Enset Research and

Development
r
Experiences
r
in Ethiopia

Edited by
Mohammed Yesuf (PhD)
Tariku Hunduma (MSc)

E 1A R M oA SA RI
 Enset Research and
Development Experiences
111 Ethiopia

Proceedings o f Sunset XaiionaS Workshop

1 9 - 2 0 A u g u st 2 0 1 0 ,

W o lk itc , E th io p ia

In d ite d In9

Mohan?.3m><B Yesuf* ( P h D )
T a r i k i i Ilsiiidiiiiia ( M S c )

Ethiopian Institute of Agricultural Research

Ministry of Agriculture
Southern Agricultural Research Institute
2012
© E IA R , 2012
h . " ) r '/ \ / - 21)04

Website: http:/ /www.eiar.eov-ot

Tel.: J (251-1)11-646 26 33
Fax: ] (251-1) 11-646 12 94
P.O.Box 2003, Addis Abeba

IS B N : 978-99944-991-4-4

T h e \ ievvs
i exp ressed in the p roceed in g are of the au th o rs and not necessarily
the E h iop ian In stitu te o f A gricultural R esearch (EIA R).

Citai ion
M oh aJnm ed Y. and T ariku H. (eds.). 2012. Enset R esearch and D ev elop m en t
E xp erien ces in E th io p ia. P roceed in gs of E n set N ational W o rk sh o p , 19-20
A ugu st 2010, W o lk ite, Ethiopia.
 C o n ten ts

Acronyms i

Preface iii

Acknowledgments iv

Opening Speech
H.E A to T efera D erbew v
M inster, M in istry o f A gricu ltu re

Technological Challenges of Sustainable Enset Farming System:

For Enhancing the Production of Food/Fiber and Industrial Outputs
T ay e BezuneJi 1

Past Research Achievement and Existing Gaps on Enset

(Ensete ventricosuni (welw.) Cheesman) Breeding
M ik ia s Y esh itila, an d Z erilm n Yem atazv 22

Research Experiences on Enset Biotechnology in Ethiopia: A Review

T esfa y e D isa sa a n d M u lugeta D iro 37

Research and Development Experiences on Enset Agronomy

M ulugeta D iro a n d A dn iasu T seg a y e 4S

Research Achievements, Experiences and Future Direction on

Bacterial Wilt of Enset
F ikre H an d oro, T ariku Hundunm a n d E n d ale H ailu 64

Experience on Bacterial Wilt Management in Eastern Africa:

The Case of Banana
E shetu A hnted a n d M o h a m m ed Y esu f 74

Regional Strategy for Banana/Enset Bacterial Wilt

Management in East Africa: The Road Map for the National Action Plan
T esfahun F cn ta a n d E ld a d K arm u ra 87

The Essence of Domestic Quarantine against Enset Bacterial

Wilt during Technology Dissemination in Ethiopia
D ereje G orfu 97

Mole Rat and Porcupine as Production Threat and their

Management Options
L eu lseg ed B eg a sh a w , W aga M azen g ia a n d M ik ia s Y esh itela 110
Experiences on Enset Technology Dissemination and Value Chain Analysis
G eiah u n Degii 130

Research Experience and Achievements on Traditional

Enset (Ensete ventricosuni) Fermentation Processes
T a iik u H undum a 145

Research Experiences on the Development of Enset

Processing Equipment
F riiiu K eleniu 162

Research Achievements and Experiences on Enset Food Products

M tilugeta T ean iir a n d A d a n e T ilahun 17S

Gender Differentials in Enset Farming and Processing

Y eshi ChicJie 196
 Acronyms

AARC Areka Agricultural Research center

AFLP Amplified Fragment Length Polymorphic DNA
ARDPLC Agriculture and Rural Development Partners Linkage Council
ASARECA Association for Strengthening Agricultural Research n Eastern a
Central Africa
BARNESA Banana Research Network for Eastern and Southern Africa
BBW Banana Bacterial Wilt
BBWCI National Banana Bacterial Wilt Control Initiative
CBOs Community Based Organizations
CBT Community Based Trainers
CSA Central Statistical Agency
DAE District Agricultural Extension
DNA Deoxvriboneucleic Acid
DRC Democratic Republic of Congo
EBW Enset Bacterial Wilt
„ ECA Economic Commission for Africa
EFS Enset Farming System
EHNRI Ethiopian Health and Nutrition Research Institute
* EIAR Ethiopian Institute of Agricultural Research
ELISA Enzyme-Linked Immunosorbent Assay
ENC Enset Nursery Center
FAO Food and Agriculture Organization
FFS Farmer Field Schools
FRG Farmers Research Group
FTC Farmers Training Center
FYM Farmyard Manure
HARC Holetta Agricultural Research Center
IAC Inter Academy Council
IAR Institute of Agricultural Research
I IT A International Institute of Tropical Agriculture
INIBAP International Network for the Improvement of Banana and Plantain
IPM Integrated Pest Management
ISSR Inter-Specific Simple sequence Repeat
ITK Indigenous Technical Knowledge
LAB Lactic Acid Bacteria
MARC Melkassa Agricultural Research Center
MDG Millennium Development Goal
MoA Ministry of Agriculture
NAADS National Agriculture Advisory Services (Uganda)
* NARO National Agricultural Research Organization (Uganda)

[i]
NBRP National Banana Research Program (Uganda)
NGOs Non Governmental Organizations
NMM Net Marketing Margin
NPN Non-Protein Nitrogen
PCR Polymerase Cham Reaction
PDC Participatory Development Communication
PSCP Producers' Share of the Consumers Price
RAFD Random Amplified Polymorphic DNA
RCBP Rural Capacity Building Project
Rep-PCR Repetitive Sequence-Based PCR
rDNA recombinanat DNA
RTPC Rural Technology Promotion Center
SNNPRS Southern Nations, Nationalities and Peoples Regional State
TGMM Total Gross Marketing Margin
UNDP United Nation Development Program
WHO World Health Organization
YPGA Yeast Peptone Glucose Agar
 Preface

Enset (Ensete ventricosum) is the main crop of sustainable and indigenous cropping
system in Ethiopia that ensures food security for several millions of people.
Approximately 20% of the total population in Ethiopia depends upon enset for food.
Moreover, different parts enset are also widely used as feed, fiber and construction
material. Apart from its multipurpose use the enset plant has cultural and socio­
economic value mainly in the south and south-west parts of Ethiopia. Despite the
enormous and diverse importance of enset in the Ethiopian agriculture, a number of
production constraints are recorded, among which enset bacterial wilt caused by
Xanthomonas campastris pv. Musacearum is in the forefront. The National Agricultural
Research System in collaboration with local and International organizations made efforts
to study the sustainable cropping system of enset thereby to ensure food security.

A National workshop on enset research and development was organized with the
k objective to review and compile research and development experiences with special
focus on enset bacterial wilt. Participants of this National workshop include policy
makers, prominent agricultural scientists, researchers, industrialist and enset farmers.
The workshop mainly dealt with review papers presentation and live exhibitions. Paper
presentations were focused on enset improvement, agronomy, diseases and insect pests,
processing and utilization, technology transfer, role of gender on enset farming and
future areas of interventions towards enset improvement and sustainable enset bacterial
wilt management. The exhibition part includes achievements on enset biotechnology,
enset food products and utilization, enset mechanization and industrial products of
enset.

In all cases, achievements, gaps and future areas of intervention were discussed. The
identified challenges and gaps will give basis for further exploitation of enset to the level
of its potential. Particularly strategies for sustainable management of enset bacterial wilt
were suggested and designed based on the available experiences at national and regional
level. Strategies for sustainable management of banana bacterial wilt in East African
countries were also reviewed for further adoption against enset bacterial wilt. It is hoped
that this proceeding will serve as valuable source of information for researchers, policy
makers and teaching institutions on the major production constraints and distinctive
characteristic of enset cropping system.

E d ito rs

[iii]
 Acknowledgements

T he organizers o f the w orkshop would like to acknow ledge the Ethiopian Institute o f
Agricultural R esearch (E IA R ) and M inistry o f A griculture (M o A ) m anagem ent for makinu
available all the necessary facilities and providing logistic services during the conduct o f the
w orkshop. Southern A gricultural R esearch Institute and A reka A gricultural R esearch Center
are also deeply acknow ledged for their active involvem ent on the organization o f the
w orcshop. Sp ecia l thank goes to Dr. Abera D eressa (Form er State M inister, M o A R D ), Dr.
Solom on A ssefa (D irecto r G eneral, E IA R ) and Dr. A defris T eklew old (D irector. Crops
R esearch , E IA R ) for their unreserved support and all the valuable inputs they provided that
helpsd us to m ake the w orkshop a success. Last but not least the organizers would like to
acknow ledge the Rural C apacity Building Project (R C B P ) o f the M in istry o f A griculture and
Rural D evelopm ent for financing the workshop.
 O p en in g S p ee ch

I I .E Ato T cfera D erbeu

M inster, M inistry o f Agriculture

Oil behalf of the Ministry of Agriculture and Rural Development and on my behalf, I
would like to extend my sincere appreciation to the organizing committee of this
important enset national workshop.

Ladies and G entlem en,

Enset, a banana-like plant is known to exist in Asia and Africa. But it has been
known to be cultivated as a food source only in Ethiopia. Enset is the main crop of a
sustainable indigenous Ethiopian cropping system that ensures food security in a
country that is food deficient.

Enset based farming is an indigenous and sustainable agricultural system in the

country. Approximately more than 20% of Ethiopia's population depends on enset
for food, feed and fiber. Over 80% of enset produced in the country is grown in the
south and southwestern parts of the country. Enset is a multi-purpose and multi-year
crop. Due to its drought tolerance, enset plant is regarded as a priority crop in
Ethiopia, where it makes a major contribution to the food security of the country. It
has been known to buffer food deficit during dry spell and recurrent drought.

