3rdTaro Symposium 3rdTaro Symposium
Transcription
3rdTaro Symposium 3rdTaro Symposium
3rd Taro Symposium 21 - 23 May 2003 - Tanoa International Hotel, Nadi, Fiji Islands Edited by Luigi Guarino, Mary Taylor and Tom Osborn Proceedings of an International Scientific Meeting jointly organised by the Secretariat of the Pacific Community and the International Plant Genetic Resources Institute third taro symposium 3rd Taro Symposium 21–23 May 2003 Tanoa International Hotel, Nadi, Fiji Islands Edited by Luigi Guarino, Mary Taylor and Tom Osborn Proceedings of an International Scientific Meeting jointly organised by the Secretariat of the Pacific Community and the International Plant Genetic Resources Institute French translations are also provided of the welcome address, two keynote addresses, titles and abstracts of contributed papers, and conference recommendations Sont également traduits en français : l’allocution de bienvenue, deux discours introductifs, le titre et le résumé des documents d’information, ainsi que les recommandations Secretariat of the Pacific Community © Copyright Secretariat of the Pacific Community, 2004 All rights for commercial / for profit reproduction or translation, in any form, reserved. SPC authorises the partial reproduction or translation of this material for scientific, educational or research purposes, provided that SPC and the source document are properly acknowledged. Permission to reproduce the document and/or translate in whole, in any form, whether for commercial / for profit or non-profit purposes, must be requested in writing. Original SPC artwork may not be altered or separately published without permission. Original text: English Secretariat of the Pacific Community Suva Subregional Office Private Mail Bag Suva FIJI ISLANDS Tel: +679 337 0733 Fax: +679 337 0021 Email: [email protected] This publication is also available from SPC as a CD-ROM and at http://www.spc.int/rgc/ Secretariat of the Pacific Community Cataloguing-in-publication data Third taro symposium (3rd : 21-23 May 2003 : Nadi, Fiji Islands) Third taro symposium /edited by Luigi Guarino, Mary Taylor and Tom Osborn (Report of meeting (technical), Secretariat of the Pacific Community, ISSN 0377-452X) French translations are also provided of welcome address, keynote addresses, titles and abstracts of contributed papers, and conference recommendations. 1. Taro – Oceania – Congresses. 2. Taro – Genetica – Oceania – Congresses. 3. Plant conservation – Oceania – Congresses. I. Title. II. Secretariat of the Pacific Community. III. Series. 633.68 AACR2 Agdex Pacific Islands 171/41 ISBN 982-00-0027-0 This publication may be cited as: Secretariat of the Pacific Community. 2004. Third taro symposium: 21–23 May 2003: Nadi, Fiji Islands. Suva, Fiji Islands: Secretariat of the Pacific Community. Printed by Quality Print Limited, Suva, Fiji Islands CONTENTS Third Taro Symposium Introduction 9 Final recommendations / Recommendations arranged by type of activity Recommandations / Recommandations (par type d’action) 10 16 Acronyms Sigles et acronymes 23 24 Welcome Hon. Jonetani Galuinadi Allocution de bienvenue 25 26 KEYNOTE ADDRESSES Taro research and development - Progress since the last Taro Symposium and challenges for the future Recherche et développement du taro – Progrès accomplis depuis le dernier colloque sur le taro et défis à relever G.V.H. Jackson Taro genetic resources for now and tomorrow: a Pacific crop Ressources génétiques d’aujourd’hui et de demain: le taro, une plante océanienne Coosje Hoogendoorn, P.N. Mathur, Ramanatha Rao and Luigi Guarino THEME 1: TARO DIVERSITY, ETHNOBOTANY AND CONSERVATION 27 30 34 40 Abstracts Theme 1 Resumés Thème 1 Networking with taro: A review of TANSAO achievements V. Lebot, J. Quero Garcia and A. Ivancic Taro diversity in a village of Vanua Lava island (Vanuatu): Where, what, who, how and why? Sophie Caillon and Virginie Lanouguère-Bruneau Applications of DNA markers to management of taro (Colocasia esculenta (L.) Schott) genetic resources in the Pacific Island region I.D. Godwin, E.S. Mace, P.N. Mathur and L. Izquierdo Using in vitro techniques for the conservation and utilization of Colocasia esculenta var. esculenta (taro) in a regional genebank Mary Taylor, Valerie Tuia, Rajnesh Sant, Eliki Lesione, Raghani Prasad, Rohini Lata Prasad and Ana Vosaki Promoting on farm conservation through taro diversity fairs in the Solomon Islands Roselyn Kabu Maemouri and Tony Jansen Home gardens and their role in the conservation of taro diversity in Vietnam Nguyen Thi Ngoc Hue, Luu Ngoc Trinh and Nguyen Van Minh Diversity and genetic resources of taro in India S. Edison, M.T. Sreekumari, Santha V. Pillai and M.N. Sheela Analysis of genetic diversity in taro in China D. Shen, D.W. Zhu, X.X. Li and J.P. Song third taro symposium 47 47 52 58 64 69 74 80 85 89 THEME 2: PESTS AND DISEASES Abstracts Theme 2 95 Resumés Thème 2 95 Characterisation of taro viruses and the development of diagnostic tests 98 R.M. Harding, P.A. Revil, G.J. Hafner, I. Yang, M.K. Maino, L.C. Devitt, M.L. Dowling and J.L. Dale The potential of the fungus Metarhizium anisopliae as a biological control agent for taro beetles 102 R.T Masamdu and N.A. Simbiken The biology of Phytophthora colocasiae and implications for its management and control 107 R.A. Fullerton and J.L. Tyson Current status of research on Rhizoglyphus mites associated with taro 112 Zhi-Qiang Zhang, Qianghai Fan, N.A. Martin and Sada Nand Lal Developing interactive diagnostic support tools for tropical root crops 114 V.dR. Amante and G.A. Norton THEME 3: PRODUCTION AND PRODUCTION CONSTRAINTS Abstracts Theme 3 Resumés Thème 3 Taro as the foundation of Pacific food security Nancy J. Pollock Taro production in Fiji, constraints and future prospects Aliki Turagakula Taro cultivation in the Marshall Islands: problems, persistence and prospects Dilip Nandwani, M.C. Cheng, Jimmy Joseph, Jabukja Aikne, Arwan Soson and Gwo-jong Moh Recent research on taro production in New Zealand W.T. Bussell, J.J.C. Scheffer and J.A. Douglas Taro production in Australia Peter Salleras Comparison of taro production and constraints between West Africa and the Pacific Kwadwo Ofori Taro production, constraints and future research and development programme in Indonesia T.K. Prana, Made Sri Prana, and T. Kuswara Taro production, constrains and research in Cuba Arlene Rodríguez-Manzano, Adolfo A. Rodríguez-Nodals, Leonor Castiñeiras-Alfonso, Zoila Fundora-Mayor and Adolfo Rodríguez-Manzano Taro (Colocasia esculenta (L.) Schott var. esculenta): production, constraints and research in Dominica and other Caribbean countries Gregory C. Robin third taro symposium 119 119 125 128 132 139 144 146 152 155 163 THEME 4: BREEDING AND DISTRIBUTION OF IMPROVED MATERIALS Abstracts Theme 4 Resumés Thème 4 Genetic diversity of taro (Colocasia esculenta (L.) Schott) assessed by SSR markers J.L. Noyer, C. Billot, A. Weber, P. Brottier, J. Quero-Garcia and V. Lebot Taro breeding programme of Papua New Guinea-achievements, challenges and constraints D. Singh, T. Okpul and D. Hunter Introduced taro cultivars – on-farm evaluation in Samoa T. Iosefa, C.J. Delp, D. G. Hunter and P. Fonoti The use of direct stolon development for mass propagation in taro (Colocasia esculenta (L.) Schott) Riki Faatonu, Philip Tuivavalagi, Winston Charles and Albert Peters Breeding Hawaiian taros for the future John J. Cho The establishment of a commercial tissue culture laboratory in the kingdom of Tonga Paul Karalus Taro breeding in India M.T. Sreekumari, K. Abraham, S. Edison and M. Unnikrishnan THEME 5: PRODUCT DEVELOPMENT AND MARKETING Abstracts Theme 5 Resumés Thème 5 Pacific taro markets: issues and challenges Grant Vinning and Joann Young Value added products from taro Richard Beyer Taro production and value adding in Palau Robert Bishop Recent developments in taro-based food products in Hawaii Alvin S. Huang, Karthik Komarasamy and Lijun He Chemical composition and effect of processing on oxalate content of taro corms E.O. Afoakwa, S. Sefa-Dedeh and E.K. Agyir-Sackey List of participants 170 170 174 181 185 189 192 197 202 207 207 210 215 222 226 230 237 third taro symposium I NTRODUCTION A s stated in the conference opening by the Fiji Minister for Agriculture, Sugar and Land Resettlement, the Hon. Jonetani Galuinadi, taro is an ancient and culturally central crop, extremely important for food security and sustainable livelihoods in the humid tropics of the Pacific and South East Asia, as well as in West Africa and the Caribbean. However, although it ranks 14th worldwide among staple crops in production, it is generally considered to have been somewhat neglected by the research community. In particular, breeding programmes are still few and no international agricultural research centre includes taro in its mandate. Having said that, there have been a number of attempts to bring specialists together and stimulate research, including two SPC seminars and two international taro symposiums. But these were in the mid 1990s, and in the wake of the recent advances in taro breeding by the TaroGen and TANSAO projects, it was thought opportune to organize this Third Taro Symposium in 2003. Over sixty taro research and development specialists from 22 countries around the world thus convened at the Tanoa International Hotel, Nadi, Fiji, from 21 to 23 May 2003. Their objectives were to review progress in taro research, analyze needs and priorities, develop a strategy for future work in taro research and development, explore new ways to use genetic diversity and improve taro quality and production, and stimulate international collaboration, information exchange and networking. The Third Taro Symposium was organized by the Secretariat of the Pacific Community (SPC) and the International Plant Genetic Resources Institute (IPGRI). The following also provided funding: the Technical Centre for Agricultural and Rural Cooperation (CTA), France’s Centre for International Research and Development (CIRAD), the Food and Agriculture Organization of the United Nations (FAO), the Australian Centre for International Agricultural Research(ACIAR) and New Zealand’s International Aid & Development Agency (NZAID) through their support for the Pacific Agricultural Plant Genetic Resources Network (PAPGREN), the Australian Agency for International Development (AusAID) through the TaroGen project, and Japan (via the Plant Genetic Diversity Group at the Gene Research Center, University of Tsukuba). The French Regional Fund provided support for translation services. After the opening by the Hon. Jonetani Galuinadi, SPC Agriculture Adviser Tom Osborn made a welcome speech on behalf of Dr Jimmie Rodgers, SPC Senior Deputy Director-General. There were then two keynote presentations, followed by some 34 presentations and a poster session arranged around the following themes: 1. Taro Diversity, Ethnobotany and Conservation 2. Pests and Diseases 3. Production and Production Constraints 4. Breeding and Distribution of Improved Materials 5. Product Development and Marketing After the presentations, there was a full day of working group discussions, which resulted in sets of recommendations for each theme that were then presented and endorsed in plenary at the end of the symposium. The papers and recommendations are presented in this volume, in the hope that they will further stimulate taro research. Taro specialist meetings in previous years Taro Leaf Blight Seminar. Alafua, Western Samoa, 1993 n Proceedings. Noumea, New Caledonia: South Pacific Commission; 1994. (Report of meeting (SPC). ISBN 982-203-502-0. Taro Seminar II. Lae, Papua New Guinea, 1995 n Proceedings. Noumea, New Caledonia: South Pacific Commission; 1996. (Report of Meeting (SPC). ISBN 982-203-519-5. First Taro Symposium. PNG University of Technology, Lae, 1993 n Book of abstract. First Papua New Guinea Symposium on Taro. Lae, Papua New Guinea: University of Technology; 1993. Second Taro Symposium. University Cenderawasih, Indonesia, 1994 n Papers and abstracts. Jackson, G. V. H. and Wagih, M. E. (eds). The Second Taro Symposium: Proceedings of an International Meeting held at the Faculty of Agriculture, Cenderawasih University, Manokwari, Indonesia, 23-24 November 1994. Lae, Papua New Guinea: University of Technology; Cenderwasih University; 1996. third taro symposium FINAL RECOMMENDATIONS Major points Gaps, opportunities Theme 1: Taro diversity, ethnobotany and conservation 1. TaroGen core has been DNA fingerprinted and is being virus indexed and will be available for distribution soon 2. Fingerprinting and indexing is also ongoing for TANSAO core • Development of capacity within the region for molecular research (USP and UNITECH) is necessary • Strengthened capacity for virus indexing needed at NARI Strategies • Integration of core collections into overall conservation and use strategy for taro using complementary approaches (national programmes, QUT, UQ, RGC, CIRAD, IPGRI) • Fingerprinting and virus testing/therapy should be closely coordinated with RGC, for which UQ and QUT can provide technical support as required Priority actions 1. Validate TaroGen and TANSAO cores and compare with other genepools using standardized molecular markers 2. Develop virus indexing and elimination capacity within region (check if CIP can assist) 3. Seek long-term funding for RGC, e.g. through Global Crop Diversity Trust • Distribution of pathogentested materials by RGC needed as soon as possible 4. Prepare catalogue of main taro diversity in region 5. Exchange clean material within region and outside (taking due account of policy implications) • RGC and USP should jointly assess the integrity of germplasm being distributed through tissue culture 6. Ensure safety duplication of collections • Validation of cores necessary 7. Molecular characterization of germplasm from India • Comparison of TANSAO and TaroGen cores would be helpful (also compare with diversity from countries not yet represented) • CIRAD and QUT should use same markers • Is there interest in developing DNA marker map(s) for Pacific taro? • Study of geographic patterns in genetic diversity would be useful 3. We know the relationship between chemotypes and taste, but we do not know the relationship between genotypes and organoleptic properties • Genetic work on understanding chemotypes (acridity, flavour, aroma, texture) is required • Molecular tools could be used: marker-assisted selection and selection of parents for breeding • Opportunity to collaborate with USP and NARI analytical labs • Need to work with farmers on taste issues • Investigation of agronomic and environmental effects on organoleptic properties necessary 10 third taro symposium • Further evaluation of core and other materials 8. Carry out participatory evaluation trials to investigate relationships among chemotypes, genotypes and organoleptic properties 9. Develop low-cost methods for evaluation of chemotypes 10. Compile information on existing traditional management practices for taro Major points 4. It is still difficult for national programmes to maintain base collections in the field or in vitro Gaps, opportunities • RGC efforts to develop reliable cryopreservation protocol for long-term conservation must be strengthened (IPGRI technical backstopping) • Until cryopreservation is developed, relative costs of conserving entire national collections in field and in vitro must be assessed, and then decisions taken as to which to pursue Strategies • Implementation of complementary approaches to conservation, including in situ as well as cryopreservation and seed conservation as main longterm ex situ methods for national base collections (national programmes RGC, CIRAD, IPGRI) Priority actions 11.Provide short-term support to national programmes for maintenance of field genebanks or in vitro conservation of base collections 12.Carry out research to develop a reliable cryopreservation protocol for taro • Rationalize national base collections according to use value (IPGRI technical backstopping) • Develop and promote regional in situ strategies 5. Seed storage presents an important opportunity for conservation and exchange of taro collections • The following questions need to be considered with regard to seed storage: 13.Carry out seed conservation research (RGC, UNITECH/ NARI/Vudal, IPGRI) - Can seed be stored longterm? 14.Study taro flowering, including induction - Is it virus free? - Is this important that we will be conserving genes and not genotypes? 6. Because of difficulties with • Documentation of IK will be ex situ conservation and important plasticity of genotypes, there is a need for further work on • Better perspective of genetic erosion is needed in situ approaches • Practical interventions must be designed, based on best practices developed in other regions as appropriate 15.Develop pilot (shop window) project for in situ management of taro genetic resources, including - PPB - Promotion (link between farmers and consumers) - Export - Terroir concept Theme 2: Pests and diseases 1. Indexing protocols are more or less complete • We must now get the germplasm to farmers • Country consultations to review survey data, and update germplasm transfer guidelines • Transfer of virus indexing technology to the region via labs at USP and UNITECH 1. SPC convenes meeting to update safe transfer guidelines 2. Obtain FQID approval to introduce plants for indexing 3. Monitor PT clones after field release 2. Need new Papuana beetle management strategies • IPM strategy using pheromones can be used • Research 4. SPC TBM to arrange field trials and chemical analyses for taro beetle control 3. Diagnostic interactive tools available for sweet potato • Could apply the model developed for sweet potato to taro • Seek donor funding for interactive diagnostic CDROM for taro 5. UQ and SPC develop project document on interactive diagnostic tool 4. Novel pest control methods becoming available • Apply novel methods to intractable pest problems e.g. alomae/bobone, Papuana beetle • Keep up-to-date with modern biotechnologies e.g. develop a taro transformation system 6. QUT/SPC/PICTs consultations on novel pest control approaches 5. More research is needed • Interaction of pests/diseases • Horizontal resistance with nutrient status unclear breeding should continue to be the preferred strategy • Post-harvest handling poor against taro leaf blight and in the Pacific other diseases in PICTs • Level of economic losses due to viruses unknown 7. CBDV epidemiological studies in PNG 8. Study interaction of pest/ diseases with nutrient status 9. Improve post-harvest handling in the Pacific (see experience from other regions, e.g. Caribbean) 10.Quantify losses due to virus diseases 11.Breed for Pythium resistance third taro symposium 11 Major points Gaps, opportunities Theme 3: Production and production constraints 1. Taro produced basically for two markets: food security and commercial market (export) 2. Production constraints often common to both systems • Many approaches will be applicable to both systems • Improvements in both systems could be obtained through greater sharing of information Strategies Priority actions • Greater exchange of information at all stages of production 1. Establish an international network to facilitate exchange of information • Greater exchange of information at all stages of production 2. Compile information on existing traditional management practices for taro 3. Information is lacking enabling producers to know what varieties to grow for what product, e.g. chip production 4. Management practices, production and consumption constraints differ around the world, but there seems to be little sharing of this information 5. Continuous availability of planting material is a problem, especially if “new” cultivars are introduced • Improvement in production possible with increased sharing of knowledge on management practices • Standardisation of management practices • Improved methods for producing and distributing planting material needed 3. Carry out national production-to-consumption constraint analysis • Further research on multiplication and dissemination techniques • Different multiplication systems should be available for different levels of production 4. Collate and review information on multiplication techniques 5. Carry out research on macro-propagation and micro-propagation techniques with focus on multiplication rates achieved, costs and applicability (focus on tuberlet formation) 6. Explore innovative avenues for distributing planting material from NARS to NGOs 6. There is a need to improve the product so that it becomes more desirable and the customer comes back for more (acridity is often a problem in product development) • Continued improvement of the crops is needed 7. Taro is underutilized for commercial production • Better linkages between producers and marketing people needed • Innovative breeding approaches 7. Participatory evaluation and selection of germplasm currently available (TaroGen and TANSAO cores, breeding lines) 8. Marker-assisted selection, in particular for edibility and taste • Appropriate technology equipment needed to assist production • Raise the “image” and awareness of taro as a food (stressing e.g. low glycemic index) • Improve agronomic and production practices 9. Marketing surveys in PICTs and Pacific Rim countries 10.Identify local criteria for product development and marketing 11.Carry out promotional campaigns 12.Explore appropriate technologies for mechanization 8. Declining soil fertility is a significant production constraint, but information is insufficient • Need information on taro production and practices as they relate to soil fertility • Make the decline in soil fertility a high priority in farming systems based projects 13.Compile information on the use of various farming practices to prevent decline in soil fertility 9. Atolls have specific requirements (nutrient deficiencies, salinity) • Must be increasingly aware of and respond to the special needs of atoll countries • Increase engagement with atoll countries on PGR issues generally 14.Investigate the possibility of screening for salt tolerance in vitro 10.Other aroids, which are important in the atolls, are very much neglected crops 15.RGC to evaluate taro varieties for atoll requirements through distribution to these countries 16.PAPGREN to address the issue of the other edible aroids 11.Taro has a low profile within the global research community 12 third taro symposium • Establish taro as an important crop within the global research community 17.Explore the possibility that the image of taro within the global research community could be raised through participation in the CGIAR system Major points Gaps, opportunities Theme 4: Breeding and distribution of material 1. Breeding programmes in Pacific, India, Hawaii progressing for specific traits using conventional, participatory and molecular approaches • Ornamental breeding • Pollen conservation for breeding programmes • Inter-generic crosses • Evaluation of core collection for target traits and use as parents Strategies • Continue conventional breeding through on-going programmes for tolerance to viruses and other important pests/diseases, adaptation, yield and quality traits Priority actions 1. Ensure sustainability of current breeding programmes in the Pacific 2. Strengthen coordination among programmes in the Pacific through networking 3. Broaden international collaboration worldwide 4. Carry out evaluation of existing collections 2. Multiplication of material is a critical issue for use of germplasm • Potential for micro• Continue research on propagation and macromacro-propagation propagation (particularly use and micro-propagation of GA-induced technique) techniques for efficient multiplication • Tonga tissue culture laboratory in operation • Linking conventional 3. Integration of molecular markers and breeding with molecular biotechnological approaches markers (molecular markerinto conventional breeding assisted selection) holds much promise Note: The following clarification was subsequently provided by Herman Francisco: “Palau questions (1) the ownership of accessions and accession derived materials; (2) the legitimacy of using farmer donated materials for commercial breeding programs; and (3) the legitimacy of using farmer donated materials for purposes other than it was donated for (i.e. TLB resistance). Palau expresses the need for ‘shared benefit agreements’ with the farmers/growers who have shared their planting materials and experience and indigenous/ privileged knowledge with researchers/breeders.” • RGC developing MTAs 6. Carry out further research on GA- induced multiplication technique 7. Explore the use of true seed • Use of molecular markers by collaborating with UQ, TANSAO/CIRAD, and University of Hawaii for marker-assisted selection. Student projects an option • Capacity building of national and regional programmes in molecular markers (for the Pacific, UQ and USP an option) 4. Issue of germplasm ownership raised by Palau 5. Review multiplication techniques • Regional agreement on access and benefit-sharing policy guided by CBD and other relevant regional/ international instruments, in particular the International Treaty on PGRFA 8. National capacity building and training in molecular techniques and implementing molecular technologies in collaboration with advanced research institutes 9. Identification of molecular markers to assist breeding 10.Molecular characterization of germplasm from India 11.Development of regional MTAs and other ABS policy instruments third taro symposium 13 Major points Gaps, opportunities Theme 5: New products and development 1. Need good quality taros at the farm gate • Can we identify and breed suitable chemotypes for present and future markets? Strategies • International network 1. Analyse the chemotypes of needed including the commercial varieties so that producing and consuming the private sector can use countries in the rest of the this information to promote world (Africa, Caribbean, products South and Central Americas, 2. Focus on quality and taste; India, China) so that use geographical indicators germplasm, information and to promote the product knowledge can be shared. combining the name of the variety, the geographical area of production, the cultivation techniques etc. for niche and high-value new markets 2. Very poor data available on production costs 3. New markets are necessary: fresh local market is fairly stable, but the fresh export market is very fragile Priority actions 3. Collect more data on costs of production • Use of all the parts of the plant (e.g. petiole and leaves) rather than focusing only the corm • Develop organic taro for niche markets for the US, Japan and Europe 4. Recognise that the markets for fresh and processed products need different plant materials and therefore distinct breeding strategies 5. Develop and implement a marketing, educational and promotional campaign to differentiate taro from other staples and position taro with healthy and culturally important foods 6. Evaluate varieties for potential in organic production 14 third taro symposium Recommendations arranged by type of activity Support conservation 1. Validate TaroGen and TANSAO cores and compare with other genepools (e.g. India) using standardized molecular markers 2. Carry out molecular characterization of Indian germplasm 3. Seek long-term funding for RGC, e.g. through Global Conservation Trust 4. Ensure safety duplication of taro core collections 5. Provide short-term support to national programmes for maintenance of field genebanks or in vitro conservation of base collections 6. Carry out research to develop reliable cryopreservation protocol for taro 7. Carry out research on flowering (e.g. induction) and seed conservation 8. Develop pilot (shop window) project for in situ management of taro genetic resources Exchange genetic material 9. Develop regional MTAs and other ABS policy instruments 10. Explore the use of true seed for exchange of genes (rather than genotypes) 11. Convene meeting to update safe transfer guidelines 12. Obtain FQID approval to introduce plants for indexing 13. Develop virus indexing and cleaning capacity within region 14. Support exchange of material within region and outside 15. Explore innovative avenues for distributing planting material from NARS to NGOs 16. Monitor PT clones after field release Evaluate and improve germplasm and products 17. Develop low-cost chemotype evaluation methods 18. Carry out participatory evaluation and selection of currently available germplasm (TaroGen and TANSAO cores, breeding lines) to investigate relationships among chemotypes, genotypes and organoleptic properties 19. Analyse the chemotypes of commercial varieties so that the private sector can use this information to promote products 20. Evaluate varieties for potential in organic production 21. Investigate screening for salt tolerance in vitro 22. Distribute taro varieties to atolls and evaluate for their specific requirements 23. Ensure continuation and sustainability of breeding programmes, including using marker-assisted selection focusing on quality and taste, recognizing distinct needs of markets for fresh and processed products Improve production 24. Collate and review information on multiplication techniques, and carry out research as necessary (e.g. GA-induced multiplication technique) 25. Carry out CBDV epidemiological studies in PNG 26. Conduct field trials and chemical analyses for taro beetle control 27. Quantify losses due to virus diseases 28. Explore appropriate mechanization technologies 29. Study interaction between pests/diseases and nutrient status 30. Improve post-harvest handling 31. Develop project document on interactive diagnostic tool for taro 32. Carry out consultations on novel pest control methods Document and exchange information 33. Compile information on the use of various farming practices to prevent decline in soil fertility 34. Compile information on traditional taro management practices 35. Prepare catalogue of taro diversity in the Pacific 36. Collect data on costs of and constraints to production in PICTs Increase awareness 37. Study the other edible aroids (e.g. in PAPGREN) 38. Carry out marketing surveys in PICTs and Pacific Rim countries 39. Carry out promotional and educational campaigns in PICTs Build capacity 40. Support national capacity building and training in molecular techniques and implementing molecular technologies Increase international collaboration 41. Establish an international network on taro to facilitate exchange of germplasm, expertise, information 42. Explore the possibility that the image of taro within the global research community could be raised through participation in the CGIAR system 43. Broaden international collaboration in taro research by linking Pacific to other taro-growing regions third taro symposium 15 RECOMMANDATIONS FINALE Principaux points Lacunes et débouchés Stratégies Actions prioritaires Thème 1 : Diversité génétique, ethnobotanique et conservation du taro 1. L’empreinte génétique de la collection noyau de TaroGen a été relevée ; l’indexation des virus est en cours. La collection pourra être prochainement distribuée. 2. La détermination de l’empreinte génétique et l’indexation des virus de la collection noyau du réseau TANSAO sont également en cours. • Il faut renforcer les capacités de la région en matière de recherche moléculaire (USP et UNITECH). • Il faut renforcer les capacités de l’Institut national de recherche agricole (NARI) en matière d’indexation des virus. • La détermination de l’empreinte génétique et les essais virologiques/la thérapie doivent être coordonnés avec le CRMG ; UQ et QUT peuvent prêter leur concours technique, au besoin. 1. Valider les collections noyaux de TaroGen et TANSAO, les comparer à d’autres banques de gènes à l’aide de marqueurs moléculaires normalisés. 2. Renforcer la capacité d’indexation des virus dans la région. 3. Mobiliser des financements à long terme pour le CRMG, par l’intermédiaire du Global Conservation Trust, par exemple. 4. Dresser le catalogue de la diversité du taro dans la région. • Distribution de matériels exempts d’agents pathogènes par le CRMG le plus tôt possible. 5. Échanger des végétaux sains dans la région et à l’extérieur (incidences politiques). • CRMG et USP devraient évaluer conjointement l’intégrité du matériel génétique distribué ayant fait l’objet d’une culture tissulaire. 6. Reproduire, pour des raisons de sécurité, le matériel génétique dans d’autres pays. • Validation des collections noyaux nécessaire. • Il serait utile de comparer les collections noyaux du réseau TANSAO et de TaroGen (et de les comparer à la diversité existant dans des pays qui ne sont pas encore représentés dans ces réseaux). • CIRAD et QUT devraient utiliser des marqueurs identiques. • Est-il intéressant de mettre au point des cartes de marqueurs ADN pour le taro océanien ? • Il serait utile d’étudier la diversité génétique en fonction de la distribution géographique. 16 • Intégration des collections noyaux dans la stratégie globale de conservation et d’utilisation du taro en adoptant des approches complémentaires (programmes nationaux, QUT, UQ, CRMG, CIRAD, IPGRI). third taro symposium 7. Réaliser des essais d’évaluation pour étudier les relations entre chimiotypes, génotypes et propriétés organoleptiques. Principaux points 3. Nous connaissons la relation existant entre chimiotypes et goût, mais pas le rapport entre génotypes et propriétés organoleptiques. Lacunes et débouchés • Travaux génétiques à mener pour une meilleure connaissance des chimiotypes (âcreté, goût, arôme, texture). Stratégies • Poursuivre l’évaluation des collections noyaux et d’autres matériels. Actions prioritaires 8. Réaliser des essais d’évaluation pour étudier les relations entre chimiotypes, génotypes et propriétés organoleptiques. 9. Développement de méthodes peu onéreuses pour l’évaluation des chimiotypes • Des outils moléculaires peuvent être utilisés : sélection à l’aide de marqueurs et sélection de parents pour la reproduction. 10. Recenser les informations sur les techniques culturales traditionnelles du taro • Possibilité de collaborer avec l’USP et les laboratoires d’analyse du NARI. • Nécessité de collaborer avec les agriculteurs pour les questions de goût. • Étude à mener sur les effets de l’agronomie et de l’écologie sur les propriétés organoleptiques. 4. Il reste difficile, pour les services nationaux, d’entretenir des collections de base en champ ou in vitro. • Renforcer les efforts que fait le CRMG pour élaborer un protocole fiable de cryoconservation pour la conservation à long terme (appui technique de l’IPGRI). • En attendant le développement de la cryoconservation, il faut évaluer les coûts respectifs de la conservation de collections nationales complètes en champ et in vitro, puis décider de la méthode à adopter. • Adopter des approches complémentaires de la conservation ; à la fois in situ, par cryogénisation et conservation des semences, comme méthodes principales de conservation ex situ à long terme des collections de base nationales (services nationaux, CRMG, CIRAD, IPGRI). 11. Assurer un soutien à court terme des services nationaux pour l’entretien des banques de gènes en champ ou la conservation in vitro des collections de base. 12. Mener des recherches afin d’élaborer un protocole fiable de cryoconservation du taro. • Rationaliser les collections nationales de base en fonction de la valeur d’exploitation (appui technique de l’IPGRI). • Élaborer et promouvoir des stratégies régionales de conservation in situ. 5. Le stockage de semences est un important moyen de conservation et d’échange de collections de taro. • Se poser les questions suivantes concernant le stockage des semences : - des semences peuvent-elles se conserver à long terme ? - Sont-elles exemptes de virus ? 13. Conduire une recherche sur la conservation des semences (CRMG, Unitech/ NARI/Univer-sité de Vudal, IPGRI). 14. Étudier l’induction de la floraison. - Importe-t-il de conserver des gènes plutôt que des génotypes ? 6. En raison des difficultés de conservation ex situ et de la plasticité des génotypes, il faut continuer à examiner des approches in situ. • Il importe de consigner les savoirs locaux. • Il faut acquérir une meilleure connaissance de l’érosion génétique. • Concevoir des interventions pratiques, fondées sur les meilleures pratiques élaborées dans d’autres régions, le cas échéant. 15. Mettre au point un projet pilote (vitrine) pour la gestion in situ des ressources génétiques du taro, notamment : • amélioration des végétaux par des méthodes participatives, • promotion (lien entre agriculteurs et consommateurs), • exportation, • notion de terroir. third taro symposium 17 Principaux points Lacunes et débouchés Thème 2 : Organismes nuisibles et maladies 1. Les protocoles d’indexation sont plus ou moins complets. • Nous devons maintenant fournir le matériel génétique aux agriculteurs. Stratégies • Consulter les pays pour examiner les résultats d’enquêtes et actualiser les directives relatives au transfert de matériel génétique. • Enseigner les techniques d’indexation des virus à la région, par l’intermédiaire des laboratoires de l’USP et d’Unitech. Actions prioritaires 1. La CPS organise une réunion pour réactualiser les directives de sécurité du transfert. 2. Obtenir l’accord de FQID pour introduire des végétaux à des fins d’indexation. 3. Surveiller les clones exempts de pathogènes après le lâcher dans les champs. 2. Nécessité d’appliquer de nouvelles stratégies de lutte contre les coléoptères du taro Papuana spp. • La stratégie de lutte intégrée faisant appel aux phéromones peut être appliquée. • Recherche. 4. TBM (CPS) organise des essais en champ et des analyses chimiques en vue de la lutte contre les coléoptères du taro. 3. Évaluer les outils interactifs existant pour la patate douce. • Possibilité d’appliquer au taro le modèle mis au point pour la patate douce. • Mobiliser des fonds auprès des bailleurs pour financer un cédérom interactif de diagnostic des maladies du taro. 5. UQ et la CPS élaborent un projet de document sur un outil interactif de diagnostic. 4. Apparition de méthodes innovantes de lutte contre les organismes nuisibles. • Appliquer des méthodes innovantes pour résoudre des problèmes difficiles de lutte contre les organismes nuisibles (viroses alomae/ bobone, coléoptères Papuana). • Se tenir au courant des biotechnologies modernes (mettre par exemple au point un système de transformation du taro). 6. QUT/CPS/pays océaniens se concertent afin d’adopter des approches innovantes de lutte contre les organismes nuisibles. • Interaction des organismes nuisibles/maladies avec l’équilibre nutritif mal connue. • Il faut continuer la sélection en vue de l’obtention d’une résistance horizontale et en faire la stratégie primordiale de lutte contre la flétrissure des feuilles de taro et d’autres maladies dans les États et Territoires insulaires océaniens. 5. Il faut approfondir la recherche. • La valorisation après récolte n’est pas suffisante dans la région du Pacifique. • Niveau des pertes économiques dues aux virus inconnu. 7. En PNG, conduite d’études épidémiologiques sur CBDV. 8. Étudier l’interaction des organismes nuisibles/ maladies et l’équilibre nutritif. 9. Améliorer la valorisation après récolte dans la région du Pacifique (comparer à l’expérience d’autres régions, par exemple aux Antilles). 10. Chiffrer les pertes dues à des viroses. 11. Sélection en vue de l’obtention de la résistance à Pythium. Thème 3 : Production et obstacles à la production 1. La production de taro est essentiellement destinée à deux marchés : sécurité alimentaire et marché commercial (exportation). 2. Les obstacles à la production concernent souvent les deux marchés. • De nombreuses approches seront applicables aux deux systèmes. • Meilleur échange d’informations, à tous les stades de la production. 1. Créer un réseau international afin de faciliter l’échange d’informations. • Meilleur échange d’informations, à tous les stades de la production. 2. Recueillir des informations sur les pratiques traditionnelles existantes de gestion concernant le taro. • On pourrait améliorer les deux systèmes grâce à un meilleur échange d’informations. 3. Les producteurs manquent d’informations sur les variétés à cultiver et les produits visés (production de chips, par exemple). 4. Les pratiques de gestion, les problèmes de production et de consommation diffèrent dans le monde, mais l’information à leur sujet circule apparemment mal. 18 • Un meilleur échange de connaissances concernant les méthodes de gestion permettrait d’améliorer la production. • Normalisation des pratiques de gestion. third taro symposium 3. Effectuer une analyse comparative des problèmes de production et de consommation à l’échelon national. Principaux points 5. Il est difficile de disposer en permanence de matériel de multiplication, surtout si de « nouveaux » cultivars sont introduits. Lacunes et débouchés • Il faut adopter des méthodes améliorées de production et de distribution du matériel destiné à la plantation. Stratégies • Poursuivre la recherche sur les techniques de multiplication et de diffusion. • Il faudrait disposer de différents systèmes de multiplication selon le niveau de production. Actions prioritaires 4. Recueillir et examiner les informations concernant les techniques de macro et micromultiplication. 5. Mener des recherches sur les techniques de macro et micropropagation, en mettant l’accent sur les taux obtenus, les coûts et les applications possibles (en particulier sur la formation des petits tubercules). 6. Explorer des moyens innovants de distribuer le matériel destiné à la plantation des NARES aux ONG. 6. Il faut améliorer le produit pour le rendre plus attrayant et pour que le client en redemande (l’âcreté pose souvent un problème de valorisation du produit). • Améliorer en permanence les cultures. • Approches innovantes de la sélection. 7. Évaluation et sélection du matériel génétique actuellement disponible avec la participation des agriculteurs (collections noyaux de TaroGen et du TANSAO, lignées reproductrices). 8. Sélection à l’aide de marqueurs génétiques, notamment en vue d’obtenir des qualités de comestibilité et de goût. 7. Le taro n’est pas assez exploité en vue d’une production commerciale. • Il faut améliorer les liens entre producteurs et spécialistes du marketing. • Rehausser l’image du taro en tant qu’aliment. • Il faut un équipement technique approprié pour faciliter la production. 9. Réaliser des études de marché dans les pays océaniens et ceux de la ceinture du Pacifique. 10. Définir des critères locaux de valorisation et de commercialisation des produits. 11. Faire des campagnes de promotion. 12. Étudier des techniques de mécanisation appropriées. 8. Le déclin de la fertilité du sol est un obstacle important à la production, mais l’information à ce sujet est insuffisante. • Nécessité d’une meilleure information concernant la production de taro et les pratiques culturales, en fonction de la fertilité du sol. • Inscrire le déclin de la fertilité du sol en tête des priorités des projets fondés sur des systèmes agricoles. 13. Recueillir des informations sur le recours à diverses pratiques agricoles visant à prévenir la baisse de fertilité du sol. 9. Les atolls ont des besoins spécifiques (carences nutritives, salinité). • Nécessité d’une prise de conscience et de la satisfaction des besoins des pays-atolls. • Mobiliser les pays-atolls autour des questions liés aux ressources phytogénétiques en général. 14. Étudier la possibilité de déterminer par criblage la tolérance à la salinité in vitro. 10. La culture d’autres aracées importantes pour les atolls est négligée. 15. Évaluation, par le CRMG, de variétés de taro répondant aux besoins des atolls et distribution à ces pays. 16. Étude, par PAPGREN, de la question des autres aracées comestibles. 11. Le taro n’a pas une bonne image auprès de la communauté internationale des chercheurs. • Faire en sorte que le taro soit considéré comme une culture importante par la communauté internationale des chercheurs. 17. Explorer la possibilité de rehausser l’image du taro parmi la communauté internationale des chercheurs grâce à leur participation au système du CGIAR. third taro symposium 19 Principaux points Lacunes et débouchés Stratégies Thème 4 : Amélioration génétique et distribution de matériel génétique 1. Les programmes d’amélioration génétique progressent dans le Pacifique, en Inde et à Hawaii, et visent l’acquisition de caractères spécifiques en faisant appel à des méthodes classiques, fondées sur la participation ou moléculaires. • Sélection à des fins esthétiques. • Conservation du pollen pour les programmes de sélection. • Croisements intergénériques. • Évaluation de la collection noyau afin de déterminer les caractéristiques requises et de l’utiliser comme parents. • Poursuivre la sélection classique au travers de programmes permanents afin d’étudier la tolérance aux virus et à d’autres organismes nuisibles et maladies importants, l’adaptation, le rendement et les caractéristiques qualitatives. Actions prioritaires 1. Assurer la pérennité des programmes actuels de sélection réalisés en Océanie. 2. Renforcer la coordination des programmes réalisés en Océanie en créant des réseaux. 3. Élargir la coopération internationale à l’échelon mondial. 4. Procéder à l’évaluation des collections existantes. 2. La multiplication est un point essentiel de l’utilisation du matériel génétique. • Possibilités de micro et macropropagation (en particulier l’utilisation de la technique d’induction par l’acide gibbérellique) en vue de rendre la multiplication plus efficace. • Poursuite de la recherche sur les techniques de micro et macropropagation. 4. Problème des droits de propriété sur le matériel génétique, soulevé par Palau. • Établissement de liens entre les techniques de sélection classiques à l’aide de marqueurs moléculaires (sélection à l’aide de marqueurs moléculaires). • Élaboration d’accords de transfert de matériel génétique par le CRMG. Note : L’explication suivante a été fournie ultérieurement par Herman Francisco : « Palau pose des questions sur 1) le droit de propriété sur les obtentions et sur les matériels dérivés des obtentions ; 2) la légitimité de l’utilisation de matériels donnés par des agriculteurs dans le cadre de programmes de sélection à des fins commerciales ; 3) la légitimité de l’utilisation de matériels donnés par des agriculteurs à des fins autres que celles qui ont motivé ce don, par exemple : résistance à la flétrissure des feuilles de taro. Palau a exprimé la nécessité de passer des « accords de partage des avantages » avec les agriculteurs/ cultivateurs qui ont partagé leurs matériels de multiplication et leur expérience ainsi que des savoirs indigènes ou privilégiés avec des chercheurs ou des obtenteurs. ». 20 third taro symposium 6. Poursuivre les recherches sur la technique de multiplication induite par l’acide gibbérellique. 7. Étudier l’utilisation de semences vraies. • Laboratoire de culture tissulaire des Tonga en service. 3. L’intégration de marqueurs moléculaires et d’approches biotechnologiques dans les techniques de sélection classiques est prometteuse. 5. Examiner les techniques de multiplication. • Utilisation de marqueurs moléculaires, en collaboration avec UQ, TANSAO/CIRAD et l’Université de Hawaii, en vue de la sélection à l’aide de marqueurs. En option, projets d’étudiants. 8. Renforcement des capacités nationales et formation aux techniques moléculaires, et mise en œuvre de ces techniques, en collaboration avec des instituts de recherche de pointe. • Renforcement des capacités des programmes nationaux et régionaux en matière de marqueurs moléculaires (pour l’Océanie, en option, UQ et USP). 9. Identification de marqueurs moléculaires pour faciliter la sélection. • Accord régional sur l’accès et la politique de partage des bénéfices, fondé sur la Convention relative à la diversité biologique et sur d’autres instruments régionaux/ internationaux pertinents, notamment le Traité international sur les ressources phytogénétiques pour l’alimentation et l’agriculture. 11. Élaboration d’accords de transfert de matériel génétique et d’autres instruments réglementaires en matière d’accès et de partage des avantages. 10. Caractérisation moléculaire du matériel génétique provenant de l’Inde. Principaux points Lacunes et débouchés Thème 5 : Nouveaux produits et développement 1. Nécessité de disposer de taros de bonne qualité à la sortie de la ferme. • Pouvons-nous identifier et sélectionner des chimiotypes appropriés pour les marchés actuel et futur ? Stratégies Actions prioritaires • Il faut mettre en place un réseau international auquel participeront les pays producteurs et consommateurs du reste du monde (Afrique, Antilles, Amérique centrale et du Sud, Inde, Chine), de manière à faire circuler le matériel génétique, les informations et les savoirs. 1. Analyser les chimiotypes des variétés commerciales, de manière que le secteur privé puisse promouvoir des produits sur la base de ces informations. 3. Recueillir davantage de données sur les coûts de production. 2. Données de très médiocre qualité sur les frais de production. 3. Il faut trouver de nouveaux débouchés : le marché local des taros frais est assez stable, mais celui de l’exportation de produits frais est très fragile. 2. Se concentrer sur la qualité et le goût ; utiliser des indicateurs géographiques pour promouvoir le produit en associant le nom de la variété, la zone géographique de production, les techniques de culture, etc. pour trouver des créneaux commerciaux et des nouveaux marchés haut de gamme. • Utiliser toutes les parties du végétal (par exemple pétiole et feuilles), au lieu de se limiter au rhizome. • Développer le taro organique pour trouver des créneaux commerciaux aux États-Unis, au Japon et en Europe. 4. Admettre qu’il existe deux marchés : celui des produits frais et celui des produits transformés, qui exigent un matériel végétal différent et qui, par conséquent, nécessitent des stratégies de sélection distinctes. 5. Mettre au point et mener une campagne de marketing, d’éducation et de promotion afin de distinguer le taro des autres cultures vivrières et de positionner le taro parmi les produits alimentaires sains et importants sur le plan culturel. 6. Évaluer les possibilités de production organique des variétés. third taro symposium 21 Recommandations par type d’action Faciliter la conservation 1. Valider les collections noyaux des réseaux TaroGen et TANSAO, et les comparer à d’autres banques de gènes (Inde, par exemple) à l’aide de marqueurs moléculaires normalisés. 2. Réaliser la caractérisation moléculaire du matériel génétique provenant d’Inde. 3. Mobiliser des financements à long terme pour le CRMG, par l’intermédiaire du Global Conservation Trust (Fonds mondial de conservation des ressources phytogénétiques) par exemple. 4. Assurer la duplication des collections noyaux de taro, par mesure de sécurité. 5. Fournir un soutien à court terme à des actions nationales d’entretien des banques de gènes au champ ou de conservation in vitro de collections de base. 6. Mener des recherches sur la floraison (induction, par exemple) et la conservation des semences. 8. Élaborer un projet pilote (vitrine) de gestion in situ de ressources génétiques du taro. Échanger du matériel génétique 9. Élaborer des accords régionaux de transfert de matériel et d’autres instruments réglementaires de l’ABS. 10. Étudier l’utilisation de semences vraies à des fins d’échange de gènes (plutôt que de génotypes). 11. Organiser une conférence pour mettre à jour les directives de sécurité en matière de transfert de matériel génétique. 12. Obtenir l’accord du FQID pour introduire des végétaux à des fins d’indexation. 13. Renforcer la capacité d’indexation des virus et de nettoyage dans la région. 14. Faciliter l’échange de matériel dans la région et à l’extérieur. 15. Explorer des moyens innovants de distribution de matériel destiné à la plantation aux ONG par les services nationaux de recherche et de vulgarisation agricoles. 16. Surveiller les clones exempts de pathogènes après le lâcher dans les champs. Évaluer et améliorer le matériel génétique et les produits 17. Mettre au point des méthodes économiques d’évaluation des chimiotypes. 18. Évaluer, par des méthodes participatives, et sélectionner le matériel génétique actuellement disponible (collections noyaux de TaroGen et du TANSAO, lignées reproductrices), afin d’étudier les relations entre chimiotypes, génotypes et propriétés organoleptiques. 19. Analyser les chimiotypes de variétés commerciales, de manière que le secteur privé puisse promouvoir des produits sur la base de ces informations. 20. Évaluer les possibilités de production organique des variétés. 21. Examiner la possibilité de criblage en vue de l’étude in vitro de la tolérance à la salinité. 22. Distribuer des variétés de taro aux atolls et évaluer leurs exigences particulières. 23. Assurer la poursuite et la pérennité des programmes de sélection, notamment à l’aide de marqueurs, en se concentrant sur la qualité et le goût, en identifiant les besoins particuliers des marchés en produits frais et transformés. Améliorer la production 24. Recueillir des informations sur les techniques de multiplication, les examiner et mener des recherches, au besoin (technique de multiplication induite par l’acide gibbérellique, par exemple). 25. Mener des enquêtes épidémiologiques sur le CBDV en Papouasie-Nouvelle-Guinée. 26. Réaliser des essais au champ et des analyses chimiques en vue de la lutte contre les coléoptères du taro. 27. Quantifier les pertes dues aux viroses. 28. Étudier les techniques de mécanisation appropriées. 29. Étudier l’interaction des organismes nuisibles/maladies et de l’équilibre nutritif. 30. Améliorer la valorisation après récolte. 31. Produire la documentation relative au projet d’outil interactif de diagnostic du taro. 32. Conduire des consultations sur des méthodes innovantes de lutte contre les organismes nuisibles. Documentation et échange d’informations 33. Recueillir des informations sur le recours à diverses pratiques agricoles visant à prévenir le déclin de la fertilité des sols. 34. Recueillir des informations sur les pratiques traditionnelles de culture du taro. 35. Établir le catalogue de la diversité du taro dans la région. 36. Recueillir des données sur les coûts et les problèmes de production dans les États et Territoires insulaires océaniens. Information 37. Étudier la question des autres aracées comestibles (dans le cadre de PAPGREN, par exemple). 38. Conduire des études de marché dans les États et Territoires océaniens et les pays de la ceinture du Pacifique. 39. Réaliser des campagnes de promotion et d’éducation dans les États et Territoires insulaires océaniens. Renforcement des capacités 40. Renforcer les capacités à l’échelon national, et soutenir la formation aux techniques moléculaires et appliquer celles-ci. Renforcement de la collaboration à l’échelon international 41. Créer un réseau international d’information sur le taro pour faciliter l’échange de matériel génétique, d’expertise et d’informations. 42. Explorer la possibilité de rehausser l’image du taro auprès de la communauté internationale des chercheurs, en faisant participer ceux-ci au système du CGIAR. 43. Élargir la collaboration internationale en matière de recherche sur le taro en nouant des liens entre le Pacifique et d’autres régions cultivant le taro. 22 third taro symposium ACRONYMS ABS........................... access and benefit sharing ABVC....................... alomae-bobone virus complex ACIAR...................... Australian Centre for International Agricultural Research AFLP......................... amplified fragment length polymorphism AusAID..................... Australian Agency for International Development BT............................. Bacillus thuringiensis CBD.......................... Convention on Biological Diversity CBDV....................... Colocasia bobone disease virus CBOs......................... community based organisations CGIAR...................... Consultative Group on International Agricultural Research CIP............................ Centro international de la papa (International Potato Center) CIRAD...................... Centre de coopération internationale en recherche agronomique pour le développement CLB........................... Colocasia leaf blight COGENT.................. International Coconut Genetic Resources Network CRECMI................... Cooperative Research and Extension, College of the Marshall Islands CTA........................... Technical Centre for Agricultural and Rural Cooperation ACP-EU / Centre technique de coopération agricole et rurale ACP-EU CTCRI....................... Central Tuber Crops Research Institute, India DNA.......................... deoxyribonucleic acid DsMV........................ dasheen mosaic potyvirus ESC........................... Economic and Social Council/Commission on Sustainable Development FAO........................... Food and Agricultural Organization of the United Nations FQID......................... Fiji Quarantine and Inspection Division FSM.......................... Federated States of Micronesia GA............................. gibberellic acid GDP.......................... gross domestic product HPLC........................ high performance liquid chromatography ICAR......................... Indian Council of Agricultural Research ICT............................ information and communication technology IITA........................... International Institute of Tropical Agriculture IK.............................. indigenous knowledge ILL............................ Illinois, USA INCO-DC.................. International Cooperation with Developing Countries IPGRI........................ International Plant Genetic Resources Institute IPM........................... integrated pest management KSH1........................ var. Kau-Shiung 1 of Colocasia esculenta LCD.......................... least developed countries MAFFM.................... Ministry of Agriculture, Forests, Fisheries and Meteorology, Samoa MAL.......................... Ministry of Agriculture, Solomon Islands MS............................. Murashige and Skoog (medium) MTA.......................... material transfer agreement NARI......................... National Agriculture Research Institute, Papua New Guinea NARS........................ national agricultural research system NGOs........................ non-government organisations NZAID...................... New Zealand International Aid and Development Agency ODA.......................... overseas development assistance PAPGREN................. Pacific Agricultural Plant Genetic Resources Network PGRFA...................... Plant Genetic Resources for Food and Agriculture PGR........................... plant genetic resources PICTs......................... SPC Pacific Island member countries and territories PICs........................... SPC Pacific Island member countries PMN.......................... Planting Materials Network, Solomon Islands PNG.......................... Papua New Guinea PPB........................... participatory plant breeding PRA........................... participatory rural appraisals third taro symposium 23 PSB-G2..................... a variety of Colocasia esculenta PT clones................... pathogen tested clones QUT.......................... Queensland University of Technology R&D.......................... research and development RAPD........................ random amplification of polymorphic DNA RGC.......................... Regional Germplasm Centre, Secretariat of the Pacific Community, Suva ROC.......................... Taiwan/Republic of China SARS........................ Severe Acute Respiratory Syndrome SPC........................... Secretariat of the Pacific Community RMI........................... Republic of the Marshall Islands TaBV......................... Taro bacilliform virus TANSAO................... Taro Network for South East Asia & Oceania project TaroGen..................... Taro Genetic Resources: Conservation and Utilisation project TaVCV...................... Taro vein chlorosis virus TBM.......................... SPC Taro Beetle Management project TGRC........................ Taro Genetic Resources Committee TIP............................ Taro Improvement Project, Samoa TLB........................... taro leaf blight UNITECH................. Papua New Guinea University of Technology UQ............................. University of Queensland UN............................. United Nations US............................. United States of America USDA........................ United States Department of Agriculture USP........................... University of the South Pacific WSSD....................... World Summit for Sustainable Development ACRONYMES ET SIGLES ATM.......................... accord de transfert de matériel CBDV....................... rhabdovirus responsable de la maladie « bobone » du taro CGIAR...................... Groupe consultatif pour la recherche agricole internationale CIRAD...................... Centre de coopération internationale en recherche agronomique pour le développement CRMG....................... Centre régional du matériel génétique FAO........................... Organisation des Nations unies pour l’alimentation et l’agriculture FQID......................... Service fidjien de contrôle phytosanitaire IPGRI........................ Institut international des ressources phytogénétiques IPM........................... lutte intégrée contre les organismes nuisibles NARES..................... services nationaux de recherche et de vulgarisation agricoles NARI......................... Institut national de recherche agricole (Papouasie-Nouvelle-Guinée) PAPGREN................. Réseau océanien des ressources phytogénétiques agricoles QUT.......................... Université de technologie du Queensland TANSAO................... Taro Network for Southeast Asia and Oceania (Réseau de recherche sur le taro pou l’Asie du Sud-Est et l’Océanie) TBM.......................... Projet de lutte contre les coléoptères du taro (CPS) UNITECH................. Université de technologie de Papouasie-Nouvelle-Guinée UQ............................. Université du Queensland USP........................... Université du Pacifique sud 24 third taro symposium WELCOME Hon. Jonetani Galuinadi Minister for Agriculture, Sugar and Land Resettlement, Fiji Islands T aro – or dalo, as we call Colocasia esculenta in Fiji – is an ancient crop, and a key component of farming systems in many parts of the lowland tropics in the Pacific, South East Asia, West Africa and the Caribbean. It ranks 14th worldwide among staple crops, with 9 million tonnes produced globally from some 2 million hectares. It is particularly important in the Pacific. Indeed, in many Pacific countries, including Fiji, it is considered an essential component of every meal. The corms are baked, roasted or boiled, and have great importance as a gift on formal occasions, and the leaves also represent a significant source of vitamins, especially folic acid. In addition to being an important traditional food crop, taro is a significant export commodity in a number of countries, including Fiji. The crop does face important problems though, not least the fungal disease taro leaf blight, which devastated production in Samoa and threatens other Pacific Island countries. Here in Fiji we take this threat very seriously, and have begun to put in place – together with SPC – mechanisms for, hopefully, preventing the history of taro leaf blight in Samoa repeating itself here. However, the recent rebound in Samoan taro cultivation is evidence that these various production and other problems can be overcome, in particular through the management, deployment and use of genetic diversity. Unfortunately, the taro genetic diversity is fast disappearing from many parts of the world, for example due to dietary changes and urban migration, as well as pests and diseases. In the Pacific, through the AusAID funded TaroGen project, a regional taro core collection has been established at SPC to help address this issue. These recent trends make it important to review the challenges faced by taro farmers worldwide, as well as the successes of research over the past half a decade since the Second Taro Symposium, held in Indonesia in 1994. That is the rationale for this Third Taro Symposium, as jointly organized by Secretariat of the Pacific Community, IPGRI, Food and Agricultural Organization, CIRAD and the Ministry for Agriculture, Sugar and Land Resettlement, with the additional support of CTA and Japan. The Third Taro Symposium will review progress in taro research. It will also analyse needs and priorities, develop a strategy for future work in taro research and development, explore new ways to use genetic diversity and improve taro quality and production, and stimulate international collaboration, information exchange and networking. The symposium will see a presentation of papers on a wide range of topics, from genetic diversity to production constraints to breeding and the production and marketing of new products. While I am saddened by the fact that our Chinese colleague has not been able to attend due to the SARS epidemic, I do appreciate that we have representation not only from most Pacific Island countries but also from our neighbours Australia, New Zealand and Indonesia. There are also are taro workers here from India and Ghana, and we have received papers from Vietnam and Cuba, so we have people representing all the major taro growing areas of the world. On behalf of the Ministry and the Fiji Government, I wish you all a productive symposium and a pleasant stay in our country. third taro symposium 25 ALLOCUTION DE BIENVENUE M. Jonetani Galuinadi Ministre de l’Agriculture, du Sucre et de la Tépartition foncière des Îles Fidji L e taro Colocasia esculenta – ou « dalo » selon l’appellation fidjienne – est un végétal cultivé depuis des temps immémoriaux, et l’un des piliers des systèmes agricoles de nombreuses régions des basses terres tropicales d’Océanie, de l’Asie du Sud-Est, d’Afrique occidentale et des Caraïbes. Il vient au quatorzième rang des cultures de base du monde, et sa production mondiale s’élève à neuf millions de tonnes, sur quelque deux millions d’hectares. Il revêt une extrême importance dans de nombreux pays océaniens, y compris les Îles Fidji, où il est considéré comme un composant essentiel de chaque repas. Les cormes sont cuits, grillés ou bouillis, et on en offre en guise de cadeau lors de cérémonies. Les feuilles sont également riches en vitamines, en particulier en acide folique. Culture vivrière traditionnelle importante, le taro est aussi un produit d’exportation pour un certain nombre de pays, dont les Îles Fidji. Cette culture se heurte toutefois à de graves problèmes, en particulier la flétrissure des feuilles de taro, qui a ravagé la production au Samoa et menace d’autres pays insulaires océaniens. Ici, aux Îles Fidji, nous prenons cette menace au sérieux, et nous avons commencé à prendre, en collaboration avec la CPS, des dispositions qui, nous l’espérons, éviteront que l’épidémie de flétrissure des feuilles de taro qu’a connue le Samoa ne se répète chez nous. Toutefois, la récente relance de la culture du taro au Samoa prouve aussi que l’on peut surmonter ces problèmes, notamment grâce à une bonne gestion, une implantation adéquate et l’exploitation de la diversité génétique de ce végétal. Malheureusement, cette diversité est en voie de disparition rapide dans de nombreuses parties du monde, notamment sous l’effet des nouvelles habitudes alimentaires et de la migration vers les villes, des organismes nuisibles et des maladies. En Océanie, une collection noyau régionale de taro a été créée à la CPS, grâce au projet Ressources génétiques du taro : conservation et utilisation, financé par l’AusAID, afin d’y remédier. Face à ces évolutions récentes, il importe de réexaminer les obstacles rencontrés par les cultivateurs de taro dans le monde entier et de dresser le bilan des avancées réalisées grâce à la recherche, au cours des cinq dernières années écoulées depuis le deuxième colloque sur le taro, tenu en Indonésie en 1994. Telle est la raison d’être de ce troisième colloque sur le taro, organisé conjointement par le Secrétariat général de la Communauté du Pacifique, l’Institut international des ressources phytogénétiques (IPGRI), la FAO, le Centre de coopération internationale en recherche agronomique pour le développement (CIRAD) et le ministère de l’Agriculture, du Sucre et de la Répartition foncière des Îles Fidji, avec le concours du Centre technique de coopération agricole et rurale (CTA) et du Japon Le troisième colloque sur le taro a pour objectifs de faire le point sur les progrès accomplis en matière de recherche sur le taro, d’analyser les besoins et les actions à mener en priorité, d’élaborer une stratégie d’orientation des travaux de recherche et de développement, d’explorer de nouveaux moyens de mettre en valeur la diversité génétique et d’améliorer la qualité et la production du taro, et enfin de stimuler la collaboration à l’échelon international, l’échange d’informations et la création de réseaux. Plusieurs exposés seront présentés sur un large éventail de thèmes, allant de la diversité génétique aux obstacles à la sélection et à la production et à la commercialisation de nouveaux produits. Je regrette que notre collègue chinois ne soit pas en mesure de participer à ce colloque en raison de l’épidémie de SRAS, tout en me félicitant du fait que non seulement la plupart des pays océaniens sont représentés ici, mais également nos voisins — l’Australie, la Nouvelle-Zélande et l’Indonésie. Des spécialistes du taro sont en outre venus d’Inde et du Ghana, et nous avons reçu des contributions du Vietnam et de Cuba. Les participants représenteront donc l’ensemble des grandes régions du monde qui cultivent le taro. Au nom du ministère et du gouvernement fidjiens, je vous souhaite à tous un colloque fructueux et un agréable séjour dans notre pays. 26 third taro symposium K EYNOTE ADDRESS 1 Taro Research and Development — Progress since the last Taro Symposium and Challenges for the Future G.V.H. Jackson Consultant Honourable Minister, country representatives, participants from universities, SPC and other agencies, distinguished guests, ladies and gentlemen. Thank you for inviting me to the Third Taro Symposium. This morning, I will provide some background on taro research and development, focusing on the Pacific, summarise recent progress, provide a perspective on current needs, before posing some thoughts for the future. Background It is 10 years ago to the month that taro leaf blight spread to the Samoan islands. That singular event had profound consequences. It wiped out taro production in those countries and an important food and valuable export was lost. Although a tragic and I think avoidable event, nevertheless it focused attention on the crop as never before. We would not be meeting here today if it had not happened. What have we learnt since that time? For me, there are three critical lessons: • we need to coordinate our efforts, collaborating at the national, regional and international levels; • modern biotechnologies are essential to any taro improvement programme; and • the model developed to address taro improvement has potential for other root crops. Before we go on to look at how recent R&D needs have been addressed, let me give a brief history of R&D in taro and other root crops over the last 30 years or so, to put more recent events in perspective. I am going to start in 1975, when SPC hosted a meeting in Fiji, bringing countries together for the first time to develop a regional strategy. That meeting reviewed what was taking place, and suggested the benefits of a collective approach. It led to a series of UNDP/FAO regional projects in root crops, starting in 1978 and lasting for more than a decade. The UNDP/FAO projects were regional in that similar activities took place on the same crops in several countries at the same time. In reality, they were country-specific, with little sharing of resources, information and results. The breeding strategies, virus indexing and conservation techniques were all inadequate in some way, and major issues, those relating to taro leaf blight, rationalisation, conservation and sharing of collections, were not resolved. As the UNDP/FAO projects ended, work on Papuana beetle began under an EU regional research programme. That research continues today, but during the early 1990s funding for other taro research was difficult to obtain — that is, until the outbreak of taro leaf blight in Samoa in 1993. That event, more than any other, showed the need to look at the entire crop gene pool, not just that present locally. In other words, it was time for a paradigm shift. Formulation of regional R&D strategies While work started in Samoa, focusing on fungicides to contain the problem, SPC, UNITECH, USP and other agencies hosted a series of meetings, the two of note being the 2nd Taro Symposium in Indonesia in 1994 and the Taro Seminar II in Papua New Guinea in 1995. It was at the 2nd Taro Symposium that we heard of networks — networks for taro leaf blight, networks for taro and yam improvement, and networks for genetic resource conservation and use. For the first time, the potential of molecular tools for describing germplasm and supporting breeding programmes was recognised. At the Taro Seminar II, all the elements to address taro leaf blight and to prevent further loss of varieties in the region were put together in one programme. It focused on plant breeding, molecular markers to rationalise germplasm collections, new methods of virus indexing to enable sharing within and between gene pools, and the development of a regional germplasm centre for rapid multiplication as well as conservation. It was realised that there was much to gain from collaboration; however, implementation had to deal with the reality of low investment in R&D in the Pacific. Research in agriculture is absent in many countries, and meagre at best. To overcome the constraints, national institutions were supported to accelerate work already in progress, regional institutions were involved in project implementation to ensure sustainability, and a consortium of technical expertise was employed, with wide donor involvement. third taro symposium 27 In mid 1998, TaroGen was established with AusAID funding, and a little later, ACIAR projects were set up at Queensland University of Technology and the University of Queensland on taro viruses and DNA fingerprinting. In addition, NZAID commissioned HortResearch to meet pathology needs. As Pacific Island countries were seeking innovative ways to collectively address taro leaf blight in the region, another project was considering how best to bring Asian countries into a network for taro improvement. This project, TANSAO, funded by the EU, began in 1998, with 5 Asian and 2 Pacific partners, and technical support from CIRAD and the Department of Plant Breeding, Wageningen University and Research Centre. Its objectives were similar to those of TaroGen: to collect, describe, rationalise, conserve and exchange taro germplasm for use in plant breeding programmes. Progress on regional R&D priorities So what progress has been made? To date, close to 4,000 accessions have been collected and described from five Asian and nine Pacific Island countries. From these, core collections — 170 for Asia and 210 for Pacific Islands that are representative of the genetic diversity of the regions — have been established in germplasm centres in Indonesia and Fiji. From molecular studies, we know that there are two gene pools: Asia and Pacific, where perhaps independent domestication has occurred. We know, too, that overall, the diversity of the diploid taro is rather low — just six different isozyme patterns represent 50% of the accessions. Diversity is greater in Southeast Asia than in the Pacific, near the putative centre of origin of the species, where there has been less improvement by farmers. Among the countries, highest diversity is found in Indonesia, where the two gene pools overlap. In most countries, wild and cultivated forms are genetically similar, again suggesting that crossing between plants has not been common. From the results, it was possible to suggest a PNG origin for germplasm in other Pacific Islands. In fact, the diversity in PNG encompasses that of the entire Pacific region. Some work has also been done on the taro leaf blight pathogen. DNA analysis has shown numerous strains in the Asia-Pacific region, but their relationship to pathogenicity is unknown. This work has concluded that plant breeders should start with cultivars rather than wild types — a fact well known to PNG and Samoa. Other work has detected only one taro leaf blight mating type in the Asia-Pacific region. A variety of methods has been assessed to devise a comprehensive conservation strategy. The advantages and disadvantages of in vitro and field genebanks, and the costs associated with each, have been compared. TaroGen and ACIAR analyses have shown that the cost of an in vitro collection of 200 accessions is just over F$10,000 per year. To reduce costs, ways of increasing the interval between sub-cultures are under investigation, as well as methods of cryopreservation. In situ conservation studies have begun. Studies in Solomon Islands looked at factors affecting farmers’ decisions to maintain taro varieties. Ideas for strengthening in situ conservation were tested, taking examples from other regions. Studies in Vanuatu compared two sites: one isolated, the other less so. Isolation was not the most important factor affecting conservation: a traditional system operates for maintaining and naming taro, possibly seedlings, found when garden land is reclaimed from forest. The TANSAO core collection has been shared throughout Asia. The varieties have been released in Vanuatu, but are being held in vitro in PNG and SPC because of quarantine concerns. The Pacific core collection has not yet been distributed, nor have any breeders’ lines. Many are now in quarantine with AQIS, and will soon be indexed by QUT. Taro can now be indexed using sensitive tests for all known taro viruses, except for the rhabdovirus that causes bobone and is implicated in alomae. Work on this virus continues, as well as the analysis of diseased plants collected during field surveys throughout the Pacific last year. A major component of TaroGen has been breeding, a more recent activity for TANSAO. In PNG, three lines from the 2nd cycle have been released, and the programme is now in the 5th cycle, with other releases imminent. In Samoa, breeding with the assistance of a breeders’ club under a Taro Improvement Programme with USP, MAFFM and farmers has shown the wisdom of using cultivars as the original parents, in this case, introductions from Micronesia and the Philippines. To date, six lines have been released from the 1st cycle, and others are expected to follow shortly from a programme that is now in its 4th cycle. Interest is also centred on Vanuatu, where crosses have been made between Asian and local varieties. Current taro needs What I have just outlined shows that we have made substantial progress in taro R&D in recent years. The question now is what still needs to be done. Well, core collections need to be validated — moved around the countries and evaluated. Countries that still maintain large taro collections should consider putting in place a controlled reduction in the number of accessions. And before there can be a comprehensive taro conservation strategy for the Asia-Pacific region, more studies are needed on seed storage, cryopreservation and in situ conservation. Also, agreement is required to place core collections with the FAO Global System on Plant Genetic Resources. This would safeguard the collections and ensure monitoring of their use. 28 third taro symposium For breeding, the question remains whether there is need for three separate programmes in the Pacific. Economies of scale could be achieved by combining two or all of these programmes. Whatever the decision, there is need to use germplasm from both the Asia and Pacific gene pools to broaden the genetic base of the programmes and, for this, quarantine and intellectual property considerations will need to be addressed. TANSAO has closed, and TaroGen will come to an end later this year, so in any priority setting, perhaps it’s timely to ask how the ultimate beneficiaries have fared from the research that has been done. Except in Samoa, farmers, NGOs, CBOs etc. have not been involved in the R&D process to date, or only marginally so. Many countries do not have policies to take advantage of the results of TaroGen or TANSAO. Most farmers in the Pacific, if asked, would probably have little or no idea about taro leaf blight, or the new taro varieties. There is need to get the varieties to them. In this connection, more work is required on in vitro multiplication, and the useful field multiplication techniques developed by MAFFM/FAO in Samoa need to be shared with other countries. TaroGen was about Conservation and Use, and while much progress has been made in relation to Conservation, the same cannot be said about Use, that is, use by farmers. This brings me to thoughts for the future. Thoughts for the future This meeting gives us a unique opportunity for a stocktake to see what is needed and where we might go from here. I urge you to use this meeting to re-establish who the stakeholders are and to re-validate how each can be better involved in any future initiative. I return to my opening statement, when I said that the critical lessons for me are three-fold: the benefits of broad collaboration, the use of new biotechnologies and the potential of the taro model for other root crops, sweet potato and yams in particular. I believe the major challenge is to take the concept of a taro network a step further and form a root crops network for the Pacific, linking with countries in other regions and the international institutes that specialise in the crops of concern. The significance of root crops to Pacific communities is so great that we cannot risk leaving their improvement to chance occurrences, such as an outbreak of taro leaf blight. In a recent paper, I and others referred to Doug Yen’s statement 25 years ago, that there are research problems in relation to crops of economic and agronomic importance to the region, which require inputs that would be difficult for individual countries to sustain (Ward and Yen, 1980). To overcome this, he proposed a regional research institute. We know this did not come about. But we now have the experience of TaroGen, TANSAO and networking — a model that gives us a cost-effective alternative, particularly for countries with small research capabilities. We need to capture this concept and place it in a more permanent framework, with a broader focus on root crops improvement for the long term. There are parallels in other Pacific programmes, which have benefited from the coordination of policy measures and the implementation of strategies both across the region or within sub-regions. A body of national policy makers, serviced by a secretariat within an existing regional organization is, in my view, required to continue the collaboration, and to coordinate further efforts on the priority root crops of this region. Whatever is decided, let the recommendations be realistic to the needs and resources of stakeholders. In relation to this, I leave you with a conclusion from the 1975 SPC Regional Meeting on the Production of Root Crops, just in case some of us are around in 28 years! And I quote: The first and essential agronomic activity is to describe the cultivars. This therefore is the collection and classification of the principal root crops over as wide an area as possible. This activity is presumed to be largely achievable in one year. There would be an intensive multiplication of materials … We can say today that we have completed an “essential agronomic activity” for one crop, but it has taken us a little longer than anticipated. Let us now finish the job for taro and meet the challenge for other root crops, making sure that we do not forget the “intensive multiplication of materials” so that farmers benefit. Reference Ward, J.G. and Yen, D.E. 1980. Pacific production systems. p. 67–100. In: Ward, R.G. and Proctor, A.S. (eds). South Pacific agriculture: Choices and constraints: South Pacific agricultural survey 1979. Asian Development Bank, Manila. third taro symposium 29 PRINCIPAL EXPOSÉ THÉMATIQUE Recherche et développement du taro Progrès accomplis depuis le dernier colloque sur le taro et défis à relever G.V.H. Jackson Expert-conseil Monsieur le Ministre, Mesdames et Messieurs les représentants des États et Territoires, d’universités, de la CPS et d’autres organisations, chers invités, Mesdames et Messieurs, Je vous remercie de votre invitation au troisième colloque sur le taro. Ce matin, je vais faire le point sur les travaux de recherche et de développement en matière de taro menés principalement dans la région du Pacifique. Je voudrais aussi décrire brièvement les progrès accomplis récemment, donner un aperçu des besoins actuels, avant de suggérer quelques orientations. Contexte Il y a dix ans ce mois-ci que la flétrissure des feuilles de taro a commencé à se propager aux îles samoanes. Cet événement insolite a eu des conséquences profondes. Il a eu pour effet de mettre fin à la production de taro dans ces pays et de faire disparaître un produit important pour l’alimentation et l’exportation. Malgré son caractère tragique et, je pense, évitable, cette péripétie a attiré pour la première fois toute l’attention sur cette culture. Si elle ne s’était pas produite, nous ne serions pas réunis ici aujourd’hui. Qu’avons-nous appris depuis lors ? À mon avis, nous pouvons tirer trois leçons essentielles : • nous devons coordonner nos efforts, collaborer à l’échelle nationale, régionale et internationale ; • les biotechnologies modernes sont au cœur de tout programme d’amélioration du taro ; • le modèle mis au point pour assurer l’amélioration du taro pourrait être appliqué à d’autres légumes-racines. Avant d’examiner la réponse apportée aux besoins actuels en matière de recherche et de développement, retraçons brièvement l’historique de la recherche et du développement consacrés au taro et à d’autres légumes-racines au cours des trente dernières années, de manière à replacer les événements récents dans leur contexte. Je commencerai en 1975, date à laquelle la CPS a organisé aux Îles Fidji un colloque qui a réuni, pour la première fois, des représentants des pays afin d’élaborer une stratégie régionale. Les participants ont fait le point sur les événements et envisagé les avantages d’une approche collective. À la suite de cette réunion, le PNUD et la FAO ont lancé plusieurs projets régionaux sur les légumes-racines, mis en œuvre à partir de 1978 et exécutés sur plus de dix ans. Les projets PNUD/FAO étaient de portée régionale, dans la mesure où des activités similaires concernant des cultures identiques étaient déployées dans plusieurs pays simultanément. En réalité, ces projets étaient axés sur des pays particuliers, et il n’y avait pas de partage des ressources, des informations ni des résultats. Les stratégies d’amélioration génétique, l’indexage des virus et les techniques de conservation étaient inadéquats à plusieurs points de vue, et les grandes questions, liées par exemple à la flétrissure des feuilles de taro, à la rationalisation, à la conservation et au partage de collections, n’ont pas été résolues. Au terme des projets PNUD/FAO, un programme régional de recherche, financé par l’Union européenne, fut entrepris à propos de la lutte contre le coléoptère du taro (Papuana sp.). Ces travaux de recherche se poursuivent aujourd’hui, mais, au début des années 1990, il fut difficile de mobiliser des fonds pour mener d’autres recherches sur le taro… jusqu’à ce que la flambée de flétrissure des feuilles de taro éclate au Samoa, en 1993. Plus que tout autre, cet épisode a montré la nécessité d’étudier l’ensemble du patrimoine phytogénétique, et pas seulement le patrimoine génétique des espèces locales. Autrement dit, il était temps de changer d’optique. Formulation de stratégies régionales de R&D Tandis que les travaux menés initialement au Samoa se concentraient sur la lutte fondée sur les fongicides, la CPS, l’Institut universitaire de technologie de Papouasie-Nouvelle-Guinée, l’Université du Pacifique sud et d’autres instances organisèrent d’autres réunions, les deux principales étant le deuxième Colloque sur le taro tenu en Indonésie en 1994 et le deuxième Séminaire sur le taro, tenu en Papouasie-Nouvelle-Guinée en 1995. C’est au cours du deuxième Colloque sur le taro que nous entendîmes parler de réseaux pour la première fois – réseaux de lutte contre la flétrissure des feuilles de taro, réseaux d’amélioration du taro et de l’igname, réseaux de conservation et d’utilisation des ressources phytogénétiques. Pour la première fois, on reconnut les possibilités offertes par les outils moléculaires pour la description du matériel génétique et la réalisation de programmes de sélection. 30 third taro symposium Lors du deuxième Séminaire sur le taro, tous les volets de la lutte contre la flétrissure des feuilles de taro et de la prévention de l’appauvrissement de la diversité dans la région ont été fusionnés en un seul programme. Celui-ci était axé sur l’amélioration phytogénétique, les marqueurs moléculaires qui permettent de rationaliser les collections de matériel génétique, les nouvelles méthodes d’indexage des virus autorisant des échanges au sein de banques de gènes ou entre banques de gènes et la création d’un centre régional du matériel génétique pouvant assurer la multiplication rapide et la conservation du matériel. Les participants prirent conscience des avantages d’une collaboration ; toutefois il fallait résoudre le problème réel de l’insuffisance des investissements dans la R&D dans la région du Pacifique. Nombreux sont les pays où la recherche en agriculture est inexistante, ou, au mieux, peu active. Pour surmonter les obstacles, les institutions nationales ont bénéficié d’un soutien pour accélérer les travaux en cours, les institutions régionales ont participé à la mise en œuvre de projets à long terme, et un groupe d’experts techniques a été constitué, grâce à un large concours financier des bailleurs de fonds. En 1998, en milieu d’année, le projet TaroGen (Ressources génétiques du taro : conservation et utilisation) a été lancé avec le concours financier de l’AusAID ; par la suite, le Centre australien pour la recherche agricole internationale (ACIAR) a conduit des projets sur les virus et la détermination de l’empreinte génétique du taro à l’Institut universitaire de technologie du Queensland. En outre, la NZAID a chargé HortResearch (Institut néo-zélandais de recherche sur l’horticulture et l’alimentation) d’étudier les problèmes phytopathologiques. En même temps que les pays océaniens cherchaient des moyens novateurs de remédier collectivement à la flétrissure des feuilles de taro dans la région, un autre projet portait sur les possibilités de créer un réseau de pays asiatiques d’amélioration du taro. Ce projet, TANSAO (Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie), financé par l’Union européenne, fut entrepris en 1998 avec cinq partenaires asiatiques et deux partenaires océaniens, avec le concours technique du CIRAD (Centre de coopération internationale en recherche agronomique pour le développement) et du Département d’amélioration phytogénétique de l’Université et du Centre de recherche de Wageningen. Ses objectifs étaient les mêmes que ceux de TaroGen : collecter, décrire, rationaliser, conserver et échanger du matériel génétique du taro destiné aux programmes de sélection phytogénétique. Progrès accomplis en matière de R&D régionale Dans ces conditions, quels sont les progrès accomplis ? Jusqu’à présent, près de 4 000 obtentions végétales ont été collectées et décrites dans cinq pays asiatiques et neuf pays insulaires océaniens. Sur ce chiffre, des collections noyaux (170 pour les pays asiatiques et 210 pour les pays océaniens), représentatives de la diversité génétique dans ces régions, ont été créées dans des centres de conservation du matériel génétique d’Indonésie et des Îles Fidji. Les études moléculaires nous ont montré qu’il existe deux pools génétiques, l’un en Asie et l’autre en Océanie, où une domestication autonome s’est peut-être produite. Nous savons, d’autre part, que, dans l’ensemble, la diversité du taro diploïde est assez faible : à peine six profils isoenzymatiques différents représentent 50 % des obtentions. La diversité est plus grande en Asie du Sud-Est qu’en Océanie, près du centre d’origine présumé des espèces, et là où les agriculteurs se sont moins livrés à des travaux d’amélioration génétique. C’est en Indonésie, où les deux pools génétiques se chevauchent, que l’on observe la plus grande diversité. Dans la plupart des pays, les formes sauvages et cultivées sont similaires du point de vue génétique, ce qui, là encore, laisse à penser que le croisement de végétaux n’a pas été beaucoup pratiqué. D’après les résultats obtenus, on peut penser que c’est en Papouasie-Nouvelle-Guinée que le matériel génétique trouvé dans les autres pays océaniens trouve son origine. En fait, la diversité observée en PapouasieNouvelle-Guinée est représentative de celle de tout le reste de la région du Pacifique. On a aussi étudié l’agent pathogène responsable de la flétrissure des feuilles de taro. L’analyse des empreintes génétiques a mis en évidence de nombreuses souches dans la région Asie-Pacifique, mais on ignore leurs rapports avec la pathogénicité. Il ressort de ces travaux que les sélectionneurs devraient commencer par des cultivars, plutôt que des types sauvages, méthode bien connue en Papouasie-Nouvelle-Guinée et au Samoa. D’après d’autres recherches, un seul type responsable de la flétrissure des feuilles de taro aurait été détecté dans la région Asie-Pacifique. On a évalué diverses méthodes en vue d’élaborer une stratégie exhaustive de conservation. On a ensuite comparé les avantages et inconvénients des banques de gènes conservées in vitro et sur le terrain, ainsi que leurs coûts respectifs. Des études réalisées par TaroGen et l’ACIAR ont montré que le coût d’entretien d’une collection in vitro de 200 obtentions s’élève à peu près à 10 000 F$ par an. Pour réduire ces coûts, on est en train d’examiner les moyens d’augmenter l’intervalle entre sous-cultures et les méthodes de cryoconservation possibles. Des études sur la conservation in situ ont été entreprises. Aux Îles Salomon, on a étudié les facteurs qui influent sur les décisions prises par les agriculteurs quant à l’entretien de variétés de taro. En s’appuyant sur des exemples d’autres régions, on a mis à l’épreuve des suggestions visant à l’amélioration de la conservation in situ. À Vanuatu, on a comparé deux sites, l’un isolé, l’autre moins. L’isolement n’était pas le principal facteur ayant une incidence sur la conservation : un système traditionnel permet d’entretenir et de nommer les variétés de taro, éventuellement les jeunes plants que l’on trouve lorsque des terres horticoles sont récupérées sur les forêts. La collection noyau de TANSAO a été distribuée en Asie. Des variétés ont été commercialisées à Vanuatu, mais sont conservées in vitro en Papouasie-Nouvelle-Guinée et à la CPS en raison de problèmes de quarantaine phytosanitaire. La collection noyau océanienne n’a pas encore été distribuée, non plus que des lignées sélectionnées. De nombreuses third taro symposium 31 lignées sont en quarantaine auprès du Service australien d’inspection et de contrôle zoo- et phytosanitaire (AQIS) et seront prochainement indexées par l’Université de technologie du Queensland (QUT). L’indexage du taro peut désormais être fait, à l’aide de tests précis, pour tous les virus connus sauf le rhabdovirus responsable des viroses « Bobone » et « Alomae ». Les recherches sur ce virus se poursuivent, de même que l’analyse des végétaux atteints, récoltés au cours d’enquêtes menées l’an dernier sur le terrain dans l’ensemble de la région. Volet important du projet TaroGen, la sélection est une activité plus récente pour TANSAO. En Papouasie-NouvelleGuinée, trois lignées du deuxième cycle ont été distribuées, et le programme en est maintenant au cinquième cycle. D’autres lignées seront prochainement distribuées. Au Samoa, la sélection, pratiquée avec l’aide d’un club universitaire d’obtenteurs de variétés de taro participant à un projet d’amélioration du taro, et en collaboration avec l’Université du Pacifique sud, le ministère de l’Agriculture, des Forêts, des Pêches et de la Météorologie, ainsi que des agriculteurs, a montré combien il est sage d’utiliser des cultivars comme parents originaux – en l’occurrence des introductions provenant de Micronésie et des Philippines. Jusqu’à présent, six lignées issues du premier cycle ont été distribuées, et d’autres lignées devraient prochainement suivre, issues d’un programme qui en est à son quatrième cycle. Les besoins actuels en matière de taro Ce que je viens de dire montre que, ces dernières années, nous avons accompli de grands progrès en matière de recherche et développement du taro. La question qui se pose maintenant est de savoir quel chemin il reste à parcourir. Il faut valider les collections noyaux, les distribuer dans les pays et les évaluer. Les pays qui entretiennent encore de grandes collections de taro devraient envisager de réduire progressivement le nombre d’obtentions. En outre, avant de mettre en œuvre une vaste stratégie de conservation du taro dans la région Asie-Pacifique, il faut étudier plus avant le stockage des semences, la cryoconservation et la conservation in situ. Il faut également solliciter un accord pour intégrer les collections noyaux au Système mondial FAO de conservation et d’utilisation des ressources phytogénétiques pour l’alimentation et l’agriculture, ce qui permettrait de préserver les collections et d’en surveiller l’utilisation. En ce qui concerne la sélection, il reste à savoir s’il faut définir trois programmes séparés dans la région. La fusion de deux programmes ou de tous ces programmes permettrait de réaliser des économies d’échelle. Quelle que soit la décision prise, il faut utiliser le matériel génétique issu des pools génétiques d’Asie et d’Océanie pour élargir la base génétique des programmes ; à cet effet, il faudra régler les problèmes relatifs à la quarantaine et à la propriété intellectuelle. Le projet TANSAO est terminé, et le projet TaroGen prendra fin cette année. Pour fixer des priorités, le moment est donc venu de se demander quels sont les avantages retirés par les bénéficiaires finals des recherches effectuées. Sauf au Samoa, les agriculteurs, les ONG, les organisations communautaires, etc., n’ont pas participé aux travaux de recherche et de développement jusqu’à présent, ou de façon marginale seulement. De nombreux pays n’ont pas de politique en matière d’exploitation des avantages issus de TaroGen ou de TANSAO. Si on leur posait la question, la plupart des agriculteurs océaniens n’auraient probablement aucune idée quant à la flétrissure des feuilles de taro ou aux nouvelles variétés de taro. Il faut leur distribuer ces variétés. À cet égard, il faut encore étudier la multiplication in vitro, et les techniques utiles de multiplication en champ mises au point par le ministère de l’Agriculture, des Forêts, des Pêches et de la Météorologie et la FAO au Samoa doivent être communiquées à d’autres pays. Le projet TaroGen concernait la conservation et l’utilisation. Bien des progrès ont été accomplis dans le domaine de la conservation, mais on ne peut en dire autant de l’utilisation par les agriculteurs. Cela m’amène à parler de l’avenir. Matière à réflexion Ce colloque nous a donné une occasion exceptionnelle de dresser le bilan des actions requises et possibles. Je vous invite à saisir cette occasion pour dresser à nouveau la liste des parties prenantes et réexaminer la manière dont chacune d’entre elles pourrait participer plus efficacement à des actions futures. Je répète ce que j’ai dit au début de mon intervention : à mes yeux, il y a trois leçons essentielles à tirer : les avantages d’une vaste collaboration, la mise à profit des nouvelles biotechnologies et la possibilité d’appliquer le modèle relatif au taro à d’autres légumes-racines, en particulier la patate douce et l’igname. Je pense que la principale difficulté consiste à faire progresser l’idée d’un réseau consacré au taro, et de mettre en place un réseau des légumes-racines en Océanie, en nouant des relations avec des pays d’autres régions et avec les organismes internationaux spécialisés dans les cultures en question. L’importance des légumes-racines pour les populations océaniennes est telle que nous ne devons pas laisser des aléas tels qu’une épidémie de flétrissure des feuilles de taro compromettre la chance d’améliorer ces cultures. Dans un article récent, j’ai moi-même, ainsi que d’autres auteurs, fait allusion à la déclaration de Doug Yen, il y a vingt-cinq ans, qui évoquait les problèmes posés par la recherche consacrée à des cultures présentant un intérêt économique et agronomique pour la région (Ward and Yen, 1980). Cela demande en effet aux différents pays de contribuer à des travaux qu’il leur est difficile de mener à long terme.1 Pour surmonter cet obstacle, Doug Yen proposait de créer un institut de recherche régional. Comme vous le savez, celui-ci n’a jamais vu le jour. Mais nous sommes 32 third taro symposium maintenant forts de l’expérience de TaroGen, de Tansao et de la mise en place de réseaux. Ce modèle nous apporte une solution rentable, en particulier pour les pays qui ont des capacités de recherche restreintes. Nous devons reprendre cette idée et la replacer dans un cadre plus permanent, davantage axé sur l’amélioration pérenne des légumes-racines. Il existe d’autres projets océaniens parallèles qui ont profité de la coordination de mesures et de la mise en œuvre de stratégies dans l’ensemble de la région ou à l’échelon des sous-régions. À mon avis, il faudrait constituer un groupe de responsables politiques nationaux, assisté d’un secrétariat, au sein d’une organisation régionale existante, afin d’assurer la poursuite de la collaboration et de coordonner les efforts qui seront déployés à l’avenir en faveur des légumes-racines essentiels de cette région. Quelle que soit la décision prise, faisons en sorte que les recommandations soient suffisamment réalistes pour répondre aux besoins des parties prenantes, compte tenu de leurs ressources. À cet égard, je vous laisse réfléchir à une recommandation formulée en 1975, lors de la conférence régionale de la CPS sur la production de légumes-racines, au cas où certains d’entre nous seraient encore en activité dans vingt-huit ans ! Je cite : La première activité agronomique importante à mener consiste dans la description des cultivars. Il s’agit de la collecte et de la classification des principaux légumes-racines dans la zone la plus vaste possible. Cette activité est censée être réalisée en majeure partie en un an. Il faudrait procéder à une multiplication intensive des matériels… Nous pouvons dire aujourd’hui que nous avons mené une activité agronomique importante pour une culture, mais que cela nous a pris un peu plus de temps que prévu. Finissons ce travail pour le taro, et relevons le défi pour d’autres légumes-racines, en veillant à ne pas omettre la multiplication intensive des matériels, de manière que les agriculteurs puissent en bénéficier. References Ward, J.G. and Yen, D.E. 1980. Pacific production systems. p. 67–100. In: Ward, R.G. and Proctor, A.S. (eds). South Pacific agriculture: Choices and constraints: South Pacific agricultural survey 1979. Asian Development Bank, Manila. third taro symposium 33 KEYNOTE ADDRESS 2 Taro Genetic Resources for Now and Tomorrow: A Pacific Crop Coosje Hoogendoorn1, P.N. Mathur1, Ramanatha Rao1 and Luigi Guarino2 1 International Plant Genetic Resources Institute 2 Secretariat of the Pacific Community Summary Small island developing states (SIDS) are internationally recognized as a special category of developing countries. They are hot spots for evolution: while the diversity on each island might be limited, isolation often leads to unique evolutionary patterns, for both wild biodiversity and domesticated species. The conservation and use of this biodiversity is hampered by the fact that the small size of SIDS makes local science and technology institutions relatively expensive. At the same time, SIDS and their biodiversity are particularly threatened, for example by climate change. The Pacific area is characterized by consisting largely of SIDS. Taro has a long history in the Pacific. It is an indispensable component of the local diet, with important cultural functions. It is, however, very vulnerable, because it is vegetatively propagated and quite low in genetic diversity in many places, and can be wiped out by pest and diseases, as happened in 1993 when taro leaf blight destroyed the crop on Samoa, and threatened to spread throughout the Pacific. The back-up collections of taro and breeding programmes were not able to provide good quality resistant material in response. Fortunately, the tradition of cooperation between states in the Pacific helped to set up the TaroGen project with support from Australia, which in the five years of its existence has provided a framework for developing an excellent collection of taro genetic resources and two breeding programmes, in Papua New Guinea and on Samoa. IPGRI has collaborated closely with TaroGen, and similarly now with its successor, PAPGREN, which not only works with taro, but also other typical Pacific crops. Through PAPGREN and other projects many challenges remain to be tackled. Taro is a Pacific crop with an international future, but fulfilling its promise requires socio-economic and marketing research. Only through partnerships and collaboration between the different states will it be cost effective to maintain a taro collection in vitro, which is needed to allow exchange of materials between states as well as the rest of the world. Traditional Pacific diets that include taro are changing to unhealthy, global ones. Taro can be part of a rich diet, both from a nutritional and a cultural point of view, but this will require selection of new genotypes with better micronutrient content and storage and cooking characteristics that will fit well in a Pacific diet of the future. Such initiatives are in line with the global strategy for SIDS, as developed by the UN, and approaches supported by SPC and IPGRI. Small island developing states The United Nations has recognized small island development states (SIDS) as a special case for both protection of the environment and development. In 1994, a special action plan was approved at a conference in Barbados (United Nations General Assembly, 1994). Currently, 41 countries are recognized as SIDS, amongst which are 15 Pacific countries. The main points of the Barbados Plan of Action were reinforced by the World Summit on Sustainable Development in Johannesburg in 2002. As far as the Pacific is concerned, aspects of implementation have been further discussed at the Pacific preparatory meeting in Samoa in August 2003 (United Nations, 2003). SIDS are characterized by a number of typical problems, such as those described in Box 1. Islands, and certainly those SIDS which still have much traditional countryside and wild areas, are recognized niches for evolution of land organisms, including crops. The isolation and small size of islands often results in very specific genotypes that are very different from related genotypes and species on other islands or continents. However, on the island itself, levels of biodiversity might be quite limited. This is true for both wild and domesticated species. 34 third taro symposium Box 1: Typical characteristics of small island developing states, as recognized by the UN Barbados Plan of action (United Nations General Assembly, 1994) • Small population • Lack of resources • Remoteness • Susceptibility to natural disasters • Specially susceptible to climate change • Excessive dependence on international trade • Vulnerability to global developments • Lack of economies of scale • High transportation and communication costs • Costly public administration and infrastructure Pacific island countries (PICs) range from very highly developed and relatively rich to ones classified by the UN and/or OECD as least developed countries (LCD). Both national GDP and ODA received vary tremendously among countries. Some of the islands have a governmental link to industrialized countries, such as France (New Caledonia) and New Zealand (Cook Islands), which influences the level of prosperity on the islands and usually signifies strong and direct links with science and technology organizations in the developed country concerned. As far as biodiversity is concerned, it has been recognized (United Nations ESC/Commission on Sustainable Development, 1999a) that the biodiversity on many SIDS, including that used in agriculture, forestry and fisheries, is unique. However, this biodiversity is threatened by industrial and population growth, and by the planned and accidental arrival of aggressive exotic (alien) species, which is mainly due to the very significant increase in travel. As far as agricultural biodiversity is concerned, it is generally recognized that there is a need for improved local germplasm, with an infusion of “exotic” germplasm if needed. To be able to face all these challenges to SIDS in the Pacific, new and region-specific science and technology are needed. However, indigenous knowledge is also under threat (United Nations ESC/Commission on Sustainable Development, 1999b), and there is a significant “brain drain” from many islands to less isolated science and technology environments. In addition, since to be effective science requires increasingly larger infrastructure, in general the critical mass of scientists and science and technology facilities on SIDS is now too small. Because most SIDS do not have the resources to “go it alone”, a regional approach is considered most likely to offer a solution. As far as science and technology for agriculture and biodiversity in the region are concerned, the Pacific region is the archetype of an island region. National institutions are often small and underfunded. The region, however, has a long tradition of effective regional collaboration, of which the Secretariat of the Pacific Community (SPC) is an excellent example. As far as genetic resources conservation of indigenous crops is concerned, it must be remembered that many indigenous Pacific crops are not seed crops, but rather vegetatively propagated (taro, yams, sweet potato, banana, breadfruit, etc.), and therefore cannot be conserved in classical seed genebanks, but need to be conserved in field genebanks. Field genebanks can be much more expensive to maintain than seed genebanks, and they are particularly prone to pests and diseases. Many collections have been lost in the past. The main complementary technologies for improved conservation and exchange of vegetatively propagated indigenous genetic resources, i.e. in vitro culture and cryo-preservation, are not yet very well developed for the crops concerned and/or in the region, with regard to either human resources and infrastructure. To improve this will require major efforts, both public and private, in the years to come. Taro genetic resources Taro (Colocasia esculenta) is thought to have originated in Southeast Asia and the Pacific. Some anthropologists believe that Colocasia might have been the first irrigated crop and that the ancient “rice” terraces of Asia were originally constructed for Colocasia (Plucknett, 1976). The domestication of the crop is thought to have occurred about 4000 to 7000 years ago. Taro then spread to other regions, about 2500 years ago to China and to Egypt, and slightly later to West Africa (where it is known as cocoyam). Much more recently, it spread from there to the Caribbean, as food on slave ships. It is now also grown commercially in countries such Australia and New Zealand. Taro is one of the most important crops in Pacific Island countries, where it plays an important role both as starchy staple and a leafy vegetable. Worldwide, it is the fifth most consumed root crop (FAOSTAT 2000), with over 25% produced in Oceania and Southeast Asia. The importance of the crop goes beyond its contribution to nutrition and income; in many Pacific Island countries, taro plays an important cultural role as it forms an integral part of customs and traditions. Taro is one of the oldest crops in this region, probably reaching the Polynesian islands 2000 years ago. There is now evidence to suggest that most cultivars found throughout the Pacific were not brought by the first settlers from the Indo-Malayan region as previously suggested (Kuruvilla and Singh, 1981; Léon, 1977; Plucknett et al., 1970), but may have been domesticated from wild sources in Melanesia (Lebot, 1992; Matthews, 1990, 1991, 1995; Yen, 1991a, 1991b, 1993). From there, cultivars were taken eastwards to Polynesia during prehistoric migrations, with a progressive decline in their number and diversity (Lebot, 1992; Yen, 1993; Yen and Wheeler, 1968). third taro symposium 35 Box 2: Overview of some genetic resources collections for taro Asia TANSAO India China Bangladesh Japan 2298 400 242 150 120 Pacific TaroGen Hawaii 2418 140 West-Africa ‘Cocoyam’ Cameroon IITA 70 60 Caribbean Cuba USA 40 60 Total ≈8000 Information compiled from TaroGen and TANSAO reports, IPGRI collections database and recent taro publication (Eyzaguirre et al., 2004). While it is clear that the crop is still growing in popularity at the global level, the prospects for further development in the Pacific region are constrained by small-scale production, high costs and difficulties in accessing information and markets. Taro cultivation requires specialist skills and it is a crop than can be seriously affected by disease. Because of its special requirements, Pacific farmers increasingly abandon the crop for alternatives that require less specialist skills and time input. This has also resulted in significant loss of taro genetic resources. The global distribution of taro as a popular food crop has resulted in a lively interest in the crop from the agricultural scientific community. However, coordinated networks for genetic resources conservation and use, i.e. TANSAO and TaroGen, have concentrated very much on and in the centre of origin of the crop. These two networks at the moment are responsible for assembling the most extensive taro ex situ collections, each consisting in 2003 of more than 2,000 accessions, with relatively limited overlap (Vicent Lebot, pers. comm.). Box 2 gives an overview with approximate number of these and other generally known taro collections in the world. Box 3: Ex situ collections in the Pacific as collected by TaroGen (SPC 2002) Cook Islands 18 Fiji 72 New Caledonia 82 Niue 25 PNG 859 Samoa Solomon Islands Tonga 15 824 9 Vanuatu Palau 502 12 The most important centres of taro diversity in the Pacific are found on the larger islands, in particular in Indonesia and PNG. Continuous selection by farmers over the long history of the crop has resulted in very interesting variation of landraces, but apart from the cultivated genotypes, the area is also home to relatives of taro, such as C. esculenta ‘aquatilis’ and feral populations. Disaster struck in 1993 when taro leaf blight, caused by Phytophthora colocasiae, started to spread, and caused great damage, for example on Samoa. An export industry worth US$7-10 million at the time was destroyed, including the livelihoods of many smallholders on the island. The disaster that hit Samoa was one of the main factors that let to the establishment of the TaroGen project in 1998. In 2003, after only five years, TaroGen and its partners have been able to collect, document and safeguard 2418 accessions (see Box 3). While the participating Pacific Island countries (PICs) all endeavour to establish effective field collections of their own materials, it has been recognized that it will be very difficult to maintain all genotypes in an effective way and on a long term basis, with the limited resources likely to be available for this work, after the project ends at the end of 2003. Therefore, a core collection has been identified of about 164 genotypes (Mace et al., 2004), and it is envisaged that together the PICs will undertake to support the long term conservation of this core collection in vitro and possibly in cryo-preservation, and will make the core collection available for wider use through virus cleaning and good documentation. 