3rdTaro Symposium 3rdTaro Symposium

Transcription

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
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
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
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
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.
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
• 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
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
• 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
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
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
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).
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.
• 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.
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.
17
Principaux points
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.
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.
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.
• 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.
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.
19
Principaux points
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.
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
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
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
• 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.
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
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
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
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.
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
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.
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
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.
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
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
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
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.
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
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
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).
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
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
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
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.
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
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
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
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.
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
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.
56 p.
45
Kuruvialla, K.M. and Singh, A. 1981. Karyotypic and electrophoretic studies on taro and its origins. Euphytica 30:405–
412.
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.
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.
46
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
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
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.
49
Promoting on farm conservation of taro
through diversity fairs in the Solomon Islands
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
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.
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.
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.
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
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).
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
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.
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
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.
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.
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
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
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
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
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
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.
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
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.
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
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
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.
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.
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.
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.
68
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
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
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).
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
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.
73
Theme One Paper 1.5
Promoting on farm conservation of taro through
diversity fairs in the Solomon Islands
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
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
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
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.
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
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.
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.
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
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.
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
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
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
Theme One Paper 1.7
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.
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
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
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
Theme One Paper 1.8
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.
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
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
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
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.
93
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.
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 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
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.
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.
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
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
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
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.
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.
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
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
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
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.
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
Theme Two Paper 2.3
The biology of Phytophthora colocasiae
and implications for its management and control
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
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
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
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
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.
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
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.
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
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.
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
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.
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