Ladies and G entlem en,

Because enset is very important crop, now its cultivation is extending to the other
parts of the country, where it was not cultivated widely before. One author wrote
that "enest plant is everything for the farmers"; it is their food, cloth, beds, houses,
cattle-feed and plates. A book, "Lost Crops o f Africa'' mentioned enset as "possibly no
plant on earth can match enset for the number of products it provides for the poor
people". Without doubt, 1 dare to say, the contribution of enset plant to the growing
economy of the country is tremendous especially towards the food security program.

Ladies and G entlem en,

Though enset is such a valuable crop plant, its production is alarmingly getting
threatened by various pests. Among others, bacterial wilt of enset is becoming the
major concern as an important threat to enset production. In the past, various
research attempts were carried out to control these economically important pests.
Despite all the attempts, the problems continue on unacceptable scale and looks for
integrated approach of all concerned bodies and designing of possible means to
tackle it. This National workshop on Enset Research and Achievements is timely and
appropriate to offer good opportunity to discuss all pertinent issues about this very
[v]
important crop plant, and design future strategic directions by involving all
concerned stakeholders. Hopefully, this workshop will create a conducive platform
to design research priorities, technology dissemination, and community mobilization
towards sustainable management of enset production threats with major emphasis
on Dacterial wilt.

Distiiufiiislictl workshop Participants.

Ladies and (■cntlcmcn,
I am fully confident that your deliberations today and tomorrow will be able to build
a re.tional and very clear objective analysis of all issues about enset and contribute to
this noble goal by realizing the contribution of enset plant to the food security of the
country.

Finally, I wish you all success in your deliberations and hope your discussions will
be enjoyable and fruitful. With these brief remarks, I declare this National Workshop
on Enset is officially open.

I thank you all.

 Technological Challenges of ?
E n s e t F a n n in g System : F o r E
P ro d u c tio n o f F o o d / F ib er and In d u s tr ia l O u tp u ts

Tayc Bezimeli
ASDAA Africa International, P.O.Box 342S2, Addis Abcba,
Tel. 0116 464371; E-mail: ta\/e56@gmail.com

In tro d u c tio n

The Enset Farming System (EFS, here after) is among the backbone of Ethiopian
agricultural economy. The crop accounts for the basic livelihood from 16 to 18 % of
Ethiopian Population As basic diet for millions of people at various agro-ecologies,
EFS has much influence on the development of rich and diversified cultures in the
south and south-western Ethiopia. The early evolution of EFS is the .least
documented. Since the last two decades, considerable attention has been given to EFS
both at national and international levels.

The enset agro-ecology ranges from l500 -1800m above sea level where the crop
growth cycle takes 4-5 years under relatively warm climate; to highland regions 1850-
3000m above sea level where the crop growth cycle is extended by 2 to 3 years
compared to relatively low elevation and warm climatic regions. Along these climatic
zones, the EFS integrates livestock husbandry with the cultivation of stimulant
crops, root and tuber and spice crops in the mid altitude; cereals, pulses, oil crops,
cooi season vegetables and temperate fruits in the high land regions (Taye
1966/73/96 and Westphal 1975).

Regions where enset is used as major staple food are among the most densely
populated in Ethiopia. In the Guraghea area, for example, 50 to 200 inhabitants per
square kilometer while over 300 inhabitants per square kilometer were inhabited in
Kembata, North Omo and Woliata areas. In pursuit for improving the livelihood of
millions of people at various agro-ecologies, the EFS has evolved environment
friendly agricultural practices based on the recycling of renewable resources and
economic complementarities among its sub-systems of production (i.e. crops,
livestock , fruits, vegetables, stimulant crops, and spices)

Hence, the technological challenges of sustainable enset farming system with more
emphasis on enhancing its food and fiber production and industrial outputs are
* summarized in seven different sections.

Section one describes the enset farming system as a bio-dvnamic sustainable

-* agricultural practice supporting the livelihood of about 20% of Ethiopian
population.. This section also emphasizes the need for scientific classification of
 2 Technological Challenge* o f Sustainable Enset Fanning Si/ston

various potential clones/varieties based on morphological, taxonomic and

horticultural traits to systematically register same in the National Registry.

The second section emphasizes the conventional research challenges resolved

through, agronomic, horticultural and bio-technology studies. This section also
identifies institutional and researchers challenges to respond to diversified issues and
problems of enset research and production svstem.

Section three analyzes enset research progress (over the last 25 years) based on an
extensive agronomic research evaluation on more than 600 clones collected from
ma or growing areas of the crop. Among the main research outputs included are
groupings (based on growth cycle maturity) of enset clones into early, intermediate
anc late sets enset clones, coupled with the identification of superior high yielding
enset clones for food and fiber uses. The section also reviews the need to advance
research on bio-technologv and screening cultivars resistance to bacterial wilt and
other diseases of the crop.

The fourth section notes the new era of enset research growth and expansion along
the concurrent evolution of Southern Agricultural Research Institute (SARI) and
establishment of Areka Agricultural Research Center (AARC) as well as the
emergence of new generation of researchers engaged on the improvement of the crop

The cultivation of enset based on salient analysis and perspectives is discussed in

section five, which deals with enset food production, nutritional values and the need
to employ available technology to expand manufacturing of various fiber products
both for export and domestic markets. The successful fiber manufacturing industrial
growth with abaca fiber (Musa texistalis) in Philippines is compared with enset fiber
in Ethiopia. Even though the tensile strength and quality of abaca and enset fiber are
similar, the lack of innovations by both the public and private sectors and the
traditional dependence on foreign investors to launch on such ventures, remain as
major setbacks in the transformation of agricultural technologies into agro­
industries.

Equally important is section six, covers the analysis of the current status and the need
to improve implements/devices for enset harvesting and processing. As the way
forward, the last section, section seven, brings to the forefront key technological
options for modernizing the EFS in order to expand economic role of enset.
 Tm /c Bezunelt

Botanical Characteristics, Economic and Agronomic

Classifications of Enset
Cheesman (1947) and Moore (1957) did establish basic botanical differences between
the two genera i.e. Musa and Ensete of the Musaceae family. Detail description of
Ensete spp. has been reported by Baker and Simmonds (1953). Morphological and
cytological differences (chromosome number, pseudo-stem size and color
characteristics, seed, embryo morphology and propagation method etc.) led to the
transfer of about 20 species from the genus Musa to Ensete. According to Vavilove
1951 and Simmonds 1958, Ethiopia is one of the primary center of origin and
diversity not only for a number of cereals (barely, durum wheat, teff,), some pulse
and oil crops, stimulants (Coffee arabica, Calha edulis etc) including enset (Ensete
ventricosum). The wild relatives of the latter species are known to occur at lower
altitudes in tropical African countries, including Kenya and Uganda south to
^ Mozambique. The previous collections of enset clones and morphological
classification and agronomic evaluation have evidenced the existence of several
clones throughout Ethiopia (Annual Reports, Debre Zeit and Holleta Agricultural
Research Centeres from 1969 to 1975).

The establishment of Areka Agricultural Research Center (AARC in 1994) gave new
impetus and thrust on enset research. The AARC collection of more than 600 clones
of enset from major enset growing regions, which is recorded in Annex 2, led to
undertaking agronomic and disease resistance research evaluations to identify or
develop high yielding food and fiber cultivars.

Classification o f enset eultivars/clones/or varieties

A national crop releasing procedure has been established whereby criteria foe the
var ous categories of crops have been stipulated as guidelines register elite
varieties/clones of crops in the National Crop Register. Criterion, which are
pertinent to evaluating enset clones is proposed to be included in existing guideline,
which in due course can be refined to register and release new enset clones/
varieties. This may include:
e Phenotypic similarity, color, and vegetative morphology (including plant
and pseudo stem heights and circumstances);
• Leaf sheath and leaf characteristics , color, size, shape, umber and weight;
® Corm circumference, grated weight variation, detection of true stem, or
emergence of voluntary suckers if any similar to Musa;
« Flower, fruits and seed morphology and variation
® Agronomic evaluation and grouping down to five best clones to produce
kocho, bulla and fiber.
• Agro-ecological adaptation and dominance of clones;
 4 Technological Challenges o f Sustainable Enset Fanning Si/stem

• Maturity range and cycles ( already in progress at ARC where clones

grouped as early, intermediate, and late sets based on period of years of
maturity and yield of kocho, bulla and fib e r;
• Cytological investigation to detect variability of somatic cells or
chromosomes; and
• Minimizing vernacular names and specific agronomic criteria;
• Types of plant (ornamental ) and suitable use including for medicinal
purposes; and
• Horticultural or domestication traits, yield of food and fiber.

Coiiiuncnts oil registration, naming and release o f clones/cult ivars

Regarding registration and release of cultivars/or clones among the key problem
include repeated vernacular names given to same or similar clone by different
ethnic traditional languages. Therefore, there is need to narrow down the apparent
large number of clones of enset based on morphological, taxonomic, horticultural
trai:s, growth cycle and yield performance (food and fiber). Classification of enset
based on these parameters could enable to identify clones with genetic diversity,
and also save considerable time and resources in the entire research effort for
identifying or developing high yielding food and fiber clones/ cultivars collected
from diverse sources and regions.

Based on p h en o ty p ic characteristics such as color of leaf and p se u d o -ste m as w ell as

ecological adaptation and growth performance of similar clones, but known bv
different vernacular names were identified in Wolaita, Kambata and Sidamo area
(Endale et.al., 1996).

Research and Technology Development Challenges

Conventional research challenges

Among the persistent research challenges of EFS include:

• The generation and application of more productive technologies to substantially

improve the yield of food and fiber, and also the productivity of the system in
sustainable manner;
• Copping with the food security needs for the apparent population increase;
• Enhance processing/industrial utilization to improve the output of enset value-
added products;
• Sustaining the conservation of the soil to minimize land degradation based on
environmentally sound farming practices;
• Integration of EFS production sub-systems (such as cereals, vegetables , fruits,
stimulant crops, spices and livestock production) and to enhance the recycling
of organic renewable resources for sustaining and conservation of the fertility
of the s o il;
• Improve the productivity of EFS to increase food and fiber production, and
also to create economic synergy among its sub-systems of production;
 5 Tai/c Bezuiich

• Avoid fragmentation of enset traditional farms due to increase of farm families,

by introducing favorable land policy that could increase size of enset
agriculture to comply with population growth;
• Strengthen the regional and national campaigns to effectively eradicate bacterial
wilt and other diseases that threaten EFS.