36 third taro symposium IPGRI and taro Box 4: IPGRI’s involvement in TaroGen teams and activities • Development and implementation of collecting strategies • Development of complementary conservation strategies • Descriptors for taro • Documentation system for taro collections • TaroGen core collection • Participation in the TGRC • Participation in and scientific backstopping for: - Taro breeding workshop (August 1998) - Taro planning workshop (September 1998) - Taro collecting strategy for Pacific islands workshop (December 1998) - Taro conservation strategy workshop (September 2001) • Start up of breeding programmes supported • Assisting in development of proposal for in situ conservation of taro in Vanuatu IPGRI has been involved in partnerships for the conservation and use of taro genetic resources almost since it was founded, now 30 years ago. One of the most important objectives of IPGRI is strengthening national programmes on plant genetic resources (PGR) conservation and use. IPGRI at present is directly collaborating on taro with national programmes in Nepal, China and Vietnam. Secondly, IPGRI aims to strengthen international collaboration on PGR conservation and use. The institute’s collaboration with SPC and its members through TaroGen (see Box 4), and also through the PAPGREN network (see Box 5) are illustrations of this approach. At the same time, IPGRI staff have been involved in meetings of the TANSAO network, which is coordinated by CIRAD and receives support from European Union INCO-DC Programme. A third major IPGRI objective is to actively stimulate through partnerships the development of new tools and new science for PGR conservation and use. Relevant taro examples are projects on the development of a descriptor list (IPGRI, 1999), the conference on Global Perspectives on Taro Genetic Resources (held in Japan in 2000), ongoing projects on in situ conservation of taro in Nepal and Vietnam, research on molecular markers and isozymes being carried out in China and Nepal, support for cryo-preservation studies in the Philippines and Fiji, and ethnobotanical studies in China. Future challenges In the last five years, through TaroGen and TANSAO, the conservation of taro genetic resources has made great strides forward. However, conservation of genetic resources is only the first step; its long term sustainability can only be justified through use of the conserved genetic resources, either directly or in crop improvement programmes. This means that taro genetic resources conservation needs the concurrent development of a strong, sustainable crop production system of which diversity is an integral pillar. Socio-economic studies One approach that is likely to benefit diversity and sustainable taro cultivation in the Pacific is promoting the active conservation of taro diversity through participatory plant breeding, diversity fairs, community biodiversity registers and seed banks, which have been found to be very effective for taro conservation and cultivation e.g. in Nepal and Vietnam (Sthapit et al., 2003). Genetic diversity is also relevant for the quality of the final product on the market. The development of new food and industrial products from taro is receiving more attention, and it will be important to investigate how genetic variation can contribute to quality products with a higher market value. Box 5: IPGRI’s contribution to the new PAPGREN network • Regional collaboration with SPC and its members • Networking model based on TaroGen • Successful project proposal development for NZAID, ACIAR • Participation in selection of PGR advisor; technical backstopping: - PAPGREN initiation and annual meetings - Scientific backstopping - PGR proposals for the region - National PGR stakeholder workshops third taro symposium 37 Finally, there seem to be opportunities for taro in mixed cropping systems involving coconut. This is being evaluated by COGENT in projects with farmers in Samoa, Tonga, PNG and Fiji (Batugal, 2000). Ex situ collection management There are many issues related to taro genetic resources conservation that need further studies. TaroGen has developed an excellent core collection along principles that should be of interest to other taro field collections and potentially to other vegetatively propagated crops. However, a core collection should not be static, but should be updated regularly to make sure that it continues to represent the most relevant genetic variation for its users. Methods for updating core collections (involving replacement, addition and deletion) need to be developed. While the methodology for in vitro conservation of taro is well developed, cryo-preservation protocols for taro need significant improvements to be able to be used reliably for long term storage. For crops such as taro, prone to pests and diseases, safe movement of germplasm is essential. Therefore efficient disease cleaning methods for plant material and robust test methodologies for pests and diseases need to be available. While all these factors are essential for efficient storage and dissemination, it is similarly important to have excellent characterization and utilization data available on the materials in the collections to enhance use in farmers’ fields and breeding programmes. The above approaches are classical components of ex situ conservation of genetic resources. In the near future, molecular genetics is likely to revolutionize genebank management. Apart from classical characterization data, both for characterization and use, DNA and/or gene profiles will be used to describe collections and useful alleles contained in accessions. Not only seeds and planting materials will be used for improvement programmes, but also isolated DNA (DNA banking). Selection using molecular markers is becoming widespread in breeding programmes, including those for globally relatively minor crops such as taro. Within the framework of these developments it will be also fitting to think more seriously about storing true seed, both from cultivars and wild relatives. Storing true seed is considered more cost effective than vegetative materials (Ramanatha Rao and Schmiediche, 1996). Molecular tools will make it much easier to make use of such materials for improved taro cultivation, either by means of marker aided selection, or by means of direct gene transfer. Nutrition and culture One of the major threats to taro production is changes in diet. The traditional Pacific diet, in which taro has a central place, is being replaced by a global diet. This change is coinciding with an increase in cardio-vascular diseases, diabetes and cancer in the region (Dr Lois Englberger, pers. comm.). Therefore SPC and IPGRI are proposing a project to address in combination the loss of traditional Pacific crops and the increase in heart disease and diabetes in Pacific, and in particular in Micronesia. The project aims to improve nutrition, food security and to reinforce cultural identities through an emphasis on typical Pacific crops, in particular banana, breadfruit, taro, swamp taro, and pandanus. The project would protect landraces of these crops and improve their use, select and disseminate nutritious varieties, and raise public awareness in the region about the value of traditional crops. Partnerships TaroGen has contributed much to stimulate very effective collaboration in the region with regard to collections, both ex situ and in situ (Eyzaguirre et al., 2004), and both at the regional level as well as at the national level. It will be important to ensure that trained staff is available at the key locations to continue the work, probably within the framework of a larger set of crops that together form the backbone of the Pacific agriculture. Partnerships will need to be effective at different levels and against different backgrounds. With the many small island nations, the regional level will need to play a pivotal role, reaching out both towards the national level as well as to the global level. Of course, such partnerships are not only needed for the conservation and use of taro genetic resources, but also for capacity building (including distance learning), coordinated marketing efforts and international policy developments, from the International Treaty on Plant Genetic Resources for Food and Agriculture to the support for SIDS as expressed by the UN and most recently at WSSD in Johannesburg. Taro could almost be seen to symbolize sustainable development. It is typical of the Pacific region, and has the clear potential to contribute to sustainable development socially, economically and environmentally. Because the crop has spread around the world, and has been adopted by many traditional and modern agricultural production systems, the significance of its development into a modern sustainable source of food and income will not be restricted to the region, but will be of benefit on a much larger scale. To achieve this, regional and global partnerships, such as discussed and presented at the workshop, will be essential. References Batugal, P. 2000. Sustainable use of coconut genetic resources to enhance incomes and nutrition of coconut smallholders in the Asia-Pacific region: Final project report for IFAD Grant No. 361. IPGRI–APO, Serdang, Malaysia. Eyzaguirre, P.B., Ramanatha Rao, V. and Matthews, P. (eds). 2004. The global diversity of taro, ethnobotany and conservation. IPGRI, Rome, Italy and MINPAKU (National Museum of Ethnology), Osaka, Japan. FAOSTAT. 2000. FAO statistical database: Agricultural production of primary crops. http://faostat.fao.org/faostat. 38 third taro symposium IPGRI. 1999. Descriptors for taro (Colocasia esculenta). International Plant Genetic Resources Institute, Rome, Italy. 56 p. Kuruvialla, K.M. and Singh, A. 1981. Karyotypic and electrophoretic studies on taro and its origins. Euphytica 30:405– 412. Lebot, V. 1992. Genetic vulnerability of Oceania’s traditional crops. Experimental Agriculture 29:309–323. Léon, J. 1977. Origin, evolution, and early dispersal of root and tuber crops. p. 20–36. In: Cook, J., MacIntyre, R. and Graham, M. (eds). Proceedings of the 4th Symposium of the International Society for Tropical Root Crops, Cali, Colombia, 1–7 August 1976. International Development Research Centre, Ottawa. Mace, E.S., Mathur, P.N., Godwin, I.D., Hunter, D., Taylor, M.B., Singh, D., DeLacy, I.H. and Jackson, G.V.H. 2004. Development of regional core collection (Oceania) for taro, Colocasia esculenta (L.), based on morphological and phenotypic characterization. In: Eyzaguirre, P.B., Ramanatha Rao, V. and Matthews, P. (eds). The global diversity of taro, ethnobotany and conservation. IPGRI, Rome, Italy and MINPAKU (National Museum of Ethnology), Osaka, Japan. Matthews, P.J. 1990. The origins, dispersal and domestication of taro. PhD thesis. Australian National University, Canberra. 421 p. Matthews, P.J. 1991. A possible tropical wild type taro: Colocasia esculenta var. aquatilis. Indo-Pacific Prehistory Bulletin 11:69–81. Matthews, P.J. 1995. Aroids and the Austronesians. Tropics 4:105–126. Plucknett, D.L. 1976. Edible aroids. p. 10–12. In: Simmonds, N.W. (ed.) Evolution of crop plants. Longman, Harlow, UK. Plucknett, D.L., de la Peña, R.S. and Obrero, F. 1970. Taro (Colocasia esculenta). Field Crop Abstracts 23:413–426. Ramanatha Rao, V. and Schmiediche, P. 1996. Conceptual basis for proposed approach to conserve sweet potato biodiversity. p. 8–15. In: Ramanatha Rao, V. (ed.) Proceedings of the workshop on the formation of a network for the conservation of sweet potato biodiversity in Asia, CIP, Bogor, Indonesia, 30 April–5 May 1996. IPGRI–APO, Serdang, Malaysia. Secretariat of the Pacific Community. 2002. TaroGen six-monthly report: April to October 2002. Sthapit, B.R., Subedi, A., Rijal, D., Rana, R. and Jarvis, D. 2003. Strengthening community-based, on-farm conservation of agricultural biodiversity. p. 344–353. In: The conservation and sustainable use of agricultural biodiversity: A sourcebook. Vol. 2: Strengthening local management of agricultural biodiversity. CIP-UPWARD, Laguna, Philippines. United Nations. 2003. Report of the Pacific Regional Meeting for the Review of the Programme of Action for the Sustainable Development of Small Island Developing States. Hotel Kitano Tusitala, Apia, Samoa, 4–8 August 2003. http://www.sidsnet.org/docshare/other/20030813142441_Apia_Meeting_Final_Report_8_August_2003.pdf. United Nations Economic and Social Council/Commission on Sustainable Development. 1999a. Progress in the implementation of the Programme of Action for the Sustainable Development of Small Island Developing States. Report of the Secretary-General. Addendum: Biodiversity resources in Small Island Developing States. E/ CN.17/1999/6/Add.5. United Nations Economic and Social Council/Commission on Sustainable Development. 1999b. Progress in the implementation of the Programme of Action for the Sustainable Development of Small Island Developing States. Report of the Secretary-General. Addendum: Science and technology in Small Island Developing States. E/ CN.17/1999/6/Add.8. United Nations General Assembly. 1994. Report of the Global Conference on the Sustainable Development of Small Island Developing States. Bridgetown, Barbados, 25 April–6 May 1994. A/CONF.167/9. Yen, D.E. 1991a. Domestication: The lessons from New Guinea. p. 558–569. In: Pawley, A. (ed.) Man and a half. Polynesian Society, Auckland. Yen, D.E. 1991b. Polynesian cultigens and cultivars: The questions of origin. p. 67–95. In: Cox, P.A. and Banack, S.A. (eds). Islands, plants and Polynesians. Dioscorides Press, Portland, Oregon. Yen, D.E. 1993. The origins of subsistence agriculture in Oceania and the potential for future tropical food crops. Economic Botany 47:3–14. Yen, D.E. and Wheeler, J.M. 1968. Introduction of taro into the Pacific: The indications of chromosome numbers. Ethnology 7:259–267. third taro symposium 39 PRINCIPAL EXPOSÉ THÉMATIQUE Ressources génétiques d’aujourd’hui et de demain : le taro, une plante océanienne Coosje Hoogendoorn1, P. N. Mathur1, Ramanatha Rao1 et Luigi Guarino2 Institut international des ressources phytogénétiques (IPGRI) 2 Secrétariat général de la Communauté du Pacifique 1 Résumé Les petits États insulaires en développement sont reconnus sur la scène internationale comme une catégorie à part de pays en développement. Ce sont des « points chauds » du processus d’évolution : la diversité peut être limitée sur chaque île, mais cet isolement est porteur de schémas particuliers d’évolution de la biodiversité, tant pour les espèces sauvages que domestiquées. La conservation et l’utilisation de cette biodiversité se heurtent aux difficultés liées à la petite taille des petits pays insulaires en développement, qui rend relativement onéreux la mise en place et le fonctionnement d’établissements locaux à vocation scientifique et technique. Dans le même temps, ces pays et leur biodiversité sont particulièrement menacés, par les changements climatiques, par exemple. Une des caractéristiques de la région océanienne est qu’elle regroupe en majorité des petits pays insulaires en développement. Le taro est présent dans le Pacifique depuis longtemps. Élément indispensable de l’alimentation locale, il a d’importantes fonctions d’ordre culturel. Il est néanmoins très vulnérable parce qu’il se propage par multiplication végétative, qu’en certains endroits, sa diversité génétique est très faible et qu’il peut être anéanti par des organismes nuisibles, comme ce fut le cas en 1993, lorsque le pathogène de la flétrissure des feuilles du taro détruisit les cultures au Samoa et menaça de s’étendre à tout le Pacifique. Les collections de secours et les programmes de sélection génétique ne furent pas en mesure de fournir en remplacement du matériel de qualité, résistant à cette maladie. Heureusement, forts de la tradition de coopération qui prévaut entre pays du Pacifique et avec le concours de l’Australie, nous avons établi le Projet TaroGen. Pendant ses cinq années d’existence, ce projet s’est révélé être un cadre tout à fait précieux pour l’élaboration d’une excellente collection de ressources génétiques du taro et de deux programmes de sélection génétique en Papouasie-Nouvelle-Guinée et au Samoa. L’IPGRI (Institut international des ressources phytogénétiques) a collaboré étroitement avec les responsables du Projet TaroGen. Il en va de même aujourd’hui dans le cadre du nouveau Projet PAPGREN, qui couvre non seulement le taro mais aussi d’autres espèces cultivées typiquement océaniennes. Dans le cadre de PAPGREN et d’autres programmes, nombre de défis restent à relever. Le taro est une plante océanienne appelée à un avenir mondial ; mais, pour réaliser ce potentiel, des recherches sur les aspects socioéconomiques ainsi que des études de marché s’imposent. Ce n’est que par le biais de partenariats et grâce à la collaboration entre États qu’il peut être économiquement viable de conserver in vitro une collection de taro, collection nécessaire aux échanges de matériels entre pays de la région et avec le reste du monde. Les modes d’alimentation océaniens traditionnels, où le taro était au menu, se transforment aujourd’hui en régimes universellement malsains. Le taro peut s’imposer comme aliment riche tant du point de vue nutritif que culturel, mais il faudra sélectionner de nouveaux génotypes avec une meilleure teneur en micronutriments, qui se conservent mieux et se cuisinent plus facilement, afin que le taro conserve sa place dans l’alimentation de l’Océanien de demain. Ces objectifs sont conformes à la stratégie mondiale en faveur des petits États insulaires en développement, élaborée par les Nations unies, mais aussi aux principes que défendent la CPS et l’IPGRI. 40 third taro symposium Les petits États insulaires en développement Encadré 1: Caractéristiques des petits États insulaires en développement, reconnues dans le Programme d’action de la Barbade (Nations unies, Assemblée générale, 1994) • Population limitée • Absence de ressources • Isolement • Vulnérabilité aux catastrophes naturelles • Susceptibilité particulière face aux changements climatiques • Dépendance extrême à l’égard du commerce international • Vulnérabilité aux chocs exogènes • Absence d’économies d’échelle • Coût élevé des transports et des communications • Coût élevé des infrastructures et des services publics. Aux yeux des Nations unies, les petits États insulaires en développement (PEID) représentent un cas à part tant du point de vue de la protection de l’environnement que du développement. En 1994, un plan d’action spécifique fut adopté lors d’une conférence à la Barbade (Nations unies, Assemblée générale, 1994). Aujourd’hui, 41 pays ont le statut de PEID, dont 15 pays du Pacifique. Les points saillants du Programme d’action de la Barbade ont été renforcés lors du Sommet mondial sur le développement durable qui s’est tenu à Johannesburg, en 2002. Pour ce qui est du Pacifique, la mise en œuvre du Programme a été examinée en plus grand détail lors d’une réunion préparatoire organisée au Samoa en août 2003 (Nations unies, 2003). Les PEID sont confrontés, de manière caractéristique, à un certain nombre de problèmes tels que ceux décrits à l’encadré 1. Les îles, et notamment les petits pays insulaires en développement qui disposent encore de vastes étendues rurales traditionnelles ou sauvages, sont de véritables niches d’évolution pour les organismes terrestres, dont les espèces cultivées. L’isolement et la superficie limitée des îles produisent souvent des génotypes très spécifiques, qui sont tout à fait distincts de génotypes et d’espèces apparentées d’autres îles ou d’autres continents. Toutefois, sur les îles en question, la biodiversité peut être assez réduite. Cela est vrai pour les espèces sauvages comme pour les espèces domestiquées. La situation des pays insulaires océaniens n’est pas uniforme. Elle peut aller de celle de pays très développés et relativement riches à celle d’un pays classé par les Nations unies ou l’OCDE comme étant parmi les moins avancés. Tant le PIB que le montant d’aide publique au développement varient énormément selon les pays. Certaines îles ont un lien officiel avec un pays industrialisé comme la France (pour la Nouvelle-Calédonie) ou la Nouvelle-Zélande (pour les Îles Cook), ce qui a un effet sur le niveau de vie intérieur et sous-entend généralement des relations directes et solides avec le monde de la science et de la technique dans le pays industrialisé en question. Pour ce qui est de la biodiversité, il a été reconnu (Nations unies, ECOSOC/Commission du développement durable, 1999a) que dans de nombreux petits pays en développement, elle est unique en son genre, y compris dans le cas des espèces exploitées à des fins agricoles, sylvicoles ou halieutiques. Pourtant, cette biodiversité est menacée par la croissance industrielle et démographique, et par l’arrivée programmée ou accidentelle d’espèces exotiques (étrangères) agressives (qui s’explique essentiellement par la nette augmentation des voyages). S’agissant de la biodiversité dans le domaine agricole, on s’accorde généralement sur la nécessité à disposer de matériel génétique local, amélioré, le cas échéant, par un apport en matériel exotique. Pour pouvoir relever tous ces défis auxquels sont confrontés les petits pays insulaires du Pacifique, de nouveaux moyens scientifiques et techniques, adaptés à la région, sont nécessaires. Néanmoins, les savoirs autochtones sont eux aussi menacés (Nations unies, ECOSOC/Commission du développement durable, 1999b), et la « fuite des cerveaux » est substantielle depuis de nombreux pays insulaires vers des sphères scientifiques et techniques moins isolées. Qui plus est, la science nécessitant des moyens de plus en plus importants pour être efficace, la masse critique de chercheurs et d’installations scientifiques et techniques des petits pays insulaires en développement est trop limitée à l’heure actuelle. La plupart de ces pays n’ont pas les moyens de faire cavalier seul, et une approche régionale semble s’imposer. S’agissant de la science et de la technique appliquées à l’agriculture et à la biodiversité, la région océanienne est l’essence même de l’espace de type insulaire. Les institutions nationales ont le plus souvent une taille limitée et un budget restreint. Mais on y trouve une longue tradition de collaboration régionale de portée réelle, dont un excellent exemple est le Secrétariat général de la Communauté du Pacifique. En matière de conservation des ressources génétiques, il ne faut pas oublier que de nombreuses plantes locales cultivées ne se reproduisent pas par germination mais se propagent de manière végétative (le taro, l’igname, la patate douce, l’arbre à pain, etc.). Elles ne peuvent donc pas être conservées dans une banque de semences classique, mais doivent l’être in situ, « au champ ». Les banques de gènes in situ, souvent beaucoup plus coûteuses que les banques de gènes de semences, sont particulièrement exposées aux attaques des organismes nuisibles. De nombreuses collections ont ainsi déjà été détruites. Les principales techniques complémentaires pouvant améliorer la conservation et l’échange de ressources génétiques locales à multiplication végétative, à savoir la culture in vitro et la cryopréservation, ne third taro symposium 41 sont pas encore très avancées, pour les plantes concernées et dans la région, en termes de ressources humaines et d’infrastructures. Il faudrait pour cela que les sphères publique et privée consentent un effort majeur dans les années qui viennent. Les ressources génétiques du taro Il est dit que le taro (Colocasia esculenta) provient d’Asie du Sud-Est et du Pacifique. Certains anthropologues estiment que les premières cultures irriguées étaient des cultures de Colocasia et que les « rizières » en terrasse d’Asie avaient été construites, à l’origine, dans ce but (Plucknett, 1976). Il semble que le taro ait été domestiqué il y a environ 4 000 à 7 000 ans avant d’atteindre, il y a près de 2 500 ans, la Chine et l’Égypte, puis, un peu plus tard, l’Afrique de l’Ouest. Cette plante, aliment des esclaves à bord des bateaux qui les transportaient, est arrivée beaucoup plus récemment aux Caraïbes. Elle est aujourd’hui cultivée à des fins commerciales dans des pays comme l’Australie et la Nouvelle-Zélande. Le taro est une des plus importantes espèces cultivées dans les pays insulaires du Pacifique, où il joue un rôle de premier rang en tant que féculent de base et en tant que légume feuillu. À l’échelle mondiale, il se place au cinquième rang des tubercules consommés (FAOSTAT, 2000), un quart des quantités produites l’étant en Océanie et en Asie du Sud-Est. L’importance du taro dépasse sa portée nutritive et économique. Dans beaucoup de pays océaniens, il a un rôle culturel capital car il fait partie intégrante des coutumes et traditions. Le taro est une des plus anciennes plantes cultivées de la région, puisqu’il est sans doute apparu dans les îles de Polynésie il y a 2 000 ans. Les éléments dont on dispose donnent désormais à penser que la plupart des cultivars présents dans le Pacifique ne sont pas arrivés avec les premiers colonisateurs venus de la zone indo-malaise, comme on le pensait jusque-là (Kuruvilla et Singh, 1981 ; Léon, 1977 ; Plucknett et al., 1970), mais qu’ils auraient pu être domestiqués à partir de sources sauvages en Mélanésie (Lebot, 1992 ; Matthews, 1990, 1991, 1995 ; Yen, 1991a, 1991b, 1993). De là, des cultivars ont voyagé vers l’est pour atteindre la Polynésie avec les migrations préhistoriques, leur nombre et diversité déclinant ensuite progressivement (Lebot, 1992 ; Yen, 1993 ; Yen et Wheeler, 1968). Le taro est certes de plus en plus populaire dans le monde, mais ses perspectives de développement dans la région du Pacifique se heurtent aux problèmes d’une production à petite échelle, aux coûts élevés et aux difficultés d’accès aux informations et aux marchés. La production de taro exige des compétences spécialisées, et c’est une culture qui peut être gravement affectée par les maladies. C’est pourquoi les agriculteurs océaniens abandonnent le taro au profit d’autres productions qui demandent moins de compétences spécialisées et moins de temps. Il en résulte une perte notable du patrimoine génétique du taro. Encadré 2: Quelques collections de matériel génétique du taro Asie TANSAO Inde Chine Bangladesh Japon 2 298 400 242 150 120 Pacifique TaroGen Hawaii 2 418 140 Afrique de l’Ouest Cameroun IITA 70 60 Caraïbes 40 60 Cuba États-unis d’Amérique Total ≈ 8 000 Informations issues de rapports des réseaux TaroGen et TANSAO, de la base de données de l’IPGRI et d’une publication récente consacrée au taro (Eyzaguirre et al., 2004) 42 third taro symposium Encadré 3: Collections ex situ dans le Pacifique recueillies dans le cadre de TaroGen (CPS 2002) Îles Cook 18 Fidji 72 Nouvelle-Calédonie 82 Niue 25 PNG 859 Samoa 15 Îles Salomon 24| Tonga 9 Vanuatu 502 Palau 12 L’omniprésence mondiale du taro, aliment populaire, suscite un vif intérêt chez les agronomes. Néanmoins, les réseaux de coordination qui veillent à améliorer la conservation et l’utilisation des ressources génétiques de cette plante (à savoir, TANSAO et TaroGen) ont concentré la plupart de leurs efforts autour de son aire d’origine. Ces deux réseaux sont aujourd’hui chargés de mettre sur pied les plus importantes collections in situ, chacune comptant, en 2003, plus de 2 000 obtentions, avec un chevauchement relativement limité (Lebot, communication personnelle). L’encadré 2 donne un aperçu de ces deux collections et d’autres parmi les plus connues ailleurs dans le monde, avec une indication approximative des quantités concernées. Encadré 4: Participation de l’IPGRI aux équipes et aux activités du Projet TaroGen • Élaboration et application de stratégies de collecte. • Élaboration de stratégies complémentaires de conservation. • Descripteurs du taro. • Systèmes de documentation pour les collections de taro. • Collection « noyau » de TaroGen. • Participation aux travaux du comité chargé des ressources génétiques du taro. • Participation et soutien scientifique lors de séminaires consacrés à : • la sélection du taro (août 1998), • la planification des activités en faveur du taro (septembre 1998), • la définition d’une stratégie de collecte de ressources génétiques du taro pour les îles du Pacifique, • la définition d’une stratégie de conservation du taro (septembre 2001). • Soutien au démarrage des programmes de sélection. • Aide à la préparation d’un avant-projet de conservation in situ du taro, à Vanuatu. Ce sont les plus grandes îles qui sont les plus importants foyers de diversité biologique du taro dans le Pacifique, notamment l’Indonésie et la Papouasie-Nouvelle-Guinée. La sélection effectuée en permanence par les agriculteurs au cours de la longue existence de cette plante a amené une variation très intéressante des variétés locales, mais outre les génotypes cultivés, la zone abrite également des espèces apparentées au taro, comme C. esculenta « aquatilis », et des populations sauvages. En 1993 survient une catastrophe avec l’apparition de la flétrissure des feuilles de taro causée par Phytophthora colocasiae, qui se répand et entraîne de gros dégâts, au Samoa, par exemple. C’est tout un secteur d’exportation, représentant à l’époque entre sept et dix millions de dollars des États-Unis d’Amérique, ainsi que les moyens d’existence de nombreux petits exploitants, qui sont détruits dans ce pays. Le drame qui frappa le Samoa est un facteur capital à l’origine du Projet TaroGen, en 1998. Dans les cinq ans qui suivirent, en 2003, les chargés du projet et leurs partenaires avaient déjà collecté, caractérisé et mis à l’abri 2 418 obtentions (voir encadré 3). Tous les pays océaniens participant au Projet s’efforcent certes de créer leur propre collection in situ de matériel génétique du taro, mais il sera extrêmement difficile de maintenir tous les génotypes efficacement et à long terme, étant donné le peu de moyens susceptibles d’être mobilisés une fois que le Projet arrivera à son terme, en 2003. Une collection « noyau » a donc été constituée avec près de 164 génotypes (Mace et al., 2004). Il est envisagé que les pays insulaires du Pacifique s’engagent à assurer ensemble la conservation durable de cette collection noyau, in vitro ou éventuellement par cryopréservation, et qu’ils la rendent disponible pour une utilisation plus générale, après élimination des virus et réalisation d’une description détaillée. third taro symposium 43 L’IPGRI et le taro Encadré 5: Participation de l’IPGRI au nouveau réseau PAPGREN • Collaboration régionale avec la CPS et ses membres. • Travail en réseau sur le même modèle que TaroGen. • Préparation d’un avant-projet soumis avec succès à la NZAID (Agence néo-zélandaise pour le développement international) et à l’ACIAR. • Participation au processus de sélection d’un conseiller pour les ressources phytogénétiques. • Soutien technique : – lancement et réunions annuelles de PAPGREN, – soutien scientifique, – demandes de financement relatives aux ressources phytogénétiques de la région, • Ateliers nationaux réunissant les parties concernées par les ressources phytogénétiques. L’IPGRI est partie prenante, quasiment depuis sa création il y a trente ans, à divers partenariats forgés en vue de la conservation et de l’utilisation du patrimoine génétique du taro. Au premier rang des objectifs figure le renforcement des programmes nationaux de conservation et d’utilisation des ressources phytogénétiques. L’Institut collabore directement à ce sujet avec le Népal, la Chine et le Vietnam dans le cadre de leur programme national à chacun sur le taro. En deuxième lieu, l’IPGRI oeuvre à renforcer la collaboration internationale pour la conservation et l’utilisation du patrimoine phytogénétique. La coopération mise en place entre l’IPGRI, la CPS et les membres de cette dernière par le biais du Projet TaroGen (voir encadré 4) mais aussi du réseau PAPGREN (voir encadré 5), en est la traduction concrète. Parallèlement, les agents de l’IPGRI prennent part aux réunions du réseau TANSAO, que coordonne le CIRAD et qui bénéficie d’un concours de l’Union européenne au titre du Programme INCO-DC. Un troisième grand objectif pour l’IPGRI est de stimuler activement, par le biais de partenariats, la mise au point de nouveaux outils et l’acquisition de nouvelles connaissances scientifiques en matière de conservation et d’utilisation des ressources phytogénétiques. Dans le domaine du taro, il convient de citer les projets d’élaboration d’une liste de descripteurs (IPGRI, 1999), la conférence consacrée à une réflexion mondiale sur le patrimoine génétique du taro (qui s’est tenue au Japon en 2000), des projets actuellement consacrés à la conservation in situ du taro au Népal et au Vietnam, la recherche en cours en Chine et au Népal sur les marqueurs moléculaires et les isozymes, le concours apporté aux travaux sur la cryopréservation conduits aux Philippines et aux Îles Fidji, et des recherches ethnobotaniques menées en Chine. Les défis à relever Depuis cinq ans, grâce au Projet TaroGen et à TANSAO, la conservation des ressources génétiques du taro avance à grands pas. Cependant, la conservation ne constitue qu’une première étape, la pérennité de cette entreprise ne pouvant se justifier qu’avec l’utilisation des ressources conservées, soit directement soit dans le cadre de programmes d’amélioration des cultures. C’est pourquoi la conservation des ressources génétiques du taro doit s’appuyer sur la mise en place simultanée d’un système solide et durable de production, dont le pilier central sera la diversité biologique. Aspects socio-économiques Pour promouvoir la diversité biologique et pérenniser la culture du taro dans le Pacifique, il pourrait être utile de mener des actions volontaristes telles que la sélection phytogénétique selon des méthodes dites participatives, l’organisation de bourses d’échange de matériel génétique, l’établissement de registres locaux du matériel génétique et des banques de semences, autant d’interventions qui se sont révélées très efficaces pour la conservation et la culture du taro au Népal et au Vietnam, par exemple (Sthapit et al., 2003). La diversité génétique entre également en ligne de compte dans la qualité du produit final mis sur le marché. La création de nouveaux produits alimentaires et industriels à base de taro suscite un intérêt grandissant, et il conviendra de déterminer dans quelle mesure la variation génétique peut permettre d’accroître la qualité et la valeur commerciale des produits. Enfin, des perspectives semblent se dessiner pour ce qui est de la culture associée du taro et du cocotier. Le COGENT (Réseau de matériel génétique pour le cocotier) procède actuellement à une évaluation de cette possibilité dans le cadre de projets conduits avec des agriculteurs au Samoa, aux Tonga, en Papouasie-Nouvelle-Guinée et aux Îles Fidji (Batugal, 2000). Gestion de collections ex situ De nombreux aspects de la conservation des ressources génétiques du taro méritent une étude plus approfondie. Le Projet TaroGen a mis sur pied une excellente collection noyau selon des principes qui devraient intéresser les autres collections de taro in situ, voire les autres plantes à multiplication végétative. Toutefois, une collection noyau n’est pas un élément statique ; elle doit être actualisée régulièrement afin de présenter les variations génétiques les plus pertinentes pour l’utilisateur. Des méthodes doivent être développées pour cette mise à jour des collections noyaux (remplacements, ajouts et suppressions). Les méthodes de conservation in vitro du taro sont certes bien au point, mais les protocoles de cryopréservation de cette plante, quant à eux, doivent encore être amplement perfectionnés avant 44 third taro symposium d’être utilisés de manière fiable pour le stockage de longue durée. Pour les plantes cultivées qui, comme le taro, sont vulnérables aux attaques d’organismes nuisibles, il est essentiel de veiller à l’innocuité de tout matériel génétique qui circule. Il faut pour cela disposer de moyens efficaces de « nettoyage » du matériel végétal et de tests robustes de mise en évidence de la présence d’organismes nuisibles. Tous ces facteurs sont fondamentaux pour un stockage et une diffusion corrects. Mais il est tout aussi important de disposer d’informations de qualité sur la caractérisation et l’utilisation du matériel que contient une collection, afin qu’il soit employé dans les meilleures conditions dans le cadre de programmes de sélection et par les agriculteurs dans leurs champs. Ces principes sont des éléments tout à fait classiques de la conservation ex situ de ressources phytogénétiques. Bientôt, la génétique moléculaire viendra vraisemblablement bouleverser la gestion des banques de gènes. Outre les données habituelles de caractérisation, qui permettent également de déterminer les utilisations les plus appropriées, les profils ADN et autres profils génétiques seront employés pour décrire les collections ainsi que les allèles utiles des diverses obtentions. Dans le cadre de programmes d’amélioration, non seulement les semences et le matériel destiné à la plantation seront employés, mais aussi des éléments d’ADN isolé (grâce aux banques d’ADN). Le recours aux marqueurs moléculaires est de plus en plus fréquent dans les programmes de sélection, y compris pour les cultures dont l’importance est relativement moindre comme le taro. En raison de tous ces changements, une réflexion s’impose sur le stockage des semences « véritables », que ce soit pour les cultivars ou pour des espèces voisines sauvages. On considère que le stockage de semences véritables est plus économique que celui du matériel de multiplication végétative (Ramanatha Rao et Schmiediche, 1996). Les outils moléculaires faciliteront grandement l’emploi de ces matériels pour améliorer la culture du taro, par une sélection mettant en jeu des marqueurs, ou encore par le transfert direct de gènes. Nutrition et culture L’évolution des modes d’alimentation fait peser une des menaces les plus sérieuses sur la production de taro. Le régime alimentaire océanien, dont le taro était la pièce de résistance, cède la place à un modèle mondial. Cette évolution s’accompagne dans la région d’une augmentation des maladies cardiovasculaires, du diabète et du cancer (Dr Lois Engleberger, communication personnelle). C’est pourquoi la CPS et l’IPGRI préparent un projet qui traiterait en même temps du recul des cultures océaniennes traditionnelles et de la progression des maladies cardiaques ainsi que du diabète dans le Pacifique, notamment en Micronésie. L’objectif de ce projet serait d’améliorer la nutrition et la sécurité alimentaire tout en renforçant les identités culturelles en mettant l’accent sur les végétaux cultivés les plus représentatifs de la région, dont la banane, l’arbre à pain et son fruit, le taro, le taro d’eau et le pandanus. Dans le cadre de ce projet, les variétés locales de ces espèces seraient protégées et mieux utilisées, les variétés les plus nutritives seraient sélectionnées et diffusées, et une campagne d’information permettrait de rappeler aux populations de la région la valeur des espèces qu’elles cultivent depuis toujours. Partenariats Le Projet TaroGen a beaucoup fait pour stimuler, au sein de la région, une collaboration très efficace autour des collections in situ et ex situ (Eyzaguirre et al., 2004), tant à l’échelon régional que national. Il est important de veiller à ce que des agents soient formés et soient disponibles en des points clés pour poursuivre ces efforts qui viseraient probablement la gamme plus étendue des espèces cultivées sur lesquelles repose l’agriculture océanienne. Les partenariats mis en place devront apporter la preuve de leur efficacité à divers niveaux et dans divers contextes. Pour ce qui intéresse les très petits pays insulaires, le maillon régional, véritable charnière entre la sphère nationale et la sphère mondiale, devra jouer un rôle décisif. De tels partenariats sont, bien entendu, nécessaires non seulement pour conserver et utiliser les ressources génétiques du taro, mais aussi pour créer de nouvelles capacités (mettant notamment en jeu les systèmes d’éducation à distance), coordonner les efforts de commercialisation et influer sur les politiques internationales, qu’il s’agisse du Traité international sur les ressources phytogénétiques pour l’alimentation et l’agriculture, ou du soutien exprimé en faveur des petits États insulaires en développement par les Nations unies et plus récemment à Johannesburg, lors du Sommet mondial sur le développement durable. Le taro pourrait pratiquement devenir le symbole de ce type de développement. Plante typiquement océanienne, il pourrait clairement contribuer au développement durable, sous l’angle social, économique et écologique. Cette plante s’est répandue dans le monde entier et elle a été intégrée à de nombreux systèmes de production agricole, tant traditionnels que modernes. Ainsi, sa transformation en une source pérenne de nutrition et de revenus de l’ère moderne dépassera les frontières du Pacifique pour porter ses fruits à bien plus grande échelle. Les partenariats de portée régionale et mondiale décrits et envisagés à l’occasion de ce séminaire en seront la pierre angulaire. Bibliographie Batugal, P. 2000. Sustainable use of coconut genetic resources to enhance incomes and nutrition of coconut smallholders in the Asia-Pacific region: Final project report for IFAD Grant No. 361. IPGRI–APO, Serdang, Malaysia. Eyzaguirre, P.B., Ramanatha Rao, V. and Matthews, P. (eds). 2004. The global diversity of taro, ethnobotany and conservation. IPGRI, Rome, Italy and MINPAKU (National Museum of Ethnology), Osaka, Japan. FAOSTAT. 2000. FAO statistical database: Agricultural production of primary crops. http://faostat.fao.org/faostat. IPGRI. 1999. Descriptors for taro (Colocasia esculenta). International Plant Genetic Resources Institute, Rome, Italy. 56 p. third taro symposium 45 Kuruvialla, K.M. and Singh, A. 1981. Karyotypic and electrophoretic studies on taro and its origins. Euphytica 30:405– 412. Lebot, V. 1992. Genetic vulnerability of Oceania’s traditional crops. Experimental Agriculture 29:309–323. Léon, J. 1977. Origin, evolution, and early dispersal of root and tuber crops. p. 20–36. In: Cook, J., MacIntyre, R. and Graham, M. (eds). Proceedings of the 4th Symposium of the International Society for Tropical Root Crops, Cali, Colombia, 1–7 August 1976. International Development Research Centre, Ottawa. Mace, E.S., Mathur, P.N., Godwin, I.D., Hunter, D., Taylor, M.B., Singh, D., DeLacy, I.H. and Jackson, G.V.H. 2004. Development of regional core collection (Oceania) for taro, Colocasia esculenta (L.), based on morphological and phenotypic characterization. In: Eyzaguirre, P.B., Ramanatha Rao, V. and Matthews, P. (eds). The global diversity of taro, ethnobotany and conservation. IPGRI, Rome, Italy and MINPAKU (National Museum of Ethnology), Osaka, Japan. Matthews, P.J. 1990. The origins, dispersal and domestication of taro. PhD thesis. Australian National University, Canberra. 421 p. Matthews, P.J. 1991. A possible tropical wild type taro: Colocasia esculenta var. aquatilis. Indo-Pacific Prehistory Bulletin 11:69–81. Matthews, P.J. 1995. Aroids and the Austronesians. Tropics 4:105–126. Nations unies. 2003. Report of the Pacific Regional Meeting for the Review of the Programme of Action for the Sustainable Development of Small Island Developing States. Hotel Kitano Tusitala, Apia, Samoa, 4–8 August 2003. http://www.sidsnet.org/docshare/other/20030813142441_Apia_Meeting_Final_Report_8_August_2003.pdf. Nations unies, Assemblée générale. 1994. Report of the Global Conference on the Sustainable Development of Small Island Developing States. Bridgetown, Barbados, 25 April–6 May 1994. A/CONF.167/9. Nations unies, ECOSOC/Commission du développement durable. 1999a. Progress in the implementation of the Programme of Action for the Sustainable Development of Small Island Developing States. Report of the SecretaryGeneral. Addendum: Biodiversity resources in Small Island Developing States. E/CN.17/1999/6/Add.5. Nations unies, ECOSOC/Commission du développement durable. 1999b. Progress in the implementation of the Programme of Action for the Sustainable Development of Small Island Developing States. Report of the SecretaryGeneral. Addendum: Science and technology in Small Island Developing States. E/CN.17/1999/6/Add.8. Plucknett, D.L. 1976. Edible aroids. p. 10–12. In: Simmonds, N.W. (ed.) Evolution of crop plants. Longman, Harlow, UK. Plucknett, D.L., de la Peña, R.S. and Obrero, F. 1970. Taro (Colocasia esculenta). Field Crop Abstracts 23:413–426. Ramanatha Rao, V. and Schmiediche, P. 1996. Conceptual basis for proposed approach to conserve sweet potato biodiversity. p. 8–15. In: Ramanatha Rao, V. (ed.) Proceedings of the workshop on the formation of a network for the conservation of sweet potato biodiversity in Asia, CIP, Bogor, Indonesia, 30 April–5 May 1996. IPGRI–APO, Serdang, Malaysia. Secretariat of the Pacific Community. 2002. TaroGen six-monthly report: April to October 2002. Sthapit, B.R., Subedi, A., Rijal, D., Rana, R. and Jarvis, D. 2003. Strengthening community-based, on-farm conservation of agricultural biodiversity. p. 344–353. In: The conservation and sustainable use of agricultural biodiversity: A sourcebook. Vol. 2: Strengthening local management of agricultural biodiversity. CIP-UPWARD, Laguna, Philippines. Yen, D.E. 1991a. Domestication: The lessons from New Guinea. p. 558–569. In: Pawley, A. (ed.) Man and a half. Polynesian Society, Auckland. Yen, D.E. 1991b. Polynesian cultigens and cultivars: The questions of origin. p. 67–95. In: Cox, P.A. and Banack, S.A. (eds). Islands, plants and Polynesians. Dioscorides Press, Portland, Oregon. Yen, D.E. 1993. The origins of subsistence agriculture in Oceania and the potential for future tropical food crops. Economic Botany 47:3–14. Yen, D.E. and Wheeler, J.M. 1968. Introduction of taro into the Pacific: The indications of chromosome numbers. Ethnology 7:259–267. 46 third taro symposium Theme One Abstracts Theme 1: Taro Diversity, Ethnobotany and Conservation Networking with taro: a review of TANSAO achievements V. Lebot, J. Quero Garcia and A. Ivancic The objective of this paper is to discuss the benefits of crop networking, especially as far as breeding is concerned, based on the lessons learnt from TANSAO. Isozymes and AFLP markers indicate that except for Indonesia, the genetic diversity of diploid taro cultivars is rather low. This means that crosses among accessions originating from any one country are not desirable. It is better to cross cultivars from the Asian and Pacific genepools. The international exchange of taro germplasm is therefore a prerequisite for a successful breeding programme. The question of how and what to exchange is discussed and it is recommended to exchange genes rather than genotypes. Furthermore, taro leaf blight is caused by numerous strains that are genetically variable within and between countries. These results also recommend that a long-term strategy of taro improvement has to be based on a population composed of carefully selected parents from diverse geographic origins. Taro is grown mostly for the quality of its corm and cormels and preliminary results show that all physico-chemical characteristics are variable. It is likely that cultivar selection will be efficient for improving these traits, essential for consumer acceptance and for processing both traditional and new products. The TANSAO core sample was distributed in vitro to participating countries and to SPC and is now being propagated in the field. Several thousands hybrids are being produced and it is now possible to exchange true seeds internationally. Theme 1 : Diversité, ethnobotanique et conservation du taro La constitution de réseaux d’échange : un examen des réussites du Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie (TANSAO) V. Lebot, J. Quero Garcia et A. Ivancic Cet exposé a pour objet de passer en revue les avantages des réseaux d’échange, notamment dans le domaine de l’amélioration des végétaux, et d’énoncer les leçons tirées de la création du Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie (TANSAO). Les analyses réalisées à l’aide de marqueurs izoenzymatiques et AFLP indiquent qu’à l’exception de l’Indonésie, la diversité génétique des cultivars diploïdes est relativement faible. Il en résulte que les croisements entre obtentions provenant d’un même pays ne sont pas souhaitables. Il est plus approprié de croiser des cultivars provenant des pools génétiques d’Asie et du Pacifique. Seul l’échange de matériel génétique à l’échelle internationale peut assurer le succès des programmes d’amélioration génétique. On s’interroge encore sur la nature du matériel à échanger et sur les modalités de transfert. Pour l’instant, il est recommandé d’échanger des gènes plutôt que des génotypes. En outre, la flétrissure des feuilles de taro est provoquée par de nombreuses souches génétiquement variables selon la région ou le pays. Ces données suggèrent également la nécessité de fonder une stratégie à long terme d’amélioration du taro sur une population composée de parents sélectionnés minutieusement et provenant de zones géographiques distinctes. On cultive principalement le taro pour la qualité de son corme et de ses cormelles. Or, certains résultats préliminaires ont mis en lumière la variabilité de ces caractéristiques physiques et chimiques. Il est probable que la sélection des cultivars permettra d’améliorer efficacement ces dernières, qui conditionnent l’accueil réservé par les consommateurs aux produits existants ou nouveaux ainsi que la transformation de ces produits. L’échantillon du TANSAO a été distribué in vitro à l’ensemble des pays participants et à la CPS, et il est en cours de multiplication en plein champ. Plusieurs milliers d’hybrides sont en cours de production, et il est maintenant possible d’échanger des semences vraies à l’international. Taro diversity in a village of Vanua Lava island (Vanuatu): Where, what, who, how and why? La diversité du taro dans un village de l’île de Vanua Lava (Vanuatu) : où, quoi, qui, comment et pourquoi ? Sophie Caillon and Virginie Lanouguère-Bruneau Sophie Caillon et Virginie Lanouguère-Bruneau Through a combination of agronomic and anthropological approaches, this study demonstrates that taro does not only play an alimentary role but also helps to underline the identity of the group that cultivates it. In Vêtuboso, the biggest village of Vanua Lava island (North Vanuatu), taros are planted in swamps, rivers, and irrigated pondfields. Farmers derive pride from their use of the sophisticated Cette étude, qui se veut à la fois agronomique et anthropologique, démontre que le taro ne joue pas seulement un rôle dans l’alimentation des populations, mais qu’il contribue aussi à la définition de l’identité des groupes qui le cultivent. À Vétuboso, principal village de l’île de Vanua Lava (Nord de Vanuatu), on plante les taros dans les marais, les rivières et les champs irrigués. Les cultivateurs sont fiers third taro symposium 47 traditional practices associated with taro cultivation in pondfields, where the alternation of dry and wet phases allows ecological pest control (Papuana spp), diminishes climatic variation, and adapts corm quality to the “nalot”, North Vanuatu’s traditional meal. On the western coast of Vanua Lava, farmers adopt distinct management strategies to plant 96 taro cultivars in various assortments. Through the description of both aerial and underground characteristics, the people of Vêtuboso are able to differentiate this high diversity of cultivars without synonymies and homonymies. Field surveys illustrate the variability of the taro portfolios that 12 households have adopted. Common cultivars that most farmers plant in high quantity have their specific uses in both every-day and ceremonial meals or have major agronomic qualities, whereas rare taros, unevenly distributed among farmers, are still planted because farmers are interested in experimenting with new cultivars, and in preserving their community and family heritage. de mettre en œuvre les pratiques traditionnelles ingénieuses associées à la culture du taro dans les champs irrigués où la succession des périodes sèches et humides permet de lutter de manière écologique contre les organismes nuisibles (Papuana spp), de compenser les variations climatiques et d’obtenir la qualité de corme souhaitée pour le nalot, repas traditionnel du Nord de Vanuatu. Sur la côte Ouest de Vanua Lava, les cultivateurs emploient d’autres méthodes de gestion en plantant 96 cultivars selon des assortiments divers. Grâce à la description des caractéristiques aériennes et souterraines des cultivars, les habitants de Vétuboso peuvent faire des distinctions entre les variétés très diverses sans utiliser de synonymes ni d’homonymes. Les études de terrain ont mis en évidence la variabilité des plants de taro cultivés par douze familles. Les cultivars communs cultivés en grande quantité par la plupart des exploitants sont destinés à une utilisation spécifique lors des repas quotidiens et cérémoniels, ou bien présentent d’importantes qualités agronomiques. En revanche, les cultivars de taro plus rares répartis de façon inégale chez les exploitants, sont encore cultivés car ces derniers cherchent à expérimenter de nouveaux cultivars et à préserver leur patrimoine collectif et familial. Applications of DNA markers to management of taro (Colocasia esculenta (L.) Schott) genetic resources in the Pacific Island region Applications de marqueurs d’ADN à la gestion des ressources génétiques du taro (Colocasia esculenta (L.) Schott) en Océanie I.D. Godwin, E.S. Mace, P.N. Mathur and L. Izquierdo I.D. Godwin, E.S. Mace, P.N. Mathur et L. Izquierdo Microsatellite markers have been developed and applied to rationalise the TaroGen taro germplasm collection. National collections were DNA fingerprinted where possible, although for the larger Melanesian collections (PNG, Solomon Islands, Vanuatu and New Caledonia) a 20% sample of the national collection was fingerprinted. A core collection representing 10% of the total TaroGen collection was identified for in vitro conservation. Overall, the level of taro genetic diversity was found to be highest in Melanesia, particularly in PNG and the Solomon Islands. Greatest genetic diversity was identified in accessions from the Solomon Islands, and indeed 2 rare microsatellite alleles were found only in accessions from the Solomon Islands. High levels of genetic similarity were observed in Polynesian accessions, as previously reported. Des marqueurs d’ADN microsatellites ont été mis au point et appliqués à la gestion de la banque de matériel génétique du projet TaroGen (Ressources génétiques du taro : conservation et utilisation). Lorsque cela était possible on a pris l’empreinte génétique des collections nationales mais, pour les collections mélanésiennes plus vastes (Papouasie-Nouvelle-Guinée, Îles Salomon, Vanuatu et Nouvelle-Calédonie), seul un échantillon représentant 20 % de la collection nationale a fait l’objet d’un relevé d’empreinte. Une collection noyau représentant 10 % du matériel contenu dans la banque du projet TaroGen a été identifiée pour une conservation in vitro. Dans l’ensemble, cet exercice a révélé que le niveau le plus élevé de diversité génétique du taro se trouvait en Mélanésie, notamment en Papouasie-Nouvelle-Guinée, et aux Îles Salomon. Les obtentions provenant des Îles Salomon présentaient la plus grande diversité génétique, et d’ailleurs, deux allèles microsatellites rares ont été détectés uniquement dans des obtentions provenant des Îles Salomon. Comme cela a déjà été noté, de hauts niveaux de similarités génétiques ont été observés dans les obtentions polynésiennes. 48 third taro symposium Using in vitro techniques for the conservation and utilization of Colocasia esculenta var. esculenta (taro) in a regional genebank Mary Taylor, Valerie Tuia, Rajnesh Sant. Eliki Lesione, Raghani Prasad, Rohini Lata Prasad and Ana Vosaki The fragmented nature of the Pacific Island region lends itself to regional strategies, and conservation is one such activity where regional policies and practices would seem to be the best option. If individual countries do not have sufficient resources for germplasm conservation, then it is best carried out on a regional basis. This becomes even more logical when one considers the commonality of the major crops: taro, yam, banana, sweet potato and cassava. As the majority of Pacific Island crops are vegetatively propagated, they lend themselves to conservation in field genebanks. However, the success of field genebanks in the Pacific has been limited, with climatic extremes, pest and diseases and insufficient resources causing problems. In vitro conservation can also be used for vegetatively propagated crops and this has proved to be a more secure and cost-effective method for Pacific crops. In addition, with a region composed of many islands, each with their own strict quarantine regulations, in vitro methodology can facilitate distribution, through providing access to pathogen-tested germplasm. Under the regional Taro Genetic Resources: Conservation and Utilization project (TaroGen), over 2,000 taro accessions were collected. Therefore a secure and sustainable method for conserving the selected accessions (core collections and taro leaf blight resistant breeding lines) had to be identified, which at the same time allowed dissemination of the material within the region. To meet these requirements, which would be applicable to most Pacific crops, the Secretariat of the Pacific Community (SPC) established a Regional Germplasm Centre (RGC), which was officially opened in September 1999. 650 taro accessions are currently being maintained in the RGC. 374 accessions have been selected for active distribution. These consist of core collections from both the TaroGen and Taro Network for Southeast Asia and Oceania (TANSAO) projects, and also taro leaf blight resistant cultivars and breeding lines from TaroGen. These accessions have been meristem cultured and are currently being pathogen tested. Once their pathogen free status has been verified, these accessions will be multiplied, providing sufficient numbers to interested countries for evaluation and utilization. Accessions will be prioritized for distribution, and maintained either at a temperature of 25°C to encourage active growth, or at 20°C to reduce growth (slow growth storage). Cryopreservation methodologies are being investigated so that accessions not immediately selected for active distribution can be conserved on a longterm basis. L’application de techniques in vitro à la conservation et à l’utilisation de Colocasia esculenta var. esculenta (taro) dans une banque de gènes régionale Mary Taylor, Valerie Tuia, Rajnesh Sant. Eliki Lesione, Raghani Prasad, Rohini Lata Prasad et Ana Vosaki La région océanienne, constituée d’îles parfois très éloignées les unes des autres, se prête bien aux stratégies régionales. La conservation des ressources appelle aussi l’adoption de politiques et de pratiques régionales. Lorsqu’un État ou Territoire ne dispose pas des ressources suffisantes pour assurer la préservation de ses propres ressources génétiques, il est logique de le faire à l’échelle régionale, d’autant plus que la majeure partie des États et Territoires cultivent les mêmes végétaux : taro, igname, banane, patate douce et manioc. La majorité de ces végétaux se reproduisent par multiplication végétative et se prêtent donc parfaitement à la conservation en banques de gènes en champ. Néanmoins, ce type de banques n’a pas rencontré le succès escompté dans le Pacifique. À cela, plusieurs raisons : des conditions climatiques extrêmes, diverses incursions d’organismes nuisibles et le manque de ressources. On peut aussi conserver les végétaux à reproduction végétative in vitro. En réalité, c’est la méthode la plus fiable et la plus rentable de préservation du matériel génétique des végétaux cultivés dans le Pacifique. Elle peut aussi faciliter la distribution en donnant accès aux nombreux États et Territoires insulaires de la région, dotés chacun d’une réglementation phytosanitaire stricte, à un matériel génétique exempt de tout agent pathogène. Dans le cadre du projet régional Ressources génétiques du taro : préservation et utilisation («TaroGen»), plus de 2 000 obtentions de taro ont été rassemblées. Il a ensuite fallu élaborer une méthode sûre et viable de conservation des obtentions sélectionnées (collections «noyaux» et lignées reproductrices résistantes à la flétrissure des feuilles de taro), ce qui à l’époque a permis de distribuer ce matériel aux États et Territoires de la région. Conscient de la nécessité de conserver et de diffuser le matériel génétique du taro, mais aussi de la plupart des végétaux cultivés dans le Pacifique, le Secrétariat général de la Communauté du Pacifique a créé le Centre régional du matériel génétique, inauguré en septembre 1999. Le Centre détient à l’heure actuelle 650 obtentions de taro. Sur ce total, 374 ont été choisies pour être distribuées de manière active. Elles sont issues de collections « noyaux » constituées à la fois dans le cadre du projet TaroGen et du Réseau TANSAO (Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie). Certaines sont des cultivars et des lignées reproductrices résistants à la flétrissure des feuilles de taro élaborées dans le cadre du projet TaroGen. Ces obtentions ont été produites par culture de méristèmes et sont actuellement soumises à des tests pour détecter la présence d’éventuels pathogènes. À l’issue de ces tests, elles seront multipliées afin que tous les États et Territoires intéressés puissent les recevoir en quantités suffisantes pour les évaluer et, à terme, les utiliser. Les obtentions seront destinées en priorité à la distribution et maintenues soit à une température de 25°C pour favoriser une croissance active soit à 20°C pour ralentir la croissance (stockage en croissance ralentie). Plusieurs méthodes de conservation par cryogénisation sont actuellement à l’étude de manière à pouvoir conserver durablement les obtentions qui ne seront pas immédiatement sélectionnées pour la distribution. third taro symposium 49 Promoting on farm conservation of taro through diversity fairs in the Solomon Islands Roselyn Kabu Maemouri and Tony Jansen This paper is about the experiences of two diversity fairs that were held to promote on farm conservation of taro (Colocasia esculenta) in Solomon Islands. The diversity fairs were held at the site of taro field genebanks in Malaita and Temotu provinces established under TaroGen, a regional project implemented by the Secretariat of the Pacific Community (SPC) in collaboration with national partners and funded by AusAID. (Some additional funding for the diversity fairs was provided by European Union Micro Projects Program in Solomon Islands.) Diversity fairs were organised by local farmers, with assistance from the Planting Materials Network (PMN) and agriculture department officers to coincide with the time that taros were ready to harvest from the farmer-run provincial field gene banks. The overall aims of the diversity fairs were to distribute taro planting materials back to the farmers in the province where the diversity had been collected and to raise awareness among farmers about taro conservation. In addition, extensive reference is made in this paper to information collected during participatory rural appraisals (PRA) that were carried out with groups of farmers during the process of collecting taros from farmers for the field genebanks and during discussions, both formal and informal, held before, during and after the diversity fairs. Home gardens and their role in the conservation of taro diversity in Vietnam Nguyen Thi Ngoc Hue, Luu Ngoc Trinh and Nguyen Van Minh Home gardens, known in Vietnamese as vuon nha, are patches of land of varying dimensions surrounding rural houses that are commonly planted with fruits, vegetables and root crops. This study was conducted primarily to describe the structure and compositions of home gardens, the activities of garden custodians, and the role home gardens may have in conserving taro genetic resources. Surveys of home gardens and interviews of garden custodians were carried out in four locations in the country. Participatory approaches were used to collect data. The home gardens in different zones generally differed in their size, structure, and the manner in which they were maintained. Taro was selected as a key home garden species because it is important in food security, and present in home gardens throughout Vietnam, as well as in the wider agro-ecosystem. Study results indicate that during the long history of taro cultivation, local people in Vietnam, especially women, have accumulated rich indigenous knowledge and experience in the use and management of taro genetic resources. Different varieties of taro are grown for different purposes and under different maintenance regimes. The fact that a number of varieties of taro were found in home gardens but were not present in larger fields 50 third taro symposium Promotion de la conservation du taro à la ferme au travers d’expositions sur la biodiversité aux Îles Salomon Roselyn Kabu Maemouri et Tony Jansen Cet exposé traite de deux expositions sur la biodiversité organisées aux Îles Salomon pour promouvoir la conservation du taro (Colocasia esculenta) à la ferme. Ces expositions se sont déroulées dans les provinces de Malaita et de Temotu, sur le site de banques de gènes du taro en champ créées dans le cadre du projet régional Ressources génétiques du taro : préservation et utilisation (TaroGen), mis en œuvre par le Secrétariat général de la Communauté du Pacifique en collaboration avec des partenaires nationaux et financé par l’AusAID. Des fonds complémentaires ont également été recueillis pour l’organisation des expositions, au titre du programme de micro-projets mené par l’Union européenne aux Îles Salomon. Les expositions étaient organisées par des exploitants locaux, avec l’assistance du réseau de végétaux destinés à la multiplication et d’agents du service de l’agriculture, à un moment où les taros étaient prêts à être récoltés dans les banques de gènes en champ provinciales, entretenues par les agriculteurs. Les expositions avaient pour buts la redistribution de matériels de multiplication du taro dans la province où la diversité avait été observée et la sensibilisation des agriculteurs aux méthodes de conservation du taro. D’autre part, l’exposé se réfère aux informations recueillies à l’occasion d’évaluations de systèmes agricoles selon des méthodes participatives, réalisées par des groupes d’agriculteurs dans le cadre de la collecte de plants destinés aux banques de gènes au champ ; des renseignements ont aussi été recueillis au cours de discussions, formelles ou non, avant, pendant et après les expositions. Le jardin potager et son rôle dans la conservation de la diversité génétique du taro au Vietnam Nguyen Thi Ngoc Hue, Luu Ngoc Trinh et Nguyen Van Minh Les jardins potagers, vuon nha en vietnamien, sont des lopins de terre de superficie variable situés à proximité des fermes et où l’on cultive souvent des fruits, des légumes et des légumes-racines. L’étude dont fait l’objet cet exposé visait à déterminer la structure et la composition des jardins potagers, les activités des cultivateurs qui les entretiennent et le rôle que sont susceptibles de jouer les jardins dans la conservation des ressources génétiques du taro. Des jardins potagers ont été étudiés et des cultivateurs interrogés dans quatre régions du pays. Les données ont été recueillies grâce à des méthodes participatives. D’après les résultats, la taille, la structure et le mode d’exploitation des jardins potagers diffèrent généralement d’une région à l’autre. Le taro a été choisi comme l’une des principales espèces potagères car il contribue à la sécurité alimentaire, et on le rencontre dans tous les potagers du Vietnam, ainsi que dans l’ensemble de l’écosystème agricole. D’après les résultats de l’étude, depuis l’époque lointaine où l’Homme a commencé à cultiver le taro au Vietnam, les habitants du pays, et notamment les femmes, ont accumulé un grand savoir et une riche expérience dans l’utilisation et la gestion des ressources génétiques du taro. Différentes variétés de and paddies suggests that home gardens are important sites in which to conserve the genetic diversity of taro. taro sont cultivées à différentes fins et selon différentes méthodes. La découverte dans les potagers de plusieurs variétés de taro absentes des tarodières de plus grande superficie tend à suggérer que les jardins potagers sont des lieux parfaitement adaptés à la conservation de la diversité génétique du taro. Diversity and genetic resources of taro in India Diversité et ressources génétiques du taro en Inde S. Edison, M.T. Sreekumari, Santha V. Pillai and M.N. Sheela India is blessed with great genetic diversity of taro (Colocasia esculenta (L.) Schott), which is cultivated throughout the country for of its corms, cormels and leaves. The Central Tuber Crops Research Institute, Trivandrum is the sole research institute in India engaged in the genetic improvement of tropical tuber crops, including taro. CTCRI maintains the largest germplasm collection of taro, with 424 accessions at its HQ at Trivandrum in the south and 120 at its Regional Centre at Bhubaneswar in east India. Genetic resources of taro are also collected and maintained by the National Bureau of Plant Genetic Resources (New Delhi) in its regional station at Trichur, Kerala, and by research centers and agricultural universities located in different agroclimatic zones under the All India Coordinated Research Project on Tuber Crops. At CTCRI, the accessions are maintained both in field genebanks and in vitro. Cytological and morphological characterization has been done and yield attributes have been identified. Two superior selections identified from the germplasm collections have been released (1987) by CTCRI for general cultivation under the names “Sree Reshmi” and “Sree Pallavi”. “Mukthakesi,” released in 2001 by the CTCRI Regional Centre, is tolerant to Colocasia leaf blight, a serious disease in certain parts of the country. Several economically desirable lines identified from the germplasm collections are under advanced stages of evaluation. Also, studies are in progress on the application of genetic and molecular markers for the characterization of germplasm to confirm the results obtained by systematic and morphoagronomic descriptors. Analysis of genetic diversity in taro in China D. Shen, D.W. Zhu, X.X. Li and J.P. Song Morphology, 5 isozymes, and AFLP and RAPD markers were used to analyze the genetic diversity of 28 taro accessions collected in Yunnan province, China. There were some differences in the patterns revealed by the different techniques, but significant levels of genetic diversity can still be found within the crop in Yunnan. In situ conservation in Yunnan should be considered as part of an overall strategy. S. Edison, M. T. Sreekumari, Santha V. Pillai et M. N. Sheela L’Inde est un immense réservoir de diversité génétique du taro (Colocasia esculenta (L.) Schott). Le taro est cultivé dans l’ensemble du pays pour ses cormes, ses cormelles et ses feuilles. L’Institut national de recherche sur les légumes-racines (CTCRI), situé à Trivandrum, est le seul institut d’Inde à oeuvrer pour l’amélioration génétique des légumes-racines des régions tropicales, et notamment du taro. L’Institut gère la plus riche collection de matériel génétique de taro avec 424 obtentions conservées au siège de l’institut à Trivandrum, dans le sud, et 120 stockées au Centre régional à Bhubaneswar, dans l’est du pays. Le Bureau national des ressources phytogénétiques (New Delhi) détient et conserve également des variétés de taro dans sa station de recherche régionale de Trichur (Kerala), tout comme les centres de recherche et les écoles d’agriculture situés dans différentes zones climatiques, dans le cadre du projet de recherche national coordonné sur les légumes-racines. L’Institut national de recherche sur les légumes-racines conserve ses obtentions à la fois dans une banque de gènes en champ et in vitro. Les caractéristiques cytologiques et morphologiques des stocks génétiques ont été décrites, et les attributs du rendement ont été identifiés. En 1987, le CTRCI a mis en circulation deux variétés sélectionnées dans le matériel génétique disponible afin qu’elles soient cultivées : ‘Sree Reshmi’ et ‘Sree Pallavi’. Une autre variété, ‘Mukthakesi’, mise en circulation par le Centre régional du CTCRI en 2001, est tolérante à la flétrissure des feuilles de Colocasia, un véritable fléau dans certaines régions de l’Inde. Plusieurs lignées identifiées au sein des collections de matériel génétique et présentant un fort potentiel économique font l’objet d’évaluations déjà bien avancées. En outre, des études sont actuellement en cours sur l’utilisation de marqueurs génétiques et moléculaires dans la caractérisation du matériel génétique et la confirmation des résultats obtenus à l’aide de descripteurs systématiques et morpho-agronomiques. Analyse de la diversité du matériel génétique du taro en Chine D. Shen, D.W. Zhu, X.X. Li et J.P. Song La diversité génétique de vingt-huit échantillons de matériel génétique de taro, collectés dans la province du Yunnan, en Chine, a été analysée par étude morphologique, à l’aide de cinq isoenzymes et de marqueurs isoenzymatiques AFLP et RAPD. Malgré certaines différences des caractéristiques mises en évidence selon les techniques, on constate une grande diversité génétique du taro cultivé dans la province du Yunnan. Une stratégie agricole globale devrait inclure la conservation in situ, dans cette province. third taro symposium 51 Theme One Paper 1.1 Networking with taro: a review of TANSAO achievements V. Lebot1, J. Quero Garcia2 and A. Ivancic3 1 CIRAD, Port Vila, Vanuatu CIRAD, TA 70/16, 34398 Montpellier, France cedex 5 3 University of Maribor, Vrbanska, 2000, Maribor, Slovenia 2 Introduction Taro (Colocasia esculenta) is an important food crop in many parts of the humid tropics. Germplasm collections have been made and lost several times in many countries (Jackson, 1994). Breeding programmes have been initiated in South East Asia and Oceania in order to provide farmers with improved cultivars but success has been somewhat limited. In the late 1980’s a geographical survey of the isozyme diversity existing in the South Pacific revealed that cultivars susceptible to leaf blight in Melanesia had a zymotype identical to the most popular cultivars in Polynesia, indicating that they would be affected if TLB were introduced (Lebot and Aradhya, 1991). Phytophthora colocasiae did indeed arrive in Western Samoa in 1993 and the consequences were severe. It therefore became clear that if taro breeding was going to have any future, it was important to exchange genetic resources in order to broaden the genetic base of the crop, which can be assessed using molecular markers. The idea of a taro network was born of this isozyme study, but it took several years before TANSAO, the Taro Network for South East Asia and Oceania, could be designed and implemented with the support of the European Union. TANSAO was established in 1998 with the overall objective to enhance the competitive position of taro in the rainfed cropping systems of SE Asia and Oceania. This would be achieved by improving quality and resistance to pests and diseases and by increasing the efficiency of production. Major constraints for taro breeding programmes are lack of knowledge of, and access to, the genetic diversity among cultivars, especially as regards disease resistance and the agronomic and processing value (Ivancic and Lebot, 2000). This situation calls for a regional and collaborative approach. However, exchanging taro germplasm can be dangerous since it can spread viruses, which severely decrease yield. Propagation via in vitro culture can produce pathogenfree cultivars, but a certification programme and strict quarantine is required to distribute safely this material on an international scale. Within TANSAO, the collaboration of scientists located in Indonesia, Malaysia, Papua New Guinea, Philippines, Thailand, Vanuatu, Vietnam, and Europe (France and The Netherlands) allows the development of a safe distribution system via a transit centre located outside the area of production. The specific objectives of the first phase of TANSAO (98-02) were: 1. 2. 3. 4. 5. to characterise approximately 2,000 accessions and analyze their genetic diversity with morpho agronomic traits and molecular markers (isozymes and AFLP); to distribute a core sample of selected cultivars to collaborating countries; to identify sources of resistance to TLB and to introduce them via targeted crosses; to assess the genetic diversity of P. colocasiae throughout the region using molecular markers; and to study the physico-chemical characteristics of the corms of selected genotypes. The aim of the present paper is to discuss the lessons learnt from TANSAO so that appropriate future developments may be designed. Materials and methods Descriptions were carried out of 2,298 accessions originating from 8 countries using 23 standardised morphological descriptors (IPGRI, 1999). Each trait was scored with qualitative data and national databases were developed. More than 2,080 accessions were characterised using six isozyme systems (Lebot et al., 2000; Prana et al., 2000) and zymotypes identified. Each country selected a core sample composed of only locally preferred cultivars representing approximately 10% of the total number of accessions. More than 200 selected genotypes were tissue cultured and indexed for DMV (dasheen mosaic virus) in Wageningen, the Netherlands, and five in vitro plantlets per genotype were sent to LIPI in Bogor, Indonesia, for multiplication and distribution to other countries. AFLP analysis was conducted on the core sample (Kreike et al., 2003; Lebot et al., 2003a). In Vanuatu, 378 cultivars were grown in a common plot. Their corms were boiled and submitted to a blind panel test composed of ten participants and their eating quality was scored (Bourrieau, 2000). The physico-chemical characteristics of 31 cultivars were analysed (Lebot et al., 2003a) and results were compared to consumers taste. 52 third taro symposium Overall, 94 isolates of P. colocasiae were collected on susceptible and tolerant cultivars. Protein extracts were analyzed with 8 isozyme systems. RAPD markers were used to study the variation among 20 isolates (Lebot et al., 2003b). Results 1. Morpho-agronomic description An extensive amount of morpho-agronomic variation was observed (Table 1). Taro germplasm is maintained ex situ in field collections, a system that presents a number of constraints. An attempt to rationalise the collections was conducted. Multivariate analyses of the data did not produce a useful picture. Many of the characters, although relevant for differentiating morphotypes (i.e. colours, shapes), were found to have limited informative value on the genetic structure of the collections. The stratification of the collections using a branching method based on six morphological characters, however, allowed the development of core subsets (Lebot et al., 2000). AFLP markers were used in Vanuatu to analyse the extent of genetic variation captured in each core subset and it was demonstrated that, although the collection has a narrow genetic base (Lebot and Aradhya, 1991), this branching approach assembled a significant amount of genetic diversity. AFLP fingerprints also demonstrated that no duplicates were found within core subsets (Quero Garcia, 2000). Table 1: Geographical distribution of major morpho-agronomic traits (percentages of accessions in participating countries) (Lebot et al., 2000) Country PH VNT H MY ID PG VU Total No. of accessions 172 350 300 135 685 278 378 2298 Dasheen 80.8 40.0 78.3 6.7 98.7 98.6 100 80.5 Eddoe 16.9 35.7 21.7 93.3 1.2 0.4 0 15.4 Flowering 46.5 14.0 6.8 13.4 19.2 100 41.0 31.8 Stolons 59.3 9.1 99.7 94.1 77.7 77.7 32.0 62.2 Early maturing (<8 months) 93.6 63.5 79.0 4.4 83.5 99.7 0 67.2 Very good taste 5.2 11.4 0.3 5.9 1.5 2.8 2.9 3.8 Excellent taste 1.2 2.3 0 0 0 1.8 4.0 1.3 Medium size (0.5-2 kg) 11.6 3.7 78.3 42.2 76.5 98.9 59.8 58.8 Unbranched shape 99.4 50.6 18.7 91.1 87.0 86.7 70.4 70.9 White flesh colour 47.7 96.0 3.7 63.0 39.7 16.9 42.1 43.2 2. Isozyme fingerprinting Considering the complexity of the zymograms, no genetic interpretation was attempted (Lebot et al., 2000; Prana et al., 2000). Indonesia, Malaysia, Thailand and Vietnam were found to host significant allelic diversity. In comparison, the countries located in the Pacific (the Philippines, Papua New Guinea and Vanuatu) appear to have limited allelic diversity (Table 2). Only six zymotypes represent more than half of the total accessions and 21 zymotypes represent more than the two thirds (70%) of the total number of accessions studies. The genetic base of taro in these seven countries is thus narrow. Although morphologically similar, wild taros assemble most of the allelic diversity revealed with isozymes. Table 2: Isozyme variation in South East Asia and Oceania (Lebot et al., 2003a) Country ID MY TH VN PH PG VU Total Accessions 688 57 322 210 198 452 154 2081 Zymotypes 194 30 64 74 10 51 8 319 % unique 72 23 50 53 40 39 0 Zymotype index1 0.28 0.52 0.20 0.35 0.05 0.11 0.05 1 0.15 No. of distinct zymotypes divided by no. of accessions studied Overall, 168 cultivars were selected for the core sample: 54 from Indonesia, 15 from Malaysia, 19 from the Philippines, 35 from Thailand, 29 from Vietnam and 16 from Vanuatu. It was decided not to include in the core sample cultivars originating from Papua New Guinea because of the presence of ABVC, which represents a serious constraint to international exchange. The number of cultivars per country was based on the total number of accessions and on an assessment of the genetic variation based on isozyme analysis (Lebot et al., 2000). third taro symposium 53 3. AFLP fingerprinting This study included 181 diploids and 36 triploids (Kreike et al., 2003). Three AFLP primer combinations generated a total of 465 scorable amplification products that were used for data analysis. The gene diversity within these groups was computed (Table 3). Except for Indonesia, the genetic diversity of the diploid cultivars within these countries is rather low. This confirmed the results obtained with isozymes showing that the wild forms present significant genetic diversity although the sample studied was small. Most of the dasheen cultivars are diploids, but some triploid dasheens were found in Thailand. Most of the eddoes are triploids, but a few diploids were found in the Philippines. Table 3: Number of accessions analyzed for AFLP and gene diversity within countries (Kreike et al., 2000). Country TH MY VN ID PH PG VU Diploid cultivars 33 13 3 50 31 35 16 Gene diversity 0.03 0.07 0.05 0.11 0.08 0.06 0.05 Triploid cultivars 2 - 24 10 - - - Gene diversity 0.007 0.13 0.14 Wild accessions 16 11 3 8 - - - Gene diversity 0.19 0.08 0.10 0.15 4. Physico-chemical characteristics of the corms Results of the physico-chemical analyses of 31 cultivars from Vanuatu are presented in Table 4. Except for the temperature of gelatinisation, all characteristics are very variable. Table 4: Physico-chemical characteristics of the corms of 31 selected cultivars (all as % DM, except the temperature of gelatinisation in °C) (Lebot et al., 2003a). DM STA AMY PROLIPMINGLUFRU Minimum 12.5 36.6 3.4 3.7 0.5 1.6 SAC MAL GEL 0.1 0.1 0.8 0.0 79.1 83.6 Maximum 55.9 77.9 12 15.8 1.5 6.6 2.7 2.6 8.7 0.2 Mean 27.9 65.5 8.2 6.5 0.7 3.4 0.6 0.6 3.4 0.1 80.8 Stand. Dev. 10.8 9.3 2.1 2.3 0.3 1.1 0.5 0.6 1.9 0.1 1.2 CV % 38.8 14.2 25.0 34.9 37.1 31.2 90.9 97.6 56.7 40.5 1.4 Coefficient of correlation with taste1 0.4* 0.4* 0.4* 0.09 -0.3 -0.5** -0.3 -0.4* -0.2 -0.02 -0.3 1 r value at 5% of significance is 0.355 (*) and 0.451 at 1% (**) DM=dry matter, STA=starch, AMY=amylase, PRO=proteins, LIP=lipids, MIN=minerals, GLU=glucose, FRU=fructose, SAC=saccharose, MAL=maltose, GEL= temperature of gelatinisation These traits are genetically controlled and cultivar selection will be efficient for improving them. Chemotypes are in agreement with indigenous knowledge, which claims that different cultivars have to be prepared in different ways in order to be palatable. The results obtained from the blind panel tasting were very consistent (Bourrieau, 2000): when a cultivar was considered “excellent”, this was true for all testers. In Vanuatu, good taste is associated with high dry matter, starch and amylose contents and low mineral and lipid contents (see coefficients of correlation in Table 4). 54 third taro symposium 5. Genetic diversity of P. colocasiae Accessions maintained in the national germplasm collections were scored for their resistance to TLB and significant variation was found for this trait within and between countries (Table 5). Table 5: Geographical distribution of the tolerance to leaf blight caused by P. colocasiae (percentages of accessions in countries surveyed) (Lebot et al., 2003b) Country PH VN TH MY ID PG VU Total No. of accessions 172 350 300 135 685 278 378 2298 Very susceptible 4.1 0 0 0 0.2 0 - 0.4 Susceptible 21.5 1.7 94.7 0 65.4 100 - 45.8 Tolerant 73.8 34.9 0.3 4.4 32.3 0 - 20.8 Resistant 0.6 41.1 5.0 43.7 1.5 0 - 10.0 Immune 0 22.3 0 51.9 0 0 - 6.4 Not determined - - - - 0.7 - 100 16.6 The variation studied in 94 isolates of P. colocasiae originating from Indonesia, Papua New Guinea, the Philippines, Thailand and Vietnam, using 8 isozyme systems revealed 52 zymotypes. A core sample of 20 isolates were analysed with RAPD markers and clear bands differentiated isolates exhibiting identical zymotypes. Results, presented in Table 6, indicate that taro leaf blight is caused by numerous and distinct strains that are genetically variable within and between countries. The geographical distribution of zymotypes shows that none is common to two countries. Although the differences in pathogenicity are not yet established, different P. colocasiae genotypes are likely to recombine and to evolve rapidly as this species is heterothallic. Table 6: P. colocasiae isolates studied for isozyme variation (Lebot et al., 2003b) Country ID VNPHTHPGTotal No. of isolates 21 5 15 45 8 94 No. of distinct zymotypes 3 2 9 33 3 40 No. of unique zymotypes 3 2 9 33 3 Discussion In most countries, farmers traditionally maintain a wide range of cultivars, including poor and excellent ones and, of course, they always cultivate more plants of the cultivar they consider as being the best for their personal uses and/or for marketing. This appreciation is based first of all on the taste and secondly on the agronomic performance of the cultivar. In Papua New Guinea, for example, cultivar Numkowec is popular because of its palatability, the same is true for Sakius in Vanuatu and for Bentul in Java. In all countries, not many cultivars are “very good” (Table 1) and few are considered as being “excellent”. The ones that are highly appreciated are widespread and some genotypes have been distributed over great distances. Depending on locations, up to 50 distinct cultivars are used. However, as soon as an opportunity occurs, exotic cultivars are introduced into the farmers’ portfolio and can become predominant if their quality is superior. Taro is essentially cultivated for the quality of the corms (although in some cases leaves, petioles, stolons and spathes are also consumed). Unfortunately, morphological variation is not correlated with the characteristics of the corm and it is difficult to screen germplasm for characteristics related to quality. Farmers use morphotypes as markers to trace the non-visible useful traits of taro, the chemotypes, which are highly variable (Table 4). Wild forms often present morphotypes similar to cultivars (Ivancic and Lebot, 1999). Considerable time and effort was invested in TANSAO to fully characterise the germplasm collections with standardised morphological descriptors (IPGRI, 1999). However, the return on investment was limited because of the poor genetic value of these characters. These coded descriptors might be of interest to curators but they are not to breeders because they are not related to the traits that need to be genetically improved (Ivancic and Lebot, 2000). Studies will have to be conducted to determine if some can be used as visual clues for screening seedlings at an early stage, i.e. to detect the pink flesh colour of the corm (Ivancic et al., 2003), a component of the ideotype in the Pacific. In most areas, traditional cultivars appear to share a common genetic background. It is likely that some of them have been diversified morphologically via vegetative propagation but it is also possible that the few that resulted from sexual pollinations were crosses between two clones (i.e. Sakius x Sakius), resulting in an increased level of homozygosity. We do not yet know what heterosis could do for taro. We do know, however, that most hybrids that have been developed in Hawaii, Samoa and Vanuatu during the last two decades are not significantly superior to their parents in growth vigour. third taro symposium 55 The use of two different markers was found useful. Isozymes are cheap, codominant markers that are useful to screen several thousand accessions. However, they tend to produce an under-estimation of the genetic diversity existing within taro collections. On the other hand, AFLP are dominant and powerful markers, but which cannot be applied to more than a few hundred accessions because of the work load involved. In the case of TANSAO, the combination of the two markers was found to be a cost efficient way of screening accurately large numbers of accessions in order to extract rationally a core sample. The cultivars representing the TANSAO core sample are the best genotypes existing in SE Asia and the Pacific. They have been tissue cultured in WAU, the Netherlands, propagated in vitro in LIPI, Indonesia, and distributed throughout the region. They are now established in field collections in each participating country and are used for direct distribution to farmers after vegetative propagation or for breeding purposes. A set of the core sample has also been given to the Secretariat of the Pacific Community (Suva, Fiji) for propagation and distribution to the South Pacific Islands countries. Results obtained with molecular markers demonstrate the existence of two distinct genepools. If further studies were conducted, it is possible that others might be identified (i.e. in India or China). Within each genepool, genetic distances between cultivars are narrow. This implies that crosses between accessions originating from only one country are not desirable. It is assumed that crosses between cultivars from neighboring countries (i.e. Vanuatu and Papua New Guinea) will also not create genetically diverse offspring. With the TANSAO core at their disposal, breeders in participating countries can now develop crosses between distant but complementary genotypes. Except from the Pacific, where cultivars appear to have been selected in a TLB-free environment, in most countries there are both susceptible and resistant cultivars (Table 5). It is difficult, however, to assess the relative importance of local variation in environmental factors, genetic variation within taro and variation among the different strains of P. colocasiae. The results obtained from the genetic diversity study of P. colocasiae suggest that a long-term breeding strategy has to be based on a wide genetic base composed of carefully selected parents from diverse geographic origins in order to concentrate multigenic resistance in progenies. This type of resistance is likely to result from crosses between tolerant genotypes that exhibit significantly distant molecular fingerprints and that originate from areas where TLB is induced by different strains. Breeding taro is fairly easy (Wilson, 1990; Ivancic and Lebot, 2000) and a single cross can produce several hundred seedlings. Very soon breeders are overwhelmed by thousands of hybrids to maintain, evaluate and select. The trick lies in the right choice of parents and of their characteristics. Taro was domesticated for its corm properties and should be genetically improved for these same properties: the problem lies in how to improve major physico-chemical traits such as dry matter, starch, amylose and other components that contribute to quality. There are several major constraints that must be overcome in order to fully utilise the potential of taro for processing. Taro corms do not present a uniform shape at harvest, thus making it difficult for mechanical peeling and marketing. Internal color of raw taro corms ranges from white, yellow, pink, to a combination of colors. The texture of corms varies after cooking, and some varieties are acrid. In most countries, accessions are well characterised and can be used within each country in crosses with varieties from the core sample. The crosses should focus on a few traits important in each country, taking a few of the exotic accessions available. National programmes can make exchanges with others, and this will have the advantage of broadening further the genetic base across the region. Exchanges can best be made of genes in the form of botanical seeds, as it is no longer necessary to exchange genotypes now that the core sample has been distributed. True taro seed (TTS) hold the key to rapid improvement of taro anywhere in the world. They can be generated in large quantities and present the advantage of maintaining genetic diversity, in contrast to the selection of a relatively small number of clones. Seeds are likely to facilitate international transfer of germplasm as they act as a filter for most viruses. The use of TTS and farmers’ early involvement in taro breeding can be used to exploit genotype-environment interaction through decentralised evaluation and selection. In some countries, taro cultivation is declining and the species is losing its competitive position in traditional cropping systems, being replaced by cassava, cocoyam and/or sweet potato (Singh et al., 2001). Diets are also evolving rapidly, and taro is being replaced by rice and other cereals. TANSAO emphasises collaboration between participants to target the rapid improvement of taro quality and resistance to pests and diseases in order to increase the efficiency of production. Unlike some other crop networks, TANSAO focuses on breeding rather than on conservation of the existing germplasm. The first phase of the network dealt with the necessary screening of the accessions. Now that this has been achieved, the exchange of genes (TTS) is the priority. Conclusion Germplasm collections should be downsized because they include numerous duplicates and/or closely related genotypes, they are difficult to maintain and often include accessions with poor breeding value. As controlled crosses can rapidly generate thousands of hybrid plants, priority should be given to the characterisation of collections and the selection of suitable parents. Stratifying is a relevant approach and should take into consideration a few major agronomical traits as well as chemotypes to compose core subsets from which parents can be chosen. Chemotypes represent the most important traits for consumers and growers, and should be the focus of genetic improvement. Aerial parts have limited informative value and less work should be devoted to the description of morphotypes. On the other hand, more investment should go to the chemical characterisation. Once the ideotype is defined and parents 56 third taro symposium identified, visual morphological clues should be used to rapidly screen the numerous seedlings emerging from breeding programmes. Heritabilities of DM, starch and amylose contents need to be studied. It is quite obvious that product development will be necessary for the continued cultivation of taro, but it will be achieved only if suitable chemotypes are developed. Populations composed of selected genotypes from diverse geographical origins exist and can now be used by breeders. Genes can be exchanged via TTS and this approach will contribute greatly to taro genetic improvement. Hopefully, farmers will obtain what they deserve in the near future. Acknowledgments TANSAO is a project funded by the INCO programme of the European Commission, Directorate DGXII. Thanks are due to T. Gendua, H. van Heck, H. Hue, N. Kreike, M.S. Prana, T. Okpul, J. Pardales, M. Thongjiem, N. Viet and T.C. Yap. References Bourrieau, M. 2000. Valorisation des racines et tubercules tropicaux pour l’alimentation humaine en Océanie: Le Cas du laplap au Vanuatu. ENSIA–SIARC Thèse de Master en Sciences, Génie Agro-alimentaire méditerranéen et tropical, Montpellier. 60 p. IPGRI. 1999. Descriptors for taro (Colocasia esculenta). International Plant Genetic Resources Institute, Rome, Italy. 56 p. Ivancic, A. and Lebot, V. 1999. Botany and genetics of New Caledonian wild taro, Colocasia esculenta. Pacific Science 53(3):273–285. Ivancic, A. and Lebot, V. 2000. Taro (Colocasia esculenta): Genetics and breeding. Collection ‘Repères’, CIRAD, Montpellier. 194 p. Ivancic, A., Quero Garcia, J. and Lebot, V. 2003. Development of visual tools for selecting qualitative corm characteristics of taro (Colocasia esculenta (L.) Schott). Australian Journal of Agricultural Research 54:581–587. Jackson, G.V.H. 1994. Taro and yam genetic resources in the Pacific and Asia: Report prepared for ACIAR and IPGRI. ANUTECH Pty Ltd, Canberra. 73 p. Kreike, N., van Eck, H. and Lebot, V. 2003. Genetic diversity in taro (Colocasia esculenta (L.) Schott) from South East Asia and Oceania. Theoretical and Applied Genetics. In press. Lebot, V. and Aradhya, M. 1991. Isozyme variation in taro (Colocasia esculenta (L.) Schott) from Asia and Oceania. Euphytica 56:55–66. Lebot, V. et al. 2000. Genetic variation in taro (Colocasia esculenta) in South East Asia and Oceania. p. 524–533. In: Nakatani, M. and Komaki, K. (eds). Proceedings of the Twelfth Symposium of the International Society for Tropical Root Crops: Potential of root crops for food and industrial resources. Tsukuba, Japan, 10–16 September 2000. ISTRC. Lebot, V., Gunua, T., Pardales, J.R., Prana, M.S., Thongjiem, M., Viet, N.V. and Yap, T.C. 2003a. Characterisation of taro (Colocasia esculenta (L.) Schott) genetic resources in Southeast Asia and Oceania. Genetic Resources and Crop Evolution. In press. Lebot, V., Herail, C., Gunua, T., Pardales, J.R., Prana, M.S., Thongjiem, M. and Viet, N.V. 2003b. Isozyme and RAPD variation in Phytophthora colocasiae Raciborski isolates from South East Asia and Oceania. Plant Pathology 52:303–313. Prana, M.S., Hartati, N.S., Prana, T.K. and Kuswara, T. 2000. Evaluation of genetic variation in taro (C. esculenta (L.) Schott.) collected from West Java using isozyme markers. Annales Bogoriensis 6(2):80–87. Quero Garcia, J. 2000. Étude de la structuration de la variabilité génétique du taro. Mémoire de DEA, Institut National Agronomique de Paris Grignon. 50 p. Singh, D., Okpul, T., Iramu, E., Wagih, M. and Sivan, P. 2001. Breeding taro for food security in PNG. p. 749–757. In: Bourke, R.M., Allen, M.G. and Salisbury, J.G. (eds). Food security for Papua New Guinea: Proceedings of the Papua New Guinea Food and Nutrition 2000 Conference, PNG University of Technology, Lae, 26–30 June 2000. ACIAR Proceedings No. 99. Wilson, J. 1990. Taro breeding. Agro-Facts, IRETA Publication No 3/89, USP Alafua Campus. 51 p. third taro symposium 57 Theme One Paper 1.2 Taro diversity in a village of Vanua Lava island (Vanuatu): Where, what, who, how and why? Sophie Caillon1 and Virginie Lanouguère-Bruneau2 1 IRD, Orleans, France 1, rue de l’église, 45 390 Ondreville sur Essonne, France ([email protected]) 2 Introduction Taro (Colocasia esculenta) is an essential staple food on the west coast of Vanua Lava, a northern island among the 80 islands that make up Vanuatu. As Barrau (1983:14) emphasizes – “aliments are not only good to eat, they are also ‘good to think’” – taro in Vanua Lava is not just as food but also reveals the identity of the group who cultivates it. As Haudricourt (1964:93) has written, this society could be described as a “taro civilisation”. Vanua Lava is the biggest island of the Banks groups (331 km²) and has a population of 2000 inhabitants divided into two language groups: the Vera’a, found in an area restricted around the village of Vatrata, and the Vurës, on the rest of the islands. As the dominant language, Vurës will be used here for local terms. Surveys were mostly done in Vêtuboso, the biggest village of the island, with 610 inhabitants in 2001 (Hess, pers. comm. 2001). A four-hour walk from the airport (19 km), the village lies among mountains at an elevation of about 150 m. The people of the west coast of Vanua Lava mainly plant taro, whereas in the other islands of the Banks group, yams (Dioscorea spp) and taros are mixed in gardens or cultivated as mono-crops. Moreover, Vanua Lava is today the only island to practise a cultivation system based on irrigation (Lanouguère-Bruneau, 1999). Its inhabitants are famous for their know-how of taro cultivation in pondfields, in which are grown a large number of taro cultivars. In this paper, we will focus on the reasons leading to the maintenance of such diversity. Vanua Lava, a mosaic of socially valued gardens Three water-based systems of taro cultivation may be recognized: irrigated pondfields (qêl, pl. qêlaqêl), in stony rivers (mat [river]) and in mud along rivers or swamps without drainage (mat boak ou boak [mud]). These three systems protect young taros from being damaged by papuana beetles (Papuana inermis and P. huebneri), which are common on dry land. These systems coexist in six large areas named rot covering 17.3 ha (area calculated by software by François Bonnot of CIRAD-CP from GPS data). The extent of these taro pondfields is great compared to other islands (only 1.7 ha on the west coast of Santo – Walter and Tzerikiantz, in press). Pondfields are composed of flat terraces whose shapes follow the contour lines and whose heights depend on the slope. They are partitioned by stone and mud walls (30 cm high ) into rectangular ponds named qêl (mean surface area 87 m2, n=57) that are themselves generally divided into two sub-ponds called tin (mean surface area 42 m2, n=119), separated by mud walls (10 cm high) and each cultivated by a family or person. The pondfields are put through alternate wet and dry phases, and taro is finally harvested on dry soil. The duration of the phases depend on the weather, soil quality, and taro properties but also farmer needs, as changing the rhythm of wet and dry can modify maturation rates. Generally, taros are harvested after one year of cultivation, but if needed the harvest can begin after eight months. A pondfield can be planted for dozens of years without yield decreasing, as the taro plant nutrients are mostly contained in the water (Claus, 1998). To restart a fallow pond, weeds are simply cleared, although, according to villagers, slash-and-burn was practised in previous generations. Thus, compared with rainfed cultivation, irrigated taro can be described as intensive and sustainable, as taros are growing quicker, the number of cultivation cycles on a same area is higher, fallows are reduced and burn-offs are not practised. Together, the three taro garden types, in pondfields, in rivers and in swamps, allow a regular harvest throughout the year of a great quantity of corms. However, though they may coexist in an area, they are not assigned the same social value. Taro pondfields: a man’s pride Taros harvested from pondfields are the pride of Vêtuboso’s men. In contrast, growing taro in rivers is easy, worthy of Wômôdô, the orphan that “does not know how to work”. Swamp cultivation is the least prestigious and no founder myths are associated with it. A man who cultivates a high number of terraces according to ancestral rules will be admired for his working capacity and his know-how. Indeed, no maintenance is needed in rivers and swamp gardens, whereas pondfields require constant 58 third taro symposium work based on knowledge transmitted by men. This knowledge covers such diverse activities as drain excavation and maintenance, wall building, irrigation and pest control, but also includes customs, bans and magic. Men can prove their superiority in pondfield management by comparing their ability to grow taro, which depends on the knowledge they acquire through the years. As an example, a kind of taro called sestan is known to be a good rivertaro but with poor yields in pondfields. When two men want to compare their knowledge, a competition is organised: only the one who knows the appropriate magical leaves will harvest the biggest corm. Knowledge is not revealed because it shapes the social status of men. Moreover, the best taros are harvested from the irrigated system, their corms being harder and tastier. But the quality depends also on how the farmer deals with the wet and dry alternations, in order to avoid deterioration caused by the papuana beetle and unwanted alterations in corm texture, such as soft inner and hard outer part (mötöltöl) or too hard inner section (te¾urnur). At the time of gatherings, a man will share the taros he has cultivated with other members of the community, who will thus be able to judge his know-how. Thus, pondfields are valued for the importance and the secrecy of the knowledge needed to grow the best quality taros in them. As irrigation is a speciality of two villages on the island, Vatrata and Vêtuboso, this cultivation system constitutes the pride of their inhabitants, especially men, the knowledge keepers in this regard. A diversity of cultivars and farmers On the west coast of Vanua Lava, 96 cultivars of taros have been characterised (agro-morphological description and origin stories), thanks to individual surveys made with questionnaires (68 rapid and 12 in-depth interviews) and to multiple informal conversations in Vêtuboso and Vatrata and their pondfields. In comparison, 5 Alocasia macrorrhiza names and 4 Xanthosoma sagittifolium types have been identified. In this paper, the term “cultivar” refers to a group of individuals presenting morphological and agronomical characteristics sufficiently remarkable for the interviewed farmer to assign them a name recognized by the community as a whole. We are not using the term “variety” or “race” because these cultivars are “a clonal replication of exceptional individuals” (Zohary, 1984). As in Gaua and Mota (Vienne, pers. comm.), the people of Vanua Lava have origin stories dealing with the context of “apparition” of each cultivar on the island. These stories were documented from four chiefs of Vêtuboso well known for their knowledge. New taros can be discovered in three different ways. The majority (47% of taro cultivars for which the story is known) have appeared after burning weeds in pondfields fallows from five to dozens of years old. Judging from the descriptions of the young shoot and the emergence conditions, these taros may have resulted from seeds arising from sexual reproduction (Caillon et al., 2003). Secondly, importing taros from another island has brought 38% of the cultivars. This important exchange of germplasm illustrates that the number of morphotypes owned by each inhabitant continues to be increased by travel. The last source of new cultivars is random somatic mutations. A new morphotype can be a chimera produced by colour changes from a known type (15%). When a farmer finds a new “variant” in a population, he monitors it to judge its qualities when mature, evaluates its potential additional value in his taro portfolio, to finally decide to keep or reject it. In Vêtuboso, this event is so rare that a farmer will always conserve the new type. In general, the name attributed to a cultivar corresponds to its origin story. If, by clearing a new terrace, the farmer finds a taro morphologically distinct from the already known and named ones, he can baptise his discovery with his own name preceded by qiat min (taro of) or wot min (born under). These prefixes, however, are slowly erased by the passage of the years. Taro names also identify places where the plant appeared (name of a river or hill) or originated from. If imported from the Banks, the original vernacular name will be conserved. In other cases, the taro will be named according to the island of origin. Chimera will take the name of the mother plant they are derived from corrected by an adjective reflecting changes in colours. Lamkör or mal is added when the stem is darker, qagqag in the opposite case. When the corm is striped (fibers) with a colour different from the parenchyma, gatgat is added, but if it is only spotted, lörlör is used. Conscious synonymy is possible for taros recently brought from the same island; they may have the same eponym of the island, though morphologically distinguishable in the field. However, there is no synonymy among the 96 Vanua Lava and Banks taro cultivars collected and described in this study (Caillon and Malau, 2002). Genetic fingerprints constructed with AFLP markers show that local identification criteria are sufficiently precise to avoid homonymy; that is, two distinct but morphologically similar cultivars will not be grouped under the same name (Caillon et al., 2003). Farmers’ strategies to manage diversity To understand the diversity managed at a village scale, 56 pondfields managed by 12 farmers were visited to count the number of plants of each cultivar and to determine areas planted. We concentrate on pondfields, noting that in three rivers in which 20 cultivars were inventoried, a single one, recently introduced, took up 45% of the total area planted with taro. The twelve farmers were chosen to give a range of ages, histories, families and characters. On average, each is planting a mean of 20 cultivars and 957 plants (Table 1). In these 56 pondfields we found growing 51 of the 96 cultivars known in the village. However, only six, called “common” (in descending order, rov, marêwasalav, lantar, wasanto, vinmötöl and wêvê) represented 83% of the taros planted, and at least 5% each of the portfolio of each farmer. At the opposite end, 40 cultivars, said to be “rare”, represented less than 1% of taros owned by each farmer. This disequilibrium between varieties in a space organised by third taro symposium 59 humans can be compared to the distribution of individuals among wild species in nature, whereby communities tend to be composed of a small number of common species and numerous rare species (Krebs, 1994). Whereas most villagers own the same common taros, rare cultivars are distributed more heterogeneously, favouring exclusivity and the valuing of difference. Table 1: Taro portfolio of the 12 farmers: taro cultivation area, number of plants, numbers of cultivars, conservation effectiveness (number of cultivar x 100 / number of plants) and varietal diversity (Shannon-Wiener information statistic) Area: m2 (% of total) Number of plants (% of total) Number of cultivars Effectiveness H (per farmer) Eric 7.3 (15.0%) 1448 (16.8%) 20 1.4 1.98 Hervé 1.1 (2.3%) 192 (2.2%) 9 4.7 1.58 Arthur 3.8 (7.8%) 805 (9.3%) 27 3.4 2.28 Bertrand 0.2 (0.4%) 34 (0.4%) 6 17.6 1.54 12.4 (25.5%) 2479 (28.8%) 27 1.1 2.21 Pascal 2.7 (5.5%) 481 (5.6%) 14 2.9 2.21 Wendy 0.8 (1.6%) 81 (0.9%) 9 11.1 2.08 Achille 3.4 (7.0%) 312 (3.6%) 22 7.1 2.25 Henri 22.1 (45.4%) 2652 (30.8%) 46 1.7 2.20 Luc 0.5 (1.0%) 163 (1.9%) 13 8.0 1.58 Martin 3.4 (7.1%) 1718 (19.9%) 10 0.6 1.42 Anatole 1.1 (2.0%) 259 (3.0%) 15 5.8 2.04 Quentin 2.3 (4.7%) 465 (5.4%) 22 4.7 2.30 Total 48.7 (100%) 8610 (100%) 51 Mean 6.0 956.7 19.8 4.8 2.0 Eric’s family1 Eric’s family includes Eric, the father, and his children Arthur, Hervé and Bertrand. The married children, Pascal and Wendy, are counted apart as they have their own family. 1 To hierarchically classify the qualitative value of the twelve cultivar portfolios, two diversity indices were used: the number of cultivars per farmer (cultivar richness) and the Shannon-Wiener index1 (Krebs, 1994). The latter, based on both the number of cultivars and their relative abundance, allows us to respond to the question “With how much difficulty could we correctly predict the cultivar name of the next individual collected?” The higher the uncertainty, the greater the diversity. This function was calculated for the portfolio of each farmer (i.e. the calculation is based on the proportion of the ith cultivar in the gardens of one farmer and not in those of all the interviewed farmers). Thus, this index gives information on how each farmer is managing his own portfolio, disregarding other members of the community. According to Table 1, Hervé, Wendy and Martin plant the smaller number of cultivars (we will not discuss Bertrand’s case as he is only 8 years old). Lack of time seems to be the cause of such drastic selection. Hervé, a 18 year old single man, prefers easy and quick growing species to liberate his time for sporting activities. Wendy, married in another village, only rarely visits her parents in Vêtuboso, and Martin is a big producer of copra (whose plantation is a 6 km walk away) and of kava that he sells in a bar he has opened in the provincial administrative centre, Sola. Henri is not only the greatest taro planter but also has an incomparable enthusiasm for conservation, with his 46 cultivars. He is a real “collector”, never throwing out a cultivar even if it does not correspond to his agronomic or gustative expectations. However, as illustrated by the medium rank of his Shannon-Wiener index (H), he is planting few plants per rare cultivar, with 71,3% of his space planted with three cultivars, marêwasalav, rov and lantar. His conservation effectiveness, that is the number of cultivars divided by the total number of plants, is weak compared to the others. At the opposite end, Quentin, a single young man, is only planting 22 cultivars, rigorously selected, in more homogeneous proportions (highest H) and with a medium efficiency. Having fewer cultivars to manage, he has more time and space to multiply every cultivar. He is a “selector”. These two extremes lead us to define what might constitute a “good conservator”. In much the same way as an ex situ collection, the “collector” maximises the number of cultivars in a restricted space. In contrast, the “selector” limits the risk of losing material and so genetic diversity in case of bad weather or social conflict, by having more plants of each. The collector and selector strategies are thus complementary; the “good conservator” will be the one who conserves a great number of cultivars in numerous copies. The goal is to maximise and equilibrate the two diversity indices. In our case study, the adolescent Arthur, with 27 cultivars and an H index of 2.28, is the best in situ diversity curator. In spite of the rich portfolio planted by households, the majority of space is taken up by only a small proportion of common taros, whereas the heterogeneous distribution of rare taros forms the foundations of taro diversity. This diversity is differently managed by farmers: somewhere between pure “collectors” and pure “selectors”, good “conservators” are identified with respect to two distinct diversity indexes, the number of cultivars and the Shannon-Wiener index. 60 third taro symposium Why such a diversity? We can understand the reasons that lead farmers to select a small group of common taros (rov, marêwasalav, lantar, wasanto, vinmötöl and wêvê) by studying their agronomic performance and their response to culinary preparations. To understand why farmers conserve marginal and rare cultivars we will need to go beyond the alimentary value of taro, and discuss its role as the bearer of a social identity. Efficient taros Villagers are much more sensitive to the agronomic performance of their cultivars than they generally admit. When interviewed, villagers prefer to talk about the organoleptic and the social value of each cultivar, rather than about growing characteristics, a topic judged disrespectful to the taro. But among the six common taros, five are selected for their agrosystem adaptation, for their precocity or for their yield. Rov, the most famous taro on the island, represents 24% of all the taros in pondfields. It is the cultivar that is best adapted to alternate irrigation. Marêwasalav (19%) is selected for its long corm which grows in six months instead of twelve. Vinmötöl (8%), wêvê (6%) and wasanto (45% of the taros in rivers) are well-known for their big corm. Taros used in daily food Taro is the staple food of the people of Vêtuboso (1.9 kg/person/day, based on the amount eaten by a household of 7 people in 16 days). Corms are most frequently boiled in water accompanied by island greens (Abelmoschus manihot). Taros can also be cooked in coconut milk (wôrqarqar) or mixed with grated coconut albumen (bigtöw). All the cultivars can be boiled; the hard ones tend to be preferred in spite of a longer cooking time (the dry matter content being higher). However, the elderly without teeth will choose soft corms. If over-boiled, taros are doughy and are described by the term ¼êditdat. While hunting wild pigs in the forest or working in taro paddy fields, people roast taros (tun) or cook them in bamboo (bônësnës). Nine dry taros (mamas) – not too hard, not too soft – have been selected for their roasting qualities as quick to cook and tender to eat (mako, marêwasalav, marêwasalav mamê, qiatgôl, siritimiat, titiritowetam, tortor, vinmötöl and wederebiliag). More occasionally, to mark particular days, such as Saturday morning gatherings, a visitor’s arrival or departure or a family feast, taros will be prepared in nalot or grated in laplap. The nalot (löt) that the Banks islanders are famous for is a pudding principally prepared with taros. Tubers are firstly baked or boiled, than peeled, before being mashed on a flat ovoid wooden dish (tabê) with a hard wooden stick (vötulöt) especially devoted to this preparation. When tubers are transformed into a compact and elastic dough, this is flattened with a piece of coconut shell. The nalot is then cut with special carved knife (meteges) previously reserved to socially distinguished men. A good nalot is an elastic pudding made with hard taros said to be ta¾a¾al (high in dry matter content), such as rov and lôkreg. To facilitate the working of the dough, these cultivars are mixed with softer ones like lantar. Rov is preferred to lôkreg because of its hard texture; it also has a smell and a flavour that please the lovers of nalot. Taros cultivated in pondfields are preferred for the preparation of nalot not only because their texture is hardest but also because of the social value of that agricultural system. The laplap, another dish typical of Vanuatu, is prepared by grating raw tubers or fruits in laplap leaves (Helicona indica) and baking them in a stone oven. In Vanua Lava, the taro laplap does not have the same attractiveness as in the rest of Vanuatu. To avoid the irritation caused by calcium oxalate particles while grating corms on tree fern (Cyathea spp), soft taros or mölumlum (high in water content) are used for laplap. Ten cultivars (lantar malgias, lantar lamkör, relenman, sarê, sesta¾, suwbê, wakata mamê, wamal, wasanto, wasanto mamê, wêvê and wotliev) fulfil these conditions, in particular if they are planted in rivers or simple flooding, as they are very watery. Thus, in Vêtuboso a great diversity of dishes is prepared with only two ingredients, coconut and taro. Nuances in the preparation are due to side dishes – fresh coconut milk, made into a white cream or in red grains, nangae (Canarium spp), or bush nut (Barringtonia edulis) grated dry or green – and to differences in the quality of the taro paste depending on the choice of cultivar. Taros for celebrations, ceremonies and magic The presence of taro, either baked or as nalot is essential for all the ceremonies or great gatherings. When a family wants to celebrate an important event, the friends and relatives who are invited bring raw or cooked corms to contribute. However, guests cannot offer all cultivars. For instance, marêwasalav is a taro whose corm reaches in six months the size that other cultivars need ten months to reach. Nevertheless, the farmer needs to wait until ten months for the corm to have a “real” taro taste and not a “yam taste”. If a guest brings a marêwasalav taro to a celebration, the host will suspect the corm to be immature. Being very easy to cultivate, marêwasalav will certainly be supplied in large quantities by the host, because it becomes soft when baked. Thus, guests have to bring either more rare or better (i.e. harder) taros to diversify gustatory pleasures. During those celebrations, some nalot are more valuable than others. Thus, the one named lötnemere¾, made with dry nangae, is a must. Other kinds of nalot, chosen according to taste and family habits, can also be added. In past times, when men were still distinguished in grade-hierarchies, two cultivars, named tortor and mako, were reserved for men. Men isolated in the men’s house, who wanted to take the first grade in the Soq-hierarchy, roasted third taro symposium 61 these cultivars. When the isolation time ended, the initiated man would prepare a nalot on a small size nalot dish, with a stick smaller than the normal size. Ingesting the nalot would then mark the grade-taking. Which nalot was involved is controversial. Present custom chiefs say it was a lötnemere¾ nalot, made with boiled taros and dry nangae. On the other hand, an older man who has lived through the period when this institution still existed, says that a wageretow nalot made with roasted taros and dry bush nuts was used for such an occasion. In the first case, the nalot is elastic, prepared with hard taros like rov; in the second case, the choice of soft taros holding a small amount of water gives a soft consistency looked for by elderly people with no teeth. It can be underlined that cultivars used in the wagaretow nalot are all ancient taros noted in custom stories either because they were brought by a mythical hero (Wômôdô or Biliag), or because they are considered by everyone in the village as the first taros of the island. Some cultivars, now forgotten by a large part of the village’s population, as they are not used anymore, are said to have magical power. In ancient times, a man bewitched with black magic would have to eat a raw tuber of a taro named dogon, for his family to know his chances of survival. Only if he ate it without throwing it up would he survive. Thus, social life needs a constant supply of taros and nalot adapted to every circumstance. Some cultivars are preferred for celebrations, ceremonials and magic for their taste and texture, but also for the social value they are granted. A heritage of taros Of the 96 cultivars, 78.4%, even if comparatively less represented, are the mainstay of taro diversity of the village and of the island. It is thus essential to understand why farmers conserve these cultivars that, according to what they say, are not particularly efficient or pleasant to eat. Rare taros selected by the twelve farmers were introduced from another island (25%), were found in a fallow pond (25%), are linked to a founder myth (22.5%), or are marked by a noteworthy morphological or agronomic characteristic (12.5%). In fact if a man, or more rarely a woman, gives his name to a new taro that he or she has discovered in a fallow pond, his descendants will conserve it as part of their heritage. In Vanuatu, there are not so much property rights but usufruct rights. An individual owns what he plants and not the soil that nourishes the crops. The new taro holds the seal of its discoverer. The farmer will plant it, multiply it and distribute it with attention as his “invention”, as the range of its dispersion will be the measure of his renown while alive and after his death. Finally, farmers like to preserve cultivars whose names appear in founder myths (ta¾evsos, bulalef, burmatan, qiatgôl and qiatqet) or other “custom” stories (¼ôvôl, rêgêt, wasê, wemenriver and wederebiliag). The taro becomes an illustration, in the absence of writings, of a story that a father will tell with pride to his children. The taro, through its imagery, is thus a travel souvenir, a parental inheritance or an illustration in a story. For all these reasons are the rather less agronomically or culinarily favoured cultivars still maintained, though by fewer people and in smaller numbers. Conclusion This “civilisation of taro” has created an incredible diversity of agrarian landscapes (pondfields, managed rivers and swamps) and cultivars. The 96 cultivars grown on the west coast of Vanua Lava have been selected for their agronomic performance and their adaptation to culinary requirements, but also for the social value given by the story of their origin. As in Ambae, Maewo (Bonnemaison, 1974a, 1974b), Pentecost (Walter, pers. comm., 2003) and Tanna (Bonnemaison, 1987), taros are described according to their magical, customary or culinary uses, and social heritage. The taro field, the plant and its transformation are inextricably bound together for the people of the west coast of Vanua Lava. The hardest taros, needed for the nalot, are particular cultivars cultivated in pondfields under alternate wet and dry phases. At the opposite end, soft cultivars, which are used grated in laplap, are grown in continual inundation. We thus need to have an integrated vision of taro in the social and ecological environment, the “terroir”. The richness of the taro heritage of these villagers lies not just in their diversity of cultivars but in the associated diversity of landscapes, practices, uses and beliefs. Notes 1. The Shannon-Wiener index (H) is calculated according to the formula: from i=1 to N (number of cultivars), H=- Σ (pi)(ln(pi)) where pi is the proportion of the ith cultivar in the ponds of each farmer. References Barrau, J. 1983. Les Hommes et leurs aliments: Esquisse d’une histoire écologique et ethnologique de l’alimentation humaine. Messidor/Temps Actuels, Paris. Bonnemaison, J. 1974a. Espaces et paysages agraires dans le nord des Nouvelles-Hébrides: L’Exemple des îles Aoba et Maewo (étude de géographie agraire). Journal de la Société des Océanistes 44:163–232. Bonnemaison, J. 1974b. Espaces et paysages agraires dans le nord des Nouvelles-Hébrides: L’Exemple des îles Aoba et Maewo (étude de géographie agraire). Journal de la Société des Océanistes 45:259–281. Bonnemaison, J. 1987. Tanna: Les Hommes lieux. ORSTOM, Paris. 