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.
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
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
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
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.
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
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
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
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
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
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
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
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.
127
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
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.
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
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.
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.
131
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
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.
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.
133
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
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).
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
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.
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
Figure 7: C. esculenta var. PSB-G2
(tissue culture)
Figure 9: Taro aphids (Aphis gossypii) and mites
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.
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
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.
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
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.
143
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
“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!
145
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
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.
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
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
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
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.
151
Taro production, constraints and future research and
development programme in Indonesia
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
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
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
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.
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.
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.
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
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.
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
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
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
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.
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
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.
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.
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
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.
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.
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
*
*
* – 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
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.
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
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.
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
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
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.
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
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.
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.
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
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.
175
Table 1: Accessions used in this study (according to Kreike et al. 2003)
176
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)
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
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.
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
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
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.
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
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
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.
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
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.
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
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.
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
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.
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
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.
191
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
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.
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.
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.
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
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
Niue
53
7
13.2
T6
47
6
12.8
Veo
12
2
16.7
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
Lauloa Keokeo
27
0
0
Van 26
63
13
20.6
Veo
28
1
3.6
Kai Ala
84
8
9.5
T2
12
0
0
Van 96
15
3
20
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 Eleele Naioea 9]
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
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.
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.
196
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).
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
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.
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
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.
201
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
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
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.
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
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.
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
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.
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
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
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
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.
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
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
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
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.
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
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
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
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.
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
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.
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)
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
221
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
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.
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
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.
225
Recent developments in taro-based
food products in Hawaii
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
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.
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