Challenges tor collaborating researchers and scientific institutions

Among the key challenges for collaborating researchers and development
institutions/organizations for improving production and productivity of enset crop
in lude:

Systematic/ scientific classification of enset clones based on botanical

characteristics, agronomic and horticultural traits, instead of vernacular names ;
Based on agronomic and horticultural traits and yield performance to register, name
and release cultivars/or varieties along the guideline provided by Scientific and
Technical Committee of Enset ;
To minimize vernacular names to released cultivars unless cleared through above
mentioned scientific committee;
Retain few high yielding cultivars for food and fiber for each major enset growing
region in order to expand clones/varietal trials;
Strengthen mechanical engineering research for radical improvement of enset
processing implements/devices/ machines as well as to develop marketable
fermentation vats/containers; and
Improve the fermentation process by also advancing microbiological research in
collaboration with universities and other stakeholders.

T h e \eed for E nset Research Expansion and Growth

During the last two decades (1990 —to present), there has been considerable
expansion and growth on enset research partially due to:
• Concurrent institutional development of Southern Region of Agricultural Research
Institute (SRARI) and the Areka Agricultural Research Center (AARC) ;
• New generation of scientists/researchers engaged on various aspects of enset
research and development;
« The establishm ent of botanical collection of several clones of enset from major
enset growing regions at AARC (Annex 2) led to an extensive agronomic
evaluation with major emphasis on varietal development to improve the yield of
food and fiber and to screen resistant cultivars to bacterial wilt diseases as well as to
control other diseases and pests.

Agronomic/horticultural research potential

Similar to other root and tuber crops, enset is largely cultivated from 1500-2850
meters altitude where 700-1200mm rainfall and 16-26 degrees centigrade
temperature prevail (Table 1). Systematic farm level survey is required to
quantify the area of cultivation occupied by the crop. The establishment of Areka
Agricultural Research Center (AARC) gave new impetus and thrust on enset
 6__________Technological Challenges o f Sustainable Enset Farming System

research. The AARC botanical collection of large number of clones/or land

races from major enset growing regions advanced agronomic and disease
resistance research evaluations in order to identify or develop high yielding food
and fiber clones/cultivars.

Atnafua and Endale, 2008, reported the following encouraging research outputs
(particularly at AARC). These include:

• Grouping of the enset clones/land races (based on growth cycle, yield of kocho and
fiber):
, E arly Set: reported superior clones with the highest kocho yield ranging from
310 to 600 q h a^ y r1 and bulla yield from 30 to 6 0 q h a 1y r ‘
o T h e in term ed iate Set: identified 6 clones that gave kocho yield from 350 to
510 q h a -'y r1; also reported 5 clones that gave bulla yield from 31 to 45 q h a 'y r
i-/
o Late Set: It is likely this group of enset clones mature 2 to 3 years later than
earliest set. Reported kocho yielcf varied from 200 to 350 qha 'y r 1and reported
bulla yield varied from 15-23 qha ‘yr *.

• Selecting high yielding of kocho clones (14); bulla clones (6) and relatively fiber
clones (7). This research finding is equally important to expand the cultivation of
superior enset clones identified for food and fiber production.
• The above sources of research findings also indicated the best fiber and bulln
yielder have been obtained from earl) set of enset clones;
• The above study also showed that clones that gave high yield of kocho were
also observed as good yielders of fiber.

The research progress made to group enset clones based on maturity, yield of kocho,
fiber and bulln could save considerable time and resources in the agronomic
evaluation to a great extent, and to identify genetic variation among clones to
some extent.

The likely threats to enset agriculture

The incidence of bacterial disease has threatened enset agriculture to a great extent
(Yirgou et al., 1968 and Ashagari, 1980). Fungus diseases such as Fusnrium oxysporuni
which threatened enset agriculture in the 1970's has been the major occupation of
research. Fortunately, farm level awareness of the disease lessened the threat of this
disease on the cultivation of the crop. The AARC current research activities of pests
have focused on key diseases (bacterial wilt, virus and fungus) and other pests
(insects and nematodes). This research effort should be further strengthened by
availing high level and appropriate lab equipment.

To develop some enset clones resistant to bacterial wilt and other diseases, there is
a need to broaden the genetic base of the genus by establishing world collection of
about 20 known spp (Cheesman 1947) Annex 1.
 7 Tn\/c Bcztuich

It is apparent that the research work at AARC has made progress to screen enset
cultivars resistant to bacterial wilt and other diseases based on the evaluation of
germplasm collected from major growing areas of the crop as well as using botanical
seedlings. In due course, such research thrust could avail to farmers resistant
cultivars to bacterial wilt and other diseases.

The second threat to enset agriculture has been in densely populated areas (for
example Wolaita, Garno, Kembata about 300 people/square km), that contributed to
the increase of members of farm families, which further contributed to the
fragmentation of traditional farms.

An alternative short-pathway to control bacterial wilt disease is to implement a

Disease Eradication Campaign recently revitalized and endorsed during the
workshop (which involves farmers, extension workers, researchers, and district
ad ministrators).
♦
Biotechnology research
Selection and agronomic evaluation of elite clones for food and fiber yield has
progressed at Areka Agricultural Research Center (AARC). The research work has
become complex process due to the occurrence of genome size variation similar to
enset's relative genus Musn diploids in which DNA markers can be employed to
identify clones resistant to bacterial wilt. The molecular characterization research
work started by Almaz (2008) is good beginning to streamline the genetic
relationship and variation among Ensete veiitricosum clones in the identification of
genes for resistance to bacterial wilt. Equally important research undertaking on the
genus not only the collection and establishment of about 20 spp. of Ensete from
Africa and Asian countries in addition to ventricosiun clones; but also to collect and
establish selected botanical seedlings in order to broaden the genetic base essential
for screening and transferring genes identified for resistance to bacterial wilt and
other diseases. Inter-genus and spp. gene transfer employing micro-organisms could
also be pursued.

it must be noted that tissue culture research work by itself should not be taken as the
only success in bio-technology research, but as modality of propagation of high
yielding disease resistant cultivars being pursued to increase large number of enset
plants both at Areka and Holleta research centers. This research work and the
employment of DNA markers should be supported (fund and facilities) to strengthen
Motechnology research on the genus Ensete to enhance the identification of genes
. resistant to bacterial wilt and other diseases, but also to establish the genetic
variability among high food and fiber yielding clones. Although
agronomic/horticultural research has identified superior enset clones for food and
* fiber production, their genetic variability and effects will also need to be studied.
 8__________ Technological Challenges o f Sustainable Enset Fanning Si/stem

Table 1. Climatic adaptation of major root, tuber and corm crops in Ethiopia

Temperature
Annual rainfall Range
Scientific Name Common name Altitude ( M) ( mm) (°C)
Ensitfe ventricosum Enset 1500-2850 700-1200 16-26
Solarium tuberosum Irish Potato 1450-2400 600-1200 16-24
Ipomoea batatas Sweet Potato 1400-2000 550-800 18-27
Colccasia esculenta Taro 1400-1900 600-850 20-27
Diosoorea abyssinica
(and D. ulbifera) Yam 1300-1900 500-800 22-27
Cocrinia abyssinica ‘Anchote’ 1500-2000 700-1200 20-25
Source: Westphal 1975 and Taye Bezuneh 1996

Bio-dynamics of* Enset Cultivation

In pursuit of improving the livelihood of millions of people along various agro­
ecologies, the EFS more titan any other crop, has evolved environment friendly
agricultural practices based on the recycling of renewable resources and economic
complementary among its sub-systems of production (i.e. cereals, livestock, fruits,
vegetables, stimulant crops, and spices) as discussed below. The EFS is among the
backbone of Ethiopian agricultural economy taking into account the following salient
analysis.
• Food security and livelihood perspectives:
o The livelihoods for 16-18% of the Ethiopian population in food security and
income generation depend on EFS;

o Regions where enset is used as major or co-staple food are among the most
densely populated in Ethiopia, as already mentioned in the introduction part

o It has been established that only 13 to 25 enset plants could provide the yearly
supply of food per person;

• From socio-cultural diversification perspectives: as basic or co-staple diet for

millions of people of different ethnic groups, the EFS has influenced the
development of rich and diversified cultures.
• The EFS has integrated the cultivation of cereals, root, tuber, stimulant crops, spices
and even livestock production to sustain both livelihoods and income generation.
• The EFS is economically more viable and durable farming practice and it has
already integrated conservation of soil fertility through the use of manure,
cereal/legum e rotation, and crop residue input into the soil and
• The EFS bio-dynamism has incorporated diversified cropping systems along various
agro-ecological zones (Westphal 1975 and Stanley 1965), which enabled 6-12
harvests for cereals; 4 to 8 harvest for cotton; 1 to 3 harvests for fruits; 8 to 12
harvests for vegetables; 3 to 5 harvests, for papaya; 4 to 5 harvests for established*
coffee or one harvest for newly established coffee; orchards before the completion
of the growth cycle of the enset plant (i.e. h arvesting), and
• Bl-Yearly production of large animals such as dairy and beef and yearly production'*
of small animas such as poultry, sheep etc.
 9 Tni/c Bczuncli

Food Production stud \utritionaI Value

Food production potential

Over the last three decades, the total area covered with the enset crop has grown
from 65,000 ha (Stanleyl965) to about 300,000 ha (CSO 2009/2010) The Southern
and Oromia Regions produce close to 80% of the crop. The kocho ai d bulln yield of
enset clones seem to vary considerably. Table 2 compares the yiek of four clones
previously studied that gave yield at the level of 190 to 3000 q h a-'y r1 kocho and less
than 30 qha-'yr' bulln (Taye Bezuneh 1966/1973/84 ). Since the establishment of
AARC several superior food and fiber yielding cultivars (such as Digomerza ,
Godera, Henuvva and Zerbo) that gave yield over 430 q h a-'y r1 kocho and at least 40
qh a-'yr1 bulln were identified (Atnafua and Endale,.2008). Systematic planting or
intercropping of these new cultivars at their respective agro-ecologies, and concurrent
improvement of enset harvesting and processing implements /devices could easily
triple the volume of enset production.