680 p. 62 third taro symposium Caillon, S. and Malau, E.F. 2002. Coconuts and taros from the West Coast of Vanua Lava (Vanuatu): An ethnoagronomic inventory. Port Vila, Vanuatu. 53 p. Caillon, S., Quero Garcia, J., and Lebot, V. 2003. Taro (Colocasia esculenta) diversity in a village of Vanuatu: A multidisciplinary approach. Third Taro Symposium poster. CIRAD, Nadi, Fiji. Claus, J.-C. 1998. Les tarodières irriguées de Futuna. CNEARC, ESAT, Montpellier, Territoire des Iles Wallis et Futuna. 169 p. Haudricourt, A.G. 1964. Nature et culture dans la civilisation de l’Igname: L’Origine des clones et des clans. L’Homme 4(1):93–104. Krebs, C. 1994. Ecology: The experimental analysis of distribution and abundance. HarperCollins College Publishers, New York. 801 p. Lanouguère-Bruneau, V. 1999. Les tarodières irriguées de l’île Vanua Lava: Une marque identitaire dans le système social inter-îles (îles Banks-Vanuatu). JATBA 41:61–91. Walter, A. and Tzerikiantz, F. In press. La tarodière irriguée: Un système d’agriculture diversifié. Zohary, D. 1984. Modes of evolution in plants under domestication. p. 579–586. In: W.F. Grant (ed.) Plant biosystematics. Academic Press, New York. third taro symposium 63 Theme One Paper 1.3 Applications of DNA markers to management of taro (Colocasia esculenta (L.) Schott) genetic resources in the Pacific Island region I.D. Godwin1, E.S. Mace1*, P.N. Mathur2 and L. Izquierdo1# School of Land and Food Sciences, The University of Queensland, Brisbane, Australia 2 IPGRI South Asia Office, New Delhi, India *Current address: Hermitage Research Station, 604 Yangan Road, Warwick 4370, Australia # Current address: Centre for Plant Conservation Genetics, Southern Cross University, Lismore 2480, Australia 1 Introduction Taro, Colocasia esculenta (L.) Schott, is one of the most important staple foods of Pacific Island countries, where it plays an important role both as a root crop and as a leafy vegetable. The genetic diversity of the crop has been characterised to date largely by morphological and cytological variation (Yen and Wheeler, 1968; Kuruvilla and Singh, 1981; Tanimoto and Matsumoto, 1986; Coates et al., 1988), and it has been observed that Polynesian cultivars are highly morphologically variable in contrast to the phenotypic homogeneity of the wild populations of Melanesia. It is thought that the high level of phenotypic variation is due to a high rate of vegetative propagation and, consequently, of somatic mutations. This would suggest that the majority of the cultivars in Polynesia are clones of a common source, and a recent study using isozymes (Lebot and Aradhya, 1991) indicated that there was very little genetic variation between the Polynesian cultivars, in contrast to the Melanesian and Asian cultivars. The results from a molecular study of taro genetic diversity, using RAPDs (Irwin et al., 1998), confirmed that although the cultivars in the Pacific region exhibit remarkable morphological variation, the genetic base appears to be very narrow. Such a limited genetic base leaves the crop very vulnerable to disease epidemics, such as the taro leaf blight outbreak in Samoa in the early 1990s, and insect damage. Consequently, germplasm collections from around the region have been undertaken to augment existing national collections and to safe-guard threatened and useful germplasm for use in regional breeding programmes. Some 1500 accessions are currently recognised by the Taro Genetic Resources Network (TaroGen), which aims to establish a regional genebank with a core collection, representative of the genetic diversity found within all the national collections. Figure 1: 527 accessions received for DNA fingerprinting 64 third taro symposium To date, studies directed at the identification of redundant germplasm in the Pacific Island national collections have utilised biogeographic, agronomic and phenotypic characterisation. Increasingly, the characterisation of germplasm collections also utilises molecular techniques e.g. Hokanson et al., 1998; Teulat et al., 2000; van Treuren et al., 2001. The emergence of PCR-based markers, such as Simple Sequence Repeats (SSRs), Amplified Fragment Length Polymorphism (AFLPs) and Random Amplified Polymorphic DNA (RAPDs) offers the opportunity for more fine-scale genetic characterisation of germplasm collections than previously possible, due to their high levels of polymorphism, their occurrence throughout the genome, their ease of detection and the additional advantage that many of the complications of environmental effects acting upon characters is avoided by looking directly at variation controlled at the genetic level. Of the molecular techniques available, SSRs are fast emerging as the marker of choice for many plant breeding applications, due particularly to their co-dominant nature, transferability, reproducibility and amenability to high through-put. Microsatellite markers have previously been isolated from taro (Mace and Godwin, 2002) and a set of polymorphic markers identified through screening with a limited range of genotypes from the Pacific Island region. Here, we report on the use of seven polymorphic microsatellite markers to evaluate genetic diversity and subsequently rationalise ten national collections from the Pacific Island countries (Figure 1). Marker-assisted rationalisation of taro genetic resource collections and the establishment of a regional core collection are discussed. Materials and methods 1. Plant material and DNA extraction The national taro collections included in this study comprise 1623 accessions (see Table 1), and were collected as part of the Pacific Island Country TaroGen network. From the entire collection, 527 accessions were fingerprinted (28% overall). The entire national collections of the Polynesian countries, Fiji and Palau were included in the fingerprinting study, with the small discrepancies between the collection size and the fingerprinting subset being due to either samples being destroyed in transit to the University of Queensland, Australia or samples being missing or too small at the time of collection. The larger national collections of Papua New Guinea, Vanuatu and New Caledonia were first analysed using passport and morphological data (Mace et al., 2004), to select 20% of the most diverse accessions within each group to be further analysed using molecular markers. The Solomon Islands collection had to be re-collected following the loss of the original collection during the political instability in 2000. Twenty per cent of the total collection was randomly sampled and sent to UQ for fingerprinting. From each of the 511 accessions, 50 mg of leaf material were collected and immediately frozen in liquid nitrogen. DNA was extracted using DNeasy® 96 Plant Kit (QIAGEN). DNA was eluted in 2x50 μL sterile distilled water and stored at 4°C. DNA concentration was measured both on a fluorometer (Hoefer TKP 100) following the manufacturer’s instructions, and by agarose gel (0.8%) electrophoresis. 2. SSR-PCR and electrophoresis Seven SSR primers were selected for use (Table 1), based on preliminary assays of amplification and product length polymorphism in taro genotypes (Mace and Godwin, 2002). Methodologies used have been previously described (Godwin et al., 2001; Mace and Godwin, 2002). Table 1: List of SSR primers used; repeat motif, oligonucleotide primer sequences, PCR annealing temperatures, expected PCR product size, number of alleles, and Polymorphism Information Content (PIC) scores. SSR ID SSR Primer sequence (5’-3’) Annealing temp. (°C) Allele size range No. alleles Diversity uq84-207 (CT)18 Fwd: aggacaaaatagcatcagcac Rvs: cccattggagagatagagagac 65.0 197-217 7 0.449 uq110-283 (TGA)6 (TGGA)4 Fwd: agccacgacactcaactatc Rvs: gcccagtatatcttgcatctcc 66.0 250-287 8 0.297 uq73-164 (CT)15 Fwd: atgccaatggaggatggcag Rvs: cgtctagcttaggacaacatgc 66.0 146-164 6 0.489 uq55-112 (CAC)5 Fwd: cttttgtgacatttgtggagc Rvs: caataatggtggtggaagtgg 65.0 112-136 3 0.089 uq88B-94 (CAT)9 Fwd: cacacatacccacatacacg Rvs: ccaggctctaatgatgatgatg 62.0 94-108 6 0.465 uq97-256 (CA)8 Fwd: gtaatctattcaaccccccttc Rvs: tcaaccttctccatcagtcc 66.0 248-256 5 0.332 uq91-262 (TG)6(GA)4 Fwd: gtccagtgtagagaaaaaccag Rvs: cacaaccaaacatacggaaac 65.0 258-262 3 0.267 third taro symposium 65 Data analysis Banding patterns observed at a particular locus were recorded as a presence/absence matrix. Similarity matrices were calculated from these data based on different measures; Nei and Li’s (1979) definition of similarity: Sij = 2a/(2a + b + c), where Sij is the similarity between two individuals, i and j, a is the number of bands present in both i and j, b is the number of bands present in i and absent in j, and c is the number of bands present in j and absent in i; this is also known as the Dice coefficient (1945); Jaccard’s coefficient (Jaccard, 1908): Sij = a / a + b + c; the simple matching (SM) coefficient (Sokal and Michener, 1958): Sij = a + d / a + b + c + d, where d is the number of bands absent in both i and j. Cluster analyses were performed on the similarity matrices using the unweighted pair group method with arithmetic averages (UPGMA) and dendrograms constructed from these analyses. Cophenetic correlation values were calculated to evaluate the robustness of the resulting tree topologies. All analyses were conducted using the NTSYS-pc software, version 2.02i (Rohlf, 1999). Results and discussion In total, 38 alleles were amplified from the seven SSR loci across the 511 taro genotypes included in this study. No locus was monomorphic across the entire collection, however two alleles (5% of total) were found to be monomorphic, and 36 (95%) were found to be polymorphic. Table 1 lists the total number of alleles per locus, the allele size ranges and the locus diversity values obtained. An average of 5.4 alleles per locus was observed and the gene diversity values ranged from 0.089 for uq55-112 to 0.489 for uq73-164. The SSRs were informative in revealing genetic differences within and among the different countries. Table 2 details the genetic variation observed within each country; the Solomon Islands collection revealed the highest proportion of polymorphic loci (1.0) and the highest average number of alleles per locus (5.3). In contrast, the lowest average number of alleles per locus was 3.86, observed in the collections from Palau, the Cook Islands and Tonga. The Solomon Islands collection also accounted for the highest proportion of the total number of alleles observed (0.974). Allele frequency variation across all loci and across countries clearly revealed a number of rare alleles (frequency ≤ 0.05) present in the germplasm collections, e.g. uq84-217, uq110-279, uq-110-281, uq88B-102, uq97-246 and uq91-262, two of which were found only in accessions from the Solomon Islands. Table 2: Genetic variation within the ten country collections analysed across 7 SSR loci. Country na Pb Ac PAd 11 0.86 3.86 0.711 Papua New Guinea 163 0.86 4.86 0.895 Solomon Islands 99 1.00 5.30 0.974 Vanuatu 89 0.86 4.14 0.763 New Caledonia 18 0.86 4.43 0.816 Fiji 71 0.86 4.29 0.789 Samoa 26 0.86 4.43 0.816 Cook Islands 15 0.86 3.86 0.711 Niue 24 0.86 4 0.737 Tonga 12 0.86 3.86 0.711 Palau a: Sample size b: Proportion polymorphic loci c: Average alleles per locus d: Proportion of total number of alleles observed Cluster analyses (UPGMA) were performed using the similarity matrices with the highest correlation coefficient (Jaccards’s similarity coefficient for all countries excepting PNG and Tonga, for which the Simple Matching coefficient revealed the highest value) based on the proportion of shared alleles across the 7 SSR loci. The cluster analyses were carried out on SSR data sets for individual countries and additionally on a combined data set, across all countries, in order to ensure that between country duplicates were not included in the final core set. There was a significant level of duplication existing within and between accessions collected from Polynesian countries for example (Figure 2). For the collections from PNG, Vanuatu, New Caledonia and the Solomon Islands, fifty percent of the total number of accessions fingerprinted were selected for inclusion in the suggested final core. For the remaining six countries (Fiji, Palau, Niue, Tonga, Cook Islands, Samoa), 10% overall of the total number fingerprinted were selected for inclusion in the suggested final core. In both cases, this was achieved by subdividing the dendrogram into sub-clusters, and selecting one or more accessions from each cluster, based on the level of diversity within each country and also the cluster analysis of the entire data set. Care was also taken to select the accessions containing the rare alleles identified. We have successfully selected a core collection from the Pacific Island countries involved in the TaroGen Network. This core collection is currently being conserved in vitro at the Regional Germplasm Centre at SPC in Suva, and is a resource for the region, subject to accessions being indexed for virus contamination. Further work is required to assess the level of genetic diversity which has been captured using this method, and there are questions to resolve as to the complementary nature of DNA fingerprint and agro-morphological data (as discussed elsewhere by Okpul et al., 2004 unpublished). There is also a need to study the genetic integrity of germplasm which has been conserved in vitro. 66 third taro symposium Figure 2: Dendrograms showing duplication within and among Polynesian taro accessions References Coates, D.J., Yen, D.E. and Gaffey, P.M. 1988. Chromosome variation in taro, Colocasia esculenta: Implications for its origin in the Pacific. Cytologia 53:551–560. Dice, L.R. 1945. Measures of the amount of ecologic association between species. Ecology 26:297–302. Godwin, I.D., Mace, E.S. and Nurzuhairawaty, N. 2001. Genotyping Pacific Island taro (Colocasia esculenta (L.) Schott) germplasm. p. 109–128. In: Henry, R.J. (ed.) Plant genotyping: The DNA fingerprinting of plants, CABI International, Wallingford, England. Hokanson, S.C., Szewc-McFadden, A.K., Lamboy, W.F. and McFerson, J.R. 1998. Microsatellite (SSR) markers reveal genetic identities, genetic diversity and relationships in a Malus x domestica borkh. core subset collection. Theoretical Applied Genetics 97:671–683. Irwin, S.V., Kaufusi, P., Banks, K., de la Peña, R. and Cho, J.J. 1998. Molecular characterisation of taro (Colocasia esculenta) using RAPD markers. Euphytica 99:183–189. Jaccard, P. 1908. Nouvelles recherches sur la distribution florale. Bulletin de la Société Vaudoise de Sciences Naturelles 44:223–270. third taro symposium 67 Kuruvilla, K.M. and Singh, A. 1981. Karyotypic and electrophoretic studies on taro and its origins. Euphytica 30:405– 412. Lebot, V. and Aradhya, K.M. 1991. Isozyme variation in taro (Colocasia esculenta (L.) Schott.) in Asia and Oceania. Euphytica 56:55–66. Mace, E.S. and Godwin, I.D. 2002. Development and characterisation of polymorphic microsatellite markers in taro, Colocasia esculenta (L.) Schott. Genome 45:823–832. Mace, E.S., Mathur, P. N., Godwin, I.D., Hunter, D., Taylor, M.B., Singh, D., DeLacy, I.H. and Jackson, G.V.H. 2004. Development of regional core collection (Oceania) for taro, Colocasia esculenta (L.), based on morphological and phenotypic characterization. In: Eyzaguirre, P.B., Ramanatha Rao, V. and Matthews, P. (eds). The global diversity of taro: Ethnobotany and conservation. IPGRI, Rome, Italy and MINPAKU (National Museum of Ethnology) Osaka, Japan. Nei, M. and Li, W.-H. 1979. Mathematical model for studying genetic variation in terms of restriction endonucleases. Proceedings of the National Academy of Sciences 76(10):5269–5273. Okpul, T., Mace, E.S., Godwin, I.D. Singh, D., and Wagih, M.E. 2004. Evaluation of variability between promising breeding lines and cultivars of taro (Colocasia esculenta (L.) Schott.) using Inter-Simple Sequence Repeat (ISSR) and agro-morphological characterization. Submitted. Rohlf, F.J. 1999. NTSYS-pc: Numerical taxonomy and multivariate analysis system. Version 2.02i. Exeter Software, New York. Sokal, R.R. and Michener, C.D. 1958. A statistical method for evaluating systematic relationships. University of Kansas Science Bulletin 38:1409–1438. Tanimoto, T. and Matsumoto, T. 1986. Variations of morphological characters and isozyme patterns in Japanese cultivars of Colocasia esculenta Schott and C. gigantea Hook. Japan Journal of Breeding 36:100–111. Teulat, B., Aldam, C., Trehin, R., Lebrun, P., Barker, J.H.A., Arnold, G.M., Karp, A., Baudouin, L. and Rognon, F. 2000. An analysis of genetic diversity in coconut (Cocos nucifera) populations from across the geographic range using sequence-tagged microsatellites (SSRs) and AFLPs. Theoretical Applied Genetics 100:764–771. van Treuren, R., van Soest, L.J.M. and van Hintum, T.J.L. 2001. Marker-assisted rationalisation of genetic resources collections: A case study in flax using AFLPs. Theoretical Applied Genetics 103:144–152. Yen, D.E. and Wheeler, J.M. 1968. Induction of taro into the Pacific: The indications of chromosome numbers. Ethnology 7:259–267. 68 third taro symposium Theme One Paper 1.4 Using in vitro techniques for the conservation and utilization of Colocasia esculenta var. esculenta (taro) in a regional genebank Mary Taylor, Valerie Tuia, Rajnesh Sant, Eliki Lesione, Raghani Prasad, Rohini Lata Prasad and Ana Vosaki Regional Germplasm Centre, Secretariat of the Pacific Community 1. Regional Germplasm Centre and taro conservation The use of tissue culture for conservation of some of the Pacific Island crops began in the mid-1980s when assistance was provided by international agencies for collecting root crops. By 1987, collections existed in many of the countries, and efforts were made to characterize and evaluate them. There was an obvious need to share this collected germplasm, but in a region composed of island countries, each with its own quarantine regulations, only distribution of pathogentested germplasm could be considered safe. Tissue culture satisfied the requirements for safe distribution. Consequently, in vitro laboratories were established within regional institutes, and pathogen-testing schemes developed. In addition, there was a perceived need for conservation of Pacific Island crops not in the mandate of any of the International Agricultural Research Centres (IARCs). For example, none of the IARCs has an international mandate to conserve taro. Similarly, with yams, the International Institute for Tropical Agriculture (IITA), Nigeria, maintains collections of the African yams, but has no Pacific yams in its collection. A regional tissue culture laboratory was established at the then South Pacific Commission (SPC) in Suva, Fiji in the mid 1980s, and in the late 1980s a regional tissue culture laboratory was established at the University of the South Pacific (USP), Samoa, funded by the European Union. These two regional laboratories were useful in demonstrating the important role tissue culture can play in conservation and utilization of plant genetic resources. Tissue culture provides methodologies which can safely maintain crop germplasm for varying periods of time, and can optimize propagation rates, thereby providing greater numbers of plants for farmers to access. Pathogen-tested germplasm facilitates safe distribution of plant material. At the same time as understanding the role tissue culture can play in any PGR management strategy, the need to adopt a regional approach to PGR conservation was gaining consensus. At a meeting of the ministers of agriculture of six ACP (African, Caribbean, and Pacific group of States) countries held in Fiji in 1997, the following resolution was endorsed: Conserving genetic diversity is the key to crop performance, and thus its neglect could imperil agriculture. Linked to this is the need to protect and utilize plant genetic resources, so that there is equitable sharing of benefits. The honorable ministers of agriculture are urged to put in place, both in their countries and through regional cooperation, policies to conserve, protect, and best utilize their plant genetic resources. If individual countries do not have sufficient resources for adequate germplasm conservation, then it is best carried out on a regional basis. A regional, cooperative approach becomes even more logical in light of the commonality of the major crops of the region. In response, SPC’s Regional Germplasm Centre (RGC) was officially opened in September 1999, supported through projects funded by the Australian Agency for International Development (AusAID) Australian Centre for International Agricultural Research (ACIAR), and the European Union (EU). 2. Taro conservation strategy within the RGC Although the conservation of genetic resources through a regional centre allows pooling of resources, there is still a need for cost-effectiveness. Simply maintaining large collections in vitro under standard conditions would strain resources. Under TaroGen, some 2418 accessions were collected; from these, a core collection of approximately 190 accessions was identified. This core collection will be maintained under slow growth conditions. The countries participating in TaroGen have expressed their support for the continuation of cryopreservation research. Optimizing existing cryopreservation protocols to achieve improved recovery rates across a wider range of cultivars would enable the non-core cultivars also to be conserved. TaroGen has also generated improved lines from the breeding programmes in Papua New Guinea and Samoa. These, together with cultivars from Federated States of Micronesia and the Philippines with some resistance to taro leaf blight, and the core sample from the TANSAO collections, have to be available for active distribution, and so will be maintained in active growth. 2.1Slow growth storage 2.1.1Introduction In vitro conservation methodologies offer an opportunity to preserve taro within a controlled environment, thereby eliminating risks from pest and disease outbreaks, and climatic extremes. Although various methods exist for reducing third taro symposium 69 the growth rate of plants in vitro, the most widely applied slow growth storage technique is temperature reduction, often combined with a decrease in light intensity or culture in the dark There are reports in the literature of taro being stored for more than eight years at 9°C in total darkness, with transfer intervals of approximately three years (Bessembinder et al., 1993). Similarly, Staritsky et al. (1986) reported that taro (Colocasia esculenta) could be conserved for three years at 9°C, and still be viable. Research carried out in the regional tissue culture laboratory at USP, Samoa, demonstrated that temperature reduction was the most practical method for slowing down the growth rate of taro. Taro could be maintained at 20°C, for 9 to 12 months, without subculturing, depending on the variety. Other parameters, besides reduced temperature, were also investigated. These were reduced light and supplementing the culture medium with osmoticums. The inclusion of mannitol in the culture medium did suppress growth, but some morphological changes in the resulting plantlets were observed. In addition, when mannitol was used with cultures initiated directly from the field, a phytotoxic effect was observed. As part of the TaroGen project, a pilot study was carried out for an in vitro genebank using temperature reduction to inhibit growth rate. The overall objective was to assess and demonstrate the technical and logistical aspects of establishing and operating an in vitro active genebank (IVAG), using taro as a model. The specific objectives were: • To select a sample of Fijian taro accessions, and to process these samples into in vitro storage under conditions of slow growth. • To provide a cost analysis for in vitro maintenance of taro (costs were also calculated for a field genebank) • To determine the needs for laboratory facilities, equipment, consumable items, and technical staffing involved throughout the operation of the IVAG. 2.1.2Methodology 44 accessions were selected from the Fiji national field genebank maintained at Koronivia Research Station. All accessions were multiplied in vitro to obtain five replicates per accession. Shoot-tips of approximately 1cm in size were excised from the cultures, after the multiplication cycle. Prior to their excision, all cultures were grown on basal medium, without any growth regulators for one month, to reduce any possible carry-over of growth regulators into the IVAG. In the IVAG, taro cultures were grown in 100 ml glass jars (Cospak), containing 20 mls of Murashige and Skoog basal medium (1962), supplemented with 3% sucrose, benzylaminopurine (1.0 mg/l) and napthaleneacetic acid (0.3 mg/l). Cultures were maintained at a temperature of 20°C under a daylength of 16 h and a light intensity of 40 µmolm-2 s-1. Some guidelines were necessary to determine when the survival of an accession in the IVAG was under threat. Viability was defined as cultures that had either (a) not grown from initiation, or (b) had outgrown the culture container and the culture medium, and were starting to senesce, or (c) were affected by any of the viability factors (shoot tip necrosis, stunting, contamination, and senescence). If three or more cultures were not considered viable for any of the reasons stated, that accession would be replaced with five new replicates, either generated from cultures in the IVAG, or from cultures maintained outside of the IVAG. The accessions in the genebank were characterized after six months of storage. The parameters selected for this process were: sucker number, callus formation, rooting, hyperhydricity, stunting, leaf shape. The cost of maintaining the IVAG was determined using spreadsheets devised by an ACIAR funded project, “Economics of preserving genetic diversity in Papua New Guniea” specifically for the cost analysis of the SPC-RGC. These sheets cover all costs for in vitro conservation and distribution of taro at SPC. A summary table provides an annual estimate of variable costs, medium term variable costs, fixed costs and total costs for maintaining the in vitro taro collection at SPC, and allows users to examine the cost budget without having to view the entire spreadsheet. Costs are estimated for the whole taro collection, per accession, and per plant replicate. A simpler system was also devised so that all inputs in the IVAG were recorded at the time of carrying out the activity. This shows what resources (labour, equipment, consumables) are required for the different operations, and also confirmed the cost analysis from the spreadsheets. It also gives an indication of immediate costs, without taking into account fixed costs. 2.1.3Results and discussion When an in vitro genebank is established, one technique has to be selected and applied to all the different genotypes under investigation. Genotype can significantly affect response to the selected slow growth methodology. In the study on a cassava IVAG, the International Centre for Tropical Agriculture (CIAT) found that of 48 varieties, 50% needed subculturing after one year of storage, 18 after 15 months, and six after eight to nine months (IBPGR/CIAT, 1994). In this study, 70% of the taro cultivars required subculturing after six months; the remainder could be maintained in culture for a further three months. Those accessions requiring earlier subculturing were generally more vigorous, and so the culture medium had been depleted, leading to nutrient deficiency. In addition, with older cultures, defoliation and senescence were a more common occurrence, increasing the chance of fungal contamination from rotting leaves in the culture vessel. No changes were observed in shoot number, callus formation, rooting, hyperhydricity, stunting and leaf shape to indicate a problem with genetic stability. 70 third taro symposium Endogenous bacterial contamination created some problems in the IVAG. The use of antibiotics is not a viable option for cultures in a slow growth storage system, because of the problems of resistance, of the possible encouragement of genetic change, and of the added cost. The recommended strategy for addressing endogenous contamination is to screen all plant material, prior to initiation into tissue culture. A microbial detection medium can be used, and planting material testing positive rejected. The costs for maintaining one accession for one year were calculated at US$49. This is a relatively high figure, but a close study of the costs involved show that the fixed costs are the major contributor to the total, including also opportunity costs. Variable and medium variable costs only account for 16.7% of the total cost, and labour accounts for 28.3% of the total. The cost of in vitro storage from the CIAT IVAG study was US$26.22 per cassava accession, and this lower cost probably reflects economies of scale, because of CIAT’s size of operation. A full report on this study is available in the Taro Conservation Strategy Report, September 2001 (AusAID/SPC Taro Genetic Resources Conservation and Utilization). The major recommendations from the study were as follows: • Slow growth storage has to be considered with other conservation methodologies, as part of a complementary conservation strategy. • The length of the subculture period is influenced by genotype, and this has to be accounted for in the allocation of resources. • Accessions should be screened for bacterial contamination prior to being introduced into in vitro storage. • A full cost analysis should be carried out to ensure that the resources are available to sustain all activities. • Good documentation is important, but depends to a large extent on the size of the collection and the resources available. It can consists of simple cards or a computerized system. 2.2Cryopreservation Cryopreservation has been recognized as a practical and efficient technique for long-term storage of vegetatively propagated plants, requiring minimum space and relatively low costs. In addition, several studies have shown that genetic integrity is maintained after recovery from cryopreservation. (Mannonen et al., 1990; DeVerno et al., 1999; Cote et al., 2000). The availability of a cryopreservation protocol would facilitate the conservation of the non-core accessions from the TaroGen project. Vitrification is a cryopreservation method which has worked well with tropical species. The plant cells are osmotically dehydrated in a highly concentrated vitrification solution, which enables direct immersion in liquid nitrogen. Vitrification is achieved through using a glycerol-based, low toxicity solution (PVS2), which sufficiently dehydrates cytosols, without causing injury. As a result, they are converted into a stable glass when plunged into liquid nitrogen. The method used with taro was developed in Japan (Takagi et al., 1997), initially for Colocasia esculenta var. antiquorum, but was later used with var. esculenta. The method was tested with three different cultivars (TNS, CPUK and E399) from the RGC. Of the different stages in the protocol, modifications in the preculture, preconditioning and dehydration stages are most likely to have an impact on success. Optimizing exposure to PVS2 (30% (w/v) glycerol + 15% (w/v) ethylene glycol + 15% (w/v) DMSO + 0.4 M sucrose in MS) is an important step for cryopreservation by vitrification, as plant tissues can suffer from phytotoxicity if doses are too high. Experiments were carried out to determine the optimum conditions for all of these stages. Different conditioning treatments were found to be optimal for the three cultivars, while the same vitrification protocol was equally successful for all. For two cultivars (E399 and CPUK), the optimum conditioning treatment was preculturing shoot tips from three month old in vitro plants on solidified Murashige and Skoog basal medium (MS) with 0.3 M sucrose in the dark for 16 hr at 25oC. For the third cultivar, the optimum treatment was preconditioning donor plants on solid MS, supplemented with 90 g/l sucrose for seven weeks. Shoot tips from these plants were excised and directly cryopreserved without any preculture. The optimum vitrification protocol was loading the shoot tips with a solution of 2 M glycerol plus 0.4 M sucrose for 20 min at 25oC. This was followed by dehydrating with PVS2 for 12 min at 25oC prior to quick immersion in liquid nitrogen. Thawing was done by rapidly shaking the shoot tips in a water bath at 40oC for 1min 50 sec followed by rehydrating in liquid MS with 1.2 M sucrose for 15 min at 25oC then plating on recovery medium. Shoot tips resumed growth within a week and developed into plantlets six to eight weeks later without any callus formation. Average recovery rates for the three cultivars were 21, 29 and 30%. However, up to 71, 85 and 100% success rates were achieved in individual trials. Experiments were carried out to investigate whether an encapsulation dehydration methodology would improve recovery rates from cryopreservation. In addition, a newly developed protocol, which is a combination of the two methods, vitrification and encapsulation/dehydration, was also tested. With these methodologies, however, no shoot tips recovered. Research is continuing with the vitrification methodology to see if it can be extended to other cultivars, and if the rates of recovery can be optimized. A full report on the vitrification methodology is available (“Cryopreservation of in vitro grown shoot tips of tropical taro, Colocasia esculenta var. esculenta, by vitrification” by Rajnesh Sant, Mary Taylor and Anand Tyagi). third taro symposium 71 3. Utilization Utilization of the taro accessions in the RGC by Pacific Island countries requires that all accessions are pathogen tested, and then rapid multiplication of selected accessions. The virus indexing procedures have been developed by the Queensland University of Technology, through funding from ACIAR. All accessions have been meristem cultured (0.81.0 mm) in the RGC. The procedure used for the virus indexing process has been to test the leaf tissue (in vitro), where possible for TaBV, DsMV, TaRV, and TVCV, and the whole plant again for these viruses and CBDV. Only suckers derived in vitro from an original meristem, which has tested negative for viruses, will be considered as negative for that viruses. These suckers will provide the source from which all distributed material will be obtained. The RGC is using a taro multiplication system developed in the tissue culture laboratory at USP, Samoa from two Masters’ projects (Palupe, 1997; Tuia, 1997). This system has been evaluated against other multiplication protocols, and to date has proved optimum. The protocol, based on a Murashige and Skoog basal medium supplemented with 3% sucrose and 8g/l agar, (MS) consists of three stages: Stage 1: MS + 0.5 mg/l TDZ Stage 2: MS + 0.8 mg/l BAP Stage 3: MS + 0.005 mg/l TDZ This method was recently compared with one developed by Chand et al. (1999). In this method, the MS medium is supplemented with 150 mls of deproteinised coconut water (DCW). The protocol, based on this medium, consists of two stages: Stage 1: MS + 150 mls DCW + 0.6 mg/l TDZ Stage 2: MS + 150 mls DCW + 1.0 mg/l TDZ With both methods, the cultures are grown in McCartney bottles, at a temperature of 25°C under a light intensity of 50 µmolm-2 s-1 and a photoperiod of 16 h. Cultures are transferred to the next stage at three weekly periods. The two methods gave significantly different results after nine weeks of culture for three different cultivars (Table 1). The comparison was carried out on three different cultivars from the RGC, one from the Cook Islands (CPUK) and two from Fiji (TNS and Hybrid). Table 1: Sucker numbers obtained from two different in vitro multiplication systems Cultivar CPUK TNS Hybrid Method Explant nos at start of Stage 1 Sucker nos at end of Stage 1 Sucker nos at end of Stage 2 Sucker nos at end of Stage 3 RGC 7 19 21 212 Chand 7 13 9 48 RGC 7 13 20 77 Chand 7 8 12 25 RGC 7 11 11 38 Chand 7 6 9 12 The RGC TDZ method is therefore the protocol being used for multiplying up the taro accessions for distribution. For countries requesting taro, proliferating cultures will be provided to those with tissue culture laboratories. For those countries without tissue culture capacity, rooted plantlets will be provided. Although the multiplication method being used by the RGC has optimized propagation rates compared to what is possible in the field, the effect of the genotype is still apparent, as illustrated by the results in Table 1. Research continues in the RGC on taro multiplication to improve multiplication rates. 4. Conclusion Regional in vitro collections provide the means to safely conserve selected accessions, such as core collections and/or breeding lines, in a region composed of small island countries with strong quarantine concerns and limited resources. At the same time, in vitro technology facilitates regional access to these accessions. However, it is merely one component of a complementary conservation and utilization strategy. Other conservation methodologies also have an important role to play. Field genebanks can still be used by countries to conserve specific accessions of particular importance, for example elite breeding lines or cultivars preferred by farmers. They can be important working collections, promoting evaluation, and at the same time having an educational role. Once the conditions determining optimum seed storage are defined, seed conservation could allow countries to maintain taro genes (as opposed to genotypes) at low cost, especially from less used cultivars. It will also allow countries with no breeding programme to evaluate seedlings and to select those best suited to their needs. On farm conservation needs to be further investigated. Since it is impossible to collect and conserve ex situ all taro genetic diversity, supporting farmers’ efforts to maintain large numbers of varieties in their fields could facilitate the long-term maintenance of a much wider range of genetic diversity, and at the same time benefit taro growers directly, through value adding and other strategies. References Bessembinder, J.J.E., Staritsky, G. and Zandvoort, E.A. 1993. Long-term in vitro storage of Colocasia esculenta under minimal conditions. Plant Cell, Tissue and Organ Culture 33:121–127. 72 third taro symposium Chand, H., Pearson, M.N. and Lovell, P.H. 1999. Rapid vegetative multiplication in Colocasia esculenta (L.) Schott (taro). Plant Cell, Tissue and Organ Culture 55:223–226. Côte, F.X., Goue, O., Domergue, R., Panis, B. and Jenny, C. 2000. In-field behaviour of banana plants (Musa AA sp.) obtained after regeneration of cryopreserved embryogenic cell suspensions. Cryo-Letters 21:19–24. DeVerno, L.L., Park, Y.S., Bonga, J.M. and Barrett, J.D. 1999. Somaclonal variation in cryopreserved embryogenic clones of white spruce (Picea glauca (Moench) Voss.). Plant Cell Reports 18:948–953. IBPGR/CIAT. 1994. Establishment and operation of a pilot in vitro active genebank: Report of a CIAT-IBPGR collaborative project using cassava, Manihot esculenta Crantz, as a model. IPGRI, Rome/CIAT, Cali, Colombia. Mannonen, L., Toivonen L. and Kauppinen, V. 1990. Effects of long-term preservation on growth and productivity of Panax ginseng and Catharanthus roseus cell cultures. Plant Cell Reports 9:173–177. Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays with tobacco tissue cultures. Physiologia Plantarum 15:473–497. Palupe, A. 1997. The rapid multiplication of taro Colocasia esculenta (L.) Schott var. esculenta by in vitro shoot tip culture. M.Agr. thesis. University of the South Pacific, Samoa. 135 p. Staritsky, G., Dekkers, A.J., Louwaars, N.P. and Zandvoort, E.A. 1986. In vitro conservation of aroid germplasm at reduced temperatures and under osmotic stress. p. 277–284. In: Withers, L.A. and Alderson, P.G. (eds). Plant tissue culture and its agricultural applications. Butterworth, London. Takagi, H., Thinh, N.T., Islam, O.M., Senboku, T. and Sakai, A. 1977. Cryopreservation of in vitro grown shoot tips of taro (Colocasia esculenta (L.) Schott) by vitrification: 1. Investigation of basic conditions of the vitrification procedure. Plant Cell Reports 16:594–599. Tuia, V.S. 1997. In vitro multiplication of taro (Colocasia esculenta var. esculenta (L.) Schott). M.Agr. thesis. University of the South Pacific, Samoa. 109 p. third taro symposium 73 Theme One Paper 1.5 Promoting on farm conservation of taro through diversity fairs in the Solomon Islands Roselyn Kabu Maemouri and Tony Jansen Kastom Gaden Association, Honiara, Solomon Islands Introduction This paper is about the experiences of two diversity fairs that were held to promote on farm conservation of taro (Colocasia esculenta) in Solomon Islands. The diversity fairs were held at the site of taro field genebanks in Malaita and Temotu provinces established under TaroGen, a regional project implemented by the Secretariat of the Pacific Community (SPC) in collaboration with national partners and funded by AusAID. (Some additional funding for the diversity fairs was provided by European Union Micro Projects Program in Solomon Islands.) Diversity fairs were organised by local farmers, with assistance from the Planting Materials Network (PMN) and agriculture department officers to coincide with the time that taros were ready to harvest from the farmer-run provincial field gene banks. The overall aims of the diversity fairs were to distribute taro planting materials back to the farmers in the province where the diversity had been collected and to raise awareness among farmers about taro conservation. In addition, extensive reference is made in this paper to information collected during participatory rural appraisals (PRA) that were carried out with groups of farmers during the process of collecting taros from farmers for the field genebanks and during discussions, both formal and informal, held before, during and after the diversity fairs. Background Solomon Islands is geographically in the heart of Melanesia, with Papua New Guinea to the west and Vanuatu to the south-east. It is the third largest archipelago in the Pacific, with a land area of 28,900 km2, and is made up of at least 922 islands, including 12 major mountainous islands, with the tallest mountains over 2000 m above sea level. There are over 90 languages in Solomon Islands among the population of 430,000, which is ethnically 90% Melanesian, with some Polynesian islands, and also a number of Micronesian settlements. Approximately 85% of the population lives in rural villages on customary land under traditional tenure, practicing semi-subsistence agriculture based on various forms of shifting cultivation. Current practices include a mixture of short and long fallow fields where different crops are grown to make use of different levels of soil fertility and cropping conditions. Taro in Solomon Islands Taro has been an important crop for the people of Solomon Islands for thousands of years. It has very significant cultural meaning and value, having been used in the past – and still in the present – for bride price, as compensation, in pre-Christian religious ceremonies, in feasts and for sharing. It continues to be important for food security, for various cultural purposes, and for income (through barter and the formal cash economy), as it is generally the most highly valued root crop. Taro cultivation is embedded in the culture and way of life of Solomon Islanders, who hold much traditional knowledge about the crop. Taro is also associated with “kastom” power, sorcery and other traditional beliefs, including numerous taboos and rituals. Taro diversity is very high in Solomon Islands in terms of numbers of landraces. However, there is evidence from molecular studies done by TaroGen partners and others that genetic diversity is actually rather limited. Solomon Islands can be considered to be part of the Melanesian centre of origin of taro, and there is some anecdotal evidence collected by the authors that farmers are continuing to develop new taro landraces through traditional practices. These include recognition of occasional somatic mutations and of naturally-occurring hybrid seedlings. The changes currently occurring in taro cultivation and taro diversity are complex and are closely related to the many different influences affecting Solomon Islands society in general. In general, however, taro cultivation is in decline, due to the following: cultivation and production constraints (related to population growth, changed settlement patterns and land degradation); pest and disease (especially taro beetle and various viruses); changing consumption patterns (particularly the move to processed white rice, noodles and white flour); and economic and market forces. As land use has intensified, largely due to population growth, it has become increasingly difficult for farmers to produce taro. As a result, taro has largely been replaced by sweet potato (Ipomoea batatas), which is now the most important staple crop. A number of other new arrivals have also become very important in some places, e.g. cassava (Manihot esculenta), kangkong taro (Xanthosoma sp.) and pana (Dioscorea esculenta), but taro remains a very important crop. In certain areas it is still a major staple, particularly in low population density areas and the highlands of islands that are settled in the interior. Promoting on farm conservation The Planting Material Network (PMN) is a national network of farmers with an interest in food security at the village level. It is supported by Kastom Gaden Association, an indigenous Non-Government Organisation registered 74 third taro symposium as a charitable trust in Solomon Islands. PMN promotes on-farm conservation of crop diversity and provides a number of services for farmers. It facilitates the exchange of seeds and planting materials, disseminates information through a newsletter, and provides training and facilitates exchanges between farmers. PMN decided in 2001 to hold a number of taro “diversity fairs” in an attempt to promote better management and conservation of taro diversity on farm. The diversity fairs were modeled on experiences of other countries, but were adapted to the local situation. On farm conservation and off farm (or ex situ) conservation are distinct but complementary approaches to maintaining and using the genetic diversity of cultivated crops. On farm conservation programs use various different tools and interventions to encourage farmers to continue looking after and using crop diversity, and thus prevent it from disappearing from their fields and gardens. This allows for continued evolution of the crop within the environments and farming systems where it was developed and continues to be used. Ex situ conservation of taro in Solomon Islands has to date consisted of the establishment of national collections stored in field gene banks under the management of the Ministry of Agriculture. Similar to the experiences of other countries in the region, field gene banks have proven to be very difficult to maintain in Solomon Islands. This is due to the high expenses involved, the difficulty of managing pest and disease, decline in soil fertility and poor management in general. This is demonstrated by the total loss of collections made by the Ministry of Agriculture (MAL) in 1994 and in 1999/2000, and similar difficulties faced by PMN in maintaining collections in 2001. Yet farmers continue to maintain a large amount of diversity on farm despite the increasing constraints mentioned above. Indeed, during PRAs conducted by PMN, evidence was collected that indicates some farmers, particularly older women in Choiseul province, are finding and generating new taro landraces through careful observation and collection of natural taro crosses that are occurring in their bush fallow gardens. PMN decided that on farm conservation by farmers themselves may well be the most sustainable conservation strategy for taro, with the highest likelihood of success and results that would directly benefit farmers. After the loss of a previous field collection undertaken by the Department of Agiculture as part of the TaroGen project, PMN agreed to organise a recollection of taro for TaroGen. This was done with the proviso that a diversity fair would also be held and no attempt would be made to maintain the field gene banks beyond one season, given previous negative experiences. These steps were followed in implementation of the project: 1. Training and planning 2. Collection in four provinces 3. One season field genebank 4. Holding the diversity fair Each of these steps is described below, followed by recommendations and a discussion of the lessons learned. Training and planning In order to carry out the taro recollection, including describing the taros using morphological descriptors and basic passport data, training was provided by the TaroGen team leader (Dr G.V.H. Jackson) to the four local collection team leaders (Tony Jansen and Roselyn Kabu Maemouri from Kastom Gaden Association/PMN, Rex Filia from the National Agriculture Training Institute, Malaita, and Jean Galo from the Ministry of Agriculture and Livestock’s Quarantine Division). This group was then responsible to facilitate 2-3 day training and planning workshops in each of the four provinces with selected PMN members and farmers (about 15-30 attended each workshop). During the training, the collection team was selected. We also selected villages or areas where the collection would take place, and the site for the field gene banks. Groups of farmers then carried out the collection of taro in each province. The collection teams were also trained in the use of three PRA tools (a brainstorm list using focus group discussions, a historical matrix, and garden cycle diagrams) that were used to collect data about taro in the past and present. Three days proved to be a very limited time for the training workshops but nonetheless good results were achieved, with many taro landraces and considerable information collected. However, as the data were collected using a PRA approach, there were some limitations. For example, it was not always possible to make comparisons between provinces or to disaggregate data by gender. Also, the data pertains to only a limited number of sites. More research will be needed to confirm if the findings are applicable in other provinces. The collection Four provinces were selected for recollection of taro: Malaita, Guadalcanal, Temotu, and Choisuel. Due to limited funds, all nine provinces could not be included. The selected provinces included two provinces with large populations where taro was known as an important crop and two smaller provinces on either end of the country. It was found that taro cultivars are usually named by farmers according to criteria such as morphological characteristics, place of origin, and name of person who discovered or introduced the taro. Sometimes, names are given that are associated with the situation or circumstances that led to the acquisition of the taro landrace. For example, a commonly cultivated cultivar in Temotu is named “selfis”(selfish) because one of the farmers who grew it refused to share it with others. The variety originated from Makira, but its Makira name is unknown. Farmers’ names for taro third taro symposium 75 landraces commonly change as they are moved from place too place. Because of such situations, collecting duplicates was inevitable, both within collections and between provinces. Another important issue identified during the collecting is that farmers acquire new landraces regularly. This appears to lead to the gradual displacement of older landraces over time. Farmers are curious to try new landraces, and if those landraces persist beyond first experimentation it is usually because they have good or superior taste characteristics under the local conditions. This observation has important implications for on farm conservation, and also for the prevention of the spread of virus and other diseases between islands. In the end 824 taro accessions were collected in the four provinces. Table 1 compares the 2001 collection with the1999 collection. Table 1: Landraces collected by province in 1999 and 2001 Province 1999 collection (MAL) 2001 collection (PMN) Malaita 173 313 Kwara’ae, Lau bush and coastal Tombaita bush and coastal Guadalcanal 10 220 Guadalcanal plains bush and coastal Longu (north east) bush and coastal villages Choiseul 72 245 Six areas in north-west and south west coastal Temotu 291 46 594 (all S.I.) 824(4 provinces) Total 2001 collection areas Santa Cruz island mainland The full extent of taro diversity in each province has not yet been collected. Indeed, farmers in all provinces reported that they had “hidden taros” that they would not share with just anyone. These hidden taros are rare landraces that have often been in the hands of one tribe for generations. One farmer in north Malaita, for example, had a landrace that he could trace back through oral history for 18 generations. Field genebanks Following the collection in the four provinces, field genebanks were established for each province. The planting of taro in the field genebank was done using traditional methods from that area. This typically involved the clearing and burning of organic matter, followed by planting of tops of taro corms using a digging stick, and other local methods such as interplanting with a local ornamental plan (Coleus sp.) in Choiseul Province. In certain areas, various tabus are involved with taro cultivation, especially in Malaita, including abstinence from sexual intercourse and consumption of certain foods such as turtle and mangrove fruits prior to working in taro gardens. The sites for the field genebank had to have: • fertile soil and no diseases (such as alomae), according to local farmers’ knowledge. • easy access for monitoring and the diversity fair. • a reliable group of farmers who agreed to take responsibility to maintain the genebank until harvest. Agreements were made between PMN and the community or person who would be responsible for the field genebank. The agreement involved payments for care of the taro and labour involved in planting, weeding, and harvesting. Payment was necessary because the farmers would not own the harvested taro corms, as they would be distributed during the diversity fair (as in Malaita) or sold (as in Temotu). In Malaita, the community at Busurata/Kwalo, in Central Kwa’arae, was selected to maintain the genebank. They are well know to KGA and very active members of PMN. In Temotu, Lazarus Kope from Nea village was selected by the MAL officers in Temotu because this he is one of their best contact farmers. In both provinces, the taro was well maintained by the selected local farmers. This was in marked contrast to the collections established in Guadalcanal and Choiseul, which were lost. In Guadalcanal, the genebank was looked after by a rural training centre called St. Martin’s. The genebank was not managed properly and the students were not involved as planned. As a result, the collection died during the dry season, when hand watering and careful tending was required but there was not enough labour. In Choisuel, the genebank was to be looked after by the extension service. Despite funds being available for labour and other necessary expenses, the collection was lost due to poor management and damage by wild pigs. PMN has subsequently done its own recollection of taro for Choisuel in 2002, with 180 landraces collected and a diversity fair planned for later in 2003. This collection is looked after by Sam Moroto, a farmer from Poroporo village and one of the more active PMN members. There is an important lesson here: farmers can be much better at maintaining and looking after field genebanks, as long as they are given training in labelling and laying out. TaroGen identified a core collection of 10-20% of the accessions from the provincial field genebanks in Choiseul, Malaita and Temotu while all three were still alive, and this is now in tissue culture at the Regional Germplasm Centre (RGC) at SPC, Suva, Fiji. This core collection was selected at random as the morphological descriptors recorded at the time of collecting taro in the field proved unreliable. 76 third taro symposium The diversity fair Initially there were plans to hold four diversity fairs. In the end, only two were held due to the loss of the field genebank collections before they were ready for harvest in two of the four provinces. From the outset, PMN had decided to try to link the diversity fairs to a high-profile harvesting and distribution of the material form the provincial taro collections, to attract farmers’ interest and provide a focus for the event. The aims of the diversity fairs held in Temotu and Malaita were thus to: • bring farmers from around the province together to learn about the diversity of taro in their provinces • promote the sharing of knowledge and experiences about taro growing among farmers • distribute the taro varieties in the genebank back to the farmers. Malaita is a long (about 180 km), narrow, mountainous (up to about 1000 m above sea level) island, with a narrow coastal plain in most areas. It has over 100,000 people, making it the most populous province in the country. Many of these people live in the interior of the islands, particularly in north and central Malaita. Temotu Province is located in the eastern part of Solomon Islands and is made up of many islands, including some inhabited by Polynesians. Temotu is situated some 350 km east of the main Solomon Islands chain. With a total area of 865 km2, only 18,300 hectares is of agricultural potential. In 1991, the total population was 16,850, with annual growth rate of 2.8%. Population density is 19 people per km2 in total, but only about one person per hectare of land of agricultural potential land (ITTA). Temotu is known for its good taro varieties and also for the absence of taro diseases like alomae, bobone and taro leaf blight. During the collection of the taro and the initial training and planning workshop, a diversity fair committee was established in each province. This consisted of about 8 people, including the farmer who would look after the genebank. Funds was budgeted for the diversity fair in each province and the committee had to work in line with this budget allocation in developing their plans. The committees met and discussed what activities would be carried out during the fair. They were responsible for coordinating the setting up of the stalls, the decoration of the venue with leaves, flowers and plants, the transport arrangements, food preparation, and the accommodation arrangements. They aimed to see that all the farmers who came to the fair would be satisfied and happy to take part, and that the event would be worthwhile to farmers. Committee members raised awareness and promoted the taro fair in their communities and through the media. This was the first time that diversity fairs have been held in Solomon Islands, so the committee members had to develop their own ideas and hope that they would be accepted and well received by farmers. There is a tradition of trade shows and produce competitions, so the fairs included this type of activity. In Temotu Province, the local provincial office of MAL was responsible for organising the committee, which included most of the MAL staff and a few farmers, while in Malaita the committee was dominated by local farmers (men and women), but included the team leader from PMN and MAL. In Malaita, invitations were sent out to farmers a month before the fair by the organising committee. They were given to the farmers who had given suckers during the collection of taro and to members of PMN. In Temotu, promotion was done a couple of weeks before the fair in the villages around Santa Cruz Island and in the station, through notices and pin-ups sent out in villages. What happened Because the taro collections were planted at different times, in Malaita, they were ready to harvest in April 2001, while in Temotu the taros were not ready until July. More than 200 farmers attended the fair at Kwalo/Busurata village. They were from villages in North Malaita (Bitaama, Takwa, Silolo, Mana’avu, Suava area) and Central Kwara’ae (Busurata, Kwalo, Gwonafu, Aisiko and surrounding villages in Kwalo). We also had students attending from one of the rural training centres in Malaita (Airahu Training Centre) and from a youth training program of Kastom Gaden Association. Farmers from North Malaita came one day before the fair because they travelled about 80 km by truck. About 150 people from Santa Cruz villages came to the fair in Temotu. Due to limited funds, other islands were not included (it is a one or two day boat ride between Santa Cruz and the Reef Islands and Polynesian outliers). Various local representatives of government ministries were involved in the program in Temotu, including Agriculture and Livestock, Police, and Health and Medical Services. The Premier of the province was also present and gave an encouraging speech. In Malaita, there was much less involvement of government departments. In preparation for the day, stalls were built for all the taro varieties that are grown in the field genebank. In Malaita, all the taros (over 300 accessions) were harvested and a bundle of each variety was tied together with leaves and placed on the stall. In Temotu, each of the 46 varieties was planted in a polythene bag well before the show, and five suckers with the corm were laid along the side of each bag with the name of the variety. This allowed farmers to see a living taro with leaves for easy identification. The stalls at both fairs were designed so that people could walk around easily and see all the varieties. third taro symposium 77 Various different activities were organised by the committees: • official opening • speeches by invited guests • cooking and taste competitions • diversity prizes • group discussion by the farmers • entertainment (in Malaita, a local type of traditional music called panpipes using bamboos) Farmers also brought their own varieties of taro and other root crops to the fair, which were displayed. During the viewing of the stalls, farmers tried to guess the local names of the taro varieties according to their local knowledge. Farmers were surprised to see so many different varieties of taro were displayed. One farmer from the north of Malaita said that she found in the collection some taros that had disappeared from her coastal area, still grown by farmers from the highlands of central Malaita. Cooking and tasting of the taros from the field genebank was one of the most popular activities. Taste is considered more important than other characteristic such as yield or size of the corm. In Malaita, the organisers divided the farmers into groups of 3-4 people. Each group took about five different varieties of taro and cooked them in the traditional way using bamboo. The taros are peeled and sliced into halves or quarters depending on their size. The bamboo is spilt in sections with an opening on one end, and it is then filled with the pieces of taro and closed with a leaf. Then a fire is prepared and the bamboos are put over the fire and slowly turned regularly to prevent them from burning until they are cooked. When all the groups were ready, they all brought the cooked taros together in one place and cut open all the bamboos. About 175 different varieties were cooked and everybody took turns to taste many varieties. The taste was scored, and all results recorded by one person in each group. Many landraces are popular and there was no obvious favourite. There were significant differences between the taros of north and central Malaita taro, with farmers from both areas rushing to get taros from the other area. Some of the taros have different names but are actually the same, while others are unique to each region. Farmers from north Malaita commented that they had never seen so many taros from central Kwara’ae, and vice versa. Many farmers thought that north Malaita taros have the best taste, but the ones from central Malaita do not have as many diseases. In Temotu, 10 varieties were selected for the cooking competition. Five varieties were cooked and five were baked. Ten varieties were cooked - five baked in traditional stone oven and five boiled in water. The participants tried to guess the names of the varieties. Anyone who chose at least five correctly received a prize from SI$50-100. Twenty men and women competed in the competition. Only five participants got the names of the varieties right. They were given small cash prizes. In Malaita, a group discussion also took place after the cooking. We divided active taro farmers into two groups of men and women to discuss some of the traditional knowledge related to taro. The discussion revealed that both men and women play an important role in growing taro. In the culture in Malaita, men are responsible for taro growing, but now it is women who do most of the work, including clearing of taro garden sites, especially in coastal villages where the bush fallows are short, with not many big trees. The farmers discussed taro poison (taro sorcery) and taboos associated with taro. Pest and disease was also discussed, especially traditional ways of managing bobone and alomae. Sorcery is considered a big problem with taro cultivation and requires expert knowledge to overcome these problems. In Temotu also, both men and women are involved in growing taro. This contrasts with Choiseul, where women are the taro growers and the holders of taro knowledge. Various speakers gave talks at the fairs, including provincial and traditional leaders, NGO representatives (local and international) and key farmers. They explained the importance of on farm conservation, where farmers have continuing access to their planting materials, and encouraged farmers to maintain taro varieties and not let them disappear due to new crops, introduced crops, and pest and disease. Farmers responded that a lot of the diversity of taro has been lost because of pest and disease, but the fairs did appear to make many farmers very interested to start collecting taro varieties. Prizes were given out to three farmers who brought the most varieties to the fair. Johnson Ladota from Masilana village in the highland of north Malaita took the first price for bringing 13 varieties of taro. One variety called “binalofo” in the local language had been growing for 17 generation in his family. The other two prize winners were two young women, Elsie Siale from Mana’abu village and Freda Siuta from Bita’ama village. The three farmers were given t-shirts and cash amount of SI$30 each. This is to encourage farmers to participate in such event and also to maintain their taro varieties. Only 10 farmers brought taro varieties to the show. This was fewer than expected, as some of the farmers did not get the message in time. Also, some had not harvested their taros yet, and many are in any case reluctant to share their “hidden” varieties. However, over time farmers might be more willing if they felt there was enough prestige associated with winning prizes at the fair. It is expected that more farmers will bring their own planting materials at future fairs now they understand the purpose of the event and the possibility of winning prizes. In Temotu, in the late afternoon the taros were sold to farmers as tubers and suckers to plant. The tubers were sold for SI$0.50-4.00, depending on the size. The Selfish taro was the most popular. People liked it because it has a good taste and it grows really very well in the area, without any diseases. The money that the organising committee raised from the 78 third taro symposium sale of taros was also used for diversity fair expenses. In Malaita, farmers were allowed to choose five varieties each to take home with them. This was recorded in a register. After everyone had been through the line, farmers were allowed to take any of the remaining material. Conclusions The cost of the diversity fairs was rather high at up to SI$14,000 each (US$2,000 approx.). This included transport for participants because the distances are large and traveling by boat expensive. However, despite the expense, both PMN and the participating farmers felt it was a worthwhile thing for farmers to come together in this way, and certainly a more cost-effective exercise than ex situ collections that end up being lost anyway. Farmers went away with new diversity and an enhanced awareness of its importance, which bodes well for continued and sustainable conservation of taro on farm in these provinces. At the national level, the diversity fairs generated widespread media interest. This resulted in farmer groups and provincial authorities from two other provinces requesting PMN to hold diversity fairs in their province. Many farmers and the public in general were inspired by the stories of the diversity fairs and started to collect varieties on their own. Diversity fairs have the potential to strengthen on farm conservation if they are part of an integrated program such as that offered by PMN. On farm conservation is probably the only way that crop genetic resources will be maintained in the long term in a form accessible to farmers given limited resources. As farming systems change, wider and more integrated interventions may be needed to provide other incentives for the maintenance of crops that farmers are slowly shifting away from. The maintenance of crop genetic resource should be encouraged where it can also help to meet other developmental needs. This can be done by strengthening traditional uses and values of the crop as well as providing new opportunities such as processing or new markets. The diversity fair concept will need to evolve and adapt to the many different cultures and farming systems of Melanesia if it is to be a success. A participatory approach to planning and implementation, with farmers at the centre is more likely to allow this to happen and create events that have real value for local farming communities. Recommendations • There should be more diversity fairs held on a regular basis not only with taro but also other root crops. • Fairs should take up to 3 days to allow more activities, such as speakers on different topics and small workshops run by farmers on how they manage pests and diseases etc. • There should be better recording of the distribution of taros from the fair to monitor how farmers maintain the varieties over time, and better recording of discussions by farmers, perhaps in local languages. • Events could be organized in combination with another gathering, e.g. provincially appointed days for sports, cooking, and competitions. These events bring people together from the whole province to attend and would save on costs as well as involving a wider sector of the population. • Combining the diversity fair with a field genebank was a good way to generate interest in the fair. Without the field genebanks it appears that very few farmers would have brought their own planting materials to share. More awareness prior to the event could overcome this problem. • Farmers are much better at maintaining field collections and organising events than the Department of Agriculture, training centres or NGOs. A partnership approach between the NGO and farmers worked well. The support NGO proved more effective than MAL. • The diversity fair day encouraged farmers to look after their taro diversity and reintroduce traditional valves and knowledge. It was an inspiring event for all involved that reminded them of the wealth of their ancestors. Further reading Bonie, J.M. 1993. Improved Temotu traditional agriculture. Agriculture Extension Services, Temotu Province, Solomon Islands. Jansen, T. 2002. Hidden taro, hidden talents: A study of on-farm conservation of Colocasia esculenta (taro) in Solomon Islands. Presented at SPC TaroGen meeting on on-farm conservation. Lebot, V., Simeoni, P. and Jackson, G. 2001. Networking with food crops: A new approach in the Pacific. In: Wells, K.F. and Eldridge, K.G. (eds). Plant genetic resources in the Pacific: Towards regional cooperation in conservation and management. ACIAR, Canberra. third taro symposium 79 Theme One Paper 1.6 Home gardens and their role in the conservation of taro diversity in Vietnam Nguyen Thi Ngoc Hue1, Luu Ngoc Trinh1 and Nguyen Van Minh2 Plant Genetic Resources Centre, Vietnam Agricultural Science Institute, Van Dien, Thanh Tri, Hanoi, Vietnam. 2 Plant Oil and Perfumes Research Institute, 171-175 Ham Nghi Street, Ho Chi Minh City, Vietnam 1 Introduction Taro is a common root crop in Vietnam, where it is used in various ways not only by people in rural areas, but also by urban dwellers. Its corm or cormel is used as a staple food, while the stem is fed to pigs, and the stolon of one variety is chopped and boiled to make a medicine for constipation. There is a variety which produces a tasty young leaf and a petiole which are stir-fried with garlic to make a special dish eaten on Festival days or cooked in special soups such as “lau” and “bun moc”. Truong Van Ho et al. (1994) and Nguyen Thi Ngoc Hue (2000) noted that taro plays an important part in the culture of the people in some parts of the country. Freshly harvested corms are a preferred gift by rural people when they visit relatives in towns or cities on special occasions. Some varieties with particularly tasty corms are grown commercially on a large scale in Vietnam. Taro is also commonly grown in home gardens. Eyzaguirre (1996) mentioned that the genetic diversity present in home gardens is closely linked to the multiple and varied uses of the plants by traditional households, and the level of genetic variation in garden-planted taro appears higher in places where its importance is greater than in areas where the crop is not highly regarded. However, taro genetic resources face various threats in Vietnam (Nguyen Thi Ngoc Hue, 2000), for example traditional varieties are being replaced by short duration crops in some areas. In view of these observations, the role of home garden in the conservation of taro needs to be seriously evaluated. This paper discusses the role of the home garden and its custodians in the conservation of taro genetic resources in Vietnam. Materials and methods PRA surveys were conducted in 180 home gardens in 4 ecological regions in the country: the Red River Delta (Nhoquan; 60 sites), the Northern Midlands (Nghiadan; 30 sites) suburban Ho Chi Minh City (Thuan An; 60 sites) and the Mekong River Delta near Can Tho (ChauThanh; 30 sites). The surveys aimed to gather information on home garden size and structure, on the measures instituted by householders to protect home garden plants, on the number of taro varieties present in the garden and on the manner in which taro is being kept in the garden in association with other plants. Interviews involved the home garden owners, both men and women, and attempted (i) to identify the family member(s) in charge of taking care of the garden, (ii) to discuss the aim of maintaining the garden and the reasons for including taro in the garden, and (iii) to determine the names of the taro varieties grown in the garden, their uses and the length of time they have been planted in the garden. Results and discussions Description of the home gardens structure and composition Home gardens make an important contribution to the livelihood of Vietnamese farmers, generating income and improving the material and cultural living standard of rural people. The total area of home gardens in Vietnam is estimated at approximately 200,000 ha, or around 4% of the cultivated area. The size of home gardens is lowest in the Red River Delta, with the average size 150m2, and largest in the Central Platen (West Highland), with the average size 0.5 ha. The majority of the home gardens surveyed in Vietnam were located less than 10m from the residence of the garden owner. The reasons are to make the garden easily accessible to any member of the household for harvesting, and to watching over the garden and drive away foraging animals like pigs and poultry. Also, the plants in the garden can contribute in beautifying the landscape immediately surrounding the house. Throughout Vietnam, a bare dirt or paved area opens directly in front of the house, with ornamentals bordering the open space or lining a walkway, including flower species such as the Hue lily (Polianthes tuberosa) and the hibiscus (Abelmoschus moschatus tuberosus), whose flowers are often placed daily on the household shrine. Medicinal and herb species are often located on the edge of the open porch area, for easy access. Fruit trees are mostly located behind the house, with a few choice species in the front, often the bigger fruit trees (>5m tall) such as eggfruit (Lucuma mammosa) that also provide shade to the cleared house yard. Gardens in lowland areas may have canals that are dug to create raised beds of land, providing good drainage in the rainy season and maintaining water supply in the dry season. In this way, diverse types of crop species can be accommodated into small niches. Fruit trees are planted on the higher ground and interspersed with lower-story crops such as Colocasia esculenta, Xanthosoma sagittifolia, Capsicum sp., or Mentha sp. The ditches are either used for drainage or irrigation depending on the region of Vietnam (primarily drainage in the Mekong Delta) and the season of the year. Ditches or ponds can accommodate crops that need more moisture, such as 80 third taro symposium Ipomoea aquatica, Eichhornia crassipes, and Sagittaria sagittifolia (Hodel et al., 1999). There is usually a more open area of the home garden where vegetables requiring light such as Solanum undatum or Brassica oleracea are grown. Bamboo (Bambusa multiplex) often provides a side or back border, with cacti commonly used as a lower fence for the front of the yard. Home gardens in Vietnam can be classified in two general categories based on primary production system, crop composition and structure: 1. Home gardens with fruit trees (South Vietnam) with pond and covered livestock area (RRD and Central Vietnam) or with some vegetables. 2. Home gardens with mixed planting of crops, including fruit trees, medicinal plants, vegetables, spices, root crops and ornamentals The percentages of home gardens with mixed plantings are 85.8%, 93,3%, 91.4,% and 50% in Nho Quan, Nghia Dan, Thuan An and Chau Thanh, respectively (Table 1). Taro as garden plant One hundred of the total number of gardens in Nghia Dan and Thuan An contain two or more taro varieties. All the gardens in Chau Thanh contained two taro varieties. In Nho Quan, the proportion of gardens with one, two, or >2 varieties was 46.7%, 13.3%, and 40.0%, respectively (Table 3). In home garden at all study sites, taro occupies the shaded or moister portion of the garden. Feral taro grows along household drains, canals or artificial waterways. With the long history of home garden taro farming, many different local names of taro varieties have developed. In Nho Quan, some varieties are named according to their special characteristics. For instance, one variety is called “Docmung,” which refers to the fact that the petiole is used. The farmers explained that this variety is so named because its petioles are an excellent vegetable and can be cooked together with other vegetables and fish or meat to make a special food (Lau). “Bac ha” is another name of a taro cultivar, this time referring to the silver color of the petiole. “Khoai nuoc” or “Ngua” are common names of feral taro. These names mean “ water taro” or “ itchy taro”. “Sap”, meaning white, soft and sticky as paraffin has a sticky corm with a yellow waxy color of the root flesh. “Tamdaoxanh” is named according to the place from which this variety was first introduced by a French person. “Sen” is the name of the lotus plant. A local name that is not understood even by the gardeners themselves is “Tau.” This name is said to have simply been adopted from the ancestors. Six varieties of taro were found in Nho Quan. Garden custodians can readily identify all of them. Similarly, the number of taro varieties observed in Nghiadan, Thuanan and Chauthanh were 4, 6 and 5, respectively. Some varieties of taro were found in home gardens but were not present in the larger fields and paddies (Table 2). The other plants planted in home gardens vary between the four sites. In Nho Quan, the commonly observed plants grown with taro are luffa (Luffa cylindrica) sweet potato (Ipomea batatas), lablab (Dolichos lablab), watercress (Ipomea aquatica) and ginger (Zingiber officinale). In Nghia Dan, yam (Dioscorea alata), sweet potato and Alpinia tonkinensis were the crops most prevalently planted together with taro. In both sites of the North, the lemon(Citrus aurantifolia), Pherynium parviflorum, eggplant (Solanum melogena) and chili (Capsicum annum) are also common. In Thuan An and Chau Thanh, taro is always grown under the shade of fruit trees with different kinds of spices and tropical vegetables, such as eggplant (Solanum sp.), chilles (Capsicum sp.) and bitter gourd (Momordica charantia) Role of home garden custodians Preliminary results show that men make most of the decisions relating to cultivation of fruits and ornamentals (Luu Ngoc Trinh et al., 2000). In contrast, spices, vegetable and root and tuber crops seem to be the domain of women. These crops are the most important for home consumption and are used daily in meal preparation by women. In most cases, it is the mother who originally planted taro in home gardens at all four sites. At Nho Quan, almost all (95.4%) of households obtained their first planting materials of taro from relatives, the rest from market. In contrast, in Nghia Dan, half of garden owners received taros from relatives. Different types of taro are cultivated for the use of different parts of the plant. Also, different ethnic groups have their own way of preparing taro and there are special occasions and festival when the products of taro are particularly relished (Table 2). The extent of cultivation and distribution of taro cultivars vary with ecology, farmers’ specific preferences, socioeconomic conditions, market forces and cultural values. All of the gardens in Nho Quan, Nghia Dan and Thuan An include taro for food and feed purposes, but in 90% of the home gardens of Chau Thanh taro is grown for income, and only in 10% for food and feed. third taro symposium 81 Table 1: Feature of the home gardens in Vietnam Total number of home gardens (%) Nho quan Nghia dan Thuan an Chau thanh Distance from the owner’s residence (m) 1. Less than 10 100 100 94.3 100 2. 10-20 0 0 5.7 0 3. More than 20 0 0 0 0 Approximate areas (m2) 1. Less than 500 6.7 0 0 0 2. 500 - 2000 73.3 36.7 45.7 4.8 3. 2000- 4000 20 56.6 34.3 23.8 4. More than 4000 0 6.7 20.0 71.4 Condition of the HG 1. Area is shaded by tree 100 100 100 100 2. Soil is moist 0 0 0 0 3. Soil is well drained 90 0 0 0 4. Area not shaded by tree 0 0 0 0 1. With fence all around 67.5 50 65.7 80.0 2. Animal nearby are tied 0 20 20.0 0 32.5 30 14.3 0 0 0 0 20.0 93.3 100 94.3 20.0 6.7 0 5.7 80.0 0 0 0 0 Safety measures provided 3. Owners watch the garden 4. Others Planting pattern 1. Mixed planting of crops without proper planting distance and arrangement 2. Systematically planted with proper planting distance and arrangement 3. Others Home garden composition 1. With fruit tree 2. With pond and covered livestock areas 3. With vegetables 4. With forest tree 5. Mixed planting of crops 0 0 0 50.0 14.2 6.7 5.7 0 0 0 0 0 0 0 2.9 0 85.8 93.3 91.4 50.0 0 Portion where taro is planted 1. Shaded 80.0 0 80.0 2. Moist area 10.0 50.0 20.0 0 0 50.0 0 100 4. Others, dry part Table 2: Main morphological characteristics and use value of some taro varieties observed in home gardens in Vietnam Local name Morphological characters Distribution Uses Nuoc tia Plant height 100-120 cm, dark purple petiole, dark green leaf, yellow junction color, ellipse corm, itches, some and very short stolon Drainage canals. Pig feed Nuoc xanh 80-95 cm, green petiole, green leaf, pink margin leaf, small oval corm, very itches, many and long stolon Drainage canals, moisture area Pig feed Bac ha 50-100 cm, round, light green leaf, whitegreen petiole with glaucous, very small corm, itches, rare and short stolon Shade area, along lower fence for the back of the house Petiole used as vegetable for “lau,” “bun moc” Tam Dao xanh 70-130 cm, glaucous green petiole, big green leaf, corm with cylinder shape, white flesh and white apex, no stolon Uplands, under fruit trees Food and fodder 82 third taro symposium Stolon used for cooking “bup khoai so kho tuong” and “khoai nau me” Tam dao tia 70-100 cm, cup-shaped dark green leaf, dark purple glaucous petiole, oval corm, long oval cormel with purple fiber flesh, no stolon Drainage canals near well Petiole used as vegetable and pig feed, cormel for soup Mon do 80-100 cm, dark green leaf with purple junction, red purple petiole, very small oval corm, no stolon Near well Medicine for stomach ill and dysentery Cao 100-140 cm, green leaf and petiole, round corm Uplands with fruit tree Food and pig feed Mon Sap 70-100 cm, green leaf, green petiole, round corm with many oval cormels, Dry area with fruit tree Food and feed Doc mung 80-100 cm, green leaf, green with purple stripe petiole, purple junction, small corm, itches Drainage canals near well Food and feed Mon ngot 80-110 cm, green leaf, red junction and top petiole color, big round corm. Under shade of fruits Food, cake Khoai So 80-100 cm, green leaf, green with purple stripe, small round corm Intercropping with sweet potato Boiled, soup Tim 90-110 cm, big, long, lumpy corm, purple flesh Shade area Food, cake Ngua 120-140 cm, small leaves, purple junction color, oval corm very itches, long stolon Drainage canals, beside ponds Animal feed Sen 90-100 cm, large leaves, green light violet towards upper end, big oval tuber, glutinous Intercropping with spices Sweet soup Tau 90-100 cm, glaucous green petiole, big and green leaf, oval corm, elliptical cormel, yellow flesh Under shade of fruits Sweeten soup Table 3: Features of home gardens with taro Total number of home garden (%) Nho quan Nghia Dan Thuan an Chau thanh 47 41 2 48 42 10 56.31 33.63 10.06 54,5 39.0 6.5 52.2 15.8 32.0 58.2 22.0 19.8 57.1 20.0 22.9 43.7 31.3 25.0 90 8 2 100 0 0 100 0 0 10.0 90.0 0 Person who originally planted taro 1. Father 2. Mother 3. Other household member 20.2 72.5 7.3 10.0 90.0 0 43.0 57.0 0 40.0 60.0 0 Source of original taro planting materials 1. Other taro grower 2. Relatives 3. Market 0 95.4 4.6 50.0 50.0 0 50.0 21.4 28.6 20.0 0 80.0 Years of maintaining taro in garden 1. About 1 year 2. Less than 5 year 3. Five years or more 0 0 100 20.0 50.0 30.0 0 28.6 71.4 0 0 100 Number of taro maintaining in garden 1. One 2. Two 3. More than two 47.7 13.3 40.0 0 70.0 30.0 0 78.6 21.4 0 100.0 0 Frequency of harvesting taro for vegetable 1. Almost daily 2. Weekly 3. Monthly 4. Yearly 5. Have not harvested yet 40.0 60.0 0 0 0 0 70.0 0 30.0 0 35.7 64.3 0 0 0 0 0 0 100 0 Person responsible for care of the garden 1. Husband 2. Wife 3. Other, parents Gender distribution in decision making 1. Female 2. Male 3. Both Purpose in planting taro 1. For food and feed 2. For income 3. Other third taro symposium 83 In Nho Quan and Thuan An the vast majority of home gardens had been growing taro for five years or more. In Nghia Dan, 20% of home garden had taro for just a year, and only 30% of home garden maintained taro for more than 5 years (Table 3). The majority of home garden custodians in three sites claimed that they do not have problem with taro as a garden plant because it is easy to grow, propagate and maintain. The only constraint mentioned was drought. In contrast, in Chau Thanh the custodians complained that they have problems with pests and taro leaf blight. In Nghia Dan and Nho Quan, the majority of garden owners do have a preferred variety, whereas in Thuan An two thirds of gardeners do not have any preferred varieties. The qualities garden owners look for are ease of cooking, delicious taste of corm and stolons and big corms. The majority of garden owners in Nho Quan and Thuan An use petioles and stolons weekly, whereas in Nghia Dan, only 30% of gardens owners harvest petiole and stolon every week (Table 3). Conclusion The composition and structure of Vietnamese home gardens vary markedly depending on agro-ecology, market forces and traditional culture. However, taro is commonly grown in home gardens throughout the country. The area under taro in home gardens is small but sustainable and taro plays a significant role in household food security. Preparation, choice of variety, cooking style and eating time vary with ethnicity and ecology. During the long history of taro cultivation, local people in Vietnam have accumulated a rich store of indigenous knowledge and experience in the use and management of taro genetic resources. Different varieties of taro are grown for different purposes and under different maintenance regimes. There seems to be little genetic erosion of taro in home gardens in comparison with its situation as a field crop. The fact that a number of varieties of taro were found in home gardens and were not present in the larger fields and paddies of the wider agro-ecosystem suggests that home gardens are good sites in which to conserve the genetic diversity of taro and home garden can play a complementary role in conservation for taro genetic resources. Acknowledgements Dinh Van Dao from VASI, Nguyen Hong Tin from Cantho University and Phan Thi Chu from Phuqui Fruit Research Center are duly acknowledged for data collecting. We are also indebted to the donors who made the study possible through their financial support, Germany’s GTZ/BMZ (Deutsche Gesellschaft fur Technische Zusammenarbeit/German Federal Ministry for Economic Cooperation and Development). References Eyzaguirre, P. 1996. IPGRI work on the ethnobotany and economics of the conservation and use of plant genetic resources. IPGRI–APO Newsletter 20:1–2. Hodel, U., Gessler, M., Cai, H.H., Thoan, V.V., Ha, N.V., Thu, N.X., and Ba, T. 1999. In situ conservation of plant genetic resources in home gardens of southern Vietnam. IPGRI, Rome. Luu Ngoc Trinh et al. 2000. Technical report of in situ Project LoA98/108 for 2000 year. Nguyen, T.N.H. 2000. Taro diversity and use in Vietnam. p. 12–17. In: Zhu, D., Eyzaguirre, P.B., Zhou, M., Sears, L. and Liu, G. (eds). Ethnobotany and genetic diversity of Asian taro: Focus on China. IPGRI–CSHS, Rome. Truong Van Ho et al. 1994. Root and tuber crop genetic resources in Vietnam. p. 167–173. In: Proceedings of the International Workshop on Plant Genetic Resources, 15–17 March 1994. MAFF Research Council, Japan. 84 third taro symposium Theme One Paper 1.7 Diversity and genetic resources of taro in India S. Edison, M.T. Sreekumari, Santha V. Pillai and M.N. Sheela Central Tuber Crops Research Institute, Trivandrum, India Introduction Taro (Colocasia esculenta (L.) Schott) is a traditional crop with a long history of cultivation in Asia and the Pacific. It is widely cultivated in India, where it is usually grown as a subsistence or semi-commercial crop in the homestead garden for its cormels, petiole and leaves. The common (local) names of taro in different parts of the country are: arvi (Hindi), chempu (Malayalam), seppan kizhangu (Tamil), kachchi (Kannada), chamadumpa (Telugu), alu (Marathi) and kachu (Bengali). There is little statistical data available on the area and production of taro in India, but taro commands a higher price than cassava or sweet potato. Efforts toward the genetic improvement of the crop are still meagre. The Central Tuber Crops Research Institute (CTCRI), under the Indian Council of Agricultural Research (ICAR), Ministry of Agriculture, Department of Agricultural Research and Education, has included taro in its mandate from its inception in 1963, though it was a lower priority than cassava and sweet potato. Germplasm collections of the crop were assembled from various parts of the country and research programmes were initiated mainly on conservation and basic research. From the end of the 1980’s, when the germplasm banks were completed, research projects were initiated at CTCRI for the improvement of the crop. Germplasm collections Plant genetic resources researchers are trained in the conduct of germplasm exploration trips and collecting criteria during short courses at the National Bureau of Plant Genetic Resources Institute, New Delhi, a sister institute under ICAR. The National Bureau of Plant Genetic Resources (NBPGR) organizes field collection trips for all crops, including tropical tuber crops. Since CTCRI is the sole research institute in India dealing with research and development of tropical tuber crops, researchers and technical staff of CTCRI join in such trips arranged for tuber crops collection. Several collection trips have so far been undertaken, resulting in the procurement of a total of 4210 accessions of various tuber crops at CTCRI, including 424 of taro (Table 1). Table 1: Number and source of taro germplasm accessions maintained at CTCRI Region No. of accessions South India 148 Central India 78 North India 84 North-East India 114 Conservation and characterization The accessions are brought to the institute in the form of corms and/or cormels and are raised initially in the nursery and later transplanted to the field. Five to 10 plants per accession are usually maintained for ex situ conservation in field genebanks. Irrigation, weeding, earthing up, etc. are carried out as and when necessary and the material is maintained from year to year by timely transplanting. The collections have been characterized morphologically using revised IPGRI descriptors lists for taro (Unnikrishnan et al., 1987, 1988). A summary of the variation obtained is presented in Table 2. There are only a few accessions of wild forms in the collections (4.7%). The cultivated material includes the dasheen and eddoe botanical varieties, as well as intermediate types. These could be hybrids between the two botanical varieties, or accessions that are difficult to classify because of the unusual shape of their corms. Both dasheen and eddoe types were found in all parts of the country. A wide spectrum of variability was evident with regard to almost all characters, but Lebot et al. (2000) have reported that morphologically variable taro varieties might show a narrow genetic base with limited allelic variation. It is probable that sexual recombination among the cultivars is very rare and the few that naturally set seeds might be due to self-pollination. Even though the pigmentation on different parts of the plant varies a lot, it is likely that very few genes are involved. Moreover, identical morphotypes have different names in different places within country. third taro symposium 85 Table 2: Distribution of morpho-agronomic traits in Indian taros Character Trait Percentage 1 Germplasm type Cultivated Wild 95.3 4.7 2 Botanical variety Dasheen Eddoe Intermediate 28.6 55.5 15.9 3 Plant type Erect Spreading 65.3 34.7 4 Stem girth High Medium Low (> 20 cm) (10-24 cm) (< 10 cm) 24.4 60.6 15.0 5 Tillering nature High Medium Low (> 6) (3-6) (1-3) 2.8 25.3 71.9 6 Leaf arrangement Clockwise Anti-clockwise 55.7 44.3 7 Leaf orientation Semi erect Drooping Cup-shaped 52.3 31.5 16.2 8 Leaf margin Entire Undulate 39.9 60.1 9 Leaf margin colour Green Purple 78.1 21.9 10 Sinus colour (upper) Yellow Purple 56.6 43.4 11 Sinus colour (lower) Yellow Purple 19.3 80.7 12 Petiole colour Green Purple (different shades) 54.4 45.6 13 Sheath colour Green Purple Mixed 47.9 23.6 28.5 14 Flowering nature Flowered Not flowered 14.0 86.0 15 Maturity Early Normal Late 18.7 56.3 25.0 16 Corm shape Cylindrical Round Conical Club-shaped Elliptical Multishaped Rhizhomatous 41.3 7.7 14.0 8.0 0.3 28.0 0.7 17 Cormel yield Low Medium High (200 g) (200-400 g) (> 400 g) 39.7 41.0 19.3 18 Corm yield Low Medium High (> 300 g) (300-500 g) (> 500 g) 21.6 63.2 15.2 19 Cooking quality of corms Good Poor 18.1 81.9 20 Cooking quality of cormels Good Poor 68.3 31.7 21 Keeping quality of corms Low Medium Long (< 15 days) (15-30 days) (> 30 days) 71.3 15.2 14.5 22 Keeping quality of cormels Low Medium Long (< 15 days) (15-30 days) (> 30 days) 21.5 58.4 20.1 23 Tolerance of colocasia leaf blight Susceptible Tolerant Resistant (<25 weeks) (26-28 weeks) (> 29 weeks) 60.6 39.4 0 .0 Ploidy level Chromosome counts taken from metaphase plates of root tip cells revealed that diploids (2n = 28) and triploids (2n = 42) occur in Indian taros in almost equal proportion. The frequency of the ploidy types varied among the different zones of the country, however. Although both types occur in all regions, diploids predominate over triploids in southern India (Table 3), while triploids out-numbered diploids in the north (Sreekumari and Mathew, 1991). Several factors are known to influence the frequency of polyploids in different eco-geographical regions. Zeven (1980) pointed out that 86 third taro symposium polyploids generally are larger and have greater adaptability, which apparently enable them to thrive at higher latitudes and altitudes. As in the case of Indian taros, Zang and Zang (1990) also observed a greater percentage of triploid forms in the higher altitude regions of China. In India, selection during domestication would have been for greater numbers of cormels, as only these are used, mostly as a vegetable, where cereals form the staple food. It is therefore probable that the nature of the staple food determines the preponderance of one form over the other in the different areas of cultivation of this crop, both within India and also globally. Table 3: Distribution of diploid and triploid taros in different regions in India (% of total) South India Central India North India Diploid Ploidy level 33.37 11.29 1.67 North-East India 3.76 Triploid 13.85 14.64 7.11 14.18 Chromosomal variations Karyotype analysis revealed some structural differences in the chromosomes of many accessions. Chromosome size has not undergone noticeable change, but a considerable degree of heterogeneity exists in regard to the distribution of various chromosome types, i.e., m, sm, st and t-types (Sreekumari, 1993). Comparative performance of diploids and triploids Initial observations for yield suggested that triploids were superior to diploids in several characters. To confirm this, field experiments were conducted. It was established that triploids in general differ significantly from diploids for such characters as plant height, tillering, habit, number and size of leaves, and corm and cormel yield. This implies that for selecting high yielding types in taro, it is desirable to consider the triploids rather than diploids (Sreekumari and Thankamma Pillai 1994). The same was found to be true in cassava, which showed significant increase in tuber yield and starch content in artificially produced triploids (Sreekumari et al., 1999). Flowering Flowering was scarce, irregular and seasonal. However, frequency of flowering was higher among diploids, starting in the middle of June and lasting to the middle of September (Sreekumari and Thankamma Pillai, 1994). The inflorescence of diploid and triploid plants could be distinguished easily by the size and length of sterile appendage, both larger in triploids. Diploids were fertile but natural seed set was observed only rarely. The cause of sterility of the triploids was studied in detail (Sreekumari and Mathew, 1993). Tolerance to Phytophthora colocasiae Taro leaf blight caused by Phytophthora colocasiae can be a serious disease of taro in India. Its occurrence is correlated with weather conditions. It usually is not a threat to taro cultivation. However, in extended periods of rainfall and high humidity the disease spreads, causing considerable damage. The majority of the accessions were tolerant to leaf blight but none was found to be resistant. Agro-economic evaluation and variety release Elite cultivars were selected from the germplasm collections by conducting field evaluation trials. The most desirable ones, based on specific characteristics (early maturity, ideal plant type, good cormel shape, disease tolerance, good cooking quality etc.), are subjected to further evaluation. The following set procedure is followed for the release of elite varieties for general cultivation: 1. Identification of desired type through germplasm evaluation 2. Unreplicated row trial (20-30 plants per row) 3. Replicated row trial (20-30 plants per row per replication) 4. Preliminary yield trial (RBD, 3 replications) 5. Advanced yield trial (RBD, 3 replications, 2seasons) 6. On-farm trial within the state (10 locations, 2 seasons, local variety as check) 7. Approval from Scientific Research Committee (SRC), CTCRI (appropriate name given to the variety) 8. Submission to the State Variety Release Committee 9. Approval by the Committee 10.Multiplication to generate sufficient planting material 11. Official release of the variety for general cultivation within the state For release of the variety at the national level, uniform regional trials are conducted in different zones (usually undertaken by All India Coordinated Research Project on Tuber Crops) and approval is then obtained from the Central Variety Release Committee. It generally takes 5-6 years for an elite line to reach the variety release stage. Based on the procedure for state level release, four taro varieties have been released in the country. They are two high yielding, good cooking quality triploid selections, “Sree Reshmi” and “Sree Pallavi” from the CTCRI, Trivandrum; one blight tolerant third taro symposium 87 variety, “Muktakeshi”, from the Regional Centre of CTCRI at Bhuvaneswar, India; and another high yielding variety, “Kovur”, from Andhra Pradesh Agricultural University. The released varieties are triploids, which seem to be preferred for their smaller size and better keeping quality (Velayudhan et al., 1991). Conclusion The 424 indigenous edible taro accessions maintained and evaluated at CTCRI are a representative collection for the country. In addition, NBPGR also maintains germplasm collections. However, the main thrust of NBPGR is on exploration and collecting, whereas CTCRI is the only research institute in India devoted to root crop utilization and improvement, in addition to collecting, conservation, cataloguing and evaluation. Significant genetic variability was found among these collections. The occurrence of diploids and natural triploids in almost equal proportion, the superiority of triploids for several characters, especially tuber yield, and the significance of fertile diploids for the production of true seeds and seedling progeny are all important features of the Indian taro collections. Altogether four superior taro selections have been released in India for general cultivation. However, lack of exotic collections from a different genetic base is a major constraint to taro breeding in India. The procurement of pathogen-free core collections from the South Pacific countries would be valuable for a better implementation of the genetic improvement programme of taro in India. References Lebot, V., Hartati, S., Hue, N.T., Viet, N.V., Nghia, N.H., Okpul, T., Paradales, J., Prana, M.S., Prana, T.K., Thongjiem, M., Krieke, C.M., Van Eck, H., Yap, T.C. and Ivancic, A. 2000. Genetic variation of taro (Colocasia esculenta) in South East Asia and Oceania. In: Nakatani, M. and Komaki, K. (eds). Proceedings of the Twelfth Symposium of the International Society for Tropical Root Crops: Potential of root crops for food and industrial resources. Tsukuba, Japan, 10–16 September 2000. ISTRC. Sreekumari, M.T. 1993. Cytomorphological and cytogenetic studies in edible aroids. PhD thesis. University of Kerala, Trivandrum. Sreekumari, M.T. and Mathew, P.M. 1991. Distribution of diploid and triploid taro in India. Journal of Root Crops 18(2):132–133. Sreekumari, M.T. and Mathew, P.M. 1993. Meiosis in triploid taro (Colocasia esculenta (L.) Schott). Journal of Cytology and Genetics 28:7–11. Sreekumari, M.T. and Thankamma Pillai, P.K. 1994. Breeding barriers in taro (Colocasia esculenta (L.) Schott). Journal of Root Crops. 20(1):60–63. Sreekumari, M.T., Jos, J.S. and Nair, S.G. 1999. ‘Sree Harsha’: A superior triploid hybrid in cassava. Euphytica 106:1– 6. Unnikrishnan, M., Thankamma Pillai, P.K. and Vasudevan, K. 1988. Evaluation of genetic resources of taro (Colocasia esculenta (L.) Schott). Journal of Root Crops 14(1):27–30. Unnikrishnan, M., Thankamma Pillai, P.K., Vasudevan, K., Nayar, G.G., Jos, J.S., Thankappan, M. and Palaniswami, M.S. 1987. Genetic resources of taro. Technical Bulletin Series 8, Central Tuber Crops Research Institute, Trivandrum, India. Velayudhan, K.C., Muralidharan, V.K., Amalraj, V.A., Thomas, T.A. and Rana, R.S. 1991. Studies on the morphology, distribution and classification of an indigenous collection of taro. Journal of Root Crops 17(2):118–129. Zeven, A.C. 1980. Polyploidy and domestication: The origin and survival of polyploids in cytotype mixtures. p. 385– 408. In: Lewis, W.H. (ed.). Polyploidy: Biological relevance. Plenum Press, New York. Zhang, G. and Zhang, D. 1990. The relationship between geographic distribution and ploidy level of taro, Colocasia esculenta. Euphytica 47:25–27. 88 third taro symposium Theme One Paper 1.8 Analysis of genetic diversity in taro in China D. Shen, D.W. Zhu, X.X. Li and J.P. Song The Institute of Vegetables and Flowers, Chinese Academy of Agricultural Sciences, Zhongguancun South Street 12, Beijing 100081, China Introduction Taro (Colocasia esculenta (L.) Schott) is one of the major starchy food plants of the world. Not just the corm, but also the leaf, petiole, and even the flowers are eaten as vegetables in China. Yunnan province, which lies in the southwest of China, is the main centre of taro diversity and production in China. Kuruvilla and Singh (1981) first used protein electrophoresis to study wild and cultivated taro in India. Tanimoto and Matsumoto (1986) analyzed the diversity of the POD (peroxide) and EST (esterase) isozymes among Japanese taro varieties. Lebot and Aradhya (1991) analyzed 7 isozyme systems for over 2,000 accessions collected from more than 20 countries, but only one accession from China was included. Therefore, genetic diversity studies of Chinese material are urgently needed to study the history, evolution and classification of the crop in China and to develop strategies for collecting and conservation of its genetic resources. 1. Materials and methods 1.1Materials 28 taro accessions were collected from 14 counties and cities in Yunnan province. Their names, origins, use, and some botanical characters are listed in Table 1, which is based on data from the Kunming Institute of Botany, Chinese Academy of Sciences. 1.2Isozymes Root tips were used to extract peroxidase (POD) and cytochrome oxidase (COD), and the first new leaves were used for superoxide dismutase (SOD), polyphenol oxidase (PPOD) and esterase (EST). Isozyme extracts were prepared using a modified Bouquets buffer. Polyacrylamide gel electrophoresis was employed. The extraction buffer, gel concentration and composition, and dyeing method are described by Li et al. (1998). 1.3DNA markers Following the method of Colosi and Schaal (1993), 0.1 g of freeze-dried taro leaves were ground in 5 ml microtubes. Immediately after grinding, genomic DNA was extracted in SDS extraction buffer using the phenol-chloroform method, as described by Qi et al. (1995). RAPDs: DNA/EcoR+Hind provided by Huamei Biotechnology Company were used to produce random amplification products. The PCR reaction mixtures were incubated in a PTC-200 Thermal Controller (MJ Research, Inc.) programmed for 40 cycles at 94°C for 2 min, 92°C for 20 seconds, 36°C for 40 seconds, 72°C for 80 seconds, and 72°C for 2 minutes after the last cycle. Amplified products were separated by 1% agarose gel electrophoresis. The gels were stained with ethidium bromide and photographed with black and white film under UV light. AFLP: The AFLP technique is described by Shen (2000). The reaction mixtures were assayed on 6% polyacrylamide gel electrophoresis with a Li-Cor IR2 DNA automated sequencer 4200L. 1.4Data analysis For data analysis, presence of a band was scored as 1 and absence as 0. Cluster analysis was carried out using the Unweighted Pair-Group Average with the Statistica software. Unclear bands were omitted from analysis. 2. Results 2.1Isozymes The EST zymograms of the 28 taro accessions can be seen in Figure 1. The bands of EST were the clearest of the five isozymes. Two main zones of EST activity could be seen. The bands in the anodal zone were stronger than those in the cathodal zone. In total 20 EST bands were observed. No band was common to all the accessions. There were 20 EST zymograms among the 28 accessions. third taro symposium 89 Figure 1. EST zymogram of 28 taro accessions Most active bands of the POD system were distributed at the cathodal end (Figure 2). 17 bands were obtained from all accessions. Only 1 band was common to all accessions. 22 zymograms were identified. 8 zymograms were shared by 2 or 3 accessions. It was noteworthy that one sample (97126) was the only one showing 6 bands in the anodal zone. Figure 2. POD zymogram of 28 taro accessions Compared with other four isozymes, SOD showed much less diversity (Figure 3). In the cathodal zone, there were two weak active zones. 4 bands were common to all accessions. One zymogram included only two accessions (98008 and 98087). The 28 accessions grouped in just three zymograms. Figure 3. SOD zymogram of 28 taro accessions The PPOD system revealed 19 bands altogether, distributed evenly from the anode to cathode. There were 22 zymograms and no common bands (Figure 4). 90 third taro symposium Figure 4. PPOD zymogram of 28 taro accessions The COD system gave similar results to POD, though the bands had different migration rate. There were 11 bands and 20 zymograms among all accessions (Figure 5). Figure 5. COD zymogram of 28 taro accessions 2.2RAPD 19 polymorphic primers were identified among 100 random primers. They are listed in Table 2. Figure 6-10 shows the RAPD results obtained with random primers OPN07, OPO01, OPQ05 and OPQ20. 183 amplification products were obtained, among which 161 were polymorphic. Polymorphic percentage was high as 88.0%. There were 9.6 amplification products per primer on average. Table 3 summarizes the results. Table 2: The primers used and the number of their polymorphic sites Total sites Polymorphic sites OPN07 Primer No. CAGCCCAGAG 16 16 100 OPN09 TGCCGGCTTG 9 8 88.9 OPN10 ACAACTGGGG 6 5 83.3 OPN14 TCGTGCGGGT 10 9 90.0 OPO01 GGCACGTAAG 12 10 83.3 OPO03 CTGTTGCTAC 15 15 100 OPO07 CAGCACTGAC 11 10 90.9 OPO18 CTCGCTATCC 11 10 90.9 OPO19 GGTGCACGTT 12 12 100 OPP02 TCGGCACGCA 6 3 50.0 OPP03 CTGATACGCC 7 7 100 OPP14 CCAGCCGAAC 7 7 100 OPP15 GGAAGCCAAC 4 3 75.0 OPP16 CCAAGCTGCC 11 10 90.9 OPP20 GACCCTAGTC 10 9 90.0 OPQ04 AGTGCGCTGA 7 6 85.7 OPQ05 CCGCGTCTTG 11 9 81.8 OPQ06 GAGCGCCTTG 5 2 40.0 OPQ20 TCGCCCAGTC 13 10 76.9 183 161 88.0 Total Sequence Polymorphic percentage third taro symposium 91 Figure 6. The electrophoretic pattern of RAPD with random primers OPN07, OPO01, OPQ05, OPQ20 The accession numbers from left to right are as follows: 98001, 98003, 98006,98008, 98019, 98024, 98025, 98026, 98030, 98033, 98040, 98042, 98044, 98053, Marker, 98057, 98061, 98069, 98070, 98075, 98083, 98084, 98087, 98089, 98108, 98114, 98121, 97122, 97126 2.3AFLP The data used for cluster analysis were from the primer combinations listed in Table 3. 184 amplification products were identified, 169 polymorphic (91.8%). Table 3: The base sequences of 3 primer combinations of AFLP in taro Primer code Sequence M47 5’- GATGAGTCCTGAGTAACAA- 3’ M49 5’- GATGAGTCCTGAGTAACAG- 3’ M59 5’- GATGAGTCCTGAGTAACTA- 3’ M60 5’- GATGAGTCCTGAGTAACTC- 3’ E-AA 5’- GACTGCGTACCAATTCAA- 3’ E-TT 5’- GACTGCGTACCAATTCTT- 3’ E-TG 5’- GACTGCGTACCAATTCTG- 3’ 2.4Comparative analysis of the results of isozyme, RAPD and AFLP Some comparative data on the results of isozme, RAPD and AFLP analysis of the 28 taro accessions are summarized in Table 4. The polymorphic percentages of the three methods were all higher than 85%, with isozyme the highest, then AFLP and RAPD that giving the lowest percentage of polymorphic products. AFLP was the most efficient method, with 61.3 amplification products per primer, even with some uncertain bands being omitted. Table 4: Summary data on the three genetic diversity methods Methods Number of isozymes or primers Number of total products Average number of products per isozyme or primer Number of polymorphic products Polymorphic percentage (%) Isozymes 5 86 17.2 81 94 RAPD 19 183 9.6 161 88.0 AFLP 3 184 61.3 169 91.8 The results of cluster analysis show that the 28 accessions are divided into two main groups. The first group includes only accessions 98087 and 98008, which are quite distinct from all other accessions. Accession 97126 also showed significant differences in isozyme composition, mainly due to POD and COD, but not in AFLP and RAPD. As POD and COD are both isolated from roots, it might be that this accession was not particularly distinct at the DNA level, but showed differences in enzyme expression due to some environmental or other factor acting on the roots. Comparing the RAPD diagram to the isozyme results, it was found that there were 10 accessions for which there were significant discrepancies (97126, 98069, 98057, 98024, 98006, 98114, 98061, 98121, 98083 and 98070). The relationships among the other accessions were consistent. Comparing the three diagrams, the relationships of four pairs of accessions were consistent, i.e. 98001 and 98053, 98030 and 98033, 98025 and 98026, 98089 and 97122. 92 third taro symposium 3. Discussion The results of this study showed significant genetic diversity among 28 taro accessions collected in the Yunnan province of China using isozyme, RAPD and AFLP markers. The materials were collected from a wide range of environments and communities, and are probably representative of the genetic diversity of taro in the province. Compared with some other crops, many traditional taro landraces are still cultivated and used in Yunnan because of their good adaptation to marginal environments and minimal management. But taro genetic diversity is nevertheless facing the danger of erosion due to rapid and continuing socio-economic, policy and land-use changes. Being a vegetative propagation crop, taro is relatively difficult to conserve. Some collected accessions have been lost, especially many wild taros. In situ conservation should be investigated in Yunnan province. Three different methods were used to study genetic diversity in taro, with somewhat different results. Isozymes are proteins and results could be affected by changes in gene expression during the course of growth and development, unlike polymorphism at the DNA level, AFLP and RAPD. The fragments of genomic DNA involved in the RAPD and AFLP methods were different. The length of DNA fragments assayed by RAPD was 300-2000 bp, while those assayed by AFLP were mostly 50-500 bp. Therefore, the AFLP method can detect more variation (Tohme et al., 1996). Two of the accessions, which were from taro species other than Colocasia esculenta – 98087 (C. gigantea) and 98008 (Alocasia macrorrhiza) – were found to be genetically close to each other, and distinct from the true taro accessions (C. esculenta). This supports the results of Yoshino (1994) and Yoshino et al. (1998) which indicated that C. gigantea might be more closely related to A. macrorrhiza than to C. esculenta. The experiment of Yoshino et al. (1998) showed that C. gigantea might be the result of natural hybridization between A. macrorrhiza and C. esculenta taro, but more research is needed to clarify this. References Colosi, J.C. and Schaal, B.A. 1993. Tissue grinding with ball bearing and vortex mixer for DNA extraction. Nucleic Acids Research 21:1051–1052. Kuruvilla, K.M. and Singh, A. 1981. Karyotypic and electrophoretic studies on taro and its origin. Euphytica 30:405– 413. Lebot, V. and Arakhya, K.M. 1991. Isozyme variation in taro (Colocasia esculenta) from Asia and Oceania. Euphytica 56:55–66. Li, X.X., Shen, D., Zhu, D.W. et al. 1998. Analysis of correlation between ethnobotany and molecular diversity of taro (Colocasia esculenta L.) in China. In: Zhu, D., Eyzaguirre, P.B., Zhou, M., Sears, L. and Liu, G. (eds). Ethnobotany and genetic diversity of Asian taro: Focus on China (Proceedings of the Symposium on the Ethnobotanical and Genetic Study of Taro in China: Approaches for the Conservation and Use of Taro Genetic Resources, Shangdong, China, 10–12 November). IPGRI–CSHS, Rome. Qi, X.Q., Zhu, D.W. et al. 1995. RAPD analysis of genomic DNA of selfing lines in Chinese cabbage and purple rape. Acta Horticulturae Sinica 22(3):256–262. Shen, D. 2000. The analysis of genetic diversity of germplasm resources in taro in Yunnan province. Dissertation. Graduate School in Chinese Academy of Agricultural Sciences, Beijing. Tanimoto, T. and Matsumoto, T. 1986. Variation of morphological characters and isozyme patterns in Japanese cultivars in Colocasia. Japanese Journal of Breeding 2 (36):100–111. Tohme, J., Gonzalez, D.O., Beebe, S. and Duque, M.C. 1996. AFLP analysis of gene pools of a wild bean core collection. Crop Science 36:1375–1384. Yoshino, H. 1994. Studies on the phylogenetic differentiation in taro, Colocasia esculenta Schott. PhD thesis. Kyoto University, Japan. 