise to equivalent volumes at the same diameter even though the boles themselves may be shorter. However, form factor analyses did not reveal any consistent patterns in tree form with forest type or species and were rejected as a possible means of forming volume equations for Ghana.

CONCLUSION

The consistency of the Forest Inventory Project data was good and the amalgamation into fairly large areas and species groups for application of volume equations was considered acceptable. The volume equations produced appear to be robust and have small errors.

Commomvealth Forestry Review Volume 73(1), 1994 23

Establishment of clone banks from plus trees selected in the OF1 international pine trial network A.J. POTTlNGER Oxford Forestry Institute, South Parks Road, Oxford OX1 3R6, UK

SUMMARY

In the early 1970s the Commonwealth Forestry Institute (CFI) began distributing seed of Pinta caribaea var hondurensis Barr. &Golf., P. oocarpa Schiede and P. patula Schiede & Deppe ssp. tecunumanii (Equiluz & Perry) Styles for what was eventually to become a network of over 600 field trials in 65 countries throughout the tropics, designed to evaluate provenance differences and estimate genetic parameters. A selection of these trials was assessed in the late 1970s and early 1980s for tree form and productivity and results distributed to collaborators. In 1986 the re-named Oxford Forestry Institute (OFI), with the financial assistance of Shell Research Ltd, began aproject designed to assemble approximately 150 plus trees from throughout the trial network in clone banks, with the objective of supplying clonal material of the plus trees selected to collaborators for use in their own breeding programmes. Scions from 73 trees ofpinus caribaea var hondurensis, 43 trees of P. oocarpa and 35 trees of P. patula ssp. tecunumanii selected from trials in 13 countries were established successfully in the OFI glasshouse at Oxford. Following multiplication of new shoots by hedging, scions for grafting were despatched from Oxford to seven tropical countries during 1988-89. Clone banks of the selected trees were subsequently established in Brazil, Zimbabwe and Australia, and further consignments of scions were despatched between 1990-93. Factors influencing the success of graft establishment, multiplication and distribution are discussed. After establishment these clone banks will be able to provide scions for distribution to pine breeders throughout the tropics.

Keywords: Pinus caribaea var. hondurensis, Pinus oocarpa, Pinus patula ssp. tecunumanii, scions, clone bank.

INTRODUCTION

The CommonweaIth Forestry Institute (CFI) began its investigation of genetic variation in Central American pines following arequest made by the Eighthcommonwealth Forestry Conference in 1963 to coordinate a study of the races and provenances of Pinus caribaea Morelet (B.C.F.C., 1962). Up to that time P. caribaea had shown promising rates of growth as an exotic species in several countries on lowland tropical sites with soils of poor nutrient status, yet most introductions were from the same seed source and little was known about the potential of the species for improved productivity. Initial work by CFI concentrated on a study of taxonomic and silvicultural characteristics (Lamb, 1973), followed by exploration of the species' natural distribution and a comprehensive collection of seed from different populations. P. oocarpa Schiede was later included in the programme as this species, which replaces P. caribaea at higher altitudes, also held promise for use in exotic plantation forestry. Seed was sent to the UK for storage by the Forestry Commission from where it was distributed in small quantities to forestry researchers throughout the tropics and sub- tropics for theestablishment of provenance trials. Between 197 1 and 1982 over 600 trials were established in 65 countries, 12 of which later formedthe subjectforan assessment programme, for growth and form, coordinated by the re-med Oxford Forestry Institute (OFQ (Gibson, 1982). The results not only provided evidence of the superiority of certain seed sources but also illustrated patterns-of ge&tic variation, and yielded genetic parameters for later use in tree improvement programmes. One of the major futdings of the ~ g r a m r n e ~ w a s the superior

performance of several of the provenances of P. oocarpa, which upon closer investigation were subsequently re-classified as P. patula Schiede & Deppe ssp. tecunumanii (Equiluz & Perry) Styles. In addition tothe assessment programme coordinated by OH, individual collaborators were encouraged to evaluate their own trials thus providing a broad overview of provenance performance over many site types. A thorough review of the results from this network is provided by Birks and Barnes (1990).

Although the scope for the trials to compare the performance of different provenances was reduced following the onset of competition, a large proportion continued to be managed and maintained in good condition. In view of this and the relatively poor access which burgeoning tropical pine improvement programmes had to germplasm of both superior provenances and promising individuals, these trials appeared to offer the potential to supply useful germplasm to tree breeders throughout the tropics. Furthermore, not only did the trials contain a wide rangeof welldocumentedpopulations whichexhibited promising growthrates, but also, furthercollection workin Central America revealed that several of the provenances represented in the trials were becoming increasingly threatened in their native ranges. The OF1 pine network therefore represented a valuable ex situ conservation resource of some of the most endangered provenances of both P. caribaea and P . oocarpa.

A project was formulated by OFI to capitalize on the genetic resource stored in the trials and make it available to the collaborators in the O R pine network. The objective of the

24 A l . Pottinger

project was to establish a number of clone banks throughout the tropics containing the best individuals from the pine network from which scions could then be supplied to tropical pine improvement programmes. The project began in 1986 and was fundedby ShellResearch Ltd, one of thecollaborators in the pine network.

The programme was planned to include the following stages: selection of suitable trials for inclusion in the project, plus tree selection and scion collection, propagation and multiplication of theclones inoxford, anddistributionof scions andestablishment of clone banks in the tropics. Although both the plus trees and tree improvement programmes, which wouldeventually benefit from the project, were situated in the tropics it was decided to cany out the propagation and multiplication stages in Oxford due to the ease of importing tropical pine scions into the UK and the experience of coordinating the pine network at OFI.

A literature search coupled with discussions with various forestry researchers highlighted the fact that while successful transportation of pine scions within a country was frequently achieved, the approach proposed in the project, with transfer of scions from several tropical countries for propagation in a temperate country followed by despatch to the tropics for grafting, had not previously been attempted. Experimental graftingat OFIof scions which hadbeen imported fromzimbabwe had. however, been achieved on a small scale and suggested that the programme was technically feasible.

Each of the stages of the project is discussed below.

SELECTION OF COLLABORATORS

Collaborators from the pine programme in 13 countries were contacted todiscusstheirpossibleparticipationintheprogramme, based on knowledge of the good condition of their trials, and all agreed to donate scions from selected trees to the project. These organizations and the number of plus trees from which scions were collected are listed in Table 1.

PLUS TREE SELECTION

Due to the absence of genetic information relating to individual trees (the provenance trials compared bulk collections from the native range), plus trees were selected by the author on a visual basis, with emphasis placed on the traits affecting timber production andquality whichare under strongestgeneticcontrol, i.e. stem and branch form. A selectionintensity of approximately 1:800 (or two trees per hectare) was used as a guideline in an attempt to provide ramets of the 200 clones which could be housed adequately in the OF1 glasshouse. While this selection intensity was generally met, in some cases there were more than two exceptional trees per hectare and it was felt that the network of collaborators would benefit fromtheir inclusion. This approach was taken in part due to the large degree of variation encountered in provenance trials emanating from bulk collection from the wildand wouldnot havebeenconsidered if theexperiments were based on seed of more uniform quality from improved sources. A list of the number of trees selected with each research organization in the programme is presented in Table 1.

TABLE 1 . Domr organizations and number of selected trees established in Oxford

Number of trees selected

Organization Countrl. Pinus Pinus Total caribaea Pinus patula

var. oorarpa ssp. hondurensis tecunurnanii

Conservation Australia 7 Commission

of the Northern Territory

Queensland Australia 2 1 21 Forestry Service

Comphania Brazil 3 1 2 6 Agrofomtal Monte Alegre (CAFMA)

Instituto Brazil 3 5 15 23 Florestal de Sao Paulo

Carton de Colombia 1 4 5 Colombia

Instituto Colombia 5 5 Nacional de 10s Recursos Naturales

Renovables y del Ambiente (INDERENA)

Fiji Pine Ltd

Kenya Forest Research Institute

Forest Research Institute of Malawi

Ministry of Natural Resources

Division of Forest Science & Technology

Tanzania Forest Research Institute

Royal Forest Department

Forest Department

Forest Department

Forestry Commission

Fiji

Kenya

Malawi 2 7 4

Surinam 4

South 8 13 1 Africa

Tanzania 3

Thailand 6 5 1

Uganda 2 2 3

Zambia 2 3

Zimbabwe 7 5

TOTAL 73 43 35 151

Establishment of clone banks 25

SCION COLLECTION AND TRANSPORT

Scions from the selectedtrees werecollectedby experienced tree climbers who cut shoots approximately 30cm in length from the upper third of the crown.

Althoughroutine operational grafting results are not usually published, it was g e n e d y agreed amongst experienced grafters encountered at the various research organizations visited that the chances of establishing tropical pine grafts are improved when the scion is approaching the periodof maximumshoot extension, i.e. the bud has broken but the shoot has not extended any more than 2cm. As plus trees were selected from trials in 13 countries, however, it was not always possible to visit them when this type of growth was evident. This situation was complicated further in some of the more tropical sites where growth was almost continuous and trees in the same trial exhibited different phenological stages.

Following collectionofthe scions it wasessential tominimize theeffectsofwaterandnutrient losses. This wasattempted firstly by keeping them cool and damp by immediately placing them in plastic bags containing moistenednewspaper, secondly by putting the bags in an insulated box containing frozen artificial ice- packs, and lastly, by transporting them as quickly as possible to the UK for grafting.

Boxes of scions were despatched to the UK by air freight under licence from the Ministry of Agriculture, Fisheries and Food (MAFF), generally within 24 hours of collection which should have ensured that the scions arrived in the UK within 72 hours of collection. Unfortunately, in practice, several problems arose including flights being delayed, missed connections, documentation being lost, and industrial disputes involving airlines andlorcargo handlers in both the country of despatch and theUK. Due to the need to visit trials in severalcountries on each overseas collection trip it was only possible to accompany five of the 22 consignments of scions despatched to the UK for grafting. While most arrived on time, unfortunately, some took up to Xdays in transitandconsequently arrivedinpoorcondition.

GRAFTING AND PROPAGATION

Scions were transported to Oxford usually within 12 hours of arriving in the UK, althoughin some cases bureaucratic problems resulted in delays of over 24 hours. Upon arrival they were usually soaked for 12 hours in sterilized water toencourage water uptake, and then soaked for five minutes in an insecticide (Malathion) followed by immersion in a fungicide (Benlate) solution prior to grafting in accordance with phytosanitary regulations specified by MAFF.

Scions weregraftedontopot-grownmtstocksofP. caribaea, P. oocarpa, P. patula ssp. recunumanii and P. radiata D. Don., of between 18 months and three years of age, using tip cleft grafting. This method was chosen as it had already been proven successful with a pilot consignment of scions from Zimbabwe and is the approach widely used for grafting pines in the tropics. Five ramets of each clone were established in most cases.

Following grafting, scions were enclosed in apolythene bag, tied around the stem of the rootstock to maintain ahigh humidity, and then covered with a paper bag to reduce the amount of

sunlight entering and hence temperature build-up inside. Daily checks were made of each ramet and any indications of fungal growth dealt with by spraying with Benlate. After about six weeks (or earlier if the ramet was exhibiting rapid growth) the polythene bag was loosened and progressively opened over the following thtee weeks in order to harden-off the ramet until, finally, the bags were removed.

Grafting success with different consignments ranged from 0% to 100% of the clones received. Such variation was not related to the country from which scions were collectedper se, but was the result of the effects of a range of interacting factors, each of which is discussed below. Unfortunately, the large number of factors involved combined with the relatively small number of consignments received meant that accurate analysis oftherelativeimportanceofeach factorwas impossible. Despite this, general trends associated with successful and unsuccessful graft establishment appeared clear and are highlighted.

Scion storage and transit time

The lackof a true dormant phase in the growth cycle of the pines inthis programme meant that, unlike grafting with many temperate species, it was imperative to reduce to a minimum the time between cutting scions from plus trees in the field and grafting onto the rootstocks in Oxford. The grafting results obtained indicated that high levels of establishment (over 90%) were possible with scions which had been in transit for up to 7 days as long as they were kept cool and damp. If, however, they had not been stored properly prior to despatch, or insufficient coolant accompaniedtheconsignment, then levels of graft establishment below 10% were sometimes experienced, even with scions which had been in transit for only 48 hours.

Phenological stage of the scion

Although the greatest grafting success was achieved with scions which hadjust begun to exhibit shoot elongation, it was evident that the phenological stage of the shoot was of secondary importance when compared with the overall vigour of the scion. Success rates over 90% were achieved with scions exhibiting all stages of growth from resting bud through to full elongation (as long as the new shoot was removed) if the scions were taken from vigorous parent trees and arrived in Oxford in good condition.

Season of grafting in the UK

There was amarked seasonality in temperature and light intensity intheglasshouseusedforgrafting. Averagemonthly temperatures ranged from a minimum of 18°C in January to a maximum of 2X°CinJuly, withabsoluteminimaand maximaof4"C and 38°C respectively. Although 400W high pressure sodium lights were employed for six months of the year, there was still a great variation in theamount of light available tothe ramets throughout the year. The highest establishment rates were achieved during the warmer and brighter months between April and September but it was not always possible to despatch scions at this time. It was discovered, however, that, providing there were no other major limiting factors affecting grafting success, such as poor

26 A J. Pottinger

storage and long transit times, it was possible to achieve establishment levels of over 70% even during the coldest and darkestmonths betweenNovember andFebruary despite the fact that at this time the rootstocks were growing at their slowest.

Size, vigour and position on parent tree of scion

Operational grafting programmes with tropical pines attempt to use scions collected from the ends of vigorously growing branches in the uppermost region of the crown. Although all of the plus trees were between 14 and 17 years old, site variation meant that their heights varied from 16-29m. This resulted in a marked variation in the size and vigour of scions collected due to both the vigour of the tree and the ability of the tree climber to reach the upper crown. As expected, grafting results indicated that greatest establishment success was achieved with straight, rapidly growing scions, while the thinner, more bent, and less vigorously growing scions were generally more difficult to establish as shown by poorer grafting success, an increased time to begin shoot elongation, and poorer subsequent development.

Although it was difficult to draw any firm conclusions regarding the effect upon grafting success of position on parent treeof the scion it was evidentthatthe detrimental effect of taking scions from lower in the crown was negligible in the largest and most vigorous trees, as long as the scions came from within the top one-third of the crown. The poorest scions were generally those which came from smaller trees, which were generally less vigorous, and most difficult to climb to the top. The effect of low vigour was therefore compounded by the relatively low position on the crown from which scions could be collected.

Soaking prior to grafting

Once a scion is cut from aparent tree water is lost from the shoot through the effects of transpiration and evaporation. While the visual effects of this are apparent in drying of both shoot and needles, its effect upon graft establishment was unclear. In order to investigate the influenceupon graft establishmentof attempting to replenish water lost during transportation, five consignments of scions were divided into two groups upon arrival in Oxford. One group was grafted as soon as possible after arrival while the other was soaked in sterilized water for 12 hours immediately prior to grafting. Unfortunately, the variation encountered in scions when they arriveddue to the confoundingfactors mentioned above made comparison between consignments difficult. The results, however, appeared to indicate thatthere was no advantage in immersing the scions when they were vigorous and the transportation time hadbeen 72 hours or less. In situations where transportation times were longer, the scions generally appeared tobe dry upon arrival in Oxford andcomparative trials suggested that immersion led to improved grafting success in some cases.

Species, vigour and age of rootstock

A pilot grafting experiment prior to the initiation of the project revealed that a high level of grafting success could be achieved using a tip-cleft graft onto 4-year-old rootstocks of P. caribaea. For the purposes of the project, however, it was moreconvenient

to use rqtstocks between 18 months and 3 years in age which required less glasshouse space. Initially, the three taxacollected were grafted ontortrotstocks of the same species but subsequent experimental m n g onto other species indicated that each could be grafted successfully onto either of the other species. Further trials revealed that more vigorous growth of ramets was achieved when using eitherP. taedaL. or P. radiata as rootstocks, a finding that was perhaps due to the superior growth of these warm temperate species under glass at Oxford in comparison with the pines from Central America.

The vigour of the rootstocks used in the project proved to be a factor of fundamental importance both to the success of graft establishment and tothe rate of subsequent growth of ramets, and consequently every effort was made to produce and maintain rootstocks in good health and vigour. While the most vigorous scions could be grafted with success rates over 60% onto rootstocks exhibitingbelow averageratesof growth, less vigorous scions frequently failed to establish unless grafted onto the most vigorous rootstocks.

GROWTH OF THE RAMETS AND PROPAGATION

The need to cut backnew growth produced on the ramet in order to produce a cluster of new shoots for export as scions, termed hedging, meant that detailedcomparative growth measurements between ramets werenottaken. Ingeneral, however,asexpected, growth rate between clones exhibited greater variation than within clones. The fastest growing clones increased in height by about 30cm during the most rapid growth period of May to September, while the slowest grew as little as 5cm during the same period. As aprinciple objective was to produce scions for distribution to collaborators, this extremely slow rate of growth in comparison with rates experienced in the tropics imposed a severe limitation on the speed at which vegetative propagation could take place. The number of usable shoots for scions produced by hedging varied fromone to six, although this didnot appear to be related to the vigour of the ramet.

DISTRIBUTION OF SCIONS TO COLLABORATORS

Discussions were held with collaborators in the programme in order to select suitable sites forthe establishment of clone banks containing the selected trees. Sites for each species were subsequently selected in Brazil, Colombia (2), the Congo, Zimbabwe, Thailand (2), Australia and Fiji, based on local and regional interest in the different species. Each clone bank was planned to have two roles. Firstly, for the host it was to provide immediate access to arange of germplasm for incorporation into their own breeding programmes, which would otherwise have been unavailable to them, and secondly, each clone bank was intended to act as a source of scions from which other pine breeders could obtain breeding material.

Scions cut from ramets in Oxford were prepared for despatch to the clone bank host by immersion in sterilized water for six hours followed by appropriate treatment with fungicide and insecticide depending upon the i m p o r t r e w n s o f the recipient country. The base of each shoot was then inserted in a small amount of highly absorbent, inert, extruded volcanic tock

Establishment of clone banks 27

(Rockwool) saturated with sterilimd water, and then placed in a small polythene tube which was subsequently sealed to produce an air-tight container. Scions were then placed in an insulated box kept cool by the use of artificial ice packs and taken immediately to the airport.

ESTABLISHMENT OF RAMETS IN CLONE BANKS

Trials at Oxford indicated that scions collected in this manner and stored at around 4°C could be grafted successfully, with an establishment rate in excess of 8O%, up to one weekaftercutting. This situation was, however, never achieved when the scions were despatched overseas, where grafting success ranged from 8% to 58%. These low rates of establishment were due to the effects, either in isolation or combined, of air freight handling errors, flight delays, problems with customs in the recipient country, and problems with grafting and aftercare. Failure to achieve acceptable establishment levels required further consignments of scions to be sent and placed a severe demand upon scion production in Oxford, which, due to the relatively slow growth rates experienced in the glasshouse, became increasingly difficult to satisfy. Although more scions were sent to each clone bank, continuing problems with low rates of establishment meant that it was impossible to continue to supply more scions to all collaborators. Very reluctantly, further scion supply was restricted to thoseclone banks which had been most successful, as it was decided that any further success with the programme depended upon the establishment of alarge proportion of clones in a smaller number of places rather than have them distributed thinly amongst the nine clone banks. This resulted in limiting the despatch of further consignments of scions to the clone banks hosted by the following organizations:

- QueenslandForestry Service, Australia(Pinus taribaea var. hondurensis and Pinus patula ssp. tecunumanii)

- Instituto Florestal de Sao Paulo, Brazil (Pinus patula ssp. tecunumanii)

- Forest Research Centre, Zimbabwe (Pinus oocarpa and Pinuspatula ssp. tecunumanii).

The first scions were sent to these organizations from Oxford in 1989, and subsequent consignments arrived between 1990 and 1993. The composition of the clone banks is illustrated in Table 2. Distribution of scions from these clone banks to other collaborators in the programme is expected to take place within the next few years.

TABLE 2. Clones established at each clone bank

Pinus P~nus Organ~zatton Country . carrbaea Prnus patula

var. oocarpa ssp. hondurensrs tecunumanii

Queensland Australia 61 16 Forestry Servtce

CONCLUSIONS

This programme has shown that anetworkof experimental trials has the potential, following its initial assessment objectives, to provide a valuable source of germplasm toresearchers elsewhere. Furthermore, the technical feasibility of distributing scionwood from such trials has been illustrated clearly. The factors which proved to have the greatest limiting effect upon successful graft establishment were related to the logistics of between-country transfer of scionwood, and the subsequent management of the grafts. The project indicated both the importance and possibility of controlling these elements of the work.

The continued spirit of collaboration between researchers in the OFi pine evaluation network played a key role during the project, and there is, therefore, good reason to expect that the further distribution of scions from the clone banks established to date will provide a valuable input to pine breeding programmes throughout the tropics.

ACKNOWLEDGEMENTS

Shell Research Ltd provided the funds to support the project. with additional support from ODA under research scheme no. R.4584. The original concept for the project came from Dr J.W. Stead of Shell, and Dr R.D Barnes of OFI.

I would like toexpress my particular appreciation to Dr R.D. Barnes for his assistance throughout. Acknowledgement is due to N.F. England, Dr J.G. Purse and K.F. Richardson of Shell for their inputs during the project. Dr A.J. Simons and Dr D. Skidmore of the University of Cambridge contributed valuable advice on transportation of scions, and J.L. Scanlon provided excellent technical back-up during the final stages of the work.

The input of all of the collaborators involved in the project, without whose help and interest the project could not have taken place, is gratefully acknowledged.

Finally, I would like torecord my special appreciation to J.L. Steel for the valuable contribution which she made to all aspects of the work.

REFERENCES

B.C.F.C. (1962). Eighth British Commonwealth Forestry Conference, East Africa, 1962. Proceedings, Committee Reports and Resolutions of the Conference.

BIRKS, J.S. and BARNES, R.D. (1990). Provenance variation in Pin~ts caribaea, Pinus oocarpa and Pinus patula ssp tecunumanii. Tropical Forestry Paper no. 21. Oxford Forestry Institute. 40pp.

GIBSON, G.L. (1982). Genotype-environment interaction in Pinus caribaea. Commonwealth Forestry Institute. 112pp.

LAMB, A.F.A. (1973). Fast-growing timber trees of the lowland tropics. No. 6. Pinus caribaea volume 1. Oxford, UK. Commonwealth Forestry Institute. 272pp.

Institute Rorestd Brazil 27 de Sao Paulo

Fomst Regnvcb Zimbabwe 35 3 1 Centrt:


Germplasm health and disease indexing with particular reference to forest trees in tropical countries

MICHAEL H. IVORY1 and PAUL B. ToMPSETT2 l Oxford Forestry Institute, University of Oxford, Oxford, UK Royal Botanic Gardens, Kew, Wakehurst Place, Ardingly, Haywards Heath, Sussex, UK

SUMMARY

This paper reviews firstly, the dissemination of phytopathogenic organisms either on, in, or together with, various forms of forest tree germplasm. It concludes that relatively few dangerous pathogens are transmitted by tree germplasm in the form of cleaned seed, pollen, or tissue cultures, but that many are disseminated by vegetative tissues, such as cuttings and scions. Secondly, it reviews traditional and modem procedures for the detection and quantification of phytopathogenic organisms which form the basis of disease indexing. Lastly, the paper contains a brief rksum6 of seed health testing and plant quarantine in relation to transfers of tree germplasm and concludes that rapid, cheap screening and diagnostic techniques are needed to ensure that the currently increasing transfers of germplasm are securely achieved.

Keywords: Tree germplasm health; disease indexing.

TRANSMISSION IN GERMPLASM

The major type of germplasm utilized for forest tree species is that of 'seed' (either true seed, or seed inside the fruit with or without remaining flower parts). Other types of germplasm include pollen. tissue or cell cultures, and vegetative tissues such as cuttings and scions. These can be infected with microorganisms in the following ways: ( 1 )internally within the living cells; (2) externally in the dead outer tissues; or (3) externally on the germplasm surface. Organisms carried with associated impuri- ties, such as pieces of cones, needles, leaves and fruits are not dealt with here as they should be excluded by modem seed cleaning procedures. Some larger pathogenic organisms, such as insects. may themselves also transmit smaller pathogenic organisms (viruses or mollicutes): this method of dissemination is also not discussed herein. This review mainly addresses the transmission of pathogenic organisms which give rise to disease in the carrier host, or other hosts. However, many other weak pathogens can destroy germplasm in storage, especially seed and pollen, particularly if storage conditions are suboptimal or the gemlplasm is senescent. Examples of potential transmission are given below when possible.

Trunsrnission in tree 'seed'

Some members of all the major groups of pathogenic microorganisms and pests can be disseminated with seed (Neergard, 1969, NobleandRichardson, 1968, Richardson, 1979, Anderson, 1986) including some which cause disease of forest trees.

1. Insects. Tree seeds are often infected with small insects which destroy their contents (Eungwijarnpanya and Hedlin, 1984;

Southgate, 1983; Sen-Sarma, Thakur and Sehgal, 1988; Singh and Rana, 1990) but few of these cause disease on other parts of growing plants. Many more pests are transported on seeds with attachedparts of fruits and flowers, or with abundantextraneous matter.

2. Nematodes. Tree seeds are not known to transmit plant pathogenic nematodes; such transmission has, however, been reported for some cereal crops (Mathur and Lal. 1989).

3. Fungi. Tree seeds are often infected by large numbers of fungi (Sen-Sarma, Thakurand Sehgal, 1988; Mittal and Wang, 1987; Yuan, Old and Midgley, 1990; Singh and Rana, 1990), most of which are saprophytes living indead tissues. However, a few of these appear to act as weak pathogens causing decay of the seed during storage, especially when held under poor storage conditions; subsequently some of these organisms may cause damp- ing-off in the young germlings produced (Ivwy, 1987). A few fungi are also known to invade the internal living tissues of tree seed and subsequently to give rise to disease of the tree after germination of the seed; examples are the systemic Fusarium wilt of Mimosa (Gill, 1968) and Chestnut Blight (Lanier, Joly, Bondoux and Bellemere, 1976). Some non-systemic disease fungi, such as Glomerella cingulata, which causes leaf and pod spots on Acacias (Sinclair, Lyon and Johnson, 1987) andcertain Fusarium species causing a pod rot on Leucaena spp. (Lend, 199 1) can also spread in this way. Many other fungi which give rise to dormant propagules (e.g. Macrophominnphaseotina and the grass pathogen Claviceps purpurea) or long-lived reproductive structures (e.g. Sphaeropsis sapinea on cone scales of Pines

Germplasm health and disease indexing 29

and Mycosphaerella species on Pine needles), may also be disseminated in these forms within impure batches of seeds (Ivory, 1987)resulting in their transmission across international boundaries.

4. Bacteria. Tree 'seed' are frequently heavily contaminated with bacteria, which are generally found in all dead tissues and onplant surfaces. Mostof these are harmless withvery few plant pathogenic bacteria known to be disseminated on tree seed; examples which have been reported include Xanthomonas campestris pv. (pathovar) juglandis, which causes bacterial blight of Walnut (Lanier et al., 1976; Noble and Richardson, 1968; Sinclair, Lyon and Johnson, 1987) and Pseudomonas fluorescens, which causes bacterial pod rot of Leucaena leucocephala (Lend, 199 1).

5. Mollicutes. Tree seeds are not known to transmit plant pathogenic mollicutes (Mycoplasma-like organisms, SpiroplasmasandRickettsia-like organisms) (Smithetal., 1988, Sinclair et al., 1987), although these organisms are often systemic in other plant parts.

6. Viruses. Despite the frequent systemic occurrence of viruses in plants and trees, few are transmitted in tree seed (Bos, 1977). Those that are transmitted in this way are mostly from the Nepovirus and Cucumovirus groups and have been found in seeds of Ash, Birch, Elm, Ginko, Paulownia, Poplar, Privet, Walnut and fruit trees such as Prunus spp. (Cooper, 1979,1988; Cooper, Massalski and Edwards, 1984; Cooper, Kelley and Massalski, 1988; Lanier et al., 1976; Murant, 1989; Smith, Archer, Dunez, Lelliot and Phillips, 1988). They are mostly canied internally in live cells of the embryo anddevelop to infect the germinating plant systemically.