Enset propagation practices and its planting svstem (involving traditional stages of
nurseries) have been reviewed extensively (Mulugetta el al. 1996/2008). Series of
enset propagation studies including tissue culture have established the pathway for
rapid and large scale multiplication of enset planting materials (Mulugetta et al.
2002/2003) that should be explored.

Table 2. Kocho. bulla and fiber production of some enset clones

Vernacular Kocho Bulla Fiber

Names (q /h a 'y r1) (q/ha-’y r 1) ( q /h a 'y r')

Ferezae + 247 33 389

Ado+ 192 29 460
Midasho + 304 27 559
Tuzuma + 386 26 380
Henuwa 440 43 317
Godera ++ 520 NA 424
Zerbo ++ 420 47 362
Djgomerza++ 550 57 406
Gena +■+ 410 NA 554
Sources: Taye Bezuneh 1973/84 for + and Atnafua Bekele et al. 2008 for ++
NA = Not available

\ u trits o n a l value
Table 3 summarizes the nutrient value of enset (ENI, 1981). The result of the study
shows that the carbohydrate content that of bulln is much higher and its protein and
calcium much lower than kocho. Previous studies indicated that the carbohydrate
content of the clones evaluated, such as Ferezae, Ado and Tuzuma therefore, varied
- from 178 qt/ha to 293qt/ha/ (Taye, 1984). Accordingly, the energy yield of enset is
estimated from 140,000 to 220,000 kilo joule /ha (ENI 1981). However, the processing
of kocho to bulln reduces the calcium content to 65 mg/100 gm and concurrently
 10 Technological Challenges o f Sustainable Enset Fanning System

raises the carbohydrate content of bulla to 54.4% per 100 gms from 41.3% (Table 3).
The traditional belief that using special enset cultivars for food preparation can help
to heal fractured bones seem to have some validity, since the calcium content of
enset food is comparatively higher than most cereals.

The 1990's extensive survey on vitamin A deficiency (which causes night blindness)
that involved on more than 6600 children covered four ecological zones including
enset cultivated area. The surprising finding of the survey has been that the
overall diet of the enset zone has the highest concentration of beta- carotene or
vitctmin A source that prevents night blindness. This national survey findings led to
recommendation for vitamin A deficiency control program throughout the country,
except or outside the enset zone (EN1 Survey referred in Lost Crops of Africa:
Volume LI, pp 179-1802). This has been attributed to vegetable gardens usually
affiliated with EFS and its leaf-sheath fresh leaves cooked for consumption.

Table 3. Nutrient composition of two ensete flour products (percent per 100 g dry weight)

Nutrient Kocho Bulla

Water 56.3 45.25
Carbohydrate 41.3 54.40
Protein 1.2(3.35) 0.25
Fat 0.2 0.1
Iron 5.3 mg/100 g 7.7 mg/100 g
Calcium 195 mg/100 g 65 mg/100 g
Thiamin Mg/100 g 0.03
Riboflavin Mg/100 g 0.04
Niacin Mg/100 g 0.025
Source: Ehiopian Nutrition Institute 1981

Enset food contains from 1.72 % to 6.2% protein, with mean value of 3.35% after two
mo riths of fermentation (Table 4). In general, enset food of fermented pulp has
relatively higher lysine than most cereals. However, methionine and iso-lucine are
the first and second most deficient essential amino acids, respectively (EFS, 1981).
The quality of protein of kocho improves with length of fermentation. The effect of
length of fermentation on protein content of kocho food is summarized below.

• Protein content changed from 3.60*/- before fermentation to 3.35% after

fermentation;
• The length of fermentation improves the quality of protein as determined bv
amino acid profiles;
• The food or kocho protein becomes generally higher in lysine than most
cereals; and
• The food or its diet contain high calcium content that accrued in traditional
treatment/ claims of healing fractured bones (Fig. 1)
 11 Taye Bezuneh

Enscte ventricosum clones and Finger Millet (also known as Dagussa vernacular name
in Amharic) are important sources of calcium compared to other staple cereals such
as teff, wheat and maize (Fig. 1, synthesized data, 1981).

Table 4. The effect of length of fermentation on protein content of kocho

Protein (%)
Fermentation period Minimum Maximum Average
Fresh 1.72 6.00 3.68
1 month 1.69 4.69 3.35
4.5 months 1.88 4.44 3.24
5.5 months 1.91 4.25 3.10
Source: Abrham Besrat et at. 1979

/Ca mg 10CH g

450

Teff Maize Wheat Enset Dagussa

Fig.1. The calcium content of some food crops in Ethiopia

Source: Taye Bezuneh ( Synthesized data 2010)
 12 Technological Challenges o f Sustainable Enset Farming System

Agro -processing/ value-added products

o Fiber production for industrial use

Data on enset fiber production in general and its yield per unit of land is virtually
lacking. The fiber yield of some enset clones varies from 3.8 to 5.9 quintals/ha (Table
2). Over the last two decades, enset research has advanced in the identification of
high food and fiber yielding enset cultivars. In case of fiber yield, since 1973 Ado, and
Midasho gave yield of 4.60 and 5.59 qha^yr.1 Likewise, Gena and Godera also gave
fiber yield from 4 to 5.5 qh^ay-1 (Atnafua et al. 2008).

Enset and abaca fibers have similar strength and quality standard as summarized in
Annex 3. The Philippines in particular used the abaca (Musa texitnlis) fiber to expand
its manufactured industrial product that also attracted export markets to Europe,
North America and other countries.

Despite the research findings (four decades ago) the enset fiber is of high quality and
similar standard to that of abaca fiber. However, very limited industrial
manufacturing did take place in Ethiopia due to lack of innovations by both the
public and private sectors.

o The rationale and challenge

Enset agriculture being the livelihood of 16 to 18% of Ethiopian population, its
cultivation is well established for several decades at various agro-ecologies. As major
and co-staple food, the crop also provides fiber/starch/paste which can be
transformed for manufacturing several products. It is estimated that about
60,000,000 kg fibers is imported yearly into Ethiopia for industrial uses. The local
supply of enset fiber is estimated about 15,000,000 kg. The lessons/experiences of
Ethiopia's industrial growth show too much dependence on foreign investors which
has been slow and least productive. This scenario can be improved by:

• Creating enabling environment (land access and fund, loans) and organizing
cooperatives comprised of young generation including college graduates in the
major areas of enset cultivation;
• Supporting industrial level of enset production, for example, cooperatives in major
enset growing areas may have members above 500 with at least 3 ha of land
available to each member;
• Expanding the production of newly identified high fiber yielding enset clones by
respective cooperatives and large scale private sectors producers;
• Attracting private investors to establish the fiber and starch/paste industries;
• Organizing the cooperatives as out-growers to supply enset fiber raw material to the *
manufacturing industries; and
• Establish Eco-Tourism micro-industries at the most sub-urban centers and villages.
.
Industrial growth of this nature could be the basis for establishing the Eco-Tourism
oriented fiber and food processing micro-industries at most attractive sites
 13 Taye Bczuncli

throughout enset growing regions (i.e. Woliso, Wolekitie, Endibir, Awassa, Yirgalem,
Dilla, Welaita Sodo, parts of Oromia Region etc). These Eco-Tourism oriented micro­
industries can manufacture enset fiber based products, and also generate income
from sales of products, as well as from visits of sites by tourists.
Based on the above mentioned enabling policy, to organize cooperatives and also
support the private sector involvement, Ethiopia in due course must export
manufactured products including those listed below similar to what the Philippines
did to all continents over the last five decades.

o M anufacturing products
Banana and enset fibers have high tensile strength, light weight, and good moisture
absorption. These fibers have insulating, thermal conductivity, biodegradable
characteristics. Similar to abaca, enset fiber can be used for manufacturing the
following products: Bags of all sorts including ladies bags; paper making; starches,
paste for textile and other industries; car seats, cushion and pillows; coats and jackets
of various quality and sizes; shirts, dresses, trousers and hats; mats of various sorts
and sizes; home furniture and folding dividers, dinning sets, hanging shades; chairs,
beds, tables; twines, cordages and lamp shades; carpets of various sorts; hand
woven rugs; making garments; curtains of different colors; jewelry of various
designs; suit cases of various sorts; roof insulations and tiles if further processed and
integrated with tiles; shower & bath unit through further processing and integration
with other fibers; making handicrafts and home decorative; wide variety of
applications in making various products like paper bags, filter paper, greeting cards,
decorative papers, pen stands and many more. These products have potential
markets and hence concerted efforts should focus in the development of elite clones
for the development high quality fiber.

o The developm ent of appropriate harvesting and processing technologies

Food and fiber extraction from the crop has always been poor both qualitatively and
quantitatively. Research should also focus to substantially improve harvesting and
processing machines/or equipment to increase outputs of both food and fiber as well
as to concurrently relief women from such tedious and cumbersome labor. Annex 4
summarizes some of the progresses made and the need for further improvement
required on enset harvesting and processing implements/ devices. However, the
decorticating of enset pseudo-stem, pulp and the pulverization of corm require the
following changes that:
® The nature of enset harvesting, decortications of the pseudo-stem and leaf-sheath
require more power (energy or machine) and change of size of the implement;
® The design should focus on mechanization that could lead to industrial use of enset
food and fiber products at community level or by investors who could afford to buy
improved devices / or equipment use with motors;
® Strengthening mechanical engineering and food technology researches be considered
crucial;
 14__________Teclniological Challenges o f Sustainable Enset Farming System

Thus, the improvement of harvesting and processing implements/devices require

the concurrent research in mechanical engineering and food science technologies,
concerted effort should be exerted to further improve the decorticators that separates
pulp from leaf-sheath; pulverizer that grates the corm into fine pieces; shredder that
chips the fiber in the fermented kocho; and kneader to squeeze out water from
fermented kocho.

o Standardization of the ferm entation process

The fermentation process of enset food can be standardized to substantially increase
the recovery of fresh weight by at least 50% to fermented and dried products. Such
an increase of recovery will improve the volume of the production of kocho and bulln.
Atriafina et.al. (2008) did undertake extensive trials to standardize quality
fermentation using different types of vats including plastic material. Kocho stored in
the traditional way practiced by farmers apparently gave the highest quality both in
texture and color compared to all other storage after 10 months of fermentation
per .od. The study also showed kocho stored in plastic materials also sustained better
quality. However, it is advisable not to use plastic material based on recent findings
that warrant leakages of chemicals from plastic is hazardous to health.