58 p. Yoshino, H., Toshinori, O. and Makoto, T. 1998. An artificial intergeneric hybrid between Colocasia esculenta (L.) Schott and Alocasia macrorrhiza (L.) G. Don. Monocots II. Paper given at the Second International Conference on the Comparative Biology of the Monocotyledons and Third International Symposium on Grass Systematics and Evolution, Sydney, 27 September–2 October 1998. third taro symposium 93 third taro symposium 98040 98042 98044 98053 98057 98061 98069 98070 98075 98083 98084 98087 98089 98108 98114 98121 97122 97126 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 Ku biu (Hongyu) Bi ge a na (Ziyu) Bu le na (Heiyu) Huo pe (Ziyu) Byong ma byong na (Ziyu) Ziyu Dishuiyu Luyu Baiyu Caiwen Yeyu Rentouyu Hongyu Hongyayu Chushuiyu Tuanyu Duoya Lugeng Yeyu Yeyu Xiaoluyu Honggeng Ziyu Baigeng Ziyu Kua mo mei (Baiyu) Baiyu Heigengyu Yeyu Jiangcheng County Jiangcheng County Jinghong City Jinghong City Jinghong City Gengma County Gengma County Gengma County Gengma County Zhenkang County Yongde County Shidian County Yingjiang County Yingjiang County Yingjiang County Yingjiang County Tengchong County Baoshan City Fugong County Fugong County Fugong County Lushui County Lushui County Yunlong County Maguan County Maguan County Yimen County Baoshan City C C C C C C C C C W C C C C M M M C C C C C C W C W C C Hani Hani Jinuo Dai Hani Han Han Wa Wa -- Han Han Han Han -- -- -- Han Lisu Lisu Nu Lisu Lisu -- Han -- Han Han Edible Edible Edible Edible Edible Edible Edible Edible Edible No use Edible Edible Edible Edible Edible No use -- Edible Edible Edible Edible Edible Edible No use Edible No use Edible Edible Use Cormel Cormel Cormel/corm Cormel Cormel/corm Cormel Peduncle Cormel Cormel -- Cormel/corm Inflorescence, cormel Cormel, corm Cormel, corm Cormel, corm -- -- Cormel Cormel/corm Peduncle Cormel Cormel Cormel -- Petiole -- Cormel Multiple corm Part of plant used Farmland,dry land Farmland,dry land Farmland,swidden Farmland Swidden Dry land Farmland Farmland Farmland Forest Farmland,dry land Farmland Swidden Swidden Swidden Farmland Gutter edge Farmland Dry land Dry land Farmland,dry land Farmland,dry land Farmland,dry land Gutter edge Home garden Forest/gutter edge Farmland,dry land Farmland,dry land Ecosystem W – Wild, species found in natural habitats without human intervention. M – Managed, species affected by partial human intervention, thus subject to a degree of farmers selection. C – Cultivated, species grown in human-managed habitats, propagation is entirely subject to and dependent upon farmers’ selection. 98033 11 98030 98026 98025 98024 98019 10 9 8 7 6 5 4 Laiyuhe Maguan Yeyu 98006 3 98008 Yimen Qingyu Baoshan Dayutou 98003 94 2 Place of origin Source 98001 Domestic name Nationality 1 No. Sample no. Table 1: The background information of 28 taro materials No No No No No No No No No Yes No No No No No Yes Yes No No Yes No No No Yes No Yes No No Running stem Light purple Green, purple spot Purplish black Light purple Purplish black Purplish black Greyish white Green Green, purple sheath Green, purple sheath Light purple Light red Green, red spot Greenish white Green, purple spot -- Greenish white Light purple Purple Green Greenish white Purplish black Light purple Greyish white Light purple Green, purple spot Greenish white Color of lower part of petiole Purple Dark purple Purplish black Dark purple Purplish black Light red Greyish white Green Geenish white Light red Geenish white Dark red Geenish white Green, red spot Light red Light red -- Light red Greenish white Greenish white Light red Light red Purplish black Purple Greyish white Greenish white Greenish white Greenish white Color of upper part of petiole Dark purple Dark purple Dark purple Dark purple Purplish red Dark red Greyish white Green Greenish white Dark red Light red Dark red Red spot Light red Light red Dark purple -- Purple Greenish white Light red Light red Light red Purplish black Dark purple Greyish white Light red Light red Light red Color of the middle of leaf Purplish red Purple Purple Purple Purplish red Purplish red Greyish white Red spot Greenish white Purplish red Greenish white Purplish red Red spot Purplish red Red spot Purple -- Greenish white Greenish white Purple Greenish white Greenish white Purple Purplish red Greyish white Purplish red Greenish white Purplish red Color of the middle of leaf Purple Purple Purple Purple Purple Purple Greyish white Greenish white Greenish white Purple Greenish white Purple Purple Purple Purple Purple -- Greenish white Purple Purple Greenish white Greenish white Purple Purple Greyish white Purple Greenish white Purple Color of leaf edge Theme Two Abstracts Theme 2: Pests Diseases and Theme 2 : Organismes nuisibles et maladies Characterisation of taro viruses and the development of diagnostic tests Caractérisation des virus du taro et mise au point de tests diagnostiques R.M. Harding, P.A. Revil, G.J. Hafner, I.Yang, M.K. Maino, L.C. Devitt, M.L. Dowling and J.L. Dale R.M. Harding, P.A. Revil, G.J. Hafner, I.Yang, M.K. Maino, L.C. Devitt, M.L. Dowling et J.L. Dale The movement of taro germplasm among Pacific Island countries for breeding and other purposes is currently restricted due to the presence of viral diseases. We have been characterising the viruses infecting taro to allow the development of sensitive, specific and reliable diagnostic tests for use in an indexing program. Prior to the commencement of our research, four viruses were reported from taro, namely dasheen mosaic potyvirus (DsMV), taro bacilliform virus (TaBV) (a putative badnavirus) and two putative rhabdoviruses, colocasia bobone disease virus (CBDV) and taro vein chlorosis virus (TaVCV). Of these viruses, only DsMV had been characterised. During this project we have (i) investigated the sequence variability in the coat protein-coding region of DsMV isolates from throughout the Pacific and used this data to develop both a PCR and serology-based diagnostic, (ii) characterised a PNG isolate of TaBV and showed it was definitively a badnavirus, (iii) investigated the sequence variability in TaBV isolates from throughout the Pacific and developed a PCR-based diagnostic test, (iv) detected and partially characterised a previously undescribed reovirus from taro and developed a PCR-based diagnostic test, (v) partially characterised the genome of TaVCV and developed a PCR-based diagnostic and (vi) partially characterised the genome of CBDV. Using these diagnostic tests, we have surveyed ten Pacific Island countries for the presence of viruses. À l’heure actuelle, la circulation de matériel génétique du taro entre pays océaniens à des fins de sélection et autres est entravée par la présence de viroses. Nous avons caractérisé les virus qui infectent le taro afin de pouvoir mettre au point des tests diagnostiques sensibles, spécifiques et fiables, utilisables dans un programme d’indexage. Avant d’entreprendre notre recherche, quatre virus du taro avaient été signalés : le potyvirus de la mosaïque du taro (DsMV), le virus bacilliforme du taro (TaBV) (supposé être un badnavirus), ainsi que deux présumés rhabdovirus : le virus de la maladie de Bobone du taro (CBDV) et celui de la chlorose des nervures du taro (TaVCV). Parmi ces virus, seul DsMV a été caractérisé. Au cours du projet, nous avons : 1) étudié la variabilité de la séquence dans la région du codage des protéines sur des isolats de DsMV provenant de toute l’Océanie, et nous avons utilisé ces données pour mettre au point un diagnostic fondé sur l’amplification en chaîne par polymérase (PCR) et la sérologie ; 2) caractérisé un isolat de TaBV provenant de PapouasieNouvelle-Guinée, et montré qu’il s’agissait assurément d’un badnavirus ; 3) étudié la variabilité de la séquence d’isolats de TaBV provenant de toute l’Océanie et mis au point un test diagnostique fondé sur l’amplification en chaîne par polymérisation ; 4) détecté et caractérisé en partie un rétrovirus du taro qui n’avait jamais été décrit auparavant, et mis au point un test diagnostique fondé sur l’amplification en chaîne par polymérisation ; 5) partiellement caractérisé le génome de TaVCV, et mis au point un diagnostic fondé sur l’amplification en chaîne par polymérisation, et 6) partiellement caractérisé le génome de CBDV. Nous avons mené une enquête, dans dix pays océaniens, pour déceler la présence de ces virus à l’aide de ces tests diagnostiques. The potential of the fungus Metarhizium anisopliae as a biological control agent for taro beetles Le champignon Metarhizium anisopliae, agent potentiel de lutte biologique contre les coléoptères du taro R.T Masamdu and N.A. Simbiken R.T Masamdu et N.A. Simbiken Application of the fungus Metarhizium anisopliae under grass mulch around taro to control taro beetles (Papuana woodlarkiana, Coleoptera: Scarabaeidae) led to significantly lower damage and increased yield. The potential of the fungus as a biological control agent is discussed. L’application de champignon Metarhizium anisopliae sous le paillis herbeux entourant le taro pour lutter contre les coléoptères du taro ( Papuana woodlarkiana, Coleoptera: Scarabaeidae) a permis de réduire considérablement les dégâts et d’augmenter le rendement. Les possibilités de recours à ce champignon comme agent de lutte biologique sont décrites. third taro symposium 95 The biology of Phytophthora colocasiae and implications for its management and control La biologie de Phytophthora colocasiae et ses effets sur les stratégies de gestion et de lutte R.A. Fullerton and J.L. Tyson R. A. Fullerton et J. L. Tyson Defining features of the life cycle of Phytophthora colocasiae include: capacity for endemic survival during extended dry periods, the ability to sporulate and infect within the same night, sporangia germinating to produce multiple zoospores under cool conditions, a very short life cycle (less than 3 days), and the ability to progress from endemicity to a destructive epidemic within days of onset of wet weather. These characteristics determine, to a large degree, the efficacy of various strategies available for its control. Field sanitation (through the removal of spotted leaves) is effective in small plots at the endemic phase but is less effective, and largely impractical, in large plots under epidemic conditions. Fungicidal control will be most effective during the endemic phase when inoculum levels are relatively low, and between-plant spread is limited, but less effective in the epidemic phase. The use of tolerant cultivars with horizontal resistance is the only viable long term solution. Where climatic conditions favour frequent disease epidemics, visual assessments in the field are relatively effective in detecting resistant genotypes in populations of progeny. Under conditions of high temperatures and dry weather, field assessments can be supplemented by laboratory assays using inoculated leaf discs and defined incubation conditions. Field evaluation of elite selections in different climatic zones is essential to ensure the reliability of horizontal resistance in epidemic prone localities. Parmi les éléments caractéristiques du cycle de vie de Phytophthora colocasiae, on retiendra sa capacité de survie endémique pendant de longues périodes de sécheresse, sa capacité à sporuler et à infecter les végétaux au cours de la même nuit, la germination de ses sporanges capables de produire des zoospores multiples en période fraîche, son cycle de vie très court (inférieur à 3 jours) et sa capacité à passer d’un état endémique à un comportement épidémique destructeur quelques jours à peine après le début d’une période humide. Ces caractéristiques conditionnent en grande partie l’efficacité des stratégies mises en œuvre pour lutter contre la maladie. Le nettoyage des cultures (ramassage des feuilles tâchées) est efficace sur de petites superficies, pendant la phase endémique, mais inefficace et malaisé lorsque les cultures sont plus étendues et que la maladie est entrée dans une phase épidémique. L’usage de fongicides est particulièrement efficace au stade endémique, lorsque les niveaux d’inoculum sont relativement faibles et que la propagation entre plantes est limitée. Les fongicides perdent de leur efficacité en période d’épidémie. L’utilisation de cultivars tolérants présentant une résistance horizontale est la seule solution viable à long terme. Lorsque les conditions climatiques sont propices au déclenchement de flambées épidémiques de la maladie, les inspections visuelles dans les cultures sont assez efficaces pour détecter les génotypes résistants au sein de différentes populations d’une progénie. Lorsque le temps est chaud et sec, on peut compléter les inspections visuelles par des tests en laboratoire sur des disques foliaires inoculés placés en incubation selon une procédure prédéfinie. Les inspections sur le terrain des variétés élites sont essentielles pour s’assurer de la fiabilité de la résistance horizontale dans les régions sensibles aux épidémies. Current status of research on Rhizoglyphus mites associated with taro État actuel des recherches consacrées aux acariens du taro Rhizoglyphus Zhi-Qiang Zhang, Qianghai Fan, N. A. Martin and Sada Nand Lal Zhi-Qiang Zhang, Qianghai Fan, N.A. Martin et Sada Nand Lal This project on taro mites was initiated by Secretariat of the Pacific Community to facilitate the export of taro from Fiji to New Zealand. Identification of mites intercepted from taro originating from Fiji and also collected in the field shows that three species of Rhizoglyphus, three species Schwiebia and one species of Tyrophagus are associated with roots of taro. Rhizoglyphus minutus is by far the most common species on taro. Its quarantine risk is discussed and a pest risk assessment is prepared for this species. Ce projet de recherche sur les acariens du taro a été lancé à l’initiative du Secrétariat général de la Communauté du Pacifique pour faciliter les exportations de taro des Îles Fidji vers la Nouvelle-Zélande. L’identification des acariens décelés sur des taros en provenance de Fidji et prélevés sur le terrain a permis de déterminer que trois espèces de Rhizoglyphus, trois espèces de Schwiebia et une espèce de Tyrophagus sont présentes sur les racines de taro. Rhizoglyphus minutus est de loin l’espèce la plus commune. Les risques phytosanitaires que cette espèce est susceptible de présenter font actuellement l’objet de recherches. Une évaluation du risque phytosanitaire est en cours. 96 third taro symposium Developing interactive diagnostic support tools for tropical root crops V. dR. Amante and G. A. Norton Information and communication technology (ICT) offers exciting new possibilities for providing diagnostic support tools for farmers, advisors and others interested in tropical root crops. A CD-ROM that provides diagnostic support for those attempting to diagnose disorders in sweetpotato crops has recently been released. This CD-ROM has been funded by the Australian Centre for International Agricultural Research (ACIAR) and has involved collaboration among The University of Queensland, The International Potato Centre (CIP) Office in Indonesia, and PhilRootcrops in the Philippines. The process of developing this diagnostic support tool is described. Four major steps are involved: (1) collection of information in text and graphic form, (2) development of fact sheets (3) construction of the diagnostic key and (4) technical and field testing. Construction of the key is discussed in detail, including the development of a list of Possible Causes to be included in the key, formulation of Possible Observations and states that describe the sick or damaged plant, and scoring Possible Causes against Possible Observations. Field testing is an important part of developing computerassisted tools, and the response to a prototype by users from the Philippines, Indonesia and Africa, is described. The possibility of developing a similar diagnostic key for taro and yam problems in Asia, Africa and the Pacific regions is discussed. Élaboration d’outils interactifs d’aide au diagnostic applicables aux légumesracines des régions tropicales V. dR. Amante et G. A. Norton Les technologies de l’information et de la communication ouvrent de nouvelles perspectives prometteuses en matière d’outils d’aide au diagnostic pour les cultivateurs, les conseillers et les autres parties intéressées par les légumes-racines des régions tropicales. Récemment a été publié un cd-rom mettant une aide au diagnostic à la disposition de ceux qui tentent d’identifier les problèmes affectant les patates douces. La production de ce cd-rom s’est faite grâce au soutien financier du Centre australien pour la recherche agricole internationale (ACIAR) et grâce à une collaboration entre l’Université du Queensland, le Bureau indonésien du Centre international de la pomme de terre (CIP), et le Centre de recherche et de formation sur les légumes-racines des Philippines (PhilRootcrops). Cet exposé décrit le processus d’élaboration de cet outil d’aide au diagnostic tout au long des quatre grandes étapes : 1) collecte des informations sous forme de textes ou de graphiques, 2) rédaction de fiches techniques, 3) construction de la grille de diagnostic, et 4) essais techniques et en plein champ. Cet exposé entre dans les détails de la mise au point de la grille, notamment l’élaboration d’une liste de Causes possibles à inclure dans la grille, la formulation d’Observations possibles d’états qui décrivent la plante malade ou endommagée, et la mise en parallèle des Causes possibles avec les Observations possibles. Les essais en plein champ sont un volet essentiel de l’élaboration d’un outil assisté par ordinateur, et l’exposé décrit les réactions des utilisateurs de ce prototype aux Philippines, en Indonésie et en Afrique. Cet exposé aborde également la possibilité de mettre au point une grille de diagnostic similaire appliquée aux problèmes affectant le taro et l’igname en Asie, en Afrique et en Océanie. third taro symposium 97 Theme Two Paper 2.1 Characterisation of taro viruses and the development of diagnostic tests R.M. Harding1, P.A. Revil1, G.J. Hafner2, I. Yang1, M.K. Maino3, L.C. Devitt1, M.L. Dowling1 and J.L. Dale1 Plant Biotechnology Program, Science Research Centre, Queensland University of Technology, Brisbane, Australia 2 PANBIO Limited, Windsor, Brisbane 4030, Australia 3 University of Technology, Lae, Papua New Guinea 1 Introduction Taro (Colocasia esculenta) is an important staple food grown throughout many Pacific Island countries and ranks 14th worldwide amongst staple crops. Despite its agronomic significance, taro cultivation has declined over the past thirty years due to pest and disease problems. Viruses are one of the most important pathogens of taro, with some infections resulting in severe yield reductions and plant death. There are reports of four viruses infecting taro, namely Dasheen mosaic virus (DsMV), Colocasia bobone disease virus (CBDV), Taro bacilliform virus (TaBV) and Taro vein chlorosis virus (TaVCV) (Brunt et al., 1990; Pearson et al., 1999). Apart from DsMV, these viruses have been poorly characterised and there is confusion in the literature concerning their distribution and the symptoms associated with infection. DsMV is an important virus affecting taro which has been found wherever taro is grown and infects both the edible and ornamental aroids (Zettler and Hartman, 1986, 1987; Jackson, 1980; Shaw et al., 1979). Infected plants usually display a conspicuous feathery mosaic pattern although cultivars vary considerably in symptom expression. The main effect of virus infection is a reduction in corm size and quality, with yield losses of up to 60% having been reported (Zettler and Hartman, 1986). DsMV has been well characterised, but these studies have been based on isolates originating from the USA and Taiwan. A commercially available ELISA is also available but the cost of this test precludes its widespread use in the Pacific. TaBV is thought to occur in combination with CBDV to cause “alomae” disease (James et al., 1973), although there is still a considerable amount of confusion regarding the etiology of this disease. Alomae disease is considered the most destructive virus disease of taro and has only been reported from the Solomon Islands and PNG (Jackson and Gollifer, 1975; Rodoni et al., 1994). Symptoms include crinkling of young leaves, which fail to develop normally, the presence of thickened veins and lamina, shortening of the petioles and the presence of irregularly shaped outgrowths on the petioles. Infected plants ultimately die due to the development of a systemic necrosis (Rodoni et al., 1994). Infection of taro with CBDV alone is thought to result in the disease known as bobone, a disease only reported from PNG and the Solomon Islands, which is characterised by stunting, leaf distortion and presence of galls on the petioles (Jackson, 1978). In contrast, infection with TaBV alone is thought to result in a range of mild symptoms including stunting, mosaic and down-curling of the leaf blades (Jackson, 1978). Based on sap dips, CBDV appears to be restricted to PNG and the Solomon Islands while TaBV appears to be widely distributed throughout the Pacific (Gollifer et al., 1977). These tests are insensitive and unreliable. Prior to the commencement of this project, neither TaBV nor CBDV had been characterised and no reliable and sensitive diagnostic tests were available for either virus. TaBV was tentatively classified as a badnavirus, primarily based on morphology of virions in sap dips from infected plants and preliminary transmission experiments that indicated that the virus is spread by mealybugs. Brunt et al. (1990) reported that CBDV was possibly a rhabdovirus, as it possessed morphologically characteristic bullet-shaped or bacilliform particles measuring 300-335 x 50-55 nm. A second putative rhabdovirus, TaVCV, causes a distinctive vein chlorosis in diseased taro and is thought to occur in Fiji, Vanuatu, Tuvalu, the Philippines and possibly PNG (Pearson et al., 1999). This virus is presumed to be different from CBDV based on the morphology of the virions in sap dips and the symptoms caused. Like CBDV, this virus has not been characterised and a sensitive and reliable diagnostic test is not available. The presence of taro viruses currently restricts the international movement of taro germplasm. This has serious implications, since many countries are denied access to agronomically elite lines, including selected traditional cultivars and lines produced in breeding programs. To address this problem, we have been characterising the viruses infecting taro to allow the development of sensitive and reliable diagnostic tests. In the short term, such tests will be useful to determine the geographic distribution of the viruses and allow informed decisions to be made by countries regarding the risks of taro importation. In the longer term, the inclusion of such tests for all taro viruses in an indexing scheme will enable the safe exchange of virus-tested taro germplasm between countries. Materials and methods DsMV and TaBV: The sources of virus isolates and protocols used in the characterisation of DsMV and TaBV are outlined in Maino (2003) and Yang et al. (2003a, 2003b), respectively. 98 third taro symposium TaRV: Taro leaves showing a variety of disease symptoms were collected from PNG, Solomon Islands, Vanuatu, Samoa, New Caledonia, French Polynesia, Fiji, New Zealand and Vietnam. DsRNA was purified from leaf and petiole samples using a modification of the method of Choi and Randles (1997). Purified dsRNA was separated using polyacrylamide gel electrophoresis and visualised using ethidium bromide. The single primer amplification technique (SPAT) (Lambden et al., 1992; Attoui et al., 2000) was used to obtain the full-length nucleotide sequence of the viral dsRNA segments. PCR reactions were performed using the Expand™ Long Template PCR system (Roche) according to the manufacturer’s protocol. The amplified products were cloned and sequenced and the sequences were compared to all known viruses on the GenBank database. Sequences were assembled, analysed and compared with other viral sequences using the Lasergene™ (DNAstar) molecular analysis software package. TaVCV: Taro leaves showing vein chlorosis symptoms were collected from Fiji, and virions were purified using modifications of a previously published method (James et al., 1973). Viral RNA was extracted from the partially purified virus preparations and cDNA libraries were generated using the Invitrogen cDNA synthesis protocol. The resulting clones were sequenced and these sequences were screened for similarities to known viral sequences. PCR primers were subsequently designed from those sequences that showed similarity to plant rhabdovirus sequences. RT-PCR was then used to amplify additional viral sequence. In an attempt to obtain N-terminal amino acid sequence, purified viral proteins were separated by polyacrylamide gel electrophoresis and sequenced using Edman degradation. CBDV: Sequences were amplified from a PNG taro sample exhibiting alomae disease symptoms using SPAT (Lambden et al., 1992; Attoui et al., 2000). Clones were screened for similarities to known viral sequences, and primers were designed from those sequences that showed similarity to plant rhabdoviruses. RT-PCR was then used to amplify additional viral sequence. Results DsMV: The coat protein (CP)-coding region of 16 DsMV isolates from PNG, Samoa, Solomon Islands, French Polynesia, New Caledonia and Vietnam were amplified by PCR, cloned and sequenced. Based on the nucleic acid sequences, a reverse transcriptase PCR (RT-PCR)-based diagnostic test was developed which was able to detect a wide range of DsMV isolates including those from Australia, New Zealand, Fiji, French Polynesia, New Caledonia, PNG, Samoa, Solomon Islands and Vanuatu. When the amino acid sequences of the entire CP-coding region were compared with each other and with published DsMV sequences, the maximum variability was 21.9%. When the core region of the CP was analysed, the maximum variability dropped to 6%, indicating that most variability was present in the N terminus of the CP. The sequence of PNG isolate P1 was most similar to all other sequences. Due to the extensive variability over the entire CP-coding region, the core region of the CP of PNG isolate P1 was cloned into a protein expression vector, expressed as a recombinant protein and used as an antigen to generate antiserum in a rabbit. In western blots, the antiserum reacted with the expected size bands of approximately 45-47 kDa in extracts from purified DsMV and from known DsMV-infected plants from PNG, while no bands were observed using healthy plant extracts. The antiserum was subsequently incorporated into an indirect ELISA. This procedure was found to be very sensitive and detected DsMV in sap diluted at least 1:1,000. Using western blot and ELISA formats, the antiserum was able to detect a wide range of DsMV isolates, including those from Australia, New Zealand, Fiji, French Polynesia, New Caledonia, PNG, Samoa, Solomon Islands and Vanuatu. These plants were verified to be infected with DsMV by RT-PCR. TaBV: The complete nucleotide sequence of a PNG isolate of TaBV was determined and comprised 7458 bp. The size and genome organisation of TaBV was similar to that of most other published badnaviruses. The putative amino acid sequence of TaBV open reading frame (ORF) 3 contained motifs that are conserved among badnavirus proteins, including aspartic protease, reverse transcriptase (RT) and ribonuclease H (RNaseH). These results confirm that TaBV is a pararetrovirus of the genus Badnavirus, family Caulimoviridae. We investigated the sequence variability in the putative RT/RNaseH and the C-terminal CP-coding regions from TaBV isolates collected throughout the Pacific. When the RT/RNaseH-coding regions of 22 TaBV isolates from Fiji, French Polynesia, New Caledonia, PNG, Samoa, Solomon Islands and Vanuatu were examined, maximum variability at the nucleotide and amino acid levels were 22.9% and 13.6%, respectively. Within the CP-coding region of 13 TaBV isolates from Fiji, New Caledonia, PNG, Samoa and the Solomon Islands, maximum variability at the nucleotide and amino acid level was 30.7% and 19.5%, respectively. Based on the sequences of the TaBV RT/RNaseH-coding region, we have developed a PCR-based diagnostic test that specifically detects all known TaBV isolates. A sequence showing approximately 50% nucleotide identity to TaBV in the RT/RNaseH-coding region was also detected in all taro plants tested and may represent either an integrated sequence or the genome of an additional badnavirus infecting taro. TaRV: While screening taro germplasm for new viruses using dsRNA analysis, we detected and partially characterised a new reovirus infecting taro, which we have tentatively named taro reovirus (TaRV). This virus appears to have a genome comprising 10 segments of dsRNA. We have obtained the complete genomic sequences of two dsRNA genome segments (S3, S4), while partial sequences have been obtained for another three dsRNA segments (S1, S2 and S10). Comparison of the full-length sequences of S3 and S4 with known reoviruses indicated that the taro reovirus was a new member of the Orzyavirus genus. Based on comparisons with other reoviruses, degenerate primers were designed to conserved regions of S4 to amplify a 1.7 kbp fragment. These primers have been successfully used to detect the reovirus from taro plants collected PNG, Solomon Islands, New Caledonia and Vanuatu. We investigated the sequence third taro symposium 99 variability in the 1.7 kbp amplicon derived from the PNG, Solomon Islands, New Caledonia and Vanuatu TaRV isolates in order to develop a more specific PCR test. A maximum of 16% variability at the nucleotide level was detected in this region among these isolates. Based on these sequences, a specific PCR-based diagnostic assay for this virus has now been developed. TaVCV: Two clones were initially obtained that showed greatest homology to two plant-infecting rhabdoviruses, namely Rice yellow stunt virus (RYSV) and Sonchus yellow net virus (SYNV). Additional sequence was obtained by RT-PCR and we have now characterised approximately two thirds of the TaVCV genome. We have the complete sequence of genes encoding the matrix and glycoproteins, and partial sequence of the L-gene (polymerase). Further, we have identified conserved intergenic sequences while N-terminal sequencing of a virion protein has identified the amino-terminus of the viral glycoprotein. CBDV: We have obtained three clones that show greatest homology to the P3, glycoprotein and L-protein genes of Northern cereal mosaic virus (NCMV), a plant-infecting rhabdovirus. These sequences have no homology to TaVCV, indicating we have identified a second rhabdovirus in taro. This virus may be CBDV, or an as yet unidentified rhabdovirus. Primers have been designed to amplify additional sequence and a PhD student has been appointed to continue this work. Virus survey During 2002, taro virus surveys were conducted in Vanuatu, Samoa, American Samoa, Fiji, PNG, Solomon Islands and New Caledonia. Samples were also provided from Micronesia and the Cook Islands. All samples were indexed for DsMV, TaBV, TaRV and TaVCV using the newly developed molecular-based diagnostic tests. However, since a molecular-based test had not been developed for CDBV, the detection of this virus was based on the presence of typical bobone symptoms. The results of the surveys are presented in Table 1. Discussion We have developed sensitive and reliable diagnostics for four of the five viruses known to infect taro, namely DsMV, TaBV, TaVCV and TaRV. Further, we are in the process of developing such a test for CBDV. Using these tests, we have conducted surveys in various Pacific Islands countries to determine the distribution of these viruses. TaBV was found to be widely distributed throughout the Pacific, which is consistent with previous observations based on electron microscopy. The symptoms associated with TaBV are still unclear because many of the plants testing positive for this virus were also shown to be co-infected with other viruses. This situation is further complicated by the large number of different taro cultivars grown in the Pacific. In general, however, infection of taro with TaBV alone appears to result in only mild symptoms, as previously described (Jackson, 1978). An interesting finding from our research was the detection of a TaBV-like sequence in all taro plants tested, including symptomless plants and those indexed as TaBV-free. Analysis of these sequences amplified from PNG and Vanuatu taro indicated they were only about 50% similar to the nucleotide sequence of TaBV. It is not yet known whether this sequence is integrated into the taro genome or whether it represents the RT/RNaseH-coding region of an additional badnavirus infecting taro. However, the ubiquitous nature of the sequence strongly suggests it is an integrant, similar to that reported for Banana streak virus (BSV) (Geering et al., 2001). Further research will now be required to resolve this issue. Table 1: Distribution of taro viruses throughout the Pacific TaBV DsMV TaRV TaVCV CBDV PNG √ √ √ √ NT Fiji √ √ √ √ NT Vanuatu √ √ √ √ NT Samoa √ √ X X NT American Samoa √ √ X X NT Tonga √ √ X X NT Solomon Islands √ √ √ √ NT New Caledonia √ √ √ NT NT Micronesia √ √ X √ NT Cook Islands √ √ X X NT √ = detected; X = not detected; NT = Not tested, as suitable diagnostic test not available at present (however, taro plants with bobone disease, thought to be caused by CBDV, were only observed in PNG and the Solomon Islands) Analysis of partial genome sequences of TaVCV and CBDV has indicated that these viruses are definitive, yet distinct, rhabdoviruses. Virus surveys revealed that TaVCV was present in several Pacific Island countries. In almost cases, infected plants showed the distinctive vein chlorosis symptoms previously reported (Pearson et al., 1999). Although a sensitive diagnostic test had not been developed for CBDV at the time of the surveys, bobone symptoms, which are thought to be caused by CBDV, were only observed in PNG and the Solomon Islands. However, since the diagnosis of virus infections based on symptoms is unreliable, further testing of all taro plants surveyed will be necessary when a 100 third taro symposium sensitive diagnostic is available. The storage on silica gel at QUT of all samples collected in the surveys will facilitate later testing as new diagnostics are developed. The development of such a test for CBDV is extremely important since the spread of this virus from PNG and the Solomon Islands to other countries would have potentially devastating consequences. The introduction of CBDV would not only result in the spread of bobone disease throughout the region but, due to the widespread distribution of TaBV, may also result in the spread of the lethal alomae disease. It is fortunate that the movement of taro material prior to quarantine restrictions did not disseminate CBDV. However, considering the relatively mild nature of TaBV symptoms compared with those of alomae and bobone disease, it is likely that alomae and bobone diseased material would probably have been considered unsuitable for use as propagules. DsMV was found to be distributed throughout the Pacific. Although typical feathery mosaic symptoms were associated with many infections, the virus was also detected in a large number of plants that appeared healthy, and plants that were infected with other viruses. TaRV is a previously undescribed virus infecting taro. Although detected in several countries throughout the Pacific, the symptoms associated with infection are unknown since the virus was detected in apparently symptomless plants and in plants that were co-infected with other viruses. Therefore, the impact of this virus on taro is unknown. The suite of newly developed diagnostic tests is currently being used to screen a core collection of Pacific and Asian taro germplasm for viruses. This will enable the long-term storage of virus-indexed germplasm, which may be accessed buy users for breeding or other purposes. References Attoui, H., Billoir, F., Cantaloube, J.F., Biagini, F., de Micco, P. and de Lamballerie, X. 2000. Strategies for the sequence determination of viral dsRNA genomes. Journal of Virological Methods 89:147–158. Brunt, A.A., Crabtree, K. and Gibbs, A.J. 1990. Viruses of tropical plants. CAB International, Wallingford, UK. 707 p. Choi, Y.G. and Randles, J.W. 1997. Microgranular cellulose improves dsRNA recovery from plant nucleic acid extracts. Biotechniques 23:610–611. Geering, A.D.W., Olszewski, N.E., Dahal, G., Thomas, J.E. and Lockhart, B.E.L. 2001. Analysis of the distribution and structure of integrated banana streak virus DNA in a range of Musa cultivars. Molecular Plant Pathology 2(4):207– 213. Gollifer, D.E., Jackson, G.V.H., Dabek, A.J., Plumb, R.T. and May, Y.Y. 1977. The occurrence and transmission of viruses of edible aroids in the Solomon Islands and the Southwest Pacific. PANS 23:171–177. Jackson, G.V.H. 1978. Alomae and bobone diseases of taro. Advisory leaflet No. 8, South Pacific Commission, Noumea, New Caledonia. Jackson, G.V.H. 1980. Diseases and pests of taro. South Pacific Commission, Noumea, New Caledonia. 52 p. Jackson, G.V.H. and Gollifer, D.E. 1975. Disease and pest problems of taro (Colocasia esculenta (L.) Schott) in the British Solomon Islands. PANS 21:45–53. James, M., Kenten, R.H. and Woods, R.D. 1973. Virus-like particles associated with two diseases of Colocasia esculenta (L.) Schott in the British Solomon Islands. Journal of General Virology 21:145–153. Lambden, P.R., Cooke, S.J., Caul, E.O. and Clarke, I.N. 1992. Cloning of non-cultivatable human rotavirus by single primer amplification. Journal of Virology 66:1817–1822. Maino, M.K. 2003. The development of a serological-based diagnostic test for Dasheen mosaic potyvirus (DsMV). Pearson, M.N., Jackson, G.V.H., Saelea, J. and Morar, S.G. 1999. MSc thesis, School of Life Sciences, Queensland University of Technology: Evidence for two rhabdoviruses in taro (Colocasia esculenta) in the Pacific region. Australasian Plant Pathology 28:248–253. Rodoni, B.C., Dale, J.L. and Harding, R.M. 1994. Review of alomae disease of taro. Papua New Guinea Journal of Agriculture, Forestry and Fisheries 37:14–18. Shaw, E.D., Plumb, R.T. and Jackson, G.V.H.. 1979. Virus diseases of taro (Colocasia esculenta) and Xanthosoma sp. in Papua New Guinea. Papua New Guinea Agriculture Journal 30:71–97. Yang, I.C., Hafner, G.J., Dale, J.L. and Harding, R.M. 2003a. Genomic characterisation of taro bacilliform virus. Archives of Virology 148:937–949. Yang, I.C., Hafner, G.J., Revill, P.A., Dale, J.L. and Harding, R.M. 2003b. Sequence diversity of South Pacific isolates of taro bacilliform virus and the development of a PCR-based diagnostic test. Archives of Virology 148:1957–1968. Zettler, F.W. and Hartman, R.D. 1986. Dasheen mosaic virus and its control in cultivated aroids. Extension Bulletin No. 233, ASPAC Food Fertilizer Technology Center, Taiwan. 13 p. Zettler, F.W. and Hartman, R.D. 1987. Dasheen mosaic virus as a pathogen of cultivated aroids and control of the virus by tissue culture. Plant Disease 71:958–963. third taro symposium 101 Theme Two Paper 2.2 The potential of the fungus Metarhizium anisopliae as a biological control agent for taro beetles R.T Masamdu1 and N.A. Simbiken2 National Agricultural Research Institute, Lae, Papua New Guinea PNG Coffee Research Institute, P.O. Box 105, Kainantu, Papua New Guinea 1 2 Introduction Taro beetles, from the genera Papuana and Eucopidocaulus (Coleoptera: Scarabaeidae), are one of the main constraints to taro yield and quality in Papua New Guinea and some Pacific Island countries. Adult taro beetles damage the underground corms by chewing and burrowing into them (Figure 1), creating tunnels. In severely damaged plants, the tunnels run together to form large cavities, allowing secondary rots to develop (Thistleton, 1984), resulting in low quality corms for consumption and marketing. Taro beetles also damage related edible aroids, such as Xanthosoma sagittifolia, Alocasia macrorhiza, Cyrtosperma. chamissonis, A. camponatus, other root crops, including sweet potato, yam and Irish potatoes, and even banana stems in a similar manner (Thistleton et al., 1995). Infestation in plants less than two months old may lead to plant death. The beetles rarely feed on corms exposed above the soil. Figure 1: Taro beetle damage to taro corms Though several approaches have been evaluated, no single, sustainable, economical and effective method of taro beetle management has been identified (Masamdu et al., in press). The fungus Metarhizium anisopliae var. anisopliae is a common and widely used biological control agent (Milner, 1992a; Rath, 1992,), and was one of the potential biocontrol agents for taro beetle identified in earlier studies (Theunis et al., 1996). There are several commercial products available of this pathogen for control of target pests, for instance Biogreen and Biocane for management of the plague locust (Locusta migratoria) and cane grubs (Dermolepida, Lepidiota Antitrogus and Rhopaea) in Queensland (Milner, 1992b), respectively. This paper presents the results of further field evaluation of the potential of M. anisopliae as a control agent for taro beetles. Materials and methods Experimental methods Two field trials were set up in Papua New Guinea near the town Lae at two sites, Situm and Bubia between the months of March and December 2000. The most popular local taro variety ‘Numkowec’ was used in the trials. The first trial, at Situm, about 25 east of Lae, had three objectives: 1. to assess the compatibility of M. anisopliae with the insecticide Chlorpyrifos 2. to assess the effectiveness of M. anisopliae in comparison to a known chemical 3. to assess if mulching with grass after application enhanced fungal survival and effectiveness. 102 third taro symposium There were 8 treatments in this trial, replicated three times in a randomized block design. Planting distance was 70 cm between plants and 80 cm between rows. Plots were 4.2 x 4.8 m2 in size and contained 25 test plants each excluding guard rows. The eight treatments were: 1. 2. 3. 4. 5. 6. 7. 8. Control M. anisopliae Chlorpyrifos as Suscon Blue granules Chlorpyrifos as Lorsban EC liquid drench Suscon Blue and M. anisopliae M. anisopliae and Lorsban Lorsban and Suscon Blue Lorsban, Suscon blue and M. anisopliae. Grass mulch was applied in all treatments at planting and the mulch was maintained throughout the cropping season of six months. The second trial, established at Bubia, 20 km northwest of Lae, was to determine the most suitable method of M. anisopliae application and consisted of four treatments: 1. 2. 3. 4. planting hole broadcasting over the surface broadcast and planting hole control The treatments were replicated four times in a latin square design. All treatments had grass mulch over the plots. The plant spacing, plot sizes and density were the same as the first trial. Fungus treatments The M. anisopliae strain TB101 applied in both trials was obtained from BioCare (Australia). In the first trial, 95 gm of fungus product were applied by hand by spreading in the planting hole. The taro sucker was then inserted into the hole and covered with soil and grass mulch. In treatments involving the chemical, this was applied first and the fungus immediately after. In the Bubia trial, the fungus was applied either in the planting hole as just described, broadcast over the plot or both, depending on the treatment. Chemical treatments The two chemical treatments were Chlorpyrifos formulations of Suscon Blue 14% granules (obtained from Ramu Sugar Ltd) and Lorsban 40% EC (AgMark PNG). Chlorpyrifos was applied at a rate of 0.1% a.i. ha in planting holes, before planting. Harvesting Taro beetle damage was assessed in each trial at crop maturity. Plant height, number of leaves and corm weight were recorded, and the following assessed: 1. severity of damage (SOD) 2. percentage of the corm removed by beetle feeding (PCR) 3. number of live and dead adult (with sex) and immature beetles with and without fungal infection. SOD is a qualitative measure for marketability and consumption and can take the following values: 0 = no damage 1 = damaged but saleable 2 = damaged, not saleable but edible 3 = heavy damage, but suitable for animal consumption 4 = heavy damage, completely riddled, not fit for animal consumption PCR is a quantitative estimate of the physical damage done by the beetle, and is recorded using a scale ranging from 0 (no damage) to 13 (more than 90% damage). The physical yield loss is then calculated using the PCR value and the weight of the corms (Appendix 1). Statistical analysis The effect of the fungus on the beetle was measured by the reduction in the amount of damage recorded on the taro third taro symposium 103 corms (PCR) as a relative measure. The estimated damaged weight (EDW) is an estimate of the amount of corm that was eaten by taro beetles. The estimated undamaged weight (EUW) is a estimate of what the corm would have weighed if undamaged. The number of dead beetles with fungal growth was a relative measure and was not statistically analysed as beetles are mobile and a substantial population of infected beetle could have migrated. The amount of damage to taro corms was analysed using analysis of variance (ANOVA) within each trial using the Minitab Statistical Package. Results In the first trial, the M. anisopliae treatment had the least damage (2.79%), but there were no significant differences among the other treatments even when M. anisopliae was combined with chlorpyrifos (Figure 2a). The percentage of marketable corm weight was also higher (Figure 2b). The results from the second study (Figure 3) show that the mode of application of the fungus does not affect the damage and marketable yields significantly. However the fungus treatments had lower damage and higher percentage of marketable corms then the control. Figure 2a: Percentage damaged corms per treatment at Situm Figure 2b: Percentage marketable corms per treatment at Situm Figure 3a: Percentage physical loss (EDW) at Bubia 104 third taro symposium Figure 3b: Percentage marketable corms by weight at Bubia Discussion The two trials have demonstrated that Metarhizium anisopliae has the potential to reduce taro beetle damage and increase marketable weight. The first trial showed that the treatment where the fungus was applied on its own resulted in significantly lower damage than all other treatments, and the marketable yield was as good or better than most of the other treatments, including the control. However, the synergistic effect of the fungus when combined with other control methods, for instance insecticides, requires further evaluation. The site where the trial was conducted receives up to 3,500 mm of rainfall per annum and during the course of the trial heavy rainfall caused some flooding in the vicinity of the trial. The results demonstrated that when applied under grass mulch the fungus can remain effective during such adverse weather conditions. The second experiment demonstrated that M. anisopliae will reduce damage no matter what the application technique. Broadcasting was the easiest method of application. Ultraviolet light and rainfall can adversely affect fungal viability, but the grass mulch prevented this. The field application methods needs refining, however, because the technique of grass mulch is labour intensive. The fungus used in this experiment had been stored in the freezer for three months before field application and this may have reduced its viability and effectiveness. Using fresher fungal spores could increase effectiveness and viability of the fungus and hence further decrease damage. A low cost production and field dispersal method of the biocontrol agent should be developed to improve effectiveness and sustainability. M. anisopliae remains the most promising biological control agent for taro beetles. The fungus infects and kills all beetle stages except eggs and therefore has the potential to be used in reducing populations by its application in breeding sites and in area wide management programmes. Other field studies at Bubia have confirmed that the fungus can remain infective three years after release into the soil (Masamdu and Simbiken, unpublished data). Acknowledgements The European Union funded this study through the Pacific Regional Agricultural Programme, 7-RPR–325, Project no. 5, Control of taro beetles. The Secretariat of the Pacific Community Plant Protection Service implemented the project and BioCare (Australia) who produced and provided the fungi are all hereby acknowledged. References Milner, R.J. 1992a. The potential of Metarhizium anisopliae for the control of scarab pests of groundnuts in Myanmar (Burma). p. 277–280. In: Glare, T.R. and Jackson, T.A. (eds). Use of pathogens in scarab pest management. Intercept Limited, Andover, Hampshire. Milner, R.J. 1992b. The selection of strains of Metarhizium anisopliae for the control of Australian sugar cane white grubs. p. 209–216. In: Glare, T.R. and Jackson, T.A. (eds). Use of pathogens in scarab pest management. Intercept Limited, Andover, Hampshire. Rath, A.C. 1992. Metarhizium anisopliae for the control of the Tasmanian pasture scarab Adoryphorus couloni. p. 217–228. In: Glare, T.R and Jackson, T.A (eds). Use of pathogens in scarab pest management. Intercept Limited, Andover, Hampshire. Theunis, W., Aloali’i, I., Masamdu, R. and Thistleton, B.M. 1996. Pathogens tested on taro beetles and their potential for biological control. p. 9–24. In: Glare, T.R and Jackson, T.A (eds). Proceedings of the 3rd International Workshop on Microbial Control of Soil Dwelling Pests. Lincoln, New Zealand, 21–23 February 1996. Microbial Control Group, AgResearch, Lincoln, New Zealand. Thistleton, B.M. 1984. Taro beetles. Harvest 10:32–35. Thistleton, B.M., Aloali’i, I., Masamdu, R. and Theunis, W. 1995. The biology and control of taro beetles Papuana sp. (Coleoptera: Scarabaeidae) in the South Pacific. Paper given at SPC/DAL/ACIAR/FAO/UNITECH/IPBGRI Taro Seminar, University of Technology, Lae, Papua New Guinea, 26–30 June 1995. 11 p. third taro symposium 105 Appendix 1: Percentage corm removed (PCR) scale Scale % Midpoint 0 no damage 0 1 <1 0.5 2 1- 5 3.0 3 6-10 8.0 4 11-15 13 5 16-20 18 6 21-30 25.5 7 31-40 35.5 8 41-50 45.5 9 51-60 55.5 10 61-70 65.5 11 71-80 75.5 12 81-90 85.5 13 91-100 95.5 Estimated damaged weight (EDW) This is an estimate of the actual amount of corm removed. It is calculated from the weight (w) and the midpoint of the percentage corm removed class (a) from the above table. The formula is: EDW=wa/(100-a) Estimated undamaged weight (EUW) An estimate of the undamaged corm weight (EUW) of the corm can be calculated from: EUW=(100w)/(100-a) 106 third taro symposium Theme Two Paper 2.3 The biology of Phytophthora colocasiae and implications for its management and control R.A. Fullerton and J.L. Tyson Horticulture and Food Research Institute of New Zealand, Mt Albert Research Station, Auckland, New Zealand Introduction Taro leaf blight (TLB), caused by Phytophthora colocasiae Raciborski, is the most destructive fungal disease of taro (Colocasia esculenta (L.) Schott). It is considered to have originated in South East Asia (Trujillo, 1967; Zhang et al., 1994) and is widely distributed throughout the tropical regions of the world (CMI, 1997). Typical symptoms are large, necrotic, zonate spots on the leaves, often coalescing to destroy large areas of leaf. The margin of the lesion is marked by a white powdery band of sporangia and numerous droplets of orange or reddish exudate. The disease can cause rapid and complete defoliation of susceptible varieties. Under some circumstances, the disease can also invade harvested corms and cause heavy losses during storage (Jackson and Gollifer, 1975). The organism has a very limited host range, most commonly affecting species of Colocasia, Xanthosoma (though Xanthosoma saggitifolia is immune) and Alocasia macrorrhiza. It has also been recorded as the cause of foot rot on Piper betle (McRae, 1934). In this paper, key features of the biology of the organism are outlined and their implications for various control options are discussed. Life cycle and epidemiology Rainfall, humidity and temperature are the key factors controlling the disease cycle and epidemiology of P. colocasiae. The primary reproductive unit of P. colocasiae is the sporangium. It is convenient to take this as the starting point for the life cycle. Germination Sporangia require free water in which to germinate and can germinate in either of two ways, depending on the temperature. 1. Indirect. Under “cool” conditions (20-22°C), cytoplasm within each sporangium differentiates into 15 to 20 zoospores. The terminal pore of the sporangium dissolves and the zoospores ooze out and swim off into the water film. This is a very rapid process. Zoospore formation usually commences within 15-20 minutes of being cooled at 20°C. From the first signs of movement within the cytoplasm to zoospore release takes less than a minute. Within about 10 minutes the zoospores settle onto the leaf surface, lose their flagella and form a rounded cyst. Cysts germinate to form a fine germ tube within 5-10 minutes. This mode of germination provides for an up to 15-fold increase in inoculum, allows dispersal in dew or rainwater, and, because new infections can be initiated within an hour of a sporangium being formed, the fungus can continue to sporulate and infect during short periods of leaf wetness. 2. Direct. Under warmer conditions (28-30°C), sporangia germinate directly by germ tubes that can infect the leaf. This is a slower process than zoospore production as it can take 5-6 hours for a sporangium to germinate. The proportion of sporangia germinating directly is generally much lower than for those forming zoospores (Putter, 1976; Trujillo, 1965). Infection Infection can occur on both surfaces of the leaf. Germ tubes can either penetrate the epidermis directly or enter via stomata and spread inter- and intra-cellularly through the leaf tissue. Most infections occur between midnight and dawn, with the majority over the period 2400-0200 hr (Putter, 1976). The conditions of cooler temperatures and free moisture from rain or dew over that period are also those that promote zoospore production by sporangia. Daytime infections only occur during continuously wet conditions. Symptom development Initial symptoms appear within 36 hours of infection as small water-soaked flecks on the leaf surface. The fungus is normally most active during the night and each morning lesions have a distinctive water-soaked margin of newly invaded tissue bearing a white mass of sporangia, and orange liquid droplets. Under dry conditions the water-soaked margin dries out during the day and the process is repeated the next night. During cooler, rainy days the lesions can also continue to expand during the day. The growth of the fungus within the tissues of the plant is strongly affected by temperature. The optimum temperature for growth of the fungus in vitro is approximately 25°C. In detached leaf tissue, the rate of symptom development is third taro symposium 107 greatest at temperatures in the range 25°C-30°C; at 35°C symptom development is halted (Fullerton and Tyson, 2001a). During hot, dry weather it is common for lesions in the field to stop expanding, and for the necrotic centres to drop out. Many of these “shot holes” expand no further; others will resume development (often from one point at the margin) under conditions of heavy rain. The most rapid expansion of lesions occurs when cool, showery weather allows fungal growth in tissues both night and day. Sporulation Sporulation normally occurs only in the zone of active fungal growth at the margin of the lesion. Sporangiophores emerge either through the stomata or directly through the epidermis on both surfaces of the leaf. Under optimum conditions (relative humidity approaching 100%, temperature 20-22°C) sporulation can take place at the margin of a lesion in less than 3 hours (Trujillo, 1965). Sporulation normally peaks between midnight and dawn with no sporangia produced during the day (Putter, 1976). Dispersal and spread Sporangia and zoospores in rain splash and wind blown spray are the principal sources of spread within and between plants. Rivulets of dew and exudate droplets carrying sporangia and zoospores are important mechanisms of spread on and between leaves of the same plant (Putter, 1976). Sporangia are not released into the air on drying. Long-distance dispersal of the organism occurs only by movement of infected plant material (leaves, petioles, infected corms). Survival The primary mode of survival is the continuous recycling of the pathogen, often on single plants within the crop. This is accomplished by the ability of the fungus to sporulate and reinfect within the same night utilising dew as the moisture source. It is not well adapted to long-term survival in the absence of living host material. The fungus does not survive as hyphae in soil (Sitansu, Ghosh and Pan, 1994). Sporangia on leaves dehydrate rapidly during the day. Sporangia in vegetative material (e.g. tops used for planting) seldom survive more than a few days, though some have been shown to survive for up to two weeks (Gollifer, Jackson and Newhook, 1980). Under normal circumstances large numbers of sporangia are washed to the soil. Most of these discharge zoospores or lyse within the first five days. However, a small proportion develops thick walls, forming chlamydospores that are able to survive in soil for up to three months (Quitugua and Trujillo, 1998). The importance of soilborne chlamydospores in the epidemiology of the disease has not been established but they could allow survival of the pathogen between crops. In situations where vegetative material dies off due to drought or cold conditions, the fungus most likely survives between seasons as vegetative mycelium in infected corms (Butler and Kulkarni, 1913). Endemicity and epidemic development The first lesions to occur in a new crop will be the result of inoculum carried from lesions in adjacent plots or wild plants. In the absence of regular rainfall, conditions favourable to reinfection occur on most nights ensuring regular cycling and survival on infected plants (endemicity). Those plants represent foci of infection scattered throughout the crop. When those conditions are supplemented by daytime rainfall, the disease can rapidly increase to epidemic proportions, spreading both on and between all plants throughout the crop. (Trujillo, 1965) determined that epidemics will occur when night temperatures and relative humidity are optimal for 6-8 hours for 3-4 consecutive days, and light rains or dews prevail in the morning. Under conditions of endemic survival, the distribution of infected plants in an area, and the severity of symptoms on those plants, is often irregular. With continuous night time sporulation and infection during the endemic phase it is possible for some plants to become severely diseased while there may be little or no disease on plants immediately adjacent. On infected plants it is common for the older (lower) leaves to be more severely affected than younger leaves. This is due to a constant supply of inoculum deposited by runoff water or dew from above, a more favourable microclimate for the fungus lower in the canopy, and also because the less waxy cuticles of older leaves allows better adhesion of spore-bearing water drops. Heterothallism and genetic variability P. colocasiae is a heterothallic fungus, requiring the presence of opposite mating types (A1 and A2) for the formation of oospores. In other Phytophthora species, sexual reproduction is associated with increased genetic variation, including increased variability in virulence and aggressiveness. The oospores may also provide a source of long-lived inoculum. Strains of both mating types have been found in Hainan (Zhang et al., 1994). In a recent study of strains from the Pacific region (Tyson and Fullerton, 2003) only one mating type (A2) was found throughout the region, including Guam, Hawaii, Indonesia, Philippines, Papua New Guinea and Samoa. Strains of neuter (A0) mating type (no oospores formed with either tester) were found also from Indonesia, Thailand and Papua New Guinea - the majority of isolates from PNG were A0. While oospore formation can be readily induced between opposite mating types in culture there is no evidence that this occurs regularly in nature. Despite the apparent lack of a sexual cycle, P. colocasiae has a high degree of genetic variability. Isoenzyme and RAPD analyses of strains from Thailand, Vietnam, Philippines, Indonesia and Papua New Guinea (TANSAO, 2001) have revealed extensive genetic variation amongst strains both within and between countries. Furthermore, the different genotypes (classified by zymotype) were unique to each country suggesting that the fungus has the capability for 108 third taro symposium significant genetic change in the absence of sexual recombination. Although pathogenic variability may be inferred from a high degree of variability determined by enzyme or molecular analysis, this has not yet been demonstrated. Implications for control strategies The epidemiological characteristics of P. colocasiae, viz. capacity for endemic survival under dry conditions and rapid transition to epidemic development under wet conditions, have an impact on the likely success of control measures. Exclusion. The organism is unlikely to be dispersed over long distances by fungal propagules. Outbreaks of the disease in new areas distant from known centres of infection are most likely due to the introduction of infected planting material. In countries that do not have the disease, constant vigilance is needed to ensure that it is not imported. Sanitation. (Putter, 1976) showed that the removal of infected leaves was highly effective in controlling the disease in subsistence taro gardens, particularly when plots were well separated from one another. This strategy would be most effective when the disease is in an endemic phase, with a relatively low and restricted incidence. When the disease is in an epidemic phase, the removal of all leaves with lesions would quickly lead to almost complete defoliation of the crop, with consequent effects on yield. This was in fact the experience of growers in Samoa (Adams, 1999), and sanitation was therefore largely abandoned as a disease management strategy. Fungicides. Successful control of taro leaf blight is technically possible with fungicides. While a range of fungicides have been evaluated, mancozeb (e.g. Dithane M45), copper (e.g. copper oxychloride), metalaxyl (e.g. Ridomil MZ – containing mancozeb) and phosphorus acid (e.g. Phoschek) are amongst those most commonly recommended. Mancozeb and copper have protectant activity only. Metalaxyl and phosphorus acid are specific for Phytophthora (and Pythium) diseases; metalaxyl is prone to the development of resistance. However, results with chemical control can be variable. Jackson et al. (1980) found that mancozeb did not control the disease in Solomon Islands. Trujillo (1996) reported that copper gave little control in Hawaii. The efficacy of fungicidal control of any foliar disease is strongly dependent on the severity of the disease at the time, and the prevailing weather conditions. In principle, fungicides are most effective when the target disease is present at low incidence, thereby limiting inoculum levels in the crop. When the disease enters an exponential phase, efficacy of control is reduced. On that basis, fungicides might be expected to be most effective against taro leaf blight when applied regularly during its endemic phase. Because of the rate at which taro leaf blight can progress from an endemic to an epidemic state, however, and the frequency of epidemic promoting conditions in many localities, fungicidal control can be both difficult and expensive. For example, in Samoa fungicides were frequently applied at higher rates and with greater frequency than label recommendations in an attempt to maintain control (Adams, 1999; Semisi, Mauga and Chan, 1998). In subsistence agriculture, fungicidal control is rarely a viable option. Resistant cultivars. Resistant cultivars represent the only sustainable solution to taro leaf blight in most production systems. To ensure maximum durability of resistance, the TaroGen project (AusAID/SPC Taro Genetic Resources: Conservation and Utilisation, 1998 –2003) adopted a strategy of breeding for horizontal resistance utilising recurrent mass selection techniques (Anon., 1998). The merit of that course is supported by the recent demonstration of a high degree of genetic variability of the organism within and between countries (TANSAO, 2001). Robinson (1998) proposed three rules for breeding for horizontal resistance: 1. Screen for yield, as this is correlated with freedom from parasites. 2. Inoculate to ensure that the high yield is due to resistance and not due to chance escapes. 3. Use the “one pathotype” technique to ensure resistance is horizontal. To date there is no evidence of strain-specific resistance in taro, or of matching pathotypes in P. colocasiae. Because horizontal resistance is not pathotype specific, failure to identify different pathotypes is not a limiting factor to the strategy. A major challenge however, is the reliable identification of the least susceptible individuals in a population for use in the next cycle of crossing. In areas where weather conditions favour frequent epidemics, field evaluations by eye remain the most convenient method of selection. This technique has proved to be very successful in the breeding programme in Samoa. While there is the chance of selecting escapes in single genotype populations, these can normally be eliminated in the second round of evaluations when small clonal plots can be assessed. In areas where there are often extended periods of endemicity, the irregular distribution of the disease in the field can cause difficulties in discriminating between “susceptible” and “resistant” genotypes, and “escapes”. Under those conditions more objective screening methods may need to be employed. In a comparison of different methods (inoculations in screenhouse, nursery, field and “water-bed”), Ivancic et al. (1996) found that only the water-bed method allowed detection of differences in resistance and susceptibility in breeding progeny. The method involved floating leaves of test plants on the surface of a water bath in which leaves with sporulating lesions had been previously been washed. This method has disadvantages in that the concentration of inoculum is unknown and there is a natural bias towards selection of plants with immunity (or vertical resistance). Wall et al. (1998) used spray inoculation and high humidity incubation in a greenhouse and rated plants on percentage of leaf area damaged after 6-8 days. This method proved to be able to distinguish the most resistant clones within the third taro symposium 109 test population (29 cultivars). This method is less practicable for screening large populations as all plants would have to be propagated in pots for testing. Fullerton et al. (1999, 2000) developed a field inoculation method using tabs of blotting paper infiltrated with sporangia and fixed to the leaf overnight with wide adhesive tape. The tape and tabs were removed the following morning and numbers of lesions and lesion size measured after three days. The method was effective for identifying hypersensitive reactions (HR), indicative of vertical resistance. However, field reactions of non-HR plants varied between repeat inoculations. Laboratory studies on leaf discs showed that the rate of lesion development was strongly affected by temperature and that lesion development slowed above 30°C and ceased by 35°C. Thus, assessments by this method are strongly influenced by daily weather and therefore unreliable. A laboratory test using leaf discs held on water agar augmented with wetting agent and benzimidazole (to prevent senescence), inoculated with agar plugs of P. colocasiae and incubated at 25°C in the dark gave consistent and reliable results (Fullerton and Tyson, 2001b). The method was able to differentiate between the most susceptible and least susceptible genotypes and was also highly effective in identifying hypersensitive reactions. The technique could also be used to screen genotypes generated in a country that does not have leaf blight provided that fresh leaves can be rapidly delivered to a laboratory where testing can be carried out. With horizontal resistance breeding strategies, it is normal to generate many progeny of good agronomic quality but differing widely in their degree of disease resistance. Such a range of material provides the opportunity to match the degree of resistance to the potential risk of disease. For example, selections of moderate resistance but of superior agronomic quality could be utilised in climatic zones less prone to epidemics. For this reason “genotype x environment” evaluations are critical to ensure maximum efficiency and output of the breeding programmes. References Adams, E. 1999. Farmers use both chemical and cultural methods to control TLB. South Pacific Agricultural News (IRETA) 16:1, 4, 7. Anon. 1998. Taro genetic resources: Conservation and utilisation. p. 21. In: Proceedings of AusAID/SPC Taro Breeding Workshop, Suva, Fiji, 26–28 August 1998. Secretariat of the Pacific Community, Noumea, New Caledonia. Butler, E.J. and Kulkarni, G.S. 1913. Colocasia blight caused by Phytophthora colocasiae Rac. Memoirs of the Department of Agriculture in India, Botanic Series 5:233–261. CMI. 1997. Commonwealth Mycological Institute distribution maps of plant diseases, Map no. 466, Edition 3: Phytophthora colocasiae. Commonwealth Agricultural Bureau, Wallingford, Oxfordshire. Fullerton, R.A. and Tyson, J.L. 2001a. p. 7. In: Plant pathology progress report: TaroGen Review, March 2001. HortResearch, Auckland. Fullerton, R.A. and Tyson, J.L. 2001b. p. 8. In: Plant pathology progress report: TaroGen review, November 2001. HortResearch, Auckland. Fullerton, R.A., Tyson, J.L. and Gunua, T.G. 1999. p. 13. In: Plant pathology progress report: TaroGen annual review, 22 October 1999. HortResearch, Auckland. Fullerton, R.A., Tyson, J.L. and Iramu, E. 2000. p. 13. In: Plant pathology progress report: TaroGen review, 27–29 November 2000. HortResearch, Auckland. Gollifer, D.E., Jackson, G.V.H. and Newhook, F.J. 1980. Survival of inoculum of the leaf blight fungus Phytophthora colocasiae infecting taro, Colocasia esculenta in the Solomon Islands. Annals of Applied Biology 94:379–390. Ivancic, A., Kokoa, P., Gunua, T. and Darie, A. 1996. Breeding approach on testing for resistance to taro leaf blight. p. 93–96. In: Jackson, G.V.H. and Wagih, M.E. (eds). The Second Taro Symposium: Proceedings of an international meeting held at the Faculty of Agriculture, Cenderawasih University, Manokwari, Indonesia, 23–24 November 1994. Cenderawasih University and the Papua New Guinea University of Technology, Lae. Jackson, G.V.H. and Gollifer, D.E. 1975. Storage rots of taro (Colocasia esculenta) in the British Solomon Islands. Annals of Applied Biology 80:217–230. Jackson, G.V.H., Gollifer, D.E. and Newhook, F.J. 1980. Studies on the taro leaf blight fungus Phytophthora colocasiae in Solomon Islands: Control by fungicides and spacing. Annals of Applied Biology 96:1–10. McRae, W. 1934. Foot-rot disease of Piper betle L. in Bengal. Indian Journal of Agricultural Sciences 4:585–617. Putter, C.A.J. 1976. The phenology and epidemiology of Phytophthora colocasiae Racib. on taro in the East New Britain province of Papua New Guinea. MSc thesis. University of Papua New Guinea. Quitugua, R.J. and Trujillo, E.E. 1998. Survival of Phytophthora colocasiae in field soil at various temperatures and water matric potentials. Plant Disease 82:203–207. Robinson, R. 1998. Horizontal resistance and its relationship to plant pathosystems. In: Proceedings of AusAID/SPC Taro Breeding Workshop, Suva, Fiji, 26–28 August 1998. Secretariat of the Pacific Community, Noumea, New Caledonia. 110 third taro symposium Semisi, S.T., Mauga, T. and Chan, E. 1998. Control of the leaf blight disease, Phytophthora colocasiae Racib in taro, Colocasia esculenta (L.) Schott with phosphorous acid. Journal of South Pacific Agriculture 5:77–83. Sitansu, P., Ghosh, S.K. and Pan, S. 1994. Effect of temperature, moisture and soil amendment on the survival ability of hyphae of Phytophthora colocasiae in soil. Journal of Mycopathological Research 32:59–65. TANSAO. 2001. Taro: Evaluation and breeding for rainfed cropping systems in South East Asia and Oceania. INCODC: International Cooperation with Developing Countries. 207 p. Trujillo, E.E. 1965. The effects of humidity and temperature on Phytophthora blight of taro. Phytopathology 55:183– 188. Trujillo, E.E. 1967. Diseases of the genus Colocasia in the Pacific area and their control. IV 13–IV 19. In: Proceedings of the International Symposium on Tropical Root Crops, Vol. 2, University of the West Indies, St Augustine, Trinidad, 2–8 April 1967. University of the West Indies, St Augustine. Trujillo, E.E. 1996. Taro leaf blight in Micronesia and Hawaii. p. 41–43. In: Taro Leaf Blight Seminar: Proceedings. Alafua, Western Samoa, 22–26 November 1993. Unpublished. Tyson, J.L. and Fullerton, R.A. 2003. Mating types of Phytophthora colocasiae strains from the Pacific region, India and South-East Asia. Abst. 28.26, 359. In: 8th International Congress of Plant Pathology: Abstracts of offered papers. Christchurch, New Zealand, 2–7 February 2003. Wall, G.C., Wiecko, A.T. and Trujillo, E.E. 1998. Evaluation of resistance to taro leaf blight in 29 Colocasia esculenta cultivars. Phytopathology 88:S123. Zhang, K.M., Zheng, F.C., Li, Y.D., Ann, P.J. and Ko, W.H. 1994. Isolates of Phytophthora colocasiae from Hainan Island in China: Evidence suggesting an Asian origin of this species. Mycologia 86:108–112. third taro symposium 111 Theme Two Paper 2.4 Current status of research on Rhizoglyphus mites associated with taro Zhi-Qiang Zhang1, Qianghai Fan1, N.A. Martin2 and Sada Nand Lal3 Landcare Research, Auckland, New Zealand Crop and Food Research, Private Bag 92169, Auckland, New Zealand 3 Secretariat of the Pacific Community 1 2 Introduction Mites (Acari) are small arthropods found in a diverse range of habitats. Many species of mites are common on crops and in stored products. Some of these are serious pests, whereas others are predators capable of reducing populations of mite and insect pests. Mites of the genus Rhizoglyphus are commonly associated with plants with bulbs, corms and tubers. Over 50 species have been named worldwide but their taxonomy is in a state of confusion (Diaz et al., 2000). Two species, Rhizoglyphus echinopus and R. robini, are known to cause damage, directly by feeding and indirectly by spreading plant pathogens, to a variety of crops (e.g. onions, garlic and other vegetables) and ornamentals (lily and other flower bulbs) in greenhouses and in the field around the world (Fan and Zhang, 2003). Most other species Rhizoglyphus are restricted in distribution and have been recorded from a narrow range of hosts. Their biology and economic importance are poorly known (Diaz et al., 2000). Rhizoglyphus mites are frequently intercepted in New Zealand and Australia on bulbs, corms and tubers of crops originating from various Pacific nations. Rhizoglyphus minutus is among the mites frequently intercepted on taro originating from Fiji. Because of the lack of knowledge of its presence in New Zealand and Australia and its potential damage to root crops in these countries, R. minutus is considered a quarantine risk species and contaminated produce is subjected to fumigation, which has negative economic consequences, as well as environmental and human health concerns. This project on taro mites was initiated by Secretariat of the Pacific Community (SPC) and conducted by two New Zealand crown research institutes (Landcare Research and Crop & Food Research) to facilitate the export of taro from Fiji to New Zealand. The project aims to (1) conduct a survey of taro mites (Rhizoglyphus) in Auckland, mainly on taro, and also on dahlia, freesia, gladiolus, hyacinth, iris, narcissus, orchid, and tulip, (2) provide identification of mites intercepted from taro originating from Fiji and also collected in the field, and (3) prepare a Pest Risk Assessment (PRA) for New Zealand Ministry of Agriculture and Forestry on behalf of SPC. Materials and methods Several hundred specimens of Rhizoglyphus mounted on glass slides were examined. Most of these are from the following collections: New Zealand Arthropod Collection in Landcare Research, Auckland, New Zealand (NZAC); the National Plant Pest Reference Laboratory, Ministry of Agriculture and Forestry in Lincoln and Auckland, New Zealand (NPPRL); Australian Quarantine and Inspection Service (AQIS). Some were fresh specimens collected from taro and other plants in Auckland. All specimens were studied using an interference-phase contrast microscope. The PRA was prepared based on published literature and the results of this study. Results Mites on taro from Fiji Some 260 mite specimens from taro originated from Fiji were examined. These included three Rhizoglyphus species: Rhizoglyphus minutus (186 specimens), Rhizoglyphus tsutienensis (13) and Rhizoglyphus robini (1). In addition to Rhizoglyphus species, we also found three species of Schwiebia (60) and one species of Tyrophagus (1). All species belong to the mite family Acaridae. Schwiebia were sometimes mis-identified as Rhizoglyphus. Rhizoglyphus minutus outside Fiji This species was first discovered from taro in Niue (Manson, 1972). We also studied specimens of Rhizoglyphus minutus collected on taro from Samoa (4) and Tonga (3). A single specimen, collected from soil in which Camellia had been growing in New Zealand (New Plymouth), was confirmed as Rhizoglyphus minutus. Rhizoglyphus in New Zealand Over 500 specimens of Rhizoglyphus from onions, flower bulbs and seeds in New Zealand were examined. R. minutus and R. tsutienensis were not found among the specimens. Rhizoglyhus robini and Rhizoglyphus echinopus are most common. R. ranunculi is only known from Ranunculus in New Zealand. Several recent samples of taro grown in Auckland revealed no R. minutus. Further efforts in collecting more samples are being made. 112 third taro symposium Discussion This study clearly shows that the so-called “taro mite” is not just Rhizoglyphus minutes, as was previously believed, although this is indeed the most common species intercepted on taro from Fiji. The confirmation of specimens of R. minutus from Tonga, Samoa and New Zealand indicates that this species may be more widely distributed in the Oceania than previously suspected. The single record of R. minutus from New Plymouth, New Zealand is not strong enough to exclude this species from the list of quarantine species for New Zealand. Further efforts are being made to find this species in Auckland. Since taro has been imported into New Zealand from Fiji for over 30 years, it is very likely that R. minutus has been introduced many times and might have become naturalized in New Zealand if it was able to survive the climate here. The survey in Auckland is therefore important. Although little is known about the biology of Rhizoglyphus minutus, it is not known to cause economic injury to taro plants in Fiji, Niue, Tonga and Samoa. It is possible that this species is a postharvest species and thrives in store houses or containers. Rhizoglyphus tsutienensis was previous known only from bulbs of several host plants in Taiwan (Ho and Chen, 2000, 2001). It is quite rare (<5%) among the intercepted material from taro in Fiji. Based on the specimen records, Rhizoglyphus minutus have been intercepted on taro from Fiji since early 1970s, and so have the two species of Schwiebia. However, Rhizoglyphus tsutienensis, which was first described from Taiwan in 2000, was seen only in intercepted material during 2001-2002. It is possible that this species was only recently introduced to Fiji from Asia. Another species, Rhizoglyphus longispinosus, has been described from Taiwan, attacking Colocasia formosana. It is only known from Taiwan. It may be of interest to note that Rhizoglyphus setosus is known from Fiji in association with mealybugs on pineapple. In Hong Kong, this species was found on taro (Manson, 1972). This species has not been found on taro from Fiji, despite the large number of samples examined. Acknowledgements This project could not have started without the enthusiasm and support of Parmesh Chand (Pacific Islands Trade & Investment Commission, Auckland), Trevor Crosby (Landcare Research) and Mick Lloyd (Plant Protection Service, Secretariat of the Pacific Community, Suva). The last two also contrinuted to ideas through stimulating discussions and Parmesh kindly collected samples of taro in Auckland. Grace Hall and Leonie Clunie (both of Landcare Research) collected mite samples from various plants in Auckland. References Diaz, A., Okabe, K., Eckenrode, C.J., Villani, M.G. and O’Connor, B.M. 2000. Biology, ecology, and management of the bulb mites of the genus Rhizoglyphus (Acari: Acaridae). Experimental and Applied Acarology 24:85–113. Fan, Q.-H. and Zhang, Z.-Q. 2003. Rhizoglyphus echinopus and Rhizoglyphus robini (Acari: Acaridae) from Australia and New Zealand: Identification, host plants and geographical distribution. Systematic and Applied Acarology, Special Publication 16:1–16. Ho, C.C. and Chen, W.H. 2000. A new species of Rhizoglyphus Claparede (Acari: Acaridae) infesting bulbs from Taiwan. Chinese Journal of Entomology 20:347–351. Ho, C.C. and Chen, W.H. 2001. A new species of Rhizoglyphus Claparede (Acari: Acaridae) from Taiwan infesting the taro and giant alocasia. Plant Protection Bulletin Taipei 43:47–49. Manson, D.C.M. 1972. A contribution to the study of the genus Rhizoglyphus Claparede, 1869 (Acarina: Acaridae). Acarologia 13:621–650. third taro symposium 113 Theme Two Paper 2.5 Developing interactive diagnostic support tools for tropical root crops V.dR. Amante and G.A. Norton Centre for Biological Information Technology, Level 6 Hartley-Teakle Building, The University of Queensland, Brisbane 4072, Australia Introduction Sweet potato, taro and yam are traditionally important crops in tropical agriculture. Until recently, they had been grown mainly for home consumption and to provide a source of additional cash. In the case of sweet potato, recent developments in the Asian region, which recognise its growing importance for food and industrial processing as well as for production of processed feeds, tend to create a new market. A similar trend is noted for taro and yam. To exploit and cope with this market opening, production systems are being changed from small to large scale operations. However, the consequence of changing production systems not only involves an increase in production inputs and a change in cultural management to obtain higher production but these changes also modify the crop’s macro and micro environments. With an increased production area and more extensive cultivation, an increase in pest, disease and other problems can be expected. These problems need to be addressed to ensure these changes in production systems are beneficial and sustainable. The worst thing that could happen is to give hope to small farmers for a better income and then allow them to experience and cope with the problems that come with a larger scale of production of which they will have difficulty in handling. Diagnosing crop problems is something that most farmers in Asia, the Pacific and Africa find difficult. Technical help is sometimes available from government institutions but, more often, they do not have the required specific skill and expertise and pass on the task to experts located in major towns or cities. More often than not, problems are either identified only after the crop has been too severely affected to merit management or they have not been identified at all because affected plants were not in a good condition for diagnosis. Advances in information and communication technology (ICT) provide an opportunity to bring expertise closer to the local people who are directly involved in crop production. The email list-server – PestNet – that has over 400 subscribers in the Pacific region, is one way in which IT is linking experts, extension agents and other parties (see www.pestnet. org). These advances in ICT also allow the development of creative and innovative teaching, learning and diagnostic support tools that could result in a better understanding of crop problems, their identification and management. This in turn could lead to higher yields, more efficient production and consequently increased farmers’ income. Figure 1: Sweetpotato DiagNotes - a diagnostic-support tool produced by the Centre for Biological Information Technology, The University of Queensland, Australia. 114 third taro symposium Dichotomous and Matrix keys There are two types of identification and diagnostic keys – dichotomous (or pathway) keys and matrix keys. Dichotomous keys are the traditional keys used for identification of biological organisms. A dichotomous key presents the user with a hierarchical set of questions or “couplets”, concerning more general to more specific features of the specimen to be identified or diagnosed. Depending on which of the couplets they choose, users are either directed to another couplet or to a solution – an identification or a diagnosis. A major problem with dichotomous keys is the “unanswerable couplet” problem. If the user is unable to decide which couplet best describes the specimen, he or she is unable to continue. By contrast, matrix keys are more flexible than dichotomous keys since they allow the user to consider and choose features describing the specimen in any order they wish. This allows users to ignore features that are not clear and still be able to get a reliable diagnosis or at least a short list of likely causes. Lucid Professional is a matrix key system. For more information go to: www.lucidcentral.com. This paper describes the process involved in developing a CD-ROM diagnostic-support tool for sweet potato, the Sweetpotato DiagNotes (Figure 1), aimed at educating and supporting those directly or indirectly involved in managing the crop. It is suggested that this sweet potato diagnostic support tool could provide a model for a similar product for taro and yam. Developing “Sweetpotato DiagNotes” Four major steps were involved in developing Sweetpotato DiagNotes - a sweet potato diagnostic support tool - (1) collecting relevant information in text and graphic form, (2) developing fact sheets (3) constructing the key and (4) technical and field testing. 1. Collection of information in text and graphic form Information about the sweet potato crop was obtained from collaborating international experts from the International Potato Center (CIP) in Indonesia and Peru, Louisiana State University in the United States of America (USA), Natural Resources Institute in the United Kingdom (UK), The University of Queensland (UQ), Australia and the Philippine Rootcrops Research and Training Center (PhilRootcrops), Philippines. These collaborators were requested to provide information based on a common format and were frequently consulted to confirm or verify details. Further information was obtained from the literature, and web searches and from the authors’ knowledge and experience of the crop. 2. Developing the fact sheets and other relevant information There are two fact sheets for each pest, disease and nutrient disorder included in Sweetpotato DiagNotes. The first fact sheet provides (Figure 2) a summary of diagnostic information, including common signs and symptoms and, when available, information on conditions that aggravate or reduce the occurrence of the disorder. The second fact sheet, which is linked to the diagnostic summary, gives more detailed information on taxonomy, economic importance, geographical distribution, morphology, signs and symptoms, biology and ecology, confusion with other symptoms, detection and inspection, management and references. Both fact sheets contain images, most of which were taken in the field. In addition to fact sheets, general information about the crop and its production is included in the product. Figure 2: Each disorder has a fact sheet providing a diagnostic summary. third taro symposium 115 3. Constructing the key Lucid Professional is a software tool developed by the Centre for Biological Information Technology (CBIT) at UQ for creating and publishing interactive identification or diagnostic keys (See the Box for more information on the two types of identification and diagnostic keys). This software tool, which was used to develop the matrix key for diagnosing sweet potato problems, consists of a Builder, which allows keys to be easily constructed, including the incorporation of multimedia (images, audio and video) and a Player, which allows users to operate the key. When a user selects in the key those features they have observed in the “sick” crop, the Player filters these information, reduces the list of Possible Causes and lists the eliminated problems under Unlikely Causes (Figure 3). Figure 3: The Lucid Player allows users to operate the key. It has 4 windows: clockwise from top left: Possible Observations, Observations Chosen, Possible Causes and Unlikely Causes. Constructing the key used in Sweetpotato DiagNotes involved the following: 3.1. Developing a list of Possible Causes or problems A list of insect and mite pests, diseases and nutritional disorders affecting the sweet potato crop was made. The project staff and collaborators discussed and deliberated on the final list of Possible Causes or problems to be included in the key based on their current importance in the countries/regions covered, the potential for introduction and increased importance of these problems in these countries/regions and the existence of sufficient information on these problems. 3.2. Determining the Possible Observations associated with a “sick” crop Possible observations and “states” that characterise the identified problems were determined. The list included all the descriptions or interpretations of signs and symptoms as well as the conditions where the problems occur. A lot of deliberation was involved on how possible observations and “states” were to be presented. Possible observations were put in question form and “states” under each possible observation were made to be easy to understand and be able to describe what the users observe or think they observe in the field in order to allow for a quicker and easier identification of the problem using the Lucid Player. Hence, aside from using simple words to describe the “sick” plant and the field conditions, notes and images were further added (Figure 4). A glossary was also included to help with difficult or technical words used in the key and the fact sheets. 116 third taro symposium Figure 4: Notes on possible observations provide users with useful help in describing the features of a “sick” plant, such as leaf colour changes. 3.3. Interpreting information for scoring using a data matrix Lucid Builder was used to build the key. This software allows the key builder to score the matrix that lies behind the diagnostic key in a number of ways. Possible Observations are related to Possible Causes by a score in the key’s matrix (Figure 5). The score is “common”, when the feature in question is a common, frequently exhibited expression or condition of that particular cause; “rare”, when a specific feature may or may not be exhibited or expressed, depending on factors such as cultivar and environment; and “misinterpreted”, when users may incorrectly select a feature (e.g. choosing reduction in leaf size rather than leaf curling). Figure 5: Construction of the sweet potato key involved scoring Possible Causes against Possible Observation states using a data matrix provided by the Lucid Builder (blue= common; green= rare). 4. Field testing Field testing of diagnostic support tools is critical in providing the authors with insights into how the product will perform with respect to critical success criteria - namely: technical accuracy, relevance and acceptability (Table 1). Sweetpotato DiagNotes was field tested in the Philippines, Indonesia and Africa. 4.1 Technical accuracy The technical accuracy of the key was assessed by experts in a workshop held at PhilRootcrops, Philippines, and by distributing a prototype CD to reviewers from different countries. During the testing, one suggested weakness of the key was the difficulty associated with situations where there is more than one causal agent affecting a particular specimen (e.g. corky lesions on the stem and round holes in the leaves caused by leaf and stem scab and tortoise beetle, respectively). In response to this, the key has been modified so that when there are no possible causes left, the users is prompted to consider that there is unlikely to be a single problem but that the cause is most probably due to two or more different agents. One way to deal with this situation is to key out the observed symptoms separately, and refer to the fact sheets to confirm the results of the diagnosis. In general, reviewers found that the key, with the functions and special features it provides, was able to provide a valuable tool for diagnosis. third taro symposium 117 Table 1: Matrix used for sweet potato key field testing in the Philippines. Criteria Indicator Methodology Technical accuracy Software performance Technical content The key and other features of the CD were reviewed by experts based on guidelines provided. Relevance Usefulness of the key and the information contained in the CD to the users. The use of the key was monitored for a whole cropping season under different testing/learning situations such as: • field/office • training • classroom • farmer field schools • library/resource context A simple monitoring tool was developed and used for each situation. Acceptability Ease of use of the key and comprehensibility of information This was tested together with relevance, in this case with the focus on the ease of using the key and how user friendly it is. 4.2 Relevance The key and other information contained in the CD are only relevant or useful if they meet the requirements of the target users and the conditions in which they would use the product. For this reason, Sweetpotato DiagNotes was developed in accordance with user requirements and specifically focussed primarily on extension workers in Asia. The general profile of these extension workers (e.g. computer literacy, technical knowledge, knowledge of technical language and proficiency in the English language) was considered in the design and the language used in the key. As well as providing diagnostic support and training for farmers and advisors, Sweetpotato DiagNotes also constitutes a useful learning tool for students and researchers in getting acquainted with the crop. Feedback from extension workers who field tested the Sweetpotato DiagNotes during a two-day training in the Philippines and field testing in Indonesia and the Philippines during the actual growing season indicated that some problems were particularly difficult to diagnose and modifications were made to the key to address these issues. More generally, there is often apprehension about the relevance of computer-based tools for rootcrop growers when the majority of users would not have access to computers. Acquisition of computers, however, is a top priority among all local government offices in the Philippines and other third world countries. They are also becoming a necessity among households particularly with students and office workers. In the near future this is unlikely to be an important constraint. 4.3 Acceptability The training and field testing sessions in the Philippines and Indonesia and some feedback from Africa indicated strong support for Sweetpotato DiagNotes. The main reasons for this appear to be the ease of use of the key, the use of non-technical words in the key and fact sheets and the preference of target users for more image-based rather than textbased materials. Extension workers said this tool would enable them to diagnose crop problems quicker and to easily learn more about them. These extension workers would like similar tools to be available for other crops, including taro and yam. Discussion The continuing change in production systems of tropical root crops such as sweet potato, taro and yam brought about by expanding markets aggravates or creates pests, diseases and soil problems. There is an increasing need to educate extension workers and farmers who are directly involved in production to enable them to cope with these problems. One way to do this is to take advantage of the advances in information and communication technology and develop interactive diagnostic support and learning tools such as Sweetpotato DiagNotes and “RiceDoctor” – a similar key that is being developed for rice growers, in association with Philrice and the International Rice Research Institute. As shown for Sweetpotato DiagNotes, these products, with user-friendly interactive diagnostic keys, fact sheets, notes and images, make diagnosis and learning about the crop and its problems much quicker and easier for extension staff. Such tools are particularly appropriate for the South Pacific, where books and advice from experts are not readily available. Acknowledgements We wish to acknowledge the other authors of Sweetpotato DiagNotes – Jane O’Sullivan, Elske van de Fliert and the late Jose Pardales, Jr. We thank Erlinda Vasquez of PhilRootcrops, Elske van de Fliert of CIP-Bogor and James Okoth of FAO-Uganda, for their effort and feedback during the field testing in the Philippines, Indonesia and Africa respectively. Their involvement in the project has enabled this paper to be written. We also wish to thank the Australian Centre for International Agricultural Research (ACIAR) for the funds that made this project possible. References Amante, V.R., Norton, G.A., O’Sullivan, J.N., van de Fliert, E. and Pardales, J. Jr. 2003. Sweet potato DiagNotes: A diagnostic key to sweet potato problems. CD-ROM. Centre for Biological Information Technology, University of Queensland, Brisbane. 118 third taro symposium Theme Three Abstracts Thème 3: Production and Production Constraints Thème 3 : Production et obstacles qui l’entravent Taro as the foundation of Pacific food security Le taro, garant de la sécurité alimentaire de l’Océanie Nancy J. Pollock Nancy J. Pollock Taro remains a strong cultural symbol for many Pacific nations, both practically and ideologically. It has long been the basis of food security with its many varieties providing alternatives in a disaster (Pollock 2002). Its perpetuation as a local food is based on a deep knowledge of both planting and processing techniques, as well as very discriminatory tastes. Support that enables availability for consumption to continue is vital. Taro has a major role to play in the future food supply of many Pacific island states. Le taro demeure un puissant symbole culturel pour de nombreux pays océaniens. Il a longtemps été le garant, idéologique et concret, de leur sécurité alimentaire, ses nombreuses variétés offrant des aliments de substitution en cas de catastrophe (Pollock, 2002). Sa pérennité, en tant qu’aliment local, repose sur une connaissance approfondie des techniques de plantation et de transformation, et sur des qualités gustatives très particulières. Il est capital de soutenir la production de taro à des fins de consommation. Le taro est appelé à jouer un grand rôle dans l’approvisionnement alimentaire de nombreux États et Territoires océaniens. Consumption has been threatened in the past by natural disasters and disease such as taro blight, as well as by colonial cash cropping. Pacific peoples developed techniques for dealing with the natural hazards, but were less successful in stemming the onslaught of cash crops that infiltrated large areas of their best lands. Today that onslaught continues through the import of western foods, such as rice and flour based foods, as well as fast foods. These are fast to prepare, and sell at prices that undercut farmers’ costs of getting taro to markets. In this paper I argue that support for taro’s place as the most significant local food crop is integral to the re-establishment of food security in Pacific nations. Local foods not only are essential to maintain a healthy diet, developed over many years, but they also empower households through subsistence farming. Techniques for overcoming the natural hazards, developed over time, present a lower risk than the price fluctuations and negative values of some imported foods. For the poor, diversity of local foods provides a cheaper, and more readily available alternative to imported foods. Taro thus contributes to greater security, both nutritionally and economically. La consommation de taro a été maintes fois menacée, dans le passé, par des catastrophes naturelles et par des maladies telles que la flétrissure des feuilles du taro ainsi que par l’introduction d’autres cultures de rente par les puissances coloniales. Les Océaniens ont mis au point des techniques de lutte contre les risques naturels, mais n’ont pas aussi bien réussi à enrayer l’introduction de cultures de rente qui ont infiltré de vastes zones de leurs terres les plus fertiles. À l’heure actuelle, cette invasion se poursuit, sous forme de produits occidentaux d’importation, tels que le riz et les aliments à base de farine, ainsi que les produits de restauration rapide. Rapidement préparés, ces derniers se vendent à des prix inférieurs aux frais de commercialisation du taro qu’encourent les agriculteurs. Dans cet exposé, je montre qu’il faut rendre au taro son importance en tant que culture vivrière locale garante de la sécurité alimentaire des nations océaniennes. Les produits alimentaires locaux ne jouent pas seulement un rôle essentiel dans le maintien d’une alimentation saine établie au fil des ans, mais ils permettent aussi aux ménages d’acquérir une certaine autonomie, grâce à l’agriculture vivrière. Les techniques élaborées au fil du temps pour lutter contre les catastrophes naturelles présentent moins de risques que les fluctuations des prix et les prix peu élevés de certains produits importés. Pour les personnes démunies, la diversité des produits alimentaires locaux offre une solution de substitution moins coûteuse et plus accessible aux produits importés. Le taro contribue donc à accroître la sécurité alimentaire et économique. Taro production in Fiji: Constraints and future prospects La production de taro aux Îles Fidji: obstacles et perspectives Aliki Turagakula Aliki Turagakula Taro (Colocasia esculenta (L.) Schott), also known in Fijian as “dalo”, is an edible aroid and the most important Fijian staple, grown mostly in the wet zones of Fiji for its traditional and economic importance. Commercial taro production mainly involves export to niche markets in New Zealand, Australia and the West Coast of the United States of America. These niche markets are characterised by high population density of Polynesian Pacific Islanders Le taro (Colocasia esculenta (L.) Schott), aracée comestible également connue aux Îles Fidji sous le nom de « dalo », est le principal aliment de base fidjien. Il est surtout cultivé dans les zones humides de Fidji, où il revêt une grande importance traditionnelle et économique. La production commerciale de taro vise principalement l’exportation vers certains créneaux de Nouvelle-Zélande, d’Australie et de la côte ouest des États-Unis d’Amérique. Ces marchés third taro symposium 119 who are the biggest consumers. Their favourite pink taro variety is known as Tausala ni Samoa. Taro production in Fiji is mainly characterised by smallholder farms, seasonal plantings, traditional hill slopes cultivation, low crop yields, high post-harvest losses, taro beetle damage, and inconsistency of supply. Demands for fresh produce and processed products for exports and food security for domestic consumption are very high. Supply of taro to meet these demands poses a great challenge to the taro industry in Fiji. Future prospects for the taro industry in Fiji look bright, with many opportunities for value-adding. Food security demands in Fiji are a serious concern especially for alleviation of poverty. Genetic erosion of taro cultivars, due to increasing consumer and market preferences, occurrences of natural disasters and biotic agents, is of serious concern. étroits se caractérisent par une forte densité de Polynésiens, principaux consommateurs de taro. Leur variété favorite est le taro à chair rose appelé Tausala Ni Samoa. Aux Îles Fidji, les spécificités de la production de taro, pratiquée par des petites exploitations agricoles, sont des plantations saisonnières, une culture traditionnelle sur les pentes des collines, de faibles rendements, des pertes élevées après récolte, des dommages causés par les coléoptères du taro, et l’irrégularité de l’approvisionnement. Les exigences imposées aux produits, frais et transformés, destinés à l’exportation, et à l’innocuité alimentaire des produits consommés dans le pays sont très élevées. L’approvisionnement en taro répondant à ces exigences constitue un véritable défi pour la filière du taro à Fidji. Les perspectives sont toutefois encourageantes pour la filière et les possibilités de valorisation nombreuses. Les conditions d’innocuité alimentaire exigées à Fidji posent de graves problèmes, en particulier dans le cadre des mesures d’atténuation de la pauvreté. L’érosion génétique des cultivars de taro sous l’effet des préférences de plus en plus marquées des consommateurs et des marchés, des catastrophes naturelles et des agents biologiques, est préoccupante. Taro cultivation in the Marshall Islands: Problem, perseverance and prospects La culture du taro aux Îles Marshall : difficultés, persévérance et perspectives Dilip Nandwani, M.C. Cheng, Jimmy Joseph, Jabukja Aikne, Arwan Soson and Gwo-jong Moh Dilip Nandwani, M. C. Cheng, Jimmy Joseph, Jabukja Aikne, Arwan Soson et Gwo-jong Moh Taro is a significant food crop in the Marshall Islands. Various local and introduced cultivars of taro, viz Colocasia esculenta (kotak), Cyrtosperma chamissonis (iraj) Alocasia macrorrhiza (wild wot) and Xanthosoma sagitifolium (wot or wuthin kabilon) are widely cultivated and consumed in the Marshall Islands. The increased use and dependence on imported foods throughout the Marshall Islands over the past two decades has had a profound effect on production and consumption patterns of taro. Production of taro has fallen dramatically in recent years in response to the increased access to imported staples. Marshall Islands are free of some serious pests and diseases of taro such as Phytophthora colocasiae, taro leaf blight. However, the country is vulnerable to many serious pests and diseases due to the easy access via air and sea transportation. Limited land resources, low elevation, salt spray, poor and nutrient deficient soil and non-availability of planting material has resulted in insufficient production of taro. The use of tissue culture to introduce pathogen-tested material into the country would significantly reduce the possibility of introducing new pests and diseases. The paper describes the results obtained with the field evaluation of two new varieties of C. esculenta var. Kau-Shiung 1 (KSH1) and PSB-G2 (tissue culture) introduced in the Marshall Islands from Taiwan/Republic of China (ROC) and Secretariat of the Pacific Community (SPC). Le taro est l’un des principaux végétaux cultivés aux Îles Marshall. On y cultive et consomme différents cultivars de taro locaux et introduits : Colocasia esculenta (‘kotak’), Cyrtosperma chamissonis (‘iraj’), Alocasia macrorrhiza (‘wild wot’) et Xanthosoma sagitifolium (‘wot’ ou ‘wuthin kabilon’). Au cours des vingt dernières années, la consommation accrue d’aliments importés et la dépendance qu’ils ont engendrée dans l’ensemble du pays ont eu de profondes répercussions sur les habitudes de production et de consommation du taro. Les volumes produits ont notamment chuté de manière vertigineuse au cours des dernières années au profit des denrées importées. Les Îles Marshall sont exemptes de certains organismes très nuisibles tels que Phytophthora colocasiae, la flétrissure des feuilles de taro. Néanmoins, le pays reste vulnérable face à leur éventuelle introduction par voie aérienne et maritime. La superficie limitée des terres arables, la faible altitude, les projections d’eau de mer, la faible teneur du sol en nutriments et l’insuffisance de matériel végétal limitent la production de taro. Le recours à la culture tissulaire, qui permettrait d’introduire dans le pays du matériel exempt de tout agent pathogène, réduirait considérablement les risques d’introduction de maladies et d’organismes nuisibles nouveaux. Cet article décrit les résultats obtenus à la suite d’une évaluation en parcelle de deux nouvelles variétés de C. esculenta var. Kau-Shiung 1 (KSH1) et PSBG2 (culture tissulaire), introduites aux Îles Marshall par un groupe d’experts de Taiwan et du Secrétariat général de la Communauté du Pacifique. 120 third taro symposium Recent research on taro production in New Zealand Bilan des dernières recherches menées sur la production de taro en Nouvelle-Zélande W. T. Bussell, J.J.C. Scheffer and J.A. Douglas W. T. Bussell, J.J.C. Scheffer et J.A. Douglas Research on taro production in New Zealand in the past decade has shown that it is not possible to grow Pacific Island cultivars with large (> 1 kg) corms and high dry matter content (c. 30%) in this country. However, smaller corms (< 400g) with high dry matter content (c. 30%) can be grown. The main environmental constraint for these cultivars is probably low temperatures. Japanese cultivars, with small corms and cormels, a dry matter content of c. 20% and an acceptable eating quality for Asian people but probably not for Pacific Island people, have been successfully grown in trials in New Zealand. Commercial production is in its infancy and a number of unresolved problems still need to be overcome. These include weed control, sprouting of secondary cormels, and post-harvest infection of cormels by bacteria and fungi. High quality leaves for palusami and other dishes have been very successfully grown in plastic tunnel houses during late spring, summer and autumn. Les recherches menées sur la production de taro en Nouvelle-Zélande au cours des dix dernières années ont montré qu’il n’est pas possible de faire pousser des cultivars océaniens présentant de grands cormes (plus d’un kilo) et une teneur élevée en matière sèche (30 % environ) dans ce pays. En revanche, on peut faire pousser des cormes moins lourds (moins de 400 g) à teneur élevée en matière sèche (30 % environ). La principale difficulté posée par l’environnement tient probablement aux basses températures. Des cultivars japonais, à petits cormes et cormelles, à teneur en matière sèche d’environ 20 % et d’une qualité gustative acceptable par les Asiatiques – mais probablement pas par les Océaniens – ont été cultivés avec succès dans le cadre d’essais réalisés en Nouvelle-Zélande. La production commerciale ne fait que commencer, et il reste à surmonter un certain nombre de problèmes, notamment la lutte contre les plantes adventices, la germination de cormelles secondaires et l’infection des cormelles par des bactéries et des champignons après la récolte. On a réussi à faire pousser des feuilles de grande qualité, utilisées dans la préparation du palusami et d’autres plats, dans des serres en plastique, à la fin du printemps, en été et en automne. Taro production in Australia La production de taro en Australie Peter Salleras Peter Salleras Australian consumers have access to taro imported from Fiji plus an estimated 1500 tonnes of local product. Most Australian taro production occurs in the warmer east coast regions, under irrigation. Growers have access to very efficient communication and transport systems. Obstacles growers face include weed control, relatively old/poor soils, erratic rainfall, and insect and vertebrate pests. The many forms of taro offer us a multitude of tastes, textures, consistencies and growing requirements to work with. There are many talented people associated with the crop, from the paddock to the plate. The prospects for the Australian taro industry are excellent. Les Australiens consomment du taro importé des Îles Fidji et environ 1 500 tonnes de taro local. Le taro australien est surtout cultivé sous irrigation dans les régions de la côte orientale, plus chaudes. Les cultivateurs bénéficient de systèmes de communication et de transport très efficaces. Ils se heurtent à certains obstacles : lutte contre les plantes adventices, sols relativement vieux et pauvres, irrégularité des pluies, insectes et ravageurs vertébrés. Les nombreuses variétés de taro offrent une multitude de goûts, de textures, de consistance et de qualités essentielles dont on peut jouer. Cette culture occupe de nombreuses personnes de talent, depuis le « terrain jusqu’à l’assiette ». Les perspectives de la filière australienne du taro sont excellentes. Comparison of taro production and constraints between West Africa and the Pacific Étude comparée des productions de taro et des contraintes observées en Afrique occidentale et dans le Pacifique Kwadwo Ofori The highest contribution of taro in dietary energy is in the Pacific Islands, but the largest area of cultivation and highest production is in West Africa. These two regions have their similarities and differences with respect to production systems, constraints to production and prospects for improved production levels. In the Pacific Islands, diseases, pests, use of low yielding cultivars, poor crop husbandry and socio-economic problems such as scarcity of land, shortage of farm labour and urbanization adversely affect taro production. In West Africa, the factors militating against growth of the taro industry include erratic rainfall, pests and diseases, limited allocation of Kwadwo Ofori C’est dans le Pacifique que la contribution du taro à l’apport énergétique quotidien est la plus élevée, mais l’Afrique occidentale l’emporte en termes de superficies cultivées et de volumes produits. Les systèmes de culture, les contraintes et les perspectives d’amélioration des niveaux de production des deux régions présentent des similitudes et des différences. Dans les îles du Pacifique, plusieurs phénomènes grèvent la production de taro : la présence d’organismes nuisibles, l’utilisation de cultivars de faible rendement, de mauvaises pratiques culturales et des problèmes socio-économiques tels que le manque de terres exploitables, la pénurie de main d’œuvre third taro symposium 121 resources, increased dependence on cereals for dietary energy, unfavourable competition of taro against other tuber crops, inefficient marketing and limited and uncoordinated research on taro. Taro plays an important role in staple food supply to populations in both regions. Policy makers should make use of their comparative advantages to facilitate increased taro production for food sufficiency, food security, improved livelihood and socio-economic status of large vulnerable populations in these regions. A unified effort aimed at exploiting each other’s experiences would provide a useful impetus for increased production. Research should aim at sustainable technologies that bring improvement strategies to the farmers’ level. This calls for research activities which recognize the farmer as a close collaborator rather than as a client. Development of novel food and non-food products with clearly defined markets would stimulate intensive and/or expanded production of taro in the Pacific and West Africa. agricole et l’urbanisation. En Afrique occidentale, les facteurs entravant la croissance du secteur du taro sont liés à l’irrégularité des pluies, la présence d’organismes nuisibles, le manque de ressources, la dépendance accrue vis-à-vis des céréales qui fournissent une grosse part de l’apport énergétique, la concurrence défavorable entre le taro et d’autres légumes-tubercules, des méthodes de vente inadaptées et le manque de recherches coordonnées. Pourtant, le taro occupe une place importante parmi les denrées alimentaires essentielles des deux régions. Les responsables politiques devraient mettre à profit l’avantage comparatif de leurs pays pour encourager la production de taro afin de renforcer leur autosuffisance et leur sécurité alimentaires, et d’améliorer le quotidien et la situation socio-économique des populations vulnérables, nombreuses dans ces régions. Des initiatives conjointes visant à mettre en commun les expériences de chacun stimuleraient efficacement la production de taro. La recherche devrait se concentrer sur la mise au point de techniques durables susceptibles d’améliorer les pratiques culturales, par le biais d’activités conférant aux cultivateurs un rôle de collaborateur plutôt que de simple utilisateur. L’élaboration d’aliments et de produits non alimentaires nouveaux destinés à des marchés clairement définis stimulerait la production intensive ou extensive de taro dans le Pacifique et en Afrique occidentale. Taro production, constraints and future research and development programme in Indonesia Production de taro, contraintes et perspectives en matière de recherche et de développement en Indonésie T.K. Prana, Made Sri Prana, and T. Kuswara TANSAO - short for Taro Network for South East Asia and Oceania - is a joint project on taro involving two European countries (France and Netherlands), five Asian countries (Indonesia, Malaysia, Thailand, Philippines, and Vietnam), and two South Pacific countries (Vanuatu and PNG) launched in 1998 for the period of three years. Through the project, various studies were conducted and interesting results were obtained. A total of over 700 samples were collected from various parts of the country, representing 181 zymotypes. Lots of morphological and physiological variations were observed among the samples and some promising cultivars/clones were identified too. The project has also managed to initiate coordination among institutions dealing with taro in the country which would be quite useful for strengthening collaborative effort in the future. As taro is not very much used as staple food in Indonesia (except in West Papua/ Irian Jaya and the Mentawai Islands) a different strategy should be set up in working out research and development plan for the future. This, among others, includes product development, breeding and selection of cultivars suitable for the various types of product, and promotion of tarobased products. 