Transmission in tree pollen

A few members in each of two major groups of pathogenic microorganism (bacteria and viruses) are known to be disseminated by plant pollen. They are amongst the smallest in size and many are also transmitted by seed (Lanier et al., 1976).

1. Bacteria. Bacterial blight of Walnut caused by Xanthornonus campestris pv. juglandis gives rise to a blight of stems, foliage and nuts. It perennates in buds and male catkins giving rise to infected pollen whichcan infect a healthy mothertree through the female flower andcanalsogiverise toinfectednuts (Smithetal., 1988, Lanier et al., 1976; Sinclair et al., 1987).

2. Viruses. Although most viruses become systemic in infected plants, very few become transmitted to the pollen which the plant produces (Mandahar, 1990). Those which have been detected in tree pollen are mainly from the ~ e ~ o v i r u s a n d ~ l a r v h groups and have been detected in the pollen of Ash, Birch, Elm, Hazel, Horse Chestnut, Lilac, Walnut and agricultural trees such as Prunus spp. (Cooper, 1979; Cooper et al., 1984; Cooper et al.,

1988; Mandahar and Gill, 1984; Murant, 1989; Sinclair et al., 1987; Lanier et al., 1976). 3. Insects, nematodes,fungiandmollicutes. Insects, nematodes, fungi and mollicutes have not been reported to be associated with pollen.

Transmission in graft scions and in cuttings

Each of the major groups of pathogenic microorganisms contain many members which can be transmitted in tree shoots used for graft scions or for cuttings (Lanier et al., 1976). However, the larger organisms such as insects and fungi are often readily visible and can thus be easily eliminated from consignments of such materials. Reduction in the size of the samples employed by reducing shoots to buds, or even to meristems, greatly reduces the risks.

1. Insects. There are many insects which can infest plant shoots -and all stages from egg to adult have been found. The smallest insects or youngest stages of development can be very difficult to detect as was the case of the accidental introduction of Pine Woolly Aphid into Central and East Africa (Odera, 1974).

2. Nematodes. Some nematodes which invade the shoots and foliage of plants can be readily transmitted in shoots used for propagation purposes. An example is that of the Pine Wood Nematode (Bursaphelenchusxylophilus), which affects several Pines in Japan, Taiwan, Korea and Hong Kong (Ivory, 1987) and is open to transmission in this way.

3. Fungi. Many fungi whichinvade the shoots and foliage of trees can be readily transmitted in shoots used for propagation purposes. However, most can be readily detected by visual examination andrejected, except in the caseof systemic diseases such as Fusarium Wilt of Mimosa and Sissoo, etc., and of diseases which have extended symptomless incubation periods e.g. Cyclaneusma minus on needles of Pinus spp. (Ivory, 1987).

4. Bacteria. Phytopathogenic bacteria such asX. campestris pv. juglandis can infect all parts of Walnut host trees including the shoots and buds. Use of buds and shoots for propagation purposes can therefore readily disseminate the disease.

5. Mollicutes. These organisms are usually systemic throughout the aerial parts of plants and are thus likely to infect all shoots of infected trees used for propagation. Most mollicutes pathogenic to trees are known to be readily disseminated with graft scions and cuttings (Raychaudhuri and Varma, 1988; Tsai, Chen, Shen and Jin, 1988; Chiykowski, 1988).

6. Viruses. Most plant viruses are present systemically in their host and are thus likely to be present in all shoots used for propagation derived from infected trees. Most tree viruses are known to be readily transmi,tted in graft scions (Cooper, 1988; Uyemoto, 1989; Sinclair, Lyon and Johnson, 1987; Lanieretal.,

30 M.H. Ivory and PB. Tompsett

1976), a characteristic used in tha traditional transmission bioassay procedure.

Transmission in tissue cultures

Most pests and pathogens, ifpresent, become readily apparent in tissue culture dishes because of the favourable environmental conditions under which the cultures are grown. Tissue cultures kept for some time are therefore often regarded as pathogen free unless obviously infected. However, some microorganisms can co-exist with the host cells in tissue cultures and are therefore disseminated to all the derived cultures and to any plants produced from them. Mollicutes and viruses are not easy to observe directly and modem techniques are required for their detection.

1. Fungi. Fungi can contaminate tissuecultures, giving little sign of their presence (Duhem, Mercier andBoxus, 1988), especially those derived from symptomless endophytes. Noexamples have been found for the dissemination of tree pathogens in this manner, however.

2. Bacteria. Bacteria are frequent contaminants of tissue culture (Duhem, MercierandBoxus, 1988; Scortichini, 1988) but, as for fungi, pathogenic species are not known to spread in this manner. They can be endophytic in origin.

3. Mollicutes. If present in the host cells from which the cultures are derived, these organisms will contaminate all the cells and tissues which are subsequently produced (Chiykowski, 1988). However, as seeds are not infected, cultures derived by careful preparation of embryos and other seed tissues will not normally contain mollicutes.

4. Viruses. Many viruses can co-exist with plant cells in tissue cultures which are derived from infected host tissue (Monette, 1988). Tree seeds, however, are mostly virus-free, except for those described earlier, and some cells in apical meristems may also be virus-free; these tissues are relatively safe tissues for culturing, but other tissues are frequently infected.

DISEASE INDEXING IN TREE GERMPLASM

One of the pre-requisites for any scheme to try to prevent the dissemination of diseases in germplasm transfers is to be able to detect and quantify the presence of phytopathogenic organisms in the germplasm. Methods should preferably be rapid, reliable, cheap and simple to operate if such transfers are to become commonplace. Some of the traditional and newer techniques are detailed and discussed below. Diagnostic techniques and problems for fungi, bacteria and viruses have recently been reviewed by Waller (1989), Civerolo (1 989) and Stace-Smith and Martin (1989) respectively.

Direct visualization techniques

1. Light microscopy. The presence of many diseases in plant parts can be detected using various forms of light microscopy to visualize both disease symptoms in the host and signs of the pathogenic organism itself. In many cases the organism can be identified in this way. The procedure is quick and easy, but cannot reliably detect the smallest organisms such as viruses. The technique is particularly suited for the detection of fungi (Wilson, Clement and Kaiser, 1991) and in some cases for mollicutes using darkfield microscopy or fluorescent stains (Markham, 1988; Hiruki, 1988) or forparticular virus inclusions (Christie and Edwardson, 1977).

2. Electron microscopy. The presence of many diseases can be detected in plant partsusing various forms of electron microscopy to visualize both detailed disease symptoms in the host and signs of the pathogenic microorganisms themselves. These procedures can be useful to characterize a disease or its causal organism, but are unsuited for routine screening of the whole plant because of the very small parts which must be examined serially.

Although fungi can be detected by forms of electron microscopy, these techniques are more commonly used for diseases caused by bacteria, e.g. Sumatra disease of Cloves (Bennett, Jones and Hunt, 1987), by mollicutes (Markham, 1988) and especially by viruses (Baker, Rarnsdell and Gillett, 1985; Christie and Edwardson, 1977; Cooper et al., 1984; Roberts, 1986).

3.X-ray radiography. This non-destructive procedure is particularly suited to the examination of small, valuable collections of seeds, and has been used for many years to examine for insect infestation (Kobmoo and Skeates, 1986; Wadhi, 1983; Yates, 1974). For larger batches of seed, direct examination by cutting open may be preferable, however. More recently, attempts have been made to use it for detection of fungi (Vozzo, 198 l ; Kamra, 1974).

Biological and hioassay techniques

1. Incubation on paper. This procedure is particularly suited to the detection of fungi in plant parts and especially in seeds (Lange, 1986). It allows many fungi to grow on the host tissue so that they can be visualized and often identified. It is simple, quite quick, and easily carried out on many samples, and is thus well-suited to routine use for the detection of fungi in and on seeds. When combined with surface disinfection of the seed it can also indicate whether the fungi are superficial or are carried inside the seed.

2. Incubation on agar. Incubation on agar using various partial disinfection procedures combined with the use of antibiotics is particularly suited to the detection of fungi and bacteria in plant

Germplasm health and disease indexing 3 1

parts (Dayan, 1986; Scortichini et al., 1989), and especially in seeds (Lange, 1986). These techniques can also now be used to isolate some fastidious bacteria which require specific conditions for growth (e.g. Sumatra disease of cloves) and some mollicutes, using particular modifications tothe basic procedure (Bennett, Jones and Hunt, 1987; Markham, 1988). Culturing the pathogen axenically in this way enables it to be better visualized and identified. The pathogen can also be subsequently characterized more fully and the degreeof its pathogenicity determined.

3. Transmission bioassay. This procedure has been used for many years to detect the presence of submicroscopic pathogens in plant tissues. It involves the transfer of the microorganism to a range of indicator test plants either mechanically, by grafting, or by a range of types of vector (insects, mites, nematodes, dodder, etc.). The disease can then be characterized by evaluating which test-plants are affected and the symptoms caused on them. The procedure is particularly useful for viruses and mollicutes in plant tissues (Cooper, 1979; Hill, 1984; Markham, -

1988; and Uyemoto, 1989), but is not sensitive enough for use with small seeds, and can take up to two years to complete.

Serological techniques

1 . Simple gel dijjksion tests. The formation of visible precipitin bands between antigens from pathogens, or from diseased tissues (containing the pathogen or substances produced by the pathogen), and specific antisera in gels have been used for many years to detect the presence of viruses (Uyemoto, 1989; Cooper, 1979; Uyemoto, Lowe, Schreader, Gubber and Kirkpatrick, 1989). Antiseraincentral wellson thedishareallowedtodiffuse to meet antigens proceeding from wells around it. The technique can alsobe adapted to detect bacteria (Addy, 1986) or mollicutes (Sinha, 1986; 1988). It is not very sensitive, requiring appreciable quantities of specially prepared extracts, but can be made more sensitive and quantitative by the use of photometers.

2.Agglutinatwn tests. Simple rapidagglutination testsof various types have been used to detect and identify plant viruses (Walkey, Lyons and Taylor, 1992). These have used particles such as chloroplasts, red blood cells, bacteria, latex, etc., to absorb specific polyclonal antibodies or viral antigens. These antibody-particle or antigen-particle combinations then aggluti- nate in the presence of particular viruses in crude sap extracts. The virobacterial agglutination test developed by Walkey et al. (1992) and other agglutination assays are claimed to be more sensitive and simpler to use than transmission bioassays and direct ELISA tests.

3. Immunosorbent assays. Many variations of immunosorbent assay procedures have t p n developed in recent years, with some being tested for a range of practical applications (Kulik, 1984). The procedures vary considerably in sensitivity, complexity and cost; some have become more used than others. Diffusion of antisera and antigens occurs on a sorbent and an enzyme is used to give a colour reaction. The direct enzyme-

linked immunosorbent assay (ELISA) is most widely used in seed health testing (Lange, 1986), but indirect ELISA, dot-blot ELISA and others also have potential for this purpose (Lange, 1986; Barnett, 1986; Clark and Bar-Joseph, 1984; Sinha, 1988; Cooper et al., 1984). These assays are used in the analysis of seeds and vegetative tissues particularly for viruses (Cooper et al., 1984; Maury and Khetarpal, 1 989), but they can also be used for bacteria (Lange, 1986; Mazarei, Hajimorad and Kerr, 1992) and mollicutes (Sinha, 1988; Markham, 1988).

The procedures can be made more sensitive and pathogen specific by the use of monoclonal antibodies. Development of theseantibodies is timeconsuming andexpensive, butthe assays resultingcan be veryusefu1,particularly for fungi, with potential for use in field diagnostic kits (Van Regenmortel, 1986; Bossi and Dewey, 1992; Dewey, 1992; Dewey, MacDonald and Phillips, 1989; Hampton, Ball and De Boer, 1990). Immunosorbent assays can also be made more sensitive by the use of electron microscopy (Immunosorbent Electron Microscopy) (ISEM). This technique involves particle trapping and is very suitable for seed health testing (Lange, 1986); it has been used to detect bacteria, mollicutes and viruses in batches of seed or inside individual insects (Sinha, 1988; Markham, 1988;Milne, 1986; Baker, Ramsdell and Gillett, 1985; Brlanski. Lee, Timmer, Purciful and Rajn, 1992). However, closely- related organisms may not be separable using immunosorbent assays (Sinha, 1988).

Biochemical techniques

1. Specific DNA hybridi;ation. Plant virologists employ varia- tionsof theDNA hybridizationprocedure in which asolidphase. such as nitrocellulose filters, is used (Markham, 1988; Moms, 1990). The most popular procedures include the 'Dot-blot' and 'Spot-blot' techniques and usually use virus-specific DNA probes produced by DNA hybridization (Barnett, 1986). The tests can be extremely sensitive and specific, and can utilize minute amounts of test materials (Markham, 1988). The tests have been utilized to detect mollicutes, viruses and viroids (Uyemotoetul., 1989; Uyemoto, 1989; Markham, 1988; Bamett. 1986; Kirkpatrick, Stenger, Morris and Purcell, 1987; Hull, 1984,1986; Owens and Diener, 1984) as they do not depend on the presence of an external protein coat.

2. PCR assays. These procedures do not employ specific probes and therefore can detect the presence of many microorganisms, including new ones (Barnett, 1986; Dodds, 1986; Dodds, Moms and Jordan, 1984). Non radioactive techniques which depend on the polymerase chain reaction (PCR), such as the random amplified polymorphic DNA assay (RAPD), detect dominant markers giving information on whether they are present or absent. The latter procedure has become widely employed recently to detect and identify a wide range of plants and microorganisms. Radioactive techniques using PCR procedures are also employed in recent applications such as 'tissue blot' systems.

32 M.H. Ivory and P.B. Tompsett

SEED HEALTH TESTING AND PLANT QUARANTINE

Although dispersal of pests and diseases may eventually take place to all parts of the world where their host plants are widely grown (Gregory, 1979) either by natural means or by human agency, the longer the process can be delayed the better (Ivory, 1984). Consequently, most governments have enacted legislation to protect their own important crop plant species against the early introduction of alien pests. This legislation, whether based on a local, a national, or an international level, is now drawn up under the International Plant Protection Convention and is designed to control the movement of all plant materials, including that of forest trees, together with their packing, and thus the pests and diseases contained within or attached to them (Kahn, 1977). Regulations controlling the movement of forest tree germplasm between regions, or countries, usually contain a prohibitionofplants withattached soil, and requireall vegetative materials to be transferred through quarantine stations or under permit with specified inspections and treatments. Tree seed can often be imported without difficulty following inspection and treatment with pesticides. Tissue cultures are not always disease free despite the fact that disease is usually apparent quickly (Kahn, 1979); they are thus treated as vegetative materials and can be subject to quarantine for detection of viruses and other latent systemic pests (Ivory, 1984).

Plant health regulations usually entail inspection and certifi- cation at the point of export, and similar inspection and possibly quarantine at the point of import (Raychaudhuri and Verma, 1990) with the major onus falling on the importing authorities (Hewitt and Chiarappa, 1977; Anon, 1983). These measures, or their implementation, are often inadequate to detect all pests in vegetative or seed materials, particularly when the transfers are carried out on a commercial scale (Neergard, 1977; Kamra, 1990). Transfers of smaller quantities of germplasm for scientific purposes can also pose dangers, but safeguards can be employed to counter these risks (Singh and Rana, 1990; Anon 1991, 1989; Diekmann, 1988; Fassil-Kibebew, 1987; Kahn, 1983).

The development of rapid, reliable pest detection systems should enable inspections to be more frequent and thorough, both at the point of origin (to detect most organisms whichmight be caiied with the germplasm) and at the point of entry (to provide efficient detection of known dangerous quarantine organisms).

ACKNOWLEDGEMENTS

PBT acknowledges assistance from the Australian Centre for International Agricultural Research, and the Overseas Develop- ment Administration UK, during preparation of material in this paper, some of which was prepared as part of planning for programmes and centres of the Consultative Group on Interna- tional Agricultural Research.

REFERENCES

ADDY, S.K., 1986. Serology in plant bacteriology. In VARMA, A. and VERMA, JP. (eds.). Vistas in plantpathology. Malhotra Publish- ing House, New Delhi, pp. 491-502.

ANDERSON, R.L., 1986. Checklist of microorganisms associated with tree seeds in the world, 1985. Gen. Tech. Rep. no. SE-39, USDA Forest Service.

ANONTMOUS, 1983. Exotic plant quarantine pests and procedures for introduction ofplant materials. ASEAN Plant Quarantine Centre and Training Institute, Serdang, Malaysia.

ANONYMOUS, 1989. New FAOWPGR technical guidelines for safe movement of cocoa, edible aroids, legumes, Musa, sweet potato and yam germplasm. FAO Plant Protection Bulletin, 37, 182.

ANONYMO~S, 1991. New FAOOBPGR technical guidelines for safe movement of cassava, citrus, grapevine and Venilla germplasm. FAO Plant Protection Bulletin, 39, 189.

BAKER, K.K., RAMSDELL, D.C. AND GLLLETI., J.M., 1985. Electron microscopy: current applications to plant virology. Plant Dis- ease, 69: 85-90.

BARNETT, O.W., 1986. Application of new test procedures to surveys: merging thenew with theold. In JONES, R.A.C. ~ ~ ~ T O R R A N C E , L. (eds.) Developments and applications in virus testing. Assoc. of Appl. Biol., Wellesboume, pp. 247-267.

B E N N ~ , C.P.A., JONES, P. and Hmt, P., 1987. Isolation, culture and ultrastructure of a xylemlimited bacteriumassociatedwith Surnatra disease of cloves. Plant Pathology, 36: 45-52.

Bos, L., 1977. Seedborne viruses. In H~wnt , W.B. and CHIARAPPA, L. (eds.) Plant health and quarantine in international transfer of genetic resources. CRC press, Cleveland.

Bow, R. and DEWEY, F.M., 1992. Development of a monoclonal antibody-based immunodetection assay for Botrytis cinerea. Plant Pathologj, 41: 472-482.

BRLANSKI, R.H., LEE, R.F., T ~ R , L.W., PURCIFUL, D.E. and RNN, B.C., 1982. Immunofluorescent detection of xylem-limited bacteria in situ. Phytopathology, 72: 1444-1448.

CHIXOWSKI, L.N., 1988. Maintenanceofyellows-typernycoplasmalike organisms. InHiruki, C. (4 . ) Treemycoplasmaandmycoplasrnas diseases. Univ of Alberta Press, pp. 123-134.

CHRISTIE, R.G. and EDWARDSON, J.R., 1977. Light and electron microscopy of plant virus inclusions. ~ l o h d a Agricultural Experiment Station Monograph Series no. 9, University of Florida.

CIVEROLO, E.L., 1989. Plant quarantinediagnostic problems; bacteria. In KAHN, R.P. (ed.) Plantprotectionandquarantine Vol. II. CRC Press Inc, Boca Raton, pp. 205-218.

CLARK, M.F. and BAR-JOSEPH, M., 1984. Enzyme immunosorbent assays in plantvirology. I~MARAMOROSCH, K. ~ ~ ~ K O P R O W S K I , H., (eds.) Methods in virology. vol. 7. Associated Press, New York, pp. 51-85.

COOPER, J.I., 1979. Virus diseases of trees and shrubs. ITE, Cam- bridge, 74 pp.

COOPER, J.I., 1988. Ecology of viruses in ornamental and forest trees. Acta Horticulturae 234: 359-364.

COOPER, J.I., KELLEY, S.E. and MASSALSK~, P.R., 1988. Virus-pollen interactions. In HARRIS, K.F. (ed.) Advances in disease vector research 5, Springer Verlag, pp. 221-249.

COOPER, J.I., MASSALSKI, P.R. andEDWm, M-L., 1984. Cherry leaf roll virus in the female gametophyte and seed of birch and its relevance to vertical virus transmissions. Annals of Applied Biology, 105(1): 55-64.

Germplasm health and disease indexing 33

DAYAN, M.P., 1986. Fungi associated with different forest ttee seeds of the Forest Research Institute seedbank. Embryon, 2(1): 28-39.

DEWEY, F.M., 1992. Detection of plant invading fungi by monoclonal antibodies. In DUNCAN, J.M. and TORRANCE, L. (eds.) Techniques for the rapid detection ofplant pathogens. Blackwell Scientific Publs., Oxford, pp. 47-62.

DEWEY, F.M., MACLXINALD, M.M. and PHILLIPS, S.I., 1989. Develop- ment of monociond-antibody - ELISA, DOT-BLOT and DIP-Snc~ immunoassays for Humicola lanuginosa in rice. Journal of General Microbiology, 135: 361-374.

DIEKMANN, M,, 1988. Seed health testing and treatment of germplasm at the International Centre for Agricultural Research in dry areas (ICARDA). Seed Science and Technology, 1q2): 405-41 7.

DODDS, J.A., 1986. The potential for using double-stranded RNAs in diagnostic probes for plant viruses. In Jones, R.A.C. and Torrance, L. (eds.) Developments and applications in virus testing. Assoc. of Appl. Biol., Wellesbourne, pp. 71-86.

DODDS, J. A., MORRIS, T.J. and JORDAN, R.L., 1984. Plant viral double- stranded RNA. Annual Review Phytopathology, 22: 15 1 - 168.

DUHEM, K., MERCIER, N. LE, and Boxus, P., 1988. Difficulties in the establishment of axenic in vitro cultures of field collected coffee and cacao germplasm. Acta Horticulturae, 225: 67-75.

EUNGWIJARNPANYA, S. and HEDLIN, A.F., 1984. Studies on seed insects of some forest trees. Embryon, 1: 49-56.

FASSIL-~BEBEW, 1987. Establishment of a seed health testing laboratory andquarantine systemat PGRCE. PGRCiE-ILCA-Germplasm Newsletter, 15: 3-8.

GILL, D.L., 1968. Mimosa-wilt Fusarium carried in seed. Plant Disease Reporter, 52: 949-95 1.

GREGORY, P.H., 1979. Movement of diseases between neighbouring states: some South American examples. In EBBELS, D.L. and KING, J.E. (eds.) Plant health. Blackwell Scientific Pubs., Oxford, pp. 269-273.

HAMPTON, R., BALL, E. and DEBOER, S. 1990. Serologicalmethods for detection and identification of viral and bacterial plant pathogens: a laboratory manual. American Phytopathological Society Press.

HEWITT, W.B. and CHIARAPPA, L., 1977. (eds.) Plant health and quarantine in international transjiir of genetic resources. CRC Press Inc.

HILL, S.A., 1984. Methods in plant virology. Blackwell Scientific Publications, Oxford.

HIRUKI, C., 1988. Immunofluorescence microscopy of yellows diseases associated with plant mycoplasma like organisms. In HIRUKI, C. (d . ) Tree mycoplasmas and mycoplasma diseases. Univ. of Alberta Press, pp. 193-203.

HULL, R., 1984. Rapid diagnosis of plant virus infections by spot hybridization. Trends in Biotechnology, 2: 88-9 1.

HULL, R., 1986. The potential for using dot-blot hybridization in the detection of plant viruses. In Jones, R.A.C. andTorrance, L. (eds.) Developments and applications in virus testing. Assoc. of Appl. Biol., Wellesbourne, pp. 3-12.

IvoRY,M.H., 1984. Planthealthlegislationandforesttrees. ~ B U R L E Y , J. and von CARLQW~TZ, P. (eds.) Multipurpose tree germplasm. ICRAF, Nairobi, pp. 241-248.

IVORY, M.H., 1987. Diseases and disorders of pines in the tropics. Overseas Research Pubfieation no. 31, ODA, London.

KAHN, R.P., 1977. Rant quarantine principles, methodology and suggested approaches. In HEWIIT, W.B. and CHIARAPPA, L. (4s . ) Plant healrh and quarantine in intermional transfer of genetic resources. CRC Press h . , Cleveland, 289-307.

KAHN, R.P., 1V9. Tissueculnue applications forplant quarantine. In Pmctical tissue c~lture applications. Academic Press Inc., 185- 201.

KAHN, R.P., 1983. Safeguarding the international exchange of plant germplasm. Proceedings 10th International Congress of Plant Protection, Brighton, Vol. 2; 866-872.

KAMRA, S.K., 1974. X-ray radiography of tropical forestry seed. Proceedings seed X-ray symposium, Macon, USA, 1-20.

KAMRA, S.K., 1990. Improving the forest seed situation in some African countries. In Turnbull, J.W. (ed.) Tropical tree seed research, ACIAR Proceedings Series no. 28, 126- 13 1.

KIRKPATRICK. B C , STENGER, BC., MORRIS, T.J. and PURCELL, A.H., 1987. Cloning and detection of DNA from a non-culturable plant pathogenic mycoplasma-like organism. Science, 238: 197-200.

KOBMOO, B. and SKEATES, D.A., (1986). X-radiography of tropical forest tree seed. Embryon, 2: 49-55.

KUI.IK, M.M., 1984. New techniques for the detection of seed borne pathogenic viruses, viroids, bacteria and fungi. Seed Science and Technology, 12: 83 1-840.

LANGE, L., 1986. The practical application of new developments in test procedures for the detection of viruses in seed. In JONES, R.A.C. and TORRANCE, L. (eds.) Developments and applications in virus

. testing. Assoc. of Appl. Biol., Wellesbourne, pp. 269-281. LANIER, L., JOLY, P,, BONDOUX, P, and BELLEMERE, A., 1976. Mycologie

et pathologie forestihs 11. Patholo,gie forestier. Masson, Paris. LENNE J.M., 1991. Diseases oT Leucuena species. Tropical Pest

Management, 37: 28 1-289. MANDAHAR, C.L., 1990. Virus trancmission. In MANDAHAR, C.L. (ed.)

Plant viruses vol.11: Patkolog.~. CRC Press Inc., Boca Raton, pp. 205-253.

MANDAHAR, C.L. and GILL. P.S., 1984. The epidemiological role of pollentranmii^-innot viruses. Zritschr~~furPJlan:enkrunkheiten und Pflanzenschuc. 9 1: 146-249.

MARKHAM, P.G., 1988. Detection of mycoplasmas and spiroplasmas ininsects. In HIRUKI C. (ed.) Ti.eemycoplasmsandmyc~oplasr~~a diseases. Univ. of Alberta Press, Canada, pp. 157-177.

MATHUR, V.K.andLAL, A., 1989. A simple technique forthe detection of white tip nematode (Aphelencoides hesseyi) in rice germplasm under exchange. Indian Journal of Nematology. 19: 71.

MAURY, Y. and KHETARPAL, R.K., 1989. Testing seed for viruses using ELISA. In AGRIHOTRI, V.P., SINGH, N., CHAUBE, H.S., SINGH, U.S. and DWIVEDI, T.S. (eds.) Perspectives in ph.~topathology. Today and Tomorrow's Printers and Publishers, New Delhi, pp. 3 1-49.

MAZAREI, M. km HAJIMORAD, M. and KERR, A., 1992. Specificity of polyclonal antibodies to different antigenic preparations of Pseudomonas syringae p.v. pisi strain UQM551 and P.S.p.v. syringae strain L. Plant Pathology, 41: 437-443.

MILNE, R.G., 1986. New developments in electron microscope serology and their possible applications. In JONES, R.A.C. and TORRANCE, L. (eds.) Developments and applications in virus testing. Assoc. of Appl. Biol., Wellesbourne, pp. 179-191.

MITTAL, R.K. and WANG, B.S.P., 1987. Fungi associated with seeds of eastern white pine and white spruceduringcone processing and seed extraction. Canadian Journal Forest Research, 17: 1026- 1034.

MONETIE, P.L., 1988. Grapevines. I ~ B N A J , Y.P.S. (d.) Biotechnology in agriculture and forestry no. 6: Crops 11. Springer Verlag, Berlin.

MORRIS, P.C., 1990. Biotechnology and plant protection. FAO Plant Protection Bulletin, 38: 25-37.

MURANT, A.F., 1989. Nepoviruses. In KAHN, R.P. (d . ) Plant protection andquarantine. Vol 11. CRC Press Inc., Boca Raton, pp. 43-57.

NEERGARD, P., 1969. Seed-borne diseases in tree seed. Dansk Skovforen Tidsskrift, 54: 241-260.