Microbiological and biochemical studies of the fermentation process such as time-

interval studies of the chemical change could identify the kind of micro flora or
organisms associated with fermentation of enset foods, and also standardize the
process. Food science and technology research is necessary to make enset processing
hygienic and systematic. There is also role for microbiologists to identify the micro
organism by creating pure cultures to produce quality and standardized kocho.

Agricultural and mechanical engineers and food technologists will need to be

provided with financial support and equipment/implements that will substantially
improve the fermentation containments and process. Microbiologists also require
modern facilities, and financial support to identify useful microorganisms to
improve quality of the fermented product and to standardize the process.

o Outputs of enset food processing

These include: Flour for preparing bulln; kocho for preparing bread and different
foods, porridge and dumplings; amitcho (starch cooked like potato); corm grated or
cut into pieces for preparing various foods; starch/paste useful for cloth and book
binding, textile, leather, food, garment finishing, and printing industries.
 15 Tai/c Bezuneh

Recommendation
The way forward for modernizing enset production and expanding its economy

® M odernizing enset production

Although the EFS has sustained food security, there is a need to expand its economy
by m odernizing its planting system and the integrated development of the fiber and
food processing industries. Modernizing EFS could be done through:
o Creating more jobs to young generation of farmers including college graduates;
o Minimizing fragmentation of traditional FFS due to increased farm family size
o Expanding the EFS economy

• Im proving the available technologies for expanding the Enset Economy

include:
a. Promote high yielding food and fiber clones

b. Develop industries that will utilize enset fiber in large scale for both export
and domestic markets;

c. Sustain the highly developed Farming System to support the livelihood of

16 to 18% of Ethiopian population; and

d. Strengthen the EFS, which has evolved environment friendly agricultural

practices based on the recycling of renewable resources and economic
complementarities among its sub-systems of production (i.e. crops,
livestock , fruits, vegetables, stimulant crops, and spices )

• Some Pertinent Hints for the Developm ent of Econom ically V iable Enset
Farm ing System:
Certain important components to modernize EFS planning at major enset
growing regions are to:
a. Put in place favorable land use practice to sustain productive EFS.

b. Organize member of the new generation of farmers cooperatives with

members who systematically inter-crop enset and coffee for
improving production and productivity in order to triple food and
fiber production and increase income by several folds from the
interplant cash crop; and

c. Systematically intercrop the enset plant (600 plants/ha) for food/

fiber; and coffee as cash crop (400 plants/ha).

Collectively/ these cooperatives could serve as important source of fiber for the
industry as well as to expand Eco-Tourism oriented micro-industries. At mid altitude
 16 Technological Challenge* o f Sustainable Enset Funning System

1500 to 2000m above sea level optimum yield for both Enset and Coffee is likely to
be obtained.

• M odernizing enset food and fiber harvesting and processing

To rurther upgrade slightly improved enset harvesting and processing implements,
survey at farm and research levels is necessary particularly to:
a. Assess the technological fitness and acceptance of improved and previously
distributed devices;
b. Further improve food and fiber processing implements or devices based
on previous experiences / lessons, and
c. Involve the private sector and youth engineers to manufacture and market
improved enset extraction devices.

• Eco-Tourism oriented fiber and food processing m icro-industries

In many respects, the growth of manufacturing industry in Ethiopia has been slow
compared to several other countries. The typical case has been the expansion of
abaca manufacturing industry over 50 years ago in the Philippines compared to that
of enset fiber industry in Ethiopia during the same period. The high standard
quality of enset fiber (similar to abaca) has been determined since the last four
decades (Taye 1963/73), its industrial potential has yet to be fully explored in
Ethiopia (Annex 3).

The modern cooperatives mentioned above if expanded throughout the enset

growing regions could serve as an important source of raw material for fiber and
food processing industries. Through industrial innovation, the enset fiber could also
blend with fibers of other crops such as flax, cotton, ramie, bamboo fiber, to
manufacture clothes such as dresses, jackets, coats, shirts, hats, light shoes, car seats,
cushion, pillows etc.

Modernization of enset fermentation process could also enhance the processing of

kocho and bulln to be further processed into several food byproducts.
 17 Taye Bezuneh

Annex 1.

Major spp. of the genus Ensete

• Ensete gilletttie is largely adapted in drier areas in West and Central Africa
from Sierraleon, Nigeria, Camerroon, to Angola ;
• Ensete perrieri is largely distributed on the island of Madagascar;
» Most species of Ensete are widely distributed across Africa except forEnsete
glaucum and E. superbum which are of Asian origin;
® Ensete calosperm um is widely distributed in the Pacific, Fijii and in New
Guinea;
• E. facundum and bagshawei largely distributed in Eastern Africa
including in Uganda;
• Ensete am oldianum largely distributed in tropical Africa ;
• Ensete schw einfuthii distributed in Sudan ;
• Ensete ruandense adapted in most warm clim ates originally observed in
Rwenzori;
• Ensete rubronenvalum adapted in Rwanda, Congo and Rwenzori.
• Ensete w ilsoni d istrib u ted in southern China;
• Ensete buchanani mostly distributed in the highlands of Southern Africa;
• Ensete davyae largely distributed in M ozam bique;

Source: Chesman EE 1947 Classification of the bananas , I The genus Ensete Horan
Kew Bulletin No 2

Annex 2
The Extent of Number of "Enset or Isset" Cloneserred in major growing
regions of the crop.

Area Number of clones referred

Sidama 97
Gedio 42
Wolaita 77
GamoGofa 196
Kembata 142
Guraghe 70
Borena 28
Source : Atnafua Bekele et nl,. ,2008. Published in Root and Tuber Crops page 157
 18 Technological Challenges o f Sustainable Enset Farming $i/stcin

Annex 3
Comparative tensile strength and relative quality of some commercial fiber crops

Scientific Name Type of fiber Fiber strength Rem arks

(gms per
denier)
Musa textalis Abacca 7.0 Fiber of high quality and
standard .manufactured
extensively to produce ropes,
mats of various sorts, lady
bags, jackets, car seats etc
These products have attracted
export markets to North
America, Europe, Middle East
countries etc..
Eusate Enset 5.0 Fiber of high quality and
ventricosuni standard ,the last four decade
s in Ethiopia, nindustrial
manufacturing not expanded
due to limited innovations bv
both public and private sectors
Agave sisalana Sisal 4.4 Fiber strong and durable and
industria 11v manu factu red
SanrkrT'iere spp Sanseviera 4.5 Same as above
Again? spp Henequen 3.3 Same as above

Source: Taye Bezuneh (1972/73) Debre-Zeit Agricultural Research Center

 19 Taye Bezuneh

, Annex 4
Status of the improvement of the enset processing devices implements
Types of implement extraction part Remark
Improvement of the traditional implement
a). Scrapper or Pseudo-stem started in 1970.The Rural Technology
decorticator and leaf-sheath Promotion Center distributed 180 of the
scrappers to Enset growing areas (Guraghea,
Awassa, Sidama, Kambata, North Omo, and
Areka). The improved decorticator or
scrapper however required more force and it~
is also time consuming with the improved
scrapper or decorticator 48kg/hr of the enset
pseudo-leaf-sheath has been extracted
compared to 33Kg/hr with traditional
implement.

b). Kneader or Water Thirty-two of water squeezing devices which almost

Pressing or Squeezing removed un wanted water from enset food were
Device distributed to Guragea zone, Kambala and North Omo.
Improved implement relatively hygienic, can be operated
by men also and can squeeze up to 110 Kg/hr compared
to 70 Kg/hr with traditional implements
c) Pulverizer Some progress has been made in the extraction of final
pulverized material of the corm. However, the implement
has been extremely time consuming
d) Shredder Cutting fibers of the fermented enset dough of tolerable
size. Apparently no device has been developed to
properly carry out the practice and to relief women from
this tedious work
Source: D eribe K ifle 1996
 20 Technological Challenges o f Sustainable Euset Fanning Si/stem