122 third taro symposium T.K. Prana, Made Sri Prana et T. Kuswara Le sigle TANSAO désigne le Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie, un projet conjoint lancé en 1998 pour une période de trois ans. Y ont participé deux pays européens (la France et les Pays-Bas), 5 pays asiatiques (l’Indonésie, la Malaisie, la Thaïlande, les Philippines et le Vietnam) et deux pays du Pacifique Sud (Vanuatu et la Papouasie-NouvelleGuinée). Les études réalisées dans le cadre du projet ont produit des résultats intéressants. Au total, plus de 700 échantillons, représentant 181 zymotypes, ont été prélevés dans différentes parties de l’Indonésie. On a constaté de nombreuses variations morphologiques et physiologiques entre les échantillons. Le projet a également permis de sélectionner des cultivars/clones très prometteurs. Il a aussi marqué le début d’un programme de coordination entre les différents organismes de recherche sur le taro, et devrait contribuer au renforcement des liens de collaboration futurs. Le taro n’occupant pas une place centrale dans l’alimentation des Indonésiens (sauf en Papouasie occidentale/Irian Jaya et dans les îles de Mentawai), il faudra recentrer les activités de recherche et de développement du réseau pour les années à venir sur l’élaboration de nouveaux produits, l’amélioration et la sélection de cultivars adaptés et la promotion de produits fabriqués à base de taro. Taro production, constraints and research in Cuba Le taro : production, contraintes et recherche à Cuba Arlene Rodríguez-Manzano, Adolfo A. RodríguezNodals, Leonor Castiñeiras-Alfonso, Zoila Fundora-Mayor and Adolfo Rodríguez-Manzano Arlene Rodríguez-Manzano, Adolfo A. RodríguezNodals, Leonor Castiñeiras-Alfonso, Zoila Fundora-Mayor et Adolfo Rodríguez-Manzano In Cuba there there are two types of “malanga” under cultivation: the so called “malanga”, corresponding to Xanthosoma spp, and the “malanga isleña,” or “taro” (Colocasia esculenta (L.) Schott). Though basically used in the same way, preference for these two crops varies among different regions of the country and different ethnicities. This clearly affects taro demand and production in Cuba. The highest taro production was reached in 1979, with 224 700 tons, decreasing considerably to 4640 tons by 2000. Production has stayed at this level for the past two years. However, the Ministry of Agriculture is interested in increasing the area cultivated with taro in the country. The main constraint for this is financial, and this is also the main factor contributing to the yield decrease. Nevertheless, the production obtained by small farmers in “conucos” (home gardens), has been significant, as revealed by the information compiled through interviews carried out in three regions of the country. The last decade’s Urban Agriculture Program promoted the crop in intensive gardens, under organic conditions, using the traditional clones preferred by local people. In this context, an intensive “seed” production system was developed, supported partially by biotechnologies for accelerated multiplication. Research on the crop includes germplasm collecting, introduction of clones, conservation, characterization and evaluation to increase the efficiency of breeding and production programs, and studies on the origin and evolution of the species. Recent evidence for the presence of a new wild stoloniferous taro, in the eastern provinces of Cuba, opens new research horizons regarding the introduction and evolution of the crop in the Caribbean region. On cultive deux types de «malanga» à Cuba : la «malanga» ou Xanthosoma spp, et la «malanga isleña», également appelée «taro» (Colocasia esculenta (L.) Schott). Bien que les deux espèces soient utilisées de manière similaire, chacune a la préférence de régions et de groupes ethniques distincts. Ce phénomène influe fortement sur la demande et la production de taro à Cuba. En 1979, le pays a atteint une production record de 224 700 tonnes avant de chuter pour ne plus atteindre que 4 640 tonnes en 2000. La production a stagné à ce niveau au cours des deux dernières années. Toutefois, le ministère de l’Agriculture s’emploie à augmenter les terres consacrées à la culture du taro dans le pays. Le principal obstacle qu’il rencontre est de nature financière, et c’est également ce qui contribue au déclin du rendement. Toutefois, la production des petits cultivateurs dans leurs «conucos» (jardins potagers) reste importante, comme l’indiquent les informations recueillies au cours d’entretiens réalisés dans trois régions du pays. Le programme de développement de l’agriculture en milieu urbain mis en œuvre ces dix dernières années a fortement encouragé la culture intensive, dans les potagers, des variétés les plus prisées, selon des méthodes organiques à l’aide des clones traditionnels appréciés par la population locale. Dans ce contexte, un système de production intensive de semences a été mis au point, en partie étayé par des biotechnologies permettant d’accélérer la multiplication. Parmi les activités de recherche engagées, citons la constitution d’une banque de matériel génétique, l’introduction, la conservation, la caractérisation et l’évaluation de clones destinés à accroître l’efficacité des programmes d’amélioration et de production et, enfin, la réalisation d’études sur l’origine et l’évolution des espèces. D’après des données récentes, on aurait observé une nouvelle variété de taro sauvage stolonifère, dans les provinces orientales de Cuba, ce qui ouvre de nouvelles perspectives aux chercheurs en ce qui concerne l’introduction et l’évolution de cette culture dans la région des Caraïbes. Taro (Colocasia esculenta (L.) Schott var. esculenta): Production, constraints and research in Dominica and other Caribbean countries Le taro (Colocasia esculenta (L.) Schott var. esculenta): Production, contraintes et recherche à la Dominique et dans d’autres pays de la région caraïbe Gregory C. Robin Gregory C. Robin Scarring, caused by the removal of suckers attached to the main corm at harvest and determining the optimal time for harvesting corms are the two major constraints to taro production, marketing and export in Dominica and other taro producing islands of the Caribbean. Experiments addressing the above examined the effects of plant depths and spacing on suckering and the effects of plant age on yield, maturity, quality characteristics, shelf life and palatability. The experiments were conducted during the wet and dry season in Grand Bay (agro-ecological zone A2, average annual rainfall 2400 mm) and Wet Area (zone D3, average annual rainfall 5300 mm). Results showed that the average number of suckers per taro corm was more Les deux principales difficultés liées à la production, la commercialisation et l’exportation du taro cultivé à la Dominique et dans d’autres pays producteurs de la région caraïbe sont la présence de cicatrices provoquées par la suppression de drageons sur le corme principal, et l’évaluation de l’âge optimal de récolte des cormes. Au cours d’expériences visant à remédier à ces difficultés, on a étudié, d’une part, la corrélation entre la distance et la profondeur de plantation et la pousse de drageons et, d’autre part, le rapport entre l’âge et le rendement, la maturité, les caractéristiques qualitatives, la durée de conservation et la palatabilité du légume. Les expériences ont été réalisées pendant les saisons sèche et humide à Grand Bay (zone third taro symposium 123 in Wet Area when compared to Grand Bay. Correlation and coefficients of the regression between suckering and scarring in Wet Area were r = 0.8647 (p<0.001) in the wet season and r = 0.4971 (p<0.01) in the dry season. In Grand Bay r = 0.7128 and r = 0.7351 in the wet and dry season respectively were significant (p<0.001). These correlations indicate that factors that reduce suckering would also reduce scarring. For corms harvested between 6 and 12 months, corm weight increased from 706 g to 1265 g in Grand Bay and 560 g to 1094 g in Wet Area. Corm shelf life increased from 17.5 days to 36.9 days in Grand Bay and 14.3 days to 33.8 days in Wet Area. Palatability was best (3.8%) when corms were harvested at 8 months in Grand Bay and at 10 months (4.1%) in Wet Area. Corm dry matter content was highest (41.9%) at 7 months in Grand Bay and (44.5%) for 8-month-old corms in Wet Area. An analysis of the producer, exporter and consumer requirements led to the selection of the following parameters: weight, shelf life, specific gravity, protein content, dry matter and palatability. Using these parameters, the optimal time for harvesting taro corms was determined as 8 and 10 months in Grand Bay and Wet Area respectively. 124 third taro symposium agro-écologique A2, avec une pluviométrie annuelle moyenne de 2 400 mm) et à Wet Area (zone D3, avec une pluviométrie annuelle moyenne de 5 300 mm). Les résultats ont révélé un nombre plus élevé de drageons sur les cormes de taro de Wet area que sur ceux de Grand Bay. La corrélation et les coefficients de régression entre la pousse de drageons et l’apparition de cicatrices à Wet Area ont été établis à r = 0,8647 (p<0,001) pendant la saison des pluies et à r = 0,4971 (p<0,01) pendant la saison sèche. À Grand Bay, les résultats r = 0,7128 et r = 0,7351, obtenus respectivement pendant la saison des pluies et la saison sèche, sont significatifs (p<0,001). Ces coefficients indiquent que les facteurs ayant freiné la pousse de drageons réduiraient aussi l’apparition de cicatrices. En pesant des cormes de 6 à 12 mois, on a déterminé qu’en moyenne, un corme passait de 706 g à 1 265 g à Grand Bay et de 560 g à 1 094 g à Wet Area. On a également observé une augmentation de la durée de conservation des cormes : de 17,5 à 36,9 jours à Grand Bay et de 14,3 à 33,8 jours à Wet Bay. L’appétibilité des cormes a atteint son niveau optimal (3,8 %) à 8 mois à Grand Bay et à 10 mois (4,1 %) à Wet Area. La teneur en matière sèche a atteint son niveau optimal chez les cormes de 7 mois, dans la zone de Grand Bay (41,9 %), et chez les cormes de 8 mois, à Wet Area (44,5 %). Une synthèse des qualités jugées essentielles par les producteurs, les exportateurs et les consommateurs a abouti à l’élaboration d’une grille contenant les paramètres suivants : poids, durée de conservation, gravité spécifique, teneur en protéines, matière sèche et palatabilité. Cette méthode a permis de déterminer l’âge de récolte optimal des cormes : 8 et 9 mois, à Grand Bay et Wet area, respectivement. Theme Three Paper 3.1 Taro as the foundation of Pacific food security Nancy J. Pollock Development Studies, Victoria University, Wellington, New Zealand E sau a le fuauli e to’a ai le moa [The taro (fuauli) will always bring with it a repleted/well satisfied feeling as expressed by the word to’a in the moa (stomach)] Courtesy G.A. Hunkin 15/4/03 Introduction Taro remains a strong cultural symbol for many Pacific nations, both practically and ideologically. It has long been the basis of food security with its many varieties providing alternatives in a disaster (Pollock, 2002). Its perpetuation as a local food is based on a deep knowledge of both planting and processing techniques, as well as very discriminatory tastes. Support that enables availability for consumption to continue is vital. Taro has a major role to play in the future food supply of many Pacific island states. Consumption has been threatened in the past by natural disasters and disease such as taro blight, as well as by colonial cash cropping. Pacific peoples developed techniques for dealing with the natural hazards, but were less successful in stemming the onslaught of cash crops that infiltrated large areas of their best lands. Today that onslaught continues through the import of western foods, such as rice and flour based foods, as well as fast foods. These are fast to prepare, and sell at prices that undercut farmers’ costs of getting taro to markets. In this paper I argue that support for taro’s place as the most significant local food crop is integral to the re-establishment of food security in Pacific nations. Local foods not only are essential to maintain a healthy diet, developed over many years, but they also empower households through subsistence farming. Techniques for overcoming the natural hazards, developed over time, present a lower risk than the price fluctuations and negative values of some imported foods. For the poor, diversity of local foods provides a cheaper, and more readily available alternative to imported foods. Taro thus contributes to greater security, both nutritionally and economically. Taro and biodiversity Taro cultivation has been purposefully perpetuated over 3000 years in the Pacific. Not only did Pacific peoples bring with them out of South East Asia chosen varieties of taro, but they extended the range of Colocasia species by adding other species, particularly Alocasia, Cyrtosperma and much later Xanthosoma. By carrying gifts of taro plants as they voyaged and visited with neighbouring islands they further diversified those plants that had desirable attributes for taste, seasonality and environmental suitability (Pollock, 1992). Since taro can be reproduced only vegetatively, human selection criteria have been the basis for the spread of these plants. By carrying whole plants they were able to eat the corms, and any young leaves, and plant the tops in their next place of land-fall. Controlled planting thus represented cultural choice which has been ever enlarging as populations settled and then exchanged taro as gifts. The human element in the spread of taro across the Pacific was thus in marked contrast to the spread of seed crops such as rice and wheat. The number of varieties of taro was found remarkable by early writers on the botany of Pacific islands, such as Seemann (1862) for Fiji and Handy et al. (1972) for Hawai’i. The 72 varieties noted in Hawai’I enabled that population to provide for the many eventualities that could reduce their food supply, such as disease and drought, as well as to increasingly diversify the taste of taro, e.g. fermented as poi. The loss of those varieties is imminent though projects that preserve some of those remaining varieties, such as the garden in Manoa, are very timely. Various species of taro contribute vitally to the starch staple food supply. The term “taro” is used to refer to four variants., of which Colocasia is the most widely accepted. In addition Dioscorea yams, sweet potatoes and more recently cassava were important root foods in Pacific societies. Add to those breadfruit, bananas, and pandanus, and we have a wide array of starchy foods which were available for selection by the cooks (see Pollock, 1992: Appendix A). This diversity provided the major basis of food security. Colocasia taro is pre-eminent amongst these, along with Dioscorea yams, in providing both symbolic value, as well as pragmatic value as foods. “A major concern is to maintain the diversity of local food plants before the associated planting knowledge is lost” (Pollock, 2002:279). Local knowledge Vegeculture techniques as well as associated planting developments were vital to maintaining the food supply. The concept of “vegeculture” as developed at the Osaka symposium (Yoshida and Mathews, 2002) includes consumption third taro symposium 125 criteria, selection of varieties for consumption, local naming of varieties, processing required by species and varieties, and suitability for various cultural occasions. Planting developments range from the links between wet and dryland taro, through to plantation agriculture, a well as selection of varieties that suit consumption needs, i.e. seasonality, soil suitability, and uses of other parts of plants, leaves, stems for alternative (non-food) needs. These all contributed to the biological diversity that was part of the planting developments over time (Pollock, 1992) that established a firm basis for food security until recently (Pollock, 2002). As knowledge of the local environment, including rainfall, periodicity of cyclones, and droughts, soils was passed on through the generations, so farmers adapted those techniques to suit local circumstances. Ensuring a food supply was the pre-eminent goal for survival. Knowledge regarding taro production was passed on over many generations. Some still survives, but much is being lost. The development of irrigation for some varieties of taro further enhanced the diversity, as wetland taro met tastes that differed from those of upland taro. On atolls, the depth of the pits in which taro grew was based on observation and failures due to changes in the salinity of the ground water (see Wiens, 1962). Cyrtosperma (babai) became a highly prized crop with secret cultivation techniques in the Gilberts/Kiribati (Luomala, 1974). Local perceptions of taro and other food plants are not always directly translatable into English. Local values of taro have not been identified clearly enough by early writers to give us the information we need today to regenerate a “taro culture” for each island society. So the holders of that knowledge who are still alive are important to the revival of taro as a foodstuff. Planting calendars, naming of plants, and parts of plants, and local knowledge of pests and diseases, and how to treat them all need urgent records. Together this body of knowledge needs to be kept alive to support taro’s future. Resilience in the face of disasters that severely reduce taro production have been noted (e.g. Connell 1978 for Solomons). More recently a study of the socio-economic consequences of the 1993 taro leaf blight that devastated Samoa’s crop showed how those farmers coped, using their local knowledge (Naidu and Umar, 2001) Drawing on their traditional modes of adjustment, they substituted other foods such as Taamu (Alocasia) and Taro palagi (Xanthosoma) as well as bananas and some rice. Farmers reported they increased their fishing, both for food for the household as well as for income. Samoan people had a positive attitude, and found alternative foods for the time of the blight, as well as alternative sources of income. “Thus the traditional diversified farming system, adaptability to different crops and food as well as food preparation practices together with the willingness to explore new livelihoods are the keys to sustainability of agriculture-based communities” (p. 21). Recent information suggests that taro is back at the head of the menu. Past experiences have carried them through this trauma with their major food crop. A concerted effort is needed to preserve the knowledge of the old varieties of Colocasia taro, as the information disappears with former farmers. Included in that knowledge is the selective features of each variety, its growing patterns, and resiliencies to wind and salt water inundation etc. Recipes are also vital, as certain taros were grown for specific feast occasions, or for specific high ranking persons. Certain taros are believed to have particular healing characteristics. Those living away from their home Pacific island have stated their longing for a particular variety of taro to “ease the stomach”. Taro fills the stomach as it eases longings. Healthy eating Taro has been reaffirmed as a healthy component of Pacific diets by nutritionists and health specialists (see Malolo et al. 1999). Its starch granules are smaller than other starches and thus more readily absorbed by the young, the elderly, and those with stomach illnesses. It has a beneficially low value on the glycaemic index (1999:25). It is high in fibre, contains a fair amount of protein and other valuable micro-nutrients. From a Pacific perspective it is healthy because it makes the eater feel full and satisfied (Leota in Pollock and Dixon, 1995). For one hundred years (1860s through l960s) regrettably Europeans derided taro as a primitive food that should be replaced by bread and potatoes. Those early Europeans were unfamiliar with the root and tree crops found in the Pacific so considered them part of the “uncivilized culture”. That the root crops grew easily, and thus did not necessitate “hard work” to maintain the crop also added to this image. Early European settlers pushed their own foods, namely bread and potatoes, as the “good foods” that should replace taro etc. Though they were not successful in banishing taro from the diet they left a legacy that taro was inferior as a foodstuff (and likewise the stigma of agricultural labour) (Pollock 1989 for Fiji). The traditional diet of many Pacific island households relied heavily on the starchy component, such as taro, yams, breadfruit etc. One or two of these formed some 80% of daily intake, with the balance provided by fish, or coconut, or other addition. (This emphasis was and still remains the basis of diets throughout South East Asia. ) Taro is kakana dina – the real food. Eaten together with an accompaniment (I coi), it becomes a meal. Alone it is just a snack. Taro is thus a means of satisfying hunger both physically and mentally. Taro is good food, real food. Taro leaves, especially the young ones, are a highly valued accompaniment, particularly in Samoa and Tonga. Mixed with coconut cream, or wrapped around some corned beef, they provide both additional nutrients, as well as a strong identity as a Pacific island foodstuff. Palusami as the latter dish is known in Samoa, has increased its value particularly for overseas Samoans, who regard it as a “traditional” dish associated with the homeland (see Appendix for a song reflecting this euphoric value). The availability of taro leaves to wrap foods is also a sign of good husbandry, that the people are using their land to provide for everyone’s needs, the household and the extended family. In this sense the whole taro plant, leaves and corm, is a representation of social relationships, from gift giving to community support. 126 third taro symposium Taro was aptly suited to cooking in the earth oven. For daily household use, whole corms were roasted alongside other roots and breadfruit, together with some fish. The corms were cooked whole. They had to be thoroughly cooked in order to avoid giving the eater an “itchy mouth” due to the acridity of raphide crystals. Cooking root crops this way was economical in both use of time, and of firewood. For special occasions, taro could be processed into other forms. By grating the root and adding coconut cream, and wrapping the portions in a banana leaf, a range of “puddings” was created. These too were cooked in the earth oven. They have been labelled as “desserts” in Malolo et al. (1999:31), but for most Pacific societies the meal consisted only of the starch and its accompaniment, one course in English terminology. These prepared package foods were a specialty for feasts and to honour high ranking guests. They provided yet another diversification of taro for consumption. Concern for the maintenance of food supply was ever present in pre-contact times. Taro gained eminence as the “food of honour” that represented the strength and well-being of households. Even today taro should be the main contribution from households to a communal event throughout Polynesia – rice is an everyday food, and does not carry those social significances. Taro, in the form of poi, served together with fish, is still symbolic of Hawaiian foods. Taro remains the icon of Pacific well-being, even though introduced foods are readily available. Such healthy living based on taro can only be maintained if taro is produced and marketed sustainably. Scarcity pushes up the price in the local market-place so only those with sufficient cash can afford to buy it. Competition from imported foods in the market and from cash crops on the land is leading to deteriorating health. The taste for taro persists mainly for older people. Its value in everyone’s diet would not only enhance their health, but also serve to strengthen their identity with Pacific island values. Conclusions Taro is a sad example of the implications of the loss of bio-diversity. As a locally grown foodstuff it has the potential to supplement other foods in daily household use. But it needs support from many sectors, both government and private to ensure a ready supply is available at reasonable cost. The taste for it is still there. The store of local knowledge which supported such a diverse range of tastes of this root, and the agronomic techniques that enabled its provisioning is diminishing rapidly. Growing taro for subsistence or cash is vital to the future security of food supply. It will continue to be eaten alongside rice, but if a household can dig up two or three taro corms for the evening meal, she will save her cash for other expenditures, and the rice will remain for another day. A healthy lifestyle includes both enjoying the taro produced in the home gardens, as well as the work in planting, weeding and harvesting it. And sharing the delicious food with other members of the community and beyond also gives “added value”. Restoring the value of taro by maintaining its diversity, supporting local knowledge of its place in culture, including production techniques, will increase the placement of taro at the centre of a healthy lifestyle. With households empowered through their taro culture, we have a firm base for moving towards increasing food security. That old time security can be recaptured to take a new place within the ways of living of the twenty first century. References Connell, J. 1978. The death of taro: Local response to a change of subsistence crops in the northern Solomon Islands. Mankind 22:445–452. Handy, E.S.C. and Handy, E.G., with Pukui, M. 1972. Native planters in old Hawaii: Their life, lore and environment. Bishop Museum Press, Honolulu. 641 p. Hunkin, A. 2002. Palusami. Poem/song. Luomala, K. 1974. The Cyrtosperma systemic pattern: Aspects of production in the Gilbert Islands. Journal of the Polynesian Society 83(1):14–34. Malolo, M., Matenga-Smith, T. and Hughes, R. 1999. The staples we eat. Secretariat of the Pacific Community, Noumea, New Caledonia. 97 p. Naidu, V. and Umar, M. 2001. Surviving the blight: Socio-economic consequences of taro leaf blight (TLB) disease in Samoa. Institute for Reseach, Extension and Training in Agriculture, Alafua, Samoa. 35 p. Pollock, N.J. 1989. The early development of housekeeping and imports in Fiji. Pacific Studies 12(2):53–82. Pollock, N.J. 1992. These roots remain. Institute for Polynesian Studies and University of Hawaii Press, Honolulu. 298 p. Pollock, N.J. 2002. Vegeculture as food security for Pacific communities. In: Yoshida, S. and Matthews, P. (eds). Vegeculture in Eastern Asia and Oceania. National Museum of Ethnology, Osaka, Japan. Seemann, B. 1973. Viti: An account of a government mission to the Vitian or Fijian Islands, 1860–1861. Dawsons of Pall Mall, Folkestone. 447 p. Wiens, H.J. 1962. Atoll environment and ecology. Yale University Press, New Haven. 532 p. Yoshida, S. and Matthews, P. (eds). 2002. Vegeculture in Eastern Asia and Oceania. National Museum of Ethnology, Osaka, Japan. 335 p. third taro symposium 127 Theme Three Paper 3.2 Taro production in Fiji: Constraints and future prospects Aliki Turagakula Ministry of Agriculture, Sugar and Land Resettlement, Fiji Islands Introduction Taro is grown in Fiji mainly for its edible corm (underground stem) and also for its highly nutritious leaves and young tender stems. It is normally cultivated under upland and dryland conditions on gentle hillslopes and fertile alluvial soils of the wet zones. It is propagated vegetatively by using suckers and tops (headsetts), sometimes overmatured corms are sliced to small setts for rapid seed multiplication in the nursery. The main planting season starts in September and ends in March and this falls within the rainy and warm season, even though plantings can be done throughout the year depending on the variety and the management practices. It normally takes 9 to 12 months to mature for the main export variety, Tausala ni Samoa, and some improved varieties mature earlier after 7 - 8 months. Taro is the most important Fijian staple and also has its traditional significance in the chiefly system of marriages, funerals and religious gatherings. In the recent years, taro has gained tremendous economic importance as a source of income generation and foreign exchange earnings in the form of exports. Taro industry overview Taro is mainly grown at the subsistence level under the traditional method of cultivation. The current policies on export-led growth has encouraged the development of niche markets in countries such as New Zealand, Australia and the West Coast of the United States of America where the Pacific islanders, mainly of Polynesian origin, are most densely populated and predorminantly the biggest consumers of taro. Taro is mainly exported to these niche markets as fresh corm produce and on a smaller scale processed taro is also exported. Commercial taro production, which mainly involves specialised non-village farmers, is concentrated in the Central Division for its fertile land and its close proximity to the Suva and Nausori municipal markets and export trading facilities, and in Taveuni in the Northern Division which supplies about 70% of the total production of the export variety, Tausala ni Samoa, with high post-harvest losses due to poor shipping and handling practices. The increasing demand in urban and export markets particularly in New Zealand has led to increased production to commercial levels. Commercial taro production for exports has also opened up in other outlying islands of Lomaiviti, Kadavu and Lau provinces due to improvements in shipping and storage facilities, transportation and telecommunication services. Some areas of Sigatoka in the intermediate rainfall zones (1,500mm/yr) have also opened up for commercial taro production for exports and occurrences of dry spells with lack of irrigation have contributed to low crop yields and inconsistent supply. The main export variety, Tausala ni Samoa, commands a farmgate price of F$0.80-1.20/kg. Export potential for processed taro has given pressure on the demand for volume and consistency of supply. These markets have become increasingly sophisticated and more competitive for high quality products which when delivered on time command reasonable prices. Demands for fresh taro in the domestic markets, as a source of food security, have also increased as a result of shortage of supply from the predorminantly subsistence growers, increasing urbanisation and seasonal plantings. National taro production (1990 - 2001) National production of taro in Fiji from 1990 to 2001 is shown in Table 1 and the geographical distribution of production is shown in Table 2. These figures include both export and domestic supply. Exports range between 20-30% of national production. Table 1: National taro production in Fiji (1990 - 2001) Year 1990 91 92 93 94 95 96 97 98 99 00 01 Tonnage 8780 8080 5876 5329 8810 21926 22613 23350 25625 25907 36612 27705 Output value F$’000 2634 3232 2938 4263 7342 8406 27135 28020 30750 31088 43934 33246 Source: Ministry of Agriculture, Sugar and Land Resettlement (MASLR), Statistics Division 128 third taro symposium Table 2: Geographical distribution of taro production in Fiji (1998) Division Previous existing area (ha) New planting (ha) Number of farmers Total harvest (ha, tonnes) Area remaining on ground (ha) Central 313 478 2915 431 3995 Western 11 24 97 24 193 10 Northern 804 724 7855 1225 9108 304 Variety: Tausala ni Samoa 315 Eastern 341 379 1262 127 458 309 Total 1469 1605 12129 1807 13754 938 Variety: Mixed varieties Central 543 775 5321 718 7712 517 Western NA NA NA NA NA NA Northern NA NA NA NA NA NA Eastern 312 59 3100 381 2892 287 Total 855 834 8421 1099 10604 804 Grand Total 2334 2439 20550 2906 24358 1742 Source: Extension Division Report, MASLR (1998) Time of planting Main season plantings begin in September and ends in March and this coincides with the rainy and warm part of Fiji’s climate. Offseason plantings begin from April to August when the weather is dry and cool. Most plantings are done in the main season resulting in peak production from June to September for most traditional varieties including Tausala ni Samoa and declines during the rest of the year. Improved varieties which are planted in both seasons are of early maturity with higher crop yields than Tausala ni Samoa and are also exported. Tausala ni Samoa variety is highly susceptible to dry and cool spells and is recommended for main season planting, however, it can maintain premium quality for longer storage periods than most other varieties. In the dry and intermediate rainfall zones plantings on flatland are done with sprinkle irrigation in both seasons, however, traditional hill slopes plantings are done mainly in the main season. Production methods Yields of selected taro varieties were compared under two production methods, traditional hill slopes and mechanised flatland. The yield results of these production methods are shown in Table 3. Table 3: Effect of traditional and mechanised production on taro yield (1991) Cross x variety Clone number Flatland, Koronivia 30.3 Corm yield (tonnes/ha) Flatland, on farm 23.9 Hillslope, on farm 18.0 R16 x Tausala NS 160/32 Maleka Dina Vavai dina x Samoa Hybrid 123/70 30.0 25.0 18.0 R16 x TN Mumu 160/31 Vulaono 29 27 23 S. Hyb x Toakula 110/6 28 24 18 TN Mumu x Tausala NS 191/37 Wararasa 32 25 20 Samoa Hybrid 22 11 18 Tausala NS 13 14 20 Source: Turagakula, Research Division, MASLR (1998) Wararasa showed significant commercial features with yield potential of 30-32 tonnes/ha, early maturing at 7-8 months, large elliptical corms of 1-2 kg, high dry matter content at 30-35% at maturity and profuse suckering ability at 5-6 suckers/plant. Tausala ni Samoa has a potential yield of 12-13 tonnes/ha, it is late maturing at 9-12 months, corms weigh 0.7-2 kg, dry matter content at around 30% and low suckering ability at 3-4 suckers/plant. Flatland mechanised cultivation produced high yields and normally involves two ploughing, two harrowing and ridging 1 metre apart. Suckers are planted in the furrow at 60 cm apart and density of 16,500 plants/ha. The traditional method involves slashing, clearing with cane knives and holes are prepared manually 1 metre apart with a large 2 metre long stick with the pointed end in the soil. The bottom of hole is usually wider in diameter than the top. third taro symposium 129 Economic returns Economic returns of four main cultivation methods of taro were compared and results are shown in Table 4 below. Table 4: Economic returns of four main cultivation methods of taro in Fiji Traditional Animal Drawn Small tractor Large tractor 1889 610 537 413 - 373 207 38 3400 2030 1484 838 Manhours/ha Animal and tractor hours/ha (operator) Production cost/ha F$ Yield (tonnes/ha) 16 12 22 25 Gross returns/ha F$ 17,600 13,200 24,200 27,500 Net Returns/ha F$ 14,200 11,170 22,716 26,662 Rates: Labour rate = F$1.80/hr Animal/small tractor = F$2.50/hr Large tractor = F$3.00/hr Taro farmgate price = F$1.10/kg Source: Turagakula, Research Division, MASLR (1998) The scale of production increases with yields under mechanisation and high returns as compared to the traditional method which is limited in scale and yields. Marketing system There are four main systems of marketing taro in Fiji. (1) Farmer sells at farmgate. The farmer sells his taro produce at his farmgate to middlemen and exporters who are based at the main commercial centres such as Suva, Lautoka, Nadi and Savusavu. Taro sold through this system is mostly exported and requires careful post-harvest handling. Prices are determined by weight of corm with two inches of petioles attached to the corm. Exporters usually have their own contracted farmers who supply to them or they also have their own back-up farms themselves. (2) Farmer sells to the market vendor. Taro farmers who are close to the market sell directly to the market vendors at the market place. Taro is often harvested on Fridays and sold at the market on Saturdays. Market vendors often buy direct from the truckloads early on the Saturday morning before the produce is offloaded from the truck. Prices are negotiated whether in bundles of taro or by weight. (3) Farmer sells own produce at the marketplace. The farmer is allocated a space in the marketplace through the approval of the market master and the market vendors association. In most municipal markets farmers are allowed to sell their own produce only on Fridays and Saturdays and from Mondays to Thursdays, vendors will buy the taro from the farmers and they are the only retailers on the market. Farmers often do their own price research at which they sell their produce. (4) Farmer sells to exporters. Farmers sell export grade taro directly at the exporters shed after prior negotiations. Prices often fluctuate according to exporter’s grading standards. When the price is very low, farmer decides to sell to the vendors at the marketplace. Production constraints Major constraints affecting taro production in Fiji are summarised below: • • • • • • • lack of planting materials; low yields under the traditional system of cultivation, need for mechanisation; lack of qualified research staff particularly on pest and diseases, food technology and mechanisation; lack of credit facilities for the agricultural sector; high post-harvest losses, 60%; threats of Taro beetle and Taro Leaf Blight; lack of industry standards. Future prospects Market demands are high for the export and domestic market. Demands for fresh produce are faced with stiff competition with other countries, however, Fiji has the potential to increase the volume to cater for its export supply as well as for its food security needs. Challenges facing the Taro Industry in Fiji require smart partnership between Government and the private sector including the farmers and exporters and an intensive research programme in collaboration with stakeholders of the industry and regional institutions. Management of plant genetic resources in terms of conservation, utilisation, multiplication and distribution of genetic materials to prevent further genetic erosion and meeting market demands require careful considerations for funding and training of personnel. Whilst production of 130 third taro symposium taro in Fiji have been managed quite effectively promotion of this commodity as well as other agricultural commodities ha been seriously neglected within the system and other non-agricultural commodities have been given more serious attention. References Bamman, H. 1998. Fresh produce market information services: TCP/FIJ/6712. Technical Report, FAO. Beever, D.J. 1998. An evaluation of the post harvest handling system for the marketing of Western Samoan taro in New Zealand DSIR. Unpublished paper. Bureau of Statistics. 1999. Export figures. Dalo profile: A programme for future development of dalo industry. 1985. Fullerton, R.A. and Purea, M. 1982. Report to research advisory committee: Non-refrigerated shipment of taro. Totokuita Research Report, Cook Islands. Grantley, C. Facilitation of Fiji’s private sector-led agricultural growth and diversification strategy. ADB Quarantine Management Review. Hunter, D. and Pouono, K. 1998. Evaluation of exotic taro cultivars for resistance to taro leaf blight, yield and quality in Samoa. Journal of South Pacific Agriculture 5:39–43. Leadbitter, N.J. 1984. Preliminary report on visit to New Zealand to study post harvest vetting in Colocasia esculenta imported from Western Samoa. Unpublished paper. Lebot, V. 1992. Genetic vulnerability of Oceania’s traditional crops. Experimental Agriculture 28:309–323. Low, J. 1986. Constraints and economic return to export marketing of taro: A case study of Western Samoa taro export in the New Zealand market. IRETA, University of the South Pacific, Western Samoa. Unpublished paper. Malaki, I. and Atkinson, W. 1998. Review of taro trade and prospects in the South Pacific. MASLR. 1998a. Extension division report. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. MASLR. 1998b. Flavour of Fiji. In: Store Promotion Report. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. MASLR. 1999a. Extension division report. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. MASLR. 1999b. Fiji Ag-trade price profile. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. MASLR. 1999c. Flavour of Fiji. In: Store Promotion Report. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. MASLR. 1999d. Statistics unit. Ag-Trade Division, Ministry of Agriculture, Sugar and Land Resettlement, Suva, Fiji. Prasad, D. Monthly FOB Prices from 1996–1999: Fiji market. Sydney. Statistics New Zealand. 1998. Turagakula, A. 1998. Current agronomic practices for commercial dalo production in Fiji. Vinning, G. 1998. Management of the strategy for agricultural growth. T.A. No. 2681 – Fiji. Asian Development Bank. Waibuta, U. 1999. Taro distribution channel. Ministry of Agriculture, Sugar and Land Resettlement, Taveuni, Fiji. third taro symposium 131 Theme Three Paper 3.3 Taro cultivation in the Marshall Islands: Problems, persistence and prospects Dilip Nandwani1, M.C. Cheng2, Jimmy Joseph3, Jabukja Aikne1, Arwan Soson1 and Gwo-jong Moh2 1 Agriculture Experiment Station, Cooperative Research and Extension, College of the Marshall Islands, Majuro, Republic of the Marshall Islands 2 Technical Mission of Republic of China, Laura Farm, Majuro 96960, Republic of the Marshall Islands 3 Agriculture Division, Ministry of Resources and Development, Majuro 96960, Republic of the Marshall Islands Introduction The Republic of the Marshall Islands (RMI), comprising of 29 atolls and 5 islands, is located in the north-west equatorial Pacific, about 3790 km west of Honolulu, about 2700 km north of Fiji and about 2200 km east of Guam (Figure 1). The atolls of the Marshall Islands, comprising over 1225 islands and islets, are scattered about in an ocean area of well over 1,000,000 km2. The total land area of the atolls is approx. 171 km2. Atolls represent the traditional desert islands. Their principal features are low altitude, (6-7 feet from sea level), salt spray, nutrient deficient soil, low fertility, and limited ground water. The conditions on the atolls impose considerable restrictions on the range of crops, which can tolerate harsh environments. The economy of the Marshall Islands is mainly United States assistance through the Compact of Free Association, extended to help the country to become self-sufficient. The current compact (financial agreement) ends on 30th September 2003 if no extension of the compact is mutually agreed upon by then. Therefore, RMI is embarking on an economic development strategy, which aims at increased agricultural production. Marshall Islands practices only subsistence farming in an agro-forestry set-up (with a multi layer canopy including taro, banana, breadfruit, coconut and other plants) due to limited land resources. Figure 1: Republic of the Marshall Islands. Courtesy of the Marshallese Cultural Society. Taro is a staple food crop in the Marshall Islands. Local and introduced cultivars of taro, viz Colocasia esculenta (L.) Schott or wet taro (kotak, fig 2), Cyrtosperma chamissonis (Schott) Merr. or giant swamp taro (iraj, Figure 3), Alocasia macrorrhiza (L.) Schott or wild taro (wild wot, Figure 4) and Xanthosoma sagitifolium Schott or dry land taro (wot/ wuthin kabilon, Figure 5) are widely cultivated in the Marshall Islands. The present degree of dependency on imported food and a growing awareness of the benefits of subsistence agriculture has generated interest among the farmers in the RMI to develop taro farms. However, shortage of elite seedlings in adequate numbers is the hurdle to achieve the goal. At present taro farms in the Marshall Islands are small and subsistence type. The traditional way of making new planting is the vegetative propagation through the shoot tips from harvested taro. This restricts the number of available planting materials at a given time. 132 third taro symposium Significant diseases of taro in the Pacific include taro leaf blight (TLB), corm rot of taro, root and corm rot, colocasia bobone disease (CBDV), taro plant hopper, and dasheen mosaic virus (DMV) (Kohler et al., 1997). Marshall Islands are free of some serious pests and diseases of taro such as Phytophthora colocasiae, taro leaf blight. However, the country is vulnerable to many serious pests and diseases due to the easy access via air and sea transportation. Cotton or melon aphid, coconut scale, brown soft scale, mealybug, Egyptian fluted scale, taro leafhopper, and spider mite cutworm are the insect pests and diseases on taro reported from the Marshall Islands (Nafus, 1996). The local population does not consider diseases on taro seriously because of the relatively smaller loss of yield and lack of awareness. Meristem culture is a proven tool to produce plants free of virus diseases (Taylor, 1998). Elite seedlings produced through tissue culture are disease free and vigorous in growth and production. TLB can spread through planting material as the fungus infects leaves and petiole. Shoot tip culture, with proper screening, can produce disease-free seedlings free of TLB. Therefore, tissue culture technique is used to produce disease free plants for distribution to farmers. This paper presents distribution, importance and field trials of introduced and tissue culture cultivars of taro in the Marshall Islands. Materials and methods C. esculenta var. Kau-Shiung 1 (KSH1) The taro variety KSH1 [C. esculenta var. esculenta] was introduced to Marshall Islands at Laura farm, Majuro atoll from Taiwan. A technical mission of the Republic of China (ROC) brought the germplasm in January 2001 and planted it at the farm. KHS1 is a leading variety that is being cultivated in Taiwan and posses gel and soft characters on eating quality. The seedlings were propagated vegetatively through corms. The adaptation of the variety in the atoll soil at the Laura farm was evaluated. The variety mainly is a corm variety with plant height of 1-2 m, which produces much bigger corm and is suitable for the irrigation fields or high precipitation areas. An experimental plot in a completely randomized block design was created. Data was collected during and after the harvesting of crop on the set parameters. Some characteristics of the variety’s performance were evaluated at the Laura farm, including evaluation conducted on the treatments of compost and fertilizers. C. esculenta var. PSB-G2 In October 2002, tissue cultures of var. PSB-G2 were obtained from the Regional Germplasm Centre at the Secretariat of Pacific Community (SPC), Fiji. On receipt of tissue cultures, the plantlets were placed in a growth room at 23±1°C temperature, 60-70% humidity, and 2500 lux light intensity for 16 hrs light period using incandescent lamps and fluorescent tubes, as they had been “traveling” for a few days and exposed to fairly hostile conditions such as darkness and temperature extremes. The plants were gently removed from the culture vessel and adhered agar was removed from the roots. Plants were washed with RO Pure water and then transferred to pots containing finely-sieved soil and vermiculite (ratio 1:1). Precautions were taken to avoid damaging the roots, then plantlets were watered without drowning the plant. Soil and vermiculite mixture was autoclaved for two hours at 121°C. Garden manure and compost was added to the soil mixture. The newly potted plants were kept in a plant growth chamber (Lab-Line Instruments Inc., ILL) for two weeks and then moved to shade house in the nursery (67% shade). Plants were covered by polythene flat covers and vented with holes in order to maintain high humidity levels. Covers were gradually removed from the plants for acclimatization to a reduce humidity level and completely after one to two weeks. Plantlets were watered minimum to avoid the increase the chance of fungal infection. Multiplication and root formation Murashige and Skoog’s (MS) medium (Murashige and Skoog, 1962) with the addition of 100 mg/l myo-inositol, 0.4 mg/l thiamine, 30 g/l sucrose and 1.75 g/l Gelrite at a pH of 5.7 was used for the multiplication of shoot bud cultures. The following three stage system and culture medium was used for maximizing shoot production of var. PSB-G2: Stage 1: MS + 0.5 mg/l TDZ + 3% sucrose Stage 2: MS + 0.8 mg/l BAP + 3% sucrose Stage 3: MS + 0.005 mg/l TDZ + 3% sucrose Hardening and acclimatization of well-developed plantlets was done as mentioned above when new tissue cultures were received. Introduction of taro germplasm Over fifty varieties of Colocasia esculenta (kotak) from the Secretariat of Pacific Community (SPC) were introduced to the Marshall Islands by the Agriculture division in 1980s. Seedlings and corms were brought in the Majuro atoll and planted in the nursery for the vegetative propagation and large- scale multiplication. A few varieties of Colocasia esculenta and Cyrtosperma chamissonis were introduced in the Marshall Islands from Kosrae state, Federated States of Micronesia (FSM) in the 1970’s and planted in Ebon atoll. third taro symposium 133 Results and discussion Colocasia esculenta var. KSH1 Kau-Shiung 1 (KSH1) is adapted in wide range of soil chemical property, which is the range of pH 4.5-9.1. It can be cultured in newly farming lands either in alkali soils, but continuous cropping will decrease the growth rate, and diseases occur easily. For the water, fertility conservation capacity and increasing the quality of taro, supplement of enough compost as base application is evidently important in the sandy soils. As many atolls of Marshall Islands have sandy and alkali soil, compost application and water irrigation or planting during the raining season should be considered. The growth duration of taro is about 8-10 months from planting to harvesting. The leaf area will reach the maximum in 5-6 months after planting, and then the corm begins to form, growing bigger subsequently. The leaf size will become small and and number less simultaneously at the harvesting stage. Evaluation of some characteristics of the variety performance in Laura farm is presented in Table 1. The evaluation was conducted on the treatments of compost and fertilizers. The growth performance of the variety on poor land (none material applied) showed short plant height, less number of petioles and less productivity on corm yield. The treatment of applying fertilizers showed enhanced growth in plant height and petiole number, and mean weight of corm per plant was 1.37 lb - that is three-fold compared to the untreated. Additional compost as base application treatment also increased the corm weight per plant - that is 3.8-fold compared to the untreated. Compost application treatment conducted in Taiwan increased not only the yield performance but upgrade on the quality. The amount of KSH1 taro seedlings propagated had been thousands. Those are planned as extension materials for outer islands in Marshall Islands. Table 1: Some characters of taro variety KSH1 performed in Laura Farm, Majuro Atoll Treatment Plant height (cm) No. of petioles (per plant) Weight of corm (lb/plant) A 111.7 6.4 1.7 B 93.9 6.9 1.37 C 61.1 5.1 0.45 A: Amount of N:P2O5:K2O is 180:100:150 kg/ha, additional compost 10,000 kg/ha as base application B: Amount of N:P2O5:K2O is 180:100:150 kg/ha C: No material applied Swamp taro was formerly a very important starchy, staple food source in Marshall Islands. It reaches heights up to 4-6m and produces huge leaves with tips that point upward. It is cultivated for its swollen, starch-rich, underground parts in man-made, muck filled, and freshwater swamps. In Marshall Islands, rainfall is also abundant throughout the year with an average annual total of 3,650 mm.The period of highest rainfall occurs between June and October, while between January and March the return of stronger trade winds results in reduced rainfall, with February being typically the driest month in Majuro. Throughout the archipelago annual rainfall decreases to the north. In the south, Ebon (4 °N) receives approx. 5,680 mm/year, while at Bikini (11°30’N) the annual rainfall is only 1,450 mm/yr. Taro needs enough water irrigation and high humidity environments. The annual rainfall should be considered as the first factor for the taro cultivation extension project. The strategy on taro extension in the Marshall Islands is summarized as follows: 1. The southern atolls get more rainfall than the north, hence the southern atolls for taro extension would be better than the northern one. 2. Planting period is suitable and cooperated with the beginning of raining season, e.g. April is a suitable month for planting in Majuro atoll. 3. Set up the dripping or sprinkling irrigation system to supply enough water while needed. Colocasia esculenta var. PSB-G2 Fungal contamination in the laboratory is relatively low, but there are always problems with bacterial contamination. Generally this tends to be bacteria originating from the internal tissues of the plant, rather than being introduced from outside. Tissue cultures received at the stage II or III were further scaled-up for the shoot bud and plantlets production. Rooted plantlets were transferred to the soil after hardening in the plant growth chamber. Plantlets produced through tissue culture performed well in the field. Plantlets established in the soil and survived up to five months after transfer. At least eleven plantlets survived from the total twelve plantlets transferred in the soil. No infection of insect pest and disease was observed in tissue culture plants; however, taro aphids and mites were observed in the control plants, raised through conventional method of propagation. Data are recorded on plant growth, corm production and quality performance. Today, one of the most intensively studied areas of tissue culture is the concept of selecting disease, insect, or stress resistant plants through tissue culture. Significant gains in the adaptability of many species have been obtained by selecting and propagating superior individuals; the search for these superior individuals can be tremendously accelerated using in vitro systems. The tissue culture facilities at CRECMI are quite satisfactory. Funding was obtained in part from the USDA Land Grant and RMI government. A nursery with screen house and polyhouse is functional with a watering system. The research facility includes a well-equipped tissue culture laboratory and growth room. 134 third taro symposium Various reports on diseases of taro in the Pacific island countries are available. They include Cladosporium colocasiae, corm and leaf rot (Marasmiellus stenophyllus), orange ghost spot (Neojohnstonia colocasie), leaf spot (Phoma spp), leaf blotch (Pseudocercospora colocasiae), corm rot (Pythium spp), root and corm rot (Hirschmaniella miticausa), colocasia bobone disease (CBD virus), dasheen mosaic virus (DMV) and brown root and collar rot (Phellinus noxius). (Kohler et al., 1997). Our survey of taro in Marshall Islands revealed that major diseases on taro in Marshall Islands are taro leafhopper affecting leaf lamina and petiole (Muniappan and Nandwani, 2002). Studies have already taken place in Micronesia on tissue culture and disease resistance of taro (Taylor, 1998). Prof. Wall at the University of Guam tissue cultured and field evaluated 29 varieties of taro collected from Thailand, American Samoa, Yap and Pohnpei, and conducted field performance evaluation in Guam. He reported 8 varieties resistant to TLB. In the South Pacific the TAROGEN project did more exhaustive studies along with ADAP partnership from the University of Hawaii. Reports on micropropagation of various taro varieties are available (Chand et al., 1999; Mixwagner, 1993; Ebida, 1995; McCartan et al., 1996; Zettler et al., 1991). In recent years significant loses have occurred in field collections of root and tuber crops maintained in many of the Pacific Island countries. The extended use of in vitro conservation is now being considered as a more secure strategy for safeguarding these genetic resources. Attempts to develop the use of biotechnology for propagation and conservation purposes are also ongoing in the South Pacific region (Taylor, 1998). Germplasm conservation Taro is a major staple in the wet lowlands of RMI. The main islands in RMI for taro cultivation are Ebon atoll, Mejit Island, Milli atoll, Majuro atoll, Namdrik atoll etc. The losses in the field collection are continual due to the drought in the past, insect pests, lack of planting material and inadequate cultivation practices. Therefore, there is a need for the conservation strategies for taro germplasm. With a tissue culture laboratory on the station the accessions could be maintained in vitro, thus adding to the security of the collection. Micropropagation Smaller farmers in RMI may encounter a shortage of planting materials. Not only can it be a problem of volume but also what is available can be heavily infected. Tissue culture in combination with an effective virus-testing system could solve this problem for the farmers. Both varieties of Colocasia esculenta, var. PSB-G2 from the SPC and var. KSH1 from Taiwan/ROC, have shown good tolerance to insect pests and diseases and are being bulked up for evaluation in the Marshall Islands. There is a need to characterize taro accessions now maintained in the field collections, and to re-establish these collections in vitro due to genetic erosion and disappearing the taro varieties. Initially the number of taro accessions introduced and maintained in the field were more than 50 in 1980’s and now there are less than 15 accessions are known and being cultivated. It was felt that the original collection could have been restored if there had been a duplicate collection maintained in vitro. Storage in vitro offers several advantages: low labour intensity; no infection from pests and diseased; no weed competition; provision of optimal growth conditions; ability to store vegetatively propagated crops; less space required. Origin of taro Cyrtosperma (iraj) is the native taro of the Marshall Islands. Colocasia (kotak) was introduced into the Marshalls from Kosrae state, Federated States of Micronesia (FSM) by the early missionaries. Xanthosoma (wot) was introduced to the islands probably 200 years ago from Hawaii. Alocasia (wild wot) is usually acrid and itchy and not popular for food consumption. According to Marshallese legend, taro was introduced to the islands from heaven when two brothers descended to earth each carrying a full basket of taro. They first visited the island of Namu where Liwaitonmour (female) lived. The two brothers wished to present the taro as a gift to her but she spurned them and their gifts so they decided to look elsewhere. Finally they came to Majuro where they planted the taro there. The young brother later moved to Aur and introduced the taro there. According to legend, Majuro atoll was the first place in the Marshall Is lands to grow taro. Decline in taro cultivation In traditional times, taro was a much more important food crop than it is today and may have been as important or even more so than breadfruit and pandanus. Today, in most of the islands taro has become secondary food crop. There appear to be a number of reasons for the decline of taro as a food crop. On some islands, it is claimed that the pigs introduced by the missionaries in the early days destroyed the taro beds. On Arno atolls, the typhoons of 1905 and 1908 are held responsible for killing most of the taro when the pits were flooded with salt water. Another reason for the decline of taro in some parts of the Marshalls was the extension of the coconut groves under the Germans. Many of the old taro pits, it is said, were planted over with coconuts for the production of copra. The emphasis also on the copra trade by the Japanese, which resulted in more cash income and sale of imported food products, continued the decline in the growing of taro. More recently, in some islands during the war years, all the taro was eaten during the last days of the war by Japanese soldiers and has never been replanted (Soucie, 1976). third taro symposium 135 Table 2: Taro distribution and Geographical characteristics of the various atoll of the Republic of the Marshall Islands Location (Lat. & Long.) Atoll No. of islets Land area (km2) Lagoon area (km2) Taro species Area Rank Area Rank A XA C CY 25 2.80 19 105.96 19 X X X X 52 14.69 3 750.29 6 √ √ √ √ 169°52’ 35 5.36 16 177.34 17 √ √ √ √ 7°10’ 171°40’ 83 12.95 4 338.69 12 √ √ √ √ Aur 8°12’ 171°06’ 42 5.62 15 239.78 14 √ √ √ √ Bikar 12°15’ 170°6’ 6 0.49 30 37.40 26 − − − − Bikini 11°30’ 165°25’ 36 6.01 12 594.14 9 X X X X Ebon 4°38’ 168°40’ 22 5.75 14 103.83 20 √ √ √ √ Enewetak 11°30’ 162°20’ 40 5.85 13 1004.89 3 − − − − Erikup 9°08’ 170°00’ 14 1.53 25 230.30 15 − − − − Jabwat 7°44’ 168°59’ 1 0.57 29 – 32 √ √ √ √ Jaluit 6°00’ 169°34’ 84 11.34 5 689.74 7 X X X X Jamo 10°07’ 169°33’ 1 0.16 31 – 33 − − − − Kili 5°37’ 169°07’ 1 0.93 28 – 31 X X X X Kwajalein 9°00’ 166°05’ 93 16.39 1 2173.78 1 X X X X Lae 8°56’ 166°30’ 17 1.45 26 17.66 27 √ X X √ Lib 8°21’ 167°40’ 1 0.93 28 – 31 X X X X Likiep 9°54’ 169°10’ 64 10.26 6 424.01 10 √ √ X √ Majuro 7°03’ 171°30’ 64 9.17 8 295.05 13 √ √ √ √ Maloelap 8°40’ 171°00’ 71 9.82 7 972.72 4 √ √ √ √ Mejit 10°17’ 170°52’ 1 1.86 22 – 30 √ √ √ √ Milli 6°05’ 171°55’ 84 14.94 2 759.85 5 √ √ √ √ Nadikdik 6°20’ 172°10’ 18 0.98 27 3.42 29 − − − − Namo 7°55’ 168°30’ 51 6.27 11 397.64 11 √ √ X √ Namorik 5°37’ 168°07’ 2 2.77 20 8.42 28 √ √ √ √ Rongelap 11°19’ 166°50’ 61 7.95 10 1004.32 2 − − − − Rongerik 11°20’ 167°27’ 17 1.68 24 143.95 18 − − − − Taka 11°18’ 169°35’ 5 0.57 29 93.14 22 − − − − Taongi 14°32’ 169°00’ 11 3.24 18 78.04 23 − − − − Ujae 9°00’ 165°45’ 14 1.86 22 185.94 16 √ − − − Ujelang 9°50’ 160°55’ 32 1.74 23 65.97 24 − − − − Utirik 11°12’ 169°47’ 6 2.43 21 57.73 25 − − − − Wotho 10°05’ 165°50’ 13 4.33 17 94.92 21 √ X X √ Wotje 9°26’ 170°00’ 72 8.18 9 624.34 8 √ √ √ √ North East Ailinginae 11°10’ 166°20’ Ailinglaplap 7°26’ 169°00’ Ailuk 10°20’ Arno X = absent √ = present − = not available A=Alocasia XA=Xanthosoma C=Colocasia CY=Cyrtosperma Conclusion Several varieties were introduced in the Marshall Islands from Fiji by agriculture division in the 1980s. However, drought and typhoon in the 1990 caused severe damage to the germplasm and no documented information is available on the introduced varieties and their characteristics. Copra, for most outer islanders still the sole means of a cash income apart from the handicraft production, has become a less and lucrative commodity due to low price in the world market. Lowering of expenditure by import substitution is a feasible option. The study shows that taro does well under coconut. It is a suitable intercrop in copra plantations. The tissue culture unit of the Agriculture Experiment Station at the College of the Marshall Islands has necessary infrastructure and expertise to carryout the above-mentioned tasks including tissue culture and large-scale propagation experiments. Acknowledgments The authors wish to acknowledge Dr Wayne Schmidt, Dr Singeo Singeru, Diane Myazoe and RMI Government for their continued and generous support for the Land Grant programs. The senior author wishes to thank to Dr Mary Taylor and Tom Osborn for providing the tissue culture germplasm of taro var. PSB-G2. Financial support received from the symposium organizers for attending the conference is gratefully acknowledged. 136 third taro symposium References Chand, H., Pearson, M.N. and Lovell, P.H. 1999. Rapid vegetative multiplication in Colocasia esculenta (L.) Schott (taro). Plant Cell, Tissue and Organ Culture 55(3):223–226. Ebida, A.I.A. 1995. In vitro propagation and in vivo establishment of the Egyptian taro, Colocasia esculenta var. esculenta (L.) Schott (Araceae). Alexandria Journal of Agricultural Research 40(3):457–474. Kohler, F., Pelligrin, F., Jackson, G. and McKenzie, E. 1997. Diseases of cultivated crops in Pacific Island countries. South Pacific Commission, Noumea, New Caledonia. 187 p. McCartan, S.A., Staden, J., and Van Finnie, J.F. 1996. In vitro propagation of taro (Colocasia esculenta). Journal of the Southern African Society for Horticultural Sciences 6(1):1–3. Mixwagner, G. 1993. In vitro multiplication of yam (Dioscorea rotundata) and taro (Colocasia esculenta L.) for planting material production [in German]. Landbauforschung Volkenrode 43(2–3):93–100. Muniappan, R. and Nandwani, D. 2002. Survey of arthropod pests and invasive weeds in the Republic of the Marshall Islands. CRECMI Publication #1. College of the Marshall Islands, Majuro, Republic of the Marshall Islands. Murashige, T. and Skoog, F. 1962. A revised medium for rapid growth and bioassays tobacco tissue cultures. Physiologia Plantarum 15:473–497. Nafus, D.M. 1996. An insect survey of the Marshall Islands. Technical Paper No. 208. South Pacific Commission, Noumea, New Caledonia. 35 p. Soucie, E.A. 1976. Taro: Tropical agriculture. Ponape Agriculture & Trade School (PATS), Ponape, Federated States of Micronesia. 157 p. Taylor, M. 1998. Biotechnology in the South Pacific island region. Acta Horticulturae 461:55–64. Zettler, F.W., Hartman, R.D. and Logan, A.E. 1991. Feasibility of producing pathogen-free aroid root crops commercially by micropropagation. p. 80–85. In: Proceedings of the 26th Annual Meeting of the Caribbean Food Crops Society, 29 July–4 August 1990, Mayaguez, Puerto Rico. Caribbean Food Crops Society, Mayaguez, Puerto Rico. third taro symposium 137 Figure 2: Colocasia esculenta (kotak) Figure 3: Cyrtosperma chamissonis (iraj) Figure 4: Alocasia macrorrhiza (wild wot) Figure 5: Xanthosoma sagitifolium (wot or wuthin kabilon) Figure 6: Corms of C. esculenta var. KSH1 Figure 8: Taro leafhopper 138 third taro symposium Figure 7: C. esculenta var. PSB-G2 (tissue culture) Figure 9: Taro aphids (Aphis gossypii) and mites Theme Three Paper 3.4 Recent research on taro production in New Zealand W.T. Bussell1, J.J.C. Scheffer2 and J.A. Douglas3 School of Landscape and Plant Science, UNITEC Institute of Technology, Auckland, New Zealand 2 New Zealand Institute for Crop & Food Research Ltd, Pukekohe Research Centre, Cronin Road, RD 1, Pukekohe, New Zealand 3 New Zealand Institute for Crop & Food Research Ltd, c/o Ruakura Agricultural Centre, Private Bag 3123, Hamilton, New Zealand 1 Introduction Recent research on taro (Colocasia esculenta (L.) Schott) in New Zealand has been partly stimulated by interest in maintaining a regular local supply of high quality ‘pink taro’ for the large Samoan population in New Zealand. Imports of this cultivar fell sharply for a time after the Samoan crop was devastated by taro leaf blight in the mid 1990’s. Trials by Bussell and Goldsmith (1998, 1999) were conducted to find if the Pacific island cultivars Niue (pink taro) and Ni Tonga (white taro) would produce large (c. 1 kg weight) corms with a high (c. 30%) dry matter content in the Auckland region. These trials were the first reported attempt to produce large corms from Pacific island cultivars in New Zealand in recent times. Pacific island taro was first introduced by Maori c. 800 years ago but they only flourished in limited locations (Best, 1925) and have not survived. Modern Pacific island taro cultivars have been grown in New Zealand for many years in the home gardens of Pacific Islanders but leaves are regularly harvested from them, making it impossible (a priori) for the large corms typical of these cultivars to develop. The retail value of imported corms of Pacific island cultivars is currently about NZ$8 million annually. The possibility of producing the more temperate commercial cultivars of lateral-cormel (side-corm) Japanese taro in New Zealand for local and export markets has stimulated recent research on this crop. A temperate climate adapted taro grown by some Maori communities in the North Island of New Zealand is thought to have been introduced by Chinese immigrants in the late 19th century (Matthews, 1985, 2002), but it has not been commercialised. Potential for commercial markets for Japanese taro were seen in the increasing numbers of Asians in New Zealand and in export to Japan in April to June when market prices in Japan are high. Japanese taro is an important traditional staple vegetable in Japan and is grown on 20,000 ha (in 1995) and produced 250,000 t. Unlike Pacific Island cultivars, most Japanese cultivars produce smaller, somewhat sweeter and stickier corms. Instead of producing one main central corm, they develop many side or ‘child’ corms from the ‘seed’ corm, a selected side corm from the previous season. This becomes the primary or ‘mother’ corm for new side corms. During one growth season, these secondary corms may also develop side-corms that are grandchild or tertiary corms. Sometimes fourth generation corms develop on the tertiary corms. The number and shape of the side corms vary depending on the cultivar. Trials have been conducted to develop suitable agronomic practices for commercial production of Japanese taro in New Zealand (Follett and Scheffer, 1996; Scheffer et al., 1999). A third motivation for research has been a desire to increase the quantity of locally grown fresh taro leaves available in New Zealand. Development of this has primarily been undertaken by Pacific Island charitable trusts based in New Zealand. This paper outlines and discusses the recent agronomic research on taro in New Zealand. Production of Pacific Island taro in New Zealand The Auckland area has a cooler and drier climate than Pacific island areas where most of the taro corms have been grown (near Apia) or are now grown (on Taveuni Island) for the New Zealand market (Table 1). In order to produce large corms of the Pacific island cultivars Niue (pink taro) and Ni Tonga (white taro), we tested agronomic practices for promoting the most vigorous plant growth possible. These included planting rooted shoots, not harvesting any young leaves, and irrigating during the driest part of the year. In a tropical or subtropical climate taro can be grown without irrigation (termed upland or dryland cultivation) in places where annual rainfall is at least 2500mm (Purseglove 1972). We therefore hoped that by applying water through overhead sprinklers at rates of about 60 mm/week (equivalent to 3120 mm/year) from 15 November (when soils were starting to dry out) to 15 April (when leaf growth had slowed considerably and rainfall was increasing) we would obtain target corm size and quality. All trial sites were on recent volcanic soils with high natural fertility. No additional nutrients were applied at any site and no deficiency symptoms were observed. third taro symposium 139 Table 1: Monthly averages (30 year means) of mean air temperature and rainfall at Albert Park, Auckland, New Zealand; Apia, Samoa; and Taveuni Island, Fiji Temperature (0 C) Rainfall (mm) Month Auckland Apia Taveuni Auckland Apia Taveuni January 19.8 26.9 27.1 65 410 276 February 20.4 26.9 27.4 96 319 185 March 19.3 26.9 27.3 91 376 255 April 16.9 27.1 26.7 117 237 173 May 14.3 26.8 25.7 124 166 154 June 12.1 26.4 25.1 141 151 108 July 11.2 26.1 24.4 141 122 83 August 11.9 26.1 24.4 139 122 114 September 13.2 26.2 24.7 101 163 96 October 14.8 26.4 25.3 97 252 115 November 16.5 26.6 25.9 89 275 127 December 18.3 26.7 26.7 88 370 154 Planting of the Niue and Ni Tonga cultivars was done in late September or early October, after the danger of frosts had passed, in rows 1m apart with 1m between plants in a row, at a density of 10,000 plants/ha and the standard density for plantings in the Pacific islands. Corms were harvested in our trials at monthly intervals from June to October, 8 to 12 months after planting. In June and July harvests corms were large (1 kg or more) but were not mature in terms of shape (the necks were wide) (Figure 1). In August to October harvests corm weight increased slightly and corm necks had become narrow and appeared similar to necks of mature imported corms (Figure 2). Corm dry matter content ranged from 15% to 23% in all harvests and these corms were regarded as too soft by Pacific Islander taste panelists. We are unable to explain why dry matter content did not reach the high levels achieved in the Pacific islands. The relatively cool summer temperatures together with the regular watering during the main part of the growing period may have affected dry matter content of Pacific island cultivars in our trials. Figure 1: Corms of cv Niue at six months after planting. Ruler in photo is 30 cm long Figure 2: Imported corm of cv Ni Tonga (left) and cv Niue (second from left); 10 month old corm of cv Ni Tonga (second from right) and cv Niue (right). Ruler in photo is 30 cm long. The Niue and Ni Tonga cultivars were grown without irrigation during summer in areas adjacent to some of our trial plantings and in the home gardens of some Pacific Islanders living in Auckland. In all locations planting material was planted in late September or early October and mature plants were ready to harvest in June. Corm weight ranged from 250-400 g. Corm dry matter ranged from 28% to 32%, close to the standard of 30% for markets in Samoa (Rogers et al., 1992). The taste of these New Zealand grown corms was considered very good by Pacific Islander taste panellists. This suggests that water management could be the key to producing large, high dry matter corms in New Zealand and further research is desirable on this point. At present, however, it is unlikely that New Zealand grown pink taro will contribute significantly to the regular supply of taro for Samoans living in New Zealand. Production of Japanese taro in New Zealand A Japanese commercial cultivar (Yamato-wase) was obtained from Kyowa Seed Co. in Japan by Crop & Food Research Ltd in June 1991, and after quarantine for 12 months, it was bulked in the field. Four agronomic trials were conducted at Crop & Food’s research station at Pukekohe, South Auckland, from 1994 to 1998 to investigate the effects of irrigation, size of planting propagule, harvest time, plant density and nitrogen nutrition on cormel yields (Follett and Scheffer, 1996; Scheffer et al., 1999). 140 third taro symposium Seven other Japanese cultivars (akame, ebi-imo, tono-imo, Ishikawa-wase, Celebes, kashira [possibly also known as yatsukashira] and one further cultivar of unknown name) were imported in 1994 (by Crop & Food Research Ltd) and in 1992 and 1995 (by P. J. Matthews). After inspections in quarantine, they were bulked up at Pukekoke and have been maintained there since. A small planting of all the available Japanese cultivars was made at UNITEC in Auckland City in 2000 to obtain preliminary information on corm weights and dry matter contents. The main aim was to find out if any of these cultivars might suit Pacific Islanders’ tastes. Three of the Pukekohe trials were conducted on a Patumahoe clay loam, the other in cold frames with a peat-pumice mixture (Scheffer et al., 1999). In the field experiments the taro was grown, partly sheltered, on flattened ridges like those used for potatoes, in 750 mm rows, with corms placed 50-100 mm deep at 300 mm intervals. Irrigation was applied with drip tape at c. 2 l/m2 every two days. The trials were planted between September and November and harvested between April and October, 6 to 11 months after planting. The crops in all trials were planted and harvested manually. After removal of intact clumps from the ground, the plants were washed with a high-pressure hose (fig 3). The side corms were then broken off and washed again. All side corms with shoots > 50 mm were trimmed leaving 50 mm (measured from the ring of buds on the corm/shoot dividing line) of the base of the shoot attached to the corm. Recordings were made of the number, weight and quality of secondary, tertiary and fourth generation cormels. The parent corms were not recorded. Trial work with the Yamato-wase cultivar has demonstrated that Japanese taro grows well in the Pukekohe region with minimum attention from October to May (Scheffer et al., 1999). With only partial shelter plants grew up to 1.2 m in height. They did show frost tenderness, a -10C screen frost causing immediate blackening of the leaves without killing the plants. This cultivar responded strongly to the application of water over summer. The total cormel (sidecorm) yield from a crop irrigated daily (at 2.5 l/m2) for 190 days was 1.9 kg/plant and significantly higher than the total cormel yield of 1.0 kg/plant of taro irrigated only for the first 80 days. This was primarily due to a significant increase in cormel number from 14/plant in the low water regime to 27/plant in the high water regime. Additional trial work at Pukekohe in 1998/99 confirmed the importance of irrigation (J. Scheffer, unpublished data). When this cultivar was harvested in early autumn (March), the total cormel yield of 0.4 kg/plant was only half that obtained in an early winter (June) harvest. Large treatment differences were also caused by the size of the planting material with small propagules (dormant cormels, 50 mm in diameter) producing 0.6 kg/plant, medium propagules (secondary cormels with trimmed tops, 55 mm diameter) 1.5 kg/plant and large propagules (mother corms with trimmed tops, 70 mm diameter) 2.2 kg/plant respectively. A high density planting of 6.7 plants/m2 produced a total cormel yield of 42 t/ha compared to 28, 30 and 38 t/ha at 2.7, 3.3 and 4.4 plants/m2. Plant density did not affect average cormel weight. Nitrogen applications up to 150 kg/ha increased total cormel yields with a slight reduction at a higher rate. Figure 3: Whole plants of cv Yamato-wase after being dug from the field and washed with a high-pressure hose. Figure 4: Cormels of cv Yamato-wase showing variations in shape and colour (see text) We suggest that, with good nutrition and soil moisture, Yamato-wase should produce about 20 cormels per plant for a total weight of 1 kg. Cormels varied in shape from round to cylindrical, and also had alternating light and brown bands giving the product an attractive appearance (Figure 4). The main factors adversely affecting cormel quality were premature sprouting, and cracking. The apical buds of most secondary corms developed large leafy shoots. Among these shoots, the shoot base was often equal in circumference to the circumference of the corm from which it developed. Such leafy secondary corms usually have thick white fleshy roots, and are not considered marketable as a vegetable, but make excellent planting material. It is understood this sort of premature sprouting of secondary corms does not occur in Japan (P. Matthews, pers. comm.). Only a few tertiary corms developed large shoots and those that develop no shoot or only a short shoot can be marketed as a fresh product. Trials have shown that tertiary corms make up about 40% of the total yield. Although many of the tertiary corms were quite small (< 30 mm), they might still be regarded as a marketable product by Japanese standards. In Japan, at all times of the year but especially in the late winter, it is common to see very small, round cormels sold already peeled and partly cooked, frozen or preserved in a salty brine. These are convenient for adding to soups for a quick meal (P. Matthews, pers. comm.). In New Zealand Japanese small taro cormels are sold in a frozen, processed form by one or two specialist shops. third taro symposium 141 The incidence of cormel cracking was high in all trials from shallow cracking just a few millimetres into the epidermis to deep cracking right into the centre of the cormel. Cracks are undesirable because they tend to be entry points for micro-organisms that cause rotting and thus reduce storage life. Cracking varied with treatment and in the irrigation experiment the incidence of cormel cracking was significantly lower in the high water regime (37%) than in the low water regime (52%). Minimising the incidence of corm cracking is an important quality issue and further research on the influence of cultivar, soil type, fertiliser use, irrigation and time of harvest is needed. Plants of the eight Japanese cultivars grown at UNITEC produced mature corms and cormels at harvest time about 6 months after planting. Mother corm sizes ranged from about 0.2 kg to 1.1 kg. The total number of cormels per plant, total corm yield and the dry matter content varied with cultivar and ranged from 5-25, from 1-3.4 kg, and from 15-22% respectively. The eating quality of New Zealand grown Japanese taro has been evaluated and found acceptable by both Asian and European customers (Scheffer, 1995; P. Matthews, pers. comm.). The soft texture of all Japanese cultivars, due to a low dry matter content, is likely to make them unacceptable to many Pacific island people in New Zealand but this has not yet been thoroughly evaluated. Commercial production of Japanese taro in New Zealand is in its infancy and to date has been limited by lack of planting material. The observations and experience from the trials show that production in New Zealand is possible and larger scale planting is needed to test the market potential both locally and overseas. There are, however, some unresolved issues. Effective weed control is essential for commercial development and currently there are no registered herbicides for use on taro in New Zealand and this needs further research and development. At Pukekohe, simazine (1 kg a.i./ha) has been used experimentally, and apparently safely, both before and after crop establishment, for residual weed control (Scheffer et al., 2000). In addition to simazine, linuron, trifluralin and pendimethalin are used in Japan (M. Kamiya, pers. comm.). In Hawaii oxyfluorfen is used as a post-emergence spray (but corms and leaves cannot be marketed for six months after the last application) and paraquat is used as a post-emergence directed spray (Hollyer et al., 1997). In Cuba, the mixture of terbutryne (1.0 kg a.i./ha) + prometryne (1.6 kg a.i./ha) applied to one to two leaf taro gave effective weed control (Carpio et al., 1982). Japanese taro grown at Pukekohe has been relatively free of fungal diseases. In one trial, the leaf fungal disease (Phyllostica colocasiae) became a problem during cold wet autumn conditions but was successfully controlled with benomyl and iprodine fungicides (Scheffer and Douglas, 2000). Dasheen Mosaic Virus, which is present on Zandedeschia in New Zealand, was identified on some Japanese taro plants (Pearson et al., 1998). These were removed from our trial plots and destroyed. Post-harvest fungal diseases, such as Fusarium solani, and bacterial diseases, particularly Erwinia carotovora, have caused high losses when left uncontrolled (Scheffer and Douglas, 2000). Cormels of New Zealand grown Japanese cultivars will need to be carefully cured before they are stored. Production of taro leaf in New Zealand Taro leaf has been harvested from the home gardens of Pacific Islanders in New Zealand for many years. A few Pacific Island trusts have been developing larger scale production of taro leaf blades in New Zealand in recent years. Their research and experience has shown that cormels (suckers) planted close together (in rows c. 15 cm apart and plants c. 15 cm in the row), and the crop grown in unheated plastic tunnel houses, will produce good quantities of young leaves in a season lasting from October to May. Conclusions Commercial production of Japanese taro is likely to gradually develop in New Zealand as a suitable range of material is now available. Production systems suitable for commercial production need to be perfected and in particular weed control and agronomic management to achieve high quality corms. This and the development of Pacific island high dry matter taro require more research. Leaves of taro produced in New Zealand are of good quality and are now widely sold in the main cities of New Zealand. Care is needed to make sure that the market does not become over-supplied. Acknowledgements Funding for part of the research described in this paper was provided by the Community Employment Group, Department of Labour and by UNITEC Research Fund. We thank Dr Peter Matthews, National Museum of Ethnology, Osaka, Japan for helpful comments and criticism. References Best, E. 1925. Maori agriculture. Elsdon Dominion Museum Bulletin No. 9. Board of Maori Ethnological Research, Wellington, New Zealand. Bussell, W.T. and Goldsmith, Z. 1998. Exotic taro in NZ. New Zealand Commercial Grower 53(8):8,10. Bussell, W.T. and Goldsmith, Z. 1999. Possibilities for production of South Pacific taro in New Zealand. Proceedings of the Agronomy Society of New Zealand 29:31–33. 142 third taro symposium Carpio, N., Meneses, E., Fieltes, A. and Coli, T. 1982. Preliminary results with mixtures of residual herbicides applied to taro (Colocasia esculenta) cv. Islena Japonesa. Ciencia-y-Tecnica-en-la-Agricultura, Viandas Tropicales 5(2):27– 36. Follett, J.M. and Scheffer, J.J.C. 1996. Japanese taro: A New Zealand perspective. International Plant Propagators’ Society Combined Proceedings 46:421–24. Hollyer, J. et al. (eds and contributors). 1997. Taro: Makua to Mokai. University of Hawaii, Honolulu. 108 p. Matthews, P.J. 1985. Nga taro o Aotearoa. Journal of the Polynesian Society 94(3):253–272. Matthews, P.J. 2002. Taro storage systems. p. 135–163. In: Yoshida, S. and Matthews, P.J. (eds). Vegeculture in Eastern Asia and Oceania. National Museum of Ethnology, Osaka, Japan. Pearson, M.N., Bussell, W.T. and Scheffer, J.J.C. 1998. New plant disease record in New Zealand: Dasheen mosaic potyvirus infecting taro (Colocasia esculenta (L.) Schott). New Zealand Journal of Crop and Horticultural Science 26:69–70. Purseglove, J.W. 1972. Tropical crops: Monocotyledons. Longman, London. 607 p. Rogers, S., Rosecrance, R., Chand, K. and Iosefa, T. 1992. Effects of shade and mulch on the growth and dry matter accumulation of taro (Colocasia esculenta (L.) Schott). Journal of South Pacific Agriculture 1(3):1–4. Scheffer, J.J.C. and Douglas, J.A. 2000. A new crop: Taro Japanese variety. New Zealand Commercial Grower 54(6):37–38. Scheffer, J.J.C., Douglas, J.A. and Triggs, C.M. 1999. Preliminary studies of the agronomic requirements of Japanese taro (Colocasia esculenta) in New Zealand. Proceedings of the Agronomy Society of New Zealand 29:41–46. third taro symposium 143 Theme Three Paper 3.5 Taro production in Australia Peter Salleras Taro Grower, Queensland, Australia “What’s the best way to cook taro?” is usually the first question people ask when they find out I’m a grower. Whilst my answers are often clumsy (I’m far from a chef), the question indicates both an interest and an obstacle for taro consumption by non-traditional users in Australia. Our consumers have access to taro imported from Fiji plus an estimated 1500 tonnes of local product. Most Australian taro production occurs in the warmer east coast regions, and in particular between Tully and Babinda in the far north of Queensland. This area is recognised as the “Super Wet Belt” of Queensland’s wet tropical region and is blessed with mild winters and good rainfall, usually averaging over 4000 mm annually. Nevertheless, all commercial taro production is irrigated. Growers have access to very efficient communication and transport systems. Refrigerated “Banana” trucks deliver freshly picked taro to southern city markets 3000-4000 kilometres away within 2 days. Although improved cultural practices will see larger individual producers emerge, the majority of growers currently work plots of 0.5 to 2 hectares. Our industry has seen a high turnover of growers with many finding it “just too hard”. Our major production limitation is labour costs. Even at $15.00 per hour it is difficult to find people willing to endure hot, hard work, made worse by the itchiness caused by taro sap. However, the high labour cost of “Large Corm” taro production will diminish in the near future. Australian taro growers have an innovative mindset, so rapid technological advances are inevitable. Obstacles growers face include weed control, relatively old/poor soils, erratic rainfall, and insect and vertebrate pests. Our main insect pests are Hawkmoth, Heliothis and Cluster caterpillars, mostly controlled with BT (Baccillus thuringiensis). Rats can inflict major damage in taro blocks, particularly where weed control is inadequate. Poison baits are the main control method, although secondary poisoning of rat predators is of concern. Feral pigs can do plenty of damage when they get a taste for taro. Trapping, shooting and electric fencing effectively minimise damage. With production advances come the threat of market gluts and poor prices – every farmer’s nightmare. However, I firmly believe demand for taro will outstrip supply for many decades on a global scale. Taro is currently as unknown to “Western” cuisine as potatoes and tomatoes were when Columbus first returned to Europe from the Americans in 1493. Historians believe acceptance of potatoes and tomatoes took considerable time not only because they were alien tastes but also because they were members of the poisonous “nightshade” family. Although taro is one of the oldest known cultivated plants, its consumption has never spread far outside the tropical regions of the world. Times change, however, and with increasing speed. Affluent consumers are increasingly demanding, and willing to pay for, new tastes and safe, healthy and ethically produced food. Taro fits the bill. It grows best under “natural” production methods. It is a physically bold and upstanding plant with leaves uniquely able to shed water almost contemptuously, yet nurse droplets like silver jewels. In the food value stakes, taro makes most other starchy vegetables and grains look like nutritional paupers. Taro is being replaced in traditional consumption areas by easier, cheaper more “Western”, trendier foods. But really we must wonder, with the rugby world cup the biggest sporting event on the planet in 2003, would mothers of players like Lote Tuqiri and Jonah Lomu have been able to raise their boys to such a level in the international sporting arena , on white rice and fast noodles? Oysters and durian may have aphrodisiac powers but taro is unchallenged for the “natural steroid” reputation. Not everyone wants to be a world class athlete, but there are dramatically increasing numbers of people suffering problems like diabetes and hypoglycemia who would benefit for the nutrition, low glycemic index (GI) and easy digestibility of taro. How do we get more people to eat taro? Years ago, I remember, seemingly every second vehicle in western Queensland was adorned with a sticker which read “YOU’RE IN CATTLE COUNTRY - EAT BEEF YOU BASTARDS!” A sticker costing a few cents on a family’s prized mobile investment can “remind” people for many years: “TARO - PEOPLE POWER ON A PLATE!” The Taro Symposium network is the perfect structure to kick off such simple and cost effective food awareness initiatives on a global scale. Even in Australia, we only need minor changes in eating trends to have substantial impacts on primary production. Australians currently eat about 60 kg of “English” potatoes annually per head and I would assume the figure for grains is considerably higher. If we consider that every 1 kg/head swing to taro equates to 20,000 tonnes, the growth potential, in Australia alone, becomes obvious. Back to “What’s the best way to cook taro?” Taro has properties which allow it to be a culinary performer everywhere from ice-cream to a roast. We recently tried a pepperoni and olive pizza with a base made from boiled and mashed taro with herbs and an egg. Excellent! Whilst today’s food scientists are increasingly allowing foods to fit lifestyles, the celebrity chefs are the real heroes of the fresh produce industry. English chef Jamie Oliver has been a major hit on Australian television with his show 144 third taro symposium “The Naked Chef”. Neil Perry, Australia’s top chef/restaurateur has “Food Source” and “Fresh and Fast” airing on the Lifestyle channel in Australia as well as in the UK and elsewhere. Neil and his crew shot a segment on taro on our farm, which will be aired in 2004. The top chefs and restaurants certainly set trends, and taro is also beginning to appear on the menus of upmarket establishments in Queensland’s main tourist areas. The many forms of taro offer us a multitude of tastes, textures, consistencies and growing requirements to work with. There are many talented people associated with the crop, from the paddock to the plate. The taro industry is bound to ‘fly’. How far depends on us, but it’s time to spread our wings! third taro symposium 145 Theme Three Paper 3.6 Comparison of taro production and constraints between West Africa and the Pacific Kwadwo Ofori School of Agriculture, University of the South Pacific, Alafua Campus, Apia, Samoa Introduction Taro (Colocasia esculenta var. esculenta) is the most important member of the aroid family in terms of production and utilization. It is also the species of highest commercial value among the aroids. Most components of the crop are consumed as food and/or feed. The highest intensity of production, utilization and dependence on taro is in the Pacific Islands. The highest areas of production, however, are in West Africa. Considering these two broad areas of cultivation, at opposite sides of the equator, with contrasting features in several factors of production, the tendency is to compare and contrast production and productivity. Beyond such comparisons, however, is the opportunity offered to learn from each others prospects and progress towards a unified strategy to improve food security, product diversification and livelihood of populations utilizing taro for food, cash and foreign exchange. Taro has a wide environmental tolerance. It is cultivated in every island group in the Pacific in almost all ecological zones. In West Africa, the crop is produced in the coastal countries with tropical wet and wet-dry conditions. Producing countries in West Africa include Nigeria, Ghana, Cote d’Ivoire, Togo, Benin, Chad, Sierra Leone, Liberia and Guinea. Most of these countries derive 30-60% of their dietary energy requirements from cereals such as rice, maize, millet and sorghum. Taro yield in West Africa and the Pacific are not at their optimum and fall well below yields reported for Asia. West Africa and the Pacific have their peculiar production environments and practices. The two areas also have some similarities, especially in constraints to production and prospects for expanded production. This paper is an attempt to review the status of taro production and to compare and contrast production practices, prospects and constraints to production in West Africa and the Pacific Islands. There is also the attempt to formulate general recommendations for overall increased production and productivity of taro in these two regions. Importance of taro in West Africa and the Pacific Taro in food system Taro plays a major role in the life of the Pacific Islands. It ranks 14th worldwide in production level, but in most Pacific Islands, it is a major component of socio-cultural, dietary and economic livelihood (Onwueme and Charles, 1994; Onwueme, 1999). Mere production figures do not convey the full picture of importance of the crop in producing countries. However, combined with figures on land availability, population and utilization, a clear picture emerges that shows the Pacific Islands having the highest intensity of production, utilization and dependence on taro for food. Commonly, production and utilization figures have been combined for taro and tannia (Xanthosoma sagittifolium) (Table 1). In the Pacific very little of tannia is utilized for food. In West Africa, however, the situation is reversed with more tannia utilized than taro, except in Nigeria. Nevertheless, taro is always listed among the staple food crops of coastal West African countries from Nigeria to Guinea. Though the bulk of taro is produced in Africa, Table 2 indicates that the Pacific countries have a higher proportion of dietary energy from taro/tannia than West Africa. Most of the crop is produced in Nigeria, Ghana and Cote d’Ivoire (Table 1). Outside of West Africa, other African producers are Gabon, Egypt, Rwanda, Burundi, Zaire, Central African Republic, Comoros Island, Sao Tome and Principe, Madagascar and Mauritius. Taro contributes significantly to food security in producing countries in both West Africa and the Pacific. It serves as an important food during the dry season or before yam and cassava harvest in West Africa. 146 third taro symposium Table 1: Production, yield and area for taro/tannia in 1990 and 2000 for leading producers in the Pacific and West Africa Area (1000 ha) Yield (kg ha-1) Production (1000 tonnes) 1990a 2000b 1990 2000 1900 2000 983 1458 5314 6058 5225 8835 150 133 11536 13936 1727 1854 738 1274 3996 5229 3130 6662 217 265 1300 1376 282 365 200 232 4500 7359 900 1707 10 4 6200 6624 62 29 2 2 6818 8333 15 20 250 571 5200 6716 1300 3835 47 48 7142 6103 337 292 2 3 9915 8384 215 160 32 31 6719 5161 41 37 6 6 6915 6150 22 32 1 4 19909 20000 30 27 4 3 6864 6476 15 26 World Asia Africa Cote d’Ivoire Ghana Guinea Liberia Nigeria Oceania Fiji Papua New Guinea Samoa Solomon Islands Tonga Sources: FAO Yearbook, Production 1990 (1991); FAO Bulletin of Statistics 2000 (2001) a b Socio-cultural value Beside its food value, taro is important in the social and cultural life of the people of the Pacific Islands. The crop features prominently in folklore, during traditional feasting and as a valuable gift. In West Africa, such prestige is attached to yam. Various parts of the taro plant are used in traditional medicine and hence has a certain amount of reverence attached to it (Onwueme, 1999). This cultural attachment is largely responsible for the existence of export market of taro in Australia, New Zealand and United States of America, where many Pacific Islanders live. Taro from West Africa is exported mostly to Europe, where again the consumers are migrants from Africa, who see themselves maintaining their culture through traditional food. Table 2: Percentage of dietary calories derived from taro/tannia in 1984 for some leading taro-producing countries in Africa and the Pacific Country Taro/Tannia Oceania 0.7 Tonga 18.1 Samoa 16.0 Solomon Island 7.7 Papua New Guinea 4.2 Africa 0.5 Gabon 4.6 Ghana 7.1 Source: Adapted from Horton, 1988. Development economy of taro Taro production generates income for several subsistence farmers in both West Africa and the Pacific. Market avenues for surpluses from subsistence production are more widespread in the Pacific, compared to West Africa, where production and consumption are rather peculiar to specific ecological zones. Taro contributes significantly to poverty alleviation for several vulnerable groups in producing regions. For some Pacific countries, taro exports form a substantial part of foreign exchange earnings. In West Africa, however, other traditional export products such as cocoa, coffee and timber exit and hence taro is classified under the non-traditional export commodities. Production environment and practices Production practices in West Africa Land resources are a major factor in the production of taro. Year-round taro production is possible in the tropics, provided that water is continuously available (Dudal, 1980). In most of West Africa, taro production is under rain-fed conditions. Large areas in West Africa are not suitable for upland taro production due to high risk of crop failure. Production is thus limited to rainy season in areas along streams and marshy areas, especially in valley bottoms, where water may be available for most part of the year. Taro has been grown in West Africa for several generations in southern and eastern Nigeria, southern Ghana and Benin along forest streams and in swamps (Irvine, 1974). In heavy rainfall areas of western Ghana and eastern Cote d’Ivoire, marshy areas are common, usually adjacent to coconut plantations. In south-western Ghana, for instance, the lethal yellow disease has destroyed most of the coconut trees and a booming taro industry has sprang up over the past six years. This area is wet throughout the year and hence yields are high. The farmers use very little of the crop but the bulk is sold for cash. third taro symposium 147 Suckers are the main planting materials used in West Africa. Most of the cultivars produce numerous suckers. Tops of corms are also used as planting material to a limited extent. Manual weed control with cutlass or machete is the common practice. Some farmers when overwhelmed by weeds cut down weeds and taro plants. Subsequently, the taro plants outgrow the weed re-growth and smother weeds. In Nigeria, 25% of labour needs are used for weeding (Knipscheer and Wilson 1980). There is very little herbicide usage, though it has been found that herbicides cost only 45-55% of cost of hand weeding (Abasi and Onwueme, 1984). Use of herbicide is slowly increasing in the south-western part of Ghana. After harvest of plant crop, weed control in the ratoon is rather difficult manually, due to the high plant population from the suckers. Optimum plant populations have been reported for Nigeria (16,667 plants ha-1) (Igbokwe and Ogbonnaya, 1980) and Ghana (20,000 plants ha-1) (Safo-Kantanka, 1986). Low organic matter and deficiency in major soil nutrients, without fertilizer application or manuring limit growth and yield of taro in West Africa (Igbokwe and Ogbonnaya 1980). This is aggravated by continuous production in the same location. Poor aeration of taro plants in stagnant water, results in high losses from rot diseases (Manrique, 1995). Various components of the crop are used with preference depending on the location and food habits. In Ghana and Nigeria, corms are preferred, while in Liberia, corms and leaves are used for food. Post-harvest losses are high, especially cases of corm rot. Markets are not well organized, with farmers receiving poor prices from middle men and women. Processed products have not been developed, but the potential for products such as taro chips look bright. Production practices in the Pacific Taro production in the Pacific has been well documented (Chandra, 1984; Onwueme and Charles, 1994; Onwueme 1999). Between 1990 and 2000, there has been a reduction in total production through a reduction in yield per unit area, without any significant change in area under taro cultivation (Table 1). Since productivity has not been improved or maintained, questions have to be raised on the sustainability of production practices. The bulk of taro production in the Pacific is from upland system. The rainfall here is higher than that of West Africa, often in excess of 2500 mm per annum. Flooded taro production is also used in some locations. Animal traction and tractor has been used on flat lands in Fiji. Suckers are the preferred planting materials but corm tops are also used. Farming is mostly at the subsistence level with no fertilizer use. Commercial farms however use a wide variety of compound fertilizers. In some countries, lack of fertilizer use is due to speculation that fertilizer reduces corm quality. Weed control is done manually, with herbicide and also mechanically. Pests such as taro beetle and leaf hopper and diseases such as leaf blight, alomae virus and bobone diseases reduce production levels considerably. These have led to strict quarantine regulations on movement pf planting materials across countries. These problems have also restricted markets for taro from some producing countries (Liloqula and Samu, 1996). Most of the crop is consumed in the Pacific. Taro leaves are used in most of the countries. Moist sacks and containers with dripping water are used while internally transporting taro from one part of the country to another. Even in Fiji, Tonga and Papua New Guinea, where taro comes after other crops in consumption, a large proportion is consumed locally. Poi, a sour-tasting taro paste packaged and sold commercially in stores, is produced in Hawaii. Taro chips are also produced occasionally in some countries. Constraints to production West Africa Generally, taro is extensively cropped in West Africa with very low yields. Yields are lower than those of the Pacific and Asian countries (FAO, 1987). Production is restricted to areas around streams in high rainfall areas and in stagnant water bodies. The cultivars are low yielding and also produce many suckers leading to poor yields of ratoon crop. Leaf and corm rot diseases can cause 40-90% yield loss (Doku, 1984). Corm rot in storage can result in 75% corm loss during severe infection (Nwufu and Fajola, 1981; Nwufu, 1988). Selection for non-flowering types has led to a narrow genetic base, which requires broadening. Table 3: Distribution of research papers on specific crops at symposium on tropical roots and tubers held in Ghana in 1994 Number of papers Percentage Cassava 43 46.2 Yam 16 17.2 Sweet potato 13 14.0 Cocoyam 10 10.8 Potato 9 9.7 Taro 2 2.1 Total 93 100 Source: ISHS Acta Horticulturae 380 (1994) There is limited allocation of resources by farmers and policy makers to taro industry. This in part has limited research effort on taro compared to other tuber crops in the region (Table 3). There is increasing dependence on cereals such as rice and maize to meet urban demands (Dapaah, 1994). Urbanization has also caused a shift in food habits to the exclusion of taro in some instances. This is aggravated by poor storability and lack of processing and stable taro products. There is unfavourable competition against taro among root and tuber crops for food uses. Drought and 148 third taro symposium erratic rainfall distribution are major hindrances to upland taro production, since irrigation facilities are non-existent. In Egypt, where taro is grown commercially under flood irrigation, yield of 31,00 kg ha-1 are common (Manrique, 1995; FAO, 2001). There is inefficient marketing strategies and limited and uncoordinated research. Internal displacement of farmers due to war and civil strife have also influenced staple food production and increased the reliance of food aid, which are mostly cereals. This is true for Cote d’Ivoire, Guinea, Liberia and Sierra Leone. The Pacific Lack of facilities for irrigation limits upland taro production. Drought has been a major factor that influences taro production in Tonga, leading to depressed yields and also causing a scarcity of planting material in subsequent seasons. Fallow periods are rather short or non-existent with no fertilizer or manure. The use of leaves for food, to some extent reduces yield of corm. High labour requirement compared to other tubers such as sweet potato and availability of imported food substitutes have negatively affected taro production in Papua New Guinea. Transportation of the highly perishable corm is a problem in some Pacific counties such as Cook Islands and Vanuatu. Post-harvest losses are major constraints in taro production in the Pacific, with most of the corms eaten without processing. By far the one most important factor has been disease, particularly leaf blight and alomae/bobone virus disease complex. These diseases have led to strict quarantine restriction that has cut off taro export for countries such as Solomon Islands and Papua New Guinea. The taro leaf blight brought production down to almost zero in Samoa in late 1993. This was reflected in a 96% decrease in production level, 350% increase in price of taro and 86% drop in foreign exchange from taro sales in the 1994 season (Table 4) (Tekiu, 1996). The taro leaf beetle is also a problem in several countries, especially in Papua New Guinea and Solomon islands. Table 4: Taro production, price and value in Western Samoa from 1985-1994 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 Production (1000 tonnes) 16.50 25.51 26.88 22.22 28.24 25.42 29.02 27.60 30.08 1.20 Price (WS$ /tonne) 538 851 941 986 1008 1096 1171 1350 916 3226 Value (Million WS$) 8.9 21.7 25.3 21.9 28.5 27.9 34.0 37.3 27.5 3.9 Source: Central Bank of Samoa (Tekiu, 1996) Prospects and recommendations for improved production The number and size of suckers produced is influenced to a large extent by the production system and cultural practices, given that suckering ability is highly heritable. High sucker production contributes to corm yield under flooded conditions but reduces corm yields under upland conditions (Sivan, 1977, 1980). Similarly, high levels of stolon production significantly reduces yield, though stolons could be used for food. In West Africa, where most taro is produced in marshy areas, several small suckers are produced. There are therefore large amounts of planting materials generated, but at the expense of corm yield of the plant crop. The low yield due to the use of such cultivars is obvious, especially in the overcrowded ratoon crop. Recent increase in production in West Africa is partly due to controlled planting instead of concentrating on volunteer plants. Selection of planting material and improved weed control, have also led to high yields. There is the need for selection for cultivars with low suckering ability for lowland taro production in both West Africa and the Pacific. There is also the need to research and also train farmers on cheap and simple ways of generating planting materials. A mini sett technique similar to that used in yam production should be tried. This may help produce uniform corms for the export market. Considering the small farm sizes in the Pacific Islands, integrated approaches should be encouraged. Animal manure would be used to complement fertilizer needs in an intensive production system (Wang and Nagarajan, 1984). The mature tops could be fed to pigs as silage (Steinke et al, 1984). During peak season, when prices are low, excess taro should be channeled into animal feed as part of a sustainable farming system. Chandra (1984) suggested that appropriate intensive small-scale taro production be developed for urban areas, such as home gardens to support urban food supply, in the light of high urbanization in the Pacific. There is the need improve production on a sustainable basis and this calls for continuous research activities with the farmer as a close partner. In areas where leaves are harvested for the market, genotypes with high rate of leaf production should be selected rather than developing multi-purpose types. This may call for re-evaluation of genetic resources already available and possibly new collections. Farmers may still have such varieties, which have been ignored because of low corm yields. Though there is evidence that light defoliation does not reduce corm yield (Safo-Kantanka, 1986), farmers may not know exactly when to stop defoliation. In Ghana, cocoyam rather than taro leaves are eaten. In the Pacific, however, where very little cocoyam is eaten, cocoyam leaves abound. Adoption of use of cocoyam leaves will reduce the pressure on taro leaves for food. Selection of suitable varieties should consider post-harvest and culinary characteristics in addition to yield and pest and disease resistance. Storability, levels of irritants, cooking time, flavour, corm texture, corm flesh colour, corm size and shape are important for successful marketing. In West Africa, especially in Ghana, cocoyam (new cocoyam) is produced third taro symposium 149 on a higher scale than taro because the taro is too soft for making “fufu”, a traditional dish. Furthermore, the levels of irritants are higher for taro than cocoyam varieties available. Through selection by farmers over the years, literature that indicates that taro needs to be cooked for at least 12 hours (Irvine, 1974; Manrique, 1995) have become obsolete. Instead the new varieties of taro available in West Africa, cook within 30 minutes. Most of the countries in West Africa depend to a large extent on cereals, with large supplies of maize and rice coming from imports. The cereals are easier to prepare and more convenient to store as a food security strategy. In Nigeria, “gari” a stable dried cassava product competes favourably with dry cereal as an instant food item. Development of the taro industry is one way of reducing cereal imports. Commercial outlets should be established for taro surpluses to earn much needed foreign exchange. Alternative production of rice and taro would lead to more sustainable land use than continuous rice production in some countries, such as Liberia. Originally, farming in West Africa was subsistence oriented (Okigbo, 1982; Ruthenberg, 1980). With growth in domestic industrialization, many raw materials can be sold locally. Urbanization, mobility and need for improved facilities have increased and so has need for money by farmers. In this new environment, the objectives of traditional farmers are various and include growing some commodities entirely for cash, while food crops still remain a priority. In line with this development, there should be technical food processing innovations for small scale applications to develop local commodities aimed at local and external markets. This will contribute to food security needs and also improve competitiveness and sustainability of taro production. Taro chips have been successfully produced in Hawaii and other Pacific countries (Pena, 1990). Processing taro into diverse dehydrated and stable products has also been examined in Hawaii (Moy et al., 1979). Griffin (1979) has also examined the use of taro in food starch, flour and non-food starch production. The latter was for biodegradable plastic production. Industrial uses should be pursued, sustained and expanded. The development of novel food and non-food products of taro would stimulate interest in increased production of taro, especially in West Africa, where land may be available for such competitive economic use. The wide range of genetic diversity available must be exploited by direct evaluation of cultivars for resistance to disease, yield and nutritional and cooking quality and also by genetic manipulations. The target should be to raise the world average yield to the current highest of about 13 Mg ha-1 produced in Asia. Wilson (1984) stressed on the need for taro breeders to identify and incorporate characters that would be required for adaptation in each production system. Stability and predictability of yield would then be more important in this regard than productivity per se in food security. The need to develop a modeling programme for taro is the challenge before researchers. There is very little exchange of information and germplasm between West Africa and the Pacific, especially with the increased threat of pests and diseases. Hence, progress in one region does not spread far, partly due to geographical distance. The establishment of an international taro newsletter or bulletin would promote centralization of information and dissemination to scientists, policy makers, non-governmental organizations and other stakeholders. This provision may lead to enhanced formal exchange of in vitro germplasm, exchange of scientists’ visits and farmer visits. Informal presentation of information rather than journals would readily notify all stakeholders of progress achieved by farmers and scientists. Upland taro production must be exploited in West Africa. There is the need for detailed assessment of cost of production of taro under the different production systems to determine its appropriate position in raising the economic state of farmers in West Africa. Conclusions The taro industry in both West Africa and the Pacific provide very useful sources of livelihood, especially for the latter region. However, production levels are below optimum due to several biophysical and socio-economical factors. These include low yielding cultivars, pests and diseases, poor husbandry practices, scarcity of arable land, shortage of farm labour, shift in food habit, limited research attention, limited resource allocation, inefficient marketing, lack of processing and innovative products and weak governmental support. Considering the socio-economic importance of taro in the Pacific countries and the availability of land in West Africa, a concerted effort that will be beneficial to both regions should be sought. Such regional co-operation would go a long way to improve the role of taro in providing a staple, food security and income for low socio-economic groups in these regions. References Abasi, L. and Onwueme, I.C. 1984. Herbicidal control in tannia cocoyam (Xanthosoma sagittifolium). p. 91–96. In: Shideler, F.S. and Rincon, H. (eds). Proceedings of the Sixth Symposium of the International Society for Tropical Root Crops, Lima, Peru, 21–26 February 1983. ISTRC. Chandra, S. 1984. Edible aroids. Clarendon Press, Oxford. 252 p. Dapaah, S.K. 1994. Contributions of root and tuber crops to socioeconomic changes in the development world: The case of Africa, with special emphasis on Ghana. Acta Horticulturae 380:44–49. de la Pena, R. 1990. The taro chip industry in Hawaii: Its problems and potential for expansion. p. 674–679. In: Howeler, R.H. (ed.) Proceedings of the 8th Symposium of the International Society for Tropical Root Crops, Bangkok, Thailand, 30 October–5 November 1988. 150 third taro symposium Doku, E.V. 1984. Production potentials of major tropical root and tuber crops. p. 19–24. In: Terry, E.R. (ed.) Tropical root crops: Production and uses in Africa: Proceedings of the Second Triennial Symposium of the International Society for Tropical Root Crops – Africa Branch, Douala, Cameroon, 14–19 August, 1983. IDRC, Ottawa, Canada. Dudal, R. 1980. Soil-related constraints to agricultural development in the tropics. p. 23–40. In: Priorities for alleviating soil-related constraints to food production in the tropics: proceedings (conference held in Los Banos, Laguna, Philippines, 4–8 June 1979). IRRI, Los Banos, Laguna. FAO. 1987. Yearbook of production for 1986. FAO, Rome. FAO. 1991. Yearbook of production for 1990. FAO, Rome. FAO. 2001. FAO Bulletin of statistics for 2000. FAO, Rome. Griffin, G.J.L. 1979. Non-food applications of starch, especially potential uses of taro. p. 275–301. In: Plucknett, D.L. (ed.) Small-scale processing and storage of tropical root crops. Westview Press, Boulder, Colorado. Horton, D. 1988. Underground crops: Long-term trends in production of roots and tubers. Winrock International, USA. Igbokwe, M.C. and Ogbonnaya, J.C. 1981. Yield and nitrogen uptake by cocoyam as affected by nitrogen application and spacing. p. 255–257. In: Terry, E.R., Oduro, K.A. and Caveness, F. (eds). Tropical root crops: Research strategies for the 1980s: Proceedings of the First Triennial Root Crops Symposium of the International Society for Tropical Root Crops – Africa Branch, Ibadan, Nigeria, 8–12 September 1980. IDRC, Ottawa. Irvine, F.R. 1974. West African crops. Oxford University Press, Oxford. 272 p. Knipscheer, H.C. and Wilson J.E. 1981. Cocoyam farming systems in Nigeria. p. 247–254. In Terry, E.R., Oduro, K.A. and Caveness, F. (eds). Tropical root crops: Research strategies for the 1980s: Proceedings of the First Triennial Root Crops Symposium of the International Society for Tropical Root Crops – Africa Branch, Ibadan, Nigeria, 8–12 September 1980. IDRC, Ottawa. Liloqula, R. and Samu, J. 1996. Solomon Islands country status report on root and tuber crops. Paper presented at the Expert Consultation on Enhanced Root and Tuber Crop Production Development in the South Pacific, Apia, Western Samoa, 17–19 July 1996. Manrique, L.A. 1995. Taro: Production principles and practices. Manrique International Agrotech, Honolulu. Moy, J.H., Wang, N.T.S. and Nakayama, T.O. 1979. Processing of taro into dehydrated, stable intermediate products. p. 223–248. In: Plucknett, D.L. (ed.) Small-scale processing and storage of tropical root crops. Westview Press, Boulder, Colorado. Nwufo, I. 1988. Storage of corms of Colocasia esculenta under modified environmental conditions. Journal of Root Crops 14:1–4. Nwufo, I. and Fajola, A.O. 1981. Storage rot diseases of cocoyam (Colocasia esculenta) in south-eastern Nigeria. Journal of Root Crops 7:53–59. Okigbo, B.N. 1982. Agriculture and food production in tropical Africa. Paper presented at the USAID/ADC Seminar on Improving the Developmental Effectiveness of Food Aid in Africa, 23–26 August 1981, Abidjan, Côte d’Ivoire. Onwueme, I.C. 1999. Taro cultivation in Asia and the Pacific. FAO, Regional Office of Asia and the Pacific, Bangkok, Thailand. 50 p. Onwueme, I.C. and Charles, W.B. 1994. Tropical root and tuber crops: Production, perspectives and future prospects. FAO, Rome. 228 p. Ruthenberg, H. 1980. Farming systems in the tropics. Clarendon Press, Oxford. 