NEERGARD, P., 1977. Quarantine policy for seed in transfer of genetic resources. In Hewitt, W.B. and Chiarappa, L. (eds.) Plant health

34 MH. Ivory and P.B. Tompsett

and quarantine in international transfer of genetic resources. CRC Press Inc., Cleveland, pp. 309-314.

NOBLE, M. and RICHARDSON, M.J., 1968. An annotated list of seedborne disease (2nd edit.) Phytopathological Paper no. 8, CMI.

ODERA, J.A., 1974. The incidence and host trees of the pine woolly aphid, Pineus pini (L.), in East Africa. Commonwealth Forestry Review 53(2): 128-136.

OWNS, R.A. and DIENER,T.O., 1984. Spot hybridization for detection of viroids and viruses. In MARAMOROSCH, K. and KOPROWSKI, H. (eds.) Methods in virology 7. Academic Press, New York, pp. 173-187.

RAYCHAUDHURI, S.P. and VARMA, A., 1988. Sandal spike, the present situation. In HIRUKI, C. (ed.) Treemycoplasmasandmycoplasma disease. Univ of Alberta Press, Canada, pp. 33-35.

RAYCHALIDHURI, S.P. and VERMA, J.P., 1990. Review of tropicalplant pathology Vol. 6 - Techniques and plant quarantine. Today and Tomorrows Printers and Publishers, New Delhi, 344 pp.

RICHARDSON, M.J., 1979. An annotated list of seedborne diseases (3rd Edn). International Testing Association, Zurich.

ROBERTS, I.M., 1986. Practical aspects of handling, preparing and staining samples containing plant virus particles for electron microscopy. In JONES, R.A.C. and TORRANCE, L. (eds.) Develop- ment and applications in virus testing. Assoc. of Appl. Biol., Wellesbourne, pp. 213-243.

SCORTICHINI, M,, 1988. Bacterial contaminationin plant tissuecultures in vitro. Informatore Fitopatologico, 38: 37-48.

SCORTICHINI, M., ROSSI, M.P., RICCI, B. and NDZOUMBA, B., 1989. Soyabean seed decay associated with Bacillus subtilis in Gabon. FAO Plant Protection Bulletin, 37: 87-91.

SEN-SARMA, P.K., THAKUR, M.L. and SEHGAL, H.S., 1988. Protection of forest seeds against insect pests and fungi during storage. Journal Tropical Forestry, 4: 350-356.

S~NCLAIR, W.A., LYON, H.H. and Jomso~ , W.T., 1987. Diseases of trees and shrubs. Comstock Publishing Association, Cornell University Press.

SINGH, B.P. and RANA, R.S., 1990. Germplasm exchange of multipurpose trees: an Indian perspective. In TURNBULL, J.W. (ed.) Tropical tree seed research. ACIAR Proceeding Series no. 28; 121-125.

SINHA, R.C., 1986. Serodiagnosis of diseases caused by mycoplasma like organisms. In VARMA, A. and VERMA, J.P. (eds.) Vistas in plantpathology. MalhotraPubl. House, New Delhi, pp. 513-524.

SINHA, R.C.. 1988. Serological detection of mycoplasma-like organisms in plants. In HIRUKI C. (ed.) Tree mycoplasmas and mycoplasma diseases. Univ. of Alberta Press, Canada, pp. 143- 156.

SMITH, I.M., ARCHER, S.A., DUNEZ, J.. LELLIOTT, R. and ~ I L L I P S , D.H., 1988. (eds.) European handbook of plant diseases. Blackwell Scientific Publications, Oxford.

S O L ~ G A T E , B.J., 1983. Handbook on seed insects of Acacia species. FAO, Rome.

STACE-SMITH, R. and MARTIN, R.R., 1989. Plant quarantine diagnostic problems: viruses. In KAHN, R.P. (ed.) Plant protection and quarantine Voll l . CRC Press Inc., Boca Raton, pp. 183-203.

TSAI, J.H., CHEX, Z.Y., SHEN, C.Y. and Jm, K.X., 1988. Mycoplasmas and fastidious vascular prokaryotes associated with tree diseases in China. In HIRUKI, C. (ed.) Tree mycoplasmas andmycoplasma diseases. Univ. of Alberta, Canada, pp. 69-97.

UYEMOTO, J.K., 1989. Union abberration of sweet cheny on Prunus mahaleb rootstock associated with X-disease. Plant Disease, 73: 899-902.

UYEMOTO, J.K., h, S.K., SCHREADER, W.R., GUBBER, W.D. and KIRKPATRICK, BC., 1989. New diagnosis and incidence of diseases associated with a decline of cheny trees in California

orchards. New Zealand Journal Crop and Horticultuml Science, 17: 265-270.

VAN RFGENMORTEL, M.H.V., 1986. The potential f o r u s i n g m o n a l antibodies in the detection of plant vimses. In J o m , R.A.C. and TORRANCE, L. (eds.) Developments and applications in virus testing. Assoc. of Appl. Biol., Wellesbourne, pp. 89-101.

Vozzo, J.H., 1981. Xeroradiography for seed research. In X-ray analysis of tropical forest tree seed. Publication Especial no. 35, Institute Nacional Investigaciones Forestales, Mexico, 299-306.

WADHI, S.R., 1983. X-ray radiography in plant quarantine. Journal Nuclear Agriculture and Biology, 12: 87-89.

WALKEY, D.G.A., LYONS, N.F. and T A ~ , J.D., 1992. An evaluation of a virobacterial agglutination test for the detection of plant viruses. Plant Pathology, 41: 462-47 1.

WALLER, J.M., 1989. Plant quarantine diagnostic problems: fungi. In KAHN, R.P. (d.) Plant protection and quarantine Vol 11. CRC Press Inc., Boca Raton, pp. 249-256.

WILSON, A.D., CLEMENT, S.L. and KAISER, W.J., 1991. Survey and detection of endophytic fungi in Lolium germplasm by direct staining and aphid assays. Plant Disease, 75: 169-173.

YATES, H.O., 1974. Radiography for detection and study of insects in plant seeds. Proceedings d X-ray symposium, Macon, 65-78.

YUAN, Z.Q., OLD, K.M. and MLDGLEY, S.J., 1990. Investigation of - mycoflora and pathology of fungi present on stored seeds of Australian trees. In TURNBLTLL, J.W. (ed.) Tropical tree seed research. ACIAR Proceedings series 28.

Quality and Performance You Can Count On!

. . . software from

I Forestry 4 Logging 4 Forest Products 4

Data Collection Product Inventory

Business Operations

#3 - 1441 Store Street Victoria, BC, Canada V8W 3J6

FAX: (604) 361 -9307

Commonwealth Forestry Review Volume 73(1), 1994 35

A four-phase sampling method for assessing standing volume using Landsat-TM-data, aerial photography and field assessments MICHAEL K ~ H L ' S.P.S. KUSHWAHA2 l National Forest Inventory Department, Swiss Federal Institute for Forest, Snow and Landscape Research, CH-8903 Birmensdorf, Switzerland

National Remote Sensing Agency, Hyderabad 500037, India

SUMMARY

The paper presents a method for the assessment of standing volume in extensive forest inventories. Information obtained from Landsat-TM-data, Colour Infra-Red (CIR) and black and white aerial photography and field assessments are combined using a four-phase sampling for stratification approach. The statistical and operational aspects of a four-phase inventory are shown with special reference to the determination of optimal stratum boundaries. The method was applied in a pilot survey in Kamataka, India.

Keywords: Landsat-TM, aerial photography, forest survey, four-phase sampling, stratification.

INTRODUCTION

The history of forest remote sensing in Indiais thirty years old. The first aerial photographs for forestry purposes were acquired in 1963 for Kulu Valley in Himachal Pradesh. Pre-Investment Survey of Forest Resources (renamed in 1980 as Forest Survey of India) was the first agency to use aerial photographs for forest resources inventory. Applications of satellite imagery in forestry dates back to 1975 when the first images from EROS Data Centre were acquired by the National Remote Sensing Agency (NRSA), the national level apex remote sensing organization of the Government of India. Forest Survey of India now regularly uses satellite imagery for nationwide forest monitoring biennially, and for the areas undergoing rapid changes, annually. The first satellite based nationwide forest cover mapping was accomplished by NRSA in 1983 (NRSA, 1983). The decade between 1980 and 1990 was dominated by satellite remote sensing for forest resources assessment and monitoring, wildlife habitats evaluation and fire damage assessment (Kushwaha, 1987,1990; Unni, 1990). Another milestone in the forest remote sensing in India was the Indo-FRG Cooperation Programme (from 1976 onwards) between NRSA and the German Society for Technical Co- operation (GTZ), Eschborn for development and operationalization of remote sensing in India. The forestry component of this programme during Phase 1, 2 and 3 was primarily focussed on the forest resources inventory, monitoring and timber volume estimation in close collaboration with the Photogrammetry and Remote Sensing Department of the University of Freiburg and Forestry and Ecology Division of NRSA. This paper gives operational and statistical details of a pilot timber volume inventdry carried out in Western Ghats, India during Phase 3 of the cooperation programme. The objective was to provide a statistically sound, large-scale, quantitative inventory of standing timber volume. Keeping in mind the digital image and aerial photo interpretation facilities

at NRSA, field survey capabilities of Karnataka Forest Department and the time and cost factors, it was decided to go for a four-phase inventory using Landsat Thematic Mapper imagery, B/W 1 :20,000, Colour infrared (CIR) 1 : 10,000 scale aerial photographs and terrestrial survey.

STATISTICAL CONCEPTS

Various statistical concepts can be employed to combine data from remote sensing with those from field surveys. Multi- stage sampling designs are based on dividing the population into sub-populations. Each element of the overall population can be clearly allocated to a particular sub-population. Theoretically, the population can be divided into as many stages as desired, i.e., the first stages (primary units) can be subdivided into secondary ones, the secondary ones into tertiary ones and so forth. The variables of interest, e.g., standing volume, are then determined from data at the lowest level.

Different selection methods may be applied at each stage: random sampling, systematic sampling, sampling with probability proportional to size, etc. Consequently, there is a great deal of freedom in sampling design. Nevertheless, the selection method must be taken into account in computing estimators, which means that there is no universally valid algorithm for the analysis of multi-stage sampling data. This problem has beendiscussed by Bowdenet al. (1 979), Johnston (1982), and particularly thoroughly by Saborowski (1990).

Multi-phase sampling designs, on the other hand, analyse information from every level. The data collected in the n-l phases serve as auxiliary variables, while the variables of interest are derived from data from the lowest (nth) level. The design must be such that the auxiliary variables are less expensive to

36 M. Kohl and S.P.S. Kushwaha

derive than the variables of interest. Remote sensing is the ideal solution for such purposes.

There are two types of multi-phase sampling designs. In the first type, the relationship between the auxiliary variables and the variables of interest is described by means of regressions. In this case, the design is termed multi-phase sampling with regression estimators. This approach has repeatedly proved its worth in temperate latitudes. Kohl and Sutter (1991) give an overview. Within the context of this paper those interested in more details are referred to publications on large-scale inventories employing satellite data and studies suggesting their potential application (Langley etal. 1969; Langley 1975; Schade 1980; LaBau and Schreuder 1983; Forstreuter and Kohl 1987). The concept of regression estimators requires that both the auxiliary variables and the variables of interest be derived from the same object. For forest inventories, this means that the sample plots for all the different phases must coincide absolutely. In areas with homogeneous forest growth and few features for orientation, this is a difficult condition to fulfil. In the planning of this particular inventory it was not possible to determine whether the correlation between the two sets of variables was high enough to guarantee good cost-efficiency. Together with the doubtful overlap of the sampling plots, this raised the question whether an alternative multi-phase sampling design should be employed.

In the second type. the auxiliary variable does not depend on the measurements, but is purely an indicator variable showing the stratum to which the variable of interest is to be allocated. This type is termed multi-phase sampling for stratification. Volumes determined through field surveys in the last phase are weighted according to the estimation of strata sizes. This forms the basis for the computation of the total and mean standing volumes, and their respective standard errors. The procedure has the advantage that the sample plots donot have to match exactly, as long as each particular stratum is accurately surveyed. In homogenous forest stands with difficult access it has decisive advantages.

In two-phase sampling for stratification, the entire population is regarded as consisting of N sampling units. Suppose there are n samples (randomly) selected, which can be allocated to any of the I strata of Phase 1. The I Phase 1 stratum comprises ni(i = 1,2, ... I) sampling units. Subsequent samples are then selected from each stratum.

The m, Phase 2 units are a subsample from the ni units that happen to lie in stratum I.

In the second sampling phase the varfiables of interest yij (j= 1,2 ..m,) are recorded. The mean 7,. and the variance G(;,,) are lomputed as follows:

where

m,

S: = 1 lYi, - T i p mi-l ,=,

The overall values for the inventory area are computed from

A

Y.. = N?..

In four-phase sampling desigs for stratification the mean F... and the variance of the mean V @....) are computed correspond- ingly (see Appendix).

DATA COLLECTION

The pilot inventory was conducted for a 900 km2 area in the Westem Ghats hill range falling in Kamataka state of India (see Figure I). The test area is very dissected and lies between 1 l0 to 12O north and 75" to 76O east at altitudes ranging from 400 m to 1800 m a d . As a result of spatial variations in the mean annual rainfall and its seasonal distribution, the main dry season occurring between January and March, various types of forest have developed. The dry monsoon forests on the eastern side of the range merge into the wet evergreen forests of the western side at the frontier between the states of Kamataka and Kerala.

L O - C A T I O N M A P

STUDY AREA I FIGURE 1. Study urea

A four-phase sampling method for assessing standing volume 37

The field surveys were planned and executed in close collaboration with the Kamataka Forest Department and the NRSA. The overall scheme of field plots selection is shown in Figure 2. In Phase 1 and Phase 2, nine systematically distributed points within one interpretation area were selected, and the stratum to which each belonged determined. In Phase 1, such allocation was based on individual pixels.

Sample Assessments selection

Phase 1

(Landsat)

Phase 2

(blw aerial photos 1:20,000)

Phase 3

(CIR aerial photos 1 : 10,000)

Phase 4 (field samples)

forest types non forest area evergreen forest semi evergreen forest moist deciduous forest dry deciduous forest teak plantation coffee plantration

density classes 0 - 0.2 0.21 - 0.4 0.41 - 0.6 0.61 - 0.8 0.81 - 1.0

height no. of stems crown diameter

measurements dbh upper diameter height tree species or species group

FIGURE 2. Sample selection in the four phases

TABLE 1. Frequency distribution of the Landsat classification (Phase I )

Phase 1 (Landsat) frequency %

non forest area 363 40.3

evergreen forest 95 10.5

semi-evergreen forest ' 44 4.9

moist deciduous forest 33 3.7

dry deciduous forest 166 18.4

teak plantation 31 3.4

coffee plantation 169 18.8

In the first phase the sampling points were allocated to classes based on Landsat data. It was possible to distinguish seven classes altogether: wet evergreen forest, semi-evergreen forest, moist deciduous forest, dry deciduous forest, teak plantations, coffee plantations and non-forest. The estimates of Landsat strata were based on a manual combination of digital and visual methods. The distribution of the various Landsat classes can be seen from Table 1. Some 40% of the samples counted as non-forest, dry monsoon forest and coffee plantations occupy 18% each, moist evergreen forest 10%, while 12% of the area was occupied by moist semi-deciduous forest, wet monsoon forest, and teak plantations.

TABLE 2. Frequency distribution Phase 2 Strata

Phase 2 frequency % (density classes)

non forest area 361 40.1

0 - 0.2 78 8.7

0.2 - 0.4 1 1 1 12.3

0.4 - 0.6 90 21.1

0.6 - 0.8 70 7.8

0.8 - 1.0 91 10.1

Stratification in the second phase was based on b/w aerial photos at a scale of 1:20 000. The stand density for all sample points was assessed at 20% intervals. The frequencies are shown in Table 2.

In the third phase, measurements were made on CIR photos on 5 of the 9 sample points. Tree height, tree number, and crown diameter were determined through parallax measurements or tallies. These data were used as a basis for strata. In addition, the correlation between the standing volume as determined from field surveys and from aerial photos was computed by means of regression equations. For most of the combinations of variables no significant regression model could be derived - a fact which speaks against the application of multi-phase sampling with regression estimators.

The aerial photography for Phases 2 and 3 was organized and executed by the NRSA, Hyderabad. As it was not possible to obtain aerial photographs of the entire study area, the latter was divided into comdors, from which a random selection was made. The units of interpretation in the first phase (Landsat classification') were selected exclusively from the chosen comdors. The problems inderiving estimators arising through the selection of corridors are discussedin detail by Saborowski (1990). As only the pilot inventory was subject to such restrictions, it was unnecessary to adapt the algorithms for the derivation of estimators in the final inventory.

Phase 4 comprised terrestrial measurements. All trees with adiameter above 12cm in the sampling plots, 20m X 20m, were surveyed. The parameters recorded were upper diameter, DBH, and tree height, as measured with a telerelascope. The single tree volume was computed by means of spline functions using SITCA Program developed in the Department of Forest

38 M . Kohl and S.P.S. Kushwaha

Biometry, University of Freiburg (Kleinn, 1989). The distribution of the sampling plots was made according to operational criteria. Altogether, only possible data from 82 sample plots could be included in the final analysis. This is obviously too small a sample volume for reliable, detailed conclusions on standing timber and species occurring in the inventory area. It did, however, suffice to resolve the question of a methodologically feasible inventory concept for this venture. The volumes observed lay between 0.6 and 1795 m3 per hectare, the mean being 165 m3. As the organizers of the field surveys decided to exclude trees with very irregular stem shapes, it must be assumed that the figures for standing volume, at least in the stratum dry monsoon forest, are somewhat underestimated. The statistical values for volume estimation in Phase 1 and Phase 2 strata are shown in Table 3.

TABLE 3. Frequency distribution

volume (m31ha)

n mean min. max standard error

Phase I Landsat-classes

non forest area

evergreen forest

semi evergreen forest

moist deciduous forest

dry deciduous forest teak plantation

coffee plantation

total

Phase 2 density-classes

without statement

0 - 0.2

0.2 - 0.4

0.4 - 0.6

0.6 - 0.8

0.8 - 1.0

total

As a result of the low number of terrestrial sample plots - only 82 - not all the combinations of strata were sufficiently covered. Consequently, some strata had to be merged to obtain larger sample sizes.

CONSTRUCTION OF THE STRATA

Where I strata are to be constructed, it is fairly easy - as regards minimizing variance -to determine optimum points for stratification, as long as the stratifying variables are identical with the estimator variables. In multi-phase inventories the auxiliary variables computed in the aerial photo/Landsat phase must be included in the stratification, which complicates the problem of selecting optimal stratum boundaries. According to Lachan (1985), the 'cum f' rule proposed by Dalenius and Hodges (1957) can be used for the auxiliary variable X as long as the correlation between X and y, the variable of interest, is sufficiently stringent.

The 'cum df' rule smooths the df values over all strata.

b, = stratum boundary

f(t) = density of probability.

The first step was therefore tocompute thecorrelation between the variable of interest (volumes per m3/ha) and the auxiliary variables mean tree height, number of trees, and mean crown diameter for each sample plot. The normal distribution of the variables was verified graphically and through application of the Shapiro-Wilk test (Shapiro and Wilk, 1965), while the significance of the coefficient of correlation was checked with the F-test.

Correlation coefficients were calculated for the separate forest strata and the entirety of the sample plots. In the case of the variable 'tree number', no significant correlation with that of 'standing volume' was found. 'Mean crown diameter' showed only a slightly significant correlation coefficient of 0.39. These two variables were thus discarded as auxiliary variables for stratification. Significant correlations with standing volume were found not only for 'tree height' and the collapsed data for the 79 sample plots but also for the strata wet evergreen forest, wet monsoon forest, and dry monsoon forest. Here, the correlation coefficient was always above 0.6. Nevertheless the small number of observations dictate great caution in interpreting the results and extrapolating the range of data.

Where stratification through proportional allocation is to be conducted, the optimum boundaries based on the auxiliary variable X can be computed via an iterative procedure, first presented by Dalenius (1950). The procedure begins with the random boundaries b, , . . .b,, . Where

a smaller value for b, is selected for the next iteration, and, vice versa, where

a greater one. This procedure is continued until the equation


is approximated. If the fnquency distribution of the auxiliary variables is apgiitd, this condition is fulfilled for the boundaries b, = 10.5, b2 =: 13.6, b, = 16.5, and b, = 20.1.Comparison with Table 4 shows that the difference between the strata means and the boundaries averages at 0.3. According to Cochran (1977), there should be between 5 and 10 strata In this inventory, because of the low number of te~~estrial sampling plots, the number of strata in Phase 3 was set at K = 5.

(N) within the inventory area. The forests within that area coveredsome 150,000ha. With a pixel size of 3Ox3Om, that gave about 1.67 million pixels. In estimating the total standing standing volume and its variance it must be borne in mind that N can only be roughly estimated.

TABLE 5. Distribution of thefield plots in the strata combinations

TABLE 4. Results of strata construction Phase 3

tree height (m) stratum Phase 3

n mean

I 24 9.03

2 26 12.33

3 37 15.19

4 3 1 18.18

5 23 22.43

total 141 15.45

The boundaries thus determined are used to fix the strataof Phase 3. For forest inventories, it is suggested that after the procedures described above a post-stratification be conducted, it cannot be assumed that the pilot inventory data suffice for the description of the whole range of variation in the volume/tree height relationship. Boundaries determined from such a small database do not necessarily provide an optimum scheme for stratification.

EVALUATION OF THE PILOT INVENTORY

There are 150 possible combinations for linking the I=6 Phase 1 strata, the J=5 Phase 2 strata, and the K=5 Phase 3 strata; about one third of these (n=51) rely on field survey data collected during Phase 4.34 of these 5 1 combinations are only covered by one single observation (see Table 5). Consequently, it was necessary to combine individual strata in evaluating the results of the pilot inventory.

As a first step, the Phase 2 strata were collapsed, with the result that the number of combinations sank to 45.25 of these were covered by Phase 4 sample plots, so that additional Phase 3 strata had to be summarized.

Finally, 28 combinations were available for analysis. Half of these were covered by two terrestrial sample&ts each able 6). This coverage, though low, was accepted for the evaluation of the pilot inventory, so as to avoid further collapsing of the strataand aconcomitant reduction of the possiblecombinations. That would have meant that individual phases were compiled to give a single stratum, andcdnsequently that some of them would no longer be considered. Given the low coverage, a realistic estimate of the range of variation is not feasible. The non-forest class is not considered in further analyses, as there are no volume observations in Rase 4.

Mean and total volume with their respective variances were deri~edhrnthed~orihsmentionedabove. ~orthesecom~u- tations it was necessary to to estimate the number of forest pixels

Phase l Phase 2 Phase 3 frequency

evergreen forest 0.2-0.4 0.6-0.8

Semievergreen forest 0.4-0.6

0.6-0.8

0.8- 1 .O

moist deciduous forest 0.2-0.4

dry deciduous forest

&

teak plantation

coffee plantation

0.4-0.6 0.6-0.8

0.8- 1 .O

0-0.2 0.2-0.4

0.4-0.6

0.6-0.8

0.8- 1 .0

0.4-0.6

0.6-0.8

0.8-1 .O

without statement 0-0.2

0.2-0.4

0.4-0.6

0.6-0.8


TABLE 6. Distribution of thefield plots according to strata combinations after collapsing

Phase i Phase 2 Phase 3 frequency

evergreen forest 0.2-0.8 0.8-1.0

semi-evergreen forest 0.4-0.6

moist deciduous forest 0.2-0.6 0.6- 1 .O

dry deciduous forest 0.2-0.4

0.4-0.6

teak plantation

coffee plantation

The mean standing volume was estimated to be 92 m3/ha, with a standard error of 9 rn7/ha. That gave an interval of confidence of 74 m3/ha to l l 0 m3/ha. Given the figures above, the total standing volume worked out to be 147 mio. m3, with a standard error of 14 million m'. The 95% confidence limits were thus computed at 1 19 m' and 175 m' respectively. The figures for the mean and total standing volume refer solely to the forested areas.

DISCUSSION

Through the 4-phase pilot survey in Kamataka, comprising a combination of L a n d s a m data, B/W and CIR aerial photos, and field surveys, it was possible to develop an operational concept for a 4-phase survey to determine standing volume. In this concept it was possible to utilize the logistics and the extensive knowledge of the Kamataka Forest Department.

This sampling design can only be optimized after detailed analysis of costs and the variation within phase and strata combinations (see Scott and Kohl, 1993, and Kohl and Sutter, 1991). One of the major aspects for optimization will be the sample volume in each of the phases.

If an optimum inventory design c& be developed, it must, of course, be compared to 2- and 3-phase designs, not only in terms of cost efficiency for the determination of standing volume, but also in the light of the conditions in the inventory m and the inventory aims; here the registmtion and documentation of forest dynamics should not be neglected.

The inventory design described here has already been widely accepted by the forest service in India. For instance, the KamatakaForest Department is planning alarge-scale inventory of the whole state based on this design (Akbar Sha, 1989).

CONCLUSIONS

The inventory of the growing stock in tropical forest regions has been one of the most difficult tasks. Most of the earlier works on timber inventory worldwide have depended on two-phase sampling design using aerial photographs and ground surveys. The advent of satellite remote sensing has added a new dimension to inventory design reducing considerably the time and costs. The present four-phase pilot statistical sampling design was devised to further economize the overall timber volume assessment operations without compromising on the accuracy of the estimates. The results of this exercise matched fairly well with the estimates obtained through a two-phase inventory carried out earlier by Karnataka State Forest Department. The design of the inventory is particularly suitable for large forest areas on complex terrain where locating a sample ground plot through trian- gulation methods is often extremely difficult.

ACKNOWLEDGEMENTS

This study has heavily depended on the laboratory analysis and field surveys jointly planned and carried out together with Prof. G. Hildebrandt, Dr. H. Schrnidtke (Freiburg University), Dr. N.V. Madhavan Unni, Dr. C.B.S. Dutt, Mr. K.S. Murthynaidu (NRSA) and anumber of Kamataka Forest Departmentofficers. Their contribution is gratefully acknowledged. Thanks are also due to Dr. Ch. Kleinn of Forest Biometry Department, Freiburg University for making available his SITCA program for volume computations.

Prof. R. Schlaepfer, director of the Swiss Federal Institute of Forest, Snow and Landscape Research, Birmensdorf, gener- ously supported this project by making resources at the institute available for the final analyses.

REFERENCES

AKBAR SHA, A., 1989. Remote sensing in forestry: Proposal for introduction of forest resource information centre in KFD, unpublished manuscript, 121989.

BowDEN,D.C.,DD(ON,G.E.,FRAYW, W.E.,GRAYBWF.A.,JEYARA~AM, S.,J~~~T~N,D.C.,KENI',B.M.,LABAU,V.J.~~~R~~ER~~,E. 1979. Multi-level sampling designs for resource inventories, Dept. of Forest and Wood Sc~emxs, C o l d o State University.

COCHRAN, W.G., 1977. Sampling Techniques. John Wiley and Sons, New York.


D m s , T., 1950. The problem of optimum -cation, Skand. Akt. Tids*, 203-21 3.

DALENR~S, T. aMi HoDoe~, J.C., 1957. The choice of stratification points, Skand. M. Tidsktj?, 198-203.

F-W. andKom,M. 1987. Ein4-phasigesStichprobendesign zur Stratifikation unter Verwendung von Landsat-Daten, Lufibildern und terrestrischen Aufnahmen. AFJZ 158 (Y3): 26-30.

JOHNSTON, DC., 1982. Theory and application of selected multilevel sampling designs. Ph.D. Thesis, Colorado State University.

KLEINN, Ch., 1989. Zur Berechnung von Einzelbaumvolumen, Mitt. Abtl. Forstl. Biom. der Universitat Freiburg, 198911.

K0~1, M. and S u m , R., 1991. Application of Brial photographs in the estimation of standing volume in the Swiss NFI, in: KOHL M., F'ELZ D.R.(eds.): Forest inventories in Europe with special reference to statistical metholls, Proceedings, Int. IUFRO S.4.02 and S.6.04 Symposium, May 14-16,1990, Birmensdorf, 176-191.