References
Abrliam Besrat, Haile Mehansho, and Taye Bezuneh . 1979. Effect of varietal difference and
length of fermentation on protein quantity and quality of enset. Nutrition Report
International 20(2):245-50.
Abate, T., Hiebsch, C., Brandt, S., and Gebremariam, S.,( Eds). 1996. Enset-Based Sustainable
Agriculture in Ethiopia. Addis Ababa: Institute of Agricultural Research.
Almaz N.2001. Molecular Characterization of enset clones published in Root and Tuber
Crops 2008, pp 160-161 (Ed.) Gebremedhin Woldegiorgis et nl)
Ashagari D. 1985. Studies on bacterial wilt of enset ( Ensete ventricosum) and prospectsfor its
control. Ethiop. /. Agric.Sci.7(1) 1-14.
Atnafua Bekele, Ermayas Tesfaye, Endale Taboge, Mikias Yeshitila, Mulugeta Diro and
Yohannes Tilahun. 2008. Enset Variety Development published in Rooyt and Tuber Crops
(Ed) Gebremedhin weldegiorgis et al. ) pp157-193
Baker, R.E.D, Simmonds, N.W. 1953. The genus Enset in Africa. Kew Bulletin 3: 405-416.
Chesman EE. 1947. Classification of the bananas , I The genus Ensete Horan Kew Bulletin No
2.
Deribe Kifle. 1996. Development and Promotion of P.NSET Processing Equipment pp 321 -323
in Enset Based Sustainable Agriculture in Ethiopia edited by Tsedake Abate et al ENI
(Ethiopian Nutrition Institute), 1981. Expanded Food Composition Table for Use in
Ethiopia. Addis Ababa: ENI.
F AO/W HO (Food and Agriculture Organization/World Health Organization), 1973. Hnerg\
and Protein Requirements. Report of Joint Expert Committee. Rome: FAO.
Fekadu, D., 1996. Potential of Enset (Ensete Ventricosum) in Ruminant Nutrition in Ethiopia,
Part III. M .Sc. Thesis, Sw edish U niversity of A gricultural Sciences.
Makiso, T., 1980. Agronomic report for 1979. Pub. No. 63. Wolaita Agricultural Development
Unit.
Mul ageta Diro and Endale Tabogie. 1994. Enset (Enset ventricosum): Surveying Base Cropping
Systems in some Enset Growing Regions of 1 thiopia. Paper presented at the Second
Ethiopia Horticultural Workshop, 1 to 3 Dec., Addis Ababa, Ethiopia
Mulugeta Diro, Mikias Yeshitila , Ermias Tesfaye, and Endale Tabogie. 2008. Enset Agrononu
published in Root and Tuber Crops pp 195-208 (Ed) Geberemedhin
W oldegiorgis et al.)
Moore, E.H. 1957. Musa and Ensete: The Cultivated Bananas Baileya, 175 ,pp189-193
Rah:nato, D., 1996. Resilience and vulnerability: enset agriculture in southern Ethiopia. In
Abate et al,(d).
Shack, W., 1966. The Gurage: A People of the Enset Culture. Oxford: Oxford University Press.
Shigeta, M., 1991. The Ethnobotanical Study of Ensete (Ensete ventricosum) in Southwestern
Ethiopia. Ph. D. dissertation, Center for African Area Studies, Kyoto University.
--------------- , 1990. Folk in situ conservation of enset (Ensete ventricosum [Welw.] E.E.
Cheesman): Towards the interpretation of indigenous agricultural science of the Ari,
southwestern Ethiopia. African Studies Monographs, Kyoto, 10:93-107.
Simmonds, N.W. 1958. Ensete cultivation in the Southern highlands of Ethiopia: A review.
Tropical Agriculture 35:302307

_______________ , 1960. Notes on banana taxonomy, IV, Recent records on Ensete in

Asia, Kew Bu 11.14,198-212
 21 Ttu/c Bezuneh

Smeds, H, 1955 Ensete Planting cultivars of Estern Sidamo, Ethiopia, Acta Geographica,
Helansiki, 13, pp 40
Stanley,S. 1966. Ensete in Ethiopian Economy. Ethiopian Geographical J. IV, 20-32
Taye Bezuneh, 1996. An overview on enset research and the future technological needs for
enhancing its production and utilization. In Abate et al, (Eds).
--------------- , ---------------- , 1984. Evaluation of some enset ventricosum Clones for yield with
emphasis on the effect of length of fermentation on carbohydrate and calcium content.
Tropical Agriculture. 61 (2) 111-116.
___________________ , 1971, The role of Musaceae in Ethiopian Agriculture . The genus
Ensete . Technical Communication , International SOC. Hor.Sci.21:97-101
Taye Bezuneh, Asrat Feleke, and Regassa Bauie, 1967. The Cultivation of the Genus Ensete
in Ethiopia. Soil and Crop Science Society of Florida Proceedings 27:133-41.
Terefe B., and Tabogie, E. 1989. A review of the available research recommendations and
future strategies on enset. In Proceedings of 19th National Crop Improvements
Conference, April 1987. Addis Ababa: Institute of Agricultural Research.
Tesiaye, S., and Haile, B. 1995. Diagnostic Survey Report of Enset Based Farming Systems of
Maraka Woreda, North Omo. Awassa: Institute of Agricultural Research. Mimeo.
Vav ilov, I. 1926. Studies on the Origin of Cultivated Plants. Leningrad: Institute of Applied
Botanical Plant Breeding.
Wo detensaye, Asnaketch. 1997. The Ecology of Ensete ventricosum in Ethipoia. Doctoral
dissertation. Department of Soil Sciences, Swedish University of Agricultural Sciences.
Acta Universitatis Agriclturae Sueciae, Agraria 78.
Westphal, E. 1975. Agricultural Systems in Ethiopia. Wageningen: Centre for Agricultural
Publishing and Documentation.
 Past Research Achievement and Existing
Gaps on Enset {Erntete ven& r&casum
(welw.) Checsinan) Breeding
M ikias Y eshiiila1, and Zerilinii Ycmataiv1
Southern Agricultural Research Institute
Areka Agricultural Research Center, P.O.Box 79, Areka

A b s tra c t
Enset (Ensete ventricosum (Welw.) Chessm an) is a perennial, herbaceous anil
monocarpic crop belonging to the family M usaceae (Westphal 1975). Enset based farming
system is an indigenous and sustainable agricultural system in Ethiopia, and the crop is
highly related with sustainable agriculture and food security, and thus to the economic,
cultural and social life o f different ethnic groups. Despite, the main/ uses o f enset which
make the crop to be highly related to sustainable agriculture, there are a lot o f biotic and
abiotic factors still existing threatening its production and productivity. In the past
various research activities have been conducted to generate new production technologies
and to promote them to the end users. Many o f the researches on enset have been
concentrated on agronomic studies. Concerning issues in relation to disease management
options, only few research findings on pathogen survival, means o f dissemination,
pathogenicity test, sanitary control, and identification o f some tolerant clones and enset
root mealy bug control measures were conducted. On the genetic improvement o f enset
only sonic limited works were conducted so far: attempts were made to collect and
preserve all the possible enset germplasms in the country and currently, a total o f 623
enset germplasms from 12 major enset growing areas o f Ethiopia have been collected and
conserved ex situ by the Southern Agricultural Research Institute o f Areka Agricultural
Research Center. The other breeding research finding was the M orphological
characterization o f enset clones based on phenotypic, quantitative and qualitative traits
has been done at Areka Agricultural Research Centre which considered the majority o f
enset accessions from the six major enset growing areas o f Southern region o f Ethiopia.
The result o f the d iv ersity index showed that low levels o f diversity existed among the
279 Enset ventricosum accessions based on the frequency o f the phenotypic characters
considered. Attempts were also made to develop regression model which, non-
destructively, predicts yield o f enset with better precision and simplifying yield
evaluation in experiments and also solve difficulties in estimating kocho yield in the
assessment o f production balance in enset production region o f the country. The
experiment zoas carried out at Areka Agricultural Research Center's on-station site on a
total number o f 32S enset plants from the six major enset growing areas o f Southern
Ethiopia. Plant height and pseudosteui circumference were found to be the best non­
destructive enset kocho yield predictors with R-squarc (0.78) value for estimating
fermented squeezed kocho yield with the model equation o f FUNK= -25.76 + 5.35PH +
20.01 PSC describing the relationship o f fermented unsqueezed kocho as a function o f
enset plant height and pseudostem circumference measurements. The other achievement
which has a paramount contribution in enset research and development zoas the
development o f enset varieties fo r best quality and yield o f kocho which zoas conducted on
 23 M ikias Yeshitila and Zcrihun Ycmataw

n total o f about 400 enset accessions collected from the major enset growing zones
(Kembata, Daioro, Gamo-Gofa, Wolaita, Sidaiua and Gurage) o f SNNP Region o f
Ethiopia. The experiment resulted in the identification and release o f six enset varieties,
namely Yanbule, Geioada and Endale from early maturing (3 to 4 years) and Kelisa,
Zerita and M esena from late maturing (4 to 5 years). Future breeding researches should
also focu s on undertaking further work in exploiting the previously untraced enset
growing areas o f the country to collect all the possible enset accessions and preserve them
fo r future use and to guarantee and ensure the availability o f these resources in the future
with back up support o f in vitro (cryopreseroation and slozo growth) techniques.
Development o f a well established taxonomic classification and descriptor list and
undertaking molecular characterization o f all the available enset clones using recent
molecular techniques should also be given due emphasis. In addition, more new enset
varieties with merits o f good culinary qualities and disease resistance especially to
bacterial wilt o f enset should be developed through cross breeding, mutation and genetic
engineering to achieve food security and ensure sustainability.

Introduction
Enset (Ensete ventricosum (Welw.) Chessman is a perennial, herbaceous and
monocarpic crop belonging to the family Musaceae (Westphal 1975). It is a large
plant with a height usually grown about 3-8 m. The pseudostem (false stem), which
consists of a system of tightly or loose clasping and lax overlapping leaf-sheaths,
dilated at the base and having circumference and height ranges of 0.55-2.27m and
0.7-3.6 m, respectively. The length and width of matured leaf are in the range of 0.7-
4.82 m and 0.17-0.97 m, respectively (Yeshitila and Diro, 2009a).

Enset corm is typically, stout tuberous or fleshv with conically growing structure. It
bears lateral and apical buds at the centre of the junction of corm and pseudostem.
Corm is also the source of kocho and it is also used as fermenting agent or starter.
Corm weight of matured (flowered) enset plant after grated was in the range of 12.5-
75.0 kg/plant. Botanically true stem remain short at the base of central shoot at
vegetative stage. When plant enters generative stage the true stem grows fast. At
maturity the true stem bears peduncle and inflorescence at the top, which are
pendulous. Enset flowers are many to each bracts in two rows, those at the lower
inflorescence are mostly female, the middle are hermaphrodite and while the upper
inflorescence are male. The fruits of enset are dry and leathery except their size the
fruits resemble banana fruits. Enset produces hard and blackish seeds (Tabogie,
1997)

There are six commonly recognized species of Ensete, E. superbuni and E. glnucuni
grow wild in Asia, E. perrieri in Madagascar and E. gilletii, E. homblei and f
ventricosum in eastern Africa (Simmonds, 1962). Some species of Ensete are also
reported to grow in North America (Chessman, 1947). Ensete ventricosum is
considered to be the only wild species growing in Ethiopia (Simmons, 1956). Ensete
ventricosum was previouslv cultivated only in the south and south-western parts of
 24 Past Rcscarcli Achievement and Existing Gaps on Enset Breeding

Ethiopia but the recurrent droughts have led to the expansion of enset cultivation to
other parts of the country (Birmeta, 2004). The optimal conditions for enset
cultivation occur at 2000-2750 m with 1100-1500 mm rainfall, a temperature range of
10-21 °C and a relative humidity of 63-80% (Brandt ct id, 1997). Enset often grows
best in acidic, heavy clay soils that retain high levels of organic matter when
manured (Shank, 1994).