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.
229
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
(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
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.
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
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
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
(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
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
LIST OF PARTICIPANTS
Australia
Dr Grahame Jackson
24 Alt St, Queens Park
Sydney
NSW 2022
Tel: (612) 93878030
Fax: (612) 93878004
Prof. David James Midmore
Central Queensland University
Rockhampton
QLD 4702
Tel: (617) 49309770
Fax: (617) 49309255
Dr Richard James Milner
CSIRO
81 Birriwa Rd
Bungendore
NSW 2621
Tel: (612) 62369212
Fax: (612) 62464042
Mr David Hicks
NORADA
NSW Agriculture
P O Box 703, Richmond
Sydney
NSW 2753
Tel: (614) 29645691
Fax: (612) 45701314
Mr Phillippe Jean Petiniaud
NQ Taro Growers
P O Box 283
Babinda
QLD 4861
Tel: (617) 40672078
Fax: (617) 40672078
Mr Jeffrey William Daniells
Queensland Department of Primary Industries
P O Box 20
South Johnstone
QLD
Tel: (617) 40641130
Fax: (617) 40642249
Mr Peter Louis Salleras
Taro Growers Australia Inc
P O Box 1095
Tully
QLD 4854
Tel: (617) 4058 6104
Fax: (617) 4068 6104
Ms Vilma Amante
University of Queensland
59 Akala Street, Camp Hill
QLD 4152
Tel: (617) 33952159
Dr Ian Godwin
School of Land and Food Sciences
University of Queensland, St Lucia Campus
QLD
Tel: (617) 33652141
Fax: (617) 33651177
Mr Hunter Laidlaw
School of Land and Food Sciences
University of Queensland, St Lucia Campus
QLD
Tel: (617) 33651518
Fax: (617) 33651177
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
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
Commonwealth of Dominica
Mr Gregory Carlson Robin
Caribbean Agricultural Research & Development
Institute
P O Box 346
Roseau
Tel: (1767) 4482715
Fax: (1767) 4485690
Cook Islands
Mr William Wigmore
Director of Research
Ministry of Agriculture
P O Box 96
Rarotonga
Tel: (682) 28711, 26720
Fax: (682) 21881
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
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
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
Mr Aliki Turagakula
Principal Research Officer (Agronomy)
P O Box 77
Nausori
Tel: (679) 347 7044, 3390739
Fax: (679) 347 7546
Ms Joann Young
Private Mail Bag
Raiwaqa
Tel: (679) 3384233
Fax: (679) 3385234
Ms Mereia Fong
P O Box 77
Nausori
Tel: (679) 3477044
Fax: (679) 3477546
Mr Moti Lal Autar
P O Box 77
Nausori
Tel: (679) 3477044
Fax: (679) 3477546
238
Mr Graeme Stephen Thorpe
Balthan International (Fiji) Limited
P O Box 1228
Suva
Tel: (679) 3398912, 3398915
Fax: (679) 3398916
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
Mr Apaitia Macanawai
P O Box 17703
Suva
Tel: (679) 3383582
France
Ms Sophie Farah Caillon
IRD/CIRAD
5 Rue du Carbone
45072 Cedex 2
Orleans
Tel: (675) 9839282
Fax: (675) 9839247
French Polynesia
Dr Charles Garnier
Service du developpement rural
B P 100
Papeete
Tahiti
Tel: (689) 574004
Fax: (689) 574690
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
Hawaii
Dr John J. Cho
University of Hawaii
P O Box 269
Kula
Hawaii 96970
USA
Tel: (1808) 8781213
Fax: (1808) 8786804
India
Dr Sivasubramanian Edison
Director
Central Tuber Crops Research Institute
Trivandrum 692017
Kerala
Tel: (91471) 2598431
Fax: (91471) 2590063
Dr M.T. Sreekumari
Senior Scientist
Central Tuber Crops Research Institute
Trivandrum 692017
Kerala
Tel: (471) 28551
Fax: (471) 2590063
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
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
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
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
Dr Zhi-Qiang Zhang
Landcare Research
P B 92170
120 Mt. Albert Rd
Auckland
Tel: (649) 8154200
Dr William Thomas Bussell
UNITEC Institute of Technology
Private Mail Bag 92025
Auckland
Tel: (649) 8154321 ext. 7802
Fax: (649) 8154346
Dr Nancy J. Pollock
Victoria University, Wellington
12 Pingau Street
Paekakariki
Tel: (644) 9058687
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
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
Mr Robert Victor Bishop
BOA - FAO
P O Box 460
Koror 96940
Tel: (680) 4888051
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
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
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
Mr Roy T. Masamdu
National Agricultural Research Institute
P O Box 1639
Lae
Tel: (675) 4751189
Fax: (675) 4722242
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
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
240
Dr Kwadwo Ofori
Plant Breeder
School of Agriculture
Private Mail Bag
Apia
Tel: (685) 21671
Fax: (685) 22933, 26232
Mr Anthony Palupe
School of Agriculture
Private Mail Bag
Apia
Tel: (685) 21671
Fax: (685) 22933
Solomon Islands
Ms Roselyn Kabu Maemouri
Kastom Garden Association
P O Box 742
Honiara
Tel: (677) 39551
Fax: (677) 39551
Ms Ellen Iramu
Ministry of Agriculture & Livestock
P O Box G13
Honiara
Tel: (677) 27987
Fax: (677) 27380
Tonga
Dr Pita Taufatofua
Ministry of Agriculture and Forestry
P O Box 14
Nuku’alofa
Tel: (676) 32125
Fax: (676) 32253
Paul Karalus
Managing Director
Pacific Biotech Limited
P O Box 1039
Nuku’alofa
Tel: (676) 29884, 8781657
Fax: (676) 29885
Vanuatu
Dr Vincent Lebot
CIRAD
P O Box 946
Port Vila
Tel: (678) 25947
Fax: (678) 25947
Mr Abel Tapisuwe
Foundation for the People of the South Pacific (FSP)
P O Box 951
Port Vila
Tel: (678) 22915
Fax: (678) 24510
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
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
(IPGRI)
Dr Coosje Hoogendoorn
Via Del Tre Denari 472 A
Maccarese
Rome 00057
Italy
Tel: (06) 6118200
Dr Prem Narain Mathur
Regional Office for Asia, the Pacific & Oceania
IPGRI South Asia Office
NASC Complex, Pusa Campus
New Delhi
India
Tel: (9111) 25827596
Fax: (9111) 25819899
Dr Ramanatha Rao
Senior Scientist - Genetic Diversity/Conservation
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
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
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
Secretariat of the Pacific Community (SPC)
Mr Tom Osborn
Agriculture Advisor
Private Mail Bag
Suva
Fiji
Tel: (679) 3370733
Fax: (679) 3370021
Dr Mary Taylor
Regional Germplasm Center (RGC) Advisor
Mr Luigi Guarino
Plant Genetic Resources Advisor
Dr Danny Hunter
DSAP Team Leader
241
Dr Davinder Singh
Plant Breeder
C/o NARI
P O Box 1639
Lae
Papua New Guinea
Tel: (675) 4751198
Fax: (675) 4722242
Mr Tolo Iosefa
TIP Manager
Alafua Campus
Private Mail Bag
Apia
Samoa
Tel: (685) 21671
Fax: (685) 22933
Dr Mick Lloyd
Plant Protection Advisor
Dr Richard Davis
Plant Virologist
Mr Stephen Hazelman
Extensionist
Mr Sada Nand Lal
Entomologist
Mr Fereti Atumurirawa
Taro Bettle Technician
242
Mr Takaniko Ruabete
Plant Pathology Technician
Ms Valerie Tuia
RGC Curator
Mr Eliki Lesione
Senior Lab Technician, RGC
Mr Rajnesh Sant
Senior Lab Technician, RGC
Ms. Raghani Prasad
Graduate Assistant, RGC
Ms Rohini Prasad
Laboratory Assistant, RGC
Ms Ana Vosaki
Laboratory Assistant, RGC
Ms Laisa Tigarea
Programme Secretary
Ms Vandna Lal
Project Assistant

3rdTaro Symposium 3rdTaro Symposium

Transcription

Documents pareils

Abortion overview - Classes

statistiques des ventes dans notre industrie sales statistics in our

Revue de presse ANGLAIS Semaine du 31 octobre au 06 novembre

surfboard - Villard de Lans

[Descartes e il Seicento 5] Jean-Robert Armogathe, Vincent Carraud

View - Prince George Digitization

Revue de presse ANGLAIS Semaine du 24 au 30 octobre 2011

4228 Holland Dr Des Moines, IA 50310 | Home for Sale | Real

Imagine you are working for the Scottish Tourist Board and you are

harwood plaza- ajax - aug 3-16