KUSHWAHA, S.P.S., 1987. Applications of remote sensing in forest survey and management. Proc. lX International Symposium on Ecological Management of the Tropical Ecosystems, 11-16 December, Varanasi, India.

KUSHWAHA, S.P.S., 1990. Forest type mapping and change detection from satellite imagery, ISPRS Journal of Photogrammetry and Remote Sensing, 45 , 175-181.

LABAU, V.J. and SCHREUDER, H.T., 1983. A multiphase, multiresource inventory procedure for assessing renewable resource and monitoring change. Proc. Int. Con$ on Renewable Resources Invento- ries for Monitoring Change and Trends, Covallis, 456-459.

LACHAN, R., 1985. Optimal Stratum Boundaries for Shellfish Survey, Biometrics 41, 1053-1062.

LANGLEY, P.G., ALDRICH, R.C. and HEZLER, RC., 1969. Multistage Sampling of forest resources by using space photography - an Apollo 9 case study, Vol.Z:Agr., Forest. andSensor Studies, Proc. 2nd Ann. Earth Resources Aircraft Program Review, NASA MSC, Houston, Texas, 1-2 1.

LANGLEY, P.G., 1975. Multistage variable probability sampling, Ph.D.Thesis, Ann Arbor.

NRSA, 1983. Nationwide mapping of forest and non-forest areas using Landsat false colour composites for the periods 1972-75 and 1980-82. Project Report, National Remote Sensing Agency, Hyderabad, India.

SABOROWSKI, J . , 1990. Schatzung von Varianzen und Konfidenzintervallen aus mehrstufigen Stichproben am Beispiel von Waldschadeninventuren, Schriften aus der Forstlichen Fakultat der Universitat Gottingen und der niedersachsischen Forstlichen Versuchsanstalt, Band 99, 135 S..

SCHADE, 1980. Ein mehrphasiges Stichprobensystem fiir forstliche Grossraum-inventuren, gestiitzt auf Landsat-MSS-Daten, Luftbilder 1:50,000 und erganzende terrestrische Messungen, Diss. Universitiit Freiburg.

S c m , C.T. and Kohl, M., 1993. A method for comparing sampling design alternatives for extensive inventories, Mitteilungen der Eidgenossischen Forschungsanstalt fur Wald, Schnee und Landschafr, Birmensdorf, Band 68, Heft 1,62 p.

SHAPIRO, SS. and W ~ K , M.B., 1965. An analysis of variance test for normality (complete samples), Biometrika 52,591-61 1.

UNNI, MADHAVAN N.V., 1990. Space and forest management, in: RAO, U.R. (4.): Space and forest management, Proceedings of the 4Ist IAF Congress, Dresden, 6-13 October, 1990.

APPENDIX

Algorithms for the calculation of the mean %...and the variance of the mean Q?. ..) in Qphase sampling for stratification.

JI

R.,. = 2 WIJ fij. J-1

% - number of elemenb in the m, selected phase 2 strata


K,,

bijk = number of elements in phase 4 stratum ijk

Cijk = number of elements in the bij in phase 3 stratum k

m, . - b. K" + IJ " c ( Wijk Yijk. YI]

mij (bij - 1) k=1


Variation of cypress aphid (Cinara cupress0 (Buckton) attack on the family Cupressaceae

JOHN F . OBlRl Kenya Forestry Research Institute, PO Box 204 12, Nairobi, Kenya

SUMMARY

Twenty-four different species of conifer trees within the Cupressaceae family were studied for their growth rates and trends of tolerance against the cypress aphid Cinara cupressi (Buckton). The study undertaken at Muguga arboretum, Central Kenya over a continuous two year period indicates a wide range of tolerance among the species. Thuya spp. and Cupressocyparis leylandii (Jackson andDallimore) were most tolerant while Widdn'ngtonia and Callistris were the least. Among the cypresses, Cupressus torulosa (Don), C.funebris (Endl.) and C . arizonica (Greene) were very resistant whereas C. benthamii (Endl.), C. lusitanica (Miller) and C. lindleyi (Klotsch) were susceptible. In view of the growth vigour and resistance towards the aphids, Cupressocyparis leylandii appears the best alternative species to C. lusitanica which is infested and currently widely grown as a plantation species. Results of the hybrid species also indicate that resistance breeding through hybridization may offer a viable long- term solution to the aphid menace.

INTRODUCTION

Trees within the family Cupressaceaecomprisethe mostcommon and important coniferous plantation species in Kenya. Among these, Cupressus lusitanica (Miller) and Juniperus procera (Hochst. ex Endl.) (East African pencil cedar) make up most of the industrial and indigenous forest cover totalling over 350,000 hectares. Cupressus lusitanica is the main species used for timber, construction and as an ornamental live hedge. Juniperus procera, which is termite resistant, is the only suitable tree for poles and fencing posts in the termite prone areas of East Africa. These species have lately been seriously attacked by the cypress aphid Cinara cupressi (Buckton). This pest is one of the large number of species in the genus Cinara (Homoptera: Lachnidae). It occurs in two forms, wingedand wingless types, bothof which suck treefluidsfmmthevascularsy stemcausing tissuedessication, branch diebackand finally tree death. The adults and nymphs of the wingless type are the most destructive as they exist in numerous populations. They feedincolonies, producing copious amounts of honey dew which covers the branches and stems. This honey dew provides a growth medium for black sooty moulds that blanket the foliage and branches. Some trees show signs of tolerance towards these aphids. However, their annual growth increment is greatly impeded as the foliage required for photosynthesis is either lost or covered with sooty moulds (Ciesla 1991). Besides CiLpressus and Juniperus species, this pest dso attacks bees of genera Calliwis, Wiiddingtonia and Thuya. This aphid occurs natively in Europe, Asia and the h4iddle East, where it is checked by natural factors including weather, parasitaids and predators (Ciesla 1991). Its firstrecord in Africa was in Malawi in August 1986 (Chilima 1991). It later m(~ednwthwzvds toTanzania in 1987, Bun& 1988, Rwanda

and Uganda 1989 and Kenya 1990 (Ciesla 1991). It also spread from Malawi to Zambia, Zimbabwe and Zaire in 1990.

Over75,OOOhaof C. lusitanica have been attacked in Kenya, 15,000 in Tanzania and 4,600 in Uganda. In Rwanda and Burundi a total of over 60,000 ha have also been infested. In Malawi, the estimate of financial losses on the standing crop by 1990 was over US$3 million (Odera 1991).

Variation in resistance against this pest has-been observed both among the generaand within the species of the Cupressaceae family (Claude and Faustin 1991, Odera 199 1). This trend is important in selecting for resistance breeding as the species and individual trees showing a high combining ability for disease resistance provide a basis for a viable hybridization programme (Zobal and Talbert 1984).

The present study was undertaken at Muguga arboretum near Nairobi, Kenya. Itexamines the patternofcypress aphiddarnage to trees between March 1990, when it was first reported, and January 1992. The trend of tolerance within species and their hybrids is also evaluated. Alongside resistance breeding, the paper reviews an option of concurrent selection on growth vigour of Cupressus spp.

MATERIALS AND METHODS

Muguga arboretum is situated on a narrow ridge of land at an elevation of about 6,800 ft. It lies at 1°13' South and 36O38' East. The soils are a fertile dark loam derived from volcanic quartz. Theclimateisessentially tropical withamean annual temperature of 18°Candmeanannualrainfallof 1 1OOmmdistributedbirnodally with peaks in April and December.

44 John F. Obiri

TABLE 1. Aphid damage score, coefficient of variation (cov), average height, diameter at breast height (dbh), and mean annual increment (mai) of six genera (24 species) of the Cupressaceae family

Species No. of Aphid damage score % Average Average mai plots crown height (m) dbh (cm)

1990/91 1991192 cov foliage at 15 yrs Height dbh Mean Mean recovery (cm@)

Thuya orientalis Thuya occidentalis Cupressocyaris leylandii Juniperus procera Callitris endlicheri Callitris intratropica Callitris hugelii Callitris collumelaris Callitris pressi CaNitris rhomboidea Widdringtonia cupressoides Widdringtonia whytei Widdringtonia juniperoides Cupressus cashmeriana Cupressus funebris Cupressus arizonica Cupressus torulosa Cupressus sempervirens Cupressus nevadensis Cupressus (hybrids) Cupressus guadalupensis var. Forbessi Cupressus lindleyi Cupressus lusitanica Cupressus benthamii

Mean 3.48 3.31

LSD (P = 0.05) 0.64 0.55

* No recovery observed as trees experienced more damage - Multiple stems

TABLE 2. Aphid damage score and percentage of crown attacked

A total of 23 species and one hybrid planted in the arboretum were assessed for growth rates and damage due to cypress aphid. These comprised ten species of Cupressus, six of Callitris. three of Widdringtonia, two of Thuya and one each from Juniperus and Cupressocyparis (Table 1). The hybrid was a cross between C. lusitanica and C . arizonica. The 23 species and the hybrid were planted out in a total of 45 plots, each plot consisting of 49 trees (7 X 7). The number of plots per species is indicated in Table 1. For each plot, the heights and diameters at breast height (dbh) of the trees were measured at the age of 15 years. The trees that had multiple stems were not assessed for dbh. The extent of attack and damage by Cinara cupressi was determined on a five-point scale as indicated on Table 2 (Odera 1991). The damage assessments were done in January 1991 and January 1992. This was during the middle of the dry season of November to March when trees suffered most from moisture stress and aphid populations were high.

%age of Aphid Remarks crown damage

attacked score

0 1 Healthy tree. No symptoms associated with cypress aphid

1 - 25 2 Light infestation. Yellowing to Light attack browning of branches and twigs

limited to bottom of the crown.

25 - 50 3 Moderate infestation. Yellowing1 Moderate brown over the bottom of the crown

extending to 50% of it.

50-75 4 Heavy infestation. Yellowing to H e w browning and death of branches and

twigs affecting up to 75% of the crown. Copious sooty moulds on lower brrtnches.

75-100 5 Severe infestation. Yellow to browning and death of twigs snd branches extending to the entire crown. Tree dead ot dying.

Variation of cypress aphid attack 45

RESULTS AND DISCUSSION

Ail the 24specKs showed widespreadvariationsintheirresistance towards the aphids. Analysis of variance between these species indicated significant differences at the 5% level. Within the species there were idsonoted variations. Theobservedcoefficient of variationin aphid damage s c m s of the 199 1/92 season varied from 8% to 53% within the species.

The results of f able 1 indicate that the most severely attacked genera of the Cupressaceae were Widdringtonia and Callitris. These had an average aphid damage score of 4.7 and 3.6 respectively in 1991192. The infestation on these two genera varied from heavy to severe with 75-100% of the crowns dying. Among the Widdringtonia species, W. juniperoides (Endl.) and W. whytei (Rendle) were severely affected as their crowns completely died. However, W. cupressoides (Endl.) experienced slightly less damage. Similarly, Widdringtonia spp. have been adversely affected in Southern Africa. This is especially true of W. nodiflora (Mlanje cedar), which is Malawi's national tree (Chilima 1991).

Callitris species showed less susceptibility to aphids compared to Widdringtonia. Cailitris rhomboidea (R. Brown), C. pressi (Miq.) and C. collumelaris (F. Muell.) were the most susceptible whereas C . endlicheri (Parl.) showed moderate susceptibility. Cupressocyparis leylandii f Jackson and Dalliinore) had an average aphid score of 1.8 in the first year of attack. This species which is a hybrid between Cupressus macrocarpa (Hart.) and Chamaecyparis nootkatensis (D.Don) (Ovens er al. 1964) shows high tolerance towards the aphids. Similarly Juniperusprocera with an average aphid score of 1.8 displayed a high tolerance.

Besides having a high tolerance, C. leylandii has a fast growth rate indicated by its mean annual increment (mai) of 0.94m for height compared toother cypress species displayed in Table 1. This high tolerance and appreciable growth rate places C. leylandii as a probable alternative species for replacing C. lusitanica. Alongside C . leylandii, Cupressus torulosa (Don) (Himalayan cypress) should likewise be considered as another alternative to C. lusitanica as it also has good tolerance and growth rate.

Thuya occidentalis (L.) and T. orientalis (L.) showed the most resistance. This genus of the Cupressaceae suffered least from the aphid attacks with the cumulative effect of the second year attack showing only 15% of the crown W i g damaged.

The Cupressus species showed the greatest variability in tolerance of the aphids. They are divided into two major categories. The first comprises the most tolerant species which are C. Cashmeriana (Royie), C.funebris (Endl.), C . arizonica (Greene) and C. torubsa. The second category of C . benthamii @&l.), C. lusitunica, and C. lindleyi (Klotsch) are the least tolerant. Cupressw sempervirens (L.) and C, nevadensis (Abrams) displayed moderate tolerance.

The species thet showed high resistance towards aphids during the f h t year of attack tended to succumb more in the

year. This~mostnoti~~~MeinThu)ra,Cllpressocyparis species. By oontrast, Callitris and Cupressw

&most-thefht yearshowed

appreciable recovery in the second. This recovery, which was noted in terms of new crown foliage, ranged from 15% for C. lusitanica to as high as 35% for C. torulosa. In some plots a few trees maintained a green crown throughout the first and second year of aphid attack. These could form the potential clones for resistance breeding. The level of aphid infestation and damage alternate with the seasons. Colonies of Cinara spp. tend to be more abundant and damaging during the dry season (Browne 1968, Kfir etal. 1985). During the rainy seasons, the population levels show a marked drop at several sites in Kenya (Owuor, pers. comm.). It is at this time that Cupressus and Callitris spp. tendtogrow newfoliageatafasterratethanJuniperusandThuya spp. Hence in the second year of attack, the latter two species experienced more stress as they had limited foliage. The results in Table 1 indicate that the most tolerant cypress species (C. funebris, C. arizonica and C . torulosa) had the lowest heights and diameters, while the most susceptible group of species (C. benthamii, C . lusitanica and C. lindleyi) had the greatest heights anddiameters. The latter group have been preferred for plantation establishment because of their enormous sizes and large wood volume yield. However, this factorhasbeengreatly downplayed by their poor resistance to aphids. They could be improved by outcrossing them with the more resistant group of cypress species, which even though yielding less wood volume would provide the desirable traits of resistance.

Observations on the hybrid plot of crossing between the tolerant C. arizonica and the susceptible C. lusitanica produced hybrids of moderate tolerance. These hybrids offer a 'compromise' tree of moderate growth rate and appreciable tolerance.

The distinct variation of the cypress into the tolerant and susceptible groups can be compared with the work of Dyson and Herbin (1%9), who on the basis of minor alkane patterns in the composition of leaf wax, classified the cypress into two major groups: the New World and the Old Worldcypress species. The New World species comprise C. lusitanica, C , benthamii and C. lindleyi. These are susceptible to the aphids. They have similar alkane components that differ from other species (Dyson and Herbin 1969). Their similarity tends to agree with modem botanical views that they are only three forms of one species, C . lusitanica. The Old World species, comprising C. torulosa, C. funebris and C. sempervirens show high tolerance.

Thecategorization of cypress into the tolerant and susceptible groups can be further distinguished by the presence of man001 and torulosic acid in the resins of Old World species and their absence from New World species (Swain 1963). These aphid tolerant Old World cypress species could best be utilized in an interspecific hybridization programme with the faster growing New World species.

CONCLUSION AND RECOMMENDATIONS

Coniferous trees of the Cupressaceae family form the most importantexotic softwood in Eastern Africa. Cupressuslusitanica

46 John F . Obiri

alone accounts for over 160,000 ha which have been infested and damaged by the cypress aphids. The present survey shows that the speciesof this family display widevariationin theirresistance to the aphids. Thuya appears to be the most resistant genus while Widdringtonia is the most susceptible. Among the Cupressus, the Old World group of species (C. torulosa and C. arizonica) are more promising for resistance breeding than the New World group as they show more tolerance. However, the faster growth rates in the New World group (C . lusitanica, C . benthamii and C . lindleyi) should encourage hybridization trials within the two groups. This could yield hybrids with suitable traits of growth and resistance. Growing of species like C . torulosa that are less sensitive to the aphid should be encouraged on a larger scale.

Cupressocyparis leylandii has a high tolerance and better growth ratecompared tomost cypress species. It should be more widely tested through vegetative propagation. Selection of resistant clones within species like C. lusitanica should also be carried out to propagate clones that will tolerate the aphids. However, this approach will require evaluation and testing through progeny trials before recommendation for widescale planting.

ACKNOWLEDGEMENTS

The author wishes to acknowledge the valuable support services rendered by the staff of the Tree Improvement Division of the Kenya Forestry Research Institute; also the Director of KEFRI for permitting the writing of this paper.

REFERENCES

BROWN, F.G. 1968. Pests and diseases of forest plantarion trees. Oxford: Clarendon Press.

CHILIMA, C.Z. 1991. The status and development of conifer aphid damage in Malawi. In: CIESLA, W.M. (Ed.) Exotic aphid pests of conifers: acrisis in African forestry. Proceedingsof aninternational workshop held in Muguga, Kenya, 3-6 June 1991. pp.64-67.

CIESLA, W.M. 1991. Cypress aphid: a new threat to Africa's forests. Unasylva 45 (167): 51-55.

CLAUDE, N.J. and FAUSTIN, M. 1991. The case of cypress attack by Cinara cupressi in Rwanda. In: CIESLA, W.M. (Ed.) Exotic aphid pests of conifers: a crisis in African forestry. Proceedings of an international workshop held in Muguga, Kenya, 3-6 June 1991. pp.76-80.

DYSON,W.G. and HERBIN, G.A. 1969. Leaf wax alkanesasa taxonomic discriminant for cypresses grown in Kenya. Phytochemistry 7: 1339-1344.

KFIR, R., KIRSTEN, F. and VAN RENSBURG, N.J. 1985. Pauesia sp. (Hymenoptera: Aphidiidae), a parasite introduced into South Africa for biological control of the black pine aphid Cinara cronartii (Homoptera: Aphididae). Environmental Entomology 14: 597-601.

ODERA, J.A. 199 1. Someopportunities formanagingaphidsof softwood plantations in Malawi. Assistance to Forestry Sector, MLW/86/ 020, Malawi, FAO, Rome.

OVENS, H.,BLIGHT, W.~~~MITCHELL,A.F. 1964. Theclonesofleyland cypress. Quarterly J. For. 58(1): 8- 19.

SWAIN, T. (Ed.)1963 Chemical plant taxonomy London: Academic Press.

ZOBEL, B. and TALBERT, T. 1984. Applied forest tree impro~~ement. John Wiley and Sons.


Provenance variation in Pinus maximinoi : a promising species for commercial afforestation in Zimbabwe

B.I. NYOKA Forest Research Centre, Box HG 595, Highlands, Harare, Zimbabwe

SUMMARY

Eight year results of a Pinus maximinoi provenance trial at three sites are reported. For volume, height and breast height diameter the two provenances San Juan and Tatumbla performed consistently well at all the sites. Although some provenances were not consistent in their performance, site X provenance interaction was not statistically significant. Site effect was pronounced, with provenances showing a threefold volume production at Chisengu B compared with Nyangui. Nyangui showed a high level of Pineus infestation but with a low frequency of cone production. Most provenances showed a consistently high frequency of forking at Chisengu A. San Juan showed a consistently high frequency of cone production, low Pineus infestation, low forking frequency but poor stem form at all sites. Growth of P . maximinoi appears to be good at altitudes of around 1350 m, where temperatures remain above freezing. The best provenances were Tatumbla, San Juan, Dulce Nombre and Cofradia.

Keywords: Pinus maximinoi, Provenance, Site X Provenance interaction

INTRODUCTION

Pinus maxirninoi H . E. Moore is a tropical pine native to Mexico and the Central American countries of Guatemala, Honduras, Nicaragua and El Salvador (Dvorak and Donahue, 1988). The taxonomy of this species, previously known as P. tenuifolia Benth. was resolved recently when the species was renamed P. maxirninoi (Stead, 1983). P. maxirninoi has a wide climatic, edaphic and altitudinal range. It is found growing at altitudes ranging from 700 m to 2400 m as1 but is concentrated between 1 100 m and 1800 m (Dvorak and Donahue, 1988).

Preliminary results of 4.5-year-old trials in Zimbabwe and elsewhere indicate that this tropical pine has potential for commercial afforestation (Crockford et al. , 1991). Kietzka (1988) reported that P. maxirninoi showed better growth than P. elliottii and equal to that of P . patula in four-year-old trials in South Africa. In terms of under-bark dry weight of wood, Ladrach (1 986) reported that P. marirninoi was more productive than either P. kesiya or P . patula. P. maxirninoi has not only shown good growth but also ahigh wood density compared with P. patula (Ladrach, 1986; Wright, 1989).

The need to maintain a broad genetic base by species diversity necessitates continuous evaluation of alternative species or provenances. Increasing the number of suitable species will alsoimprove productivity on some sites by facilitating better species-site matching. .

The aims of this study were:

3. To identify the best provenances on which to concentrate research efforts.

MATERIALS AND METHODS

This trial comprised 14 test seedlots (13 provenances of P. maximinoi and one provenance of P. pseudostrobus) over three sites, namely Nyangui and Chisengu A and B. Besides being at ahigher altitude, Chisengu A is also amore exposed site. Rainfall and temperature figures for the two Chisengu sites are from a meteorological station which is 10 km from either site, and therefore are only approximate. All the seed was provided by the Oxford Forestry Institute. Table 1 lists the provenance and site details andTable 2 shows details of the threeZimbabwe test sites. The trial design was a Randornized Complete Block with six replications per site. All 14 provenances were represented at Chisengu A and B and 12 provenances at Nyangui. Although the layout varied from site to site, the plot size was consistent at 36 trees per plot with 16 measured trees.

The measurements taken at eight years were as follows: height in metres; diameter at breast height (dbh) in centimetres; stem formonascoreof 1-7, with 1- worstand7-best (Barrett and Mullin, 1968); susceptibility to the pine woolly aphid (Pineus pini), on an 0-3 scale (0 -no infestation, 1- light, 2- moderate and 3 -heavy); cones; forking and foxtailing, as presence or absence. These measured traits together with thi derived over-bark volume are reported in this paper.

~t the two-Chisengu sites, t h e ~ i n a s de 0ro was Tog~ninfOnnationon~mvenancevariatiw inP'mariminoi excluded from the analysis as most of the measured trees were under Zimbabwean conditions. fillers of a different seedlot. Also, the Tecpan provenance of P. ~ o ~ & ~ t h e i n f l u e n n : e o f siteongro~andproducti~~~~ pseudostrobus represented at all three sites, was excluded from of this species. the analysis of variance, but shown in the tables of means.

48 BJ. Nyoka

TABLE S . Provenance details

Provenance Country Code Altitude Latitude Longitude Rainfall (m) (NI (W) (mm)

Cofradia Dantali Dipilto Dulce Nombre Guinope Jinotega La Fortuna Loma de Ocho Minas de Oro Rafael San Juan Tatumbla Volcan Yali

Tecpan*

Honduras Nicaragua Nicaragua Honduras Honduras Nicaragua Honduras Honduras Honduras Nicaragua Guatemala Honduras Nicaragua

Guatamala

COFR DANT DIP1 DULC GUIN JINO FORT OCHO MINA RAFA JUAN TATU Y ALI

TECP

* P . pseudostrobus provenance

TABLE 2. Details of test sites

Site Altitude Latitude Longitude Rainfall Mean (m) (mm) temperature

Chisengu A 1550 19'54's 32"53'E 1410 16 'C Chisengu B 1350 19"54'S 32'53'E 1410 16 "C Nyangui 1845 18"Ot'S 32'47'E 1530 12°C

TABLE 3. F-values for combined analysis on height, DBH, volume and form

Source DF Height DBH Volume Form

Site 2 253 ** 166 ** 316 ** 16.8 ** ReplicatesISite 15 4.4 ** 3.4 ** 2.5 ** 3.1 ** Provenance 9 1.4 ns 4.6 * 5.1 * 4.1 * Site*Provenance 18 0.6 ns 0.6 ns 1.1 ns 1.4 ns Error 135

ns = Not Significant * = p< 0.05 ** = p< 0.01

RESULTS

Significant differences between provenances weredetectedonly for height at Nyangui, dbh at Chisengu A and form at Chisengu B and Nyangui (Figures S , 2 and 4). Ranked means of the provenances for height and dbh, volume and stem straightness are shown in Figures 1,2,3 and 4.

Dulce Nombre performed well at the two Chisengu sites but was below average at Nyangui. For the three main traits, (height, dbh and volume), San Juan and Tatumbla performed consistently well at all three sites. Although not significantly different from other provenances, Dantali and Jinotega were consistently

ranked last at all sites. Tecpan, a P. pseudostrobus provenance was just below average at Nyangui but ranked last at the two Chisengu sites.

Most of the provenances were consistent at all three sites, as evidenced by the site X provenance interaction, which was not significant (Table 3). This suggests that the best provenances performed consistently well across sites. Site effect was, however, very pronounced ( k O . 0 1 ) forthe main traits. Tatumbla for example yielded more than three times rbe volume at Chisengu B as at Nyangui (Figure 3).

Provenance variation in Pinus maximinoi 49

Chisengu A

- 9.OTATO

- NAN - DULC,RAFA JINO - 8.5 OCHO YALI. GUIN 1 COFR - DANT

- 8.0

- 7.5 TECP

- 7.0

- MINA

- 6.5

k 12.5 TATU. JUAN

DANT. DIP1

12.0 JINO. FORT YALI

C 1 1.5 RAFA .

l 10.0

9.5 TECP

Sig. ** s.e. 0.51

FIGURE 1 . Proportionately ranked means for height in metres at three sites. Dashed lines span provenances which did not differ signijcantly at 5%.

Nyangui Chisengu A Chisengu B

TATU 14.0

JUAN

JUAN 20.5

W H O RAFA, FORT 20.0 GUlN

19.5 T A W , DULC DIP1 OCHO

DANT. TECP 19.0 FORT COFR. GUlN

YALI. DIP1 JINO, DANT

RAFA 18.5 YALI

DANT

18.0

JINO 17.5

TECP

Sig. ns

From 2. Proportionately ranked means forBremt Height Diameter in cm at three sites. Dashed lines span provenances which did not d~ffer sign@canrly at 5%.

SO B.I. Nyoka

70.0 TATU

I - DULC I 1 - COFR I - 250 ' - JUAN I -- TATV I I - 240 I - FORT I - RAFA I - OCHO I - GUlN I - 2U) I I I - YAU I - 220 l I - nNO I - DANT. DIP1

100 TECP

m 15.5 1

FIGLIRE 3. Proportzonately ranked means for volume in dm' at three sires. Dashed lines span provenances which did notdiffer significantly at 5%.

Pineus infestation at the two Chisengu sites was rather low, but quite high at Nyangui (Figure 5). The two outstanding provenances. San Juan and Dulce Nombre, were relatively Pineus

FIGURE 4. Proportionately ranked means for stem straightness at three sites. Dashedlinesspanprovenances which did notdlflersign@cantly at 5%.

free, while Dantali, Jinotega and Guinope had over 50 % infestation, with the latter showing the highest level of heavy infestation of 17 %.

m Heavy

U Moderate

m Light

, , , , . , CWR WLC WH JUAN TATU YALI

OWf OCHD N O W A TECP

Provenance

Provenance variation in Pinus maximinoi 5 1

Forking varied considerably across the three test sites, with forking frequency. The highly productive San Juan provenance Chisengu A showing a consistently high forking frequency. At had aconsistently low forking frequency at all the three test sites, this site, the Dantali provenance was highest, with a forking while Dulce Nombre and Tatumbla were moderate but variable. frequency of 40 8 (Figure 6). The Nyangui site had the lowest

.. -

COFR

Chisengu A

Chisengu B

GUlN . . OCHO

, , . , ,

JUAN TATU YALI DANT DULC FORT JlNO RAFA TECP

Provenance

FIGURE 6. Provenance variation in forking at three sites

Across the three sites, San Juan again showed aconsistently high frequency of cone production (Figure 7). Site influence was strong, with Chisengu B showing a high frequency for all provenances, and Nyangui showing a very low frequency. The

other high yielding provenances of Dulce Nombre and Tatumbla exhibited a marked variability, with less than 10 % of the trees producing cones at Nyangu

- .

m Chisengu A

LA Chisengu B m Nyangui

W T A N R N A TCCP

YALI

Provenance

52 B.I. Nyoka

Although the San Juan provenance was outstanding for most of the measured traits, its stem form was poor at all three sites. The other good provenances, Dulce Nombre and Taturnbla, had good stem form. TheTecpan provenance had good stem form at Nyangui only. Stem form was highly variable at Chisengu B.