Enset cultivation is concentrated in the highlands of south and south-western

Ethiopia covering approximately 168,000 ha and is the staple and co-staple food
source for about 10 million people (CSA, 1997). Productivity of the crop is also very
high compared to other crops but varies depending on edaphic factors, altitude,
cultural practices and varietal differences (Birmeta, 2004). Actual fresh kocho yield of
enset reported by CSA (1997) was 28.01 and 25.93 kg/plant in Oromia and SNNP
regions, respectively. From these figures the national average actual "kocho" yield in
tons ha-1 y e a r1 was found out to be 11.87, assuming a growth period of 4.5 years and
plant population of 2222 enset stands per hectare in farmers' fields. Bezunhe (1984)
also reported the average national average "kocho'' yield to be between 7.41 and 11.95
tons ha-1 year-1 assuming 2000 plants/hectare. Based on the CSA (2009) the total area
under enset production in Ethiopia is 278667.5 ha with annual total kocho yield
production of 5565898.5 qt from these figures the national average actual "kocho"
yield was found out to be about 2.0 tons ha-1 y e a r1. Fresh yield potential of "kocho"
may go up to 150 tons ha-1 y e a r1 assuming growth period of 3 years and plant
population of 2500 enset plants per hectare (calculated from total dry biomass which
is 30% of the total fresh total biomass with harvest index of 0.5) (Tsegaye, 2002).

Enset is a multipurpose crop with all plant parts being utilized for human food,
animal forage, medicinal or ornamental uses (Pijls ct n i, 1994). It has high
significance in day-today-life of the peasant households cultivating this crop as
staple food (Bezuneh, 1983; Brandt ct nl. 1997). The peasants indicate that enset is
their food, their cloth, their house, their bed, their cattle-feed and their plate (Brandi
ct d., 1997). The types of products obtained from enset are kocho, bulla and uiuicho,
which are used as human food providing mainly carbohydrate and energy. Fibre is
the by-product, which according to Bzuneh (1996) is believed to be an excellent
structure and its strength was found to be equivalent to important fiber crop Musn
tcitalis and serves as a valuable raw material in local industries. Enset starch is also
being used for textile, paper and adhesive industries in saving the foreign exchange
(Ethiopian Science and Technology Commission, 2003). Enset also plays an
important role as a feed for animals during dry spell when feed is scare. Some of the
enset clones are used as local medication for different illness such as bone fracture,
bone breakage, diarrhoea, in discharging placenta, for both human being and
animals (Tsegaye, 1991). It therefore contributes enormously to food availability and
^ economic sustainability.
 25 M ikias Yeshitila and Zcrihun Ycmataiv

In Ethiopia enset based farming system is an indigenous and sustainable agricultural

system. Due to productivity and tolerance to transient drought, enset is regarded as a
priority crop in Ethiopia. The kocho, major yield, of enset per unit area and time (in
terms of weight and energy) is much higher than the yields of any other crop grown
iii Ethiopia, and therefore, the cultivation of onset in densely populated areas under
low input conditions can sustain the population better than any other crop (Tsegaye,
2002). It has also been known to buffer food deficit during dry spell and recurrent
drought because of its resistance to a fluctuating rainfall pattern after establishment.
In case of food shortage enset plant can be harvested at any age of the plant and at
any time of the year. Under normal storage condition kocho can be stored up to 10
years without being spoiled (Bezuneh, 1983; Alemu and Sandford, 1991; Tabogie,
1997). The corm (nniicho) is also cooked and consumed in the same way with other
root and tuber crops and it is a source of food during critical food shortage in parts ol
a year. As a result, enset has tremendously been contributing to food security in the
country. The crop represents 65% of the total crop production and the highly
carbohydrate rich products serve as staple and co-staple food for about 20% ol
population of Ethiopia (Brandt et id., 1997; Spring et al., 1996; Alemu and Sandford,
1991).

The plant is also an attractive evergreen ornamental which is being introduced to

different geographical regions of Ethiopia. The' presence of 20-40 magnificently huge
enset plants framing small huts and farmyards presents an impressive picture
(Shank, 1994).

Enset has very large leaf area and the canopy is closed after plants are established
and thus it protects rainfall from splashing the soil; the leaf litter checks runoff and
also improves nutrient recycling. According to Elias (1998) soil fertility is being
maintained, and even increased, in enset fields. It was also emphasized that erosion
does not occur in these fields, because of high organic matter and a more stable soil
structure, the presence of mulch material and greater care provided by the farmers.
Woluetensay (1997) also reported that higher levels of nutrients are present in enset
fields than in non-enset fields. As a result of such multifarious uses of enset makes it
highly related with sustainable agriculture and its cultivation is deeply entwined in
the economic, cultural and social life of different ethnic groups (Birmeta, 2004).

Despite, the many uses of enset which make the crop to be highly related to
sustainable agriculture, there are a lot of biotic and abiotic factors still existing
threatening its production and productivity. In the past various research activities
have been conducted to generate new production technologies and to promote them
to the end users. Many of the researches on enset have been concentrated on
agronomic studies (Bezuneh 1996). Agronomic researches were developed mainly by
Areka, among which propagation (seed/vegetative), planting time, plant spacing,
frequency and rate of organic and inorganic fertilizer application and frequency ol
transplanting are the major ones (Diro, 1997; Belehu, 1996; Tabogie et al., 1996; Haile
 26 Past Research Achievement and Existing Gaps on Enset Breeding

et ill., 1996; Diro ct al., 1996; Tsegaye and Struik, 2000; Gebremariam, 1996).
Concerning issues in relation to disease management options, research findings on
pathogen survival, means of dissemination, pathogenicity test, sanitary control, and
identification of some tolerant clones and enset root mealy bug control measures are
among the major ones. On the genetic improvement of enset only some limited
works were conducted so far, of which the most important ones are indicated below
with their research gaps.

Previous Research Achievements on Enset Breeding:

Collection stud M aintenance o f E nset Gcnngdasin

Most of the genetic diversity of enset is traditionally maintained in situ by farmers
(Shigeta, 1991). Unfortunately, many valuable clones have been lost because of
various human and environmental factors (Gebremariam, 1996), which may have
reduced the total available genetic diversity of the crop. Nowadays, many people
have become aware of the importance of preserving genetic diversity for the survival
and continuation of any form of life that exists on earth. Data on the relative genetic
diversity within and among plant populations can have a major significance in the
preservation of genetic diversity in crops and the improvement and maintenance of
crop germplasm for sustainable agriculture (Birmeta, 2004).

Attempts were made by researchers of Areka Agricultural Research Center to collect

and preserve all the possible enset germplasms in the country and currently, a total
of 62 ^ enset germplasms from 12 major enset growing areas of Ethiopia (94, 93, 71,
43, 49, 44, 49, 35, 27, 29, 57, and 32 enset accessions from Kembata/Hadiya,
Dawro/Waka, Gamo/Gofa, Wolayta, Sidarno, Guragie, Yem, West
Shewa/Southwest Shewa, East Shewa, Kaffa, Sheka, and Jimma respectively) have
been collected and conserved cx situ by the Southern Agricultural Research Institute
ol Agricultural Research at Areka (Yeshitila, et al., 2009). According to information
obtained from questionnaires being distributed among enset growing areas, the
expected number of vernaculars in Sidama, Gedio, wolaita, Jamjem, Gomu-Gofa,
west Shewa and kembata Hadya was 97, 42, 37, 28, 106, 70 and 142, respectively.
Therefore, further work has to be done in exploiting the previously untraced ense*
growing areas of the country to collect all the possible enset accessions and preserve
them for future use.

A number of enset accessions, which could be used in future crop improvement

program for the development of resistant/tolerant to bacterial wilt and other
desirable agronomic traits, are maintained in field gene bank at Areka ARC (Yeshitila
et al., 2009) and in situ by farmers (Gebremariam, 1996). However, these accessions
are exposed to biotic and abiotic factors. The clones also need to be rejuvenated on
average once per two years to keep them in the maintenance field, which makes the
process labour intensive, apart from taking more space and time. Moreover, all enset
accession might be infected by currently devastating disease, enset bacterial due to
 27 M ikias Yeshitila and Zerihun Yeniatazv

the motility of the causative agent. In order to ensure the availability of these
resources in the future, in vitro (cryopreservation and slow growth) conservation back
up is important.

Phenotype and Genotype B ase Variability Study o f Enset Clones

Under Areka Condition
Most of the earlier research on enset has focused on agronomic studies that include
yield and productivity, plant density and intercropping (Bezunhe, 1996). Hence, only
limited research was conducted on the molecular characterization of enset clones of
which molecular characterization research conducted by Tsegaye (2002), Negash
(2001) and Birmeta (2004) are the major ones. However, only few enset accessions
from a few enset growing area were molecularly characterized in their study. In
addition, attempts have been made to document and analyse clonal identity using
farmers' classification. In these cases, clonal names reported in the literature are
associated with the limited phenotypic data provided by farmers (Shigeta, 1991).