424 p. Safo-Kantanka, O. 1986. Effect of leaf harvesting and spacing on yield of Xanthosoma sagittifolium and Colocasia esculenta. p. 94–95. In Terry, E.R. Akoroda, M.O. and Arere, O.B. (eds). Tropical root crops: Root crops and the African food crisis: Proceedings of the Third Triennial Symposium of the International Society for Tropical Root Crops – Africa Branch, Owerri, Nigeria, 17–23 August 1986. IDRC, Ottawa. Sivan, P. 1977. Effects of spacing in taro (Colocasia esculenta). p. 377–381. In: Leaky, C.L.A. (ed.) Proceedings of the Third Symposium of the International Society for Tropical Root Crops, Ibadan, Nigeria, 2–9 December 1973. IITA, Ibadan, Nigeria. Sivan, P. 1980. Growth and development of taro under dry land conditions in Fiji. IFS Report 5:167–182. Steinke, W.E., Wang, J.K., Carpenter, J.R. and de la Pena, R.S. 1984. The use of taro silage as animal feed in the Pacific. In: Chandra, S. (ed.) Edible aroids. Clarendon Press, Oxford. Tekiu, N. 1996. Tropical root and tuber crops in Western Samoa. Paper presented at the Expert Consultation on Enhanced Root and Tuber Crop Production Development in the South Pacific, Apia, Western Samoa, 17–19 July 1996. Wang, J.K. and Nagarajan, K. 1984. An integrated taro production system for the humid tropics. p. 140–148. In: Chandra, S. (ed.) Edible aroids. Clarendon Press, Oxford. Wilson, J.E. 1984. Taro and cocoyam: What is the ideal plant type? p. 151–159. In: Chandra, S. (ed.) Edible aroids. Clarendon Press, Oxford. third taro symposium 151 Theme Three Paper 3.7 Taro production, constraints and future research and development programme in Indonesia T.K. Prana, Made Sri Prana, and T. Kuswara Research Centre for Biotechnology, Bogor, Indonesia Introduction Taro, undoubtedly, once was an important staple food in Indonesia, but its role has been decreasing with the successful introduction of “new” crops like rice, maize, cassava and sweet potato. In few places, such as Irian Jaya or Papua, Mentawai Islands (West Sumatra), Sangihe Talaud (North Sulawesi) and at least in one place in East Jawa (Cemoro Sewu); however, the local people still consume taro asides from the above mentioned staple foods. Its role however has been gradually replaced by rice. This process is still going on at even an increasing speed, for some reasons. Firstly the local people concerned (Papua, Mentawai etc.) regards rice as not only better but more prestigious staple food to consume (wealthy people eat rice). Secondly, so far, the government has not been giving so much attention on taro research and development despite the fact that food diversification has always been part of the national programme. Thirdly in cases of natural disasters or food scarcity in the non rice eating areas, for a very simple classical reason that rice stock is always available and relatively (compare to root crops) more handy to transfer, the rescue team (the government) would drop rice instead of the staple foods which the local people normally consumed. What left now, in most part of the country, actually are remnants of populations of both wild and cultivated types. Even these still display a wide range of genetic variability which should be explored, rescued, conserved, and utilized to revive taro cultivation in the country should food diversification be still regarded as really important for the survival of the nation as well as the country. Though this does not necessarily imply that taro uses should be only limited to staple food. That was actually all the reasons behind the eager participation of Indonesia, in this case the Indonesian Institute of Sciences (LIPI), in the TANSAO (Taro Network for South East Asia and Oceania) project launched in 1998. During the past 5 years LIPI, in collaboration with a few other institutions like Bogor Agriculture University, agriculture services, the Research Division for Root Crops and Beans in Malang- East Jawa, and the Winaya Mukti University (West Jawa), has made a lot of achievements in implementing taro research and development programme in the country. All those results obtained could be utilized to further develop the crop through various programme. It is therefore considered quite timely now to carefully plan the future of the crop, to revive its role as an important crop commodity through systematic research and development programme. In doing so, various aspects should be taken into consideration. In that respect, four major aspects will be discussed in this paper, namely the present state of knowledge, constraints, final objective, and future R&D programme. The scope of discussion will be largely focused to Jawa and Bali islands where taro has lost its role as staple food but open a great opportunity to develop it in food industry. Present condition Despite of its long history of cultivation, taro has not been regarded as an important crop commodity in Indonesia, it ranks number 6 after rice, maize, cassava, sweet potato and even potato. It is therefore not surprising if there has been hardly any budget allocated by the government for research and development of taro, except for once i.e. during the severe economic crisis (1998-1999). During the time all food crops were regarded as important including the minor ones like arrow root (Marantha arundinacea), yam (Dioscorea spp), kana (Canna edulis), cocoyam (Xanthosoma spp). It is sad to notice that the exotic root crop, potato, has had even a better position than taro, despite the fact that it is generally regarded as a vegetable rather than staple food . Productivity of the crop in most places is low, in the Mentawai island it was reported as low as 2.5 ton/ha (Jusuf et.al., 1996) and in Irian Jaya (Papua) 3.1-3.4 t/ha as cited by Prana and Kuswara (2001). In Jawa and Bali, based on our reason observation, a higher productivity (7-11 t/ha) could be expected. In most places, except in the taro producing centers, taro is grown on ridges of the rice fields, along the irrigation canal, or in a mixed cropping system together with cassava, maize, sweet potato etc. or in between perennial crops like banana, mangoes, rambutan etc. in the home- or orchard-gardens. During the severe food crisis (1998/1999) taro was planted almost everywhere, such as in office yards, bare lands, or even along the sides of highways. Inputs (tillage, manure/fertilizer, pesticides application) have usually been minimal, since taro indeed is quite easy to grow although the yield may not be high. The number of plants planted in a particular area (land ownership) usually ranges from a few (less than 10) to ten to twenty plants. The varieties grown could be a mixture of local cultivars, indicating that most of them (the farmers) do not bother so much about the planting materials. Consequently, productivity is also low. 152 third taro symposium However, it is this type of agricultural practices that has saved taro from a much more serious genetic erosion. In the taro growing areas in Jawa, such as in Bogor and Sumedang (West Jawa), and Malang (East Jawa), the situation is completely different. There, taro is mostly grown in a mono culture system with better tillage and higher input (manure/fertilizer, and pesticide application), as well as intensive maintenance. The cultivars grown are highly selected, based on demand of the markets and superiority in certain characteristics, like good eating quality, high yield, resistance to diseases and pests etc. The popular variety in Malang for example is “Boring”, in Bogor it is “Bentul”, and in Sumedang “Semir”. In term of both quality and productivity of the cultivars are comparatively much higher. In the highly intensive cultivation practices, taro is grown in holes (60 x 60 x 60 cm) with heavy manure application, and kept in the ground for 8- 9 months before harvesting. The yield may reach 3- 4 kg per individual corm. Corms of this class usually are sold in supermarkets for the well-of customers. Presently taro production seems to have reached “its peak”. The (local) markets have been more or less saturated. Interestingly to note that in Bogor area (West Java) the price of taro corms, at farmer’s level, remain the same through out the year, irrespective of the seasons, either during the peak production season (March-June) or during low production season (November-February) as reported by Kuswara and Prana (2003). This suggested that the people in the area might not so much dependent on the crop, they only buy it at certain price level. The fact that taro based industry is hardly in existence in the area might also contribute to the special phenomena. Taros in the (local) markets are mostly sold as fresh corms, and to a limited extent tender shoots of taro are also sold as vegetable. No exports of fresh corms have been recorded so far. In such a condition obviously the market is very much limited i.e. only locally (district level), not even to a provincial level. Production constraints One of the main production constraints is problem of pests and disease. There are some pests and disease that frequently attack taro. Among the common pests are mites ((Tetranychus cinnabarinus), worm (Spodopter litura) and grasshoppers (Tarophagus proserpina). Meanwhile the most notorious disease recorded was Taro Leaf Blight (TLB) which is found almost anywhere in the country. The damage caused by the TLB is frequently quite extensive, resulting in harvest failure. Recently about 2 years ago an out break of corm rot was reported by Burdani (pers. comm., 2001) in Bogor, it is locally called “muntaber” which literary means “cholera” indicated by the rotten corm containing putrid smelly liquid. The cause of the disease remains uncertain, Burdani suggested it was caused by Fusarium spp, most probably F. oxysporium, whilst disease of similar symptoms was suggested to have been caused by Phytophthora spp or perhaps also Phytium spp (Bridge). The local popular cultivar “Bentul” in Bogor was observed to be quite susceptible to the disease and hence in some places it was gradually replaced by the more resistant cultivar “Bogor”. That was interesting because only about two decades ago “Bogor “ was gradually replaced by “Bentul” for the simple reason that the cultivar “Bogor” produces so many shoots that require tedious work to maintain (the shoots are usually removed to get better yield). Other cause of production constraint is maturity (time to harvest). The time it takes for the corms to be ready for harvest has been considered as a bit too long by most farmers in the Jawa island. Most of the superior cultivars known in the island are late maturing, which require at least 7 months before they can be harvested. This make them not so suitable for areas with 6 –7 dry months (monthly rainfall less than 100 mm), such as to be found in the southern, central and eastern parts of the island or even in the southeastern parts of Indonesia in general. This is due to the fact that taro, at least the Indonesian cultivars, to our knowledge is not tolerant to draught (Prana and Kuswara, 2001). Certainly, this dependency on rainfall, theoretically, could be overcome by proper irrigation, but in such a case taro has to compete with other more profitable crops like rise and maize. The characteristics it has, have made taro mainly occupy areas with longer (7 months or more) wet months (monthly rainfall 100 mm or over) or in well irrigated places where it has good price in the local markets so that it can successfully compete with other high valued crops. Research achievements Research and development activities on taro in Indonesia have been largely dependent on external funding, such as from the IFS, EU, CIRAD etc. As a consequence there is no guarantee of continuity of the programme. Taro project were up and down depending on availability of fund from external sponsors. Consequently, almost every time the new project has to start from the base line i.e. begins with exploration and end up usually with evaluation, a point where scientists begin to understand the potency of the crop. No chance to develop it further since the project has come to an end, meanwhile financial support from other sources are usually not available. That story goes on over and over again. The sad thing is that the germplasms collection is usually, eventually, being lost too because there is no budget for maintenance, not even from the government. The last project on taro was that done in collaboration with two countries in Europe (France, the Netherlands), 5 countries in South East Asia (Malaysia, Philippines, Thailand, and Vietnam), and two countries in the Pacific (PNG and Vanuatu), known as TANSAO project, entitled “Taro: Exploration and Evaluation for Rainfed Farming Systems”, funded by the European Union The project have been successfully implemented, but then again it had to be terminated at the moment when scientists began to understand their genetic variability and potency. The proposal submitted for a possible extension of the project for the second phase (development) failed to get funding support. It would not be third taro symposium 153 surprising, therefore, if at one time in the future the existing collection will be lost too unless maintenance cost could be made available. Through the project, Indonesia was able to collect 181 zymotypes, and identified quite a number of interesting genotypes having potential agronomic characteristics that can be utilized and develop further through cultivar improvement programme. Some tissue cultured materials of superior cultivars were also obtained from the member countries (in S.E. Asia and the Pacific) and are maintained by the Research Centre for Biotechnology in Bogor. All samples have been characterized. CIRAD has studied the zymotipic profiles of the bacteria causing TLB as well analyzing the physico-chemical characteristics of the starch (Lebot et al., 2000). Some potential hybrids have been identified and propagated. A multi-location trial is being carried out in 18 places through out Jawa with the objective to select cultivars that eventually could be released to the local farmers (in Jawa). At present LIPI in collaboration with the Bogor University of Agriculture has encouraged small scale industry to produce snacks that eventually hopefully could be exported. Future programme Realizing the limited available resources, the future programme would be focused on: 1. Product design/development towards product diversification, improved packaging and expand marketing. This will be done through collaboration with other research institutions, NGOs, and other relevant organizations (Women Organizations) and services (agriculture, industry, trade etc.). Through concerted efforts progress could be achieved much more efficiently. Nowadays many organizations, including NGOs, who are very much interested in activities that would have direct impact on poverty alleviation at the grass root level. With the decentralization programme recently introduced by the government it should be possible to encourage the local government to pay special attention on such a programme concerning with food diversification and improvement of social welfare. 2. Breeding to develop new/superior cultivars in term of productivity and product quality requiring lower inputs etc. To certain extent this may be achieved by simple selection of the existing local cultivars and many more local cultivars yet to be collected (North Sulawesi, Kalimantan and Sumatera). Collaboration in this case should be done with other research institutions in the field of agriculture, not exceptionally universities. The research centers under the Ministry of Agriculture have expertise as well as facilities in all the provinces that can be utilized for the implementation of field trials. Meanwhile the universities have lots of highly dedicated students that could be utilized to do the observations. 3. Expand marketing to both domestic and regional/international markets. This is possible since some industries are beginning to get interested in developing taro based industry. However, those are certainly too big tasks to be done solely by LIPI or by any other institution. It needs a concerted effort involving not only Research and Development Institutions but also other organizations, including NGOs concerned with the crop. References Jusuf, M.A. Basyir, Marzempi and Jonharnas. 1996. Status of taro genetic resources in West Sumatra and research accomplishment. Paper presented at the 13th Taro Symposium, Manokwari, Irian Jaya, Indonesia. 17 p. Kuswara, T. and Prana, M.S. 2003. Trade pattern of taro (C. esculenta) in Bogor, West Java, Indonesia. Unpublished. 7 p. Lebot, V., Hartati, N.S., Hue, N.T., Viet, N.V., Nghia, N.H., Okpul, T., Pardales, J., Prana, M.S., Prana, T.K., Tongjiem, N., Krieke, C.M., Van Eck, H., Yap, T.C. and Ivancic, A. 2000. Genetic variation of taro (Colocasia esculenta) in South East Asia and Oceania. In: Nakatani, M. and Komaki, K. (eds). Proceedings of the Symposium of the International Society of Tropical Root Crops: Potential of root crops for food and industrial resources, Tsukuba, Japan, 10–16 September 2000. ISTRC. Prana, M.S. and Kuswara, T. 2001. The cultivation of taro: Diversification to support national food security programme (in Indonesia). TANSAO/LIPI/EC, Bogor, Indonesia. 79 p. 154 third taro symposium Theme Three Paper 3.8 Taro production, constraints and research in Cuba Arlene Rodríguez-Manzano, Adolfo A. Rodríguez-Nodals, Leonor Castiñeiras-Alfonso, Zoila Fundora-Mayor and Adolfo Rodríguez-Manzano Institute of Fundamental Research on Tropical Agriculture (INIFAT), Calle 1. esq. 2, Santiago de las Vegas, C. Habana, Cuba Introduction Within the Araceae family there are two genera used as food in Cuba: Colocasia, commonly called “malanga isleña,” and Xanthosoma, the so-called “malanga” or “guagüí”. Pichardo (in Roig, 1965) said that the word “malanga” was adopted into the Cuban language from an African source. The integration of African and Spanish cultures in the common name of cultivated Colocasia esculenta (L.) Schott is clear, with the voice “malanga” contributed by the black slaves and the word “isleña (islander)” referring to Spaniards from Canary Islands that settled in Cuba. Though basically used in the same way, preference for these two crops varies among different regions of the country and different ethnicities (Castiñeiras et al., 2000). Taro originated in the Indo-Malay region, and it was dispersed to east and southeast Asia, the Pacific Islands and west to Madagascar and África, from where it was introduced into the Caribbean and America, according to Ivancic and Lebot (2000). Introduction routes to Cuba during colonial times in the 18th century probably included from western Africa and from the Canary Islands, through slaves and settlers, respectively. Another possible introduction route could be from the Philippines Islands, through the galleon trade route Manila–Acapulco-Havana. Lastly, it would have been introduced directly from China, through the Chinese immigrants in the 19th century and from Japan, in the years that preceded and during the Second World War. The Japanese immigrants settled in Cuba and formed a colony at Youth’s Island, to the south of Havana (Rodríguez-Manzano et al., 2001). The greatest variation in Colocasia in Cuba for is reported in the central region of the country, probably due to the fact that large numbers of Spaniards from the Canary Island settled in that area, and introduced the crop. However, new evidence indicates that the eastern region is also an important source of variability for taro, due to the discovery of a wild stoloniferous type. Given the importance of taro in the diet of the Cuban population, it is necessary to preserve its diversity, not only in genebanks (ex situ conservation), but also by means of in situ conservation, by farmers in their rural home gardens, and also in urban and peri-urban gardens (Rodríguez-Manzano et al., 2000). It is also necessary to preserve wild populations in their natural settings in Cuba. The objective of the present work was therefore to investigate the principal factors that limit taro production in Cuba. Possible future strategies are also discussed. Materials and methods Taro production on state farms was analyzed from the annual reports of the Cuban Ministry of Agriculture (Manso, 2001). In the urban setting, production is carried out with intensive, organic technology (Rodríguez-Nodals et al., 2002a). Agro morphological, botanical, cytogenetic, and isoenzymatic characterization of an ex situ collection of 42 taro clones was carried out, following the methodology proposed by Rodríguez Manzano et al. (2002). To select promising clones for use in breeding programmes and in production, the following variables were considered: the average of the all experimental and commercial yields, palatability, chromosomes number and presence of inflorescences. Information on the variability of the crop according to farmers’ perceptions was also obtained through interviews in different home gardens, selected from a project on on-farm conservation, as recommended by Castiñeiras et al. (2000). Finally, expeditions to the mountainous areas of the eastern region of Cuba were carried out, searching for new sources of variability, and surveys of farming families about their use were carried out. From the information obtained, the main factors that limit production were analyzed in the various farming systems, as well as the possible strategies for the development of new investigations about taro in Cuba. Results and discussion Taro production in Cuba The maximum taro production obtained in Cuba was 224700 t in 1979, when a great area of irrigated land was dedicated to its cultivation. Yields have diminished in the subsequent years, reaching 3850 t in 1990. From that year, the decrease in taro production was slower, and in 1998 had been reached 2590 t. From then it has registered a modest increment, and in 2000 reached 4640 t. It has remained stable in the last years. third taro symposium 155 During the years 1990-2000, Cuba crossed through a special period, as a consequence of the economic blockade imposed by the United States, and the collapse of the Socialist Block. This necessitated totally reorganizing the external trade of the country, and many sectors of the economy were affected, among them the agriculture, because of the lack of the necessary fuel to guarantee the irrigation of the crops and other key agricultural practices. Taro was one of the crops affected. The sowed area decreased considerably, causing a great deficit in the markets, with a consequent increase in prices. This situation also affected Xanthosoma, but Colocasia was more affected, mainly because it requires more irrigation water. This is the main limitation to expanding its area of cultivation, although irrigation by gravity, or other techniques that do not require high fuel consumption have successfully been used. Thus, Colocasia continues to be preferred, since the improved clones produce higher yields (more than 36 t/ha) under intensive cultivation, and with a cycle of 8-10 months, in comparison with approximately 20 t/ha over 12 months with the best clones in Xanthosoma. Some taro clones have very high potential yields of 60 and 70 t/ha, and both corms and cormels can be eaten. It has been shown that viral diseases do not affect taro yields in Cuba because of the presence of a less severe stock of the Dasheen Mosaic Virus, which protects it against the more severe one. The virus problem is more prevalent in the case of commercial clones of Xanthosoma, with a significant decrease in yields detected (Quintero et al., 1999). Important results have been obtained in Cuba with regards to the production of in vitro explants free of virus and endogenous microorganism (Mederos et al., 1995; Quintero Fernández et al., 1999; Rodríguez-Manzano et al., 2003). A methodology is also available “to vaccinate” the in vitro cultures or plants with the less severe strain of Dasheen Mosaic Virus (DMV) (Quintero et al., 1999). Main uses of taro in Cuba Taro is a common ingredient in Cuban cuisine. The corms and cormels are eaten in various ways: as a puree or cooked in pieces, with only salt, or perhaps with oil, onion and garlic; in the form of flour; in fine fried slices or fritters; as sweets and as a gelifying agent in ice creams. It is also very much used in soups, with bean stews and in a typical dish called “caldosa” or “ajiaco”, which is a meat (usually pork) broth with a mixture of taro pieces and of other species like sweet potato (Ipomoea batatas (Lim.) Lam); cassava (Manihot esculenta Crantz); yam (Dioscorea spp); banana and plantains (Musa spp). Taro is also used as a specialty food for children and old people, and also for people with digestive illnesses. Conservation and characterization of ex situ taro collection The decrease in the area of cultivation of taro and the prevalence in production of only one commercial clone (“Camerún 14”), gives the conservation and characterization of the national ex situ collection added importance (Rodríguez-Manzano, 2001). A five-stage working methodology for the characterization of taro genetic resources has been developed, which can also be applied to other root and tuber crops (Rodríguez-Manzano, 2001; RodríguezManzano et al., 2002): 1) Characterization and evaluation of the accessions using the descriptors recommended by the IPGRI, with the possible incorporation of new descriptors or modalities depending on the variability in the collection. 2) Use of multivariate statistical analysis to identify the more important descriptors determining variability, and to select the indispensable minimum descriptors for clone identification. 3) Elaboration of a key for clone identification 4) Realization of cytogenetic and molecular analyses to identify the presence of duplicates in the collection, as well as to clarify aspects of phylogeny 5) Selection of promising clones for their use in breeding programs and in production taking into consideration yield, pest resistance and quality In Cuba, the first evaluation of genotype x environment interaction in taro was carried out in five sites during three years, using four well morphologically differentiated clones: “Isleña Japonesa”; “Selección Herradura”; “Isleña Rosada Habana” and “Isleña Miranda”. In this study, genotype x environment interactions were highly significant, and also year x environment ones, but significant differences did not exist for genotype x year interactions, nor for genotypes x years x environment interactions (Rodríguez-Nodals, 1984). Starting from the morphological study of the 42 clones of the ex situ collection over 2 years, and keeping in mind the number of chromosomes and also whether inflorescences were present or not, it was possible to select 17 clones with high yields and good palatability. Among these, are the three present commercial clones: “Camerún 14”, “Isleña Rosada Habana” and “MC-2” (Ministerio de la Agricultura, 1998). Further studies on pest resistance, genotype x environment interaction and crop management, among other aspects of interest, were carried out on the other clones. Among the selected clones, 11 had a good or delicious taste, high yields and inflorescences; of them, three can be used as diploid parents (“Isleña Rosada Jibacoa”, “Rosada Sancti Spiritus” and “Panameña”), and eight as triploid parents (“Isleña Rosada #1”, “Isleña Rosada Mayajigua”, “Isleña Cienfueguera”, “Isleña Yabú”, “CEMSA 75-11”, “Camerún 14”, “Madere Blanc” and “Isleña Rosada Sabanilla”), when their female or male fertility is verified. In any case, mating barriers can also be eliminated through protoplast fusion. These clones cannot only be recommended for breeding programmes, but also to improve production. 156 third taro symposium Attempts have been carried out in Cuba to obtain taro true seeds, but it has not so far proved possible (RodríguezNodals, 1984; Rodríguez-Manzano and Rodríguez-Nodals, 2002). However, some clones have been obtained by selection of somatic mutations (Rodríguez-Nodals, 1984; Rodríguez-Manzano, 2001), as well as by irradiation, starting from in vitro cultures (Milián et al., 2001). Breeding programmes can be developed by biotechnology methods, like somaclonal variation, in vitro mutations induction, genetic transformation and protoplasts fusion, as well as by hybridization. Wild and semi-wild accessions introduced from the center of origin could also be used as gene sources for the improvement of flowering, environmental adaptability and pest and disease resistance(Rodríguez-Manzano et al., 2002). This work provides the basis for improving the clonal composition of this crop, that at the present time is quite narrow, with only one clone in production (“Camerún 14”) (Ministerio de la Agricultura, 1998), and new clones obtained from breeding also possessing the same genetic base (Milián et al., 2001). Taro production in the Urban Agriculture Program The National Urban Agriculture Group (GNAU) detected the importance of taro in urban and peri-urban farms, and the associated demand, and an intensive technology production on organic basis in small areas was proposed, with water supply and the use of traditional clones adapted to each region. This means that the crop is produced near population settlements, avoiding the expense of fuel for transportation from distant places (Rodríguez-Nodals et al., 2002a). The Urban Agriculture Guidelines of the Root and Rhizomes Subprogram describe the main aspects that should be taken into account (GNAU, 2003), and the recommended strategies for production in these systems: 1) The use of clones with local adaptation. For example: “Isleña Rosada Sabanilla” in Union de Reyes, Matanzas province; “Isleña Japonesa” in the Youth’s Island; “Isleña Herradura” to the south of Pinar del Río province; “Isleña Rosada Mayajigua” in Yaguajay, Sancti Spiritus province ; “Isleña Rosada Escambray” in the mountainous of the Guamuhaya Massif, in Cienfuegos province, “Isleña Filé”, in the mountains of Santiago de Cuba province, etc. In other cases, clones with general adaptation could be used, such as the current commercial clones: “Camerún 14” and “Isleña Rosada Habana”. 2) Seed selection and appropriate manipulation (Rodríguez-Nodals et al., 2002b). 3) Planting at the appropriate time. In Cuba the best time is between January and March, though it is also possible to plant from November to April. The planting distance depends on the size of the cormels to be used (Rodríguez-Nodals et al., 1990). 4) Appropriate fertilization, using very aggressive products and practices. For example, application of 200 t/ha of organic matter is recommended, depending on the available type of manure. This is important for the positive impact that this has since on the environment, reducing considerably the use of chemical fertilization. 5) Appropriate handling of irrigation, since Colocasia is water demanding but does not withstand flooding. 6) Cultivation practices. Cultivation should be done as necessary and possible, up to 90 days after germination, preferably with animal traction, with the objective of to maintaining the conformation of the soil, and to facilitate the development of the root system, while managing weeds. 7) Harvesting should be carried out starting 10 months after germination, with an optimum between 10-12 months. If taro is harvested after 10 months, irrigation should be maintained, always stopping it about 15 days before harvesting. In the case of “Camerún 14”, resistant to the attack of the fungus Rizogliphus, harvesting can be delayed to 14 months, without loss of quality. The network of 162 seed farms belonging to the National Movement of Urban Agriculture can satisfy the local demand for planting material production. For example, a clone with high yield potential, “Isleña Filé”, was detected in Santiago de Cuba province (Figure 1), and this was very much in demand from residents of Tercer Frente Municipality. For this reason, it was multiplied in the local seed farm. This contributes to the rescue of traditional varieties, supporting their use and demand by local farmers. Another important aspect to highlight is the possibility that Cuba has for the massive in vitro propagation of plants, because of having a considerable installed capacity, starting from the existence of eleven plant factories. At present, production is limited because of the lack of the necessary inputs to produce the necessary “in vitro seeds” for peasants in the urban and periurban areas. third taro symposium 157 Figure 1: Taro field in the Sierra Maestra slopes In situ conservation of cultivated taro in Cuba Farmers conserve a great diversity of cultivated plants species, and a high diversity of forms within each species, on their farms and orchards, ensuring the food self-sufficiency of the family. Each species and variety plays a specific role within the family alimentary economy (Eyzaguirre and Watson, 2002 ). A study was carried out of the perception of the farmers as regards taro diversity. It showed the presence of 16 distinct types in 12 home gardens in the Western and Central regions of Cuba (Castiñeiras et al., 2000; Fundora-Mayor et al., 2003). These taro clones were only present in home garden of these two regions, so they should be studied and monitored, and it is also important to educate farmers as to the importance of the material they are maintaining to ensure that it is not lost in the future. The greatest variability was found in Cienfuegos province (central region) , with 11 types, followed by Pinar del Río (western region), with 5types. The maximum variability per “conuco” was three types, in one conuco in Cienfuegos and another in Pinar del Río. In Cienfuegos, only two “conucos” of those studied did not cultivate taro; however, in Pinar del Río, 10 “conucos” did not cultivate any taro. No taro was reported in home gardens in Guantánamo province (eastern region). This coincides with reports by other authors, who affirmed that the central region is the richest in variability for this crop. Nevertheless, new expeditions carried out recently indicate that it will be necessary to continue exploring other areas in the east of Cuba, searching for new variability. Visiting producers in other provinces, in both urban and rural areas, different phenotypes were detected within the same plantation, deliberately planted together. For example, one producer has clones without pigmentation in the leaves, together with clones with purple pigmentation in the center of the leaf, and also others with purple pigmentation in the center of the leaf and in the two basal veins. It would be interesting to use both morphological and molecular markers to detect the areas with the highest genetic diversity and largest genetic erosion and to deepen or understanding of the relationship among clones in Cuba. New explorations in the Eastern Region of Cuba The eastern region of Cuba possesses few reports of taro variability. For this reason, an expedition was organized that covered poorly explored areas. New variability was indeed encountered, in particular a wild type. The wild taro type was found in mountainous areas of the eastern region, in the Sierra Maestra slopes, Tercer Frente Municipality, Santiago de Cuba province, near to Filé town, specifically in the Saltón River. It was growing in a semicaducifolious forest, under different conditions: a) in a vertical slope, near the waterfalls (Figure 2); b) in the river bed formed by calcareous rocks (Figure 3); c) growing between big rocks (Figure 4); and d) in the sandy bed of the river (Figure 5). This coincides with the habitats Matthews’ (1997) description on the main characteristics of the places where wild types of taro grow in the tropical forests of Queensland. It was very common to find these wild plants growing together with the jipi-japa plants (Carludovica palmata Ruiz et Pav.) (Figure 3). This plant apparently produces a trap effect, retaining segments of the stolons transported by the river. It is interesting to highlight that this plant is probably native to Central America (Bolivia, Perú and Chile) (Esquivel et al., 1992). 158 third taro symposium Figure 2: Taro plants in a vertical position, nearby the water fall Figure 3: Taro plants on the calcareous rocks Figure 4: Taro plants on soil, among calcareous rocks Figure 5: Taro plants along the sandy river bed It is not easy to explain the presence of wild taros in Cuba, since their recognized center of origin is in Southeast Asia, where they are found near rivers or other watercourses in humid tropical forests (Matthews, 1997). The possible hypotheses on the presence of the wild taro in Cuba are the following: 1) That they are clones escaped from cultivation, from rural areas where taro is cultivated, and during many years (more than 400 years) stoloniferous mutations have arisen, adapted to the aquatic and semi-aquatic conditions. 2) That man found taro plants in the water, and later he adapted them to the terrestrial habitat. That is to say that man’s selection could have been made in two directions: from aquatic habitats to terrestrial ones, and vice versa, from terrestrial habitats to aquatic ones. The most probable thing is that the adaptation to aquatic habitats is a re-adaptation; i.e., from aquatic habitat the species has adapted, in a natural way, by means of seed dispersion, or from the transference of tubers by man to the soil, and then it escapes from cultivation and re-adapted spontaneously to water habitats. Another approach is a re-adaptation, mediated by man, intensively cultivating them in marshy, irrigated or flooded areas. On the other hand, an introduction of a wild taro is not very probable, since stoloniferous forms have not been reported in the country either in older collections (Roig, 1913), or in the current national collections. It cannot be suggested that this wild taro originated in Cuba, since botanists do not accept a New World origin for C. esculenta. A combination of the first two hypotheses is probably the explanation. Since the introduction of taro in Cuba, new variability has been generated due to the interaction of different factors, like the vegetative reproduction, together to third taro symposium 159 man’s selection, the inability of some clones for emitting inflorescences, the sterility, the autopoliploidy, the genomic mutation and the structural changes in the chromosomes (Rodríguez-Manzano and Fundora-Mayor, 2002). All the plants of the wild taro type had stolons (Figure 6) that were more than a meter long in some cases, mainly when the plants were on the rocky bed of the river, in places where some soil exists. The plants collected in the sandy bed of the river, or among the rocks, or even in the rocky bed of the river, possessed stolons and many lengthened roots; the petioles, the sheaths and the lamina are green. The pulp of the corms and stolons are white; the buds and roots rosy and white, in different plants. Inflorescences were not observed, either in their natural environment, or after 15 months of cultivation. Figure 6: Detail of the stolons When sowing these plants in the field, they kept their stolons. In the first months of cultivation, these developed on the surface of the soil, and after five months began to go into the soil and to develop new plants. The palatability tests carried out allowed us to classify them as “delicious”, both stolons and corms. These results coincide with the surveys carried out with farmers from the area where the wild taro grows, who reported that in the “special period” of 1990-2000 they used the wild plants from the river for food and also to feed domestic animals. The discovery of the wild taro in Cuba opens new research perspectives, since up to this moment this region of the country was the one that had the least variability. Now the region becomes into an important potential source of new variability. Other aspects on the introduction of taro in the Caribbean can be clarified with molecular studies, comparing with clones from different parts of the world. The Ministry of the Agriculture of Cuba has within its objectives the conservation and use of taro genetic resources for people’s nutrition, among other species of this group (Ministerio de la Agricultura, 1999), so the results of this work contribute to the successful development of this crop. Conclusions 1. Taro production in Cuba has diminished due to the economic constraints faced by the country in the last years; stability has now been reached, due to a modest economic recovery of the sector, supported by the use of new strategies and technologies. 2. Investigations carried out on the ex situ collections, and on the in situ conservation of taro by the farmers through their use, allow us to develop strategies to avoid genetic erosion of existing variability of taro in Cuba, as well as to support the production programs and genetic improvement of this species. 3. The presence of a wild stoloniferous type of taro is reported for the first time in Cuba, in the eastern region, that presents new questions regarding the introduction and evolution of this species in the Caribbean. 4. Funding is necessary to support new expeditions, the in situ conservation of this species in different areas, and an increase in taro production in urban and periurban areas, that will help to ensure the food security of urban people, which comprise 76% of the total Cuban population. References Castiñeiras, L., Fundora, Z., Shagarodsky, T., Fuentes, V., Barrios, O., Moreno, V., Sánchez, P., González, A.V., Martínez-Fuentes, A., García, M. and Martínez A. 2000. La conservación in situ de la variabilidad de plantas de cultivo en dos localidades de Cuba. Revista del Jardín Botánico, Universidad de la Habana, XXI(1):25–45. 160 third taro symposium Esquivel, M., Nüpffer, H. and Hammer, K. 1992. Inventory of the cultivated plants. Vol. 2, Chap. 14. In: Hammer, K., Esquivel, M. and Nüpffer, H. (eds). Y tienen faxones y favos muy diversos de los nuestros ... Origin, evolution and diversity of Cuban plant genetic resources, Inst. Pflanzengenetiku Kulturpflanzenforsch, Gatersleben. Eyzaguirre, P. and Watson, J.W. 2002. Home gardens and agrobiodiversity: An overview across regions. p. 10–13. In: Watson, J.W. and Eyzaguirre, P. (eds). Home gardens and the in situ conservation of plant genetic resources in farming systems: Proceedings of the Second International Home Gardens Workshop, Witzenhausen, Federal Republic of Germany, 17–19 July 2001. IPGRI, Rome. Fundora-Mayor, Z., Castiñeiras, L., Shagarodsky, T., Barrios, O., Fernández, L., Moreno, V., Cristóbal, R., RodríguezManzano, A., García, M., Hernández, F., Giraudy, C., Fuentes, V., Sánchez, P., Valiente, A. and González, A.B. 2003. Percepción local de la diversidad infraespecífica de las especies presentes en los huertos caseros de tres zonas de Cuba. Memorias del V Encuentro de Agricultura Orgánica, La Habana. In press. GNAU. 2003. Lineamientos para los subprogramas de la Agricultura Urbana para el año 2003 y sistema evaluativo. Grupo Nacional de Agricultura Urbana, MINAGRI, Agrinfor. 96 p. Ivancic, A. and Lebot, V. 2000. The genetics and breeding of taro. CIRAD. 194 p. Manso. 2001. Informes sobre las producciones de malanga isleña en Cuba. MINAGRI, La Habana, Cuba. Matthews, P.J. 1997. Field guide for wild-type taro, Colocasia esculenta (L.) Schott. Plant Genetic Resources Newsletter 110:41–48. Mederos, V., Magaly García, Cabrera, O., López, J. de la C., Ventura, J., Diosdada Gálvez and Álvarez, M. 1995. Optimización de la tecnología de micropropagación de la malanga. Avances en Biotecnologia Moderna 3:11–18. Milián, M., Sánchez, I., García, M., Rodríguez, S., Guerra, D., Portieles, J.M., Hernández, M., Corrales, A., Lago, M.A., García, J. and Oliva, M. 2001. Nuevos clones de malanga isleña (Colocasia esculenta (L.) Schott) en Cuba. Revista Centro Agrícola 4(28):47–54. MINAGRI (Ministerio de la Agricultura). 1998. Instructivo técnico sobre el cultivo de la malanga. SEDARI/ AGRINFOR, La Habana, Cuba. 24 p. MINAGRI (Ministerio de la Agricultura). 1999. Informe sobre situación actual de los recursos genéticos en el ministerio de la agricultura de Cuba. 20 p. Quintero Fernández, S., Rodríguez Nodals, A., Rodríguez Manzano, A., Maribona, R.H., López, M., Pérez, S., Cala, M., Proenza, M., Rodríguez, A.J., Fundora, Z., Peralta, E.L., Olivera, H., Pérez, D., Pérez, O., Morales, N., Mojena, D. and Socorro, A. 1999. Recuperación del cultivo de la malanga (Xanthosoma sp.) mediante procedimientos biotecnológicos. Convención tropico/99 (Agricultura Tropical) PSM. Soft Cal. No 16. Rodríguez Manzano, A. 2001. Conservación y manejo de las plantas de reproducción asexual: Raíces, rizomas y tubérculos. p. 255–272. In: Fundora, Z., Castiñeiras, L. and Fernández, L. (eds). Lecciones de avanzada sobre conservación y manejo de Recursos Fitogenéticos. INIFAT, La Habana, Cuba. Rodríguez Manzano, A. and Fundora Mayor, Z. 2002. Variability of ‘malanga isleña’ Colocasia esculenta (L.) Schott in Cuba. Procicaribe News 10, Abril. 16 p. http://www.procicaribe.org/main/news/pn-2002-04.pdf. Rodríguez Manzano, A. and Rodríguez Nodals, A. 2002. Diversidad de la malanga isleña Colocasia esculenta (L.) Schott en Cuba. III Inflorescencias. Revista Jardín Botánico Nacional XXIII(1):119–126. Rodríguez Manzano, A., Rodríguez Nodals, A. and Quintero Fernández, S. 2000. Caracterización de germoplasma y mejoramiento participativo en especies de raíces y tubérculos alimenticios tropicales y musáceas en Cuba: Fitomejoramiento participativo en América latina y el Caribe. Programa de Investigación Participativa y Análisis de Género del GCIAI (Programa PRGA), http://www.prgaprogram.org/prga. Rodríguez Manzano, A., Quintero Fernández, S., Rodríguez Mansito, A.J. and Fundora Mayor, Z. 2003. Establecimiento in vitro de ápices de malanga (Xanthosoma sagittifolium Schott). Cultivos Tropicales 3. Rodríguez Manzano, A., Rodríguez Nodals, A., Román Gutiérrez, M.I., Fundora Mayor, Z. and Castiñeiras, L. 2001. Morphological and isozymatic variability of taro Colocasia esculenta (L.) Schott germplasm in Cuba. Plant Genetics Resources 126:31–40. Also p. 534–543 in: Nakatani, M. and Komaki, K. (eds). Proceedings of the Twelfth Symposium of the International Society for Tropical Root Crops: Potential of root crops for food and industrial resources,Tsukuba, Japan, 10–16 September 2000. ISTRC, 2000. Rodríguez Manzano, A., Rodríguez Nodals, A., Román Gutiérrez, M.I., Fundora Mayor, Z., Castiñeiras Alfonso, L. and Manzano Figueredo, M.J. 2002. Metodología para la caracterización de germoplasma y variabilidad infraespecífica de Colocasia esculenta (L.) Schott en Cuba. CENDA, La Habana, Cuba. Rodríguez Nodals, A. 1984. Mejoramiento genético de los cultivos de raíces y tubérculos tropicales en la República de Cuba. Tesis de Doctor en Ciencias Biológicas. Godollo, Hungría. 232 p. Rodríguez Nodals, A., Rodríguez Manzano, A., Quintero Fernández, S. and Sánchez Iglesias, A. 2002b. Tecnología para los huertos intensivos de raíces tuberosas y rizomas tropicales. GNAU-MINAGRI, La Habana, Cuba. 16 p. third taro symposium 161 Rodríguez Nodals, A., Batlomovna, F., García, M., Hernández, M., Lima, M., Mederos, V., Pino, J.A., Quintero, S., Rodríguez, S. and Morales, A. 1990. Recomendaciones para la multiplicación de propágalos en viandas tropicales, IDP, La Habana. 37 p. Rodríguez Nodals, A., Rodríguez Manzano, A., Sánchez Iglesias, A., Prats Pérez, A., Fresneda Buides, J., Benítez Alzola, M.I., Carrión Ramírez, M., Fraga Aguirre, N., Barrios Govín, O., Avilés Pacheco, R., Quintero Fernández, S., Chavéz Rojas, T.H. and Muñoz De Con, L. 2002a. Manual técnico para la producción de semillas en la agricultura urbana. II. Hortalizas y Propágulos. INIFAT-GNAU-PNUD, La Habana, Cuba. 80 p. Roig, J.T. 1913. Las especies y variedades de malanga cultivadas en Cuba. Estación Experimental Agronómica, Santiago de las Vegas. 21. Roig, J.T. 1965. Diccionario botánico de nombres vulgares cubanos. Vol. 2. Ed. Pueblo y Educación, La Habana, Cuba. 949 p. 162 third taro symposium Theme Three Paper 3.9 Taro (Colocasia esculenta (L.) Schott var. esculenta): Production, constraints and research in Dominica and other Caribbean countries Gregory C. Robin Caribbean Agricultural Research and Development Institute, Roseau, Commonwealth of Dominica Introduction Taro is an important staple in the Caribbean and is widely grown in St. Vincent, Jamaica, Dominican Republic and the Commonwealth of Dominica. The entire plant is used for human consumption. The corms are traditionally an important energy source and the young leaves are used as a vegetable, mainly for the preparation of a popular dish called “Calalou”. Taro is sometimes used for feeding animals, particularly pigs (Bown, 1988). Taro production is suited to the high rainfall (3800-5000 mm annually) conditions, which exist in certain locations of Dominica (Barker, 1981). Dominica, although a relatively small root crop producer, in terms of the total quantity produced, has the highest per capita production for all root crops (342 kg) among Caribbean countries (Ferguson, 1985). After Jamaica, Dominica is the second largest regional producer of taro, which is also the most important root crop produced in the country (Ferguson, 1985). The per capita production of taro in Dominica is 146 kg. This is the highest per capita production of any single root crop in the region (Ferguson, 1985). Taro is grown year round throughout Dominica, the major producing areas are Wet Area, Grand Bay, Grand Fond and Belles. The taro cultivar called “Comme” or “Common taro” is the predominant variety grown in Dominica and is recommended for export to the UK market. This cultivar forms a single corm, which tends to be oval to round in shape (Prevost, 1977; Robin, 1993). Among the known cultivars it suckers the least and therefore has the least scars (Prevost, 1977). The flesh is light blue in colour after cooking. The “White Taro” is the predominant cultivar grown in St. Vincent. It is exported both to the UK and U.S. markets. In recent years, there has been an increasing demand for taro in the drier Leeward Islands, the United States Virgin Islands (USVI), the French West Indies and the expanding ethnic market in the United Kingdom (UK) and Holland. This has resulted in a five-fold increase in taro exports during the past ten years. Though taro production and exports are on the increase, exports are still constrained by post-harvest rots due to scarring. Scarring is caused by the removal of suckers during harvest and post-harvest activities (Adams et al., 1985; Cooke et al., 1988 and Crucefix, 1990). These scars make the corms un-presentable and are sites for disease infection (Adams et al., 1986; Wickham and Elango, 1990) and rotting (Crucefix, 1990). Spacing (Cable and Asghar, 1981), depth of planting (Robin, 1990) nitrogen fertilizer (de la Pena, 1990), moisture (Ezumah and Plucknett, 1981; de la Pena, 1983; Pardarles, 1985) and weed competition (Gurnah, 1985) are all known to affect suckering. Observations have shown that in Dominica corms of the “Comme” variety have different levels of suckering, depending on location, time of planting and the cultural practices implemented. However, how these agroecological zones and cultural practices impact on suckering had not been scientifically examined. The choice of taro corms for export is normally based on physical specifications - weight between 0.9 and 1.8 kg, 10 to 10.5 cm in diameter, 15.5 to 17.0 cm long, oval to round in shape i.e. diameter/length ratio of 0.35 to 0.5 indicates the corm is “dumb-belled” shape (the least acceptable shape), 0.5 to 0.6 partially “dumb-belled” shape, 0.6 to 8.5 oval and 0.85 to 1.0 round (the most acceptable shape) (Robin, 1993), and scar and disease free (Medlicott, 1990; Crucefix, 1992). The effects of environment, location, seasonality, spacing and depth of planting on these specifications were examined (Robin, 1993). Presently, very little emphasis is placed on age, maturity, texture and taste characteristics, all of which can affect corm quality and consumer acceptance. Studies by Constantin et al. (1974), Purcell et al. (1976), Tom and Hernandez (1978) and Bradbury and Holloway (1988); indicated that the environment and degree of maturity of root crops (sweet potato, yam and Colocasia) affects their nutritional composition and yield. In Dominica, taro corms are normally harvested between 6 and 10 months after planting. However, the time to maturity for taro corms may vary from one agro-ecological zone to another. Farmers reported that in Grand Bay, with an average annual rainfall of 2400mm and a marked dry season, on soils characterized as plastic sticky clay loams without a silica pan (Barker, 1981), taro corms can mature in as early as 6 months (pers. comm.). Farmers also reported, that in the Wet Area where soils characterized as sandy clay loams, and average annual rainfall of approximately 5300 mm (Barker, 1981), taro corms mature in 8 to 9 months after planting (pers. comm.). De la Pena (1983), found that moisture levels affect the time to corm maturity. Batch exports of taro corms are not normally location specific. Therefore exported corms while appearing similar in shape, size and weight may differ in age, maturity and origin. These differences are thought to affect shelf life and eating quality, and require investigation. third taro symposium 163 Two studies were conducted. The first study investigated, how variations in crop density and planting depth affected suckering of the “Comme” taro, when grown in the wet and dry seasons, in two contrasting agro-ecological zones of Dominica. The second study investigated the effects of corm age at harvest on yield and yield characteristics, protein composition, shelf life and palatability of taro corms grown in the same agro-ecological zones. These parameters were used to determine the optimal time for harvesting corms. In this study corm age is the time between planting and harvesting of the crop. Materials and methods Experiment 1 Experiments were conducted in two of the major taro producing areas: Grand Bay and the Wet Area. Table 1 describes the two locations. Table 1: Climatic, topographic and soil data for the two agro-ecological zones in Dominica where the experiments were conducted Agro-ecological characteristics Average annual rainfall (mm) Rainfall pattern Location Grand Bay (A2) Wet Area (D3) 2400 5300 Dry season Jan. to May Mild or no dry season Temperature (°C) 27 25 Altitude (m) 235 500 Soil Type Clay loam Sandy clay loam Sand (%) 37 60 Silt (%) 23 19 Clay (%) 40 25 Bulk density (g/cc) 1.1 0.6 Porosity 0.6 0.8 The experiments examined the effects of three planting depths 20, 25 and 30 cm and three spacing - 55x55 cm (33, 025 plants per hectare (pph), 65x65 cm (23,645 pph) and 75x75 cm (17,760 pph) in a 3x3 factorial arrangement. The nine treatments were laid out as a randomized block with three replicates at each site. Plot were 3.75x4.5m, each plot contained a total of 56, 40 and 30 experimental plants, for the treatments 55x55 cm, 65x65 cm and 75x75 cm respectively. Both sites were cleared of unwanted vegetation and sprayed with paraquat (25cc/l) before planting. Suckers of the variety “Comme”, with the upper 2 to 4 cm of the corm intact, a basal diameter of 5.0 to 7.0 cm, and a mean weight of 245±15 g, were used as planting material. Suckers were selected from the most vigorous plants, cleaned of all roots, dead tissue and soil; then dipped in a solution of bleach (containing 2% sodium hypo-chlorite) for 15 to 20 minutes. Petioles were cut back to a length of 25 to 30 cm. Planting materials were all taken from the same source. Wet season plantings in Wet Area and Grand Bay were carried out in May. Dry season plantings in Grand Bay and Wet Area were carried out in December and January respectively. The experimental plots were kept weed-free during the first 3 months. Paraquat (20cc/l) was used to control weeds before canopy formation. Subsequent weddings were done manually. At 0.5 and 2 months after planting, 57g of the NPK + MgO (16:8:24:2) fertilizer was banded around the plants. Within one to two weeks after the second fertilizer application, soil was mounded to a height of 6 to 8 cm around the base of each plant petiole, by moving soil from within a radius of 30 to 35 cm around each plant. Manual harvesting with a fork and cutlass took place 9 to 10 months after planting. Daily rainfall data were collected throughout the experiment. The number of visible suckers around the main corm of all experimental plants was recorded in each plot one month before harvest. After harvest the number of scars on each corm from each sample plant in each plot was also counted. Experiment 2 Four commercially established taro farms were selected in Grand Bay and the Wet Area respectively. Suckers were used as planting material on these farms. The taro farms selected were located in relatively homogenous areas and were well managed. At the time of selection, the age of the plants on each farm was recorded. On each farm in each location a stratified random sample made up of four strata was established. There were a total of 16 strata within each location. Each stratum contained approximately 30 to 40 plants. Within each stratum, three randomly selected plants were harvested monthly. Harvesting commenced when the plants were six months old and ended when the plants were 12 months old. Twelve plants were harvested monthly from each stratified random sample. A total of 48 plants were harvested each from the Grand Bay and Wet Area locations. Data from the main corms in the 16 strata in each location were pooled together when calculating monthly means. 164 third taro symposium Corm diameter and length measurements were made using a caliper. Corm shape was approximated by the diameter to length ratio (DLR). Corms were weighed then submerged in water to measure their volumes by displacement. Corm weight was divided by volume, to determine the specific gravity. Half of these corms were then randomly selected for dry matter, crude protein and palatability studies. The other half was used for shelf life studies. For corm dry matter studies, longitudinal sections from each corm, approximately 4 cm wide and 15 cm long were peeled then grated into small bits, 10 g of bits were then placed in a pre-weighed crucible and dried to a uniform weight over 16 hours at temperatures of 100°C. The crucibles were then allowed to cool in a desiccator before weighing. After drying, dasheen samples were ground into a fine powder. Percentage nitrogen and crude protein were measured using the Kjeldahl method. Palatability tests were undertaken by a group of 15 to 20 panelists, using longitudinal sections from the same corms used for dry matter studies. The sections were peeled, and then boiled until the flesh became soft. The sections were then cut into 2.5 cm cubes, labeled with a three-digit number using a table of random numbers and then placed randomly on plates of similar size. Panelists tested each sample. After each sample was tasted, panelists were required to gargle with water in order to remove left over tastes. A scale, ranging from 1-5 (1=Dislike a lot, 2=Dislike a little, 3=Neither like nor dislike, 4=Like a little and 5=Like a lot) was used to quantitatively assess the degree of acceptance. For shelf life studies, corms were cleaned in running water 3-4 hours after harvest and then dipped for 2-3 seconds in a solution of Ridomil mbc 60wp (14 g/28 litres of water). The corms were allowed to air dry and then stored in a cool aerated room at temperatures between 26-30°C. Corms were monitored daily for incidence of softening, sprouting, fungal infections and shriveling. Corms, which showed symptoms of the above, were removed from storage, and the number of days from harvest up to the time of removal was recorded. Results and discussion Experiment 1 Cumulative monthly rainfall shown in Tables 2 and 3, were higher in Wet Area both in the wet and dry seasons. Rainfall received by the crop during the critical growth period (0-6 months after planting), were 990 mm and 1300 mm higher in Wet Area during the wet and dry seasons respectively. Rainfall distribution in Grand Bay was irregular as compared to that of Wet Area, particularly in the dry season. Table 2: Comparison of the cumulative monthly rainfall (mm) received by the taro plants during the wet season at Grand Bay and Wet Area in Dominica Months after planting Location 1.5 3 4.5 6 9 Grand Bay 340 550 980 1230 2000 Wet Area 480 1150 1700 2220 4180 Table 3: Comparison of the cumulative monthly rainfall (mm) received by the taro plants during the dry season at Grand Bay and Wet Area in Dominica Location Months after planting 3 4 5 6 7 8 9 Grand Bay 450 600 810 1200 1320 1785 2070 Wet Area 1000 1290 1775 2500 3175 3585 NA NA – not available The level of suckering and scarring was higher in Wet Area across seasons. This can be attributed to the higher and more consistent patterns of rainfall observed in Wet Area. Ezumah and Plucknett (1981) indicated that whether suckers matured to contribute significantly to yield depends on water availability. Pardarles (1979) and (1985) reported similar findings. third taro symposium 165 Correlation coefficients of the regression of the number of visible suckers around each plant with the number of scars per corm, for both locations in the wet and dry seasons are shown in Table 4. Table 4: Correlation coefficients of the regression of suckers vs. scars, for Grand Bay and Wet Area locations, during the wet and dry season in Dominica Location Season r value Wet 0.7128*** Grand Bay Dry 0.7351*** Wet Area Wet 0.8647*** Wet Area Dry 0.4971** Grand Bay ** – P<0.01 *** – P<0.001 Since positive and significant correlations were obtained between the number of visible suckers and scars, indications are, that agronomic practices which reduce suckering, will also reduce scarring. Table 5 shows that the closer spacing significantly reduced the number of suckers per plant in both locations during the dry season (P<0.05) when moisture supply was a limiting factor, and interplant competition for soil moisture and nutrients increased (El-Habbasha et al., 1976). Reduction of suckering at the closer plant spacing during the dry season in Grand Bay, also significantly reduced average main corm size and weight per plant below that of the recommended export specifications (Robin, 1993). Weight and size of corm has priority over scarring, therefore it may not be possible to produce dasheen during the dry season in Grand Bay for the export market. Table 5: Mean number of suckers per taro plant, for different spacing during the wet and dry seasons, at the Grand Bay and Wet Area locations in Dominica Number of suckers per plant Spacing (cm) Grand Bay Wet Area Wet Dry Wet Dry 55 x 55 3.4 3.5 4.0 4.3 65 x 65 3.6 4.2 4.1 4.5 75 x 75 3.9 5.1 4.4 5.7 S.E.D. (16 d.f.) 0.51 0.58 0.40 0.42 F test NS * NS * NS – Not significant * – P <0.05 In the Wet Area, as shown in Table 6, increasing plant depth significantly reduced the number of suckers per plant both in the wet and dry season (P<0.05). Bud dormancy seems to be more prolonged at deeper plantings; but how this occurs in dasheen corms is unknown. Table 6: Mean number of suckers per taro plant, for different plant depths during the wet and dry seasons, at the Grand Bay and Wet Area locations in Dominica Number of suckers per plant Depth (cm) Grand Bay Wet Area Wet Dry Wet Dry 20 3.4 4.7 4.6 5.5 25 3.4 4.2 4.9 5.1 30 3.1 3.9 2.7 3.9 S.E.D. (16 d.f.) 0.46 0.6 0.93 0.42 F test NS NS * * NS – Not significant * – P <0.05 Deeper plantings during the wet season also significantly increased average main corm weight per plant, mean corm yields and corm shape also remained oval (Robin, 1993). In the dry season, suckering was significantly influenced by variations in both plant spacing and plant depth. However, it was observed that the difference between the number of suckers (3.9) at the deepest planting (30 cm), and the number of suckers (4.3) at the closest spacing (55x55 cm) was not significant. Deeper plantings significantly increased average main corm weight per plant, mean corm yields per hectare and corm shape remained oval (Robin, 1993). Corm quality in the Wet Area was further enhanced at the deeper plantings as a result of a reduction of the number of scars per corm and maintenance of other marketable corm characteristics. 166 third taro symposium Spacing and plant depth interactions had no significant effect on the number of suckers per plant during the wet and dry seasons. Experiment 2 Table 7 shows the effects of corm age on morphological characteristics, quality, shelf life and taste. In Grand Bay taro corms between the ages of 7 and 12 months satisfied the required export weight specifications. However none of the corms satisfied the export shape specifications (DLR 0.6-1.0). Nine-month-old corms, though not oval in shape (DLR 0.558), were the closest to the required market specifications. Specific gravity (1.00) was highest for the 8-month-old corms. Bowers et al. (1964) suggested, that high corm specific gravity indicates maturity. In the Wet Area, corms between the ages of 7 and 12 months satisfied both the required weight and shape market specifications. Specific gravity (1.01) was highest at 10 months. Table 7: The effects of age (months) on corm physical and nutritional characteristics, shelf life and palatability of corms produced in the Grand Bay (GB) and Wet Area (WA) locations of Dominica Parameters Weight Length (L) (cm) Diameter (D) (cm) Shape Specific gravity (g/cc) Dry matter Protein % Shelf-life Palatability Location 6 7 8 9 10 GB 706 936 975 1145 1263 11 12 SEM 78 1178 1201 1094 1265 WA 560 933 990 1144 996 78 GB 16.9 18.3 19.9 20.4 20.2 21.4 21.4 0.4 WA 16.0 18.5 18.3 18.2 18.2 18.3 17.2 0.4 GB 9.0 9.9 10.3 11.1 11.1 10.7 11.3 0.2 WA 9.3 10.9 10.5 11.1 11.1 11.1 10.6 0.3 GB 0.53 0.54 0.52 0.56 0.55 0.50 0.54 0.004 WA 0.58 0.57 0.57 0.60 0.62 0.60 0.59 0.013 GB 0.97 0.98 1.00 0.96 0.96 0.90 0.98 0.004 WA 0.91 0.96 0.96 0.99 1.01 0.90 0.99 0.013 GB 39.2 41.9 40.6 38.2 36.0 31.0 36.4 1.4 WA - 37.3 44.5 35.9 40.3 36.7 40.1 1.3 GB 1.9 2.8 2.3 3.4 1.7 2.3 1.5 0.2 WA - 4.0 3.7 2.8 2.4 1.7 1.3 0.4 GB 17.5 19.2 21.4 24.5 25.1 36.9 - 3.1 WA 14.3 14.4 25.3 28.8 33.8 - - 3.9 2.9 0.2 2.2 0.2 GB 2.7 2.3 3.8 3.1 3.7 3.6 WA 3.5 3.0 3.4 3.5 4.1 3.5 In Grand Bay dry matter percentages (a factor which impacts on palatability) were the highest for 7 (41.9%) and 8 (40.6%) month old corms. In the Wet Area, dry matter percentages for the 8 (44.5%), 10 (40.3%) and 12 (40.1%) month old corms were the highest. Corms in Grand Bay had high dry matter percentages at an early age i.e. 7 to 8 months; whereas corms in the Wet Area seem to have higher sustained corm dry matter percentages between 8 and 12 months. The crude protein content of the corms in Grand Bay increased up to a maximum of 3.4% at 9 months and then decreased. However in the Wet Area crude protein content was at its highest at 7 (4.0%) and 8 (3.7%) months and then decreased. The percentage crude protein was higher in the younger corms; i.e. corms between the ages of 7 and 9 months. Taro however, is not considered as a protein source and therefore protein content would not be a determining factor when selecting appropriate harvest dates. The effects of corm age on corm shelf life for corms produced in Grand Bay and Wet Area, shows that there were marked increases in shelf life as the corms got older. In Grand Bay the younger corms (6 to 7 months) had a longer shelf life than corms of similar ages in the Wet Area. Whereas, in the Wet Area; the corms of 8 to 10 months had a longer shelf life when compared to corms of similar ages in Grand Bay. The effects of corm age on corm palatability for both Grand Bay and Wet Area shows corm palatability in Grand Bay was more acceptable when the corms were harvested between the ages of 8 to 11 months; the 8 month old corms having the most acceptable taste. Whereas in the Wet Area corm palatability was acceptable between the ages of 6 and 11 months; with the 10-month-old corms having the more acceptable taste. Corm taste in the Wet Area was acceptable over a longer harvest period. Reduced corm palatability was observed at 12 months for both Grand Bay and the Wet Area. In Grand Bay the average maximum corm weight per plant was obtained when corms were harvested at 12 months, the older the corm the longer the shelf life. Maximum mean dry matter and crude protein were obtained from 7 and 9 month old corms respectively; and the best tasting corms were harvested at 8 months. In the Wet Area, maximum average corm weight per plant was obtained when the corms were harvested at 10 months. The best corm shape (DLR 0.621) was also obtained when the corms were harvested at 10 months. However, maximum mean dry matter and mean crude protein were obtained when corms were 8 and 7 months old. Palatability was best when corms were harvested at 10 months. third taro symposium 167 In determining the most appropriate time to harvest corms in Dominica, consideration was given to the producers, exporters and the consumers. Dasheen producers normally use weight, followed by shape, as the main criteria for exporting dasheen. However, using weight as the number one priority would necessitate harvesting corms at 12 months in Grand Bay. The 12-month-old corms have an additional advantage of a long shelf life (38.9 days); which is favorable for export. However, the dry matter (36.4%) and the crude protein (1.5%) content of 12-month-old corms were low. In addition, corm palatability ratings of 2.9 were not highly acceptable. Since consumers and exporters are primarily concerned with a quality product; consideration should ideally be given to physical characteristics (weight and shape), nutritional characteristics (protein) and taste i.e. dry matter and palatability. Therefore, if corms are harvested when dry matter and crude protein percentages were at their maximum (i.e. 7 and 9 months respectively), and palatability was best (8 months); corm weights would not be at the maximum but within the acceptable export weight specifications. Specific gravity of 1.00 an indicator of maturity was also obtained when the corms were 8 months old. Nine-month-old corms were the closest to GRADE-A specifications (i.e. corm weight of 1145g and a diameter length ratio of 0.558) and the shelf life of all corms fell within the acceptable time frame for shipping to Europe. In Grand Bay, the best possible time for harvesting dasheen is at approximately 8 months. In the Wet Area, using weight as the main criteria for harvesting would mean harvesting corms at 10 months. The better-shaped corms were also harvested at 10 months. Since GRADE-A corms and corm palatability were best at 10 months, and dry matter percentages (40.3%) and corm shelf life (33.8 days) at 10 months were high; it seems that dasheen harvest in the Wet Area would be most appropriate at 10 months. Conclusion Reduction of suckers on the dasheen corm across seasons, as a means of improving corm appearance and prolonging shelf-life, without significantly affecting other corm export specifications of size, weight and shape negatively, is obtainable in the Wet Area of Dominica at plantings of 30 cm deep and spacing of 55x55 cm. Agro-ecological conditions does affect corm quality and maturity, therefore harvesting recommendations must vary depending on location. Harvesting dasheen corms at 8 and 10 months in Grand Bay and Wet Area respectively, assures farmers of good economic returns and the consumer also receives a quality product. Acknowledgements The author is grateful to the Caribbean Agricultural Research and Development Institute, who through the Agricultural Research and Extension Project funded by USAID, provided financial support for this research. I am also grateful to the Ministry of Agriculture for providing land for conducting the experiments and the Windward Island Banana Growers Association for providing meteorological data. Thanks are also due to The University of the West Indies for providing supervision and technical guidance and to the CARDI technicians Darnel Shillingford and Jefferson Carbon who assisted in data collection and collation. References Adams, H., Pattanjalidial and Brown, G. 1986. Final report on the increased production of aroids (tannia, dasheen, eddoe) and arrowroot in the Eastern Caribbean. Caribbean Agricultural Research and Development Institute, Botanic Gardens, Roseau, Commonwealth of Dominica. Adams, H., Pattanjalidial and Clarke, A. 1985. Effects of metalaxyl and benomyl on post-harvest decay of dasheen Colocasia esculenta (L.) Schott. p. 143–149. In: Degras, L. (ed.) Proceedings of the Seventh Symposium of the International Society for Tropical Root Crops, Gosier, Guadeloupe, 1–6 July 1985. INRA, Guadeloupe. Barker, G.H. 1981. Dominica: An agricultural profile. CARDI, St Augustine, Trinidad. Bowers, F.A., Plucknett, D.L. and Young, O.R. 1964. Specific gravity evaluation of corm quality in taro. Circular 61. Hawaii Agricultural Experimental Station, College of Tropical Agriculture and Human Resources, University of Hawaii. Bown, D. 1988. Aroids: Plants of the arum family. London: Century Hutchinson. 277 p. Bradbury, H.J. and Holloway, W.D. 1988. Chemistry of tropical root crops: Significance for nutrition and agriculture in the Pacific. Australian Centre for International Agricultural Research, Canberra. 201 p. Cable, W.J. and Asghar, M. 1981. Effect of spacing on taro yield and quality at Laloanea, Western Samoa. Alafua Agricultural Bulletin 8(3). Constantin, P.J., Hernandez, T.R. and Jones, L.G. 1974. Effects of irrigation and nitrogen fertilization on the quality of sweet potatoes. Journal of the American Horticultural Society 99:308–310. Cooke, R.D., Richard, J.E. and Thompson, A.K. 1988. The storage of tropical root and tuber crops: Cassava, yam and edible aroids. Experimental Agriculture 24:257–470. Crucefix, D. 1990. Quality assessment of horticultural produce received in the UK from member countries of the OECS. Natural Resource Institute, UK. 168 third taro symposium Crucefix, D. 1992. Quality assurance workshop standards. CARDI/Division of Agriculture, Dominica. Typescript. 7 p. de la Pena, R.S. 1983. Agronomy. p. 167–178. In: Wang, J. (ed.) Taro: A review of Colocasia esculenta and its potentials. University of Hawaii Press, Honolulu. de la Pena, R.S. 1990. Response of upland taro to pre-plant application of fertilizers and depth of planting. International Society of Tropical Root Crops. El-Habbasha, K.M., Behairy, A.G. and Abol-Magd, M. 1976. The relationship between method of propagation, plant density and growth mineral uptake and yield of dasheen (Colocasia antiquorum Schott). Scientia Horticulturae 4:15–22. Ezumah, H.C. and Plucknett, D.L. 1981. Cultural studies on taro, Colocasia esculenta (L.) Schott: The relationship of leaf development, suckering capacity and plant population with yield of taro grown under different irrigation and land preparation methods. Journal of Root Crops 7:(1) and (2):41–52. Ferguson, T.U. 1985. Root and tuber crops in the Caribbean. p. 191–219. In: Root crops production and research in the Caribbean: Proceedings of a regional workshop held in Guadeloupe, 9–10 July 1985. CIAT, Guadeloupe. Gurnah, A.M. 1985. Effects of weed competition at different stages of growth on yield of taro. Field Crops Research 10:283–289. Medlicott, A.P. 1990. Product specification and post-harvest handling for fruits, vegetable and root crops exported from the Eastern Caribbean: CATCO handbook. CARICOM Export Development Project, St Michael, Barbados. Pardales, J.R. 1979. Role of natural water stress on some agronomic characters of lowland gabi planted in the upland under different fertilizer application. VISCA Vista 2:22–23. Pardales, J.R. 1985. Effect of mulch application and planting depth on growth, development and productivity of upland taro. Annals of Tropical Research 7:27–39. Prevost, L. 1977. Dasheen (Colocasia esculenta) varietal performance carried out at the wet area experimental station in the Commonwealth of Dominica. Typescript. 7 p. Purcell, A.E., Pope, D.T. and Walter, W.M. 1976. Effect of length of growing season on protein content of sweet potato cultivars. Horticultural Science 11:31. Robin, G.C. 1990. The effect of different planting depths on size, shape and weight of dasheen corms. p. 441–446. In: Proceedings of the 26th Annual Meeting of the Caribbean Food Crops Society, 29th July to 4th August, 1990. Caribbean Food Crops Society, Mayaguez, Puerto Rico. Robin, G.C. 1993. Agronomic and post harvest studies on the production of export grade dasheen (Colocasia esculenta (L.) Schott var. esculenta) corms. M.Phil. thesis. Tom, C.S. and Hernandez, T.P. 1978. Wet soil stress effects on sweet potatoes. Journal of the American Society for Horticultural Science 103:600–603. Wickham, L.D. and Elango, F. 1990. Manual of tropical root crops: Production of marketable tubers. Prepared for a workshop on post-harvest handling of tropical root crops. CARDI, Trinidad and Tobago. third taro symposium 169 Theme Four Abstracts THEME 4: Breeding and Distribution of Improved Materials Thème 4 : Sélection Et Distribution de Matériel Génétique Amélioré Genetic diversity of taro (Colocasia esculenta (L.) Schott) assessed by SSR markers Étude de la diversité génétique du taro (Colocasia esculenta (L.) Schott) par analyse de marqueurs SSR J.L. Noyer, C. Billot, A. Weber, P. Brottier, J. QueroGarcia and V. Lebot Following the screening of 96 clones picked from a microsatellite-enriched library, 49 sequences containing repeat motifs were isolated. Fifteen primer pairs were designed. All of them exhibited polymorphism on a subset of 5 taro accessions with a number of alleles ranging from 2 to 8. Seven primer pairs were used to study the genetic diversity of 105 accessions and 100 alleles were revealed. The sample was constituted in order to cover the genetic diversity of Colocasia esculenta as previously described within TANSAO, using AFLP markers. Dendrograms were constructed using the NJTree method based on a similarity matrix computed with a Dice index. Heterozygosity levels were calculated despite of the presence of diploid and triploid accessions. The results are presented and discussed. Taro breeding programme of Papua New Guinea – achievements, challenges and constraints D. Singh, T. Okpul and D. Hunter The Papua New Guinea taro breeding programme was established in the early 1990s with limited resources but has accomplished successful outputs with recent support from the SPC/AusAID-funded TaroGen project. The programme released three leaf blight resistant and high yielding taro varieties (NT 01, NT 02 and NT 03) in 2002, and distributed planting material widely to farmers and extension agencies. In addition to benefiting PNG growers, the taro growers in other Pacific countries will also benefit from these lines, once safe international transfer is guaranteed. The PNG breeding programme has now progressed to its fourth cycle and additional superior lines are expected to be released from these advanced cycles. Major challenges for future breeding programmes will be in the areas of multiplication and distribution, safe movement of germplasm, widening the genetic base of the crop, extending the selection criteria to taro beetle tolerance and superior post-harvest traits, simplifying the breeding using molecular markers, and most importantly the sustainability of the programme. The important constraints to prevail over various challenges will be funding, liaison between the national research and extension agencies, and national scientific capacity. 170 third taro symposium J.-L. Noyer, C. Billot, A. Weber, P. Brottier, J. QueroGarcia et V. Lebot Le criblage de 96 clones extraits d’une banque enrichie en marqueurs microsatellites a permis d’isoler 49 séquences contenant des motifs répétés. Quinze paires d’amorce ont été créées. Toutes ont révélé un polymorphisme (nombre d’allèles compris entre 2 et 8) dans un sousensemble de cinq obtentions de taro. Sept paires d’amorce ont été utilisées pour étudier la diversité génétique de 105 obtentions, et ont permis de révéler 100 allèles. Cet échantillon a été constitué pour couvrir toute la diversité génétique de Colocasia esculenta telle que décrite par des membres du Réseau de recherche sur le taro pour l’Asie du Sud-Est et l’Océanie (TANSAO) à l’aide de marqueurs AFLP. Des dendogrammes ont été réalisés grâce à la méthode NJTree (Neighbor Joining Method) fondée sur une matrice de similitude calculée selon un indice de Dice. Le niveau d’hétérosygotie a été calculé malgré la présence d’obtentions diploïdes et triploïdes. Les résultats sont décrits et commentés. Programme d’amélioration génétique du taro en Papouasie-Nouvelle-Guinée : résultats, défis et contraintes D. Singh, T. Okpul et D. Hunter Le programme d’amélioration génétique du taro de Papouasie-Nouvelle-Guinée a vu le jour au début des années 90. Malgré des ressources insuffisantes, il a produit de bons résultats, grâce notamment au soutien que lui a apporté récemment le projet TaroGen (« Ressources génétiques du taro : préservation et utilisation ») financé conjointement par la CPS et l’AusAid. Le programme a abouti, en 2002, à la production et à la distribution de trois variétés de taro à fort rendement, résistantes à la flétrissure des feuilles de taro (NT 01, NT 02 et NT 03) et à une large distribution aux cultivateurs et aux organismes de vulgarisation de végétaux destinés à la plantation. Les nouvelles lignées ne bénéficieront pas seulement aux cultivateurs papous, mais aussi à l’ensemble de leurs homologues océaniens, dès qu’une procédure de transfert de matériel offrant toutes les garanties de sécurité sera mise en place. Le programme d’amélioration de PapouasieNouvelle-Guinée vient d’entrer dans sa quatrième phase, au cours de laquelle il est prévu de produire de nouvelles lignées supérieures. Les principaux défis à relever dans les années à venir seront la multiplication et la distribution du matériel, la sécurité des transferts, l’enrichissement du fonds génétique des plantes cultivées, l’inclusion, dans les critères de sélection des végétaux, de la résistance aux coléoptères du taro et de caractéristiques propices à la valorisation post- récolte, la simplification du processus d’amélioration grâce aux marqueurs moléculaires et, surtout, la pérennité du programme d’amélioration. Parmi les principaux obstacles qui risquent d’entraver la réalisation ces objectifs, citons le manque de ressources, le manque d’interaction entre les organismes de recherche nationaux et les organismes de vulgarisation, et l’insuffisance des capacités scientifiques du pays. Introduced taro cultivars - on-farm evaluation in Samoa T. Iosefa, C.J. Delp, D.G. Hunter and P. Fonoti The outbreak of taro leaf blight disease devastated taro production in Samoa, as a result several exotic taro cultivars from Palau, Federated State of Micronesia and Philippines reported to have tolerance to TLB disease were introduced to meet the local production demand. Eight cultivars from these countries were multiplied and distributed to over thirty farmers of different locations throughout Samoa to assess their adaptability to different environments and farmers’ management and their adoption by the local communities. After two years of on-farms evaluation, we found out that Palau-10 is the most resistant to TLB disease, with highest yield; PSB-G2 is best in eating quality, other Palau lines and two of Micronesian lines (Pwetepwet and Toantal) are tolerant to TLB with acceptable eating qualities. Pastora from Micronesian is the only line rejected as of poor eating quality. Palau lines (P-3, P-4, P-10, P-20) together with Toantal and PSB-G2 are well liked and widely distributed by Samoan farmers. These materials are also highly recommended for the TLB breeding program in Samoa. The use of direct stolon development for mass propagation in taro (Colocasia esculenta (L.) Schott) Riki Faatonu, Philip Tuivavalagi, Winston Charles and Albert Peters Corms of four TLB tolerant clones of taro (Fili, P10, N16 and N21) were treated by dipping in solutions of 100 ppm and 500 ppm of gibberellic acid (GA). They were then planted in pots together with a control. The experimental design used was a simple randomized block replicated four times. Direct stolon development was achieved in the GA treated corms whereas multiple suckers were developed in the control. GA at the higher concentration prolonged stolon elongation and delayed sucker differentiation and development of the terminal bud in contrast to GA at the lower concentration. GA at the higher concentration produced significantly more stolons per corm than GA at the lower concentration. Similarly, corms treated with 500 ppm GA produced significantly more nodes on the stolon per corm than those treated with 100 ppm GA. Single node cuttings from the stolon of each corm were planted out in nursery beds under bi-wall drip irrigation and gave 100% sprouts and plantlets, which were ready for transplanting Cultivars de taro introduits : évaluation à la ferme au Samoa T. Iosefa, C.J. Delp,2 D.G. Hunter et P. Fonoti Une épidémie de flétrissure des feuilles de taro a dévasté la production de taro au Samoa. En conséquence, plusieurs cultivars exotiques de taro, provenant de Palau, des États fédérés de Micronésie et des Philippines, réputés tolérants à cette maladie, ont été introduits pour répondre à la demande des producteurs locaux. Huit cultivars issus de ces pays ont été multipliés et distribués à plus de trente agriculteurs de différents sites du Samoa, afin que soient étudiées leur capacité d’adaptation à différents milieux et aux modes de gestion des agriculteurs, ainsi que leur adoption par les populations locales. Après deux ans d’évaluation sur l’exploitation, nous avons constaté que Palau-10 est le cultivar le plus résistant à la maladie et qui donne le rendement le plus élevé ; PSB-G2 présente la meilleure qualité alimentaire, et d’autres lignées de Palau et deux lignées micronésiennes (Pwetepwet et Toantal) présentent une certaine tolérance à la maladie ainsi que des qualités gustatives acceptables. Pastora, de Micronésie, est la seule lignée rejetée en raison d’une médiocre qualité alimentaire. Les lignées de Palau (P-3, P-4, P-10, P-20), ainsi que Toantal et PSB-G2, sont appréciées et largement distribuées par les agriculteurs samoans. Ces végétaux sont vivement recommandés pour le programme de sélection réalisé au Samoa pour obtenir des variétés résistant à la flétrissure des feuilles de taro. Utilisation du développement direct de stolons dans la multiplication en masse du taro (Colocasia esculenta (L.) Schott) Riki Faatonu, Philip Tuivavalagi, Winston Charles et albert Peters Les cormes de quatre clones de taro tolérants à la flétrissure des feuilles de taro (P10, Fili, N16 et N21) ont été soumis à des doses de 100 et 500 ppm d’acide gibbérellique (AG). Chaque corme traité, ainsi qu’un témoin, a ensuite été planté en pot. Le dispositif expérimental consistait en un bloc simple randomisé reproduit quatre fois. Des stolons se sont développés directement sur les cormes traités à l’AG alors que le corme témoin a produit plusieurs drageons. En présence d’une forte concentration d’AG, la croissance des stolons est prolongée, et la différenciation du drageon et le développement du bourgeon terminal sont retardés, résultats qui ne se vérifient pas en présence d’une dose plus faible. Les cormes traités à 500 ppm d’AG ont produit une quantité beaucoup plus importante de stolons et de noeuds par corme que ceux traités à 100 ppm. Des segments à un seul nœud ont été plantés dans des plateaux de pépinière et arrosés par un système de goutte-à-goutte constitué d’une gaine à double paroi. Tous ont produit des pousses et des plantules prêtes third taro symposium 171 in eight weeks. GA at 500 ppm gave a significantly higher multiplication ratio (1:134) than GA at 100 ppm (1:91) for all the treated clones. The control produced suckers with a multiplication rate of 1:4. The protocol developed by this corm-to-stolon technique is currently employed for mass propagation of taro in support of the rapid expansion of taro cultivation in Samoa. This technique could be beneficial to taro improvement programs in other parts of the world where taro is cultivated on an extensive scale. pour le repiquage en huit semaines. Pour tous les clones traités, la dose de 500 ppm d’AG a permis d’obtenir un taux de multiplication beaucoup plus élevé (1:134) que la dose de 100 ppm (1:91). Le corme témoin a produit des drageons selon un taux de multiplication de 1:4. Le protocole mis au point pour cette expérience de production de stolons à partir de cormes est actuellement utilisé dans le cadre d’opérations de multiplication en masse de taros visant à développer rapidement la production à Samoa. La technique pourrait également être employée avantageusement dans le cadre de programmes d’amélioration du taro dans d’autres régions du monde qui pratiquent une culture extensive. Breeding Hawaiian taros for the future L’amélioration génétique du taro à Hawaii : pour un meilleur avenir John J. Cho A program was initiated in 1998 to improve commercial taros through breeding by increasing resistance to pests such as taro leaf blight and aphids, increasing plant vigor and yield, and. developing new and exciting varieties for the restaurant and landscape trade. In this program, Hawaiian taro cultivars are being used to incorporate different corm colors, low acridity, soft rot tolerance, early maturation, and brilliant colors. Hawaiian taros have been found to be closely related genetically, based on RAPD studies conducted in Hawaii, thereby limiting their usefulness for our breeding program. Therefore we have introduced taro varieties from where they originated, where we should expect greater genetic diversity. Introduced taro cultivars from Micronesia, Palau, Indonesia, Papua New Guinea, Thailand and Nepal are being used to increase resistance to taro leaf blight. Our approach is to incorporate two to three different sources of resistance into our improved taros to increase the durability of resistance. Tolerance to aphids is being incorporated into commercial taros using cultivars from Micronesia and Indonesia that reduce the longevity of aphids and/or reduce the number of offspring. Several new F1 hybrids and backcross F1 hybrids have been generated and evaluated for commercial potential in 2002, and a few are currently being advance tested in on-farm tests. The establishment of a commercial tissue culture laboratory in the Kingdom of Tonga Paul Karalus Pacific Biotech Limited, a 50/50 joint venture between a Tongan company (Tupulekina Technologies Ltd. owned by Tafolosa and Paul Karalus) and a New Zealand trust of the Keymer Family was established in 1999 to provide tissue cultured plants for both the local and export markets. The laboratory has 12 laminar flow cabinets and currently has the capacity to produce over two million cultures per year. From July 2000 the laboratory has been operating commercially and currently supplies taro and vanilla cultures to the Tonga market and calla lily cultures to the export market. The advantages of micro-propagation through tissue culturing are widely known and little needs to be said on 172 third taro symposium John J. Cho En 1998 a été mis en place un programme d’amélioration génétique des variétés de taro cultivées à des fins commerciales visant à augmenter leur résistance face à la flétrissure des feuilles de taro et aux pucerons, à accroître leur vigueur et leur rendement et à obtenir de nouvelles variétés plus attrayantes aux yeux des restaurateurs et des paysagistes. Dans le cadre de ce programme, des cultivars de taros hawaïens sont utilisés pour la diversité des couleurs de leur corme, le goût peu âcre de ces variétés, leur tolérance à la pourriture molle, leur maturation rapide et leurs couleur vive. Des analyses de marqueurs RAPD ont mis en lumière une grande proximité génétique entre les différentes variétés de taro présentes à Hawaï, limitant ainsi leur utilité dans le cadre de notre programme d’amélioration. Il a donc fallu recourir à des cultivars provenant des régions d’origine de chaque variété de taro, présentant logiquement une plus grande diversité génétique. Des cultivars de taro de Micronésie, de Palau, d’Indonésie, de Papouasie-NouvelleGuinée, de Thaïlande et du Népal sont actuellement utilisés pour accroître la résistance des taros hawaiiens à la flétrissure des feuilles de taro. La procédure consiste à introduire deux à trois sources de résistance différentes dans nos taros améliorés afin d’accroître la durée de résistance. Nous augmentons la tolérance aux pucerons des variétés de taros destinées à la vente grâce à des cultivars d’Indonésie et de Micronésie, qui réduisent la longévité et la progéniture des pucerons. Plusieurs hybrides F1 et hybrides F1 issus de rétrocroisements ont été produits. Leur potentiel commercial a été évalué en 2002, et certains font déjà l’objet de tests plus poussés en exploitations agricoles. Création d’un laboratoire de culture tissulaire à vocation commerciale au Royaume des Tonga Paul Karalus Pacific Biotech Ltd., une co-entreprise détenue à parts égales par deux sociétés, l’une tongane (Tpulekina Technologies Ltd., propriété de Tafolosa et Paul Karalus) et l’autre néozélandaise (appartenant à la famille Keymer), a été créée aux Tonga en 1999 pour approvisionner les marchés locaux et étrangers en végétaux obtenus par culture tissulaire. Le laboratoire est équipé de douze hottes à flux laminaire, et peut produire plus de deux millions de cultures par an. Le laboratoire de culture tissulaire a ouvert ses portes en juillet 2000, et produit actuellement des cultivars de taro et de vanille pour le marché tongan ainsi que des cultivars d’arum destinés à l’exportation. that. It is the actual mechanics of establishing such a facility that is the main thrust of this paper. The paper draws on actual experience and highlights the general technical and financial requirements for such a facility. A final consideration is the application of tissue culturing to the production of taro cultures and the nursing-on of these for the commercial production of taro. Throughout, an effort is made to provide technical practices that could assist others in avoiding what can be painful experiences in the developing of a laboratory. Les avantages de la micro-multiplication par culture tissulaire sont bien connus et il n’est pas besoin de s’y attarder. C’est la genèse de ce laboratoire qui est l’objet de cet exposé, qui décrit cette expérience ainsi que les conditions techniques et financières de la création de cet outil. Nous nous penchons enfin sur l’application de la culture tissulaire à la production de cultivars de taro et leur utilisation en vue de la production commerciale de taro. On s’efforce constamment de proposer des modalités techniques susceptibles d’aider des tiers à éviter les inconvénients liés à la mise en place d’un laboratoire. Taro breeding in India L’amélioration génétique du taro en Inde M.T. Sreekumari, K. Abraham, S. Edison and M. Unnikrishnan The genetic improvement of taro (Colocasia esculenta (L.) Schott) is one of the major research programmes of the Central Tuber Crops Research Institute, Trivandrum, which maintains 424 edible accessions. Field evaluation of 369 accessions resulted in the identification and release of two high yielding superior varieties from the CTCRI Headquarters at Trivandrum and a leaf blight tolerant type from its Regional Centre at Bhubaneswar. The varieties developed from germplasm selections are triploids, indicating the higher yielding potential of triploid taros than diploid taros. By intervarietal hybridization in diploid taros, a large number of hybrids were produced from different combinations. Of the 4280 segregants evaluated, 802 (18.7%) were rated as above average for important attributes like dwarf plant types, higher corm and cormel yield, non-acridity, long keeping quality and early maturity. From the initially screened segregants, rigorous selection for the above attributes resulted in the identification of seven superior hybrids, viz H-2, H-3, H-4, H-12, H-13, H120 and H-160. They consistently recorded higher tuber yields and superior quality attributes during the yield trials at the CTCRI. Currently they are being evaluated in farmers’ fields for verification of their merits, prior to variety release. Besides the above high yielders, novel types like dwarfs, erect types, profusely flowering lines and CLB (colocasia leaf blight) tolerant types were identified among the segregants. With the regular flowering of the clones and the possibility of producing full sibs, half sibs and selfs, taro breeding has been pursued at the CTCRI extensively and intensively, for developing hybrid varieties combining the various superior attributes. The latest approach in taro breeding at the CTCRI is to artificially produce triploids by crossing diploids with induced tetraploids to increase productivity. For this, tetraploid taros developed by colchicine treatment are being tested for flowering, fertility, and interploid (diploid x tetraploid) compatibility for the production of triploids. M.T. Sreekumari, K. Abraham, S. Edison et M. Unnikrishnan L’amélioration génétique du taro (Colocasia esculenta (L.) Schott) est l’un des principaux thèmes de recherche de l’Institut national de recherche sur les légumestubercules (CTCRI) qui possède une collection de 424 variétés comestibles de taro. L’évaluation au champ de 369 obtentions a permis d’identifier et de distribuer deux variétés supérieures à haut rendement, issues du siège du CTCRI, à Trivandrum, ainsi qu’un type résistant à la flétrissure des feuilles de taro provenant du centre régional de l’Institut à Bhubaneswar. Les variétés obtenues à partir de matériel génétique sélectionné sont des triploïdes qui présentent un rendement potentiel supérieur à celui des taros diploïdes. De nombreux hybrides ont été produits à partir de diverses combinaisons, par hybridation intervariétale. On a estimé que, sur les 4 280 ségrégants évalués, 802 (18,7 %) présentaient des attributs importants supérieurs à la moyenne (types végétaux nains, rendement des cormes et cormelles supérieur, absence d’âcreté, capacité de longue conservation et maturité précoce). Une sélection rigoureuse effectuée à partir des premiers ségrégants criblés en vue de l’obtention des attributs précités a permis d’identifier sept hybrides supérieurs : H-2, H-3, H-4, H-12, H-13, H120 et H-160. On a enregistré, pour tous ces hybrides, des rendements plus élevés et des attributs qualitatifs supérieurs au cours des essais de rendement réalisés au CTCRI. Les agriculteurs sont en train d’évaluer ces plants dans leurs champs, pour vérifier les avantages de ces variétés avant leur mise en circulation. Outre les types précités à rendement élevé, de nouveaux types – types nains, types dressés, lignées à floraison abondante et types résistant à la flétrissure des feuilles de Colocasia – ont été identifiés parmi les ségrégants. Au vu de la floraison régulière des clones et de la possibilité de produire des pleins-frères, des demi-frères et des franc-pieds, le CTCRI a poursuivi des travaux de sélection extensive et intensive du taro afin de mettre au point des variétés hybrides associant ces divers caractères supérieurs. Récemment, le CTCRI a cherché à produire artificiellement des triploïdes en croisant des diploïdes avec des tétraploïdes induits, afin d’augmenter la productivité. A cette fin, la floraison, la fertilité et la compatibilité interploïdale (diploïdes et tétraploïdes) en vue de la production de triploïdes ont été testées chez des taros tétraploïdes élaborés par traitement à la colchicine. third taro symposium 173 Theme Four Paper 4.1 Genetic diversity of taro (Colocasia esculenta (L.) Schott ) assessed by SSR markers J.L. Noyer1, C. Billot1, A. Weber1, P. Brottier2, J. Quero-Garcia3 and V. Lebot3 2 1 CIRAD, TA 40/03, 34398 Montpellier, France Genoscope, CNS, CP 5706, 91057 Evry, France 3 CIRAD, Port Vila, Vanuatu Introduction Taro, a vegetatively propagated root crop species, is grown in the humid tropical regions and is of considerable socio-economic importance in Southeast Asia and Oceania. Breeding programmes have been initiated with national collections sharing a narrow genetic base. Breeders are now attempting to broaden their working populations and morpho-agronomic characterisation has to be followed by molecular analyses in order to provide an accurate picture of the diversity within cultivars as well as in the wild genepool. The use of biochemical and molecular markers for taro germplasm characterization is quite recent but is expensive when thousands of accessions have to be analyzed. The first isozyme studies (Lebot and Aradhya, 1991) covering a wide geographical region included 1417 cultivars and wild forms from South East Asia and Oceania. They revealed the existence of two genepools, one in Southeast Asia and the second in Melanesia, indicating the possibility of two independent domestication processes. Only 48 cultivars from Indonesia were sampled but they appeared to be the most diverse, with 80 % of dissimilarity. Within the Pacific countries, Papua New Guinea and the Solomon Islands were the most diverse, followed by Vanuatu and Fiji. Polynesian countries showed the narrowest genetic base, with most cultivars corresponding to a single zymotype. Isozymes were however, unable to differentiate the tremendous morphological variation found within this region. It was also impossible to make correlations with ploidy levels or germplasm types (wild vs. cultivated). A second study including many more cultivars from Southeast Asia (Lebot et al., 2000), along with cultivars from Papua New Guinea and Vanuatu, confirmed the original hypothesis of two distinct genepools. The genetic base of the majority of the cultivars found within this vast geographical area was found to be rather narrow since only 21 out of 319 zymotypes represented two thirds of the total number of accessions. Molecular markers (RAPD) have been used more recently to analyze a subset of 44 accessions from diverse origins (Irwin et al., 1998) but no clear geographical or morphological structure was obtained. A combination of isozymes and RAPDs was also used to study Asian taros (Ochiai et al., 2001) and the differentiation of the studied regions (Nepal, Yunnan, Japan) was obvious although the relationships between the different populations were far from being evident. Interestingly, their data gave support to an autopolyploid origin of the triploids. More reliable dominant markers (AFLP), have been used to study the diversity of a core sample including 255 accessions from seven countries (Kreike et al., 2003). Most accessions could be clearly differentiated by using three primer pairs and few duplicates were found. A differentiation between Southeast Asian and Melanesian taros was obtained, confirming the isozyme results. Thirtyeight wild genotypes were analysed and only those from Thailand (16 acc.) showed a significantly higher genetic diversity as compared to the cultivars. For Indonesia and Malaysia, cultivated and wild genotypes were not clearly differentiated, indicating a possible feral origin of some wild genotypes. Triploids were not associated to diploids and their origin remains unknown. In fact, two clusters of triploids were identified, indicating the possibility of different polyploidisation processes. In Vanuatu, AFLPs were used on a core sample (40 acc.) aiming at validating a stratification approach for germplasm collections. No correspondence was found between the structure of the dendrogram produced and the major morpho-agronomic traits (Quero Garcia, 2000). Mace and Godwin (2002) have developed a microsatellite-enriched library following the hybridization method described by Edwards et al. (1996). These microsatellite markers were tested on a sample (17 acc.) from several Pacific countries. They proved to be a valuable tool for the identification of duplicates although the geographical structure produced was not very informative, probably due to the small size of the sample. Another microsatellite-enriched library was constructed (Bastide, 2000) following a hybridisation-based capture methodology using biotin-labelled microsatellite oligoprobes and streptavidin-coated magnetic beads (Billote et al., 1999). This second source of microsatellite markers was used in order to analyze a subset of the sample previously characterized by Kreike et al. (2003). The approach presented and discussed hereafter, might be of interest to breeders because it takes into consideration heterozygosity levels. 174 third taro symposium Material and methods Plant material Plant DNA was obtained from N. Kreike (Kreike et al., 2003). Each DNA sample was diluted to a final concentration of 5 ng/µl. A subsample (Table 1) was defined in order to cover both the widest genetic diversity and the largest geographical area. The amount of available DNA was the final criteria. Five additional accessions of Xanthosoma sagittifolium were added to the sample in order to get an external reference but no scorable products were obtained. Microsatellite analysis From the microsatellite-enriched library (Bastide, 2000), 96 clones were sequenced at the Genoscope, (Centre National de Séquençage) according to Artiguenave et al. (2000). Forty-nine sequences containing repeat motifs were identified. Fifteen primer pairs were designed using Primer3 (www-genome.wi.mit.edu/cgi-bin/primer/primer3_www. cgi). All of them revealed polymorphism on a subset of 5 taro accessions with a number of alleles ranging from 2 to 8. Seven primer pairs revealing more than 10 alleles each were finally retained (Table 2). IR-fluorescent PCR reactions were performed using the following strategy. One of the PCR primers had a 19 base extension at its 5’ end with the sequence 5’-CACGACGTTGTAAAACGAC-3’. This sequence is identical to an IR-labelled universal M13 Forward (29) primer, which is included in the reaction. During the PCR, the tailed primer generates a complementary sequence to the M13 primer which is subsequently utilized for priming in the amplification reaction thereby generating IR-labelled PCR products. All PCR were produced in 10 µl containing 20 ng of DNA, 1 µl 10x PCR buffer (200 mM Tris-HCl (pH 8.4), 500 mM KCl), 200 mM of each dATP, dCTP, dGTP, dTTP, 2 mM of MgCl2, 0.05 mM of each the M13-tailed primer, 0.1 µM of the other primer, 0.1 µM of the IR-labelled (with IR700 or IR800) M13 primer and 1U of Taq DNA Polymerase (Eurobio). Primers were synthesized by Eurogentec (France) and the IR-labelled M13 primer by Biolegio (The Netherlands). Cycling conditions consisted of an initial denaturing step of 4 min at 94ºC, followed by 24 cycles of “touchdown” PCR consisting of 30 s at 94ºC, 45 s at 64ºC (reduced by 0.5ºC each subsequent cycle), and 45 s at 72ºC, 10 additional cycles consisting of 30 s at 94ºC, 45 s at 50ºC and 45 s at 72ºC, and a final elongation step at 72ºC for 10 min. All PCR reactions were performed on a Dyad 384 MJResearch thermocycler. Gel electrophoresis and visualization of the STR alleles were accomplished using a LI-COR IR2 Model 4200 automated DNA sequencer (LI-COR, Inc., Lincoln, NE). Gels were 18 cm in length, 0.25 mm in thickness and composed of 6.5% KB+ acrylamide, 7M urea (LI-COR). Runs were performed in 1X TBE buffer, at 48°C and 40 W constant. A standard size ladder obtained from amplification of known band sizes was loaded regularly. The raw data depicting the STR alleles is displayed as an autoradiogram-like image on the computer and analyzed as it. Data analysis Presence or absence of one allele at one locus was scored respectively as 1 and 0. Due to the presence of triploids and diploids in the same analysis, all present fragments are not always detected and a simultaneous fragment absence becomes significant. Therefore, the absence and presence modalities must be considered of equal weight. Considering this, we chose to calculate the genetic similarities between accessions i and j (di-j) using the Sokal and Michener index, as: di-j = (n11 + n00)/( n11 + n10 + n01 + n00), where n11 is the number of shared alleles between i and j, n10 and n01 the number of alleles present for one accession and absent for the other, and n00 the number of simultaneously absent alleles. The matrix of pairwise di-j values between individuals was used to construct a NJTree with the unweighted NeighborJoining method (Saitou and Nei, 1987). This analysis was performed with the Darwin 4.2 software (Perrier et al., 1999). Heterozygocity levels were calculated according to the hypothesis that only one locus is revealed by each primer pair. Consequently, each individual presenting more than one band level is considered as heterozygous. third taro symposium 175 Table 1: Accessions used in this study (according to Kreike et al. 2003) 176 third taro symposium Table 2: Description of the 7 primer pairs used in this study Exp. Size: expected size, based on the sequenced allele All. Nbr: number of alleles per locus He: observed heterozygocity Results One hundred alleles were identified ranging from 12 to 17 per locus with an average of 14.3 (Table 2). No correlation can be observed between the nature or the length of the repeat motif and the number of alleles. Except for IND497, which presents 4 band levels at the locus 1C06, not more than 3 alleles were scored for the triploid accessions. Accessions IDN217, 218, 331 and 453 were supposed to be diploids but presented 3 alleles at different loci. No accessions revealed a triploid pattern for the locus 1B02. Table 3: Band level distribution per locus Nbr: occurence number of one band The heterozygosity level is ranging from 41.2 to 86.7 (Table 2) with an average of 68.7. Again, no relation can be observed between this heterozygosity level neither with the size nor with the structure of the repeat motif. If true allelic frequency cannot be calculated due to the presence of triploids in the sample, the band level distribution for each locus (Table 3) can be analyzed. The situation is extreme for locus 1C06 but no locus presents a normal distribution. Figure 1: Distribution of the similarity values (Sokal and Michener index) third taro symposium 177 The similarity values are very low (Figure 1) with an average of 0.156 and a maximum of 0.25. This is summarized by the distance scale on the NJTree (Figure 2) as well as by the frequent null bootstrap values frequently observed. This observation indicates that the global meaning and the stability of this NJTtree should be taken cautiously. Nevertheless, some clusters can be identified. Accessions from Thailand are grouped and well differentiated from other origins. Accessions from Papua New Guinea (BC) and more widely from Melanesia (BC, VU) are also clustering together. New lines and cultivars (GO, GS, PRG) from The Philippines are grouped. A fourth cluster can be observed which links wild types and all triploid accessions but except for three of them (IND517, VN50 and VN276). Triploids are associated to IDN217, 218, 331 and 453, the diploid cultivars which presented triploid patterns as mentioned above. The NJTree stands global comparison with the UPGMA dendrogram of Kreike et al. (2003) based on AFLP data. Minor modifications can be observed. In the AFLP dendrogram, VN50 and VN276, mentioned above, were grouped with other Vietnamese accessions from which they are isolated here. In the NJTree only one major Indonesian group is defined instead of three with AFLP data, meanwhile remaining accessions are spread in both cases. Identical conclusions were obtained with microsatellite data analyzed by a UPGMA dendrogram based on a Dice index (data not shown). Figure 2: NJTree representation of the genetic relationships of 105 acc. based on a similarity matrix involving 100 alleles (Sokal and Michener index, Bootstrap values >50 are indicated) 178 third taro symposium Discussion Despite some changes, like the clustering of new and old accessions from The Philippines, the the NJTree based on microsatellite loci and the UPGMA dendrogram based on AFLP data (Kreike et al., 2003) give consistent results. A differentiation between Southeast Asian and Melanesian taros is observed, confirming AFLP and isozyme results. Accessions from Thailand are grouped but Indonesian accessions show a large distribution confirming again AFLP results. This similarity between AFLP and microsatellite results was not fully expected. Indeed, AFLPs are scored as dominant markers with two allelic modalities at each locus. More alleles can be detected at a single locus with microsatellites (in fact, more than with other molecular markers presently used), and this results in an average index of similarity between individuals which is generally much lower. This may explain the low correlation between this technique and others, especially when individuals are not closely related. According to Powell et al. (1996), the correlation between AFLPs and SSR is not significant (r = 0.14). In our study, however, the similarity average value of 0.156 is ranging in the same order than the values obtained with the Nei and Li index (Dice index) by Kreike et al. (2003). Similarity of organization and of index values could be explained by the fact that the accessions of the whole sample are closely related whatever may be their origin. With an average of 14.3 alleles per locus, a very high level of polymorphism is observed. Considering that the number of alleles scored for each accession do not exceed their ploidy level (except for IND497 which presents 4 band levels at the locus 1C06 and for IND217, 218, 331 and 453 clustered with the triploids), we can assume: 1- that the microsatellite markers used in this study are locus specific and, 2- that no frequent events of locus duplication disrupt the analysis. Nevertheless, our purpose was to analyze simultaneously the genetic diversity of both diploids and triploids. The Nei distance index, the most widely used for such analysis, is based on allelic frequency which calculation is biased when a triploid exhibits two alleles. As this case occurs in our study and we thus preferred to avoid the calculation of genetic distances between and within populations or geographic groups. The analysis of the band levels distribution, which can be done without restriction, gives, however, an unexpected information. Even with a high average of 14.5 alleles/locus, we observe a few (1 to 4) very common band levels and many rare ones at each locus. Even the Indonesian sample that was considered as being the most diverse (Lebot and Arahdya, 1991; Kreike et al., 2003) follows the same distribution without covering all the band levels possibilities (data not shown). Added to the high level of heterozygosity, these observations let us assume that the sample could be considered as a population issued from a narrow genetic base which would not be in a panmictic situation. This last point is, of course, in agreement with the vegetative propagation of taro and this situation confirms that the sample is far from covering the whole diversity of C. esculenta. In the very next future, we will increase the number of accessions involved in this study to cover the complete TANSAO core sample (170 acc.). Increasing the number of wild types will be also a major objective in order to elucidate their relationships with cultivars. Genetic distances between and within diploids and triploids will have to be analyzed independently, then together, in order to evaluate the bias introduced by the unknown allelic distribution induced by the presence of two band levels at one locus for a triploid. The number of microsatellite loci will also be increased. Ten loci are generally admitted as being adequate to assess all the parameters of population genetics (De Vienne, 1988; Pritchard, Stephens and Donnelly, 2000; Cornuet et al., 1999; Koskinen, 2003). The following step will be to identify the natural or breeding populations within which these parameters will be studied. Their choice will be directed by the results obtained in the present study and according to the needs of the breeders. Acknowledgements This study would not have been possible without the support of the Taro Network for Southeast Asia and Oceania (TANSAO), a project funded by the INCO programme of the European Union (DG XII) (grant no. ERBIC18CT970205). References Artiguenave, F., Wincker, P., Brottier, P., Duprat, S., Jovelin, F., Scarpelli, C., Verdier, J., Vico, V., Weissenbach, J. and Saurin, W. 2000. Genomic exploration of the hemiascomycetous yeasts: 2. Data generation and processing. FEBS Letter 487(1):13–16. Bastide, C. 2000. Création de banques microsatellites pour l’igname (Dioscorea alata, Dioscorea praehensilis et Dioscorea abyssinica) et le taro (Colocasia esculenta). Mémoire de DUT, Dépt Génie Biologique Option Agronomie. IUT de Perpignan, France. 13 p. Billotte, N., Lagoda, P.J.L., Risterucci, A.M. and Baurens, F.C. 1999. Microsatellite-enriched libraries: Applied methodology for the development of SSR markers in tropical crops. Fruits 54:277–288. Cornuet, J.-M., Piry, S., Luikart, G., Estoup, A. and Solignac, M. 1999. New methods employing multilocus genotypes to select or exclude populations as origins of individuals. Genetics 153:1989–2000. de Vienne, D. 1998. (ed.) Les marqueurs moléculaires en génétique et biotechnologies végétales. INRA, Paris. 194 p. third taro symposium 179 Edwards, K.J., Barker, J.H.A., Daly, A., Jones, C. and Karp, A. 1996. Microsatellite libraries enriched for several microsatellite sequences in plants. BioTechniques 20:758–760. Irwin, S.V., Kaufuis, P., Banks, K., de la Peña, R. and Cho, J.J. 1998. Molecular characterization of taro (Colocasia esculenta) using RAPD markers. Euphytica 99(3):183–189. Koskinen, M.T. 2003. Individual assignment using microsatellite DNA reveals unambiguous breed identification in the domestic dog. Animal Genetics 34(4):297–301. Kreike, N., Van Eck, H. and Lebot, V. 2003. Genetic diversity in taro (Colocasia esculenta (L.) Schott) from South East Asia and Oceania. Theoretical and Applied Genetics, in press. Lebot, V. and Aradhya, M. 1991. Isozyme variation in taro (Colocasia esculenta) from Asia and Oceania. Euphytica 56:55–66. Lebot, V., Hartati, S., Hue, N.T., Viet, N.V., Nghia, N.H., Okpul, T., Pardales, J., Prana, M.S., Prana, T.K., Thongjiem, M., Kreike, C.M., Van Eck, H., Yap, T.C. and Ivancic, A. 2000. Genetic variation in taro (Colocasia esculenta) in South East Asia and Oceania. p. 524–533. In: Nakatani, M. and Komaki, K. (eds). 2000. Proceedings of the Twelfth Symposium of the International Society for Tropical Root Crops: Potential of root crops for food and industrial resources. Tsukuba, Japan, 10–16 September 2000,.ISTRC. Mace E.S. and Godwin, I. 2002. Development and characterization of polymorphic microsatellite markers in taro (Colocasia esculenta). Genome 45:823–832. Ochiai, T., Nguyen, V.X., Tahara, M. and Yoshino, H. 2001. Geographical differentiation of Asian taro, Colocasia esculenta (L.) Schott, detected by RAPD and isozyme analyses. Euphytica 122:219–234. Perrier, X., Flori, A. and Bonnot, F. 1999. Les méthodes d’analyse des données. p. 43–76. In: Hamon, P. et al. (eds). Diversité génétique des plantes tropicales cultivées. CIRAD, Montpellier, France. 387 p. Powell, W., Morgante, M., Andre, C., Hanafey, M., Vogel, J., Tingey, S. and Rafalski, A. 1996. The comparison of RFLP, RAPD, AFLP and SSR (microsatellite) markers for germplasm analysis. Molecular Breeding 2:225–238. Pritchard, J.K., Stephens, M. and Donnelly, P.J. 2000. Inference of population structure using multilocus genotype data. Genetics 155:945–959. Quero-Garcia, J. 2000. Étude de la structuration de la variabilité génétique du taro. INAPG, Paris. 31 p. Saitou, N. and Nei, M. 1987. The neighbour joining method: A new method for reconstructing phylogenetic trees. Molecular Biology and Evolution 4(4):406–425. 