DISCUSSION

The growth of P. k i m i n o i was good at Chisengu B, moderate at Chisengu A but rather poor at Nyangui. The moderate performance of the species at Chisengu Acouldbe due to the fact that the site is at a higher altitude and exposed to wind damage. Also, this trial suffered from animal damage soon afterestabiish- ment. In 4.5-year-old trials Crockfordet al. (1991) found highly significant differences between provenances for volume, height and dbh at the two Chisengu sites only. They attributed the poor growth and the lack of significant differences at Nyangui to freezing temperatures experienced at that site. In the results reported here, there were no significant differences between provenances for volume at all three sites. Significant differences between provenances were detected only for height and dbh at Nyangui and Chisengu A respectively. The lack of significant differences for volume at eight years contrasts with the results of Crockford etal. (199 l), who found highly significant differences between provenances at 4.5 years. This suggests that differences between provenances may become less apparent with age. The nearly threefold volume production at Chisengu B compared to Nyangui is mostly due to freezing temperatures at Nyangui, as this species does not experience such low temperatures in its natural habitat (Dvorak and Donahue, 1988). Despite these changes at 8 years, Tatumbla, Dulce Nombre and Cofradia remained the top volume producers. In South Africa, Kietzka (1988) reported that these three provenances were the best for height, volume and dbh.

Although site X provenance interaction was not significant, the site effect itself was, suggesting a high degree of site specificity for this species. Crockford et al. (1991) found the interaction to be significant, but on a greater range of sites. High productivity with this species is thus realized with correct matching of site.

The San Juan provenance reported by both Crockford et al. (1991) and Kietzka (1988) as having poor stem form and average productivity at 4.5 years, was found in this study to produce a consistently high volume at all three sites (Figure 3). This provenance seems tohave a slow initialgrowth rate. Despite improved growth later, the poor stem form persists. Growth of Tecpan was slightly below average at Nyangui but very poor at Chisengu. P. pseudostrohus, and in particular the Tecpan is a high altitude (2200111) provenance and therefore was planted well outside its range. The stem form was however impressive at Nyangui, a site which probably approximates the provenance's natural range.

San Juan had the highest frequency of cone production, followed by Dantali and Jinotega, although the last two were variable. Although this agrees with Crockford et al.'s (1991) 7.5-year-old results from South African trials they concededthat high cone production did not necessarily imply high viable seed production.

There was no foxtailing in any of the provenances at the three sites. P. maximinoidoes not normally foxtail in its native habitat. According to Umgo and Lambeth (1988) foxtailing in this species increases more with increasing latitude and longitude of origin than with climatic factors.

Forking in this trial seems to be influenced by site. Nyangui had the lowest forking frequency for most of the provenances, while Chisengu A was consistently high for all provenances, probably due to cattle damage. The San Juan provenance had consistently low forking frequency at all test sites. Ladrach (1986) attributed site differences in forking to boron deficien- cies. From these results it is difficult to ascertain the site factor causing this.

Pineusinfestation was only noticeable at Nyangui (Figure4). Loss due to this pest can sometimes be substantial. In South Africa, Zwolinski (1988) reported annual loss of increment in volume in young P. elliottii of 19.5 % as a result of Pineus infestation. Dulce Nombre, Taturnbla and Cofradia were moderately infested while the highly productive San Juan provenance was relatively Pineus free.

CONCLUSIONS AND RECOMMENDATIONS

Although differences between provenances were not significant for most of the measured traits it is recommended that research be concentrated on the Dulce Nombre, San Juan, Tatumbla and Cofradia provenances. These provenances consistently scored higher than all others for most of the measured traits. San Juan, despite its poor stem form, is a good volume producer. Its poor stem form could be improved through selection and breeding. Individual seedlots of these provenances need to be collected so that genetic studies can be initiated. Future trials need to cover a greater number of sites to evaluate site effect properly as well as site X provenance interaction, since maximum productivity can be attained by correctly matching the species to sites. Although P. miminoiis also found in Mexico and El Salvador, all provenances used in this study were from Guatemala, Hon- duras and Nicaragua. It would be necessary to include materials from these areas in future trials. P. maximinoi seems to do well at moderate altitudes of around 1350 m. Its growth a higher altitudes, with freezing temperatures, is poor.

ACKNOWLEDGEMENTS

I am grateful to S. Kunaka, Data Processing Officer, for his invaluable advice in data analysis. I also wish to thank the Field Station Managers for the assistance in the management of the trials, and Drs R. D. Barnes and K. J. Crockford for their invaluable comments.

REFERENCES

BARREIT, R.L. and MULLIN, L.J. 1968. A review OF introduction of forest trees in Rhodesia. The Rhodesia Bulletin of Forestry Research'l: (x)227pp.

CROCKFORD, K. J.,BAYus, W. B., DUNSDON, A. J., WANYANCHA, J. M.,

Provenance variation in Pinus maximinoi 53

and WRIOHT, J. A. 1991. Provenance productivity of Pinus mimino i in Southern Africa. Comrn.For. Rev. 70 (4), 213-236.

DVORAK, W. S. and DONAHUE, J. K. 1988. Pinus maximinoi seed collections in Mexico and Central America. CAMCORE Bulletin on Tropical Forestry No. 4, 47p.

KrmucA, J. E. 1988. Pinus mimiwi: a promising species in South Africa. S. Af7. For. Jour. 145,33-38.

LADRACH, W. E. 1986. Comparisons between provenances of seven conifers in the Andean region after eight years. Carton de Colombia Research Report No. 105. 8p.

SW, J.W. 1983. A study of the variation and taxonomy of the Pinus pseudostrobus complex. Comm. For Rev.62(1):25-35.

URREGO, J.B. and LAMBETH, CC. 1988. Provenance and family in foxtai1ingofPinusmawiminoiH.E. Moore in the Andes mountains of Colombia. Carton de Colombia Research Report No. 12 1,4p.

WRIGHT, J.A. 1989. Wooddensity anddry matterproductioninaPinus mimino i trial at age seven years. (Unpublished).

ZWOLINSKI, J.B. 1988. Preliminary evaluation of the impact of the pine woolly aphid on condition and growth of pines in the Southern Cape. S. Afr. For. Jour. 153,22-26.

54 Commonwealth Forestry Review Volume 73(1), I994

RESEARCH NOTE

Comparison of the growth and yield response of Pinus patula between natural stands in Mexico and South African plantations

CELEDONIO AGUIRRE-BRAVO and SUSAN A. WINTER1

INTRODUCTION

Pinus parula Schldl. and Cham. native to Northeast Mexico (Peny 1991) is intensively managed as an exotic species in southern and eastern Africa. While the silvicultural characteristics and management of the species as an exotic are well documented (Wormald 1975), there is little growth and yield information from its natural distribution. Aguirre and Smith (1 986) compared the development of dominant height of natural stands in Mexico and African plantations.

In this note, the Richards growth function has been used to compare thegrowthand yieIdofPin~rspatulain unthinnedeven- aged natural stands with that in plantations in South Africa. Growth is comparedusing growth functions developedin South Africa for managed and unmanaged P. patuln plantations (Van Laar 1978: Von Gadow 1983). The derived models, because of their biornathematical properties (Pienaar 1965), have potential for assessing alternative options to manage natural stands and plantations of P. patula in Mexico. The density of the natural plantations is assumed to be >25,000 stems per ha.

total tree height (H), and age at breast height were determined for all trees in each plot.

The mean age at breast height varied from 5 to 43 years, volume per hectare (V) from 20 to 650m3, basal area (B) from 8 to 60 m2, and site index at base age 25 years from 23.5 to 24.3 m. Volume and site index were determined from the stem volume and index equations developed by Aguirre and Smith (1986).

Modelling Approach

Stand attributes for the Mexican data set, such as D, H, B, and V, were estimated based on the same growth functions and fitting procedures which Van Laar(1978) and Von Gadow (1979) used to model thegrowth and yieldresponse toCCT(corre1atedcurve trend) thinning experiments of P. parda plantations in South Africa. The treatments are shown as nominal numbers of trees per hectare in Table 1.

RESULTS METHODS

Dafnfiom Mexico

Threecircularunthinned sample plots of0.0125 hawereselected from sixty unthinned natural stands of similar site index. The sample plots were carefully selected for homogeneity in species composition, density and age class distribution, were not affected by insect or disease attack, were freeof density-independent mortality, not thinned during their life history and free of evidence of growth stagnation. Diameter at breast height (D),

The results are summarized in Figures 1 to 4 where mean diameter (dbh), stand mean height, basal area and residual stand volume respectively are plotted against stand age for different stand densities. The delay in the start of growth in natural stands compared with the plantation is to be noted. Mean heightgrowth is slower in the natural stands sampled but basal area for natural stands is similar to the unthinned plantations. Volumegrowth of natural stands, being distributed on many smaller trees, reaches an asymptote at the same level in moderately thinned stands in the plantations.

T.4~1.11 1 . Thinning tr-earmenis eupressed as stem numbers in South African thinning trials compared with natural stands in Mexico for P. patula

Stand Age (years) treatment 1.75 2.85 4.24 4.84 5.75 6.62 7.62 7.85 35

1 2 3 4 5 6 7 8

Natural Stands

Nominal trees per hectare (N) -

- - - -

741 - 741 494 - 741 494 37 1 741 494 37 1 741 494 37 1

- -

Growth and yield response of Pinus patula 55

STAND AGE (years)

FIGLRE 1.Mean diameter development of Pinus patula for natural stands in Mexico and plantations in South Africa.

"0 10 20 30 40 STAND AGE (years)

FIGURE 3. Mean basal area development of Pinus patula for natural stands in Mexico and plantations in South Africa.

REFERENCES

AGUIRRE-BRAVO, C. and SMITH, F.W. 1986. Site index and volume equations forPinuspatula in Mexico. Commonw. For.Rev. 65(1): pp. 51-60.

PERRY, P.P. 1991. The pines of Mexico and Central America. Timber Press. 231pp.

PIENAAR, L.V. 1965. Quantitative theory of forest growth. Ph.D. Dissertation. University bf Washington. 180 pp.

VAN LAAR, A. 1978. The growth of unthinned Pinuspatula in relation to spacing.S. Afr. For. J. 107: pp. 3-1 1.

VON GADOW, K. 1983. A model of the development of unthinned Pinus patula stands. S. Afr. J. 126: pp. 39-47.

WORMALD, T.C. 1975. Pinuspatula. Tropical forestry papers, No. 7 . Dept. of Forestry, Commonwealth Forestry Institute, University of Oxford. 150 pp.

"0 10 20 30 40 STAND AGE (years)

FIGURE 2.Mean height developnienr of Pinus patula for. natural stands In Mexico and plantations in South Africa

- 0 10 20 30 40

STAND AGE (years)

FIGURE 4. Mean volume per hectare developnzenr of Pinus patulafor. natural stands in Mexico and plantations in South Ajkica.

EDITORIAL NOTE

This note has been prepared from a larger paper by the authors. For further information please write to the authors.

' This research was undertaken while the authors were with the Division de Ciencias Forestales, Universidad Autonoma Chapingo and Programa Forestal, Colegio de Postgraduados, Mexico. The authors are currently with the USDA Forest Service, Rocky Mountain Forest and Range Experiment Station, Fort Collins, Colorado, and the USDA Forest Service, Land Management Planning, Fort Collins, Colorado.


BOOK REVIEWS

W.T. ADAMS, S.H. STRAUSS, D.L. COPES and A.R. GRIFFITHS (Eds.) American conifers, the second with design and management of seed Population genetics of foresr trees. Forestry Sciences Vol. 42, orchards and a third with conservation. The fourth paper reviews Kluwer Academic Publishers, PO Box 989, 3300 AZ Dordrecht, effects of forest management on gene pools. These papers underline Netherlands. 420pp. Dfl.2501 US$1651£87 (hardback). lSBN 0 the important uses of biochemical markers in the sustainable 7923 1857 9. management of our forests.

This book is the published proceedings of an international symposium on 'Population genetics of forest trees' in Corvallis, Oregon in 1990. It contains a keynote paper on genetic diversity by M.T. Conkle. 17 more specific papers by authors from North America, Europe and Australia and a final commentary by 1-1.-R. Gregorius and Ph. Baradat. The papers deal with the major applications of biochemical markers to forest genetics research. In the last 10 years biochemical and molecular methods have largely replaced morphology as the traits of choice for phylogenetic studies. . . -

Conkle's keynote paper provides a wider perspective of genetic variation above the species level with an interesting discussion of effects of past evolution on diversity. Biochemical markers facilitate the identification of genetic relationships and differentiation within and among populations of a species.

The next set of four papers are grouped under the topic of assessment of genetic diversity within and among species, and summarize a lot of information on genetic variation of European coniferous and broadleaved tree species, of Eucalyptus and Acacia and some tropical woody species. Woody species have more genetic diversity and partition genetic diversity within and among populations differently than species with different life forms. Gymnosperms have slightly more genetic diversity than angiosperms. Whereas breeding systems appear relatively unimportant in predicting the level of genetic diversity, species whose seeds are ingested or attached to animals have the highest diversity values (Hamrickeral.). These papers are very useful to persons involved in conservation and management of the genetic resources of woody species. Information on genetic variation needs to be incorporated into strategies for the preservation of genetic resources, as for example regionally or widely distributed tropical species are likely to be genetically distinct in different locations.

The next three papers are concerned with biosystematics and adaptive significance of biochemical markers. They are largely concerned 1~1th techniques of analyses, e.g. terpenes in a paper by Hanover, experimental design (Bush and Smouse), and interpretation of data (Strauss (,I trl.). These are important papers for active scientists working in this area.

The section on mating systems, gene dispersal, and genetic structure within populations consists of four papers. Mitton in reviewing mating systems of conifers notes heterogeneity appears at virtually ever! level and many of the assumptions employed to estimate mating system parameters areviolated innatural populations and in veed orchards. The paper by Adams provides a good review o f methods of measuring gene dispersal within populations of trees and the following paper by Ellstrand continues with a comparison of IINII 111~~1hods used toestimate gene flow among seed plant populations. Lipperson revlews theoretical andempirical resultsonspatial patterns as amounts of variation. levels of inbreeding and local degree of adaptation depend on isolation by distance and natural selection operating. These are technical papers of importance to the serious research scientist.

To the general reader the section on application of biochemical markers in forest management might be expected to be of most interest. The first paper is concerned with defining breeding zones of four NW

The last section deals with DNA as a biochemical marker; most of the book and the data presented is based on isoenzyme, terpene or electrophoretic techniques. In recent years DNA-based genetic markers (particularly restriction fragment length polymorphisins and polymerase chain reaction) have been developed and continue to be improved. The two papers are excellent but with the very rapid progress and proliferation of studies with DNA markers since 1990 the information presented is dated. This criticism is not really applicable to most ofthe book as other biochemical marker techniques have not experienced this recent rapid development. These other techniques are often as or more appropriate than DNA markers and usually substantially cheaper. In fact. one of the principal omissions was a major critique of the uses of the main biochemical marker techniques.

The aim of the book was to review the use of biochemical markers to provide a better understanding of the genetics of trees. especially of their population genetic structures, evolutionary biology. and mating systems. That objective has been achieved very competently and professionally. The book has been well edited and appears free of errors. and is reasonably consistent in style. Although the general forester or research scientist will find much valuable information in this hook. I recommend i t particularly fortheserious research workeron biochemical markers. I don't know of any other hook which deals in such detail with the methodology of biochemical markers, especially isoenzymes and terpenes, and provides such a comprehensive review of the application of these methods to forest tree species. The cost at E87 is what one can expect to pay these days for such a book.

S.R. AIKEN and C.H. LEIGII Vul>iskit~,q IYIIII for~c.st~: //I(' i'(.~~/r),yi(~u/ rr-ut~sition in Mr~la~.siu. Oxford Monographs on Biogeography No. 5, Clarendon Press, Oxford. 1992. xiii + IY4pp. f40.00. ISBN 0- 19-854242-9.

The focus of this book is on the human interaction with the forests of Malaysia, a developing nation with great biological diversity and the third highest per-capita GNP in Southeart Asia (US$1800). The stage isset withan overview ofthe structure,diversity and functioning of the rain forests in the region. The authors then discuss the processes and policies shaping anthropogenic changes to the region's forests during the past two centuries, emphasizing recent issues. They then examine the consequences of these changes on the human and natural environment and consider the initiatives necessary to direct Malaysia toward a more sustainable future.

The book, by way of introduction, gives an good summary (p. 19- 45) of rainforest structure and function in Malaysia, but this not its main theme and those looking for such a book should look elsewhere (e.g. T.C. Whitmore Tropical rain forests of the F a r East 2nd ed, Clarendon, Oxford, 1984).

The main thrust of the book is about human interactions with the forest, the consequences of past land use policies and practices. and implications and challenges for the future. It examines the impacts of the colonial administration, of peasant agriculture, rubber and oil palm plantations and of tin mining on the forest environment. FELDA

Commonwealth Forestry Review Volume 73(1), I994 57

(Federal Land Development Authority) oil palm and rubber schemes are examined in some detail. Most deforestation in Peninsular Malaysia has occurred since independence: between 1958 and 1987, forest cover declined from 74% to 368, and considerable areas of forest have been degraded.

Three silvicultural systems are reviewed: the Regeneration Improvement Felling System (RIFS) prior to Japaneseoccupation, the MalaysianUniformSystem(MUS)during 1949- 1980, andtheselective Management System(SMS)introducedin 1977. TheSMS isaflexible polycyclic system with anominalcycleof 30 yean,and is used mainly in Peninsular Malaysia. In Sarawak, fellingcycles are 25 and45 years for mixed dipterocarp and peat-swamp forests respectively, while SabahmaintainsamodifiedMUS withan 80 yearcycle. 'Government officials acknowledge that current management problems include an inadequate number of trained staff to enforce harvesting regulations, insufficient knowledgeofhow to manage the hill forests for sustainable production, and a paucity of data on the impact of different harvesting systems on the residual stands.' The authors contend that the main rationale behind the sequence of silvicultural systems - RIFS, MUS, SMS - has been the mounting economic pressure on the forests.

After many calls from concerned foresters over a long period. Malaysiaformulated anational forest policy andenshrined it in law as the National Forestry Act in 1984. Unfortunately, in the Malaysian Federation, this law is not binding on states unless gazetted as a state law, and several states have not been prepared to do this. Malaysian forests provide a wealth of products and services, but have been managed almost exclusively for industrial wood production. The authorscall for greater attention to multipurpose management including the needs of indigenous forest-dwelling people. They also draw attention to the undervaluing of non-timber forest products and the disregard for the environmental costs of timber harvesting.

Chapter 4 examines the implications of exploitation, degradation and removal of the forest. Effects on hydrology, soil, flora and fauna are reviewed. Some 343 of the 654 tree species endemic to the Malaysian Peninsula are considered endangered, and many of the 66 species confined to the tin-rich alluvial soils have not been sighted for 50 years. Many of Malaysia's plants have aconfineddistri bution: 2 16 of the 343 endangered tree species occur in only one state, and many rare plants are known from a single specimen or site. Many animal species ranging from the tiger and Sumatran rhinoceros to turtles and peripatus 'worms', arealso threatened, mostly by habitat loss, but also by hunting, fishing and collecting.

The progress of conservation efforts is outlined, and four case studies are examined in some detail: the Endau-Rompin national park proposal; the conflict between loggers and the Penan and other ethnic groupsintheBaram-Limbangdistricts; theproposedSungaiTembeling dam near Taman Negara, the only national park on the Peninsula; and the Bakun hydroelectric dam near Belaga in Sarawak.

The authors argue that five imperatives for a sustainable future in Malaysiaincludeacquiringbetterinventorydataforforestmanagement; stabilizing the population through poverty alleviation, reduced fertility and mortality, and enhancing the status of women; reforming government departments to increaqe inter-departmental cooperation and greater accountability for formulating policies and budgets that foster sustainable development; devolving power to enable greater involvement by governments, NGOs and communities and focusing more on human fulfilment and environmental integrity; reforming fiscal arrangements between state and federal governments and increasing timbermyalties. Theconflict betweenloggem andindigenous people should be resolved, more forest should be protected from logging, and small-scale forestry and agroforestry based on non- timber resources should be encouraged.

TIK book is readable and thought-provoking. It may offer an interesting insight for policy makers, planneis and land managers in

the moist tropics. Malaysia's relatively high GNP, forest cover and low population density make it rather unique in the moist tropics, but implications for policies elsewhere can be inferred.

JERRY VANCLAY

T. BANURI and F. APFFEL MARGLIN (Eds.). Who will save the forests? Knowledge, powerandenvironmentaldestruction Zed Books, 7 Cynthia St., London N1 9JF, UK. 1993. 195 pp. Paperback £ 1 1.95 (also in Hardback, £32.95). ISBN 1 85649 160 9 Pb: 1 85649 159 5 Hb.

M. COLCHESTER andL. LOHMAN. (Eds.). The struggle for landand the fate of the forests. Zed Books, 7 CynthiaSt., London NI 9JF, UK. 1 9 9 3 . 3 8 9 ~ ~ . Paperback£ 12.95 (alsoinHardback,£32.95). ISBN 1 85649 140 4 Pb; 1 85649 139 0 Hb.

Thecontrast drawn by JeffRomm (1991 )between 'forest policies' and 'policies that shape forests' suggests a theme common to these two books, though each addresses it at a different level. That theme might becharacterizedas the relationships between the distribution of power in societies and the use of forests. Whilst Tariq Banuri, Friderique Apffel Marglin and their contributors address these relationships in terms of the politics of knowledge itself, Marcus Colchester, Larry Lohman and their coauthors concentrate on the political economy of land tenure and itsconsequences. Banuri and Apffel Marglin's analysis is based on the concept of 'systems of knowledge', which they illustrate by definingcontrasting 'modern' and 'non-modem' ideotypes; the former might be approximated by technological knowledge. and the latter by the more traditional. These concepts provide a framework against which conflicts over forest use may be interpreted in terms of the assertion of one of these systems over the other, typically the modem over the non-modern, although they also observe counter- examples (e.g. the Chipko Movement) and conflicts within each. Colchester and Lohman focus their analysis on the linkages between land tenure, rural poverty and tropical deforestation, to illustrate their case for agrarian reform; indoing so, they seek torekindle debate about an issue which they see as having lapsed from development agendas. Both books have a similar structure, of introductory chapters followed by case studies, and both reach broadly similar conclusions which emphasizeempowermentof forest-dwellingordependentcommunities. However, giventhedistinct analytical frameworkofeach. they represent complementary contributions to the debate about why forests should be sistained, aid how they might be.

Who niN save the joresrs? comprises six chapters: two which introduce the analytical framework and the examples, and four which discuss case studies from northern India, northeastern Finland and the northeastern USA. The twointroductory chapters providean accessible introduction to a paradigm which may be unfamiliar to many natural scientists, one which relates knowledge - and the means by which it is acquired and expressed - to power, and how it is established and exercised. The second chapter also illustrates how the case studies might be interpreted against this framework, echoing for the particular case of forests the analysis applied to development generally in Sachs' (1992) Development Dictionary, which describes itself as A guidc to knowledge aspower. The Indian case studies concern the Hill Maria tribe (Dr Savasaachi) and the Chipko Movement (Ramachandra Guha); the Finnish (Jukka Oksa) and American (David Vail) studies consider forest-dependent communities in the North Karelia province andin the stateofMaine,respectively. Each study illustrates, variously, the editors' intent to demonstrate the consequences for forest resources


and forest users of the tensions between traditional and technological knowledge and their implementation. In the absence of a concluding chapter, it is left to Vail to attempt to draw together the somewhat disparate contributions; whilst he succeeds to some extent, there are - as he comments - 'few obvious bases for comparison' between some of the cases.

The struggle for land and the fate of the forests comprises four introductory chapters, seven case studies (Guatemala, Brazil, Zaire, Thailand, Indonesiaand the Philippines), and twoconcluding chapters, representing contributions, in addition to those of the editors, by SusannaHecht, Rachel Kiddell-Monroe and SKEPHI, Marvic Leonen, George Monbiot, Roger Plant, and a pseudonymous author. The introductory chapters focus on distinguishing the agents from the causes - and the myths from the realities -of deforestation, tracing the history of agrarian reform, and considering the circumstances of forest peoples. Each of the case studies was 'purposefully chosen to illustrate both the complexity of the problem and the diversity of situations', and each does so effectively. Although the analysis and examples are limited to the world's tropical forests, understandable given the book's origins in the WorldRainforestMovement, themoregeneral relevance of many of the issues should also be apparent; for example, Hecht's discussion of factors influencing Brazilian deforestation demonstrates the limitations of the more conventional analyses which are widely applied. The concluding chapters propose reforms to United Nations and World Bank policy and practice, and consider future options in terms which acknowledge thecomplexities of societies andeconomies.

It would be possible for some readers to take exception toelements of both books. Those familiar with the details of the case studies may disagree over detail, or draw alternative interpretations to those discussed. Foresters. in particular, might baulk at (yet more) criticism of their (usually sincere and well-intentioned) efforts, manifested in, for ~xample, Banuri and Apffel Marglin's critique of 'scientific forestry'. Natural scientistsmay finddifficulty in the frameofreference, the language and some of the imagery of the systems-of-knowledge discourse. However, I urge readers to persist; although both books suffer some of the variability of quality which characterizes many edited compilations. each presents challenges to concept and practice of which those concerned with forests should, at the very least. be aware. Foresters associated with attempts to facilitate greater local control over forests will appreciate how fundamental these challenges are.

ROMM. J. 1991. Exploring institutional ophons for global forest management In: D. Howlett and C. Sargent m). Proc Technical Workshop to q l o r e options for global forestry management. IIED, London. 186192. SACHS, W. (Ed). 1992. The development dictionary. Zed Books, London. 306 p.

P.J. KANOWSKI

K.S. BAWA and M. HADLEY (Eds.) Reproductive ecology of tropiccilforestplants (UNESCO, Man and the Biosphere Series, Vol. 7). Parthenon Publishing, Camforth, Lancs, and 120 Mill Road, Park Ridge, New Jersey 07656.1990.42 1 pp. £35. ISBN 1-85070-268-3.

This volume in the UNESCO 'Man and the Biosphere' series reports the proceedings of the five day international workshop held at the Kebangsaan University, Bangi, Malaysia in June 1987. Out of a total of 20 invited papers and 50 offered oral and poster contributions we

have here 23 p~persmangedunder six sectional headings: 1. Phenology (three papers); 2. Plant pollinator interactions, sexual systems and gene flow (six papers); 3. Seed and fruit dispersal (twopapers);4. Seed physiology, seed germination and seedling ecology (three papers); 5. Regeneration (two papers); 6. Reproductive biology and tree improvements programmes (four papers). Each section is provided with an introductory commentary, and the whole isenclosed within an introductory overview by KamalBawa,Peter Ashton andM.N. Salleh, and some concluding remarks by T.N. Koshoo.

Among the 46 contributors represented here (most papers are co- authored) there are 20North Americans, with the remainder distributed between two Australians, two British (co-authors of the same paper), four Central-South Americans, three French, one German, six Malaysians (the host country) and five Sri Lankans (co-authors of the same paper). Does this represent a crude approximation to the funding and quality research effort in tropical forest research by the countries concerned? The absence of indigenous African (the two papers devoted specifically to African topics are by the French authors), and the paucity of Latin American contributors, is regrettable.

Tropical forest 'conservation' these days tends to involve three aspects: natural parks and reserves and an emphasis on conserving the status quo; selective or sustainable logging or other exploitation of natural forests; and agroforestry, with the study of plantation species of economic value. Although each of these aspects has its own priorities, and there is certainly potential for a conflict of interests. there are also themes of common interest which can be studied to mutual benefit. One such is the advantagesof inter-digitating plantation forestry areas withremnantsofnaturalforest, whichobviously benefits forest plants and animals, but can also provide a reservoir of natural pollinators for, e.g., Bertholletia, Cedrela or Hevea crop species. The 'regeneration' of largely devastated forest areas is another topic of potential common interest, since a detailed knowledge of the reproductive biology of natural forest species may provide insights into how forests which are skewed in their composition towards trees of economic interest may be regenerated.