180 third taro symposium Theme Four Paper 4.2 Taro breeding programme of Papua New Guinea – achievements, challenges and constraints D. Singh1,3, T. Okpul2 and D. Hunter3 2 1 National Agricultural Research Institute, Lae, Papua New Guinea Agriculture Department, University of Vudal, Rabaul, Papua New Guinea 3 Secretariat of the Pacific Community Introduction In Papua New Guinea (PNG), taro (Colocasia esculenta) is second important root staple with an estimated annual production of about 436,000 t from an area of 77,000 ha (Sar et al., 1998). The crop is not only an important source of food and income, but also plays a vital role in the cultural heritage of Papua New Guineans, and is considered an essential component of many traditional ceremonies. Taro production in PNG has been continuously declining over past few years (Singh and Okpul, 2000). The declining trend in production can be mainly attributed to multiplicity of biotic and abiotic stresses. Of prime significance is taro leaf blight (TLB) disease caused by pathogen Phytophthora colocasiae, which first arrived in PNG in the 1940’s and is estimated to cause up to 50% loss in corm yield (Cox and Kasimani, 1988). Since then country opted and practiced various pronged cultural, biological and chemical control strategies, but without any noteworthy success. In 1980s, PNG commenced taro improvement programme with emphasis on TLB resistance breeding but this programme was also not successful due to lack of funds and staff changes. As an effect, the production trend continued to decline and important taro germplasm collected by farmers over years kept eroding. In some cases farmers abandoned the crop and replaced with other crops like sweet potato. Realizing the economic importance of crop, and implications of TLB on production and diversity, National Agriculture Research Institute (NARI) of PNG re-established its taro resistance breeding programme (then under the Department of Agriculture and Livestock) in 1993. The programme received a further support from the SPC/AusAID regional TaroGen project to develop taro lines with resistance to TLB, high yield and good eating quality. The present paper brings up to date achievements of NARI breeding programme, and also addresses the future challenges and constraints. Materials and methods Breeding strategy The breeding programme is based on the strategy of improving the population by adopting a modified recurrent selection approach, and has focused on incorporating horizontal resistance to TLB using a systematic cyclic strategy. Base population TLB resistance sources in the base population included a wild variety from Thailand (Bangkok), hybrids from a Solomon Islands breeding programme and three semi-wild taro varieties (Ph 15, Ph 17 and Ph 21) from PNG, and more than 50 agronomically important and popular varieties grown by local farmers. Methodological approach The entire methodological approach used by the breeding programme for the release of new varieties is outlined in Figure 1. The methodological details of each step are outlined in Singh et al. (2001) Selection criteria and trait assessments The programme is based on selection criteria, which emphasizes on identifying genotypes with moderate levels of TLB resistance, high yield and acceptable eating quality. A popular local variety (Numkowec) was included for comparison in all trials. In preliminary trials, yield was measured as corm weight (grams), but in advanced trials it was estimated as tones/hectares (t/ha). Yield stability was estimated on the basis of the Eberhart and Russell (1966) model. Severity of TLB was recorded using standard area diagram developed by Gollifer and Brown (1974). The percentage diseased leaf area (DLA) was estimated for each leaf and the mean disease rating for each plant was calculated by dividing the total of the assessments for each leaf by the total number of leaves examined as outlined by Hunter and Pouono (1998). Only genotypes with moderate resistance were further selected and genotypes with hypersensitive reactions (immune to disease or showing no signs of infection at all) or very high DLA were discarded. Eating quality was assessed in terms of eating quality score (EQS). For eating quality, five parameters (texture, acridity, fibre, aroma and colour) were evaluated. The total score (TS) was calculated for each parameter. After calculating the TS, a final score (FS) was calculated for each parameter by multiplying the TS of that parameter with its constant parameter weighting (CPW: 0.50 for texture, 0.30 for acridity, 0.10 for fibre and 0.05 each for aroma and colour). A final EQS was third taro symposium 181 then calculated by summing FS of all five parameters. The higher the value of the EQS, the better the eating quality was assumed to be. Figure 1: Schematic representation of evaluation, development, and release process for promising taro varieties Results NARI taro breeding programme has successively advanced to its fourth cycle. The progressive achievements of each of the four cycles is presented below: Cycle 1 Cycle 1 was developed in 1994 by crossing the resistant base population with superior local taro varieties. The detailed evaluation results of Cycle 1 superior lines are presented in Okpul et al. (1996). No recommendations were made from Cycle 1 since most lines retained undesirable wild characteristics, although some recombinants were partially superior. Cycle 2 Cycle 2 was created in 1996 by inter-crossing partially superior genotypes from Cycle 1. In 1998, 32 lines were selected from Cycle 2 on the basis of preliminary assessments on yield, TLB resistance and eating quality. Further testing enabled selection of 12 TLB resistant lines for inclusion in replicated advanced yield trials. The replicated trials identified seven lines (C2-E1, C2-E3, C2-E4, C2-E7, C2-E8, C2-E10 and C2-E11). These seven lines were advanced to genotype x environment (G x E) trials for assessing their adaptability under major agro-ecological sites of PNG. The details of these results are presented in Okpul et al. (2002). 182 third taro symposium The G x E trial results were reviewed by a panel of scientists and were presented to the national Taro Improvement Coordinating Committee (TICC) with a recommendation to release three lines, C2-E3 and C2-E4 and C2-E8. These lines were selected on the basis of high yield, yield stability over a range of environments, resistance to TLB and good eating quality. Full description of the three lines is presented in Table 1. A final release document paper was prepared and presented to NARI management. The release of lines was endorsed by NARI, and the lines were officially named (NT 01, NT 02 and NT 03) and released on 13 December 2001. This is a first report of any taro varieties released in PNG. Cycle 3 Cycle 3 was created by inter-crossing 21 selected Cycle 2 lines in a half-diallel design. More than 300 crosses were attempted and a population of over 10,000 seedlings was created and evaluated. Forty-nine superior lines in terms of TLB resistance, high yield and good eating quality were recovered in preliminary trials. Twenty-six lines were selected from intermediate trials and finally six lines were selected from advance trials. The selected six lines are being evaluated in G x E trials for adaptability. Any superior lines identified from these trials will be recommended for new release(s). Table 1: Descriptions of PNG released taro varieties and a popular standard cultivar Numkowec Trait Variety NT 01 (C2-E3) NT 02 (C2-E4) NT 03 (C2-E8) Numkowec (control) Yield (t/ha) 10.49 7.68 7.65 5.89 Average corm weight (g) 525 g 380 g 380 g 300 g Yield stability Taro leaf blight (TLB) TLB diseased leaf area (%) Stable Stable Unstable Stable Resistant Resistant Resistant Susceptible 8.24 7.34 7.19 15.76 Susceptible Susceptible Susceptible Susceptible Taro beetle damage (%) 19.70 19.04 18.74 19.53 Eating Quality Good Good Good Good Eating Quality Score 2.64 2.59 2.54 2.52 6 6 6 6 3-4 2-3 5-6 6-8 Taro beetle Time to maturity (months) Sucker production Growth habit Erect Erect Erect Erect Plant height Tall Medium Tall Medium Leaf lamina Light green Dark green Dark green Dark green Petiole colour Light green Purple green Purple Light green Purple Purple Purple Purple Rare Rare Frequent Frequent Cylindrical Elliptical Conical Conical Smooth Smooth Smooth Smooth Pink Pink Pink Pink 35 41 41 38 Petiole junction Flowering Corm shape Corm skin Flesh colour Corm dry matter content (%) Cycle 4 In advancing to fourth cycle of recurrent breeding, more than 200 selective partial diallel crosses were performed among 49 genotypes selected from Cycle 3 in 2001. Approximately 8,000 seedlings were generated and 237 superior genotypes were selected on basis of TLB resistance, high yield and eating quality. The selection criteria in this cycle were extended to wider adaptability during preliminary screening to widen the genetic base of population for better adaptation of the identified genotypes. Therefore, the selected 237 lines were tested under three varied agro-ecological sites to identify wider adaptable lines in addition to high yield, TLB resistance and good eating quality. On this basis, 22 elite lines were identified from Cycle 4 for undertaking G x E trials and simultaneously for creating Cycle 5. Compared to previous cycles, the genetic gains were considerably higher for Cycle 4 in terms of recovering superior progeny and traits of interest. Discussion PNG is one of the few countries in the Pacific region, where taro varieties have been successfully bred for superior yield, TLB resistance and good eating quality. The released varieties will help growers wishing to improve taro production for subsistence purposes in order to maintain a traditional food staple with cultural significance. There will also be benefits to farmers who previously grew taro for domestic markets, but in recent years have had to abandon their production because of TLB. The superior traits associated with these lines should enable successful acceptance and adoption by smallholders, subsistence and semi-commercial growers. The yield of these lines should be sustainable over time, since these varieties are derived from genetic improvement and are widely adaptable. The TLB resistance will be durable since it is based on horizontal resistance relying on additive effects of multiple genes against the pathogen. third taro symposium 183 The released lines are being multiplied at four different sites covering different regions of PNG. The material is being distributed to farmers nationally in collaboration with TICC, extension workers, non-government organizations and church groups. Till now, more than 5,000 planting suckers of each line have been distributed to farmers and communities through out PNG. In future, more lines are expected to be released from post Cycle 2 recurrent cycles. It is likely that those lines released from advanced cycles will be more superior in their attributes, especially eating quality because of the polygenic breeding approach (which relies on accumulation of superior genes from cycle to cycle) adopted by PNG programme. PNG breeding programmes could be used as a vehicle for breeding network to control TLB disease in the Pacific region. The material has been transferred to the Regional Germplasm Centre (RGC) Fiji for distribution to other Pacific island countries in the region, once safe quarantine movement is guaranteed. Taro varieties identified in PNG although successful locally, may not suit individual countries with a wide diversity of environments and cultures, and therefore regional G x E evaluations will be a challenge. Additional challenges are developing innovative ways to increase the rate of multiplication of planting material, safe germplasm exchange, capacity to meet the needs of larger number and isolated farmers for enhanced distribution of new and improved germplasm. The sustainability of breeding programme is a major challenge and participatory plant breeding can offer a scope to address this challenge as pointed earlier by Hunter et al. (2001). Future genetic gains of taro breeding will challengingly rely on widening the genetic base of the crop. Recent biochemical and molecular studies on taro germplasm from Asia and the Pacific has identified the existence of two distinct regional taro genepools representing these two regions. The level of genetic diversity of PNG germplasm compared to Asian germplasm is very low. Future taro improvement programmes will require introgression of selected exotic and local germplasm into PNG breeding programme to create genetically diverse progeny and further broaden the genetic base of the crop. Once the genetic base is broadened, focus should also be given to selection of post harvest traits like shelf life, and marketable traits demanded by the export markets. Although difficult, breeding for tolerance to taro beetle or selecting for factors reducing taro beetle attack will be a challenge. In addition, there is a considerable potential of molecular markers to simplify selections and breeding programme. The foremost constraints to prevail over various challenges are funding, liaison between the national research and extension agencies, and scientific capacity once donor project technical advice is terminated. It is however stressed that whatever the constraints are, the country has to take initiative to sustain its breeding programme, if it wants to overcome the declining production trend and explore development potential of this major crop. Acknowledgement The support of funding agencies AusAID, EU and ACIAR, and implementing agencies NARI and SPC is gratefully acknowledged for backing PNG breeding programme over a stretch of time. References Cox, P.G. and Kasimani, C. 1988. Control of taro leaf blight using metalaxyl. Tropical Pest Management 34:81–84. Eberhart, S.A. and Russell, W.A. 1966. Stability parameters for comparing varieties. Crop Science 6:36–40. Gollifer, D.E. and Brown, J.E. 1974. Phytophthora leaf blight of Colocasia esculenta. Papua New Guinea Agricultural Journal 25:6–11. Hunter, D.G., Iosefa, T., Delp, C.J. and Fonoti, P. 2001. Beyond taro leaf blight: A participatory approach for plant breeding and selection for taro improvement in Samoa. p. 219–227. In: Proceedings of the International Symposium on Participatory Plant Breeding and Participatory Plant Genetic Resource Enhancement, Pokhara, Nepal, 1–5 May 2000. CGIAR/PRGA, Cali, Colombia. Hunter, D.G. and Pouono, K. 1998. Evaluation of exotic taro cultivars for resistance to taro leaf blight, yield and quality in Samoa. Journal of South Pacific Agriculture 5:39–43. Okpul, T., Ivancic, A. and Simin, A. 1996. Evaluation of leaf blight resistant taro (Colocasia esculenta) varieties at Bubia, Morobe Province, Papua New Guinea. Papua New Guinea Journal of Agriculture, Forestry and Fisheries 40:13–18. Okpul, T., Singh, D., Wagih, M., Wiles, G. and Hunter, D. 2002. Improved taro varieties with resistance to taro leaf blight for Papua New Guinea farmers. National Agricultural Research Institute, Lae, Papua New Guinea. Sar, S.A., Wayi, B.M. and Ghodake, R.D. 1998. Review of research in Papua New Guinea for sustainable production of taro (Colocasia esculenta). Tropical Agriculture (Trinidad) 75:134–138. Singh, D. and Okpul, T. 2000. Evaluation of 12 taro (Colocasia esculenta (L.) Schott) leaf blight resistant clones for yield and eating quality in Papua New Guinea. SABRAO Journal of Breeding and Genetics 32:39–45. Singh, D., Hunter, D., Iosefa, T. and Okpul, T. 2001. Guidelines for undertaking on-farm taro breeding trials in the South Pacific. Secretariat of the Pacific Community, Suva, Fiji. 184 third taro symposium Theme Four Paper 4.3 Introduced taro cultivars – on-farm evaluation in Samoa T. Iosefa2, C.J. Delp2, D.G. Hunter1 and P. Fonoti3 Team Leader, AusAID/SPC TaroGen Project, Secretariat of the Pacific Community School of Agriculture, University of the South Pacific, Alafua Campus, Apia, Samoa 3 Ministry of Agriculture, Fisheries, Forests and Meteorology, Apia, Samoa 1 2 Introduction Taro leaf blight (TLB), a disease caused by Phytophthora colocasiae, has been present in the Pacific region since the early 1900s. It is a disease highly adapted to the wet humid environment of the region and is a major constraint for taro production in those countries where present. The most recent introduction of the disease was to the Samoan islands in 1993. Following the outbreak of taro leaf blight (TLB) disease in mid 1993, which severely affected all local taro cultivars in Samoa, the Department of Agriculture authorized importation, through tissue culture, of new cultivars which were reported to have tolerance to the disease. Among these exotic cultivars were PSB-G2 (now called Taro Fili) from the Philippines, Toantal, Pwetpwet and Pastora from the Federated State of Micronesia [FSM], and several Palau cultivars including: P-3 (Ongdibel), P-4 (Homestead), P-6 (Kerdeu), P-7 (Ochelochel), P-10 (Ngeruuch), P-16 (Meltalt), and P-20 (Dirratengadik). In June 1996, the Department of Agriculture started releasing some of the exotic cultivars to the farming community. PSB-G2 was most widely distributed to hundreds of farmers, and the FSM cultivars in fewer numbers were given to some farmers. One of the problems that contributed to the devastation of taro in Samoa in 1993 was the relative uniformity (lack of cultivar diversity) of the crop. Increasing cultivar diversity on farmers’ fields was identified by researchers at University of the South Pacific (USP) as an important future disease management strategy. Researchers were concerned that lessons had not been learned and that production might revert to the pre-1993 situation if only one or two improved cultivars were widely distributed and promoted. Discussions between researchers and farmers also revealed that some of the released cultivars had a few shortcomings including susceptibility to the disease in wetter parts of the country, low yields and poor storability. Farmers also raised concern about the length of time it was taking to get access to resistant germplasm evaluated through formal screening programmes. Researchers at USP were also concerned with the slow rate at which resistant taro was released through formal taro screening programmes and the rigorous testing over several years trying to identify a few cultivars that might be of limited relevance to farmers. There is evidence from elsewhere that much of the germplasm officially released through conventional plant breeding programs was of limited relevance to farmers, and much of the material that is rejected has been found to have subsequent acceptance among farmers (Maurya et al., 1988). The farmer participatory approach to plant breeding, adopted in this study, involving researchers, farmers, and extension staff, was considered as a means to achieve a number of objectives including: • learn more about what farmers want from improved taro cultivars by involving them in the technology development process; • use many farmers under diverse environments, providing them with a range of options so that they can select the best for their conditions. This also ensures that farmers gain quicker access to resistant taro; • increase the diversity of taro cultivars grown by farmers to minimize the risk of a repeat disease outbreak; • strengthen the linkages between researchers, extension staff, and farmers; and • make more effective use of limited time and resources of researchers and extension staff in Samoa. On-farm evaluation program with farmer participation Taro Improvement Project TIP, a farmer focus group, was initiated at USP in early 1999. The aim was to bring together taro farmers and provide them with more options for improving production and managing TLB. TIP represents a partnership between USP, Ministry of Agriculture Forestry and Fisheries (MAFFM) research and extension staff, and farmers from the islands of Upolu and Savai’i. Membership was open to all farmers who agreed to compare taro cultivars using the PPB approach and take part in focused group discussions (FGDs) on their performance and other issues. Efforts were made to ensure good geographical coverage of the islands when initially selecting farmers. Men tended to dominate the group, which is a reflection of the gender balance in taro cultivation in the country. Farmers completed an information form on attendance at their first TIP meeting that provided researchers with information on farming systems, farmers’ profiles and needs. third taro symposium 185 Participatory Rural Appraisals (PRAs): Crop-focused PRAs were conducted with farmer groups to learn more about taro production problems, perceptions of taro cultivars and criteria important in the selection of a cultivar. PRA techniques included FGDs, farm visits and observation, key informants, informal interviews and scoring and ranking exercises. The PRAs were conducted by a facilitator based at USP. The PRAs highlighted problems related to the cultivars that were available to farmers at the start of the TIP project (Table 1). Table 1: Key problems in taro production perceived by taro farmers and extension officers identified through PRAs in Samoa Rank Problem identified 1 Poor quality of taro sold in local market 2 Control of taro leaf blight 3 Inappropriate use of fungicides to control taro leaf blight 4 Short shelf-life of PSB-G2 means it is not suitable for export 5 Shortage of improved taro planting materials 6 Palatability of some Palau cultivars is not acceptable for export 7 Difficulty in identification of different Palau cultivars 8 Expense involved in maintenance of taro plantations 9 Cultivar PSB-G2 is low yielding For example, PSB-G2 was identified as low yielding and having a short shelf-life making it unsuitable for export. Because Palau cultivars had been imported illegally from nearby American Samoa farmers experienced difficulty with identification and were unsure about what was being supplied to the market. This contributed to problems with consumer reaction to quality. Some Palau cultivars had poor palatability but were still finding their way to the market. The PRAs also revealed that palatability, TLB-resistance and high yield were the most important criteria for farmers when selecting new taro cultivars (Table 2). Table 2: Ranking of important criteria for selection of new taro cultivars perceived by Samoan taro farmers Rank Criteria identified 1 Good palatability 2 Resistance to TLB 3 High yielding 4 Tender leaves for luau (traditional vegetable dish) 5 Long shelf-life 6 Vigorous growth Farmer-managed trials: The planned programme of evaluations was described at TIP meetings. Researchers provided farmers with taro cultivars with a range of characteristics and TLB resistance. It was up to each farmer to identify those cultivars that he or she preferred and that were most suitable for their environment. Farmers had the opportunity to visit a demonstration site at USP to observe the cultivars close to harvest. The data accumulated on each cultivar was discussed with farmers. The evaluation process was described and appropriate, simple on-farm trial layouts discussed. Farmers were given up to eight cultivars with 10 planting suckers per cultivar. Trial design was a simple non-replicated layout using single rows of each cultivar with farmer traditional spacing. The importance of labelling, plot maintenance and a layout plan were stressed, with no use of fungicides. On-going management of trial plots was based on normal farmers’ practices and the responsibility of farmers. Farmers were advised to establish plots in an area where taro was already growing to ensure exposure of the cultivars under test to TLB. PSB-G2 was included as the reference cultivar, as it was regarded as the best available cultivar at the time. The first farmer-managed trials were planted in July 1999. Evaluation of the trials: Monthly TIP meetings held at USP were the main forums for FGDs and other PRA exercises, although a few farmer-to-farmer visits were organized to allow participating farmers to observe other trials for comparison. Simple data sheets for vigour and disease score were explained and distributed to each farmer. The importance of data collecting was highlighted as well as the requirement to feedback information to monthly FGDs as a learning experience for the group. In FGDs, criteria such as vigour, yield, TLB resistance, suckering ability and palatability were scored using a ranking system based on 1 to 4 (unacceptable to outstanding). Farmers were also requested to notify researchers as cultivars matured so accurate yield data could be collected. All corms and planting material remained the property of farmers. All household members were encouraged to prepare and cook taro corms at home and provide information on quality. Farmers were also requested to bring corms of cultivars to monthly FGDs for assessment of quality in blind taste tests. This allowed accurate evaluations to be carried out on the effect of location and date of maturity on corm quality. A summary of the results of the preliminary trials using this approach to evaluate introduced exotic cultivars is presented in Table 3. PSB-G2 was ranked highest by farmers for palatability scoring as high as Niue would have in the pre-TLB period. Farmers also ranked Toantal and P20 high for palatability followed by Palau 10. However, PSB-G2 and Toantal ranked relatively poorly for the next two important criteria, yield and resistance to TLB, whereas Palau 10 scored high for these two criteria. Palau 20 also scored well for yield. Palau 10 scored highest for overall plant vigour followed by Pastora, Palau 20 and Palau 7. PSB-G2 ranked eighth for this criterion, while, Pastora scored very poorly for palatability. 186 third taro symposium Researchers made irregular visits to farmer-managed trial sites to collect data on the criteria outlined in Table 4. The data are based on 2 to 3 visits to each of 30 farmer-managed trials and blind taste evaluations carried out at Alafua Campus, USP. This allowed some comparison between farmer and researcher evaluation. It is interesting to note that there was general agreement between the rankings of the top cultivars in terms of palatability. PSB-G2, Toantal, Palau 20 and Palau 10 were the top scoring cultivars in both evaluations. Farmers ranked Palau 10 highest for overall vigour, yield and resistance to TLB, which corresponds to the data collected by researchers in Samoa and elsewhere. Both sets of data also demonstrate for criteria other than palatability, PSB-G2 ranks poorly. Table 3: Summary of farmers ranking of exotic taro cultivars introduced to Samoa Cultivar 1 2 Vigour Yield TLB resistance Sucker production Palatability PSB-G2 3.11 2.4 2.0 3.4 4.0 Pastora 3.8 3.3 2.9 3.2 1.6 Pwetepwet 3.4 2.9 2.7 3.8 2.2 Toantal 3.3 2.3 1.7 2.7 3.5 Palau 3 3.3 3.0 2.6 3.1 2.9 Palau 4 3.1 2.1 2.6 3.9 3.1 Palau 7 3.5 3.0 2.8 2.8 2.4 Palau 10 3.9 3.8 3.5 3.2 3.2 Palau 20 3.7 3.5 2.6 2.9 3.6 Niue (post-1993)2 1.9 2.0 1.1 1.9 1.9 Niue (pre-1993)2 3.9 3.9 - 3.1 4.0 Ranking for all criteria are based on 1 = unacceptable; 2 = okay, but not good; 3 = good; 4 = outstanding. Farmers were asked to rank Niue, the preferred cultivar of Samoans, for the criteria highlighted before and after the arrival of TLB in the country. Table 4: Summary of data collected by researchers from farmer-managed trials and palatability evaluations carried out at Alafua Campus, USP Vigour1 Yield2 TLB Severity3 Sucker production4 Palatability5 PSB-G2 4.2 0.6 9.7 4.0 3.1 Pastora 4.4 0.7 6.0 3.0 1.6 Pwetepwet 4.9 1.0 5.3 3.0 2.2 Toantal 4.5 0.7 9.0 3.0 2.8 Palau 3 5.2 1.1 3.6 4.0 2.3 Palau 4 4.8 0.6 5.8 4.0 2.3 Palau 7 4.3 - 5.0 - 2.2 Palau 10 5.2 1.0 3.4 4.0 2.3 Palau 20 4.2 1.0 6.0 3.0 2.6 Cultivar Average number of leaves per plant. Average corm weight (kg) per plant. Average percentage TLB per leaf. 4 Average number of suckers per plant. 5 Average of 20 blind taste tests made by teams of 10 to 15 tasters. Ratings are 1=unacceptable, 2=okay, 3=good and 4=outstanding 1 2 3 Conclusions Farmer participation in this on-farm program was a useful method to combine the knowledge and experiences of researchers, extension agents and farmers. This gives credibility to these conclusions. Results from different agro-climatic zones and management conditions show that all of the exotic cultivars tested adapt well to the environments of Samoa. The Palau lines P-3 and P-10 are the most TLB resistance of all, followed by FSM and Philippine cultivars. All have at least 4 to 5 leaves as they approach maturity under TLB conditions where local cultivars, especially Niue, lost most of their leaves and produced poor yields and poor eating quality even with the use of intensive fungicide applications. In terms of yield, Palau 3, P-10 and P-20 are the most vigorous in growth and usually produce big corms. PSB-G2 and Palau-4 are the lowest yielding taro. For eating quality, PSB-G2 is the most preferred and farmers always rate it excellent. Toantal is also close to excellent. The only taro rejected by the majority of farmers is Pastora. Although Pastora grows and yields well with little TLB, it has a wet texture and is rated poor for eating quality. Referring to the criteria pinpointed by farmers for improved taro, none of the cultivars evaluated contain all of the desired attributes. Growers are now back in taro production, but they are looking forward to improved characteristics demonstrated in the seedling clones coming from the ongoing TIP breeding program. third taro symposium 187 Recommendations 1. PSB-G2, the most preferred taro, is highly recommended for production throughout Samoa. 2. Palau 3, P-4, P-10 and P-20 are the most highly recommended of all 20 Palau lines. Palau 10 yields the highest and is the most resistant to TLB. 3. Toantal is the only cultivar from the FSM recommended for its good eating quality (taste). 4. All the above cultivars should be distributed widely to the farming communities. 5. All exotic taro cultivars adopted by the farming communities are recommended as potential parents for breeding programs for higher levels of resistance, yields and excellent eating quality. 6. The breeding programs with crosses of exotic taros and local cultivars should be continued to preserve the diverse germplasm, add to the horizontal resistance to TLB and to search for improved characteristics. Acknowledgements This Project is a joint research cooperation among the Taro Genetic Resources: Conservation and Utilization Project of the AusAid/SPC Program, University of the South Pacific, School of Agriculture, Alafua Campus and the Ministry of Agriculture - MAFFM Research and Extension Services of Samoa. I would also like to acknowledge the contribution of all the farmers who were involved in the evaluation of exotic taros. References Maurya, D.M., Bottral, A. and Farrington, J. 1988. Improved livelihoods, genetic diversity and farmer participation: A strategy for rice breeding in rain fed areas of India. Experimental Agriculture 24:311–320. 188 third taro symposium Theme Four Paper 4.4 The use of direct stolon development for mass propagation in taro (Colocasia esculenta (L.) Schott) Riki Faatonu1, Philip Tuivavalagi1, Winston Charles2 and Albert Peters1 1 Crops Division, Nu’u Research Station, Ministry of Agriculture, Forests, Fisheries and Meteorology, Samoa 2 Office of the FAO Sub-Regional Representative for the Pacific, Apia, Samoa Introduction Taro (Colocasia esculenta) was the major export commodity in Samoa prior to 1993. From 1994 export declined dramatically due to a severe outbreak of taro leaf blight (Phytophthora colocasiae Racib.). This epidemic caused a significant drop in production and devastated the country’s economy. All local cultivars were severely affected by taro leaf blight (TLB) and the Department of Agriculture authorized the importation, through tissue culture, of new cultivars that were reported to have a high degree of tolerance to the disease. These accessions were evaluated for high yield potential, disease reaction and good cooking quality. Accessions having very promising characteristics were then used to initiate a breeding program. The program was aimed at the recombination of all desirable characteristics so as to recover types combining high yield, a high degree of tolerance to TLB, good eating quality and tender leaves for making “pulusami”. In June 2000 the Government released two promising clones: N16 and N21 from the breeding program along with two accessions: P10 and Fili for multiplication increase of planting material and distribution to taro growers in an effort to revive the industry to the level it was prior to 1993. The multiplication rate of taro in tissue culture is low. At the Laboratory at Nu’u the average ratio is in the region of 1:3 (Palupe, pers. comm.) The mass propagation technique using “tops” under Bi-wall Drip Irrigation (BDI) gives a multiplication ratio of 1:3. with all clones. Experiments using single node cuttings from stolons derived from plants grown under BDI gave an average multiplication ratio of 1:24. with all treated clones. This paper reports the results of a new rapid multiplication technique. It is capable of generating large quantities of planting material of taro in a short time from stolons developed directly from corms without the intervention of sucker or plant development. Materials and method Corms of four TLB tolerant taro clones: Fili (V1), P10 (V2), N16 (V3) and N21(V4) were used in this study. They were dipped in solutions of gibberellic acid (GA) at concentrations of 500 ppm (T1) and 100 ppm (T2) for 10 minutes. They were then planted singly along with a control (T0) in soil in black plastic pots spaced 0.5m x 0.5m. Single node cuttings derived from stolons of each clone were planted out in a nursery bed. The soil was kept to field capacity using Bi-wall drip irrigation (BDI). The experimental design used was a simple randomized block replicated four times with two corms per plot. Results The results on mean number of stolons produced are presented in Table 1. The experimental results showed that the GA treatments produced stolons directly from corms (slide 1) and the untreated corms produced multiple suckers (slide 2). It was observed also that GA at the higher concentration (500 ppm) prolonged stolon elongation and slowed down the development of the terminal bud into a sucker. Apical sucker development took place earlier at the lower GA concentration of 100 ppm. third taro symposium 189 Table 1: Mean number of stolons per plant per variety per treatment in four replications REP I Variety REP II REP III REP IV T1 T2 T1 T2 T1 T2 T1 V1 7 14 8 14 8 15 7 T2 15 V2 7 16 9 19 9 17 8 20 V3 8 9 6 10 6 12 7 7 V4 10 9 9 15 9 15 9 11 Total 32 48 37 58 32 59 31 53 Varieties V2T2 V1T2 V4T2 V4T1 V3T2 V2T1 V1T1 V3T1 Means 18.0 14.5 12.5 9.8 9.5 8.5 7.5 7.3 Means joined by the same line are non-significant at P = 0.05 The results of the analysis of variance on stolons produced (Table 1) showed that the interaction variety x treatment was highly significant. All clones treated with GA at 500 ppm gave significantly more stolons than those treated with 100 ppm of GA except for N16. Corms of P10 when treated with 500 ppm GA produced significantly the highest number of stolons than all the other clones with the same treatment. Corms of Fili treated with 500 ppm produced significantly more stolons than corms of N21 with the same treatment. Similarly at 500 ppm N21 gave significantly more stolons than N16. In general, 500 ppm of GA produced significantly more stolons in all varieties than corms treated with 100 ppm GA. At 100 ppm GA there was no significant difference between stolon production in varieties N21 and P10. At 100 ppm GA stolon production in N21 was significantly higher than in Fili and N16. There were no significant differences in stolon production between Fili and N16 when corms were treated with 100 ppmGA. The results on the mean number of suckers produce per corm in the control are presented in Table 2. The analysis of variance on the data in Table 2 showed that the differences between the mean number of suckers produced for variety in the control were non-significant. The multiplication ratio was approximately 1:4. The mean number of single node cuttings, obtained for each clone are presented in Table 3. These cuttings gave 100% sprouts and plantlets. Table 2: Mean number of suckers produced per variety for the control Variety REP I REP II REP III REP IV TOTAL MEAN V1 3 4 3 3 13 3.3 V2 5 5 4 4 18 4.5 V3 4 3 2 3 12 3.0 V4 3 4 5 3 14 3.5 Total 15 15 14 13 57 3.57 Table 3: Number of single node cuttings produced per plant per variety per treatment in four replications REP I Variety REP II REP III REP IV T1 T2 T1 T2 T1 T2 T1 T2 V1 84 115 98 156 111 130 104 135 V2 71 128 41 163 66 145 86 117 V3 71 100 106 117 67 122 73 103 V4 118 144 103 161 103 139 147 176 Total 344 487 348 597 347 536 410 531 Varieties Means V4 V1 V2 V3 136.38 116.63 102.13 94.88 Treatment Mean T2 T1 134.44 90.56 Means joined by the same line are non-significant at P = 0.05 The results of the analysis of variance on mean number of single node cuttings showed that the F values for replication and the interaction VxT were non-significant at P0.05., indicating that variety and treatment response are independent of each other. The F values for Variety and Treatment were highly significant indicating that significant differences existed between Variety and Treatment means. The Duncan’s Multiple range test showed that variety V4 produced significantly more single node cuttings than all the other varieties. No difference existed between the means of V1 and V2, and V2 and V3 respectively. However the mean of V1 is significantly higher than V3. Treatment T2 (500 ppm GA) produced significantly more single node cuttings per corm than T1 (100 ppm GA). 190 third taro symposium GA at 500 ppm gave significantly a higher multiplication ratio (1:134) than GA at 100 ppm (1:91) with all the treated clones. The control produced suckers with a multiplication ratio of 1:4. Discussion In Samoa, farmers propagate taro traditionally by using the top portion of the corm (about 1-2 cm) and 30-40 cm of petioles. Suckers obtained from lateral shoots are also used. This system is often unable to cater for the shortfall in planting material supply. Semi-commercial and commercial farmers often encounter shortages of planting material at the time it is most needed. The major problem confronting the expansion of taro cultivation in Samoa is the inadequate supply of TLB tolerant planting material. Planting material is lost during the dry season particularly in dry areas where prolonged drought conditions may become severe in some years. Most of the crop is planted during the beginning of the wet season and at this time planting material becomes scarce following the dry season and this reduces the area cultivated to the crop caused by the short supply of planting material. During the course of project operation several multiplication techniques were experimented upon to find a technique that will give the highest multiplication rate to support large scale production of planting material of the four TLB tolerant clones. From these observation trials the technique of direct stolon development from corms using gibberellic acid gave the most promising results. Single node cuttings derived from this technique were sprouted and a large number of plantlets were produced from a single corm. This was achieved over the course of an eleven-week cycle. The average multiplication ratio obtained was 1:120. Based on the results of these preliminary investigations an experiment was designed to determine the multiplication ratios of the four TLB tolerant taro clones using two concentrations of GA. This experiment served the basis for the present study. The results of this study has shown that this technique is capable of delivering large quantities of planting material in a short time since stolons are developed directly from corms without the intervention of plant or sucker development. The physiological explanation of the effect of gibberellic acid induction on direct stolon development from corms is not entirely known. However it seems likely that GA plays a role in channeling the carbohydrate reserves in the corm into stolon development instead of sucker or plant development. The results of the present study demonstrate the effect of gibberellic acid at 500 and 100 ppm on direct stolon development in corms of the four TLB tolerant cultivars. The control treatment produced suckers only. It was observed also that GA at the higher concentration (500 ppm) prolonged stolon elongation and slowed down the development of suckers at the end of the stolon. Apical sucker development took place earlier at the lower GA concentration of 100 ppm in all cultivars. Plantlet development from single node cuttings derived from stolons showed that GA at 500 ppm concentration produced a higher multiplication ratio than GA at 100 ppm. The multiplication ratio for varieties ranged from 1:95 to 1:137, indicating that the technique of direct stolon development from corms is many times more rapid than by normal multiplication technique. In this study taro was rapidly propagated vegetatively with the use of single node cuttings derived from stolons produced directly from corms. Nodal cuttings were sprouted in nursery beds to produce plantlets. These were ready for transplanting into the field after 8 weeks. Soil moisture is an important factor in sprouting nodal cuttings and biwall drip irrigation (BDI) is used to keep the soil in the nursery bed at field capacity. The results also indicated that the multiplication rate is much higher than that obtained from tissue culture and conventional field multiplication. It is of interest to note that the multiplication rate can be further increased since more stolons are produced when stolons are harvested from the corm. This technique is new and its now possible to develop stolons directly from corms without having to rely on stolon formation from plants in the field. With the use of this innovated technique, it is now possible to generate large quantities of planting material of the four highly tolerant TLB taro lines for distribution to taro growers. The fact that this technique is complementary to the Bi- wall Drip Irrigation BDI system it is possible to mass produce planting material throughout the year under both wet and dry season. Farmers are encouraged to establish nurseries for the multiplication increase of the four TLB tolerant clones so as to expand the area under taro cultivation. In conclusion, the protocol developed by this corm-to-stolon technique is currently employed for mass propagation of taro in support of the rapid expansion of taro cultivation in Samoa. This technique could be beneficial to taro improvement programs in other parts of the world where taro is grown on an extensive scale. third taro symposium 191 Theme Four Paper 4.5 Breeding Hawaiian taros for the future John J. Cho Department of Plant and Environmental Plant Sciences, Maui Agricultural Research Center, University of Hawaii, Kula, Hawaii Introduction Around the 4th or 5th century A.D. a large double-hulled voyaging canoe originating from the Marquesas Islands laden with taro, breadfruit and other crops made landfall in Hawaii. These were the first Hawaiians. Although only a few different taro varieties were thought to have been brought into Hawaii by the first Hawaiians, over 300 hundred varieties have been documented and represent a reasonable number that were present prior to the arrival of Captain Cook in 1778 (Handy, 1940; Handy et al., 1972). It has been suggested that the large number of varieties may have been derived from genetic crosses made by old Hawaiians and/or selection and propagation of mutant clones. The number of varieties by far outnumbered any found in Polynesia where it came from. The chiefs or ali’i selected many of the varieties for specific characters such as Lehua and Pi’i ali’i that were favored and the low acrid varieties, Lauloa and Haokea, favored as medicinal or ceremonial taros. Moreover many new varieties were selected for their adaptability to different microenvironments encountered with the expansion of agricultural production beyond the banks of rivers and streams. It was in Hawaii where taro was brought to the highest state of cultivation and played an important role in the diet of the people. Upon the arrival of the first Hawaiians in Hawaii, taro or kalo, was first planted along the seacoast in marshes near the mouths of rivers (Krauss, 1993). With an increase in population to what was estimated to be about 200,000 Hawaiians, an intensification of agricultural production occurred to fulfill the demand for food. Accordingly, land in the valleys were cleared developing an elaborate system of plant production under flooded conditions in banked and terraced plots called lo’i, which required delegation of labor and movement of large amounts of water. Taro was not only cultivated under irrigated wetland conditions that are in evidenced today but archeological studies indicate that large plantings along with sweet potato probably occurred on the leeward sides of Maui and Hawaii under dry land conditions (Kirch, 1985). This production system was a key component of a resource management system of land units known as the ahupua’a. Taro production was not merely an activity of food production but was tightly interwoven into the Hawaiian culture and their legends about creation. Genetic improvement of commercial taros are needed to increase resistance to adverse environmental conditions (i.e. high salt and soil pH, low rain fall) and pests, to increase plant vigor and yield, and to develop new and exciting varieties with different colors and tastes for the growing Hawaii regional cuisine restaurant trade. Further, a few taro varieties are being grown for the ornamental-landscape marketplace and a breeding program designed to introduce new and colorful taro varieties is needed. Several of the Hawaiian taros have desirable attributes which could be combined in a number of different combinations to produce superior varieties compared with those that are presently grown commercially. For example, there are different corm flesh colors available including the orange-yellow corm found in Mana Ulu, dark purple or red found in the Lehua varieties, and white found in Moi and Haokea. Two of the varieties (Lehua and Moi) are grown commercially for the red and gray poi market. There may be a place for the development of white and yellow poi varieties. The Kai and Wehiwa varieties are known to be very tolerant or resistant to soft rots as compared to the major commercial varieties. Currently, Kauai wetland growers are experiencing severe crop losses due to a type of soft rot called pocket rot. Development of wetland poi varieties with Kai or Wehiwa variety attributes may be useful in reducing future pocket rot losses. The irritating or acrid factor that is associated with many of the major commercial varieties is one of reasons that has impeded wider acceptance and utilization of taro. Lauloa, Kalalau, and Haokea varieties are relatively nonacrid and could be used in the development of low acrid commercial varieties. Other useful attributes that could be used to improve commercial taros include early maturation (Piko Elele), and brilliant color (Ulaula Kumu). Recent biochemical and genetic evaluations are beginning to provide a basis for distinguishing Hawaiian varieties and understanding the history of taro in Hawaii and the Pacific. Hawaiian and Polynesian taros showed very low genetic variation based upon isozyme variation (Lebot and Aradhya, 1991). On the other hand, variation based upon DNA sequence using RAPD (random amplified polymorphic DNA) markers could be used to distinguish between varieties (Irwin et al., 1998). However, the majority of the Hawaiian taros were found to be closely related with about 80% DNA similarity. The narrow genetic base found in Hawaiian varieties makes sense since they were derived from only a few introduced taros. The geographic region from India to Southeast Asia is the center of genetic diversity for taro (Chang, 1958; Coates et al., 1988; Yen and Wheeler, 1968). Introduction of taro varieties from the center of diversity will provide different genes that could be used in a breeding program to broaden the genetic base of Hawaiian taros. Plant breeders have used genetic diversity in other crops that can be credited for at least one-half of a doubling in yields of rice, barley, 192 third taro symposium soybeans, wheat, cotton, and sugarcane; a threefold increase in tomato yields; and a fourfold increase in yields of corn, sorghum and potato (World Resources Institute, Agriculture and Genetic Diversity: www.wri.org/wri/biodiv/agrigene. html#risks). Also many genes for resistance to pests and tolerance to adverse environmental stress have been introduced into cultivated crops from related wild plants from its center of diversity. For example, several sources of taro leaf blight resistance appear to be available. This disease caused by the fungus, Phytophthora colocasiae is the most important disease in Hawaii and the major taro growing regions worldwide. In Southeast Asia several taro varieties are reported to either resistant or immune to the disease (Deshmukh and Chibber, 1960; Paharia and Mathur, 1964). In the Solomon Islands breeding program, Patel and Liloqula (1985) have developed advanced materials generated from a cross between TLB resistant and susceptible materials and indicate that a single dominant gene confers resistance. Vasquez (1990) in the Philippines reports three taro accessions moderately resistant and one accession highly resistant to TLB when inoculated with P. colocasiae 2 to 4 months after planting. In the Papua New Guinea breeding program, intrageneric crosses have been made between cultivated and wild genotypes with TLB resistance (Ivancic and Kokoa, personnel communication). They indicate that single and multiple gene(s) confer TLB resistance. Greenough et al. (1996) have evaluated several taro lines from Micronesia in American Samoa and Hawaii and confirmed resistance of a few lines (pers. comm.). Importation of the available resistant varieties for breeding purposes is warranted. Materials and methods Germplasm: In 1997 we started to assemble a collection of 298 taro genotypes to be used in a genetic improvement program. Several genotypes were first obtained from the University of Hawaii’s Taro Germplasm Nursery located in Kauai. These included about 70 Hawaiian varieties collected by Whitney et al. (1939), and several accessions from Asia, Indonesia, Polynesia, and Melanesia collected by Lebot (Lebot and Aradhya, 1991). Table 1: Taro (number of accessions and country of origin) used in Hawaiian breeding program Region Number North America USA/Hawaii 56 14 11 2 Micronesia Guam Palau Pohnpei Rota Saipan Tinian Yap Number Far East 63 Southeast Asia Thailand Vietnam Indonesia Myanmar Region 4 15 6 5 6 4 6 India Nepal 2 3 Polynesia Cook Islands Easter Island Niue Samoa 4 6 1 28 Melanesia Fiji New Caledonia Papua New Guinea Vanuatu 2 2 5 22 China Japan Philippines 5 20 3 Asia Many of the Indonesian accessions were shown to be TLB resistant (Java 48, Java 74, Java 75, Kuat, Ketan 36). Furthermore, several taro leaf blight (TLB) resistant genotypes were introduced in Hawaii including 2 accessions (PH15, PH21) from Papua New Guinea (Kokoa and Darie, 1992) and a wild type (Bangkok) originating from Thailand (Patel and Liloqula, 1985) in 1997, 7 accessions (Thailand, Pwetepwet, Gilin, Kugfel, Oglang, Ol, Sushi) from Micronesia (Wall and Wieko, 1998), 1 accession (C81081) from Nepal, and 15 accessions (Ngesuas-P1, Terrekakl-P2, OngdibelP3, Homestead-P4, Ochab-P5, Kerdeu-P6, Ochelochel-P7, Moalech-P8, Ngeruuch-P10, Merii-P12, Dirraiuosch-P13, Moded-P15, Meltalt-P16, Ngetmadei-P19, Dirratengadik-P20) from Palau (Hamasaki et al., n.d.; Trujillo, 1996) in 1999. In 2000, 4 aphid tolerant taro accessions that either reduces aphid longevity (Likay), reduces the number of offspring (Saipan, Rumung 1) or both (Japon) were introduced from Guam (Miller and Wall, pers. comm.). Ketan 36 was also shown to reduce aphid longevity and fecundity. Also many accessions from Southeast Asia were secured during a collecting trip in 1999. Table 1 summarizes the number of accessions and country of origin assembled for our genetic improvement program. Genetic crosses: The major objectives of our breeding program are to develop higher yielding, good tasting food taros and brilliantly colored ornamental taros with increased disease and insect resistance and increased genetic complexity. Our breeding strategy uses a modified backcross and recurrent selection approach. In this approach we first develop first generation F1 hybrids by making genetic crosses between commercial Hawaiian taro varieties and different TLB resistant taro accessions. The F1 progeny are then evaluated, selections are made for desirable horticultural characteristics, and selected BC1s are evaluated for TLB resistance, yield, and taste quality in on farm trials. Genetic crosses will be made between selected BC1 progeny with different sources of TLB resistance in an attempt to combine 2 and more sources to create more durable resistant commercial taros. third taro symposium 193 In general, the commercial type taros are used as pollen donors. The male portion of the spadix is removed when pollen is shed and applied to emasculated pest resistant recipient at anthesis by either brushing stigmas with pollen or placing the male spadix between the female flowers and the spathe. Seeds are collected from successful crosses about 1 month following pollination and germinated in peat moss trays. Fifty to one hundred germinating seedlings from each cross are then randomly selected from each cross and transplanted first into 2.5 cm Speedling flats containing peat moss, and after about 2 months, seedlings are transplanted into field plots located on the island of Maui. Approximately 6 to 8 months after transplanting, field transplants are evaluated and individuals exhibiting the best horticultural characters based upon commercial standards are selected. Results and discussion 1998 Crosses: Genetic crosses were made between three TLB resistant taro accessions (Bangkok, PH15, PH21) and seven different commercial type taro varieties (Table 2). All crosses were successful resulting in viable progeny. About 12% of the progeny were selected for further genetic improvement. Those selected produced short or no stolons, 3 to 12 suckers, and well-shaped corms. However, none of the selected F1 progeny were considered suitable for commercial production because of the small corm size. Average mature corm weights for 8 F1 (Bangkok x Niue Waula) progeny ranged from 0.7 kg to 1.9 kg, 1.2 kg for 1 F1 (Red Moi x PH15); 0.9 kg for 1 F1 (Bun Long x PH21), and 1.1 kg for 1 F1 (Piko Eleele x PH15). 1999 Crosses: In 1999, 24 successful genetic crosses were made between selected F1 progeny from our 1998 crosses and commercial type taros generating over 800 progeny. A few of the crosses made are shown in Table 3. These modified backcrosses were initiated to restore commercial type characteristics. Maui Lehua, the major commercial taro variety grown in Hawaii for poi, was used in several crosses between selected 1998 F1s. All BC1 progeny were transplanted in field plots, visually evaluated in April 2000 for desirable horticultural characteristics and 120 progeny were selected for further evaluation as possible food and 24 as possible ornamental types. Five to 10 suckers from each selected BC1s were removed and planted in field plots on April 10, 2000 and evaluated for commercial potential on March 27, 2001. Thirty-four out of the 120 BC1s were selected in our evaluations, 30 for food and 4 for ornamental uses (Table 4). Fourteen out the 30 BC1s selected as potential food taros came from one cross between F1 (Bangkok x Niue Waula 21) and Maui Lehua. These BC1s varied from one another in their corm flesh color, average corm weights and in the number of suckers produced. Seven BC1s were selected as potential ornamental taros. Four exhibited green, white and/or red striped petioles derived from (F1 [Bangkok x Eleele Naioea 9]) x Van 26 cross and (F1 [Bangkok x Niue Waula 6]) x Fasa Fa Uli cross. Table 2: 1998 crosses between 3 TLB resistant accessions and 7 commercial taro varieties (recurrent parent) and the number of hybrid progeny generated for each cross Recurrent parent No. hybrids Selected Bangkok TLB resistant parent Niue Waula 30 12 Bangkok Eleele Naioea 15 2 Bangkok Moi 3 1 Bangkok Apowale 100 2 PH15 Piko Eleele 5 3 PH21 Bun Long 2 1 PH15 Red Moi 15 2 2000 Crosses: In 2000 twenty-one genetic crosses were made between Hawaiian taro varieties (food type) and ten different TLB resistant taros from Micronesia (Pwetepwet, Olgang, Thailand, Gilin), Palau (Ngesuas-P1, Moalech-P8, Dirratengadik-P20), Indonesia (Ketan 36, Kuat, Java 74), and Papua New Guinea (PH21) that resulted in 432 first generation F1 hybrids. One hundred fourteen F1 hybrids were selected for further improvement. Seven F1 hybrids (F1 (Pwetepwet x Maui Lehua 3), F1 (Pwetepwet x Maui Lehua 9), F1 (Pwetepwet x Maui Lehua 13), F1 (Maui Lehua x Thailand 54), F1 (Maui Lehua x Thailand 56), F1 (Moi x P20-9), F1 (Maui Lehua x Sushi 7) produced large corms weighing between 2.2 to 3.8 kg per plant; these F1 hybrids are being evaluated in on farm trials for TLB resistance, yield, and taste characteristics. 194 third taro symposium Table 3: 1999 modified backcrosses between selected F1 hybrid and recurrent commercial type taro varieties, number of BC1 progeny generated and number (percent) of BC1 individuals selected for further evaluation Selected F1 Selected Recurrent parent No. hybrids F1 (Red Moi x PH15) Maui Lehua 8 0 0 F1 (Bangkok x Niue Waula 21) Maui Lehua 67 22 32.8 F1 (Bangkok x Niue Waula 21) Niue 53 7 13.2 F1 (Bangkok x Niue Waula 21) T6 47 6 12.8 F1 (Bangkok x Niue Waula 21) Veo 12 2 16.7 F1 (Bangkok x Niue Waula 21) Fasa Fa Uli 5 1 20 F1 (Bangkok x Apowale 9) Maui Lehua 40 7 17.5 F1 (Bangkok x Apowale 9) 25 No. % Lauloa Keokeo 8 2 F1 (Moi x Bangkok 2) Lehua Maoli 10 2 20 F1 (Moi x Bangkok 3) Van 49 16 2 12.5 F1 (Bangkok x Eleele Naioea 9) Maui Lehua 12 0 0 F1 (Bangkok x Eleele Naioea 9) Lauloa Keokeo 27 0 0 F1 (Bangkok x Eleele Naioea 9) Van 26 63 13 20.6 F1 (Bangkok x Eleele Naioea 6) Veo 28 1 3.6 F1 (Bangkok x Eleele Naioea 6) Kai Ala 84 8 9.5 F1 (Bangkok x Eleele Naioea 6) T2 12 0 0 F1 (Bangkok x Eleele Naioea 6) Van 96 15 3 20 F1 (Bangkok x Eleele Naioea 6) Veo 45 1 2.2 Nine modified backcrosses were also made between selected 1998 F1 hybrids and commercial taros that resulted in 205 BC1 progeny. Twenty-two BC1s were selected for further backcrosses. Twenty crosses were made for the development of ornamental taros. Nine crosses used PH21 as one of the parental lines to develop F1 populations with purple and green striped petioles. Six crosses used a Hawaiian taro, Lauloa Palakeapapamu, for its dark purple petiole color. Table 4: Attributes of thirty-four 1999 BC1 individuals selected for further evaluation for commercial food or ornamental uses. Selected BC1 BC99-2 BC99-3 BC99-4 BC99-5 BC99-6 BC99-7 BC99-8 BC99-9 BC99-11 BC99-13 BC99-24 BC99-31 BC99-32 BC99-19 BC99-1 BC99-21 BC99-25 BC99-26 BC99-34 BC99-22 BC99-30 BC99-10 BC99-15 BC99-16 BC99-17 BC99-20 BC99-12 BC99-23 BC99-14 BC99-18 BC99-27 BC99-33 BC99-28 BC99-29 Selected F1 parent F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Niue Waula 21] F1 [Bangkok x Eleele Naioea 9] F1 [Bangkok x Eleele Naioea 9] F1 [Bangkok x Eleele Naioea 9] F1 [Bangkok x Eleele Naioea 9] F1 [Bangkok x Eleele Naioea 6] F1 [Bangkok x Niue Waula 6] F1 [Bangkok x Niue Waula 6] F1 [Bangkok x Apowale 9] F1 [Bangkok x Apowale 9] F1 [Moi x Bangkok 3] F1 [Moi x Bangkok 2] F1 [Red Moi x PH15] F1 [Red Moi x PH15] Recurrent parent Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Maui Lehua Niue WBL T6 T6 T6 Van 110 Veo Van 26 Van 26 Van 26 Kai Ala Kai Ala Fasa Fa Uli Fasa Fa Uli Maui Lehua Maui Lehua Van 49 Lehua Maoli Van 4 Maui Lehua Corm Color fresh Corm wt. (kg) No. suckers Use Pink Purple Purple Purple Purple Pink Purple Purple Pink White White Pink White White White White White Pink Yellow Pink White Pink White White White White Pink Pink Pink Pink Pink Yellow White White 2.1 1.8 2.6 2.4 4.4 2.2 1.4 2.8 2.6 1.6 3.2 2.4 2.1 2.5 2.6 1.4 3.0 1.6 2.8 1.5 2.2 0.8 1.5 2.5 2.8 3.7 0.6 2.9 1.3 1.4 2.3 2.2 2.8 2.0 17.5 12.5 9.7 5.3 7.3 11 16.7 4.5 24 10 12.5 15.5 12 6.5 17.5 17 10.5 19 food food food food food food food food food food food food food food food food food food food food food ornamental ornamental ornamental food food ornamental food food food food food food food 14 13 13.3 21 6.5 13 11 25 17.5 7.7 8 14 14 13.5 third taro symposium 195 Genetic crosses between Indonesian taros (Ketan 36, Kuat, Java 74) and Maui Lehua, PH21, and Agaga resulted in F1 progeny that showed a wide variation in phenotype. Variation occurred in plant size from 0.3 m to over 2.5 m in height, leaf shape and size, and plant color. Six out of 29 F1s from a cross between Ketan 36 and PH21 were selected as possible ornamental taros. A commercial nurseryman is now evaluating these taros. This kind of segregation where individuals of the population fall beyond their parental phenotypes has been referred to as transgressive segregation. This type of segregation might have evolutionary implications, since it can affect characters of adaptive significance leading to new races or species (Schwarzbach et al., 2001). Transgressive segregation may allow individuals to occupy new ecological niches or to better compete in existing environments (Rieseberg et al., 1999). For breeding, it represents a potential source of novel variation. These transgressive segregants will be evaluated for possible sources of root knot nematode resistance, increased yields, and as potential ornamentals. 2002 Crosses: In 2002, four hundred modified backcrosses were made between F1 hybrid plants selected from our 1998 and 2000 crosses and commercial type Hawaiian taros for the development of food type taros. Again many crosses were made to generate more ornamental type taros. More than 5,000 seedlings have been transplanted in field plots for selection sometime this year. References Chang, T.K., 1958. Dispersal of taro in Asia. Annals of the Association of American Geographers 48:255–256. Coates, D.J., Yen, D.E. and Gaffey, P.M. 1988. Chromosome variation in taro, Colocasia esculenta: Implications for the origin in the Pacific. Cytologia 53:551–560. Deshmukh, M.J. and Chibber, K.N. 1960. Field resistance to blight (Phytophthora colocasiae Rac.) in Colocasia antiquorum Schott. Current Science (Bangalore) 29:320–321. Greenough, D.R., Trujillo, E.E. and Wall, G. 1996. Effects of nitrogen, calcium, and or (sic) potassium nutrition on the resistance and/or susceptibility of Polynesian taros, Colocasia esculenta, to the taro leaf blight, caused by the fungus Phytophthora colocasiae. p. 19–25. In: ADAP Project Accomplishment Report, Year 7. Agricultural Development in the American Pacific Project, Honolulu. Hamasaki, R., Sato, H.D., Arakaki, A., Shimabuku, R., Fukuda, S., Sato, D., Yoshino, R. and Kanehiro, N. Leaf blight tolerant taro variety project. http://www.extento.hawaii.edu/IPM/taro/default.htm. Handy, E.S.C. 1940. The Hawaiian planter, volume 1: His plants, methods and areas of cultivation. Bernice P. Bishop Museum Bulletin 161. Bishop Museum Press, Honolulu. Handy, E.S.C. and Handy, E.G. 1972. Native planters in Hawaii: Their life, lore, and environment. Bernice P. Bishop Museum Bulletin 233. Bishop Museum Press, Honolulu. Irwin, S.V., Kaufusi, P., Banks, K., de la Pena, R. and Cho, J.J. 1998. Molecular characterization of taro (Colocasia esculenta) using RAPD markers. Euphytica 99:183–189. Kirch, P.V. 1985. Feathered gods and fishhooks. University of Hawaii Press, Honolulu. 349 p. Kokoa, P. and Darie, A. 1992. Field screening of taro (Colocasia esculenta (L.) Schott) for resistance to taro leaf blight (Phytophthora colocasiae) in Papua New Guinea. Internal report. Krauss, B.H. 1993. Plants in Hawaiian culture. University of Hawaii Press, Honolulu. Lebot, V. and Aradhya, K.M. 1991. Isozyme variation in taro (Colocasia esculenta (L.) Schott) from Asia and Oceania. Euphytica 56:55–66. Paharia, K.D. and Mathur, P.N. 1964. Screening of Colocasia varieties for resistance to Colocasia blight (Phytophthora colocasiae Racib.) Science as Culture 30:44–46. Patel, M.Z. and Liloqula, R. 1985. Progress on breeding disease resistant taro in Solomon Islands. In: Fifth Conference of the Australasian Plant Pathology Society, Auckland, New Zealand, 20–24 May 1985. Australasian Plant Pathology Society, Auckland. Rieseberg, L.H., Archer, M.A. and Wayne, R.K. 1999. Transgressive segregation, adaptation, and speciation. Heredity 83:363–372. Schwarzbach, A.E., Donovan, L.A. and Rieseberg, L.H. 2001. Transgressive character expression in a hybrid sunflower species. American Journal of Botany 88:270–277. Trujillo, E.E. 1996. Taro leaf blight research in the American Pacific. ADAP Bulletin 1:1–3. Vasquez, E.A. 1990. Yield losses in taro due to Phytophthora leaf blight. Journal of Root Crops 16:48–50. Whitney, L.D., Bowers, F.A.I. and Takahashi, M. 1939. Taro varieties in Hawaii: Hawaii Agricultural Experiment Station Bulletin 84. HAES, Honolulu. Yen, D.E. and Wheeler, J.M. 1968. Introduction of taro into the Pacific: The indications of the chromosome numbers. Ethnology 7:259–267. 196 third taro symposium Theme Four Paper 4.6 The establishment of a commercial tissue culture laboratory in the Kingdom of Tonga Paul Karalus Pacific Biotech Ltd, Tonga Why a tissue culture laboratory for taro? While it is easy to obtain planting material from the field or from taro nurseries, and in good quantity when properly planned, it is not always the case that adequate or sufficient suitable material is available. The more common experience in the context of the Pacific Islands is that taro planting-material is often begged, stolen and even long term borrowed. This makes it difficult to put a money value on the planting material and mitigates against the sale of planting material from either the field or a nursery. Furthermore, material sold from the field, as most islands do not use nurseries, is highly variable in type of cultivar, in quality, in freedom from disease, and often the resulting crop is variable in maturation or in uniform size of the tubers. Tissue culturing in commercial taro production has its best application in establishing new taro plantations of commercial size. This is so because adequate new material of a particular cultivar can be obtained, it can be uniform in size and it can be clean of virus and disease. The cultures can also be used to establish taro nurseries from which large quantities of planting material can be obtained. Once there is a suitably large enough nursery it might be thought there is no longer a need for tissue culturing. It is, however, suggested that new clean material is necessary in a three to five year cycle. The length of the cycle may vary, but good farming practice would suggest that replacing old gene stocks with new ones would best provide against disease and deterioration of the planting material. Another good application of tissue culturing is in the preservation of bio-diversity in planting material. Where the same cultivar is used in repetitive cycles it is often advantageous to change the cultivar and tissue culturing allows good stocks to be made available quickly. Further to this is the introduction of new cultivars into commercial production. Layout and staffing within the laboratory Tissue culture laboratories come in all shapes and sizes. There is no optimum size as the purpose will often determine the size. What is proposed here is a laboratory that optimises the commercial requirements for viability as well as the operational requirements that allow for a good efficient specialisation of roles and the development of team skills in a diversity of plant types. A good laboratory should aim to be practical, have a diversity of products, be financially viable and be a continuous production entity so that it efficiently uses the physical and human resources required. While in highly developed economies labour is the single largest expense (usually around 70% of the total cost) this varies in a small island economy marked by the high costs of energy (labour is 45% of costs and utilities 25%). Plant costs (facility costs), chemical supplies, packaging and other consumables make up the balance of costs. Pacific Biotech Ltd has twelve laminar flow cabinets but has only used ten of these until now. Two of the ten are often used to fill the flasks with media and so only eight were fully productive, this will now increase to ten. Given this number and given a single six hour cutting shift it is possible to produce over two million plants in a year. Each cutter averages over three thousand cuts per shift. By running an additional shift or more shifts it would be possible to further increase production. Theoretically the facility could turn out a maximum of over six million plants per year if the growing room space was large enough to hold the production until sale or nursing out. This would assume three shifts. It is the size of the growing room(s) that determines the maximum production of the Pacific Biotech laboratory. For good hygiene most laboratories advise that a single shift where the same operator utilises the same space is the best way to ensure good hygiene and prevention of contamination. Technical expertise and management can be a single person role, but to make the role a viable one requires a certain size of operation. This is where a single shift of ten cutters makes good use of the technical and managerial expertise of a single person. This size of operation requires a further supervisor of the cutting room and an assistant to screen the flasks for cutting, to label the flasks of cut-on material and to control the storage of the flasks on the growing room shelves. To provide the media for this size of operation requires a supervisor who makes up the media, two hands to control the “cooking” of the batches, the ladelling of the media into the flasks and the sterilising of the flasks where they are re-cycled (a must in a small economy). third taro symposium 197 To provide for a constant flow of materials there is another need for a person to control stores, to order and obtain supplies and to supervise the packing and despatch of the production. A casual hand or two can handle the actual packing of the flasks into boxes and the labelling of the boxes, etc. Management of the accounts and other organisational requirements are out-sourced to a central facility for the total business activities of the Karalus family. The total staff of the laboratory is one Laboratory Manager, one supervisor and one assistant for the cutting room and growing room, ten cutters, one supervisor and two assistants for the media room and the kitchen and one supply and despatch supervisor with two casuals. That is seventeen permanent staff and two casuals. This arrangement allows for specialisation. However, being small it also allows for staff to be proficient in more than one area in the event of illness, leave periods, and seasonal production requirements. Ideally the two casual staff can also fill in when required. This size of operation allows for effective group training to be given, it also allows for good team building and offers opportunity for the development of a stratification of roles where experience and ability can be rewarded. Team building is a very important requirement for a tissue culture laboratory as the work is highly repetitive and monotonous. This size of operation allows for a diversity of plants to be produced. As each plant has different seasonal requirements it is very important to ensure that a diversity of plants provide for a constant flow of production. Pacific Biotech exports calla lily cultures from August to January and provides taro for local plantings from January to August. Vanilla is produced throughout the year. If there is only one production type a large part of the year could be taken up with down time. Should taro be the only product, however, this is lessened as taro can be taken off for most of the year and the nursing-on time varied to suit the optimum planting out times. Other products like bananas could also be produced for year round production. The other diversity is that the mix of an export line and a local line(s) reduces the risk of exposure to a single production line and to either a local or an export market. Laboratory standards Probably the biggest single requirement for a laboratory is the preservation of a sterile environment. In Tonga the laboratory manual requires all staff to shower before entering the media room, the cutting room, the kitchen and the growing room. This recognises the variety of exposures to contaminants from the outside environment and engenders good laboratory work practices for all staff. Visitors into any of these areas are discouraged and only people authorised by the Manager are permitted to enter. Visitors must also take showers if they are to enter production and growing areas. Size of the operation makes for an easy acceptance of Operational Manuals and their strictures and also for the adherence to standards and practices required by quarantine protocols where plants are exported. In summary, this section has considered the optimum size for Pacific Biotech with a single shift. It has to be stated, though, that it has taken over two years time to reach a full understanding of the work flows that exist and the most efficient, effective and economical operating practice for the laboratory. The first requirements are to insist on standards of hygiene and work practice and to control the heat and light environment to ensure contamination (either pathogenic or bacterial) does not occur. The second requirement, where initiating is also carried out (as for taro in Pacific Biotech), is to design production flows to ensure there is a smooth production schedule. This can be very difficult but is essential to the viable operation of a laboratory. The Pacific Biotech laboratory There are two distinctly different sections to a tissue culture laboratory. One is the production section and the other is the plant growing section. A production area can produce material for growing areas that can vary greatly in size. Ideally the smallest growing area allows for the full time use of a single shift of the production area. Thereafter the growing area can be expanded to take up additional shifts with each additional area taking up a further single shift. For Pacific Biotech the production area is of 100 m2. This supplies a growing area of 60 m2. The production area is considered here first. As the laboratory was purpose built it allowed for simple modules to be used in the construction. It is suggested that this is the easiest way to construct a tissue culture laboratory. Each of the modules houses a separate function but each is in close proximity to the other to allow for a production flow. If this is not possible the growing room can be separated from the production area and have its own air source. In any event, in the Pacific Islands all must be under the one roof. Pacific Biotech occupies an upper mezzanine floor in a large steel framed and clad building of 20 by 25 metres. The laboratory takes up 160 m2 and there is ample room for expansion. Being on the mezzanine floor the laboratory is separated from the other functions of the building, viz., warehousing, food and coffee processing. This arrangement also means the laboratory is not exposed to the outside extremes of heat and weather. This is extremely important. The Pacific Island environment demands that there be air conditioning to provide even temperatures (22-25ºC) and to reduce humidity. Should the laboratory walls also be the external walls this adds substantially to the cost of building the laboratory. 198 third taro symposium The external walls, the interior partitions and the ceiling of the laboratory are of 50 mm freezer panels. These are easily put together and provide excellent insulation. An extension is made of ply wood exterior cladding and gibraltar board interior cladding (again good insulation.). The entire flooring is of marine ply. These materials provide for good sealing off of the lab from exterior contaminants. Ideally there is only one entrance commonly used. There should, however, be security doors from both the “cooking” room and the cutting room to be used only in the event of fire. The following dimensions are a simple rule of thumb. The entrance way and toilet and shower block and staff locker room take up 20 m2 (20%) of the production area space. The media make-up room and chemical store room make up a further 20 m2 (20% again) of the production area. A further 20 m2 makes up the “cooking room” or kitchen which also has tubs and tables for packing and re-packing sterilised flasks, etc. A staff table is also here to allow for a tearoom. The cutting room with the laminar flow cabinets is twice the size of the other rooms at 40 m2 (40%) of the production area. While the above dimensions may be varied they have been found to be well suited to the functioning of the laboratory. It is important that the “dirty” area is very well defined and is enclosed and separated from the sterile areas. In the toilet/shower block and staff locker area, the media make up room and the kitchen there is a need for water, preferably fresh rainwater as opposed to artesian water which on atolls is very hard. Filtered water is essential for the media make up room which must also have a very stable floor to provide a firm base on which stands a delicate scales capable of weighing up to one thousandth of a gramme. The media make up room also requires adequate table and bench space, secure and dry locker space, cupboards and a refrigerator. This is for the storage of chemicals and other media ingredients. As the kitchen area has heat generated by the autoclaves or sterilisers it is best to locate this at the furthest extremity so that the air flow exits the laboratory from this room and takes the heat with it. In the laboratory of Pacific Biotech there is a single source of air. This air is filtered and then air conditioned before entering the laboratory. The air is pushed by a large fan system and enters the laboratory through ducts that flow into the laminar flow cabinets from above where it is further filtered by hepa filters. This flow of air then passes over the cutting benches such that no impurities can be picked up before cutting. On the benches there is space for electric glass bead sterilisers for the cutting instruments. The cutting room extends across the full width of the laboratory and is at one end of the laboratory. The air flow therefore enters the cutting room first as this is the room where greatest sterility is required. It is this same air flow that is passed through one half of the laboratory and into the growing room and then onwards into the kitchen. The other half width of the laboratory takes the filtered air from the cutting room into the media preparation room and then through the toilet/shower block before onward through the kitchen. All air exits through a duct at the end of the kitchen and is returned to the air conditioners and fan for a further revolution through the laboratory. In this manner the entire laboratory becomes an enclosed air system though fresh air can enter before the filters and air conditioners should leaks occur (e.g. on the doors being opened for entry to the laboratory, etc. Ideally there is an air lock in an entrance porch to reduce the chance of impurities entering by way of the outside air. The growth room is adjacent to the production area and is linked. There should be maximum air flow into the whole cross-section of it against which there are the shelves which hold the flasks of plants. To assist in this the doorway from the cutting room into the growing room is ideally kept closed as much as possible and is a full plate door (not perforated or vented). This means all air passes through a perforated wall between the cutting room and the growing room and then directly over the shelves. As heat is generated by the lights over the shelves holding the flasks of plants it is important to ensure that the air flow is ducted over the shelves and not down the corridors between the shelves. This is achieved by suspending 90% shade cloth screens down the full length of the racks of shelving. Should the flasks receive more heat than the corridors “misting” caused by condensation occurs in the flasks. It is this condensation that can cause significant bacterial contamination. Large fluctuations in heat also cause plant stress and again can result in increased contamination. A final consideration is that the power supply must be regular and should not fluctuate so as to cause equipment failure. Pacific Biotech’s laboratory is located in the Small Industries Centre in Nuku’alofa and is only a short distance from the power generators. Three phase power is the preferred power to ensure even distribution. The set-up costs and operating costs of a laboratory Pacific Biotech was fortunate to purchase an existing tissue culture laboratory from New Zealand as a going concern. This reduced the initial costs markedly. Costs here are shown in US$ and are approximate. It is assumed that a twelve laminar flow cabinet laboratory is built. There are several main components of a laboratory. There is the housing for the laboratory, the laboratory equipment itself, the supplies that are needed for the media and the plants and then the personnel for the plant. If the laboratory is free standing and is of 160 m2 the structure will cost a minimum of US$50,000. This is to provide a water tight raised slab, a solid cyclone proof structure that is secure, air tight, well insulated and built of high quality materials with large surface exterior and interior cladding panels. The air ducts are included in this. third taro symposium 199 The laboratory itself will cost a further US$50,000. This includes the laminar flow cabinets (minimum of 12). There is also the fan system for driving the air flow and an air conditioning system and a back-up system (Pacific Biotech has two main air conditioners and a smaller one for a downstairs growth room). There are large quantities of shelving and a lighting system with exterior ballasts required for the growing room. There is the set of autoclaves or large sterilisers for the kitchen area. There are the media make up facilities, kitchen facilities, water pumps and tanks and secure plumbing systems for the toilet and shower facilities. There are the accurate scales and a host of other production requirements. There are also the high quality comfortable stools for the cutting room and the overall furniture for the laboratory. Experience indicates that short cutting the equipment and furniture can cost a lot in terms of not preventing contamination. The initial supply of consumables is also large and is included in this section. It is suggested here that a further US$50,000 is required to get properly qualified people to initiate the training of staff. Pacific Biotech had the luxury of technical assistance from the Centre for Industrial Development of the European Union. A Pacific Islands Industrial Development Grant of NZ$75,000 was also from the New Zealand Government to assist in the setting up of the laboratory. The first year of operation is a non productive one as it is very difficult and slow to get a team together and to teach them the art of tissue culturing. It is suggested then that this US$50,000 may not be adequate to get the personnel side set-up. Pacific Biotech has found too that the first three years of production will not be profitable in that it takes that long to iron out the technical problems and to also establish an optimum operating regime. This is likely to be a further US$50,000 of expenditure. There will also be modifications necessary to the actual laboratory. So far then US$200,000 has been expended and the production is only half of the capacity. This is also the experience of Pacific Biotech and is a common experience in the establishment of a tissue culture laboratory. After establishment Pacific Biotech is now in a position where it has a competent and stable staff. It can do all its own initiations and is poised to achieve a viable level of production though short of the total potential production capacity. For the 2003-2004 financial year the production target is for 800,000 calla lily cultures, 250,000 taro cultures and 50,000 vanilla cultures. Capacity remains for further production beyond this. The total operating costs for a year’s operation at this level is approximately US$100,000. The breakdown of this expenditure is as follows: Management and staff Utilities (power, water, gas) Insurance, R and M, rent, depreciation Chemicals and other supplies Flasks, packaging, paper towels, etc. Communications, marketing, stationery 45% 25% 12% 7% 7% 4% In this type of business there should be at least a 25% return on capital as there are high risks and there are exposures to exchange risks, to market restraints and market demands. A 25% return on capital would add a further US$50,000 to the annual returns expected to make the enterprise an attractive investment. This means the 1,100,000 cultures sold need to yield USD 0.13 each to return the 25% return on capital. And that makes the price of Pacific Biotech products competitive in the international market. Should the surplus capacity be taken up there is little to add to costs and the yield from a single culture can be reduced accordingly. At US$0.13 there are probably few farmers who would buy tissue cultured taro plants on an on-going basis. However to initiate a nursery the price is attractive as from each culture many suckers can be obtained. It is noted, however, that in the first generation there are fewer suckers produced from tissue cultured plants. It is the second and third generations that produce a greater number of suckers, but there is a deterioration in the standard of the planting material. It is this need to renew nursery stock that makes the tissue culturing of taro an attractive proposition. Taro in tissue culture It is intended in this final section to briefly note some observations on taro in tissue culture. While not a plant pathologist and not a tissue culture expert myself there are I believe some features of tissue culturing taro that could be the study of plant scientists. It has been observed that in initiating taro that the younger more vigorous suckers provide the best material for initiations. This differs from the initiation of calla for instance where more mature bulbs provide good material from their primary sprouts. The initiation period for taro is approximately twelve weeks. In this time four cuts will be made with there being very little multiplication. Once multiplication is commenced it has been noted that by increasing the number of plants in a flask there can be very significant increases in the rate of multiplication. This suggests that taro has a communal nature in its growth. There is a need for them “to hold hands and grow in harmony”. Further to this when young cultures are taken from the flasks for nursing-on it is noted that if planted separately in cell trays the taro does not do as well as when planted in concert. Calla lilies have the same characteristic. A methodology developed then is to plant the whole of a flask (30 cultures) in a PB5 planter bag of soil and nutrient. The resulting plants acclimatise much more readily, transplant shock is minimised and growth is greatly enhanced. 200 third taro symposium After nursing out and when plants are planted in the field they need more protection from the wind and sun than suckers. This is achieved by planting the small plants in holes up to 15 centimetres deep and 15 centimetres wide. There they will survive even if smothered by soil in heavy rain. Taro has been used as an intercrop in the first year of planting out coffee trees. This has been to provide shelter and shade to young coffee trees as well as to provide some income in the early years of planting coffee. It has been noted though that taro does not do well when planted in widely spaced single rows. There is huge leaf growth but poor tuberisation. The same taro plants, however, do much better when planted densely in field areas of taro only. Again this may be due to a communal nature of taro referred to earlier. Finally there is a common belief that among Tongan farmers that taro does not respond to fertilisers. It has been noted that tissue cultured taro have responded to fertiliser with large increases in tuber size and uniformity and especially when planted communally. These observations would suggest that the use of tissue cultured taro does provide good nursery material where greater density is used and also that tissue cultured taro may produce a return from the use of fertiliser that justifies any additional expense over and above suckers used as planting material. It is hoped that these observations may be taken up for future research. third taro symposium 201 Theme Four Paper 4.7 Taro breeding in India M.T. Sreekumari, K. Abraham, S. Edison and M. Unnikrishnan Central Tuber Crops Research Institute, Trivandrum, India Introduction Colocasia esculenta (L.) Schott is a popular tuber crop consumed as a vegetable in India and South East Asia, while it is an important staple food in the Pacific region. It is commonly known as taro, dasheen, cocoyam and occasionally eddoe. It belongs to the family Araceae. Despite its importance as a popular edible tuber crop, very little attention has been devoted to the genetic improvement of taro. A lot of research on taro has been focused on chromosome studies (Yen and Wheeler, 1968; Coates et al., 1988; Sreekumari and Mathew, 1991a, 1991b). Information on the sexual potentialities of the crop has been very fragmentary, and the improvement programme has been largely dependent on the exploitation of the existing genetic variability among the cultivars (Kuruvilla and Singh, 1981; Plurre, 1984; Tanimoto and Matsumoto, 1986; Velayudhan and Muraleedharan, 1985; Unnikrishnan et al., 1987, 1988; Lebot and Aradhya, 1992). For a long time it was believed that taro plants do not flower, and therefore fail to produce seeds (Plucknett , 1970; Shaw, 1975). However, later reports from different countries indicate that many clones flower and produce viable seeds (Abraham and Ramachandran, 1960; Plucknett Shaw, 1975; Jackson et al., 1977; Jos and Vijaya Bai, 1977; Strauss et al., 1980; Ghani, 1979). Because of its long history as a cultivated crop and of the vegetative mode of propagation, it has been possible to select and preserve distinct types or varieties in taro, which are useful to man. Accumulation of such multiplicity of types must have made it possible for suitable cultivars to be selected in different areas and in various growing conditions of soil, water, altitude, temperature, planting practices, etc. Doku (1980) has pointed out that this large store of variation present in the crop would be immediately available for utilization in the raw state and in all future combinations and recombination if conditions for flowering, hybridization, seed setting and raising of seedlings are discovered. Research priorities for taro genetic improvement • Germplasm collection, maintenance, characterization and evaluation • Publishing bulletins and catalogues for easy reference • Evaluation of the collections and identification of suitable cultivars for (1) direct release based on performance in the different field trials conducted at the Institute farm and in on farm trials and (2) incorporate in various breeding programmes • Intervarietal hybridization and production of true seeds • Production of selfed seeds • Evaluation of the seedling and subsequent clonal progeny for various desirable attributes • Identification and multiplication of the selected hybrids for the conduct of field trials • Isolation of superior hybrid selections for release • Production of triploids in large scale by crossing induced tetraploids with diploids • Evaluation of the induced triploids and identification of genetically improved types. The taro gene bank The taro gene bank at the CTCRI possesses 424 accessions collected from all over India. The cormels are planted in April-May and harvested after six months. Important characters were recorded based on modified IPGRI descriptor and catalogues and bulletins were published. (Unnikrishnan et al., 1987). A wide spectrum of genetic variability was noticed among them for several characters, especially growth habit, pigmentation on different plant parts, crop duration, flowering habit, fertility, open pollinated fruit and seed set, size, shape and yield of corm and cormels, cooking quality, tolerance to leaf blight disease etc. Evaluation trials were undertaken such as row trials, preliminary evaluation trials, advanced yield trials and finally on farm trials prior to the release of superior selections. Two accessions, viz. C.149 and C.266, were thus released for general cultivation under the names Sree Reshmi and Sree Pallavi respectively from the CTCRI. A gist of the characters of these two varieties and few of the elite germplasm collections identified are given in the Table 1. Ploidy in relation to tuber yield Twenty diploids and twenty triploids were evaluated for growth performance and yield. It was observed that triploids are superior to diploids in seven of the nine characters studied. The corm and cormel yield showed very promising and impressive increase in the triploids except in the case of cormel number which was significantly more in diploid plants. 202 third taro symposium This implies that for selecting high yielding types in taro it is desirable to consider the triploids rather than diploids. The comparative performance and extent of variability noticed within the two ploidy types is given in Table 2. Table 1: Important characters of the released varieties and other elite germplasm selections Germplasm selections Description 1. C.149 (Sree Reshmi) This is a local collection and is a natural triploid. It grows to a height of 1-1.5 m and has edible petiole, leaves, corms and cormels. Cooking quality is excellent. It matures in 7-8 months and cormel yield ranges from 15-19 t ha-1. Cormels contain 14-16 percent starch and 2-2.5% protein. This was released from the CTCRI during 1987. 2. C-266 (Sree Pallavi) It is another released variety of taro from the CTCRI. It is a triploid collection from Meghalaya. Plant height ranges from 1-1.5 m and the crop duration is 7-8 months with a yield potential of 12-15 t ha-1. Only the cormels are edible. The starch content ranges from 19-23% and protein content 1.8-2.1%. 3. C-9 A triploid accession. It is an early maturing type (5-6 months) with medium height (60-80 cm). The leaves are medium broad and average yield (cormel yield) is 15 tonnes per hectare. Cormels are excellent for culinary purpose but the cormel keeping quality is poor. 4. C-189 It is a high yield triploid with an average yield of 20 t ha-1. The plants are tall types with comparatively long duration. The cormels are many and long fusiform in shape. The most attractive attribute of this selection is the long keeping quality of the cormels (4-5 months) 5. C-303 and C-384 These two diploid accessions are the only ones in the germplasm that flower almost regularly. The flowers are highly fertile resulting in open pollinated seed production. The plants are of medium type, cormel yield ranges from 10-14 t ha-1 and cooking quality is good. Both the accessions are usually incorporated in the breeding programme. 6. C-408, C-444/2 and C-481 These accessions are tall, late maturing types. They produce comparatively large main corm (>500 g) with 8-10 well-developed big cormels. They are identified as triploids, but based on tuber characters they can be treated as ‘intermediate types’ between dasheen and eddoe groups. Table 2: Comparative performance of diploid and triploid taro Sl no. Diploid Characters Triploid t value Mean CV (%) Mean CV (%) 1 2 Plant height (cm) Tiller number 69.7 ± 0.47 3.4 ± 0.14 33.14 44.12 76.3 ± 2.2 3.6 ± 0.17 31.6 52.8 2.168* NS 1.090 NS 6.276 6.149** 3 Number of leaves 8.3± 0.56 73.49 8.1 ± 2.22 60.49 4 Shoot girth (cm) 8.8 ±0.79 98.86 14.6 ± 0.46 34.93 5 Leaf length (cm) 20.5 ± 0.84 44.87 30.4 ± 0.58 21.05 9.628** 6 Leaf breadth (cm) 16.0 ± 0.67 45.63 25.4 ± 0.56 24.02 10.823** 7 Corm weight (g) 141.2 ± 8.01 62.11 203.6 ± 10.94 58.59 4.600** 8 Cormel number 18.8 ± 1.00 65.48 12.6 ± 0.70 60.16 3.050** 9 Cormel weight (g) 206.2 ± 7.69 65.48 447.1 ± 8.03 26.51 6.348** * – Significant at 5% level ** – Significant at 1% level NS – Not significant Flowering frequency and floral biology Flowering was seasonal mostly starting by the middle of June (2-3 months after planting) and lasting till the middle of September. Highest frequency of flowering was usually observed during the last week of July. Flowering is rather very low in cultivated forms. However, frequency of flowering in wild taro grown in marshy areas under waterlogged condition was notably high everywhere. The genetic resources maintained at the Central Tuber Crops Research Institute (CTCRI) showed that natural flowering is neither profuse nor predictable which might be correlated with the characteristics of the location, which is as follows: Latitude Longitude Minimum night temperature Maximum day temperature Rainfall Day length Crop season 8°40’ N 77°0’ E 19°C 33.4°C 1400-1500 mm 11 hr 23 min (shortest in Dec.) to 12 hr 39 min (longest in June) April-October In addition to the environmental factors, flowering was also influenced by the ploidy status of the genotype. A few of both diploid and triploid accessions flowered but frequency was more among diploids compared to triploids ranging from 2.5-5.0% in the former and from 0.8-2.5% in the latter. Flowering data recorded from 120 diploids and 119 triploids during three consecutive seasons is given in Table 3. third taro symposium 203 Table 3: Frequency of flowering in the two ploidy types of taro during three years in two seasons Year Season No. of accessions flowering (%) 2n 3n 1990 April planting November planting 5 (4.2) 1 (0.8) 2 (1.7) 0 (0.0) 1991 April planting November planting 6 (5.0) 0 (0.0) 3 (2.5) 0 (0.0) 1992 April planting November planting 3 (2.5) 1 (0.8) 1 (0.2) 0 (0.0) Flowering ability and floral productivity are extremely important for the conventional breeding process. In India, poor flowering is the main factor limiting planned hybridization. Ploidy variations, incompatibility, female or male sterility, disease, soil conditions, heavy rain, shade, etc. were the other factors limiting planned hybridization. Floral biology was studied and pollination techniques standardized. Stigma receptivity was found to last for a considerable length of time. Jos and Vijaya Bai (1977) observed that on the day of opening the inflorescence, the percentage of successful pollination was 85.7. The stigma receptivity was at its peak for some hours after the liberation of pollen also. Limited receptivity was noticed upto 36 hours after the liberation of pollen. and there was no seed set beyond that period. The detailed study had established that trace receptivity could be realized for 44 hours earlier to anthesis and 60 hours later to anthesis ( Sreekumari and ThankammaPillai, 1994). Intervarietal hybridization The main targets of taro breeding were: • Genetic variability • Ideal plant type • Cormel yield • Taste quality of cormels • Resistance/ tolerance to taro leaf blight • Early maturity • Longer keeping quality • Density tolerance • Incidence of flowering and floral productivity. However, as flowering was irregular, non-synchronous crosses could be made using the available fertile diploids only. For the successful breeding of novel taro varieties with new combinations, a high genetic diversity between the parents is desirable. Here, the genetic diversity of the very few flowering diploids alone could be incorporated in the breeding programme. Accessions derived from one country alone or from crosses between materials from neighboring countries are reported to be not desirable for an effective breeding programme. Generating a very diverse offspring will only be possible if crossing of cultivars from genetically diverse gene pool is carried out (Kreike et al., 2002). By doing so the chance of wild characters will be minimized in the offspring and the improvement of the taro crop can be accomplished in a very short time. Open pollination and selfing Open pollination occurred in fields planted with different genotypes. Depending on the flowering rate, synchrony etc. high open pollinated fruit and seed formation was obtained from several diploid accessions. Self pollination was also successful when manual pollination was done between plants of the same clone. However, it was less successful when mature buds of the same plants were bagged two days prior to anthesis. Hybrid progeny evaluation Mature berries were harvested and the seeds were extracted from the pulp, washed and sun-dried for two days. As taro seeds were non-dormant, the fresh seeds were sown in seed beds made of sand-soil mixture inside the glass house. Seed germination initiated within 8 to10 days and the germination percentage in different cross combinations ranged from 60.5 to 82.3. Time to germination and germination percentage did not show any noticeable difference in the cross, self or open pollinated seeds. Seedlings were transplanted to field within 90 to100 days after germination. Altogether 10,898 seedlings were evaluated during the last six years. The sexual progeny exhibited a wide spectrum of variability with regard to almost all characters: 10-15.5% of the progeny showed wild characters such as stoloniferous root and highly acrid nature of all plant parts. Such types and also other poor yielding non-edible types having acridity were discarded at the seedling stage itself and the rest were carried over for clonal evaluation. More than 4,000 sexually generated clonal progeny are at present under different stages of evaluation. Recurrent back crossing was proved to be difficult due to the poor flowering ability of the cultivars. From the first clonal generation onwards itself potential new cultivars were selected including dwarf types, blight tolerant types, high yielding good cooking quality types etc. Five hybrids having 18.5 to 22.0 tonnes per hectare yield with good cooking quality are being tested in on farm trial prior to release. It is expected that the best two of them will get released for general cultivation soon as the first hybrid varieties of taro from India. 204 third taro symposium Flowering in the sexual progeny Vigorous growth, the continuous occurrence of floral clusters, 4 –5 inflorescences per cluster, well developed spadices, a high proportion of fertile female flowers, an abundance of pollen, an intense odour a day before pollen was released, many insects inside or outside the inflorescence, good seed set and well developed fruit head are reported to be the main indices of good and productive flowering in taro (Ivancic et al., 1996). The sexual progeny showed a regeneration of sexuality with all the attributes mentioned above. This may be due to genetic segregation and the genotypic differences of the sexually derived clones. A similar regeneration of sexuality was observed in another vegetatively propagated tuber crop viz. Greater yam (Dioscorea alata) by Abraham et al. (1985). The progeny thus showed a wide spectrum of genetic variability and the much needed flowering frequency and synchrony of flowering. As all of them were diploids, highly fertile types were abundant for the use as base material for future breeding programme. This sexuality improvement in taro signifies the tremendous scope for the genetic improvement of this species. Polyploidy breeding in taro The natural triploids in taro were found to be significantly superior in yield compared to diploids. With the objective of producing artificial triploids in large scale so as to enhance the frequency of superior yielders in taro, attempts to produce induced tetraploids for interploid crossing for the production of artificial triploids are underway at the CTCRI. Induced polyploidy has been widely used as a potential plant breeding method on account of the tendency of colchiploids to manifest gigantism in plant parts (Stebbins 1950, 1971). In seed propagated plants this has only limited application because of the fall in pollen and seed fertility, due to meiotic abnormalities usual with auto-polyploids (Gottschalk, 1978), which is not a problem in vegetatively propagated crops like taro. Ten desirable accessions based mainly on cormel yield and cooking quality of cormels were selected for the induction of tetraploidy by colchicines treatment. It was successful when 0.2% solution was applied on the emerging shoot tip for 6-8 hours. The occurrence of tetraploids ranged from 0.0-31.0% in the different accessions. However all did not establish in the field. Preliminary evaluation for the identification of the higher ploids was easy from the stomatal size and the stomatal chloroplast count which showed noticeable differences compared to the control. However, the occurrence of tetraploids (2n = 56) was confirmed through cytological screening. Evaluation of the tetraploids revealed that some of them were better in performance compared to the control (Sreekumari, 1993) the result needs further confirmation. Attempts to produce triploids by crossing induced tetraplodis with diploids are underway for which induction of flowering in tetraploids is needed. Conclusions The taro breeding in India is limited to the National level research programme. Structuring the genetic diversity is necessary to optimize the use of germplasm by breeders for which molecular level screening is highly warranted. It appeared that floral attributes are unreliable to classify accessions because variation within each variety is so important that the variation between the two becomes doubtful. Shy flowering, non-synchrony and protogyny were recognized as the major breeding barriers for a planned breeding programme. Large scale production and evaluation of sexually regenerated clonal progeny revealed the tremendous scope for the genetic improvement of the crop through hybridization and selection. Indian agriculture has to be accompanied by intense genetic improvement of the crop for which it is essential to have International co-operation among taro breeders and establishing a procedure for germplasm exchange. References Abraham, A. and K. Ramachandran, 1960. Growing Colocasia embryos in culture. Curr. Sci. 29: 342-343. Abraham, K., S.G. Nair, M.T. Sreekumari and M. Unnikrishnan, 1985. Sexual regeneration in Greater yam, Dioscorea alata L. Tropical Tuber Crops. Nat. Symp. P.65-69. Coates, D.J., D.E. Den and P.M. Gaffeym 1988. Chromosome variation in taro. Colocasia esculenta : Implications for origin in the Pacific. Cytologia. 53: 551-560. Doku,E. V. 1980. Stategies for progress in cocoyam research. In. Terry E. R., K. A. Oduro and F. Caveness (eds.) Tropical Root Crops. Research Strategies for the 1980s. Proc. Triennial Root Crops Symp, ISTRC, Ibadan, Nigeria. Gottschalk, W. 1978. Open problems in polyploidy research. Nucleus. 21: 99-112. Ghani, F.D. 1979. The status of Keladi china (Colocasia esculenta) cultivation in peninsular Malaysia. IFS Provisional Rep. 5: 35-54. Ivancic, A., A. Simin, Y. Tale, 1996. Breeding for flowering ability and seed productivity of taro. In Proc. Second Taro Symp. at Indonesia . G.V.H. Jackson and M.E. Waigu (Eds.) p. 53-57. Jackson, G.V.H., E.A. Ball and J. Arditti, 1977. Seed germination and seedling proliferation of taro. Colocasia esculenta (L.) Schott in vitro. J. Hort. Sci. 52 : 169-171. Jos, J.S. and K. Vijaya Bai, 1977. Stigma receptivity in Colocasia. J. Root Crops. 2(3) : 25-28. third taro symposium 205 Kuruvilla, KK.M. and Avtar Singh, 1981. Karyotype and electrophoratic studies on taro and its origin. Euphytica. 30: 405-413. Kreike, N. van Heck. H. and V.Lebot .2002. Genetic diversity of Taro (Colocasia esculenta (L.) Schott) in South east Asia and Oceania (Communicated). Lebot, V. and K.M. Aradhya. 1992. Collecting and evaluating taro (Colocasia esculenta) for isozyme variation FAO/ IBPGR Plant Genetic Resources Newsletter. 90: 47-49. Plurre, W. K. F. 1984. Clonal variability of taro in the central highlands of Irian jaya. In. Chandra, S. (ed.) Edible aroids. Clarendon Press, Oxford. P. 173-177. Plucknett, D.L., 1970. The status and future of the major edible aroids. Proc. 2nd Symp. Inter. Soc. Trop. Root. Crops. 1: 127-135. Shaw, D.E. 1975. Illustrated notes on flowering, flowers, seed and germination in taro (Colocasia esculenta). PNG. Dept. Agric. Stock Fish Res. Bull. 13: 39-59. Sreekumari, M.T. and P.M. Mathew. 1991 a. Karyotypically distinct morphotypes in taro. Cytologia, 56: 399-402. Sreekumari, M.T. and P.M. Mathew, 1991b. Karyomorphology of five morphotypes of taro (Colocasia escuelnta (L.) Schott) Cytologia 56: 215-218. Sreekumari, M. T. 1993.Cytomorphological and Cytogenetic studies in Edible aroids. Ph. D.Thesis, University of Kerala, Trivandrum, India Sreekumari, M. T. and P. K. Thankamma Pillai, 1994. Breeding barriers in taro (Colocasia escuelnta (L.) Schott. J Root Crops 20 (1): 20-25. Stebbins, G.L., 1950. Variation and evolution in plants. Colombia University Press. NewYork. Stebbins, G.L., 1971. Chromosomal evolution in higher plants. Addison-Wesley Pub. Co., Melnopark, California. Strauss, M. S, G. C. Stephens, J. Gonzales and J. Arditti, 1980. Genetic variability in Ann.Bot.43: 603-612. Tanimoto, T. and T. Matsumoto, 1986. Variations of morphological characters and isozyme patterns in Japanese cultivars of Colocasia escuelnta Schott and Colocasia gigantia Hook Japanese Journal of Breeding , 36: 100-111. Unnikrishan,M., P.K.Thankamma Pillai, K.Vasudevan, G. G. Nayar, J.S.Jos, M. Thankappan and M.S.Palaniswami. 1987. Genetic Resources of taro. Tech. Bull. Series 8. CTCRI,Trivandrum. Unnikrishan,M., P.K.Thankamma Pillai and K.Vasudevan, 1988. Evaluation of genetic resources of taro (Colocasia escuelnta (L.) Schott). J.Root Crops.14 (1): 27-30. Velayudhan, K. C.and V. K. Muralidharan, 1985. Variability in a collection of Colocasia from the wild. Proc. Nat. Sypm. Trop. Tuber Crops- Production and Utilization. CTCRI, Trivandrum. Yen, D. E. and J.M. Wheeler, 1968. Introduction of taro into the Pacific: The indications of chromosome numbers. Ethnobotany, 7 (3): 259-267. 206 third taro symposium Theme Four Abstracts THEME 5: Product Development and Marketing THÈME 5 : Développement de la production et commercialisation Pacific taro markets: Issues and challenges Grant Vinning and Joann Young Le marché du taro dans le Pacifique : réalités et défis Grant Vinning et Joann Young Intra-Pacific trade in taro began in earnest in the early 1950s when Fiji commenced exporting to New Zealand. Samoa followed in 1957 with other Pacific Islands following suit. The Australian market was explored soon after. Whilst New Zealand and Australia continue to be the destination of choice for South Pacific taro exporters, for two very different reasons these markets cannot be taken for granted. At the same time there are two other Pacific taro importers whose markets dwarf New Zealand and Australia - Japan and the United States. For these two markets to be developed to the full potential, South Pacific taro exporters will need to take a different approach to what they have shown in the past. La commercialisation du taro entre pays océaniens a réellement commencé au début des années cinquante, lorsque les Îles Fidji se sont mises à exporter vers la Nouvelle-Zélande. Le Samoa a suivi cet exemple en 1957, puis d’autres pays océaniens. Le marché australien a été prospecté peu après. Bien que la Nouvelle-Zélande et l’Australie restent des destinations de prédilection pour les exportateurs de taro du Pacifique sud, ces marchés ne sont pas définitivement acquis, pour deux raisons tout à fait différentes. Il existe en même temps deux autres importateurs de taro océanien dont les marchés relèguent au second plan ceux de Nouvelle-Zélande et d’Australie : le Japon et les États-Unis d’Amérique. S’ils veulent exploiter tout le potentiel de ces deux marchés, les exportateurs de taro du Pacifique sud devront adopter une autre approche commerciale que celle qu’ils ont appliquée jusqu’à présent. Value added products from taro Valorisation des produits fabriqués à partir du taro Richard Beyer The key to increasing root crop consumption in general and taro in particular is through a programme of product development. Taro is a particularly suitable raw material for product development since it is bland and essentially without colour. Not only will value adding overcome the buying disincentives suffered by the raw vegetable but will provide an opportunity to add value and increase the return to local growers and entrepreneurs. By building in appropriate features to developed products it is possible to match the features of imported items and thus reduce the expenditure on imported foods. In addition, processing provides an opportunity to overcome the inconsistencies and vagaries of quarantine inspection services. Taro production and value adding in Palau Robert Bishop Taro in Palau dates back to the misty past. Taro is a prominent and identifying component of Palau’s culture. The traditional system of utilizing wetlands to produce taro is ancient, distinctive, rich and varied. Palau has a vast reservoir of traditional knowledge and skills related to taro. The three main types of taro grown in Palau are Colocasia, Cyrtosperma and Xanthosoma. The number of varieties of Colocasia currently present in Palau is estimated at about 100. Palau Community College – Cooperative Research Richard Beyer Si l’on veut augmenter la consommation d’un légumeracine en général, et du taro en particulier, il faut appliquer un programme de valorisation des produits. Le taro constitue une matière première particulièrement bien adaptée à la valorisation des produits car il est incolore et d’un goût neutre. La valorisation du taro lèvera les réticences de l’acheteur devant le légume cru tout en offrant la possibilité d’une valeur ajoutée et d’une augmentation des bénéfices des cultivateurs et des entreprises locaux. En incorporant des caractéristiques appropriées aux produits mis au point, on peut égaler les qualités des produits importés et réduire ainsi les dépenses d’importation. En outre, la transformation du taro permet de surmonter les incohérences et aberrations des services de contrôle phytosanitaire. Production et valorisation du taro à Palau Robert Bishop L’origine du taro à Palau remonte à un passé lointain. Il occupe une place primordiale et caractéristique dans la culture de Palau. Le système traditionnel, qui consiste à exploiter les terres humides pour cultiver le taro, est à la fois ancien, particulier, riche et varié. Les habitants possèdent de vastes connaissances et savoir-faire en matière de culture du taro. Les trois principaux types cultivés à Palau sont Colocasia, Cyrtosperma et Xanthosoma. On estime à une centaine le nombre de variétés de Colocasia third taro symposium 207 & Extension, maintains sixty-eight varieties. The number of varieties is rapidly dwindling. The number of varieties available during traditional times has been estimated as over 200 (least) and over 400 (most). Since traditional times, few taro varieties have been introduced from the outside, the most notable exception being varieties introduced by SPC due to their resistance to taro rot and salt water. In everyday diet, taro is being gradually replaced by rice and cassava. à l’heure actuelle à Palau. Le Collège communautaire de Palau (département de recherche et de vulgarisation en coopération) entretient 68 variétés, mais ce chiffre est en train de diminuer rapidement. Le nombre de variétés qui existaient autrefois se situait dans la fourchette des 200 à 400. Depuis lors, peu de variétés ont été introduites de l’extérieur, l’exception la plus remarquable étant celles qui l’ont été par la CPS et qui résistent bien à la pourriture et à l’eau salée. Dans la vie quotidienne, le taro est progressivement supplanté par le riz et le manioc. Recent developments on taro-based Évolution récente des produits alimentaires fabriqués à partir du taro food products in Hawaii Alvin S. Huang, Karthik Komarasamy and Lijun He Several food products containing taro as the main ingredient have been developed in Hawaii in recent years. These products were developed with dry-land taro as the raw material and the local and tourist market as the main focus point. The developmental strategy is a result of the current economical and environmental emphasis. Taro yogurt will be discussed in greater detail, as an example of how a traditional taro staple can be transformed into a health food. Other new taro products are briefly introduced to highlight taro’s diverse functions in each application. Chemical composition and effect of processing on oxalate content of taro corms E. O. Afoakwa, S. Sefa-Dedeh and E. K. Agyir-Sackey The chemical composition as well as the effect of processing on the corms of two Xanthosoma sagittifolium species and Colocasia esculenta corms was evaluated. A 3 × 3 factorial experimental design with cocoyam varieties Xanthosoma (white-flesh), Xanthosoma (red-flesh) and Colocasia, and corm section distal, middle and apical, was performed to determine the chemical composition of the corms. Oxalate content of the various corms was also evaluated and the effect of processing assessed using standard analytical methods. The mean values of the proximate composition of the three cocoyam species evaluated were: crude protein 2.98-5.50 g/100 g, total fat 0.28-0.97 g/100 g, ash 1.56-2.98 g/100 g, starch 12.23-36.04 g/100 g and crude fibre 1.11-3.00 g/100 g. The apical section of all the species had high protein content while the distal section had high levels of ash, fibre and minerals. Potassium, zinc, magnesium and phosphorus were the most abundant minerals. Oxalate compositions of the fresh samples were in the range of 253.49-380.55 µg/100 g for the Xanthosoma sagittifolium (red-flesh), 302.19-322.82 µg/100 g for the Xanthosoma sagittifolium (white-flesh) and 328.4-459.85 µg/100 g for the Colocasia esculenta. No significant differences (p≤0.05) were found between the oven-dried and solar-dried samples. However, drum drying reduced the oxalate levels by approximately 50% to average levels ranging from 99.94-191.16 µg/100 g, implying that solar, oven and drum drying techniques can be used for the development of marketable dehydrated products from taro with reduced oxalate contents. 