Although the introductory overview by Bawa et al., and also Koshoo's concluding analysis, certainly explore these three 'conservation' aspects, the bulk of the papers in this symposium volume concern studies on the reproductive ecology of natural forest communities. The book is thus a goldmine of information for researchers in tropical biology, particularly in the fields of plant - animal interactions in pollination biology and seeddispersal - seedling establishment. It would be a pity, however, if more applied biologists and foresters were put off by this apparent bias towards strictly 'biological' topics. Much of the data presented here is of fundamental importance, and will have to be taken on board by all who have an interest in sustaining forests of whatever kind in tropical regions. This book is highly recommended.

PETER GIBBS

J.M. B ~ N G A and P. VON ADERKAS In vitro culture of trees (Forestry Sciences Volume 38), Kluwer Academic Publishers, PO Box 322,3300 AH Dordrecht, Netherlands. 1992. xiv + 236pp. £42. ISBN 0 7923 1540 5.

I had mixed feelings about reviewing this book on tissue culture of trees as there seems to have been quite a number of books on this general topic in recent years including a three volume publication on Cell andTissue Culture in Forestry d t e d by the senior author in 1987. I might have been more enthusias& if thffethffehad been adifferent senior author as I thought this book might only be an abbreviated version of the 1987 publication. I was very wrong. Thisbook ga ins fmnking


authors who are experienced and active resoarchers, one of whom previously edited acomprehensive publication, and it very effectively provides an up to date account of tissue culture of trees. In fact when it was realized that I had a copy, the book was continually being borrowed by research students even though there was ready access to other tissue culture books.

The book which is typeset and hardbacked consists of eleven chapters.

In the Introductory Chapter the authors note that this book is primarily concerned with problems that specifically apply to tissue culture of tree species. Where the methodology or equipment is in normal use, the reader is referred to the appropriate literature. The authorsmake theimportant pointthat tissueculturemust beconsidered in conjunction with breeding and traditional methods of multiplicaiton such as rooting of cuttings.

The second chapter on laboratory organization and equipment (7 pages) covers a number of points for those setting up a tissue culture laboratory and a number of practical tips. It is very easy to pick out items where more discussion would have been welcome, e.g. light quality, but the authors wisely restricted the scope.

In the chapter on media preparation the authors attempt to provide information on the physiological roles of media components to assist the reader to design the most appropriate media for his species and conditions. This approach should help tissue culturalists move away from the approach of using recipes and is one of the strengths of this book. The problem of finding media suitable for a range of genotypes of the species of interest was notedas this provides unwanted selection. Some species are more sensitive than others.

The chapter on collection, sterilization, excision and culture deals briefly with the condition of the plant material where again I would have welcomed a fuller discussion of the problems. The difficulty of usingolder field grown material and browning of cultures are discussed in some detail which is very helpful.

Members at this research station found the chapter on clonal propagationparticularly valuable. The first section deals with general principles and types of micropropagation. The problem whether to clone mature ortets or to go the breeding-clone route is covered well. The section on handling cultures deals with problems of vitrification, rooting and acclimation.

Thechapter on special culture includes updates on embryo rescue, haploid culture and protoplast culture. This is followed by very short chapters on secondary metabolite production, pathology including a paragraph on quarantine, nitrogen fixation (1 page), storage of cultures and genetic modification. In the epilogue the authors note that, despite the advances in embryogenesis and in protoplast culture and manipulation, the extensive commercial application of tree tissue culture has not yet taken place, and comment that there has tobe better understanding of the genetics of crops before 'genetic engineering becomes truly effective.'

The book has a comprehensive literature reference list. Included are many photographs and line drawings which are of high quality and most do show what they are supposed to.

I consider this an excellent book for the active researcher on tissue culture of trees. It also provides an update for progress on species of interest and the subject/species index (with Latin andcommon names) is very helpful. The cost of this book is rather high but is more affordable than many other specialized books these days.

DAVW A. ROOK

P.N. BRADLEY Woodfuel, women and woodlots - Volume 1 : The foundations of a wmx&el strategyfor East Africa. Macmillan, London, 1991.338pp. ISBN 0 333 54378 5.

P.N. BRADLEY and M. HWBY (Eds.) Woodfuel, women and woodlots - Volume 2: The Kenya woodfuel development programme. Macmillan, London, 1993. 274pp. ISBN 0 333 54803 5.

This two-volume work brings together material from a pioneering exercise in research and development, the Kenya Woodfuel Development Programme (KWDP), that contributed significantly to widening the understanding of the role of trees and wood fuels in meeting rural household energy needs, and of the boundaries of the potential for agroforestry interventions to contribute to sustaining these supplies. The book covers a five year period, from 1983-88, which saw the execution of the research, and the establishment of an extension and support programme that has evolved into the present Kenya Woodfuel and Agroforestry Programme.

Theresearch phase of the study began as a followup toearlier work that had identified wood fuels as the principal rural energy option, and trees on farms as the main source of supply of these fuels. The new project was to explore the potential for increasing supplies from on- farm tree growing. At an early stage it was recognized that the woodfuel situation cannot be looked at in isolation from the rest of the farm household system, and a broader research framework was developed. This explored the multiple roles of trees on farms, alternative uses of land and other household resources, and of trees and tree products, and the impact of competing claims on household members' time, and of gender division of responsibilities and rights. on access to and use of available woodstocks.

The first discusses the conceptual underpinnings of the research model (chapters 1 to 3); describes the surveys and other studies undertaken in order to develop the necessary information (indigenous agroforestry practices and socioeconomic data. inventory of existing on-farm woody biomass, social andcultural patterns - chapters 5 to7), leading to the development of an appropriate farm typology (chapter 8); and sets out the findings and the conclusions drawn from them (chapters 9 and 10). The end product of the research was a quite detailed picture of the existence of tree stocks, and of prevailing agroforestry and wood use practices, on a range of different types of farms in the three districts within which the project was operating.

The second volume, comprising eleven papers contributed by more than twenty of the project staff and collaborators, describes how the ideas generated in the research phase were translated into practice. as aprogrammeof interventions to support agroforestry initiatives that should increase on-farm fuelwood supplies. The main focus of these chapters is on the development of agroforestry packages appropriate to the different types of farm (chapters 3 and IO), on the evolution of an extension strategy and methods chapters (4,5,7 and 9) and on the process of monitoring the impact of the programme (chapters 6 and 9).

Dealing, as it did, with a subject about which little was known at the time it started, and in which earlier work had focused narrowly on analysisof fuelwood 'shortages' and 'gaps', the project had to pioneer many of the approaches and methodological applications it used. It is a strength of the book that the authors discuss the consequent process of trial and error, and draw attention to components that proved overly costly or time consuming, to areas of study that might with hindsight have been better investigated in another way, and to issues of understanding and conflicting views that emerged as the intervention took shape. In doing so, they have enhanced its usefulness as a source of infoxmation about approaches and methods that might prove appropriate in future work.

This book is a useful addition to the literature concemed with the overlapping issues of fuelwood supply and use in rural households, and of hee management as part of farming systems. Though it had its origins in the narrow remit of encouraging farm level growing of trees


for fuel, the work reported on in the book has helped make clear the need to set research and the design of interventions into these, and other, aspects of rural livelihood systems in a much broader, holistic framework. It should be of interest to many concerned with social forestry and agroforestry, and with agricultural and rural energy programmes.

J.E.M. ARNOLD

I.R. CALDER, R.L. HALL and P.G. ADLARD (Eds.) Growth and wclter use offorestplantcltions (Proceedings of the International Symposium held at Bangalore, India, February 1991). John Wiley & Sons. 1 9 9 2 . 3 8 1 ~ ~ . £40. ISBN 0 471 93561 1.

Large-scale planting of Elcca!\ptus causes concern to local people in many tropical and Mediterranean countries. Plantations are thought to use larger quantities of inadequate water supplies and to deplete soil nutrient status. A programme funded by the British Overseas Development Administration was therefore set up to measure the water use, nutrient uptake andgrowthrates ofEuc~a1)ptus in Kamataka, southern lndiain 1987. The resultsofthese studies andof similar work from otherpartsof the world were presentedat this symposiumand are described in the chapters of this book. Coverage is thus much more specific than implied in the title. Those working on water use by forests will find much of interest here but coverage of growth is only to aid the interpretation of the water use data: there are no fundamental considerations of forest growth. The papers are grouped into four sections (Social, economic and scientific background; Forest growth; Water use studies; and Growth and water use). Inevitably for such volumes, the 30 papers vary considerably in length, standard and in the interest orrelevance of theircontent. Some are very short(1 or 2 pages) and are subjective or descriptive in approach; others are proper scientific papers combining original data with review and modelling.

Water use by forests and 'the Eucalyptus controversy'. which is a subset of this subject. are important issues. This bookis thusof interest and relevance. The southern Indian studies have not answered all the questions of the Eitcalypzus debate but, along with the Australian data some of which are also presented here, they take us close enough to an understanding to be able to guide planting decisions. Eucalypts douse a lot of water (up to c.26(X) mm per year). In the dry Indian site they used the same amount of water as the indigenous dry deciduous forest. Water use by both forests was higher than that of agricultural crops. If we put aside amenity and ecological implications, the merits of plantation forestry could bejudged by comparison ofeconomic returns from timber production versus the reducation in water yield (or rather the value of the decrease of water resource which will vary between different alternative uses). It may be unusual for such comparison to be unfavourable to forestry; where this is so, ways of improving the water use efficiency of the forests will become relevant.

As well as the issues touchedon above, the techniques and process studies which are described in some chapters are valuable. I enjoyed the paper by R.L. Hall on modification of climate by forests. There is little published on this subject and yet forests can cause significant humidification and cooling of the atmosphere on the scale of some IOOs of kms. which will benefit both agricultural productivity and living conditions. The considerations of where and when soil water availability or light interception limit growth are of considerable interest and here again, discussion reflects contemporary understanding of growth constraints. On balance I think the merits of the original, thought-provoking papers outweigh the disappointment of the short, superficial papers. I am pleased to have acopy of this book and expect to refer to it in future.

P. FREER-SMITH

DEBRA J. CALLISTW Illegal tropical timber trade: Asia-PUC$C TRAFFIC International, 219C Huntingdon Road, Cambridge CB3 ODL. 1 9 9 2 . 8 3 ~ ~ . nps. ISBN 0 9476 13889.

Debra Callister's interim report was commissioned in 1989 and published in 1992. It has been widely distributed. It tackles illegality issues head on, with many examples of alleged malpractices in Asia- Pacific countries. A significant amount of evidence is from cases which have come to court, which indicates a degree of successful vigilance. Examples range over transgressions at village level, up to large scaleinternationalinvolvement, from illegality innon-industnay non-commercial sectors - agricultural clearance, fuelwood collection, and logging by settles - to illegality in the industrial sector, in particular that proportion of industry which exports wood products. Examples include activities in areas beset by political strife.

Governments and named companies have had time to consider individual allegations and to act accordingly. The survey makes reference to Cambodia, China, India. Indonesia, Laos, Malaysia, Myanmar, Papua New Guinea, Philippines, Solomon Islands. Sri Lanka, Thailand and Vietnam. Its definition of illegality includes illegal removal of commercial or non-commercial trees, smuggling, transfer pricing, under-invoicing, wrong classification of species, and illegal processing.

All societies develop rules and regulations to stop practices by some of their members whichare harmful toothers. It is an observation, not an excuse, to say that malpractice in the forest products sector is only another example of mankind's tendency to break the rules. The real issue must be to create awareness of the losses to society through uncontrolled activities in forests. Confrontational tactics to force the pace towards more appreciation and protection of forest values are commonplace. Those of us on the receiving end may not like it very much. But any success in curbing illegal activities should be of long term benefit inensuring future maintainable supplies offorest products and a more equitable distribution of earnings from logging - not to mention the environmental gains.

Governments are now more aware of the roles of their forests. Peninsular Malaysiais noticeably stricter in its policing of forests, and Indonesia is bearing down more strongly on illegal logging. Both are aware of the limits to extractive capacity and the need to ensure the perpetuation of raw material resources for avaluable industrial sector. This is what TRAFFIC'S report would call a narrow focus, but financial rewardcan be aprime motivator forgovernments, communities and industries to want to safeguard assets as well as for wrongdoers looking for a quick buck.

The fundamental problem stems not from developing appropriate laws and policies but in making them work. It depends on the will and capacity of governments to enforce decisions. Governments in the developed world cannot make their citizens totally law abiding. In the developing world it is even more difficult. The author lists 30 recommendations. A number of them relate to strengthening the resources and motivations of Forestry Departments. Effective control starts with the standing tree, but Forestry Departments need to see sufficient share of royalties and fees associated with wood extraction coming intotheircoffers tocoverthecost of supervision and monitoring. Their staff need to be well paid and well equipped inorder to minimize the attraction of tempting offers from the less scrupulous.

The author puts considerable importance on the role of the International Tropical Timber Organization. ITTO ought to be the ideal forum for non-confrontational discussion of difficult issues like illegal logging and for the exchange of experiences in policing methods and forms of checks and balances.

Some of the evidence provided in the report is said to be anecdotal and some of it is liot very precise. Une perhaps wwld not expect otherwise, given the subject matter. Just what proportion of Asian

Commonwealth Forestry Review Vohme 73(1), 1994 61

tr& is affected may be difficult to ascertain, but it is sizeable and in the long term not good for the forest or the trade.

G.M. PLEYDELL

F.W. CUBBAGE,J. O'LAUGHLIN,C.S. BULLOCK 111 Forestre~ource policy, John Wiley & Sons, Baffins Lane, Chichester. West Sussex, P01 9 1 UD, UK. 1993.562pp, £57 (hardback). ISBN O- 47 1-62245- 1.

This book is a textbook on forest resource policy for upper level undergradwate or postgraduate students. It is mainly concerned with forestry policy in the USA. and draws on the extensive knowledge of the authors in this field. who come from the US Forest Service and Universities of Idaho and Georgia. The book is in three sections: Processes, Participants, and Programs, with three appendices dealing with legal issues and acronyms used in the hook.

The first part of the hook is mainly theoretical. It starts with a brief overview of the state of US forests, then goes on to discuss the basic elements of forestry policymaking. Once i t has set out these elements, it discusses how policy is formulated, who the main participants are. how decisions are taken, and the possible consequences of decisions. It also contains some brief examples of how recent forestry policy decisions have been taken in the USA. Because this part of the book is quite theoretical, many of the issues raised in this section are likely to be relevant to forestry policymaking in other countries.

The second section of the book deals with the participants in the policy process in greater detail. The information contained in this section is more specific to the situation in the USA. The first three chapters deal with the main US government agencies, and are unlikely to be of much interest to anyone outside the USA. The rest of the section deals with interest groups and the media. This part of the book shows quite well how interest groups operate and develop. and. by going through the history of interest groups in the USA, it show.; how policy issues can be significantly altered by their intervention\. The section on forestry and the media is also enlightening. It discusses the roles of' different media, their effectiveness at putting across different kinds of messages, and even contains some useful guidance on how to deal with the mediaand public relationsgenerally. Forestry iscurrently a topic that does seem toget a great amount of mediaattention in many countries, so this is a very useful section of the book.

The final section of thehook examines specific forestry policies in the USA in detail.Topics covered include: public ownership. multiple use managament, regulation, wildlife management and incentives for forestry. Many of the chapters start with an historical analysis of how policies in each particular field have evolved in the USA. Again most of this section is very specific to forestry policy in the USA. but the examples given could be used to illuminate forestry policymaking elsewhere. The final chapter examines forestry policies in other countries, but it contains so little material that it can't be considered anything other than a very brief foray into this subject.

On the whole, the book is well written and easy to wad. At 562 pages it isquite large, but iseasy todipinto, andcontains a very detailed index and list of contents. For most of the topics, it discusses the theory behindthem, givesabrief outlineof US history in thearea, and finishes with adescriution of the currefit situation in the US. In many cases. the text also makes very good use of examples. Very few of the topics are covered in exhaustive detail, but each chapter also contains a full list of references for further reading.

The book gives a real flavour of the sort of problems that are enccxmeid in forestry policymaking in the USA. The USA has one of the most demorratic systems of governmefit in the world, and this bodr clearly shows how this has led to the creation of many interest

groups, each trying to move forestry policy in a direction that suits them. It also gives the reader some guidance on how to deal with such influences. The text thoroughly discusses all of the participants in the forestry policymaking process but avoids, quite rightly, judging them in any way.

The book isaimedat students, andcontains aconsiderable amount of historical data from the USA which would make it of interest to students of forest history as well as policymaking. However, forestry policy is asubject of current importance to many foresters and forestry officials, andone which I suspect many have not been formally trained in. I would therefore, recommend this to anyone involved in forestry policymaking who wishes to get a clear explanation of the concepts behind policymaking. and some good examples of what is done in a major developed country.

ADRIAIL. WHITEMAK

JOHN D A R G A V E L ~ ~ ~ RICHARDP. TUCKER (Eds.) Chur~gi~ig f ~ ( ' $ ; l '

jimsrs: histor-ical pevspectiia or1 tlir ,forest e c m o t i i ~ rf thc ~Pac!'ficBa.siri. Forest History Society, Durham, N. Carolina, and Duke University Press, 1 Gower Street, London WClE 6HA. 1992. vii + 156pp. £28.45 hbk. & 12.95 pbk. ISBN 0 8223 1262 X (hbk); 0 8223 1263 8 (pbk).

'~heauthorsedit(andcontrihuteto)abookcomprising papersdiscus5ed at a conference sponsored by the Forest History Society and the lUFRO Forest History Group, as part of the Pacific Science Association's Seventeenth Congress held in Honolulu in 199 1 . The original stated purpose was to discuss growing linkage5 between tropical and temperate forests of the Pacific Basin as forest economies and trade developed in the region. 'Trade and its effect., in forests' purport to be the central themes of the papers collected. echoing 'linkages across the Pacific'. Theobjectives wereover-ambitious and the dozen papers published represent an uneven mixture of anecdote. pedantry and (too rarely! serious historical studies.

Dargavel ('Incorporating national forests into the new Pacific cconomic order: processes and consequences') provides a statistical background which is. inevitably. inadequate andout ofdate. Like most paperson forest products trade. it ignorespricesandmicro-determinants of market drive.

The next four papers concern Japan. Totman ('Forest products trade in pre-industrial Japan')gives asketchy and unreferenced history . - of pre- 1860forest exploitation anddepletion. A single sentence refen to reforestation and forest management, claiming a reversal in a 150- year downward trend in timber production after the 1770s. Henning ('The Pacificnorth-west lumbertrade and Japan 1898- 19 14' ,offers an interesting and well-researched account based on adata basecomprising records of 339 voyages from the Pacific north-west to Japan canying some250,OOOm'of 'lumber, laths and shingles' during the period. The paper hints at fascinating early differences in shipping developments between the US and Japan (steam versus sail and private versus public enterprise) but the paper is too short to do more than whet the appetite for more. Waggener and Backman ('US-Soviet roles in pacific rim softwood log tiade: the Japanese market as the competitive link') present data on changing patterns of trade. The paper is weak on other than very recent trade (post-1960) and it isdifficult tojudge the quality of the statistics (which are unsourced: and there are apparent discrepanciesbetween Figures 2 and 3 relating to net imports). Despite the more detailed analysis of recent trade, Japan's pioneering role as an 'in-transit- producerof plywood is not mentioned and the discussion of non-tariff barriers to trade is perfunctory. It is of interest that while import tariffs in Japan and 'political' restrictions on log exports from Canada are identified as distorting features, below cost timber sales in


theUSAarenot similarly viewed. In ahistorical publication, onelooks for greater objectivity. Finally, in this fmt series of papers, Kakizawa ('The current stateof forest management andindushv in the Soviet Far East and the Japanese-Soviet timber trade') provides a case study of perestroika in action in the Soviet Far East. It is much too early to discuss effects of change but this annotation is a useful record of organizational (bureaucratic) reform and the age-old socialist conflict between forest management and forest industry.

Menzies ('Sourcesof demandandcyclesof logging in pre-modern China')offers the only paper in this collection withagenuinehistorical perspective. Menziesposes aquestion as to the nature anddetermination of deforestation in southern Shaanxi province; he offers hypotheses and identifies physical and ecological constraints to explain why land conversion took place in some parts of the area and not in others; and he relates episodic development of the timber industry to cyclic demand influenced by economic conditions in north-western China and requirements of the lmperial household. Menzies has been charged withthe preparation of the foresby volume in JosephNeedham's gigantic work Science and civilization in China; the professional approach demonstrated in this paper makes this reviewer impatient for the publicationof the moreambitioustreatise. Lim Hin Fui ('Aboriginal communities and the international trade in non-timber forest products: the case of Peninsular Malaysia') is anecdotal but well researched and explodes the myth of subsistence agriculture survival. It is only marginally related to the purported themes of the Conference proceedings. Flint and Richards ('Contrasting patterns of Shorea exploitation in India and Malaysia in the 19th and 20th centuries') like Menzies pose a question - in this case the marked difference in the history of Shorea exploitation between India and Peninsular Malaysia - and proceed to explain it in cultural and economic contexts. Except for the final two paragraphs which are irrelevant, subjective and emotive, the paper reports the kind of study which can justify the existence of the Forest History Society. Tucker('Managing subsistence use of the forest: the Philippine Bureau of Forestry 1904-60') records little about either management or subsistence and has to be adjudged lightweight. Cox ('The evolution of Hawaiian forestry: the web of influences'), though brief, is objective and makes interesting reading. The overriding importance of watershed protection and the domestic drive for forest management are notable; as is the final sentence, to the effect that 'Gifford Pinchot deserves hardly more than a footnote'!

The final papers are close to home for this reviewer. Jackson ('Forest policy and trade: the New Zealand experience') provides a factual and competent outlme but a comparison with the Amazon is naive, and the 'linkages' are contrived. Events in New Zealand have been much influenced by stupid and doctrinaire market intervention and over-regulation, culminating in the biggest distortion of all, the sale of State Forest Assets. Roche('Privatizing the exotic forest estate: the New Zealand experience'), who is New Zealand's best known foresthistorian,outlinesaview ofhow and why theLabourGovernment decided to remove the natural forests from commercial management and to offer the plantation forests for sale. The account may or may not be factual (it is impossible, at this stage, to draw conclusions; but the need to rely on press reports and statements issued by the bureaucracy for information, and the scarcely credible variation in forest valuation - from one to seven billion dollars - provide no basis for a definitive history). North American studies of New Zealand's privatization programme have at least been given a geographic perspective. (See e.g. Marion Clawson, 'Public forestsin New Zealandandin the United States', Resources for the future Discussion Paper series No. RR88- 01, 1988; and R.L. Fischman and R.L. Nagle, 'Corporatization - implementing forest management reform in New Zealand' - a study of the Environmental Law Institute.) In New Zealand, it is too early to provide objective evaluation of the context or to make other than political judgements.

It has to be concluded that this volume of collected papers fails in its objectives. It does, however, emphasize the need for proper and professional studies in forest history. The successes of this book (primarily the paper of Menzies and to a lesser extent those of Henning and Flint and Richards) whet the appetite and demonstrate, perhaps, that conference proceedings (with contrived themes) do not provide the right medium or the needed discipline.

The volume is reasonably priced and is worth perusal if only for the atypical contributions noted above.

DENNIS RICHARDSON

A.J. GRAYSON Private forestrypolicy in Western Europe. CAB International, Wallingford, Oxon OX10 8DE, UK. 1993. xx + 329pp. £40 (US$76 Americas only). Hb. ISBN 0 85 198 843 1.

Private ownership is the main form of tenure of woodlands in Britain and Western Europe and the private fo~estry policies of governments are obviously of prime importance to owners and to foresters.

Over the past decade there have been significant changes in the aims of government forestry policies in Britain and the rest of Western Europe. Furtherchangescanbeexpected - for example as aconsequence of the reshaping of the Common Agricultural Policy; the advent of the single market within the European Community; and the continuing shift of emphasis from wood production towards a greater provision of non-market benefits such as nature conservation and recreation.

Against this background, and the possible increased interest from overseas investors in investing in afforestation in Britain, the Scottish Forestry Trust commissioned a study aimed at extending the debate on the foundations of forest policy with special reference to private wmdlands in Britain and a number of other countries that appear relevant to the British situation. Methods of policy formulation, trends in thecontent and balanceof objectives, and methods of implementing policy are covered. The implications of the findings of the study for the framingandimplementation ofpolicy in Britain are fully discussed.

The book is divided into 18 chapters of which 10 describe and comment on forestry policies in individual countries - Britain, Ireland, France, Belgium, the Netherlands, Germany, Switzerland, Denmark, Norway and Sweden.

These chapters are comprehensive in their coverage of relevant facts and up to date in pointing to recent trends in forest policies. In addition to basic statistics on forestry they include relevant information on matters such as legislation (current, proposed); administration; property, income, wealth and inheritance taxes; grants; extension and research. Outstanding issues being debated in the different countries are discussed.

A chapter on private forestry considerations in other selected countries gives highlights of relevant private forestry policy considerations in Australia and New Zealand. There are thumbnail sketches of the position of private forests ineasternEuropeancountries.

Thereare major chapters ondevelopment ofpolicy; implementation of policy; private forestry in a European context; current policy issues In Western Europe; conclusions on policy implementation; and implications of findings for the framing and implementation of policy ~n Britain.

Fourappendicescoverpopulationand forest statistics for countries surveyed, the European single market; the application of policy analysis; and sterling exchange rates. Each chapter and appendix is supported by an extensive bibliography.

The enormous amount of information that has beea marshalled is made more interesting and more useful than some of the compilations of data published by official bodiesby virtue of the authcu's comnknts on and evaluations of the material he h@ coll- The


i n d i v i w country chapters have been reviewed in their entirety by prominent foresters in the respective countries and the names of other ove;rseas foresters who have provided material and guidance occupy two pages. This gives the book a great authenticity as a source of facts and informed opinion.

Inhis preface, the author says that Britishforestry literature shows a marked imbalance towards biological and technical matters; the policy and economic aspects of the profession's concerns are left largely untouched by published papers. One pertinent reason, he suggests, is that those employed in the public sector are hardly encouraged to write about, still less tocriticize public policy. Another is that within the forestry profession generally, few seem to have the time and the inclination tocany out the necessary research orto use the written word to convey their thoughts. It is his hope that this work will stimulate morediscussion, reporting and debate on this most influential - - of all topics, the role of government in setting the aims and goals for foresters and the means for them to implement policy.

Based on evidence he has collected from abroad and his ideas founded on those examples, the author gives by way of conclusion twelve principal recommendations for those engaged in developing future forestry policy.

Although the study draws its material mainly from Britain and selected Western European countries, and Norway and Sweden, and is in its own right an excellent book of reference for this geographic area, it is of interest andofrelevance to amuch wider audience. It deals in an analytical and philosophical way with important questions concerning policy formulation and implementation, the content and balance of objectives of policies and addresses the economic aspects of the forestry profession's concerns. The publisher's claim that the book will interest academics and practitioners concerned with forestry, from the perspectives of economics, policy-making, planning or investment, is indeed justified.

E.G. RICHARDS

M. PENKA, M. VYSKOT, E. KLIMO and F. VASICEK Floodplain forest ecosystem 11: afer water management measures. Developments in agricultural and managed-forest ecology Volume 15B. Elsevier, PO Box 1991, 1000 BZ Amsterdam, Netherlands and PO Box 882, Madison Square Station, New York NY 10159, USA. 1991.630pp. Dfl. 450.00; US$231.00. ISBN 0 444 98756 8.

This volume reports the results of the second part of a scientific study of the effects of water regulation projects upon the floodplain forests of Southern Moravia in what is now the Czech Republic. The project waspart oftheUNESCO 'Manand the Biosphere' research programme. The fmt volume, published in 1985, described the area before water regulation. This volume describesthe changes in the ecosystem after the former natural annual inundation of the deciduous floodplain fonst and water meadows along the lower reaches of the ~ ~ j e and Morava rivers was stopped in 1972 as a result of engineering of river channels and the construction of storage reservoirs and imgation channels.

The wider interest of this volume is that it examines in depth the process of intensification of agriculture and forestry by means of flood control and irrigation which has already proceeded to a great extent in many other developed countri'es and thereby degraded or destroyed wetland ecosystems. The book will be of interest to ecologists, water managersand f m s t nseatchers and @icy makers i n ~ u r o b and other temperate regions interested in managing remnant floodplain forests for a range of objectives and also to those who wish to recreate floodplain fozest of a n&tural character.

water wgulation gs~vented spring flooding which had previously been vital in maintaining high water tables in low-lying gatts of the

forest and bringing in nutrients in silt. Habitat diversity was reduced as pools, backwaters andchannel meanders were eliminated. The soils and the plant and invertebrate communities grew more uniform and less diverse as soils became drier, better aerated and higher in available nitrogen. Small mammal communities also changed, insectivores being partly replaced by 4 - e a t e r s ; and the latter reduced the regeneration of seedling lime, hornbeam and oak. A major change was the reduced vigour of the herb layer which declined by 60-70% in biomass in 10 years in the wetter areas. Changes in composition of shrubs and tree layers also started to appear within 10 years, including dieback of poplars and ash.

Although the timber production prospects have been improved by the prevention of floods because tree canopies can be denser, a greater

of oak can be grown and the ground is easier to work, the study shows how this is at the expense of reduced biodiversity. It therefore has value as areference for management of floodplain forests elsewhere.

The book is rather heavy going with the detailed descriptive style ofEasternEuropeunleavenedby comparisons with the wider literature. Tables are frequently overladen withdata and more summary tables or figures would havebeen better. Summaries of each of the many short chapters and the final conclusions did help greatly to draw things together, however.

G. PATTERSON

R.T. WNSLEY (Ed.) The role of trees in sustainable agriculture. Kluwer Academic Publishers, PO Box 17,3300 AA Dordrecht, the Netherlands. 1992. vii + 186pp. Dfl. 125 / US$84 I UK£50. ISBN 0 7923 2030 1.

The book would be more accurately titled 'the role of trees in sustainable agriculture in Australia', as it consists of six review papers presented to a national Australian conference, 'The role of trees in sustainable agriculture', held in Albury, Victoria, in October 1991. The papers have also appeared in a special issue of the journal Agroforestry Systems (Volume 20, 1992).

The purposeof the review process was to bring together the results of agroforestry research and development throughout Australia, including previously unpublished material. Each of the papers was based on six or more state reviews, and a comprehensive coverage has been achieved with a style that is oriented to practical considerations, and a heavy reliance on grey literature. A wide range of agroforestry practices areencompassed, from timberbelts in the wet temperate zone to the useof salt-tolerant shrubs for fodder in semi-arid rangeland. The book is likely, therefore, to have considerable appeal and to stimulate readers in both temperate and tropical countries where agroforestry research and development has not yet reached the stage where farmers are adopting practical recommendations based on research results.

In terms of subject matter, the most valuable contribution is an excellentreview of tree planting for dryland salinity control. There are also focused treatments of the use of trees to provide shelter, the retention of native woody vegetation on farms and of the integration of wood production in f&ng systems. The usefulness of reviews of tree fodder and alternative products from woody perennials are, however, somewhat constrained by the patchy nature of the research and development activity in these areas.

The six reviews are brought together in an overview paper, which contrary to what is said in the preface, appears in the middle, rather than at the start of the sequence of papers. This is somewhat superfluous and repetitive given that the other papers are, in themselves, reviews. From a scientific standpoint, some of the material falls a little short of usual standards for an internationally refereed journal; output from simulation models is presented without a clear statement of all of the


assumptions involved and the input data used (e.g. p.159, 174-176); linear relationships are presented without regression coefficients or any estimateoftheirfit(e.g. p. 10,112); anddata, used to support points made in the text, are presented without description of how they were collected or any estimate of error (e.g. Figures 5 and 6, p.80-81). At one point, discussion is based upon a figure (Figure 5, p. l l l ) in which the number of agroforestry research projects (size not defined) in each of six states is plotted against an estimate of the number of farmers practising agroforestry in the state, with a straight line drawn through four of the points!

While the text and references are clearly presented, the size and definition of figures varies considerably. Some figures use a large amount of space to convey very little information (e.g. p.158, 11 l), while others, because of a combination of their small size and poor definition, are difficult to make out (e.g. Figure 2, p.91). An identical map of the agoclimatic zones of Australia appears three times in the book (p.89, 107, 121) at different sizes, underlining the fact that the book consists of a set of separate, self-standing papers, rather than an edited whole.

There is undoubted value in this issue of Agroforestry Systems being made available separately in book form, and it will appeal to people, particularly those with a practical interest in agroforestry development, in and beyond Australia, who do not have access to the journal.

FERGU~ L. SINCLAIR

B.A. ~ZICHARDSON Wood preservation (2nd e h . ) E. and F.N. Spon, 2-6 Boundary Row, London SE1 8HN. 1993. xii + 226pp. £45. ISBN 0 419 17490 7.

Tlris is a second edition of a book fmt published in 1978. Contents include chapters on preservation technology, wood degradation, preservation sy stems, preservation chemicals and practical preservation. Three appendices deal with: selection of a preservation system, wood borers, and wooddestroying fungi. Theauthor specifically statesin the preface that the book is concerned with wood preservation, not with wood deterioration, and in spite of the title of the second chapter this is largely true. An excellent survey of the history and development of preservation technology is given in the introduction to chapter 1 which sets the scene for more recent developments.

This is followed by a brief section on preservation principles and wood structure which is rather less comprehensive than it could have been; more could have been said about the problems of CCA retention in the heartwood of hardwoods whlch has received considerable attention recently. Wood deterioration is dealt with well and so are preservation sy stems whichincludemechanisms, application techniques and evaluation systems. Preservation chemicals are also well covered although a considerable amount of space is given to chemical combinations which were used historically but are no longer in use. The account of how the chemicals used today were developed over time is, however, interesting.

Some mention is made of the problems of the toxicity of some of the more effective preservatives as perceived by some environmental agencies, and the pressures that are developing to find alternatives whichare less toxic tomammals and birds, butthiscouldhavereceived rather more attention, as could progress in recent research into finding new methods of inhibiting organisms which attack wood. Thechapter on practical preservation does, however, deal with health and the. environment but concludes that most commonly used preservatives are not harmful if properly used.

A very brief section at the end of the book, entitled further &g, gives three sources which can provide lists of literature for further reading but nowhere else is there a list of references. The author has

thus covered a wide subject in such a way as to give a good overview of the whole of wood preservation, but the book is neither a technical handbook nor doesit provide the latest informationonresearch. It will, however, give the reader a useful and interesting grounding in the subject.

R.A. PLUMI?I~E

ANDREW DENNY RODGERS III. Bernhard EduardFernow: a story of American forestry. Forest History Society, Durham, North Carolina, USA. 1 9 9 1 . 6 2 3 ~ ~ . £18.95. ISBN 0 89030 047 X. (Reprint of 195 1 publication by Princeton University Press.)

'What is a forester? A Robin Hood who takes from the rich to give to the poor? These were the questions put to the young American woman, Olivia Reynolds, who attracted (and manied) the Prussian forestry student Bernhard Eduard Fernow who became the first professionally trained forester in the United States of America. 'If a certain young girl had not gone to Germany in the year 1872 and fallen in love with acertain young forestry student .... the forestry movement in this country might have been delayed for a decade or more' she modestly claimed. Femow had been selected by an uncle to inherit a large estate in Prussia on condition that he became suitably qualified and Fernow consequently spent four years at the forestry school in Miinden, Hanover. Family hopes were dashed as he followed his future wife back to the USA at a time when public debate was beginning and various Government Acts were being passed and amended to set aside forest land for the economic and protective purposes of the nation. Femow became a US citizen in 1883.

Working as a consultant engineer to mining companies and responsible for the management of forests for the production of charcoal, Fernow observed, learned and wrote about the conditions of forestry and tree growth in the eastem USA. He started writing articles in 1881 and began to support the cause that the nation's natural resources should be conse rved... 'An industry dependent on charcoal as fuel must, to be permanent, maintain large forest areas, thus benefitting the surrounding country.' ~e referred to the 'science or art of forestry' and pleadedforanemphasison 'forest management' rather than 'forest culture', probably the first use of the term in American literature, and he included the concepts of wildlife management.

At the first American Forestry Congress in Cincinnati during 1882 - - Fernow began to make contacts among politicians, industrialists and land-owners that ledCanadiandelegates to invitehimand thecongress to Montreal in 1883 and eventually he was appointed secretary. He gave a paper emphasizing the need todetermine the scientific principles of forest growth and management at a hme when the only relevant science concentrated on taxonomy a n d d e s c r i p t In his biography of Femow, Rodgers goes into great detail of the politics, persons and papers of these Congresses and of the Federal and State enactments related to forestry; many of the papers sought to educate the public as well as government.

In 1885 Fernow organized a New York State forestry association; ~t was shortlived but through it he became acquainted with Theodore Roosevelt, later to become President of the USA and a committed conservationist. Shortly them&e.rFemow, aRepublican, wasappointed as chief of the Forestry Division of the United States Depamnent of Agriculture by President Cleveland, with aDemocratic administmtion in 1886; shortly afterwards he published a major mport on 'General principles of forestry'.

During Fernow's time as Chief (Chapter 3), forestry progressed throughthree main phases -theforpstterrarvationpoky,theintroduction of management prsrtice, and aprolpammeof timbrphysics. Chapter 4 considered the new forestry Iegislatim and Chapter S the reorganization of forestry and forestadmini~tratim. All three chapters


included the minutiae of conferences and correspondence seeking to establish forestry as a major government task and policy issue with great details of the frnests and legislation of individual States.

However,unlikeGiffordPinchot, Femow wasnotapoliticianand, when a Colonel Fox proposed the idea of a demonstration forest tract in the Adirondacks and suggested that Cornell University might undertake the experiment with a new College of Forestry, which was established in 1898, Femow himself applied for the Director's post. He established the first undergraduate degree course in forestry in the USA with Filibert Roth as lecturer. This was followed closely by the Yale School of Forestry in 1900 with Henry Graves, F.H. Newell, Edward Bowers and James Toumey as staff members and Gifford Pinchot (Fernow's successor as chief federal forester) as special lecturer. Federal budgets and the demand for trained foresters rose dramatically as both public and private programmes spread in the United States and Canada, and conflict emerged between lumbermen and the proponents of scientific forestry.

Forestry education spread westwards (Chapter 7) and to Canada (Chapter 8); after having prepared a prospectus of a new course and Department of Forestry for Pennsylvania State College, Fernow became the first teacher but shortly moved toTorontoas Dean and head of the forestry school. Chapter9details hisexperience there, his views on silvicultural practice, and the historical development of forestry in Canada.

For long periods, Fernow was honorary vice-president of the American Forestry Association and member of the editorial board of the Society of American Foresters (Chapter 10). He became President ofboth the SAFand theCanadian Society of Forest Engineers in 1914. During this tenure he initiated a programme to develop forest terminology and continued to stress the importance of research in forest biology and management.

The final chapter is entitled 'Fernow, first forester of North America' but this thought permeates the whole book. Femow died in 1923 leaving a large number of students who rose to senior positions. a number of forestry colleges, departments and associations that had benefitted from his inspiration, and above all a great public and political awareness of the values of forests.

It is unusual to be asked to review a book 40 years after it was first published but this 1991 reprint of Rodgers' detailed biography of Fernow is still an unparalleled history ofaman,apublic movement and two national services. The book is a classical historical study of published and unpublished records with frequent time cross-over?; it is often difficult to trace a particular event or sequence and the chapter titles do not help. However, a republication is timely now in that it reminds us of the origin and political conflicts of the conservation movement through a study of the life of one of the key figures of the time. As forestry worldwide is undergoing major reappraisal, it is to be hoped that an update of north American forestry since 195 1 will be written around the turn of the century but there may not be such an outstanding individual to reflect the period as Bernhard Fernow.

J. BURLEY

E. R ~ H R I G AND B. ULRICH (l!!&.) Tempemte deciduous forests (Ecosystems of the World Vol. 7). Elsevier Science Publishers, PO Box 21 1,IWO AE Amsterdam, Netherlands (USAICanada: PO Box 882, Madison Squm Station, New York NY 10159), 1991. xi + 635pp. Dfl. 440 1 US$225.50. ISBN 0 7923 1643 6.

Ten authors have cmaibuted the chapters of this book which deals w i t h t e ~ d e c i d u o u s forests of north and south America, Asiaand ?&rope. Those in other parts ofthe southern hemisphere, Australia, New Zeatm&ami South Africa, are not considered on the gmnds that they belong totypes in whichperiodicdroughtra~er than w a n versus

cold seasons determines the nature of the flora and fauna. Section I (Chapters 1-6) provides a general introduction in which temperate deciduous ecosystems are described. Functional aspects, including ecochemistry andboth primary and secondary productivity are covered in Part I1 (Chapters 7-9). The final Section (Chapters 10-17) is concerned with descriptions of the regional deciduous forests, covering flora and, in mostcases, fauna. Because the regions occupied by these forests are, in general, so favourable to human settlement and have thereforebeen enormously changed, the authors have used theconcept of potential rather than actual natural vegetation. They admit that this is neither entirely satisfactory, nor free of speculation.

The book does much to summarize the vast amount of information that exists on temperate forests for most parts of the world. Among its considerable values is the information on the general characterization of temperate deciduous forests, the processes within them, and the classifications of vegetation together with usually well illustrated information on the distributions of the im~ortant species and forest types in different regionsl. For this it is an invaluable source of reference which is greatly assisted through excellent author, species -and general indexes. The book is not always easy to read - several chapters have very tediously constructed, long sentences. Readers who are not intimately knowledgeable about the classification of vegetationandphytosociology - such as this reviewer - might also find some parts difficult to understand owing to a profusion of highly technical terms. However, the book is an extremely valuable guide to understanding temperate deciduous forests.

PS. SAVILL

H.T. SCHREUDER, T.G. GREGOIRE and G.B. Woon Sampling methods formultireso~rrce forest inventory John Wiley & Sons, 1993. xv + 446pp, 24xl6cm. Index. Price £66.00. ISBN 0471552453.

The title of this book appears very opportune as increasing emphasis is being placed on the management of all resources within a forest. The authors (two from the USA and one from Australia) have written this book as a reference text for forest biometricians and for graduate students withastrong interest in becoming forest inventory specialists. They point out, quite rightly, that many forest inventory methods appear to have developed almost independently of the corresponding theory on statistical survey sampling and that there is aremarkable lack of awareness of the forestry literature by statisticians. They hope to contribute to the cross-fertilization between survey sampling and forest inventon/ methodologies that could lead to more efficient - techniques. In this respect, the inclusion of a large chapter on mensurational aspects would indicate that the book was also intended for statisticians moving into forest inventory.

After an introductory chapter which briefly outlines the history of statistical survey and forest resource sampling, there are five chapters on statistical theory in relation to forest sampling. These take up half the bookand follow the logical pattern of basic definitions, distributions and sampling theorems, followed by single level sampling strategies and then multilevel sampling. Finally, there is a chapter on model- based inference. Throughout this statistical text the needs of forest inventory are paramount both for the estimation of the current status of a forest and also for the estimation of growth. A useful section is a comparison of some selected sampling strategies - fixed-area plots versusvariableradiusplots andsampling with probability proportional to size versus Poisson sampling are two examples. Besides the sampling techniques and variance estimators that should be familiar to most foresters, the authors cover a number that will be unfamiliar such as centroid and importance sampling, ranked set sampling and the use of jacknifing and bootstrapping for variance estimation.


The chapter on mensuration, the largest in the book at nearly 90. pages, is excellent. Rather like Mike Philip's book (Measuring trees and forests), it deals first with single trees and then with stands. It concentrates very much on the sources of error and bias in the various measurement and estimation techniques. There are small sections on such topics as biomass, site assessment, regeneration, understorey vegetation and forest health. Different aspects of sample plots are described with particular attention to the adjustment for edge effects. This chapter, as with some of the others, ends with over 60 problems which this reviewer has every intention of using in his inventory classes (the answers are also given!). The next chapter gives a brief review of avariety of topics including small areaestimation, sampling wildlife and fish populations, sampling rare populations, spatial distributions and remote sensing. This is followed by a short chapter on missing data analysis, measurement errors and Bayesian and likelihood estimation. The last chapter gives some indication of future directions in multiresource sampling in forestry.

This is a very valuable book as there has been a need for a textbook on sampling that was written specifically for the forest inventory specialist. It should beemphasized thatthis is astatistical textbook and, apart from the chapter on mensuration, requires some familiarity with sampling theory to obtain the most out of it. The title, in some ways however, isalittlemisieadingas the bulkof the bookisconcerned with the more traditional aspects of forest inventory. I would have liked to have seen more on the multiresource aspect including subjects such as biodiversity. The emphasis was understandably much more temperate rather than tropical; a small section on particular problems arising in moist tropical forests would be a useful addition to future editions. Notwithstanding, it will no doubt become an essential text for those concerned with designing forest inventories and should also be on the shelf of all biometricians and statisticians who may from time to time be consulted by foresters over inventory problems.

HOWARD WRIGHT

R.E. SCHL~LTES AND R.F. RATTAUF. The healing forest: medicinal and toxic plants of the Northwest Arnazonia. Second printing 1992. Dioscorides Press (an imprint of Timber Press Inc.), 9999 SW Wilshire, Portland, OR97225, USA. 500pp. withnumerous figures andhalf-tones. $59.95 + $4 shipping and handling. ISBN 0 931 146 14 3.

This is a fascinating compendium of information on what many foresters too facilely consider to be 'minor forest products'. It is based onalmostfifty years' fieldwork by Schultes,whowasoftenaccompanied by Rattauf of the pharmaceutical industry. There is a foreword by the President of the World Wide Fund for Nature, a preface, which is a sketch of the vegetation of the region, acknowledgements and 'abbreviations and conventions'. The bulk of the book is a catalogue of the plants and their uses. arranged in alphabetical order of family, genus and species.

Each family entry comprises a description of the family, followed by generic accounts within whichindividual speciesandtheir attributes are discussed, beginning with precise bibliographical citation of the specific binomial and lists of specimens examined; there follow the uses and then chemical information. The most striking feature here is how often the last item is completely lacking, showing how much work still needs to be done to test the plants and ascertain their efficacy in biochemical terms. The family entry ends with a general bibliography in a rather abbreviated form.

The work deals with some 15 16 species in 594 genera referred to 145 families, which include fungi as well as green plants; as there are perhaps 80,000 species in the Amazonian region, this is clearly only scratching the surface, even though there has been collecting of

material and uses since von Mattius in 1820, a later figure of great importance being Spruce, who is heavily quoted throughout the text. Although formerly much exploited, especially during the rubber boom, the Colombian Amazonians have retained their language, culture and knowledge of plant properties. Although the book concentrates on themedicinal and toxic, notably thecentral significance of hallucinogens of Malpighiaceae, Myristicaceae and Solanaceae for example, there is a welath of information on their food-plants, timbers and so on. The book ends with a general bibliography and indexes of 'Symptom, Disease, and Treatment' and genera.

Although copiously illustrated, often with the ratherbolddrawings fmmearlier Schultes publications and photographs clearly taken many years ago, not one ofthese has a scale. ~lthough there are rather few typographical errors, the authors admit that the momenclature is not always up to date, and that the chemical literature scanned was principally that published between 1967 and 1984, suggesting the book was some time in the press. But these are rather piffling points when the wealth of information presented here is considered. If you need to know how to use a 'bioactive' substance, either ingested as a tea, or gargled, or as a poultice or snuff (kss often smoked or as an enema) or need toknow theconstitutionofanarrow poison, whichmay contain materials from up to 15 different species, the instructions are here. There is an understandably long piece on cocaine, for example. If you do not know what the houseplants in the genus Fittonia are used for (toothache) or when 90cm long cigars are smoked, then this is the book for you.

The book also points up topics of general ecological interest, showing that it is impossible to think of the rain forest without human beings, who selectively fell, encourage and import different plants; some of these arecultigens now, e.g. Poraqueiba sericea (Icacinaceae), an important beverage plant and ingredient in contraceptive preparations; Cecropia sciadophylla is spared in felling forest for farming as its leaf-ash is preferred as the alkaline additive to coca, which, again, is unknown as a wild plant. As with other modem ecosystems, the forest cannot be maintained in what is often called a 'pristine' state (i.e. thestate that Europeansconsiderto be 'virgin', that is, as it was when they first saw it) without the maintenance of traditional management, which means maintaining the culture of the indigenous peoples. It is obviously of world benefit to do so, as these peoples are the repositories of hard-learned knowledge of forest plants and their uses, knowledge they have graciously shared with the authors. But we still do not understand anything of the chemistry of the Brazil nut beyond that of its oil, despite the fact that it is the most commonly sold tropical American product collected in the wild; not hardly anything of the chemistry of Potalia amara, one of the most valuable and versatile of all the local remedies in the region.

In view of the knowledge of the indigenous people, we stand humbled. We are also grateful for the life's work of the authors and to Timber Press for yet another excellent work, its value being shown by the fact that it hasbeenreprinted within less than two years from fxst publication.

D.J. MABBERLEY

N.S. SUBBA RAO and C. RODRJGUEZ-BARRUECO (Eds.) Symbioses in nitrogen-furing trees. Oxford & IBH Publishing Co., 66 Janpath, New Delhi 110001, India. 1993. xv + 271pp. US $74. ISBN 81 204 0757 1.

Plants that are able to fix atmospheric nitrogen and utilize it for their own nutrition have a special advantag~over other plants, particularly on nutrient-poor sites. This abiity is mthoweverin&hkto theplant, but is acquired only via symbiosis with certain m of microorganisms. Free-living nitrogen-flxiag m i m k s may aiso costrkbutie


to soil fertility. Inch?aarinply, the potential of nitrogen-fixing woody species (whichinclude plants from the generaAlnus,Acacia, Carwrina, and Leucueiw among others) for growth on difficult sites is being recognized. Biologically fixed nitrogen can also benefit other crops grown together with the nitrogen-fixers, and may go some way to replacing expensive fextilizers. Such nitrogen-fixing trees are of paiticular importance in many tropical areas, where they provide an attractive route towards inrreased productivity.

Thisbookcontainsaseriesofreviewsby amultinational assemblage of authors. A broad range of aspects of the symbioses in which nitrogen-f*ng haes participate akaddressed: na&lly the emphasis is on the associations with nitrogen-fixing micro-organisms but it was gratifying to see due consideration being given to the mycorrhizal associations of these trees also.

The ten well-referenced chapters that comprise this book are entitled: 'Some general considerations on symbioses of nitrogen- fixing trees' (A. Quispel, C. Rodriguez-Barmeco and N.S. Subba Rao); 'Nitrogen fixation by legume trees' (J.M. Sutherland and J.I. Sprent); 'Soil constraints to the growth of nitrogen-fixing trees in tropical environments' (P. Reddell); 'Comparisons of aerial nodulation of Casuarina sp. with legume stem nodulation' (E. Duhoux, Y. Prin and Y.R. Dommergues); 'The cycad-cyanobacterium symbiosis' (N. Grobbelaar); 'Myconhizal associations and theiifunctions innodulating nitrogen-fixing trees' (M.P. Herrera, P. Salamanca and J.M. Barea); 'Utilization of vesicular-arbuscular mycorrhizal fungi in the establishment of nitrogen-fixing trees' (T. Aziz and D.M. Sylvia); 'Host specificity in Frankia symbiosis' (C. Rodriguez-Barmeco, E. Cervantes and N.S. Subba Rao); 'Contribution of molecular biology techniques to Frankia strain characterization and molecular ecology' (S. Nazaret, P. Simonet andP. Normandl: and 'Interaction of nodulated tree species with othermicro-organisms andplants' (N.S. SubbaRao).

These titles give a reasonable indication of the chapter content. Happily there is commendably little duplication of material between chapters.

A good balance is achieved between Rhizobium and Frankia symbioses, and the book provides a good overview of the topic, primarily at an 'organism' level. For completeness a chapter on the biochemistry of the nitrogen-fixing process would have been useful, whilst two chapters (on aerial nodulation and nitrogen fixation in cycads) are perhaps of rather more limited interest.

The less satisfactory aspects of this book mostly concern its production. Forabookthat is definitely not cheapthe poorreproduction of photographic plates is disappointing, and did not create a good impression. Similarly the changes in type size that occurred in a few places (perhaps where changes had been made at the proof stage) detracted from the presentation. The number of misprints also approaches the unacceptable; their abundance variesbetween chapters and suggests that proof-reading has been less than meticulous. In places these caused some annoyance, for example where there were errors in reference citations.

Despite this, the book willbe auseful source of reference to anyone wishingto build on abasic knowledge of the biology of nitrogen-fixing trees and their svmbioses. It should Drove valuable to forest scientists contemplating exploring the potential (and complexities) of planting nitrogen-fixing trees on difficult sites. Although its content would make it very suitable as an adtranced student text, I am reluctant to recanmendit as suchbecause of the many (often trivial) misprints that could prove confusing to the inexperienced. Although the book provides a very helpful overview of nitrogen fixation in trees, it ultimately failed to excite or inspire me personally.

R.B. PW\RCE

NOTICES

J.D. BRAZIER A review of tropical hardwood consumption. Building Research Establishment Report CVSfB i3. BRE, Garston, Watford WD2 7JR. 1992. l lpp. £12. ISBN 0 85 125 536 1.

This report is a short, concise and well illustrated survey of the use of tropical hardwoods in Europe with particular emphasis on the United Kingdom. It puts the industrial use of tropical hardwoods in the UK and Europe in the context of their global use and argues that the UK consumes 0.1% of total and 0.7% of industrial consumption of this timber. The author points out that the import to Europe is dominated by a relatively small number of species and of predominantly red and brown colour for the UK market. He argues that the refusal of the UK to buy tropical hardwoods would have a negligible effect on their total use and little, if any, effect on destruction of tropical forest; any effect that was produced would be likely to accelerate rather than reduce destruction because it would lower the value of the forest in the eyes of producers and therefore increase the incentive to remove the forest in favour of some other use for the land.

He advocates continued use of tropical hardwoods accompanied by encouragement to manage forest sustainably and a willingness to pay more for traditional products and forfurthermanufacturedproducts from the producer countries.

The report is a well argued, concise statement of the current position regarding the use of these timbers and the erroneous idea that using less of them will contribute anything to reducing tropical forest destruction.

R.A. P L ~ E

CANADIAN FORESTRY ASSOCIATION Old growth forests. Special Issue No.5 of pores ty on the hill, 1993.90pp.

This special issue brings together 28 short papers covering aspects of the ecology, management and conservation of old growth forests. It is presented as 'a cross-section of perspectives for informed public participation and decision-making', with the result that there is a good deal of repetition. Contributors include academics, professional foresters and their associations, ministry officials and conservation bodies.

A lengthy definition of old growth forests is proposed - essential elements include the presence of many large, full mature trees of various species, with accumulations of large dead standing and fallen trees and visible signs of decadence. The structural diversity in these dynamic ecosystems is greater than in younger forests. Of Canada's 453 million ha of forest, 92 million are described as mature or over- mature.

A number of old growth forests across Canada are described and many issues are raised and discussed at a fairly superficial level: wildlife conservation; biodiversity and forest fragmentation; logging and the use of this timber for paper-making; the sustainability of harvesting old growth forests. Several contributors call for full forest inventories and for the designation of some forests for protection from harvesting.

J.M. PIPER


N.C. JAW (Ed.) Tree improvement and provenance research.. Proceedings of a workshop held at Grass Farm, Khatipura Road, Jaipur, 20-21 April, 1992. Dept. of Forest Government of Rajasthan, Jaipur, 1993. 288pp.

This workshop was held to bring together scientists and foresters involved in tree improvement work in the arid and semi-arid zones of India. The first three technical sessions were titled Status of tree improvement research; Multipurpose species: introduction and provenance research; and Modem methods of tree improvement and clonal propagation. About nine papers were presented in each of these sessions. The fourth technical session, Futurestrategy foracoordinated approach, consisted of a discussion leading to recommendations for the development of a plan of action for various aspects requiring coordination involking exchange of resources and sharing of responsibilities.