208 third taro symposium à Hawaii Alvin S. Huang, Karthik Komarasamy et Lijun He Plusieurs produits alimentaires à base de taro ont été mis au point à Hawaii, au cours des dernières années. Ces produits, qui ont pour matière première du taro des terres sèches, s’adressent principalement au marché touristique. La stratégie de développement a été élaborée à la lumière des objectifs économiques et écologiques qui sont poursuivis actuellement. L’exemple du yaourt au taro sera décrit en détail ; il montre comment une denrée traditionnelle de base peut être transformée en aliment sain. D’autres nouveaux produits à base de taro sont brièvement décrits pour mettre en lumière les diverses fonctions que remplit le taro pour chacune des applications. Composition chimique des cormes de taro et effets de la transformation sur leur teneur en oxalate E.O. Afoakwa, S. Sefa-Dedeh et E.K. Agyir-Sackey Nous avons évalué la composition chimique ainsi que les effets de la transformation des cormes de deux espèces de Xanthosoma sagittifolium et de Colocasia esculenta. Un dispositif expérimental, établi selon un plan factoriel 3x3 [variétés de taro d’eau (Xanthosoma (à chair blanche), Xanthosoma (à chair rouge) et Colocasia) et section du corme (distale, mésiale et apicale)], a été mis en place pour déterminer la composition chimique des cormes. L’expérience a également permis d’évaluer la teneur en oxalate des différents cormes et les effets de leur transformation, grâce à des méthodes d’analyse normalisées. Les valeurs moyennes obtenues après analyse de la composition des trois espèces de taro d’eau sont les suivantes : protéine brute 2,98-5,50 g/100 g, matière grasse totale 0,28-0,97 g/100 g, cendres 1,56-2,98 g/100 g, amidon 12,23-36,04 g/100 g et fibre brute 1,11-3,00 g/100 g. La section apicale de toutes les espèces dénote une teneur élevée en protéines alors que la section distale contient une forte quantité de cendres, de fibres et d’éléments minéraux. Le potassium, le zinc, le magnésium et le phosphore sont les minéraux les plus abondants. La teneur en oxalate des échantillons frais varie entre 253,49 et 308,55 µg/100 g pour Xanthosoma sagittifolium (à chair rouge), 302,19 et 322,82 µg/100 g pour Xanthosoma sagittifolium (à chair blanche) et 328,4 et 459,85 µg/100 g pour Colocasia esculenta. L’expérience n’a pas révélé de disparités significatives (p<0.05) entre les échantillons séchés au four et ceux séchés au soleil. Néanmoins, le séchage en fût réduit d’environ 50 pour cent la teneur des cormes en oxalate, qui s’établit en moyenne entre 99,94 et 191,16 µg/100g. Les techniques de séchage au soleil, au four et en fût sont donc toutes trois adaptées à la fabrication de produits déshydratés à base de taro, de qualité marchande et présentant une teneur réduite en oxalate. third taro symposium 209 Theme Five Paper 5.1 Pacific taro markets: issues and challenges Grant Vinning1 and Joann Young2 Asian Markets Research, Brisbane, Australia Ministry of Agriculture, Sugar and Land Resettlement, Fiji 1 2 Introduction Intra-Pacific trade in taro began in earnest in the early 1950s when Fiji commenced exporting to New Zealand. Samoa followed in 1957 with other Pacific Islands following suit. The Australian market was explored soon after. Whilst New Zealand and Australia continue to be the destination of choice for South Pacific taro exporters, for two very different reasons these markets cannot be taken for granted. At the same time there are two other Pacific taro importers whose markets dwarf New Zealand and Australia - Japan and the United States. For these two markets to be developed to the full potential, South Pacific taro exporters will need to take a different approach to what they have shown in the past. New Zealand Whilst Fiji initiated taro exports to New Zealand in the early 1950s it was Samoa who more fully developed this market. Following the 1963 severe floods in Fiji, Samoa increased its production to fill the gap. Samoan exports collapsed in 1993 due to the devastation caused by the Taro Leaf Blight. Following the Samoan industry’s collapse, Fiji has recaptured the New Zealand market and now has more than 80 percent of the market. The New Zealand Samoans are the country’s dominant Islander group. Their preference is for the “Samoan Pink” variety. When Fiji stepped up production in the early 1990s, to maintain the Samoan market it called its taro Tausala ni Samoa. (Ironically, in Samoa the variety is called “Taro Niue”.) New Zealand imports around 6,000 t annually, reflecting the country’s sizeable Pacific Islander population. Moreover, this population is expected to double in the next 30 years. Detailed analysis is hampered by the lack of consistent import data and no wholesale data. Available data show that a surprising number of countries have exported to New Zealand. In the period 1992-2002, a total of 16 countries are recorded as taro import origins. A number of these origins would have been expected: Fiji, Tonga, Samoa, Niue, New Caledonia, Cook Islands, American Samoa. Some of the other suppliers are a little surprising but at least they have a taro producing tradition: Korea, Taiwan, China, Vietnam, and Philippines. Three are most surprising – Australia, Egypt, and Saudi Arabia. Explicit price and quantity import data are available for the five years to 2001. This shows that the volume of taro imports into New Zealand has been consistent around 6,000 t. In 2001 imports were 6535 t at an average CIF price of NZ$1.78. New Zealand taro imports Imports (t) CIF price (NZ$/kg) 1998 5674 1.54 1999 6516 1.55 2000 6331 1.57 2001 6535 1.78 It is a common hypothesis that second generation emigrants start to move away from their traditional foods / cuisines and instead consume the cuisine of their adopted country. Despite this, taro in New Zealand is clearly still an important part of Pacific Islander culture and cuisine. Whilst there will always be a demand for taro, it is an extremely price sensitive market because there are numerous alternatives in the form of plantains, sweet potato kumala, and even potatoes. When retail prices are between NZ$2.00-3.00/kg, demand for taro is strong. At these prices, Pacific Islanders will eat dalo several times a week. Above NZ$3.00/kg retail, demand starts to decline. At NZ$4.00/kg taro demand virtually dries up because taro ceases to be an everyday meal item. Consumption becomes limited to special occasions, such as the Sunday feast and other community-cultural events. The New Zealand market cannot be guaranteed. In mid-2002, taro mite Rhizoglyphus minutus was discovered by New Zealand Quarantine on a taro shipment from Fiji. The mite is a microscopic organism that attaches to the lower half of the corm. From Fiji’s perspective, the mite is not a pest as such because it does not damage the corm. More-over, Fiji has argued that the mite is found in almost all the islands exporting taro to New Zealand, and they have been reported to be present in New Zealand. Nevertheless, 210 third taro symposium the mite is a Regulated Quarantine Pest in New Zealand. As such, all taro imported from Fiji into New Zealand has to be fumigated with methyl bromide. This significantly reduces the products shelf life. At the time of writing, the Secretariat of the Pacific Community had instigated a major project to establish if the taro mite being imported is the same as the mite currently present in New Zealand. If it is the same type, then no further action will be taken. However, if it is a different type, then it is likely that New Zealand will require disinfestation for the mite to be undertaken before shipment from Fiji. When account is made of the transit time, this will dramatically reduce taro’s shelf life that will have major repercussions for the Fiji’s taro trade with New Zealand. Australia Despite its much greater population, Australia is a vastly smaller market for imported taro. The obvious reason is that Australia’s Pacific Islander population is much smaller than that of New Zealand. At the same time Australia is rapidly developing its own taro production. Australia produces three types of taro. All are Colocasia esculenta, one being the traditional Pacific – type taro of around 1.0 kg in size; the second being the vastly smaller 60 g type preferred by the Japanese; and the third being the 150 g smooth - skinned type preferred by the Vietnamese. This paper will use the terms Taro Pacific, Taro Supreme, and Taro Vietnam to describe these types (see Asian Markets Research, in press). Taro Pacific appears to have arrived in Australia with the Chinese joining the Gold Rushes and a little later with the Kanakas when they were blackbirded to work in the Queensland sugar cane fields. Today in North Queensland is called, variously, “Chinese taro”, “bun long” and “purple taro”. In a number of the creeks and gullies of the hilly country in the hinterland of northern New South Wales, and along the Queensland coast, taro grows as a feral plant. Taro planting material continues to be imported in an undocumented manner. In the mid-1980’s a considerable volume of Samoan Pink and other taro varieties were imported in order to generate material locally for a research project based at The University of Queensland investigating nutrition and diseases of root crops vital to the Pacific. The imports were imported pursuant to all quarantine protocols. After the project had developed enough planting material, planting materials were given to the cooperating growers and the Queensland Department of Primary Industries. It is estimated that Australia produces around 1,000 t of Pacific Taro. Nearly all of this is grown in pockets along the east coast north of just south of the Queensland-New South Wales border by Pacific Islanders, Chinese, Vietnamese and Cambodians, and an increasing number of Australians. The Queensland-based Taro Growers Association has over 40 Australian members. A great deal of the production in North Queensland is marketed in Sydney and Melbourne with some produce being shipped as far away as Perth. The volume marketed through the Brisbane wholesale market is vastly smaller, largely as a result of the existence of a large back-yard growing industry and an extensive informal marketing systems that is essentially church-member based. Australian taro production has received two major boosts over the past decade. One was the material and knowledge that came out of The University of Queensland nutrition and disease project. The second is another university-based project. Central Queensland University has received significant Rural Industry Research and Development Corporation funding to facilitate the development, inter alia, of a Taro Supreme industry focused on the Japanese market (Hassalls and Associates 2002). Many of the findings of this project – fertilizer usage, pest and disease control – have immediate application to the Taro Pacific industry. Australia currently imports around 3,000 t of taro. Whilst this is comprised principally of Taro Pacific from the Pacific, a recent survey (see Hassals and Associates, 2002) reported that Taro Supreme is imported into Australia from China as (a) frozen, peeled and stand-alone, (b) peeled in brine, (c) frozen, peeled and with other products, principally burdock Arctium lappa, lotus Nelumbo nucifera, and bamboo Dendrocalamus latiflorus. Australia’s move towards self-sufficiency in taro will continue. Wholesale prices at the moment are encouraging, especially for Taro Supreme. It is considered that Pacific exporters should no longer count on Australia as being an assured long term market. Indeed, it may be best if they viewed Australia as a potential rival for at least the New Zealand market. It is noted that over the past 18 months a number of exporters from Fiji as well as some New Zealand taro importers have visited North Queensland to assess the potential of exporting taro from there to New Zealand. Other taro markets in the Pacific Taro is a crop with a long tradition in Pacific Asia and has a significant presence in the region. China, for example, produces over 12 million tonnes. The Pacific contains a number of taro importers. The biggest market by far is Japan, followed by the United States, then Canada, with smaller volumes going into Hong Kong and Singapore. Japan Taro sato imo is a traditional Japanese crop. Whilst six types are commonly recognised, three dominate: • Ishikawa-wase: only the daughter tuber is used. By Taro Pacific standards, this is an exceptional small taro, usually around 60 g. • Dodare: only the daughter tuber is used; a little larger than Ishikawa-wase. • ereves, where the mother and daughter tubers are used. Sato imo production in Japan had consistently trended downwards since the high levels of around 500,000 t in the 1960s to 258,000 t in 1999. The reasons for the decline in taro production are common across a large number of third taro symposium 211 Japanese agricultural crops: and aging farmer population and the limited ability to off-set labour shortages through mechanisation due to the small plots involved. Throughput at the Tokyo wholesale market system has declined from 25,000 t in 1987 to 14,619 t in 2000. Japan imports sato imo in both the fresh and processed forms are consistent. As processed, taro is as stand-alone product and as a mixed vegetable with gobo, renkon, and takenoko in both brine and frozen forms. Data are available for fresh and frozen products. The data show a consistent growth pattern notwithstanding the sudden decline in fresh imports in 1997. China supplies over 90 percent of the estimated total imports of more than 100,000 t fresh equivalent. Japan sato imo imports (t) Fresh Frozen Total 1995 26862 n.a. 26862 1996 25643 58662 84305 1997 c5643 53615 59258 1998 c6148 52043 58191 1999 10321 51861 62182 2000 20344 55873 76217 2001 20254 55012 75266 CIF prices for fresh sato imo show a distinct rise from May to around September, the same period when domestic sato imo wholesale prices also rise. (Import data from Japan Tariff Association. Wholesale data is based on Dodare and Ishikawa wase and comes from Tokyo Metropolitan Government.) This is the Japanese and Chinese summer, clearly presenting a market window for southern hemisphere suppliers. Japan sato imo fresh monthly imports (yen/kg): 1997-2001 Japan sato imo monthly wholesale price, Tokyo (yen/kg): 1995-2000 United States Taro production in the United States is centred in Hawaii where it has declined over the past thirty years to just 3 975 t in 2000. Production of poi taro, the fermented mixture that comes from the pounding of a specific variety of Colocasia esculenta, is about ten times the volume of fresh or “Chinese” / “white type”. Taro is also grown in Florida: it is mainly Xanthosoma the preferred variety of Cubans. Data is not recorded. Imports of taro into the United States have shown remarkable growth both in terms of volume and CIF prices. United States taro annual imports (left=tonnes, right=$US/kg): 1981-1999 Given these growth figures, it is little wonder that the United States has been a favourite export destination. In the 20 years over which the previous graph was constructed, more than 30 countries have supplied taro. Costa Rica and the Dominican Republic provide over 90 percent of imports. Jamaica is the price leader by an exceptional margin. China, despite being a comparatively small supplier, is a price maverick, exhibiting a behaviour that indicates that its desire 212 third taro symposium for foreign currency is greater than the need to consistently build a market presence. (Information provided by the price reporting service of the United States Department of Agriculture, Los Angeles’ Seventh Street Wholesale Market.) Taro has three markets in the United States that can be categorised in terms of ethnic groups and taro types: • Pacific Islanders: pink and white Taro Pacific • Chinese and Hispanic: white Taro Pacific and Taro Supreme • Hawaiians: white and pink Taro Pacific, and, exclusively, poi There is little doubt that the Islanders, especially the significant Samoan population based around Los Angeles, preferred the Samoan Pink. Nevertheless, following the Taro Blight in 1993 when this trade all but vanished, there was a willingness to accept the white types. Potential exporters to the United States would be better off studying the techniques that Jamaica uses that make it the price leader rather than the price behaviour of the Chinese. (This includes conditioning of the harvested product, dipping, and packaging.) Canada Until the late 1980s, Canada was a solid, albeit not great, market for Islander taro. The trade centred on Vancouver because of its Islander population and more direct transport links to the region. As will be noted below in more detail, this market was destroyed by the non-commercial private trade that exploded after the first Fijian coup of 1987. The non-commercial trade can be described as between family member in Fiji and Canada plus other members of the Pacific Islander taro eating community often selling at below market prices. Other Pacific importers Within the Asian-Pacific region, taro is imported into Hong Kong and Singapore. Official data does not exist. Taiwan produces around 50,000 t, significantly down from the 80,000 t it produced at the beginning of the decade. Unlike in Japan and the United States, taro is a relatively low-priced item in Taiwan. Even at the peak months – which are highly erratic and inhibit clear interpretation – wholesale prices are still less than US$1.00/kg. Market development Three issues needed to be addressed if South Pacific taro producers are to capitalise on the market opportunities identified above. Varieties The Japanese market concentrates on the Dodare and Ishikawa wase varieties. These types are significantly smaller to what most Pacific Islanders are used to: Domestic shipping grade standards for Ishikawa wase and Dodare types of sato imo Type 2L L M Ishikawa –wase 60 g 40-60 g 20-40 g Dodare 90 g 60-90 g 30-60 g To use smaller varieties of locally available plants, e.g. dalo ni tana in Fiji, would not meet Japanese requirements. To access this market, genetic material would have to be imported and adopted to local conditions. The Kingdom of Tonga has already started this process. Frozen product Frozen taro addresses two major issues handicapping export market development: shipping and quarantine. Maintenance of the cool chain is always a problem in the Pacific with less frequent calls and the need to tranship. It is for this reason that frozen product, albeit processed, should be considered. Despite the apparent contradiction, frozen product is more easily handled than fresh as there are clear protocols and well established procedures. Being frozen and par-processed, that is at least peeled, the product escapes the rigorous attention usually accorded to a soil-based fresh product that already has a poor quarantine image. Frozen Taro Pacific is currently imported into New Zealand, Australia, and the United States with varying success. Whilst the market preference is for fresh, expatriate Islanders in the three markets have shown that they have adapted to the local pace of life and that convenience has a higher priority compared with their countries of origin. Provided the product is peeled and cut into plate size pieces that can be boiled, or preferably micro-waved, then there is a market. third taro symposium 213 Private marketing One of the major issues bedevilling taro exports from the South Pacific has been private marketing. This is trade that, whilst perfectly legal, seriously handicaps commercial trade. Private trade is where the taro is sent on either a free or possibly transport-costs recovery basis. Common examples are from family groups in the Islands to family groups in the importing country, and church groups in the exporting country to church groups in the importing country. In Niue, where taro is one of the very few export income generators, in any shipment to New Zealand up to 50 percent of the total is being given away to family members: why should families buy Niue taro, allegedly the prized taro of the Pacific, when they can get it for free. Similarly, the Otaro, Avondale and Mangere Markets in Auckland New Zealand are supplied nearly solely by free taro. This product competes directly with commercial shipments that must recover not only transport costs but also purchase prices (Vinning, 2002). A variation on this was seen in the Vancouver market just after the 1987 Fiji coup. Residents who fled the country received taro on a non-purchase price basis: whatever monies they received from the sale of the product was kept in Canada, effectively enabling them to export money. Unable to compete with such low priced competition, the commercial trade collapsed soon after (Vinning, 1998). A comparable behaviour was noted in San Francisco in 2000, again after the 2000 Fiji coup. However the size of the market has limited the impact of the practice. The vexed question regarding the private trade is just how far does a government go in protecting income earning exports against a well established social action. Other countries have trade based on citizen-to-citizen action as distinct from company-to-company action but these have always been within the parameters of requiring an export license and the concomitant transparent paper trail of monies. One of the advantages of the more distance Pacific markets of Japan and the United States it that the distances involved may militate against the widespread practice of private trade, notwithstanding that it does occur in the United States market. Food safety There are increasing incidents of disrupted trade in taro due to developed markets imposing more stringent food safety (sanitary) measures on Fijian and other Islander exporters. As small developing countries, the Pacific Islands would like to see that food safety standards are based on international standards, guidelines and recommendations such as Codex. It becomes too costly for the Islander exporters to comply with different sets of standards for different countries. However, it should be recognised that all Islander exporters face difficulties in implementing and complying with Codex standards. Under Article 9 and 10 of the World Trade Organization Agreement on Sanitary and Phytosanitary Measures, they can request for technical assistance and longer time frames to adjust and comply with Codex standards. Conclusion South Pacific taro exporters face a paradigm shift in their industry. Previously assured traditional markets can no longer be taken for granted. At the same time, there are vastly bigger markets out there in the Pacific. To assess the new markets the Pacific Islands must adopt a plethora of new procedures. These include new varieties and new transportation arrangements. There is a significant need to change the approach towards taro production by giving it the same level of scientific approach as they give towards introduced crops (see Gonemaituba and Young, IN PRESS). There is the need to recognise that the rules of trade have changed and that in the new WTO world non-tariff barriers in the form of sanitary and phytosanitary standards will play a large role. Governments must also address the private trade issue if the commercial trade that brings in crucial foreign income is to survive. The new markets are large. In most cases they are quite remunerative. Effort will be required to successfully exploit them. It is argued that the effort is worth it. Acknowledgements This paper draws on Asian Markets Research (2003). Research for that was funded partially by two Rural Industries Research and Development Corporation (Australia) projects, UCQ-13A and NAME PROJECT NUMBER. Insights for this paper draw on projects funded by the FAO in 1994 and 2002, UNDP in 2002, Asian Development Bank (1997 –1998), and by Asian Markets Research from 1995 onwards. Our also thanks to the then price reporting service of the United States Department of Agriculture, Los Angeles’ Seventh Street Wholesale Market for the price information on the US market. References Gonemaituba, W. and Young, J. Fiji. Asian Markets Research. In print. Hassall and Associates Pty Ltd. 2003. Asian vegetable industry: A situation assessment. Rural Industries Research and Development Corporation, Canberra. 82 p. Vinning, G.S. 1998. Management of the diversification of Fiji’s agricultural economy. Asian Development Bank, Manila. Vinning, G.S. 2002. The marketing of primary products from Niue. FAO CST-NIU 26/3/02. Vinning, G.S. 2003. Select markets for taro, sweet potato and yam. Rural Industries Research and Development Corporation, Canberra. 90 p. 214 third taro symposium Theme Five Paper 5.2 Value added products from taro Richard Beyer Food Scientist, Suva, Fiji Introduction The consumption of all root crops is declining throughout the Pacific region by about 10-15% per annum. This is not particularly surprising since the root crops in general, and taro in particular, are heavy, frequently dirty and are inconvenient to prepare. As lifestyles change, more women enter the workforce and the time for family dining dwindles, food choices turn toward items that are more flavoursome, single serve, and which are more appealing to all family members. The exploitation of taro as a base for the manufacture of convenience foods is long overdue. It is bland tasting, pale in colour and it is thus an excellent base to which colours, flavours and texture modifiers can be added. In addition, processing provides an opportunity to eliminate anti-nutritional and unacceptable components such as oxalate. The market for added value product from taro has been tainted somewhat by a number of substandard products such as rancid taro chips which have such a high concentration of oxalate that they have only limited appeal. Nevertheless there is enormous potential to increase the value of taro products not only by producing high quality “me too,” products but also innovative products which are completely new in their concept. Because taro is somewhat more difficult to grow than cassava and the yields are slightly lower than that of cassava, it is commonly more expensive than cassava and thus it is consumed less frequently. It has a reasonably long shelf life and thus requires little intervention to retard deterioration. It has sufficient shelf life to reach most urban and some international markets in good condition. Hence taro commands fairly consistently high prices and with modern postharvest handling techniques there is little loss. Therefore, there has been relatively little incentive to add value, or to base a processing industry on taro as a raw material. However exporters from Pacific Island Nations (PINs) report that there has been a significant shift in the attitude of the Australian Quarantine Inspection Service (AQIS) and quarantine services in New Zealand. The frequency with which they insist on the fumigation of fresh taro as it enters Australia has increased – in some cases it is fumigated because of the presence of Acarina – or carpet mite. In New Zealand all shipment of taro are being fumigated. Fumigation significantly reduces shelf life from several weeks to a few days. Declining consumption Influences such as technology, improved communications including television, national and international travel have made significant impacts on food choice. Urbanisation has distanced many indigenous people from their immediate food sources. Formal employment and denser habitation in urban centres make gardening more difficult. Local foods have been displaced by western foods, which tend to be more energy dense and are considered to have increased the rate of obesity. In Vanautu (Vanuatu Department of Health Report, 1996) and in PNG (Hodge et al., 1997; Bourke, 1982) for instance, the rate of obesity is directly linked to involvement in the cash economy with civil servants most prone to excessive weight. Suburban dwellers are more likely to indulge in high-density diets and rural dwellers least likely. The Fiji National Food and Nutrition Centre (FNFNC) (FNFNC, 1997; FNFNC, 2000) data has demonstrated enormous increases in the choices of food. Even in remote communities the average number of foods in the entire food library has risen from a total of 21 foods – all locally grown – in 1952 to 107 in 1994. Such a large increase has resulted from the advent of village stores, which have increased from seven in 1956 to 3,547 in 1996 (Fiji Bureau of Statistics, 2000). In one store that was asked to record purchases on a weekly basis 200 kg of flour, 90 kg of wheat-based noodles and breakfast cereals, 80 kg of biscuits, and 40 kg of potatoes were purchased. Such purchases have probably displaced root crops, which would have been consumed four decades earlier. In the later survey (FNFNC, 2000) only 8% of dietary energy was reported to have originated from root crops and 41% were derived from cereals. This is a significant drop from an estimated 85% energy intake from root crops in the 1950s (Parkinson, pers. comm.). Many influences have been identified but it is largely a matter of convenience since rice is lighter, cleaner, more convenient and cheaper than taro. In Fiji, the Indian Diaspora has increased the consumption of flour-based breads. Throughout the region Two Minute NoodlesTM has satisfied a growing demand for single serve foods resulting from a gradual disintegration in the frequency family dining as sport and other activities distract the young. Traditional preservation of taro In more isolated regions, taro is the raw material for the traditional fermented Polynesian product – poi (Cable, 1981; Cable, 1982). Corms are sliced or shredded and cooked until soft. The cooked product is strained through cloth in which third taro symposium 215 it is retained and allowed to ferment. The product sours as lactic acid levels increase as a result of homofermentative lactic acid bacteria activity. Historically sea and air links to isolated islands were infrequent and unreliable. Poi was an important alternative food source during periods of isolation, inclement weather patterns or low food production seasons. More recently the ready availability of a plethora of food types, the spread of television as an advertising medium and increased travel has directed food choice toward those which are more immediately appealing. The consequence is that the consumption of poi has become increasingly more restricted and is now largely confined to traditional and ceremonial feasts and to consumption by older generations. In order to attract consumers back to the root crops in general and taro in particular then a series of features must be included in any value-added products. Product development approach At the start of any product development programme it is important to evaluate the features and the benefits that follow of any product that is proposed. Taro suffers the significant disadvantage that it contains calcium oxalate. Oxalates are present in all forms of living matter and usually occur as sodium ammonium salts that are soluble. Among root crops, insoluble calcium oxalate is more common. As an insoluble salt it does not contribute to the osmotic concentration of the tuber and thus is an efficient method of calcium storage. Calcium oxalate forms insoluble needles or raphides, and they are potent irritants. They puncture the skin and cause irritation either singly or in combination with a proteolytic enzyme (Bradbury and Nixon, 1998). They may have evolved as a defence. The ingestion of high levels of oxalate may cause corrosive gastroenteritis, shock, convulsions, low plasma calcium and high plasma oxalate, renal damage and ultimately renal failure. Of the root crops grown in the PINs, giant taro and taro contain the highest levels and the concentrations are given in Table 1. Table 1: Oxalate concentrations of giant taro and Xanthosoma spp Giant taro (mg 100-1 g) Xanthosoma spp (mg 100-1g) Skin Anatomical location 310 Not consumed 10 mm beneath the skin 135 86 - 139 58 64 - 106 20 mm beneath the skin Core 74 - 112 Bradbury and Holloway (1988) It is difficult to reduce the levels of calcium oxalate in taro and giant taro. Baking and earth oven cookery tends to exacerbate the problem since moisture loss will increase the concentration. During processing, soaking in 2-3% brine for periods of up to four hours will be sufficient to replace the calcium with sodium. Sodium oxalate is more soluble and is removed during subsequent rinsing. Unfortunately the sodium will only penetrate a very short distance into the intact taro tissue. Beyer (2001a) reports that oxalate can be reduced to levels that fall below detectable limits in taro cut into 4mm shoestring strips after immersion in 3% saline prior to draining and frying. For products that are based on mashed taro, (e.g. extruded snack foods) soaking grated taro in saline solution will remove most calcium oxalate. Having overcome the disadvantage of oxalate removal, there it still becomes important to build in features which will persuade the consuming public to buy and use taro. For some, ideas tend to be generated on the basis of experience and as a result of inspiration. For others, the process of establishing a new product portfolio is more a process of deliberate thought and systematic examination of existing, successful products. A number of simple techniques can be used to generate ideas. Fundamental to generating ideas is a very clear understanding of the circumstances under which products are consumed. Since the new product is designed to attract a replete consumer, it must displace an extant commodity – the competition. By identifying the competing food then it is often less difficult to incorporate features which will persuade the consumer to abandon the existing product in favour of the new one. Ideas come from a variety of sources - many ideas will not reach commercial reality. For the entrepreneur, the process of idea generation must not stop. Simply by observing what consumers buy in markets and supermarkets and noting the eating habits of others, ideas begin to arise. There are a number of ways of generating ideas. 1) “Me too” product development “Me too” product development is simply a copy of an existing product. As entrepreneurial activity has gained momentum, there are an increasing number of products that can be copied. Frozen taro, cassava and, more recently, sweet potato are now established products. Direct copies can be made of those products. For the entrepreneur embarking on a food industry, there must be markets for theses existing products – the challenge is persuading the existing consumers to change allegiance. Consumers are familiar with the product so that less expenditure is required for launching and educating consumers than it would be for an entirely new product. 216 third taro symposium 2) Modified “me too” This is simply a modification of a product that is already successfully established in the market. Local analogues of existing products are a rich source of ideas for product development. For the PIN entrepreneur, all the potato and cereal products are potential templates for the production of root crop, plantain or breadfruit analogues. They may indeed be products that are essentially the same as the template product with a partial replacement of one or two of the ingredients. Such products have the advantage that they use local products in forms that have proven market acceptability. Since they are already familiar to consumers, they do not require the advertising support of totally new products. Furthermore, they reduce the dependency on imported foods. 3) Improved convenience of an existing product Frequently existing products are not as successful as they might be, because they are inconvenient. The root crops, for example, are far from convenient. They are heavy, sometimes dirty and require peeling and cooking. Yams may be very large, taro is difficult to peel and the latex from breadfruit may give it an unfortunate appearance. Many products have been more successful than their competitors because some steps in the kitchen preparation have been included in the new product. As a simple exercise, those seeking ideas are invited to survey a supermarket and examine the vast number of products that they do not purchase. After establishing the reasons which deter purchase, it becomes possible to define what changes are needed in that product in order to increase its convenience and encourage more frequent use. Convenience can be added by: • Altering the package size (single serve, family pack etc). • Altering the packaging design to improve convenience (re-sealable pack, ring pull can opener etc). • Peeling and preparing difficult and dirty vegetables (root crops). • Mixing a variety of raw materials that would otherwise require several shopping stops. • Adding an exciting flavour that is not readily available to the consumer. • Partial cooking to save time and encourage impulse consumption. As a further extension, convenience can be added in specifically catering to the food service industry. Under normal circumstances, Fiji receives approximately 350,000 visitors annually and the current local food input into food for the hotels and resorts is approximately 7% (Beyer and Paretti, 1998). There is thus huge potential to increase the consumption of local products if they are presented in a sufficiently acceptable form for use by the toursit industry. 4) Traditional product or dish modified for commercial sale Many products are consumed throughout the Pacific region are cooked in coconut milk. These have been successfully duplicated in a can. Although there is a small export trade for these commodities, the canning process causes significant damage to both texture and taste. Improved results are now possible with root crops packaged in coconut milk that are subsequently frozen. The products so produced can be thawed, and cooked (if necessary) using a microwave and are thus appealing to western consumers. This is especially true for consumers who frequent the communal dinner table less often, and eat “on the run.” Some examples of products have been produced for Pacific Islander communities now residing in Australia and New Zealand but there is mounting evidence that western consumers are consuming them as well. 5) Varieties Once an industry is established, it is a simple process to increase sales by increasing the market width and depth by developing varieties. Cassava chips for instance, can be made in a single flavour, or spices and colourings can be added to give an alternative variety. Varieties are very common in the food industry because the same equipment can be used for a number of products. This increases the throughput rate, which assists in reducing the fixed cost element that must be recovered per unit item sold. 6) Technology-driven product enhancement There are two aspects of technology-driven product development: • Technology in the home • Technology in the food industry Technology has made a significant difference to our daily lives. The use of refrigerators in the home and now in rural areas throughout the region has extended the shelf life of previously perishable products. Products which had a marginal shelf life (such as pickles and sauces) now have unlimited shelf life if they are kept at low temperatures. The formulation of recipes depends very much on the expected shelf life. Bulk packs of frozen foods are now possible which were not possible a generation ago simply because the product will retain its freshness for the period of consumption of the pack. Frozen cassava and sweet potato French fries are now possible and are displacing frozen potato French fries that are imported in large volumes into the PINs. third taro symposium 217 For the export trader, there is increasing use of microwave ovens in Australia, New Zealand and in the USA. The use of the microwave oven is beginning to extend to the car. Trends would indicate that communal dining is in decline as leisure activities occupy evening time and the family unit becomes less permanent. The microwave encourages expedient eating and thus single serve products. A number of technologies within the food industry have opened up a range of products that were not previously possible. Extrusion has for instance made possible a range of products that have totally different characteristics from any product previously produced. Such products include the open textured honeycomb expanded products such as snack foods TwistiesTM and BongosTM. Such technology has been used in Fiji for the manufacture of a range of extruded snack food based on the starchy staples. Cassava, taro and sweet potato have been extruded and have a bright future as a base for a whole new range of snack foods. 7) Responding to a fashion or a fad The food industry is the subject of a great number of fads and fashions. The range of confections particularly directed at the young is commonly aligned to the latest cinema or television releases. For instance, such fads as space travel results in rocket shaped confections. Computer games such as PacmanTM was followed by an extruded puffed snack food with the same name. Other more sophisticated trends include the latest nutritional fads. Many products today are produced with reduced fat, low salt, monosodium glutamate-free and cholesterol free. The processor must be aware that the scientific validity of such fads is irrelevant - the market perceptions define the buying patterns. Extruded snack foods made from root crops can follow the same trends by incorporating carrot (vitamin A) leafy green vegetables (iron) and lime juice (vitamin C). The public perception of hitherto unhealthy food can be assuaged by incorporation of some nutritional additive. 8) Reducing the cost Once products have been launched on the market, successful companies undergo constant re-appraisal of the formulation. Most attempt to reduce the cost of the raw materials by substituting or extending the expensive components with cheaper alternatives. There have been many instances of fraud and mal-practice based on watering milk and fruit juices and much of food analysis research has been stimulated by the requirement to police such activity. Codex Alimentarius to which most PINs subscribe or to which they are signatories is a series of standards that dictate the ingredients required and the additives that are permitted in foods. Many foods sold today are hidden from the view of the customer in opaque packaging. Even after preparation it is difficult to estimate the contents and thus quality, of foods such as hamburger, sausage and fish sticks. Codex Alimentarius is designed to protect consumers from excessive substitution and fraud. In the case of fresh or frozen starchy staples, there is no opportunity to substitute cheaper components since the items are clearly visible and any additive is immediately obvious. Once these vegetables are used in a less recognisable formulation such as soup base or snack food, then it becomes possible to substitute and extend. At this stage it becomes important for the developer to understand the specifications for the product that are laid down by Codex Alimentarius or by the importing country. For exporters those standards and specifications laid down by Codex Alimentarius and modified by the importing country are sacrosanct. To improve the performance of foods, a range of substances are added to food that will improve appearance, taste, texture, safety and keeping quality. Other publications are available which detail the use of these additives and the information they contain will not be duplicated here. There are a number of ingredients, which can be incorporated into foods that have the same roles as additives. These items have the advantage that they are natural foods and therefore may appear on the label as food ingredients – not additives. By using these components the PIN entrepreneur gains another comparative advantage. 9) Completely new There are still opportunities for the development of completely new products that are entirely innovative. Of the items with enormous potential are the use of breadfruit as a source of latex and the entire range of starchy staples as a source of specialty starches for confections and as sources of novel ingredients in the food industry. The removal of water for the production of flour is expensive. In most instances the flour is used as an ingredient to which water or other fluid is added. Thus it is economically sound to produce shelf stable root crop, plantain and breadfruit products that are equally useful to consumers as an ingredient but which does not have the production expense of dried flours. Starchy staple pastes can be produced and packaged in resealable tubes (similar to tomato paste tubes), for example. Snacks can be formulated and packed in tubes as paste for the consumer to simply squeeze into hot fat as required. Summary and conclusions The key to increasing root crop consumption in general and taro in particular is through a programme of product development. Taro is a particularly suitable raw material for product development since it is bland and essentially 218 third taro symposium without colour. Not only will value adding overcome the buying disincentives suffered by the raw vegetable but will provide an opportunity to add value and increase the return to local growers and entrepreneurs. By building in appropriate features to developed products it is possible to match the features of imported items and thus reduce the expenditure on imported foods In addition, processing provides an opportunity to overcome the inconsistencies and vagaries of quarantine inspection services. References Beyer, R. 1998. Multiple pasteurisation for shelf-life extension of non-acid foods packed in pouches. Unpublished. Report available. Beyer, R. 1999. Kumala as a base for chutney, sauce and jam for the provision of a vector for micro-nutrients. Fiji MAFF Cabinet Memorandum. Beyer, R. 2000. The status of the Fijian food industry. FAO, Sub-regional Office for the Pacific, Apia, Samoa. Beyer, R. 2001a. Root crop processing: A review. Secretariat of the Pacific Community, Suva, Fiji. Beyer, R. 2001b. Increasing the consumption and use of local foods: CDE Report to Fiji Ministry of Agriculture Fisheries and Forests. MAFF, Suva, Fiji. Beyer, R. and Paretti, O.J. 1998. Local food programme for resorts and hotels. Fiji Hotel Association, Suva, Fiji. Bourke, R.M. 1982. Root crops in Papua New Guinea. In: Bourke, R.M. and Kesavan, V. (eds). Proceedings of the Second Papua New Guinea Food Crops Conference, Goroka, 1980. Department of Primary Industry, Port Moresby. Bradbury, J.H. and Holloway, W.D. 1988. Chemistry of tropical root crops: Significance for nutrition and agriculture in the Pacific. ACIAR, Canberra. 201 p. Bradbury, J.H. and Nixon, R.W. 1998. The acridity of raphides from the edible aroids. Journal of the Science of Food and Agriculture 76:608–616. Burlinghame, B.A., Milligan, G.C., Apimerika, D.E. and Arthur, J.M. 1994. The concise New Zealand food composition tables. New Zealand Institute for Crop and Food Research, Wellington. Cable, W.J. 1981. The spread of taro in the Pacific. USP School of Agriculture, Alafua Campus, Apia, Samoa. Cable, W.J. 1982. Report on some recent edible aroid research and references in New Caledonia, Nigeria and parts of Asia, USP School of Agriculture, Alafua Campus, Apia, Samoa. FAO. 1981. Food loss prevention in perishable crops. FAO Agriculture Services Bulletin No. 43. Rome, Italy. FAO. 2000. Food industry surveys of some Pacific island nations. FAO Sub-regional Office for the Pacific, Apia, Samoa. Fiji Bureau of Statistics. 2000. Annual report: Government statistician. Government Buildings, Suva, Fiji. Fiji National Food and Nutrition Centre. 1997. Fiji food balance sheet. Government Buildings, Suva, Fiji. Fiji National Food and Nutrition Centre. 2000. Fiji food balance sheet. Government Buildings, Suva, Fiji. Hodge, A.M., Dowse, G.K., Koki, G., Mavo, B., Alpers, M.P. and Zimmet, P.Z. 1997. Modernity and obesity in coastal Papua New Guinea. International Journal of Obesity 19:154–161. International Conference on Food Security. 1999. Food security: The new millennium. Consumers International, Penang, Malaysia. Kay, D.E. 1973. Root crops. Tropical Products Institute, London. 245 p. Parkinson, S.V. 1982. Nutrition in the South Pacific, past and present. Journal of Food and Nutrition 39:121–125. Parkinson, S.V. 2001. Pers. comm. 19, Vuya Road, Suva, Fiji. Parkinson, S.V., Stacy, P. and Mattinson, A. 1999. Taste of the Pacific. David Bateman, Auckland, New Zealand. 112 p. Vanuatu Department of Health. 1996. Heath report: Second national nutrition survey. Vanuatu Department of Health, Port Vila. third taro symposium 219 Appendix: Composition of the starchy staples Elephant foot yam A.campanulatus Giant swamp taro C. chamissonis Giant taro A. macro-rrhiza Taro X. sagittifollium 79 266 6.8 13 N/A 1.1 0.2 0.74 69.1 480 1.12 24.5 1.01 1.46 0.10 0.87 67.1 521 1.55 27.6 0.42 0.99 0.11 1.04 70.3 449 2.15 21.5 0.96 1.85 0.10 0.92 75.4 348 0.51 16.8 1.03 2.78 0.16 0.67 77.8 336 2.24 16.6 0.14 1.45 0.06 1.36 2 17 7 415 N/A N/A N/A 0.59 0.01 N/A 32 70 115 1.8 448 8.5 0.43 0.18 3.8 0.35 0.09 8.5 53 27 6.6 530 7.9 0.40 0.19 0.52 0.17 0.09 38 44 52 30 267 12 0.83 0.07 1.57 0.62 0.10 182 16 21 72 67 3.3 0.61 0.11 2.3 0.69 0.09 97 67 47 4.1 622 12 0.51 0.18 1.05 0.31 0.17 0.011 0.07 0.02 N/A 0.32 21 0.007 0.032 0.025 0.76 0.19 15 0.005 0.024 0.032 0.80 0.33 14 0 0.021 0.018 0.48 0.46 17 0.005 0.025 0.019 0.46 0.07 16 0.07 0.06 0.05 1.2 3.8 Burlinghame et al. (1994) Bradbury and Holloway (1988) 220 Taro C. escuLenta Potato Moisture % Energy (KJ/100 g) Protein % Starch % Sugar % Dietary fibre % Fat % Ash % Minerals (mg/100 g) Ca P Mg Na K S Fe Cu Zn Al B Vitamins (mg/100 g) Vitamin A (ret.+B-car./6) Thiamin Riboflavin Nicotinic acid Pot. Nic. acid = Trp/60 Total vitamin C (AA +DAA) third taro symposium Appendix continued: Composition of the starchy staples Cassava M. esculentum Wholemeal wheat Rice (white boiled) D. esculenta Moisture % Energy(KJ/100 g) Protein % Starch % Sugar % Dietary fibre % Fat % Ash % Minerals (mg/100 g) Ca P Mg Na K S Fe Cu Zn Al B Vitamins (mg/100 g) Vitamin A (ret.+B-car./6) Thiamin Riboflavin Nicotinic acid Pot.Nic. Acid=Trp/60 Total vitamin C (AA +DAA) 74.2 406 2.06 19.3 0.55 1.15 0.06 0.82 69 509 2.3 28 N/A 0.8 0.2 0.95 12 1150 12.1 52 0.20 11.2 2.1 2.7 62.8 580 0.53 31.0 0.83 1.48 0.17 0.84 7.5 39 26 3.1 303 16 0.75 0.17 0.46 0.24 0.07 4 N/A 13 5 10 15 0.3 0.34 0.6 0.04 0.05 30 27 102 5 315 N/A 0.3 N/A 1.3 N/A N/A 20 46 30 7.2 302 6.4 0.23 0.14 0.48 0.060. 0.07 0.017 0.045 0.028 0.41 0.66 20 0 0.03 0.01 1 0 0 0 4.2 0.11 6 6 0 Tr 0.05 0.04 0.6 0.07 15 third taro symposium 221 Theme Five Paper 5.3 Taro production and value adding in Palau Robert Bishop BOA-FAO, Koror, Palau Introduction Taro in Palau dates back to the misty past. Taro is a prominent and identifying component of Palau’s culture. The traditional system of utilizing wetlands to produce taro is ancient, distinctive, rich and varied. Palau has a vast reservoir of traditional knowledge and skills related to taro. The three main types of taro grown in Palau are Colocasia, Cyrtosperma and Xanthosoma. The number of varieties of Colocasia currently present in Palau is estimated at about 100. Palau Community College-Cooperative Research & Extension, maintains sixty-eight varieties. The number of varieties is rapidly dwindling. The number of varieties available during traditional times has been estimated as over 200 (least) and over 400 (most). Since traditional times, few taro varieties have been introduced from the outside, the most notable exception being varieties introduced by SPC due to their resistance to taro rot and salt water. In everyday diet, taro is being gradually replaced by rice and cassava. Cultural significance of taro The cultural significance of taro is illustrated well by the oft-quoted proverb: “A mesei a delal a telid” which is usually translated to “The taro patch is the mother of our life.” Traditionally, taro was the most important and most prominent and most revered (prestige) food and crop in Palau. As with most Palauan proverbs, stories and legends, this proverb has different levels of meaning. A more literally translation of the proverb renders it: “The taro patch is the mother of our breath.” This implies that at the end of the last day that the last Palauan women goes to the last taro patch, Palau’s culture will have breathed its last breath. The highly productive taro patch system enabled and sustained the Palauan “way of life”. In traditional Palau, a woman’s skill level and how well and how diligently she tended her, her family’s, the clan’s and the community’s taro patches and her provision and preparation of food and other resources from the patches to others was crucial in determining her position and status in the family, clan and village. It provided one of the few vehicles for advancement and wealth accumulation. Advancement and wealth accumulation are important driving forces in Palau’s culture. To illustrate, the more adept and hard-working a woman was in the taro patch the more say she had in the selection of her husband. The taro patch was the “school” where mothers taught their family’s closely guarded secrets, especially privilege knowledge of the taro patch, The taro patch was also the “school” where mothers taught their children especially their daughters: life skills, what it is to be “Palauan”, who they are, what their roles were and social values such as deference, reciprocity, thrift, diligence, and responsibility. This traditional education was devalued, ignored and discouraged by formal educators. Moreover since younger people are increasingly disinterested in agriculture there are less and less opportunities to pass on knowledge, values and skills. Traditionally, taro was featured and required in all customs. Taro features prominently in Palau’s legends, stories, songs, chants and proverbs. Taro was the central element defining Palau’s culture. Production systems There have been observed at least eight identifiable taro production systems. 1. Mesei. The foremost was the mesei, a wetland taro production system. The taro is grown in a paddy-like system with channels and dikes for water control. Various useful plants are grown on the dikes, mainly for food, green manure, medicine, and ceremonies. The wetlands traditionally would be divided into functional sections, which would determine or indicate the u1timate use of the taro grown. According to same growers, the divisions also serve to isolate plots from diseases, as a rotation pattern, and for continuous production. The “typical” taro patch is elusive to define since differences exist between villages and individuals. The following description outlines the methods used in one patch in Ngerbeched. First the planting materia1 is prepared and placed in a channe1 or another coo1 spot. Next, the patch is weeded and the weeds placed in a pile. The top 15 cm of the soi1 are turned over and placed in a pile. The second (15 cm) layer of soi1 is turned over and placed in a different pile. The weeds are then placed in the hole followed by bundles of green manure, according to the dictum, “the more the better”. Dry or dead leaves are used only when green manures are not available. Some growers believe green manures contain the “essence” of life and dry/dead materials do not. Next, the top 15 cm inches of soil are thrown back in the hole, with the roots on the bottom. Then the second 15 cm layer of soi1 is placed upside down. Everything is then smoothed 222 third taro symposium out. This is considered very important for proper growth. The klaeb, the path for water between the different sections of a patch, is redone. The prepared plants are set out and then the mulch, usually consisting of dry banana leaves, is put in place. Intercrops commonly found in the taro patch are kangkong, Limnophila aromatica, and two types of low growing grass. Palau’s taro patch system, in my opinion, is physically the most demanding food production system in Palau; it is also a unique agroforestry system. Although the system differs from village to village and individual to individual, all the patches I have seen utilize trees/shrubs. Trees and bananas are planted or allowed to grow on the perimeter and the dikes. These trees and other plants are used for healing, food, ceremonies, building, firewood, to tell cycles, and magic. 2. Dechel. Another taro plot system observed is the dechel. It is simi1ar to mesei in that the soi1 is damp or wet, however, there is much 1ess water. Usually, some dikes and channels are made, but these are not as extensive as in mesei. The land is usually cleared of weeds first. Generally, a long narrow shove1, stick, crow bar, or a specia1ly made meta1 bar is used to make a hole. The planting material is placed inside the hole, and then the surrounding loose soi1 is stepped on. Mu1ch and organic matter are usua11y not used. Same growers believe dechel wi11 initially produce bigger tubers than mesei. However, most women state that taro grown in the mesei is preferable to that grown in the deche1. Mesei grown taro taste better and weights more than dechel grown taro. 3. Sers is loosely translated as garden or farm. This third system of taro production is on higher and dryer ground than the other two systems described above. Sometimes the land is simply slashed and burned, and the taro is planted in holes made with a pick-ax or three-prong cu1tivator. Some growers make mounds and others form rows. If mounds or rows are to be made, organic matter is added at the time they are prepared. Artificial fertilizer is used more in this system and so is mixed cropping. Some growers wi11 plant on top of the rows while others plant between the rows with or without an intercrop. Most growers considered sers grown taro inferior in taste and quality to both mesei and dechel grown taro. 4. Hybrid. A hybrid system is observed. A dike is built around a section of dechel land so that water collects within the diked area. Unlike mesei, it lacks channels. About six inches of soil is turned over and the soi1 is smoothed. A mulch of various leaves is placed on top. 5. Step terrace. There is a system of taro production on sloping land, utilizing step terracing and living fences. Trees are left at the top of the slope to retain the soi1. Taro is planted on the step terraces. Within each step terrace, a furrow is made. Planting material is laid sideways in the furrow, towards the rising sun. The reason for placing taro sideways is “taro grows up, not down.” Leaves are placed on the side of the taro in the furrow. After about two months, the furrow is covered lightly with soil. In another two months, more soi1 is mounded over the taro, “to take advantage of the increased nutrients in the upper layer of the soi1.” A mulch of cut weeds, as well as artificial fertilizer, is used. The living fence is also a source of plant food. 6. Forest land. The land that were formerly mangrove swamp or forest, are initially slashed and burned sometimes following a fallow. A knife, stick, or similar tool is used to make holes to plant Colocasia and Cyrtosperma, and the loose soi1 is stepped on. As the Cyrtosperma grows, the Colocasia is phased out. 7. Post digger. A fairly new and fast system has evolved due to an introduced tool the “post digger”. It is used on both dry and wetland. A farmers uses the “post digger” to “punch” holes about 15 to 25 cm. deep. The hole is then filled with 7.5 cm of with dry leaves or weeds and a handful of ash. The planting material is place in the hole. Next a handful of soil is placed over the planting material. As the taro grows up the remaining soil taken from making the initial hole is placed around the plant. This system was developed to save time and take advantage of “taro grows up, not down”. 8. Back-sloping terrace. In the far past, apparently taro and other crops were grown on huge back-sloping terrace systems, similar to Indonesia. Soil management Practices to improve and/or maintain soil fertility, in order of prevalence are: the addition of plant matter (especially green manure), mulches, fallowing, ash, animal manure, fish gut “soup”, lime and compost “teas”. It is considered by some farmers, in order to have adequate fertility in the initial year of a taro patch (360 cm x 360 cm), it is necessary to add 28 bundles (about 23 kg each) of elephant grass. In succeeding years, only 7 bundles would be required for each crop cycle. It is believe by many farmers that adding organic matter to the soil improves the taste of the taro. Conversely, it is believe by many farmers that artificial fertilizers “hurts” the soil, attract pests including diseases, worsens the taste of taro and “puffs” it up (like a puffed snack food). Pest management Intercropping is sometimes used to “confuse” pests. It is believed high soil fertility, additions of large amounts of organic matter, additions of lime and ash, and the use of aromatic plants prevents and deters pests. In the taro patch, the good maintenance of the furrows between patches and the ditches around the patches ensures the proper flow of water to prevent and remove diseases. Also in the taro patch, sometimes the water level is raised to drown pests. Planting materials are sometimes dipped in a solution of derris or Barringtonia or chlorine for disease and insect treatment. In past severe insect infestations were countered by a smoking and yelling involving many community members. third taro symposium 223 Growers will grow 5 or more varieties of taro as insurance against insect, disease and other pest attacks. The women in the taro patches were very surprised and saddened to here that the Samoan taro industry was based on a few “super” varieties that were wipe out by only one (“minor” to Palauans) disease. All taro growers have their favorite varieties, but they make sure to plant other varieties “just in case”. The three major pests as identified by taro growers are in descending order of importance, are: diseases, taro plant hopper; and the uek (purple swamp hen). Taro growers considered the corm rot of Colocasia, known locally as obei the most serious of the diseases. Value adding The taro’s corm, stem and leaves are use in a variety of ways for food, ceremonies and for healing. Over the years Palauans have developed a variety of ways to preserve and add value to taro. Palauans have adapted and/or adopted many techniques for taro preservation and value adding from foreign administrators and other foreigners living in Palau. Several of these traditional ways and adapted/adopted techniques will be described. Melid: The taro root is scrape clean. It is sometimes then put in a bag, froze and sent overseas. Ongat: The taro root is steamed. It is used to provide steam for the childbirth ceremony. Meliokl: The taro root is boiled. Blsiich: The taro root is boiled, pounded, and molded into a log. Sometimes coconut oil is added. Oumillum: The taro root is boiled, pounded and/or grated, wrapped in banana, ti plant, the sheath of the betelnut leaf or coconut leaves and boiled to make billum. Blillum: The billum is boiled again. It changes the taste. Mengat: The billum is smoked. It last longer and changes the taste. Delul: The taro root is boiled, pounded or grated, flattened, burnt slightly on both sides, grated (sometimes with sugar added) and then molded into balls or paddies. Cheluit: The taro root is cooked, sliced, recooked with light coconut milk. Mengerdoched: The taro root is sliced, fried, with sugar or caramelized sugar added sometimes. (chips) Mengeluomel: The taro root is wrapped and baked. The taro root and sometimes the leaves are slow-cooked or steamed inside a chicken or pig. Mengum: The taro root is bake in the ground. Telledou: The taro root is boiled; pounded, grated coconut is added and then molded. Telumar: The taro root is boiled, mashed, and then cooked with coconut milk in a large pot until it is a thick paste. Mengesureor: The taro root is cooked, sliced, and then recooked with coconut syrup. Titimel: The taro root is cooked, mashed, molded in small balls, then coconut cream or caramelized sugar is added. Chemlol: The taro root is cooked, placed with rainwater in a large loosely covered clay jar and fermented. Once fermented, drinks are made by taking small amounts from the jar and then adding water and coconut syrup. Telooch: The taro root is boiled, and pre-chewed, usually for babies and elderly. Chelbakl: The taro root is boiled, mashed or grated with caramelized sugar or coconut syrup, then wrapped. Curried/Stew: The taro root is cooked with curry or stew. Salad: Cooked taro root replaces potato in potato salad. Taro roll: The taro root is cooked, mashed, flattened, rolled with thickened coconut cream and sliced. Pandan-taro: The taro root is grated; wrap with pandan leaves to form squares and then boiled. The pandan leaves impart a slight taste. Sandwiches: Cooked and sliced taro is used in place of bread in sandwiches. Chelang: Boiled taro stems with coconut milk and sugar. Similar to rhubarb sauce. Demok: Boiled taro leaves with coconut milk. Similar to cream of spinach soup. Ngesur: Furled taro leaves tied together and cooked with pig. Chesul: The taro leaves are formed into circles and the cooked with heavy/thick coconut milk. Training and technical assistance requests Several recent participatory trainings in Palau by the Bureau of Agriculture-Food and Agriculture Organization joint project entitled “Capacity Building in Farm Management, Marketing and Value Adding for Sustainable Livelihoods” 224 third taro symposium has indicated Palauans are keen to try out new (to them) techniques for adding value to taro. Recent missions of FAO and ESCAP consultants verified these indications in interviews. Palauans what like training and technical assistance in the following areas: packaging that can withstand boiling; a “taro” cooker similar to rice cookers; taro ice cream and other taro based frozen desserts; taro wine; taro fudge; taro gelatin product similar to “jelly ace”; taro filled araban/ amban/shoban; taro bread, and instant taro powder (not poi). References Bishop, R.V. 1991. Agroforestry offers a promising future. Social Forestry Network Paper 12g:1–3. Ferentinos, L. (comp.). 1990. Sustainable taro culture in the Pacific: The farmers’ wisdom. LISA Project, Honolulu. Nero, K.L., Murray, F.B. and Burton, M.L. 2000. The meanings of work in contemporary Palau: Policy implications of globalization in the Pacific. The Contemporary Pacific 12(2):319–48. Ngiralmau, M. and Bishop, R. 1989. A report on the rapid rural appraisal of Colocasia taro agriculture in Palau. In: Vargo, A. and Ferentinos, L. (eds). Rapid rural appraisal of taro production systems in Micronesia, American Samoa and Hawaii. LISA Project, Honolulu. A story on how Palauans invented the sandwich Long ago in the misty past, long before the Earl of Sandwich was born, taro was the staff of life. One day a group of “poor” fishermen were returning from their fishing expedition. Their village was far inland. They were very hungry, so they stopped at the nearest house. They found in the house a very elderly lady cooking taro. They asked the lady to cook some fish for them. She agreed. The fishermen upon seeing that the lady was “poor” for she had no meat in her household gave her some fish for herself. The lady upon seeing the fishermen were “poor” for they had no taro, decided to give them some cooked taro. Once she finished cooking the fish for the men, she sliced several taros in half. She then put the cooked fish in the middle of the taro and tied the taro together and places them in a basket. She called the men over and gave them the basket. The men thanked the lady and continued on their way until they found a shady cool spot. They sat down to eat and opened the basket. Seeing the taro inside the basket, one man grumbled that “We gave the lady some fish and she took the ones for us also”. Another man scolded: “Let’s not assume the worse. Look! The taro is tied let’s untied them.” Upon untying them, they were very happy to see their fish inside the taro and proceeded to eat them with relish. Some morals of the story: If you help others, they will help you. The outside or the package does not necessarily indicate its content. Similar to the modern proverb: Do not judge a book by its cover. third taro symposium 225 Theme Five Paper 5.4 Recent developments in taro-based food products in Hawaii Alvin S. Huang, Karthik Komarasamy and Lijun He Department of Human Nutrition, Food and Animal Sciences, College of Tropical Agriculture and Human Resources, University of Hawaii, 1955 East-West Rd, Honolulu, HI 96822 Introduction Several food products containing taro as the main ingredient have been developed in Hawaii in recent years. In this discussion, a taro-based food product is defined as one that contains taro as the first or second in its ingredient list, which is based on weight percentage. Research and development on taro processing and product development has been on-going at the University of Hawaii since 1970 (Moy and Nip, 1983). In Table 1, a comparison has been attempted to show a shift in the taro product R&D from the first two decades to the recent years (1990 to present). This shift could be attributed to the changes in funding sources and a reflection of the economic environment. In the first two decades, for instance, these research projects were mainly funded by the U.S. Department of Agriculture (USDA), which stressed the employment of new technology and the importance of global food resources. In recent years, however, most projects have been initiated by a number of local food businesses and through grants provided by the Small Business Administration (SBA) and the Department of Defense (DOD). These new funding sources tend to direct a more practical and Hawaii-centered approach on the taro product development and stress a niche-marketing emphasis. Table 1: A comparison of two periods (1970-1990 and 1990-2003) in taro product development in Hawaii Period 1 (1970-1990) Period 2 (1990-present) Corm source wet-land taro dry-land taro Intermediate form dehydrated powder (taro flour) taro paste (poi) Process technology spray drying, freeze drying baking, freezing, cold processing End products pancake mix, drink powder, baby food bakery filling, non-dairy yogurt, frozen dessert Marketing position exporting, distant shipping local and tourist market Outcome too expensive to compete competitive in niche market One major developmental shift is to move away from the traditional wet-land taro as the raw material source for new products. Rather, the less labor intensive and easier to scale up dry-land taro has become the choice of source. Dry-land taro farming also requires less amount of agricultural water, which has also become a contentious issue in resource distribution in Hawaii. However, this does not imply a de-emphasis on the traditional poi industry, but an effort to strengthen it with a new material source (dry-land taro) and an extended line of taro products to expand its business base. Many of the new products being developed are built on the poi manufacturing structure and use this “taro paste” as an intermediate material. The taro paste is essentially a poi product with a different moisture content and textural properties. The downstream processing and end products in bakery and frozen dessert forms are also more compatible with the types of food processors that already exist in Hawaii. In the past, the making of taro flour and dehydrated taro products were aimed at an exporting market outside of Hawaii. Even though the shipping costs of dehydrated products are considerably lower than those for shipping taro corms or poi, the high costs of manufacturing and the small land area in Hawaii still render these products less competitive. The total taro planting acreage in Hawaii has been fluctuating around 1,000 acres in the last ten years, a small-scale production in comparison with some of the other taro producing areas in the world (Nakamoto et al., 1994). Accompanying a more business-oriented R&D effort is a growing emphasis on intellectual property protection in the forms of patent application and non-disclosure/confidentiality agreement. Due to these constraints, some of the discussions in this article have to be limited in details. The most exciting and far-reaching development on taro-based products has to be the creation of a taro “yogurt”. In the U.S., yogurt is defined as a dairy product and is clearly listed in the Standardized Food category. This non-dairy taro “yogurt” apparently has to be re-named. Creative names such as “Tarogurt”, “Kalurt” have been suggested. Before a consensus can be developed and reached, we will refer the product as “taro yogurt” in this discussion. The product is at its infancy, perhaps at a stage similar to that of cheese in the 19th century. However, with recent global interest in cultured foods and probiotics, we are certain that a product like taro yogurt will be developed at a much faster pace. It is a product that can be improved simultaneously at several localities for each unique taste and probiotic attributes. 226 third taro symposium Focus on taro yogurt In a 1994 article (Huang et al., 1994), we scientifically established that the Hawaiian staple, poi, is a natural fermentation product mainly caused by lactic acid producing bacteria (LAB). Several species of LABs have been isolated and identified with DNA technology in recent years. They are Lactococcus lactis, Lactobacillus plantarum, Leuconostic lactis, Tetragencoccus halophilus and Weissela confusa. They all are white pinpoint colonies, Grampositive, catalase-negative, facultative anaerobes. Their distributions in poi vary from taro sources and poi processors. Consequently, the types and number of LABs in poi can clearly affect the eating and storage quality of poi. In Table 2, the major quality control problems facing today’s poi manufacturing are listed. Most of these can be solved or minimized by inoculating pure culture lactic acid bacteria in poi. Table 2: Quality control issues in modern day poi manufacturing Problems Causes Solutions pH does not drop on first day Low lactic acid bacteria (LAB) counts Mix in lactic acid, inoculate LAB Short shelf-life, moldy in 3 days pH too high (> 4.5), low LAB counts Recover with dehydration “Funny” taste in sour poi Other bacteria outnumber LABs Select taro source, inoculate LABs Poi gels in refrigeration Solid content too high Layer water over; mix water prior to chilling In order to have a pure cultured poi product, the taro paste has to be pasteurized or cleaned prior to inoculation. Due to the stickiness and high starch content in taro corms, a successful process has been developed after testing various combinations of steps. The process we have developed and which is being used in making taro yogurt is outlined in Figure 1. Taro corms ↓ Pressurized washing (tap water) ↓ Peeling with a steam peeler ↓ Soaking in 3% lactic acid solution overnight ↓ Cut to two inch cubes ↓ Second wash with distilled water ↓ Pressure cooking of cubes at 250°f for 30 minutes ↓ Cooling in a clean chamber ↓ Grind with distilled water to a paste of 15% solid ↓ Pasteurize to a total bacteria count under 100 ↓ Cooling in a clean chamber ↓ Ready for inoculation Figure 1: Process scheme of manufacturing a pasteurized taro paste It is important to note that the process has been developed based on the taro corms available to us in Hawaii. Specifically, we used the Lehua variety taro grown in a modified dry-land farming practice that uses drip irrigation (Huang et al., 2000). In our case, the soaking of peeled corms in lactic acid solution is critical in reducing the existing LAB counts, which can be quite high in the raw material. The process may have to change and adapt when different kinds of taro corms and farming practices are encountered. However, the use of distilled water that is sterile and the precaution of cooling in a clean environment would be critical in achieving a very low total bacteria count (<100) for a successful inoculation of the pure culture LAB. We first tried commercially available LABs used for making regular yogurt or sour milk. They all have difficulties in growing in this non-dairy medium. The bacteria counts stayed at approximately 105 and pH drop stopped at 4.7, much higher than the pH value of 3.9 common in a sour poi. Sensory evaluations revealed that the inoculated poi tasted “funny” or “garbage-like”. HPLC analysis of the products showed that other acids such as acetic acid and succinic acid outpaced the accumulation of lactic acid. third taro symposium 227 We then started inoculating the isolated LABs from poi. They all grow much more vigorously than the commercial yogurt LABs in poi, the counts increasing from 105 to 109 in one day at 20°C. However, most of these species have had difficulties to sustain the growth and bacteria counts after three days. The numbers usually start decreasing after three days, apparently having exhausted the sucrose and glucose needed to sustain the growth. The species that yielded the best growth and resilience in the taro paste is the Weissela confusa we isolated and purified from poi. In Table 3, we compare the taro yogurt inoculated with this LAB and a paired testing with a taro yogurt inoculated with the commercial dairy LAB, as well as a regular yogurt. The test numbers reflect the observations aforementioned. All three products were not flavored with fruits, but the diary yogurt has sugar added. We believe that a mixture of fruits and sugar in taro yogurt would drastically improve the taste and mouthfeel. Table 3: Comparison of two taro yogurts with a regular dairy yogurt in microbial and sensory attributes Taro Yogurt A (L. bulgaricus etc.) Taro Yogurt B (W. confusa) Regular Dairy Yogurt (L. bulgaricus etc.) End pH 4.7 3.4 4.7 Total count (3 day) 3.4 × 107 5.3 × 109 1.6 × 105 Organic acids Lactic acid, acetic acid Lactic acid Lactic acid Total acidity 87 mg/100 g 122 mg/100 g 67 mg/100 g Sensory Taste Aftertaste Mouthfeel funny, garbage-like pungent, bitter starchy, grainy pure sour pleasant smooth, slimy pure sour pleasant smooth, rich, silky 2.3 5.8 6.6 Acceptability (1-9 scale, 9 being the most acceptable) The University of Hawaii has decided not to release this LAB species we isolated from poi and is contemplating licensing and other intellectual property issues. However, we are certain that there are other LABs that could be isolated from taro grown outside of Hawaii that may possess unique properties such as producing bacteriocin and better acid tolerance. These are important attributes for probiotic microorganisms and are great assets to human health. Taro ice cream (non-dairy frozen dessert) We have tried a cooked taro paste to replace milk and other dairy ingredients in a frozen dessert (ice cream-like) formulation. The formulation, composed mainly of taro, hydrogenated vegetable fat and sugar, has been tried on a commercial ice cream machine with satisfactory results. Table 4 lists a comparison of this non-dairy product with a premium ice cream and a frozen sorbet in terms of nutritional composition and sensory attributes. The comparison shows that the non-dairy taro product has a similar nutritional benefit as the frozen sorbet, but with a taste and sensory attributes similar to a high-fat premium ice cream. A number of minor ingredients have to be used in this taro frozen dessert to prevent the formation of large ice crystals. The product also takes a longer time to soften prior to serving. The technology has been filed for a U.S. patent by the University of Hawaii. Table 4: Nutritional and sensory attributes of taro ice cream, a premium ice cream and a frozen sorbet Premium Ice Cream Frozen Sorbet Fat % Attributes Taro “Ice Cream” 8% 18% Protein % 1% 4% 0% Dairy components none mostly none Sensory Creaminess After taste (descriptor) Mouth feel (descriptor) 6.3 (0.7) 6.5 (0.5) clean, starchy 5.4 (0.5) cold, melt slow 7.6 (0.5) 6.2 (0.8) cream, greasy 7.3 (0.4) rich, melt smoothly 1.2 (0.3) 5.3 (0.5) clean, watery 4.1 (0.4) cold, icy 1% 1 1 hedonic 1-9 scale, standard errors are in parentheses (n = 12) Other taro products Similarly to the application in frozen dessert, taro paste has been used in bakery filling to replace fat. In addition to reduce the fat and caloric intakes, a lot of consumers appreciate the taste of taro transpiring through the filling. Cooked taro chunks have been used in a vegetarian burger as the meat replacement. The product has been commercialized in Maui and is being distributed on the U.S. West Coast. The success of the product depends on cooking the taro chunks separately to achieve the right texture and best taste. In an experimental trial, we have tried to use dehydrated poi as replacement for wheat flour in extruded products, such as breakfast cereals and puff snacks. The dehydrated poi can make up 50% of the total formulation weight and developed right cell structure in a twin-screw high pressure extruder. Prior to the test, some extruder experts concerned that the stickiness of the poi product may build up excessive pressure in the chamber, but the tests proved that to be false. The main obstacle in this application has proved to be the raw material costs. Due to the limitation in production scale and high water content (60 to 80% on weight base) in taro, the dehydrated poi would cost 8 to 10 times higher 228 third taro symposium than wheat flour. However, for those allergic to wheat flour, extruded taro products in puff snack and breakfast cereal forms may be welcome alternatives. Conclusions Our studies have demonstrated that taro can be very versatile in food applications. It is a good carbohydrate source in lactic acid bacteria fermentation. It replaces fat in frozen desserts and bakery fillings. It is used as a meat replacement in a vegetarian burger. It can replace wheat flour in extruded products. The costs and availability of taro are the main obstacles in developing and commercializing taro products. Hopefully, with the proof in marketing success, more taro will be planted. Along with improvement in farming practices and disease prevention, a better and cheaper taro supply can be materialized in the near future. References Huang, A.S., Lam, S.Y., Nakayama, T. and Lin, H. 1994. Microbiological and chemical changes in poi stored at 20C. Journal of Agricultural and Food Chemistry 42:45–48. Huang, A.S., Titchenal, C.A. and Meilleur, B.A. 2000. Nutrient composition of taro corms and breadfruit. Journal of Food Composition and Analysis 13:859–864. Moy, J.H. and Nip, W.-K. 1983. Processed foods. p. 261–268. In: Wang, J.K. (ed.) Taro: A review of Colocasia esculenta and its potentials. University of Hawaii Press, Honolulu. Nakamoto, S.T., Wanitprapha, K., Iwaoka, W. and Huang, A.S. 1994. Cassava, ginger, sweet potato and taro trade statistics. CTAHR, University of Hawaii, Honolulu. third taro symposium 229 Theme Five Paper 5.5 Chemical composition and effect of processing on oxalate content of taro corms E.O. Afoakwa, S. Sefa-Dedeh and E.K. Agyir-Sackey Department of Nutrition & Food Science, University of Ghana, Legon-Accra, Ghana Introduction Xanthosoma sagittifolium and Colocasia esculenta are tropical root crops commonly referred to as taro. They are used as subsistence staples in many parts of the tropics and sub-tropics in Africa. They produce starch storage corms and have several genera and species throughout the world. Investigations have shown that taro contain digestible starch, vitamin C, thiamin, riboflavin, and niacin and as well as proteins and essential amino acids (Onayemi and Nwigwe, 1987). The dietary importance of root crops has led people to devise various means to determine the composition of food commodities. Several authors have evaluated the chemical composition of corms from both Xanthosoma sagittifolium and Colocasia esculenta (Wills et al.,1983; Bradbury and Holloway, 1988). It has been observed that in spite of the fact that taro corms are neglected crops, their compositional value is high with an average protein content of 6 % and 390 calories per 100g dry matter. Onwueme (1982) reported that the corms of taro show distinctive variation within the tubers, from the distal attachment to the growing apex (geotropic). In spite of these observations, little attention has been given to taro. One major limiting factor in the utilization of taro is the presence of oxalates which impart acrid taste or cause irritations when foods prepared from them are eaten. Ingestion of foods containing oxalates have also been reported to have caustic effect, exert irritations to the intestinal tract and cause absorptive poisoning (Sakai, 1979). Oxalates are also known to interfere with the bio-availability of calcium. Even though it has been reported that the traditional methods of drying taro reduce their limits (Purseglove, 1988), they do not completely eliminate oxalic acid as itching is still reported by large number of consumers (Onayemi and Nwigwe, 1987). The results obtained indicated significant levels of reduction. Available literature on the effect of processing on oxalates appears conflicting and inconclusive. This research was therefore aimed at evaluating the chemical composition of Xanthosoma sagittifolium and Colocasia esculenta corms as well as determining the effect of processing on the oxalate levels of products derived from the taro corms. Materials Preparation of fresh taro samples Fresh samples of two taro varieties, Xanthosoma sagittifolium (red and white cultivars) and Colocasia esculenta, corms were respectively harvested from local farms at Akyem-Begoro and Anyinam in the Eastern Region of Ghana and transported to the laboratory for the study. Within a day of harvest, the samples were peeled and the edible portion cut into three sections representing the distal, middle and apical parts of the corms. A sample ratio based on weight was used to divide each corm into three parts, i.e. a ratio of 2:3:1 for the distal, middle and apical sections. The above ratio was arrived at by examining the colour variation across the corm. Portions of the raw sample of the sections were blended and used for the analysis. The other portions were dried at 50°C for 24 hours using an air oven. The dried were subsequently milled into flour in a hammer mill (Christy and Norris Co., USA) to pass through a 4 mm sieve. The flour products were kept in sealed polyethylene containers for analysis. Experimental design A 3 x 3 factorial experimental design was used and the principal factors were: i. Type of cultivar: X. sagittifolium (white-flesh), X. sagittifolium (red-flesh) and Colocasia esculenta ii. Corm section: Distal, middle and apical Samples were then analyzed for chemical composition (moisture, protein, fat, ash, starch and fibre) and minerals (calcium, magnesium, zinc, iron, sodium, potassium and phosphorous). Preparation of samples to study the effect of processing on the oxalate content Fresh samples of the taro varieties were peeled and the edible portions prepared for air-oven, solar and drum drying. For the air-oven and solar-dried samples, the edible portions were sliced into sizes (20 g, 40 g and 60 g) with surface area of the sizes in the range of 1.96 x 10-3 m2 to 9.50 x 10-3 m2. The dried products were milled into flour and used for analysis. Preparations of samples for drum drying involved two forms, wet- and dry-milled processes. In the wet-milled process, the edible portion was blended in a Warring blender and mash obtained adjusted with water to obtain a 75% 230 third taro symposium (w/w) paste before drum-drying. In the dry-milled process, flour obtained from the air-oven dried samples was mixed with water to form 75% (w/w) paste and the resulting paste drum-dried. The flaky drum-dried products were further dried at 40°C for 30 minutes and milled into flour. The pre-gelled flour was packaged in polyethylene containers and used for analysis. Experimental design Xanthosoma sagittifolium and Colocasia esculenta samples were processed using three methods of dehydration, namely, air-oven, solar and drum-drying. The tubers were washed, peeled and processed as follows. a. Air-oven drying A 2x3x3 factorial experimental design was developed for the samples and replicated using a cabinet dryer (Gallenkamp, England). The factors and levels used were: i. Drying time: 12 and 24 hours ii. Size of corm: 20, 40 and 60 g iii. Type of cultivar: Xanthosoma sagittifolium (white-flesh), Xanthosoma sagittifolium (red-flesh) and Colocasia esculenta b. Solar drying A 2x3x3 factorial design was used and duplicated. The factors and their levels were: i. Dehydration time : 2 and 3 days ii. Size of corm : 20, 40 and 60 g iii. Type of cultivar: Xanthosoma sagittifolium (white-flesh), Xanthosoma sagittifolium (red-flesh) and Colocasia esculenta c. Drum-drying Double roller drum dryer using steam at 80 psi, clearance angle of 0.012 mm and rotating at the speed of 25-50 rpm was used on 2 kg samples. The products obtained were milled and used for the determination. Chemical analysis Chemical composition The samples were analysed in triplicates respectively for moisture, ash, crude fat, crude protein and fibre contents using Association of Official Analytical Chemists’ Approved methods 925.10, 920.87, 920.85, 923.03 and 963.09 (AOAC, 1990). Carbohydrate was estimated by difference. Mineral analysis Standard AOAC (1990) method was used to digest 2.0 g flour samples. One hundred millilitre (100 ml) standard solutions were prepared from the digest and used for the mineral analysis. Minerals (calcium, magnesium, zinc, iron, sodium, potassium and phosphorus) were determined using standard analytical methods. Estimation of Ca, Mg, Zn and Fe Part of the standard solution of the digest was used to determine Ca, Mg, Zn and Fe using Perkin Elmer Atomic Absorption Spectrophotometer (Model AAS-3, Carl Zeiss, Germany), with air acetylene flame at 422, 286, 720 and 722 nm respectively. Estimation of Na and K Two (2) millilitres of the digest were used to estimate sodium and potassium using the flame photometric method. Five (5) millilitres of the standard solution was placed in a beaker and the inlet tube of the photometer placed in the solution. The solution was absorbed, atomised by the flame photometer (Model PEP7, Jenway, United Kingdom) with butane gas and their quantities estimated by a detector. Phosphorus determination The method described by Watenabe and Olsen (1985) was used in determining phosphorus. Aliquots of digest were added to 1.25% P-nitrophenol in 50 ml volumetric flask and the solution neutralised with 5N HCl. The samples were diluted and reduced with ascorbic acid. Absorbance was measured on the UV/VIS/NIR Spectrophotometer (Model PU8620 Phillips, Netherlands) with 1 cm cuvette at 712 nm. Oxalate determination The AOAC (1990) Analytical Method was used to determine the oxalate content of the fresh and processed samples. The oxalate content was determined by titrating an aliquot of extracts from the homogenized samples with 0.01 third taro symposium 231 KMnO2 solution. Prior to the determination, the heavy metals in the acidified extracts were precipitated with 5 ml tungstophosphoric acid reagent and centrifuged at 1700 rpm for 15 mins. Statistical analyses The data obtained from the studies were statistically analyzed using Statgraphics (Graphics Software System, STCC, Inc. U.S.A). Comparisons between sample treatments and the indices were done using analysis of variance (ANOVA) with a probability p<0.05. Results and discussion Chemical analysis Proximate composition The proximate composition of the corms of Xanthosoma sagittifolium and Colocasia esculenta samples investigated are presented in Table 1. Mean values obtained for Xanthosoma and Colocasia samples in g/100 g dry weight basis were: crude protein 1.56-2.98, starch 12.23-36.64 and crude fibre 1.11-3.00. The moisture content of the fresh weight ranged from 57.63 to 77.41% in Xanthosoma sagittifolium (red-flesh), 54.46 to 71.97% in Xanthosoma sagittifolium (white-flesh) and 59.30 to 72.06% in the Colocasia esculenta (Table 1). For the sections (distal, middle and apical) of the varieties, high moisture was found at the apical section of the two Xanthosoma samples and distal sections of the Colocasia samples. Table 1: Proximate composition of sections of three taro varieties (g/100 g, dmb) Moisture Protein Starch Fat Ash X. sagittifolium Distal 68.54 4.09 22.65 0.60 2.98 Fibre 1.16 (red-flesh) Middle 57.63 3.96 33.61 0.43 2.68 1.70 1.77 Apical 77.41 3.94 12.23 0.74 3.93 X. sagittifolium Distal 63.48 5.50 26.97 0.58 2.38 1.11 (white-flesh) Middle 54.46 4.92 36.64 0.28 1.98 1.72 Apical 71.97 4.94 18.03 0.43 3.29 1.35 Distal 72.06 4.69 17.81 0.97 1.88 2.80 Middle 59.56 4.30 31.01 0.75 1.66 2.74 Apical 59.30 2.98 32.53 0.64 1.56 3.00 Colocasia esculenta Wide variations in moisture content were found at the different sections of all the varieties. Similar trend was found in the results obtained for fat, ash and crude fibre. The variations in the mean values of protein content of the sections of each variety were also distinct, being largest for the distal section, followed by the middle section and the apical section. All the sections had protein content of less than 5% with the exception of the distal section of Xanthosoma sagittifolium (red-flesh) variety, which had a value of 5.5% (Table 1). However, the protein content of Colocasia esculenta was slightly higher than that of Xanthosoma samples and had a higher range of values. Similar observations had been reported for the chemical composition of other taro varieties (Agbor-Egbe and Rickard, 1990). The crude protein content obtained in this study were comparable with the mean values of 5.60 g/100 g reported for sweet potato (Bradbury et al., 1988) and lower than 9.02-9.96 g/100 g reported for yams (Afoakwa and Sefa-Dedeh, 2001). The yield of starch at the sections of the varieties varied from 12.23% to 36.64%, the highest being obtained at the middle section of the Xanthosoma samples and the apical section of the Colocasia samples. Considerable variations in the distribution of starch within the corms were observed. In general, wide variations were observed in the proximate composition values obtained at the sections as well as among the varieties. These variations observed have been ascribed to differences in the genetic background as well as climate, season, and the agronomic factors (Onwueme, 1982). The high levels of starch and fibre in taro have been utilized in the preparation of various food products. Preparation of speciality foods in the prevention of allergic diseases based on carbohydrates in taro has been reported (Rehm and Espig, 1991). It has also been reported that fibre from Colocasia samples, which was incorporated into ice-cream sherbet, effectively activated the action of intestinal bifidobacteria for good digestion and vitamin synthesis (Sotozono, 1989). However, the low levels of protein in taro mean that food products from such commodities should be improved by combining them with other high-protein sources for good nutritive value. Analysis of variance on the data showed that, levels of moisture, protein, fat, ash, starch and crude fibre were significantly different (p<0.05) at the section of each variety and with the different varieties. Multiple range analysis showed that most notable sources of variation were between the Xanthosoma samples and the Colocasia samples. Mineral analysis Levels of minerals in the taro varieties studied are given in Table 2. For the Xanthosoma samples, high levels of minerals were obtained at the apical sections as compared to the distal and middle sections. With respect to the distal and middle sections, the concentrations of minerals did not show any significant variation. The Xanthosoma sagittifolium (red-flesh) variety had comparatively higher concentrations of minerals at the three sections than the Xanthosoma sagittifolium (white-flesh) variety. 232 third taro symposium Table 2: Mineral composition of two Xanthosoma cultivars and Colocasia esculenta samples (mg/100g, dry matter) Ca Fe Mg Zn K Na P 7.70 3.73 69.78 51.05 1451.39 21.57 48.87 Middle 8.53 2.62 64.48 25.75 769.61 21.07 47.71 Apical 24.33 2.81 85.03 20.87 1525.80 23.70 63.07 X. sagittifolium Distal 7.91 3.34 64.24 46.25 985.58 28.91 52.51 (white-flesh) Middle 9.55 3.36 58.86 28.33 963.05 30.27 43.04 Apical 18.99 3.89 67.57 51.09 1388.68 49.14 41.58 Distal 7.09 3.75 64.78 28.41 1004.63 34.62 60.24 Middle 4.68 2.68 48.71 16.98 763.98 29.75 54.72 Apical 5.21 3.10 57.22 17.70 835.00 28.48 61.52 X. sagittifolium Distal (red-flesh) Colocasia esculenta The results obtained for the Xanthosoma samples in this investigation agree with the observations of Lauzon and Kawabata (1988). With respect to the Colocasia esculenta high mineral values were obtained at the distal section indicating some differences in mineral distribution between the Colocasia and the Xanthosoma samples. In general, considerable variation in mineral distribution was noted at the section of the varieties. Analysis of variance conducted on the results showed that levels of minerals in the samples analysed were significant (p<0.05) at the sections of each variety with the exception of values obtained from iron. The variation of mineral distribution among the three cultivars was also significant (p<0.05). Potassium was the most abundant mineral (763.05-1451.30 mg/100 g) found with appreciable amounts noted for zinc (16.98-51.09 mg/100g), magnesium (48.71-85.03 mg/100 g) and phosphorus (41.58-63.07 mg/100 g). Iron concentration was the lowest of minerals observed in the varieties studied. Although little is known regarding the environmental and physiological processes that regulate the uptake of minerals in plants, considerable variations in mineral concentration have been generally observed. The influence of species, concentration of minerals in the soil and age of the plant has been reported (Fennema, 1988). Calcium oxalate content of taro corms Evaluation of fresh samples The presence of oxalates in foods especially taro have been associated with acridity and toxicity when such commodities are consumed. The levels of oxalates in the locally grown taro are important in the assessment of their nutritional status. The data comparing oxalate contents of sections of fresh taro samples investigated are as presented on Table 3. Mean values obtained for the sections in each variety were in the range of 253.49-380.55 µg/100 g for Xanthosoma sagittifolium (red-flesh), 302.19-322.82 µg/100 g for Xanthosoma sagittifolium (white-flesh) and 328.41-459.85 µg/100 g for Colocasia esculenta. For the Xanthosoma sagittifolium (red-flesh) variety, high oxalate content was found at the apical section followed by the distal and middle sections (Table 3). In comparing the values of oxalates obtained from the intra-sections to the whole section of the edible portions as is normally reported in literature, a variation was observed implying that oxalate values quoted in literature may not represent the optimal levels contained in the corms. Similar observations were made for the Xanthosoma sagittifolium (white-flesh) though values obtained in this case were lower. For the sections of Xanthosoma sagittifolium (white-flesh) variety, the highest oxalate level was obtained at the distal sections. Considerably high levels of oxalate were detected at the distal section of the Colocasia esculenta. Table 3: Oxalate content of fresh cocoyam samples (µg/100 g) Oxalate content X. sagittifolium Whole 309.03 (red-flesh) Distal 295.15 Middle 253.49 Apical 380.55 X. sagittifolium Whole 302.19 (white-flesh) Distal 322.82 Colocasia esculenta Middle 305.5 Apical 269.49 Whole 459.85 Distal 488.96 Middle 328.41 Apical 402.69 Oxalate values for the Colocasia samples were found to be much lower than the values in the ranges of 430-1560 µg/100 g obtained by Huang and Tanadjadja (1992) using anion exchange high performance liquid chromatography. Levels of oxalate obtained in the fresh samples of the three samples investigated were also found to be considerably higher than the reported threshold levels of 71 mg/100 g. The presence of oxalates in taro is known as to cause acridity, absorptive poisoning and binds calcium thereby inhibiting its absorption. In Ghana, Colocasia esculenta samples are third taro symposium 233 consumed as much as Xanthosoma. The appreciably low oxalate levels found in Colocasia samples grown in Ghana may account for its wide consumption especially in the coastal and forest regions. Effect of processing method on calcium oxalate content Effect of air-oven and solar drying Figure 1 shows the effect of variety and size of cocoyam slices on oxalate levels dehydrated using the air-oven method at different drying times (12 and 24 hours). For the 12-hour dried samples, the Xanthosoma sagittifolium (white-flesh) and the Colocasia esculenta samples showed decreasing effect on oxalate content as sample size was increased. The effect was more pronounced in the Colocasia samples than in the Xanthosoma samples. Ultra violet rays from the sun’s radiation reported to influence oxalate decomposition may account for the observed trends, since there was gradual increase in surface area from 20 g sample to 60 g sample. For the Xanthosoma sagittifolium (red-flesh) variety, the trend differed. Oxalate content decreased in the 40 g sample and then increased slightly in the 60 g samples. The trend of oxalate levels observed for the 24 hour dried samples was not different from that obtained in the 12 hour dried samples except that there was a steady increase in the oxalate values in the Xanthosoma sagittifolium (red-flesh) variety. The variation may be due to inherent genetic factors. As well, since the drying condition of temperature, relative humidity and air velocity were not constant during the drying periods, the variations observed could be due to such influences. Figure 1: Effect of variety and size on oxalate levels derived from 12-hour (A) and 24-hour (B) oven-dried chips The percentage retention of oxalate content for the samples are shown in Table 4. The retention values obtained for the 1 hour and 24 hour oven-dried samples ranged between 63.5-99.5% for Xanthosoma spp (red-flesh) and 49.5-86.0% for the Colocasia samples. This may be due to the effect provided by the large cross-sectional area of the corms of Colocasia, which had greater exposure to heat penetration. The analysis of variance conducted on the results of the oven-dried samples in Table 4 indicated that variety and size had significant effect (p<0.05) on oxalate levels. However, drying time did not show any significant effect (p<0.05). Table 4: Percentage retention of oxalate composition in the processed cocoyam samples Oven-dried Size (g) X. sagittifolium (red-flesh) X. sagittifolium (white-flesh) Colocasia esculenta Solar-dried 12-hour 24-hour 2-day 3-day 20 78 77.5 83.5 53.5 40 63.5 66.5 60 48.5 60 71.5 99.5 34.5 40 20 89 86.5 68.5 88.5 40 83.5 82 36.5 63.5 60 74.5 73 31.5 54.5 20 86 86 62.5 58 40 63 58 35.5 41.5 60 49.5 49.5 37 38.5 Figure 2 shows the oxalate values obtained when sizes of varieties were solar-dried for 2 days and 3 days. An average of 8 hours exposure to solar radiation per day was used. Moisture content of the solar-dried products ranged from 6-22%. This is due to wide range of relative humidity and solar radiation observed during drying. For the two Xanthosoma samples, and the drying times showed decreasing effect on oxalate levels as corm size was increased. The decreasing effect was more pronounced in the 2 day solar-dried samples as compared to the 3 day solar-dried samples. Oxalate retention was higher in the Xanthosoma sagittifolium (red - flesh) variety than in the Xanthosoma sagittifolium 234 third taro symposium (white-flesh) variety (Table 4). This may be due to high oxalate levels obtained in fresh samples as compared to the Xanthosoma sagittifolium (white-flesh) samples. The pattern of oxalate levels obtained for the Colocasia esculenta was slightly different from that of the Xanthosoma samples. Oxalate values for the 2-day solar-dried samples showed slight increases from the 40g sample to the 60g sample, whereas in the 3-day solar-dried sample decreases were observed. Low oxalate retention values were recorded for Colocasia, as compared to the Xanthosoma samples. Figure 2: Effect of variety and size on oxalate levels derived from 2-day (A) and 3-day (B) solar-dried chips In general, the pattern of oxalate levels obtained for the solar-dried samples were similar to that of the oven-dried samples except that lower oxalate values were obtained in the solar-dried samples than the oven-dried samples. This observation can be seen from the percentage retention values (Table 4). ANOVA conducted on the data revealed that variety, time and size all had significant (p<0.05) effects on the oxalate levels of the solar-dried samples. Effect of drum drying The taro samples were given two forms of treatments (dry and wet milling) and subsequently drum-dried. The oxalate levels measured and percentage retention values are presented in Table 5. Significant variations but few consistent trends were observed in the data. For the two treatments, lower oxalate values were observed for the dry-milled samples. This might be attributed to the effects of drying and milling processes which could have contributed to oxalate degradation. The oxalate contents of the products did not show any wide variation. However, comparing the results of the drumdried products to values obtained in the fresh samples showed significant changes in the rate of oxalate reduction. Table 5: Composition (µg/100g) and percentage retention of oxalates in drum-dried cocoyam products X. sagittifolium (red-flesh) Condition Oxalate composition % retention Dry milled 143.84 46.5 Wet milled 156.61 50.5 X. sagittifolium (white-flesh) Dry milled 99.94 32.5 Wet milled 159.01 53.5 Colocasia esculenta Dry milled 163.48 35.5 Wet milled 191.61 41.5 Drum-drying reduced oxalate levels by approximately 50% to average levels ranging from 99.94 to 191.61 µ/100g. However, these values were higher in comparison to the results obtained by Onayeni and Nwigwe (1987) when samples of Xanthosoma and Colocasia samples were sliced, soaked and boiled. Oxalate levels ranging from 9% to 26% were reported. High temperatures are known to cause the calcium oxalate containing cells (raphides) to collapse leading to the breakdown of oxalate structure. The mechanism of oxalate reduction by heat has not been fully understood. Generally, the rate of oxalate decomposition was higher in Colocasia, than the Xanthosoma samples. Conclusions The chemical composition of the three cultivars of taro, Xanthosoma sagittifolium (red-flesh and white-flesh) and Colocasia esculenta, shows wide variations among the varieties and across their respective corms. The Xanthosoma (white-flesh) variety had the highest levels of nutritive value. Similarly, the distal sections of the taro species studied had comparatively higher amounts proximate composition than the middle and apical sections. High levels of minerals are located at the apical sections as compared to the distal and middle sections. With respect to the distal and middle sections, the concentrations of minerals did not show any significant variation. Wide variation in oxalate contents exist third taro symposium 235 between the Xanthosoma sagittifolium and Colocasia esculenta, with the Colocasia having relatively higher oxalate levels. The various processing methods used reduced the oxalate levels by approximately 50%. The greatest reduction was observed for the drum-dried products, which reduced the oxalate contents to safer levels. Processes that eliminate oxalates in taro are critical for the development of taro food products. Therefore, drum-drying, solar and oven drying techniques can be used to develop taro products with reduced oxalate levels. These dehydration techniques are viable means of developing new products from taro, thereby increasing the market availability of taro products. References Afoakwa, E.O. and Sefa-Dedeh, S. 2001. Chemical composition and quality changes in trifoliate yam Dioscorea dumetorum pax tubers after harvest. Food Chemistry 75(1):85–91. Agbor-Egbe, T. and Rickard, J.E. 1990. Evaluation of the chemical composition of fresh and stored edible aroids. Journal of the Science of Food and Agriculture 53:487–495. AOAC. 1990. Official methods of analysis of the Association of Official Analytical Chemists. 15th edn, Vol. 2. AOAC, Washington, DC. Bradbury, J.H. and Halloway, J. 1988. The chemical composition of tropical root crops. ASEAN Food Journal 4:34– 38. Fennema, O.R. (ed.) 1988. Food chemistry. 2nd ed. Marcel Dekker, New York. 991 p. Huang, A.S. and Tanadjadja, L.S. 1992. Application of anion-exchange high performance liquid chromatography in determining oxalates in taro (Colocasia esculenta) corms. Journal of Agricultural and Food Chemistry 40(11):2123– 2126. Lauzon, R.D. and Kawabata, A. 1988. Physico-chemical evaluation of cocoyam starches. Philippines Journal of Crop Science 13:16–21. Onayemi, O. and Nwigwe, N.C. 1987. Effect of processing on the oxalate content of cocoyam. Food Technology 20:293–295. Onwueme, I.C. 1982. The tropical tuber crops: Yams, cassava, sweet potato and cocoyams. Wiley, Chichester. 234 p. Purseglove, J.W. 1988. Tropical crops: Monocotyledons. Longman Scientific and Technical, Harlow, England. 607 p. Rehm, S. and Espig, G. 1991. The cultivated plants of the tropics and subtropics: Cultivation, economic value, utilization. Josef Margraf, Weikersheim, Germany. 552 p. Sakai, W.S. 1979. Aroid root crops, acidity and raphides. p. 265–268. In: Charalambous, G. and Inglett, G.E. (eds). Tropical foods: Chemistry and nutrition. Vol. 1. Academic Press, New York. Sotozono, M. 1989. Ice cream sherbet containing taro as main raw material. Japanese Patent No. 03098539A. Watanabe, F.S. and Olsen, S.R. 1985. Test of ascorbic acid method for determining phosphorus in water and sodium bicarbonate extract from soils. Soil Science Society of America Journal 29:677–678. Wills, R.B.H., Lim, J.S.K., Greenfield, H. and Bayliss-Smith, T. 1983. Nutrient composition of taro (Colocasia esculenta) cultivars from the Papua New Guinea highlands. Journal of the Science of Food and Agriculture 34:1137–1142. 236 third taro symposium LIST OF PARTICIPANTS Australia Dr Grahame Jackson 24 Alt St, Queens Park Sydney NSW 2022 Tel: (612) 93878030 Fax: (612) 93878004 Email: [email protected] Prof. David James Midmore Central Queensland University Rockhampton QLD 4702 Tel: (617) 49309770 Fax: (617) 49309255 Email: [email protected] Dr Richard James Milner CSIRO 81 Birriwa Rd Bungendore NSW 2621 Tel: (612) 62369212 Fax: (612) 62464042 Email: [email protected] Mr David Hicks NORADA NSW Agriculture P O Box 703, Richmond Sydney NSW 2753 Tel: (614) 29645691 Fax: (612) 45701314 Email: [email protected] Mr Phillippe Jean Petiniaud NQ Taro Growers P O Box 283 Babinda QLD 4861 Tel: (617) 40672078 Fax: (617) 40672078 Email: [email protected] Mr Jeffrey William Daniells Queensland Department of Primary Industries P O Box 20 South Johnstone QLD Tel: (617) 40641130 Fax: (617) 40642249 Email: [email protected] Mr Peter Louis Salleras Taro Growers Australia Inc P O Box 1095 Tully QLD 4854 Tel: (617) 4058 6104 Fax: (617) 4068 6104 Email: [email protected] Ms Vilma Amante University of Queensland 59 Akala Street, Camp Hill QLD 4152 Tel: (617) 33952159 Email: [email protected] Dr Ian Godwin School of Land and Food Sciences University of Queensland, St Lucia Campus QLD Tel: (617) 33652141 Fax: (617) 33651177 Email: [email protected] Mr Hunter Laidlaw School of Land and Food Sciences University of Queensland, St Lucia Campus QLD Tel: (617) 33651518 Fax: (617) 33651177 Email: [email protected] Dr Rob Harding School of Life Science Queensland University of Technology, Gardens Point Campus GPO Box 2434 Brisbane QLD 4001 Tel: (617) 38641379 Fax: (617) 38641534 Email: [email protected] Dr Peter Andrew Revill School of Life Science Queensland University of Technology, Gardens Point Campus GPO Box 2434 Brisbane QLD 4001 Tel: (618) 38645218 Fax: (617) 38641534 Email: [email protected] Commonwealth of Dominica Mr Gregory Carlson Robin Caribbean Agricultural Research & Development Institute P O Box 346 Roseau Tel: (1767) 4482715 Fax: (1767) 4485690 Email: [email protected] Cook Islands Mr William Wigmore Director of Research Ministry of Agriculture P O Box 96 Rarotonga Tel: (682) 28711, 26720 Fax: (682) 21881 Email: [email protected] third taro symposium 237 Federated States of Micronesia Mr Moses Asher Micronesia Plant Propagation Research Centre (MPPRC) P O Box 1000 Tofol Kosrae Tel: (691) 3702768 Fax: (691) 3702768, 3703952 Email: [email protected] Mr Adelino S. Lorens Agriculture Pohnpei Office of Economic Affairs Pohnpei State Government P O 1028 Kolonia Pohnpei 96941 Tel: (691) 3202400 Fax: (691) 320 2127 Email: [email protected] Fiji Islands Mr Samisoni Ulitu Deputy Secretary Agricultural Development Ministry of Agriculture, Sugar and Land Resettlement Private Mail Bag Raiwaqa Tel: (679) 3384233 Fax: (679) 3385254 Email: [email protected] Mr Aliki Turagakula Principal Research Officer (Agronomy) Ministry of Agriculture, Sugar and Land Resettlement P O Box 77 Nausori Tel: (679) 347 7044, 3390739 Fax: (679) 347 7546 Email: [email protected] Ms Joann Young Ministry of Agriculture, Sugar and Land Resettlement Private Mail Bag Raiwaqa Tel: (679) 3384233 Fax: (679) 3385234 Email: [email protected] Ms Mereia Fong Ministry of Agriculture, Sugar and Land Resettlement P O Box 77 Nausori Tel: (679) 3477044 Fax: (679) 3477546 Email: [email protected] Mr Moti Lal Autar Ministry of Agriculture, Sugar and Land Resettlement P O Box 77 Nausori Tel: (679) 3477044 Fax: (679) 3477546 Email: [email protected] 238 third taro symposium Mr Graeme Stephen Thorpe Balthan International (Fiji) Limited P O Box 1228 Suva Tel: (679) 3398912, 3398915 Fax: (679) 3398916 Email: [email protected] Dr Richard Beyer Food Scientist Suva Tel: 3370709 Mr Sam Foy Chung Rootcrop Council P O Box 717 Suva Tel: (679) 3300892, 3320784 Fax: (679) 3320043 Email: [email protected] Mr Apaitia Macanawai P O Box 17703 Suva Tel: (679) 3383582 Email: [email protected] France Ms Sophie Farah Caillon IRD/CIRAD 5 Rue du Carbone 45072 Cedex 2 Orleans Tel: (675) 9839282 Fax: (675) 9839247 Email: [email protected] French Polynesia Dr Charles Garnier Service du developpement rural B P 100 Papeete Tahiti Tel: (689) 574004 Fax: (689) 574690 Email: [email protected] Ghana Mr Emmanuel Ohene Afoakwa Department of Nutrition and Food Science University of Ghana P O Box 134 Accra Tel: (233) 24685893 Fax: (233) 21500389 Email: [email protected] Hawaii Dr John J. Cho University of Hawaii P O Box 269 Kula Hawaii 96970 USA Tel: (1808) 8781213 Fax: (1808) 8786804 Email: [email protected] India Dr Sivasubramanian Edison Director Central Tuber Crops Research Institute Trivandrum 692017 Kerala Tel: (91471) 2598431 Fax: (91471) 2590063 Email: [email protected] Dr M.T. Sreekumari Senior Scientist Central Tuber Crops Research Institute Trivandrum 692017 Kerala Tel: (471) 28551 Fax: (471) 2590063 Email: [email protected] Indonesia Dr Made Sri Prana Research Centre for Biotechnology PROSEA Herbarium Bogoriense Jalan Ir. H Juanda 22 Bogor Tel: (62) 251322859, 251337762 Fax: (62) 251370934 Email: [email protected] Marshall Islands Dr Dilip Nandwani Researcher Department of Cooperative Research and Extension College of the Marshall Islands P O Box 1258 MH 96960 Tel: (692) 6252299, 6255340 Fax: (692) 6255340 Email: [email protected] New Caledonia M. Didier Varin Direction de l’Agriculture et de la Foret B P 259 98822 Poinoinie Tel: (687) 255107, 427252 Fax: (687) 255129, 427376 Email: [email protected] New Zealand Dr Robert A. Fullerton Horticulture and Food Research Institute of New Zealand P B 92169 Auckland Tel: (649) 8154200 ext. 7334 Fax: (649) 8154200, 8154225 Email: [email protected] Dr Zhi-Qiang Zhang Landcare Research P B 92170 120 Mt. Albert Rd Auckland Tel: (649) 8154200 Email: [email protected] Dr William Thomas Bussell UNITEC Institute of Technology Private Mail Bag 92025 Auckland Tel: (649) 8154321 ext. 7802 Fax: (649) 8154346 Email: [email protected] Dr Nancy J. Pollock Victoria University, Wellington 12 Pingau Street Paekakariki Tel: (644) 9058687 Email: [email protected] Niue Mr Colin Etuata Tongatule Director of Agriculture, Forestry and Fisheries Department of Agriculture, Forestry and Fisheries P O Box 74 Fonuakula Alofi Tel: (683) 4032 Fax: (683) 4079 Email: [email protected] Palau Mr Herman Francisco Chief of the Division of Agriculture & Mineral Resources Bureau of Resources and Development Division of Agriculture P O Box 460 Koror 96940 Tel: (680) 4881517, 4881475, 4888171 Fax: (680) 4881725 Email: [email protected] Mr Robert Victor Bishop BOA - FAO P O Box 460 Koror 96940 Tel: (680) 4888051 Email: [email protected] Papua New Guinea Dr Geoffrey Christopher Wiles NARI Chief Scientist National Agricultural Research Institute P O Box 4415 Lae 411 Morobe Province Tel: (675) 4721751, 4721752 Fax: (675) 4722242 Email: [email protected] third taro symposium 239 Mr Andrew Yamanea Secretary for Agriculture & Livestock Department of Agriculture and Livestock P O Box 20331 Konedobu NCD Tel: (675) 3212839 Fax: (675) 3211387 Prof. Terence Vincent Price University of Vudal Rabaul ENBP 611 Tel: (675) 9839282 Fax: (675) 9839247/9166 Email: [email protected] Mr Tom Okpul University of Vudal Rabaul ENBP 611 Tel: (675) 9839144 Fax: (675) 9839247 Email: [email protected], [email protected] Prof. Mohammed Wagih Biotechnology Centre University of Technology Lae Tel: (675) 4734455 Fax: (675) 4734477 Email: [email protected] Mr Roy T. Masamdu National Agricultural Research Institute P O Box 1639 Lae Tel: (675) 4751189 Fax: (675) 4722242 Email: [email protected] Samoa Ms Laisene Samelu Principal Horticultural Development Officer Ministry of Agriculture, Forestry, Fisheries & Meteorology P O Box 1874 Apia Tel: (685) 20605, 23416, 23246 Fax: (685) 23996/20607 Email: [email protected] Prof. Alfred E. Ebenebe Pro Vice Councellor (Alafua Campus) Cum Head of School, Agriculture University of the South Pacific Private Mail Bag Apia Tel: (685) 21674 Fax: (685) 22933, 26232 Email: [email protected] 240 third taro symposium Dr Kwadwo Ofori Plant Breeder University of the South Pacific School of Agriculture Private Mail Bag Apia Tel: (685) 21671 Fax: (685) 22933, 26232 Email: [email protected] Mr Anthony Palupe University of the South Pacific School of Agriculture Private Mail Bag Apia Tel: (685) 21671 Fax: (685) 22933 Email: [email protected] Solomon Islands Ms Roselyn Kabu Maemouri Kastom Garden Association P O Box 742 Honiara Tel: (677) 39551 Fax: (677) 39551 Email: [email protected] Ms Ellen Iramu Ministry of Agriculture & Livestock P O Box G13 Honiara Tel: (677) 27987 Fax: (677) 27380 Email: [email protected] Tonga Dr Pita Taufatofua Ministry of Agriculture and Forestry P O Box 14 Nuku’alofa Tel: (676) 32125 Fax: (676) 32253 Email: [email protected] Paul Karalus Managing Director Pacific Biotech Limited P O Box 1039 Nuku’alofa Tel: (676) 29884, 8781657 Fax: (676) 29885 Email: [email protected] Vanuatu Dr Vincent Lebot CIRAD P O Box 946 Port Vila Tel: (678) 25947 Fax: (678) 25947 Email: [email protected] Mr Abel Tapisuwe Foundation for the People of the South Pacific (FSP) P O Box 951 Port Vila Tel: (678) 22915 Fax: (678) 24510 Email: [email protected] Wallis & Futuna Mr Francois Nuttens Service d’Etat de l’Agriculture, de la Forêt et de la Pêche Services Territoriaux de l’Economie Rurale et de la Pêche B P 19 Matautu 9800 Uvea Tel: (681) 720400 Fax: (681) 720404 Email: [email protected] Donor Agencies / Organisations Australian Agency for International Development (AusAID) Mr Ian Kershaw AusAID G P O Box 887 Canberra ACT 2602 Australia Tel: (612) 262064841 Email: [email protected] International Plant Genetic Resources Institute (IPGRI) Dr Coosje Hoogendoorn International Plant Genetic Resources Institute Via Del Tre Denari 472 A Maccarese Rome 00057 Italy Tel: (06) 6118200 Email: [email protected] Dr Prem Narain Mathur International Plant Genetic Resources Institute Regional Office for Asia, the Pacific & Oceania IPGRI South Asia Office NASC Complex, Pusa Campus New Delhi India Tel: (9111) 25827596 Fax: (9111) 25819899 Email: [email protected] Dr Ramanatha Rao Senior Scientist - Genetic Diversity/Conservation International Plant Genetic Resources Institute Regional Office for Asia, the Pacific & Oceania P O Box 236 UPM Post Office, Serdang 43400 Selangor Darul Ehsan Malaysia Tel: (603) 89423891 ext. 204 Fax: (603) 89487655 Email: [email protected] Food and Agriculture Organization of the United Nations (FAO) Dr Narasimha Murthi Anishetty Senior Officer - Plant Genetic Resources Seed and Plant Genetic Resources Service (AGPS) Rome Italy Tel: (39) 0657054652 Fax: (39) 0657056347 Email: [email protected] Mr Ricki Faatonu FAO TCP/SAM/0165 (A) (Taro project, Samoa) MAFFM Vaivase-UTA P O Box 681 Apia Samoa Tel: (685) 20719, 23416 Fax: (685) 24576, 21865 Mr Winston Charles Office of the FAO Sub-Regional Representative for the Pacific Private Mail Bag Apia Samoa Fax: (685) 22126 Email: [email protected] Secretariat of the Pacific Community (SPC) Mr Tom Osborn Agriculture Advisor Private Mail Bag Suva Fiji Tel: (679) 3370733 Fax: (679) 3370021 Email: [email protected] Dr Mary Taylor Regional Germplasm Center (RGC) Advisor Email: [email protected] Mr Luigi Guarino Plant Genetic Resources Advisor Email: [email protected] Dr Danny Hunter DSAP Team Leader Email: [email protected] third taro symposium 241 Dr Davinder Singh Plant Breeder Secretariat of the Pacific Community C/o NARI P O Box 1639 Lae Papua New Guinea Tel: (675) 4751198 Fax: (675) 4722242 Email: [email protected] Mr Tolo Iosefa TIP Manager University of the South Pacific Alafua Campus Private Mail Bag Apia Samoa Tel: (685) 21671 Fax: (685) 22933 Email: [email protected] Dr Mick Lloyd Plant Protection Advisor Email: [email protected] Dr Richard Davis Plant Virologist Email: [email protected] Mr Stephen Hazelman Extensionist Email: [email protected] Mr Sada Nand Lal Entomologist Email: [email protected] Mr Fereti Atumurirawa Taro Bettle Technician Email: [email protected] 242 third taro symposium Mr Takaniko Ruabete Plant Pathology Technician Email: [email protected] Ms Valerie Tuia RGC Curator Email: [email protected] Mr Eliki Lesione Senior Lab Technician, RGC Email: [email protected] Mr Rajnesh Sant Senior Lab Technician, RGC Email: [email protected] Ms. Raghani Prasad Graduate Assistant, RGC Email: [email protected] Ms Rohini Prasad Laboratory Assistant, RGC Email: [email protected] Ms Ana Vosaki Laboratory Assistant, RGC Email: [email protected] Ms Laisa Tigarea Programme Secretary Email: [email protected] Ms Vandna Lal Project Assistant Email: [email protected]