ALAN LONGMAN Rooting cuttings of tropical forest trees. London, CommonwealthScienceCouncil:February 1993,138pp.,ISBN 085092 394 8. Price21 2.50 (US$22.50 in UNESCO Coupons). Cheques to be made payable to Commonwealth Secretariat Publications, Marlborough House, Pall Mall, London SW l Y 5HX. UK. Tel: 07 1-839 341 1. Fax: 07 1-930 0827.

This is the first volume in anew seriescalled Trq~ic~ulh.ees:propagation at~dplantin~n~unrruis which aims to promote thegrowingandplanting of trees. The series will cover all stages from genetic selection and setting up a forest nursery to planting and establishment in the field. Four more volumes are planned to complete the series viz.. Volume 2 - Raising seedlings of rropicui trees; Volume 3 - Growing good tropical trees for planting; Volume 4 - Preparing to plant tropicul trees; Volume 5 - Plutitin,q and estahlishnzent of'rropicul trees. Each manual will give clear, illustrated step-by-step instructions. They are spiral bound so that pagescan easily be photocopied for use in the field.

Chapters in this first volume include: genetic selection, stockplant management; propagation; taking cuttings; care of cuttings; and making checklists and records. The style adopted is one of questions and answers. This is effective in breaking up the text and making it an easy reference manual. The simple line drawings used to illustrate selected points are bold and clear.

Although the manual is clearly intended for the field worker, it containsinformation that could be useful to anyone concerned withthe vegetative propagation of trees in the tropics.

To be operationally successful and sustainable, cuttings programmes withmosttreesrequire meticulousadherence toasequence of techniques. Any break in the routine can bring about failure and, veq often, it is very difficult to identify the actual point offailure. This manual will play a significant role not only inestablishing routines but in ensuring their continuity as staff, labour and circumstances change.

R.D. BARNES

SW I.I)ISH PULP AND PAPER ASSOC~ATION (Skogsindustriema)Annual publication 1993: The forest cycle - piece by piece. 1993. Stockholm. 48pp. plus endpapers.

This short publication trumpets the positive features of forests and foresindustry in Sweden withemphasison sustainability, preservation of biodivenity ,development of more 'environment-friendly' products in the paper sector and maintenance of a carbon balance. But the trumpeting is not excessive and the very headings, such as 'low-

resource products' indicate where sensitive issues lie. A final section summarizes the condition of the industry in 1992 and is suppMZed by a comprehensive set of statistics for Swedish forest and wood-using indust&. Overall, the booklet provides a clear picture of main aspects of the whole sector and servesas a model for othercountries toconsider as a descriptive and informative statement of their own industry.

A.J. GRAYSON

C. W A T ~ S (Ed.) Ecological e#ects of afforestation: studies in the history and ecology of afforestation in western Europe. C.A.B. International, Wallingfonl, Oxon OX10 8DE, U.K. 1993. xi + 224 pages. £32.50; US$61.75 (Americas only). ISBN 0-85 198-8 18-0.

This book is the outcome of a meeting held in Freiburg in 199 1. 11s 13 Chapters, contributed by 24 authors, provide a series of case studies showing how theecologicaleffectsof afforestation have beenexamined using a variety of approaches from a number of academic disciplines, and describing the effects of different forms of forest management. Case studies come from the U.K. (5 papers), Italy (4 papers). Spain ( I paper), Ireland (1 paper) and Belgium (2 papers). They cover topics ranging from plant and animal communities, microfauna, stream chemistry, the history of changes following the abandonment of agricultural land, and government policies.

Most future planting in much of Europe is likely to replace arable crops or ley grassland. Much of the book concentrates on ways in which woodland habitats that have been lost on these types of ground have, to an extent, developed again, based on research on recently afforested land. For ground vegetation, the importance of diversity of scale and arrangements of stands is emphasized, as is the maintenance of long-term refuges and the avoidance, where possible, of ecological isolation and excessively intensive management.

The book is really a collection of somewhat disparate papers such as one might find in a well refereed scientific journal. It is likely to be of interest and value to those concerned with the subject but it is a pity that noattempt was made todraw the lessonsfrom all thecontributions together into a final chapter of generalized principles.

PS. SAVII-I.

PUBLICATIONS RECEIVED

BAKKER, L.M. Colonization and land use in the humid tropics of Latin America. Werkdocument KC-NBLF No.34, Stichting voor Nederlandse Bosbouw Ontwikkelings Samenwerking (BOS), Postbus 30, Wageningen, Netherlands. 1993. Gratis.

This 39 page document is a literature review of the extent, nature, effects and causes of agricultural colonization of humid tropical rain forest in Latin America. Its brevity necessitates considerable generalization.

INSTITUTE m COMMERCIAL FORESTRY ~ E A R C H ~ ? R U Q / research report 1993. ICFR, PO Box 375, Pietermaritzburg 3200, South Africa. xxxiv + 197pp. ISBN 1 86840 034 4.

UNCED Earth summit '92: the United Nations Conference on Environment and Development, Rio de Janeiro 1992. (Ed. Joyce Quarrie) The Regency Press Corporation. UK~Ernpe distributor: BookSystemsPlus, 2B PrimHall Farm, Widdmgton, Saffron Walden, Essex CB l l 3SB. 250pp. f 9.95. ISBN 0952046903.


SUMMARIES

A. FRENCH

Terminalia amazonia (Gmel.) Exell: diffusion d'une essence indigene pour le reboisement e t l'agroforesterie D. NICHOLS

Le Terminalin amazonia (Gmel.) Exell, de la famille des Combretaceae, se rencontre dans les tropiques humides, en forsts naturelles, du Mexique au Brtsil. A 1'Bge adulte, I'arbre peut atteindre une hauteur sup6rieure B 50 metres et un diambtre supkrieur B deux mbtres. Son bois de couleur plut6t jaun2tre avec des stries rouges, convient pour de nombreuses applications dans les domaines de la construction, de lamenuiserieet de I't5Mnisterie. Les arbres issus des populations naturelles ont constitut une source de bois d'oeuvre et I'essence est maintenant diffuste lors d'efforts de reboisement et dans les systkmes agroforestiers g&ce au travail des fermiers et des vulgarisateurs. I1 a ttb propose de dkvelopper la sylviculture de cette espi?ce de manikre plus globale. I1 est n6cessaire de prtserver dbs maintenant les populations naturelles de T. amazonia, surtout en ce - - qui concerne les provenances B faible production semencikre mais B haut potentiel quant B I'utilisation dans des projets de reboisement.

Estimation volumdtrique d'arbres grlce h l'utilisation d'un nouveau coefficient de forme JOSB A. ALEIXO DA SILVA, BRUCE E. BORDERS, GRAHAM H. BRISTER

L'utilisation d'un coefficient de forme cylindrique pour I'estimation du volume est discutee en ~arallble avec I'utilisation du coefficient de forme de Girard et des coefficients de forme normaux et absolus. Un nouvel estimateur du coefficient de forme cylindrique est obtenu par deux mesures du dimetre du tronc, B 15.2 cm et B 1,52 cm au- dessus du niveau du sol. Ces diambtres sont aistment mesurts par une personne debout se tenant B cat6 d'un arbre. Le coefficient de forme estim6 est utilise pour calculer le volume total du tronc de 2509 arbres distributs sur sept importants sites deprise de donnkes independantes. Trois espi?ces ont 6t6 prises en compte: Pinus raedia, P. elliortii, P . caribaea. L'estimateur obtenu a donnk d'excellents rksultats dans tous les cas. Cet estimateur pourrait s'avtrer utile lors d'inventaires forestiers en prbsence d'espkces qui ne prtsentent pas une augmen- tation de diambtre significative B la souche et pourrait ainsi parer B la ntcessitb d'btablir des 6quations de volume spkifiques.

Equations de dbtermination de volume de fats adaptdes aux essences de boisd'oeuvre des fod t s denses humides du Ghana J.L.G. WONG, H.L. BLACKETT

Des volumes de flits ont btt mesuds pour 3488 arbres comprenant 61 esfices et issus de 7 rkserves forestibres repdsentant les diffkents typologies forestibres de la r6gion de foret dense humide au Ghana. Les mesures de diambtre ont dt6 r&lis&s en utilisant des rubans sur les arbres abattus et en utilisant un relascope de Bittern, modkle gnrnd angulaire, SW les arbres debout. Les dsultats olxerms par les deux m&hode5 ont mend une bonne drence. A parlkdecesdondes, deuxtypesd'6quations& volume r6gionalesont ttCproduitsen~tilisantw&leB~treuniqueWsurla

&~llJmktne.LemodUeaWcLjraSttenutili9(~tune Bansformetion logarithmique n m a l e .

~tablissements de banques de clones d'arbres plus ~Clectiomds parmi le rdseau international d'essais dePinus de 1'Institut de Foresterie d'Oxford A. J. POTTINGER

Dans le ddbut des annCes septante I'Institut de Foresterie du Com- monwealth (CH) a entrepris la distribution de semences de Pinus caribaea var. hondurensis Barr. & Golf., P. oocarpa Schiede et P. patula Schiede & Deppe ssp. tecunumanii (Equiluz & P e w ) Styles. Ceci devait devenir par la suite un &eau de plus de 600 essais de terrain repartis dans 65 pays B travers les tropiques. Le but en Ctait d'6valuer les diffkrences de provenance et d'estimer les parametres g6n6tiques. Une partie de ces essais a tt6 6valuCe dans la fin des anndes septante et le ddbut des ann6es quatre-vingt quant B la forme des arbres et B leur productivit6. Les rCsultats ont 6t6 distribuCs entre les collaborateurs. En 1986, I'institut nouvellement rebaptisC Institut de Foresterie d'Oxford (OFI) a, avec l'appui financier de la Shell Research Ltd. entame le projet de rassembler approximativement 150 arbres plus B partir de I'entikretb du rdseau d'essai de la banque de clones. L'objectif Btait de pourvoir les collaborateurs en matCriel gCnCtique clonal des arbres plus sClectionn6s afin qu'ils puissent les utiliser dans leur propres programmes de stlection. Des greffons issus de 73 arbres de Pinus caribaea var hondurensis, 43 arbres de P. oocarpa et 35 arbres de P. patula ssp. tecunumanii et sClectionn6s parmi les essais provenant de 13 pays, ont CtC replant& avec succks B Oxford dans les serres de 1'OFI. Faisant suite B la multiplication de nouveaw rejets par bouturage, des descendants obtenus par greffage ont CtC distribuks B partir d'Oxford vers sept pays tropicaux durant la @node 1988-1989. Des banques de clones ont &C subsCquemment Ctablies au Bdsil, au Zimbabwe et en Australie et des exp6ditions supplhentaires de plants ont tt6 effectukes de 1990 B 1993. Les facteurs qui influencent la dussite d'un greffage, leur multiplication et leur distribution sont discutks. Ap&s leur mise sur pied, ces banques de clones seront capables de fournir des plants pour l'approvisionnement des reproducteurs- multiplicateurs de plants B travers les tropiques.

tat sanitaire du mattriel gtnttique et classement des maladies avec un inttrst plus particulier pour les arbres de for& en pays tropicaux. M.H. IVORY. P.B. TOMPSETT

Dam un premier temps, cet article passe en revue la dissemination des organismes phytopathogknes B la fois sur, dans ou avec les differents types de materiel g6nCtique utilises dans le cas d'arbres forestiers. 11 conclut que relativement peu d'organismes pathogbes sont transmis par l'intermtdiaire de matdriel g6nCtique d'arbres sous la forme de semences propres, pollen ou cultures de tissus, mais que beaucoup sont disdminks par la voie vegetative, tels que par plants ou greffons. Dans un deuxikme temps, on passe en revue les procedes traditionnels et modernes de detection et de quantification des organismes phytopathoghes, proctdures qui constituent la base de I'indexation des maladies. Finalement est present6 un bref rksumt? des techniques d'Cvaluation sanitaire des semences et de mise en quarantaine des plantes en relation avec les techniques de transport de mattriel genttique. L'article conclut que des techniques de depistage et & diagnostic peu coliteuses sont ntcessaires pour assurer que la diffusion de matCriel gtnttique, activitt en nette croissance pour le moment, soit conduite en toute s&uritt.

70 Commonwealth Forestry Review Volume 73(l), 1994

Une m6thode d'tkhantillonage B quatre niveaux destintk B l'estirnation du cubage des peuplements B l'aide des donnks Landsat-TM, de photographies a6riennes et d'estirnations sur le terrain M. KOHL, S.P.S. KUSHWAHA

Ce document pdsente une mtthode d'estimation de cubage n?alisCe dans le cadre d'inventaires forestiers extensifs. Les informations obtenues par les donntks de Landsat-TM, les photographies a6riennes en noir et blanc, les clichts infrarouges et les estimations sur le tarrain ont BC combinks en utilisant un tchantillonage B quatre niveaux afin d'ttablii une approche de stratification. Ce travail met en lumikre les aspects statistiques et opirationnels d'un tel inventaire et il fournit des renseignements sp6cifiques sur la dttermination de la limite optimale & la strate. La mtthcde a et6 appliqute dans un inventaire pilote B Karnataka, en Inde.

Variabilitk des attaques de l'aphidien du cyprks Cinara cupressi (Buckton) sur la farnille des CupressacCes J.A.F. OBIRI

On aBtudiBles niveauxdecroissanceet les types de toltrance de trente- quatre esgces difftrentes de coniferes choisies parmi la famille des CupressacCes en prtsence d'aphidiens du cyprks Cinara cupressi (Buckton). L'Bude a t t t conduite B I'arboretum de Mugaga, dans le centre du Kenya, durant une +r ide de deux annCes. Elle indique qu'on rencontre une large gamme de niveaux de toltrance parmi les difftrentes esgces. Thuya spp. et Cupressocyparis leylandii (Jackson et Dallimore) se sont avBrtes &re les espkces le plus tolBrantes, tandis que Widdringtonia et Callitris se sont avtrBes Ctre les moins tolBrantes. Parmi les cyprks, Cupressus torulosa (Don), C.funebris (Endl.) et C. arizonica (Greene) ttaient rtsistants 1hoh C. henthamii (Endl.), C. lusitanica (Miller) et C. lindleyi (Klotsch) Btaient sensibles. En tenant compte de la vigueur de sa croissance et de sa rksistance vis-B-vis des aphidiens, Cupressocyparis leylandii semble etre la meilleure es@ce pour remplacer C. lusitanica qui est infestke et ?I ce jour, largement dpandue comme essence de plantation. Les rtsultats obtenus sur base des es@ces hy brides montrent tgalement que la rtsistance acquise par I'hybridation est en mesure d'offrir une solution viable ?I long terme a la menace des aphidiens.

Variabilitk des provenances du Pinus maximinoi: une esp&ce pleine de promesses dans le domaine du reboisement commercial au Zimbabwe B.I. NYOKA

Cet article prBsente les rtsultats de huit annks d'essais de provenances &Pinusmariminoi. Ces essais ont t t t conduits sur trois sites. Surbase des caractt5ristiques de diamlttre B 1,3 mttre, de hauteur totale et de volume, les deux provenances de San Juan et de Tatumbla se sont comportks de manitre satisfaisante et avec constance sur tous les sites. Bien que certaines provenances aient fait preuve d'un comportement inconstant, I'interaction site X provenance ne s'est pas avtrtk etre statistiquement significative. L'effet de site Btait prononce des provenances ont monbt! sur le site de Chisengu B une production en volume trois fois supkieure B celle relevCe sur le site de Nyangui. Ce dernier a, quant B h i , montn? un taux d'infestation en Pinew Clevt mais avec une fr6quence faible de production de canes. Sur le site de Chisengu A, la piupart des provenances ont montd, d'une manitre constante, une forte tendance B produire des fourches. La provenance & San Juan a fait preuve d'une W e n c e constamment Ciev6 de

production de canes, d'un faible niveau d'infestation par Pineus, d'une faible tendance B fourcher mais d'une mauvaise forme de hunc sur tous les sites. La croissance de P. maximinoi semble &re bonne B une altitude de 1350 metres, altitude B laquelle tout gel est exclu. Les meilleures provenances sont Tatumbla, San Juan, Duke Nombre et Cofradia.

B. SPANISH

Terminalia amazonia (Gmel.) Exell: El desarrollode unaespecie nativa para reforestaci6n y agrosilvicultura. D. MCHOLS

Terminalia amazonia (Gmel.) Exell, Combretaceae, se encuentra en 10s tr6picos humedos en 10s bosques naturales desde Mtxico hasta Brasil. En su madurez, es un b l de hasta 50 m en altura, con un dihmetro de hasta dos metros. Su madera amarillenta. generalmente rayada de rojo, es apropiada para muchos prop6sitos de construcci6n y para la fabricaci6n de muebles. Los &boles de poblaciones naturales han sido la fuente de madera, y laespecie estA ahora siendo incorporada en 10s esfuerzos de reforestaci6n y sistemas agroforestales, atravts del trabajo de granjeros y extensionistas. Se sugiere que la silvicultura de esta especie se desanolle m& completamente. Existe la necesidad urgente de preservar las poblaciones naturales de T. amazonia, especialmente de proveniencias con baja producciivn de semillas pero alto potencial de uso en proyectos de reforestaci6n.

La estimaci6n del volumen del hbol utilizando un nuevo factor de forma. JOSE A. ALEIXO DA SILVA, BRUCE E. BORDERS, GRAHAM H. BRISTER.

Se discute el uso del factor de forma de cilindro en la estimacidn del volumen del kbol junto con el uso de la clase de forma Girard y 10s coeficientes de forma normales y absolutos. Se deriva un nuevo estimador del factor de forma de cilindro el cual usa dos mediciones del diimetro del tallo a 0.5 pies (15.2 cm) y a 5.0 pies (1.52 m) sobre el nivel del suelo. Estos diirnetros son medidos convenientemente por una persona parada junto a1 b l . El factor de forma estimado se "s6 paracalcular el volumen total del tallode 2509 b l e s distribuidos en siete colecciones grandes de informaci6n independientes de tres especies: Loblolly, slash y Caribbean pine. El estimador sedesempeii6 muy bien en cadacaso. Este estimador puede ser &il en inventarios de bosque para especies que no expresan un punto de engrosamiento significativo, y puede evitar la necesidad de ecuaciones de volumen especificas.

Las ecuaciones de volumen de tronco para especies rnaderables del bosque alto en Ghana. J.L.G. WONG y H.L. BLACKEl'T

Se midieron 10s voIhenes de tronco de 3488 &ales de 61 especies de siete Reservas Forestales que son representativas de 10s t i p s de bosque encontrados en la regi6n de bosque alto en Ghana. Las mediciones se hicieron utilizandocintasdedi~troen drboles faiados y elrelascopio Spiegel de Bnguloanchoen Brbolesen pieconunabuena consistencia entre las dos medicioms. A partir de estos datos, se derivaron dos conjuntm deecuaciones devd&nregiona~es utilirando un mode10 de. poder de padmetro Me0 b a d en mediciones del diametro a ta altura &l pecho. El &1o se ajust6 atilizando una transformsci6n log-normal.

72 Commomvealth Forestry Review Volume 73(1), 1994

Advice to Contributors

GENERAL

Manuscripts shouldbe typewritten, double-spaced throughout, with at least a 3 cm margin on the left. Two copies must be submitted to the Editorial Office, Commonwealth Forestry Association, Oxford Forestry Institute, South Parks Road, Oxford OX1 3RB. Authors are advised to keep a copy of their manuscripts for checking against proofs. The style of setting out, subdivision of text, and layout of tables should conform to that of a current copy of the Review.

Contributions must be original and not have been submitted for publication elsewhere.

SUMMARY

Authors must supply a summary of not more than 150 words at the beginning of the paper, and should suggest no more than five keywords.

PREPARATION FOR PUBLICATION

Each submitted manuscript will be referred to a member of the editorial board and at least one expert referee. Authors will be consulted if the paper is considered suitable for publication but alterations are thought desirable. After these alterations have been included the manuscript will be considered final.

Page proofs will be sent to authors for checking before publication. These proofs are simply for the purpose of correcting printing errors; only in exceptional cases, and at the discretion of the Editorial Board, may authors be permitted to change their text at this stage.

SCIENTIFIC NAMES

The complete scientific name (genus, species and authority, and cultivar where appropriate) must be cited for every organism at the first time of mention. The generic name may be abbreviated to the initial thereafter except where intervening references to other genera with the same initial would cause confusion. If vernacular names are employed, they must be accompanied by the correct scientific name on first use.

TABLES

Tables should be numbered in arabic numerals and each submitted on a separate sheet with an appropriate legend at the head. The preferred position in the text should be indicated.

ILLUSTRATIONS

Text-figures. The publication of papers may be delayed, and the cost is increased, when figures have to be redrawn or relettered by the printers. This can be avoided if authors will comply exactly with the following instructions.

Drawing should be made in black ink on Bristol board, tracing film, or graph with faint ruling. Photocopies are not satisfactory for reproduction. A photocopy or other suitable duplicate should, however, accompany each figure. The maximum space available on a page is 228 X 179 mm (9 X 7 inches), and the width of a single column of text is 86 mm (3.4 inches), thus for a full page figure the original should not be larger than 456x 358 mm (18 X 14 inches) and for a singlecolumn width figure theoriginal should not be larger than 172 X 358 mm (6.8 X 14 inches). Authors should indicate on the back

of figures whether they are to be reproduced at single or double column width.

Legends to figures must not be incorporated in the figures, but should be typed in double spacing on a separate sheet. Each legend should contain sufficient explanation to be meaningful without cross-referencing. The individual figures must be numbered clearly in pencil and should bear the name of the author.Photographs for plates must be of a high quality, printed on glazed paper, and should be supplied at the same size as for final reproduction, bearing in mind the dimensions given above. A separate typewritten double-spaced EXPLANATION OF PLATES should be supplied, setting out the legends for the photographs, which will be referred to as Plate 1, Plate 2, Plate 2A etc. A scale should be included in the picture wherever possible, otherwise the scale of the original should be stated in the EXPLANATION OF PLATES so that the final scale can be calculated. Photographs should be black and white; colour reproduction will not be considered, and black and white reproduction of colour photographs is possible but will give an inferior result.

ORIGINALS

If authors wish to have original illustrations returned after publication they shouldmakeaspecific request when submitting their paper for consideration.

REFERENCES

A great deal of editorial time can be taken up by revising references to conform with the style of the Review. Authors are requested to observe the following guide closely.

Cltations in the text should take the form: Smith and Robinson (1960). If several papers by the same author in the same year are cited, they should be lettered in sequence (1960a), (1960b) etc. When papers are by more than two authors, all the names should be recorded on first mention, e.g. Smith, Jones, and Williams (1960). Subsequent reference should be to Smith et al. (1960).

At the end of the paper, the list of REFERENCES must be arranged in alphabetical order w~thout serial numbering. The following standard form of citation should be used: BUNTING, G.S., 1967. The conserved type of Calliandra. Taxon, 16,

469-472. ELIAS, T.S., 1978. Mimosoideae. In POLHILL, R.M. and RAVEN, P.H.

(Eds.) Advances in Legume Systematics Part 1. Royal Botanic Gardens, Kew, Richmond, Surrey. 425pp. All citations in the text should appear in the list of references and

vice versa. The list must be typed double-spaced throughout and checked thoroughly before submission.

COMPUTER DISKS Where authors use a word-processor, they should nevertheless submit their papers in printed form only in the first instance. When a paper is accepted for publication, it will be greatly appreciated if authors can provide a disk to accompany the final version of the manuscript. If possible, the paper should be written in Wordperfect 5.1, or alternatively sent as an ASCII file, but materials in other programmes can be translated hem if necessary. There is no need to send figures in graphics programmes, but the printed figure should be of a high quality suitable for reproduction.

Commonwealth Forestry Review Volume 73(1), March 1994

Australian National University

Department of Forestry School of Resource and Environmental Management

THE, DEPARTMENT The ANU Department of Forestry has become a foundation department within the new School of Resource and Environmental Management. The Department retains its traditional strengths in forest management (including fue management, wood science and industrial technology), eucalypt silviculture and ecology, inventory and modern usage of computers in forestry. In addition students now have access to a wide range of resource subjects, including resource economics, soil conservation, landcare and rehabilitation, agroforestry, agroecology, remote sensing and geographic information systems, human sciences, and others still in the process of development.

I The Department maintains strong linkages with forestry programs in many countries. It is particularly well placed to offer graduate programs to students from these countries and to students wishing to pursue careers 'in developing countries. There are currently 51 graduate students enrolled of whom 27 are from overseas.

Postgraduate Diploma (Grad Dip) Each Diploma program is individually designed to meet the student's needs. It may be aselfcontained course with a mix of coursework, directedreadingand research to enable candidates to specialise in an area of interest or it may be preliminary work for later masters or PhD study. It is especially valuable for overseas students as it can be designed to provide an introduction to Australian forests and environmental conditions. 10 months full-time - 18 months part time Entry Qualifications -Appropriate BA or BSc

Master of Forestry (MF) A program of coursework in forest and environmental management. Can be directed to particular topics or countries. 1 year and six weeks full-time Enuy Qualifications - Appropriate honours degree or postgraduate diploma

Master of Science (MSc) Doctor o f Philosophy (PhD) A program of intensive research in forestry or environmental science. Can be directed to articular topics or countries. M I C - 1-4 years full-time - 2-4 years part-time Enuy Qualii~cations - Appropriate honours degree or po-stgraduate diploma PhD - 3 years full-time Entry Qualifications -Appropriate honours degree, postgraduate diploma or MSc

INQUIRTES: Department of Foresuy International Fax Australian National University + 61 6 249 0746 GPO Box 4 Canberra ACT 2601 International Telephone AU- + 61 6 249 2579

Commonwealth Forestry Review Volume 73(1), 1994

FORESTRY EDUCATION AT ABERDEEN The University of Aberdeen, founded in 1495, has been awarding degrees in Forestry since 1914.

BSc (Forestry) Firmly founded on a scientific base, this degree emphasises resource and business management aspects relevant to both temperate and tropical forests. The honours degree normally requires four years of study but students withdiplomas or equivalent qualifications are usually admitted direct into the second year.

MSc/Diploma in a. Forest Management or b. Agroforestry

Designed primarily for upgrading practising foresters from the tropics and sub-tropics are Postgraduate Diploma (9 months) and MSc (1 2 months) programmes. Eight course modules are selected according to individual needs. Modules are also available from other University land-use departments and Agroforestry is taught in collaboration with the Macaulay Land Use Research Institute. The MSc programmes include a dissertation over 3 months.

Course modules specific to the Forest Management MScIDiploma include eg Forest policy; Forest business studies; Forest planning systems; Harvesting; Wood Science; Tropical forest ecology and management; Environmental remote sensing. Course modules specific to the Agroforestry MSc/ Diploma include eg Rural development; Agricultural project planning; Tropical crop systems; Tropical animal systems; Soil fertility and management; Soil erosion, conservation and sustainable land use. Common modules for both Forest Management and Agroforestry students include Forest mensuration and inventoiy; Temperate silviculture; Statistics and experimental design; Tropical plantation forestry; Tree improvement; Rural environmental economics; Rural development forestry and agroforestiy I (agrisilviculture); Agroforestry 11 (silvopastornlism).

The Department also offers MSc and Postgraduate Diploma programmes in Arboriculture; information will be supplied on request.

Research Degrees These are available at both Master's and Doctorate level: a wide ranging research programme is available encompassing both temperate and tropical forestry issues.

Aberdeen and the North-East of Scotland contain a concentration of research establishments in Land Use and Environmental matters in Europe: in addition to the departments within the University there is the Macaulay Land Use Research Institute, the Institute of Terrestrial Ecology, the Rowett Research Institute and the Marine Research Laboratory; combining the interests of all of these is the Aberdeen Centre for Land Use. The University is also the base for CEMP, the international Centre for Environmental Management and Planning.

Write for further details to Dr WH Parry Department of Forestry University of Aberdeen St Machar Drive Aberdeen AB9 2UD United Kingdom

UNIVERSITY OF

Tel+44 224 272667 Fax +44 224 272685 Telex 73458 UNIABN 9

ABERDEEN

Documents

THE - Bodleian · PDF file(Vice-Residents sit acr observers on the Governing Council ... NEWS OF MEMBERS ARTICLES Terminalia ... I visited the Forest Research Institute of Malaysia