Morphological characterization of enset clones based on phenotypic, quantitative and

qualitative traits has been done at Areka Agricultural Research Centre (Tabogie, 1997
and Yeshitila and Diro, 2009b; Yemataw, 2010). However, morphological
characterization of the clones is rudimental and a well-established taxonomic
classification and descriptor list are lacking. The morphological characterization and
evaluation work in the study undertaken by Areka Agricultural Research Center w as
the first of its kind. It had considered the majority of enset accessions from the six
major enset growing areas of Southern region of Ethiopia. Thus, it is of paramount
importance in providing reliable information for future work where duplication ol
efforts that are creating complications in the conservation, improvement, and
u tilization of enset by farmers, conservationists and breeders would be avoided. In
the study two hundred seventy seven accessions of Enset ventricosum were tested at
Areka Agricultural Research Centre in a non-replicated field. The objectives of the
study were to assess the diversity of these accessions based on key morphological
descriptors and evaluate the accessions for yield and yield related traits. Data on 6
qualitative and 22 quantitative traits were collected and subjected to various
statistical analyses (Gomez and Gomez, 1984). Cluster analysis based on qualitative
characters indicated the formation of six clusters and existence of variability on leaf,
midrib and petiole traits. The result of Shannon-Weaver diversity index (H ) showed
lia t low levels of diversity existed among the 279 Enset ventricosum accessions based
on the frequency of the phenotypic characters considered. Analysis of variance for
quantitative characters indicated significant variation among the accessions in all the
22 yield and yield components of enset plant. The cluster and distance analysis of
quantitative characters pointed out the distance between most of the clusters were
highly significant (P < 0.01) indicating diversity among accessions in the different
clusters. Correlation study between various quantitative characters showed highly,
significant association among characters. Maturity time has a negative correlation
 28 Past Research Achievement and Existing Gaps on Enset Breeding

with yield and yield components of an enset plant (Yeshitila and Diro, 2009b;
Yemataw, 2010).

However, morphological characterization of the clones is rudimental and a well-

established taxonomic classification and descriptor list are still lacking. Few attempts
have also been made to document and analyze clonal identity using farmers'
classification. In these cases, clonal names reported in the literature are associated
with only limited phenotypic data provided by farmers (Shigeta, 1991). Although
assessment of morphological variation present in enset is feasible, rather its use is
limited due to the small number of phenotypic markers and the fact that they are
influenced by environmental conditions. Molecular characterization of enset clones
was conducted using AFLP (Tsegaye 2002; Negash 2001) and RAPD techniques
(Birmeta, 2004). However, enset accessions form limited enset growing areas were
considered in their studies. In the future, the development of molecular markers and
marker assisted selection methods for improved characterization of the different
Enset accessions collected from all the major enset growing areas of the country,
employing few of the different types of molecular markers that have been applied on
Musa, such as: RFLP, AFLP based on PCR amplification of selected restriction
fragments and microsatellites would have a paramount importance for future
advancement of enset research.

Development o f Enset Yield Estim ation Models

Attempts were also made to develop regression model which, non-destructively,
predicts yield of enset with better precision and simplifying yield evaluation in
experiments and also solve difficulties in estimating kocho yield in the assessment of
production balance in enset production region of the country. The experiment was
carried out at Areka Agricultural Research Center's on-station site on a total number
of 328 enset plants from the six major enset growing areas of Southern Ethiopia (59,
46, 49, 44, 49, and 81 from Gamo-Gofa, Guragie, Wolaita, Sidamo, Dawro/Waka and
Kembata/Hadiya respectively). Plant height and pseudostem circumference were
found out to be the best non-destructive enset kocho yield predictors. The R-square
value for estimating fermented squeezed kocho yield was about 0.78 with the
equation FUNK (kg/plant) = -25.76 + 5.35PH (m) + 20.01 PSC (m) describing the
relationship of fermented un-squeezed kocho as a function of enset plant height and
pseudostem circumference measurements. This research finding recommends the
prediction of fermented squeezed kocho yield from the linear relationship with the
fermented un-squeezed kocho yield with the equation FSQK (kg/plant) = 0.69FUNK
(kg/plant) (with R2=0.86). From the slope of linear relation line which was forced
through the origin the amount of kocho in un-squeezed kocho is about 69% and the
rest 31% will be the moisture. Similarly a linear model was developed to estimate the
kocho dry matter content from the un-squeezed kocho.yield using the equation Dry
matter (kg/plant) = 0.044FUNK (kg/plant). From the linear relationship (with R2 =
0.77) the dry matter content was found out to be only 4.4 kg in 100 kg fermented un­
squeezed kocho.
 29 M ikias Yeshitila and Zerihun Yemataiv

Development o f E n set Varieties

Last but not least from the past enset breeding research achievements was the
development of enset varieties for best quality and yield of kocho which was
conducted on a total of about 400 enset accessions collected from the major enset
growing zones (Kembata, Dawro, Gamo-Gofa, Wolaita, Sidama and Gurage) of
SNNP Region of Ethiopia. The accessions were evaluated, characterized and
maintained at Areka Agricultural Research Center. In the year 2006 about 9
promising enset accessions were selected and laid for further verification for their
adaptation, yield ("kocho") quantity and quality, disease and pest tolerance, at
different agro-ecologies of five research centers' on-station fields and six farmers
fields. Along with the leading center of Areka the key collaborators were the Holleta,
'imma, Kulumsa Research Centers and the Awada Sub-center and six farmers' fields
at Wonago, Chichu and Areka. The experiment resulted in the identification and
release of six enset varieties with merits of best "kocho” yield and quality and
registered by the Ministry of Agriculture and Rural Development, Animal and Plant
Health Regulatory Department in 2009 and 2010 for early and late set varieties
respectively. The varieties are Yanbule, Gewnda and Endale from early maturing (3 to 4
years) and Kelisa, Zerita and Mesena from late maturing (4 to 5 years). Some nf the
quantitative and qualitative morphological and other traits of the varieties against
local checks of early set (Local Check 1) and late set (Local Check 2) are described
below in Table 1 and 2. (Yeshitila et al., 2011).

However, in the future variety development of enset should also encompass crossing
breeding techniques, to derive enset varieties lrom crosses between established clones
of parents complementary in the trait they possess each of which could contribute a
different set of desired genes (e.g. wild with clones having good culinary quality) and
to convey selection of varieties having various characteristics by observation during
fh< growing of the genotypes in the field including tests to determine resistance,
cu inary quality, earliness etc. Further breeding techniques on mutation and genetic
engineering should also be carried out especially to develop elite clones resistant to
the overwhelming problems of bacterial wilt disease.
 30 ________ Past Research Achievement and Existing Gaps on Enset Breeding

Table 1. Average quantitative values of the six released varieties over early and late set local checks

AVERAGE QUANTITATIVE VALUES AT Early set varieties (year of release 2009) Late set varieties (year of release 2010)
MATURITY
Yanbule Gewada Endale Local check 1 Kelisa Zerita Mesena Local check 2
Pseudostem height (m)
Research (on-station) 2.35 1.72 1.98 1.32 1.60 1.66 1.58 0.98
Farmers field 1.86 1.82 1.50 1.42 1.59 1.67 1.86 1.16
Pseudostem circumference (m)
Research (on-station) 1.44 1.22 1.30 0.94 1.27 1.32 1.13 0.86
Farmers field 1.01 0.94 1.06 0.70 1.00 1.06 0.99 0.73
Leaf height (m)
Research (on-station) 4.90 4.10 4.33 3.28 3.60 3.99 3.52 2.46
Farmers field 3.93 4.05 3.58 3.04 3.40 4.02 3.86 2.86
Leaf width (rn)
Research (on-station) 1.10 0.90 0.88 0.70 0.84 0.85 0.84 0.66
Farmers field 0.75 0.78 0.71 0.64 0.74 0.80 0.78 0.63
Leaf num ber
Research (on-station) 12.00 11 11 11 11 12 11 10
Farmers field 10.00 9 11 8 12 10 10 10
Unsqueezed Kocho (t/h/y) .
Research (on-station) 31.49 22.75 26.16 14.18 23.13 24.58 19.81 10.10
Farmers field 18.96 17.17 17.77 9.38 17.15 19.01 18.59 8.33
Squeezed Kocho (t/h/y)
Research (on-station) 21.12 15.13 17.47 9.26 15.39 16.39 13.12 6.47
Farmers field 12.54 11.31 11.72 5.98 11.30 12.57 i 2.26 5.26
STARCH VALUES IN BULLA
Starch (%) 90.00 92.5 77.5 82.5 85 95
pH 5.49 5.16 5.73 5.31 5.32 5.7
Moisture (%) 12.50 13.4 15.5 13.1 15.5 14.5
Viscosity (mm2/sec) 15.70 10.69 13.06 17.28 9.27 14.68
Gelatinization (Temp oC) 75.00 75 75 75 75 75
 31 M ikias Yeshitila and Zerihnn Yemataw

VALUES INKOCHO AFTER

FERMENTATION
Moisture (%) 62.8 62.07 69.58 66.48 62.17 65.83
Fat/Ether extract (%) 0.11 0.09 0.34 0.1 0.09 0.15
Protein (N% x 6.25) 0.75 0.76 2.66 1.07 0.64 0.82
Ash (%) 1.15 1.33 1.23 1.17 1.13 1.24
Crude fiber (%) 1.59 1.37 1.34 1.76 1.52 1.3
Zinc (mg/1 OOgm) 2.65 1.7 5.34 8.05 0.76 4.05
Calcium (mg/100 gm) 207.14 264.49 232.9 229.7 190.02 223.4
Iron (mg/100 gm) 6.45 5.25 12.53 10.52 7.85 9.1
Phosphorus (mg/100 gm) 2.48 2.37 2.5 3.03 2.26 2.27
Total Carbohydrate (%) 34.35 35.14 27.51 30.49 35.09 31.48
 32 Past Research Achievement and Existing G aps on Enset Breeding

Table 2. Qualitative measurement values of the six released enset varieties

Early set varieties (year of release 2009) Late set varieties (year of release 2010)
Yanbule Gewada Endale Kelisa Zerita Mesena
Growth/leaf habit Erect Erect Intermediate Erect Erect Erect

Mature leaf shape Lanceolate Lanceolate Lanceolate Lanceolate Lanceolate Lanceolate

Pseudostem Color Medium Green with Medium Green Purple Brown Medium Green Brown (Rusty Brown) Orange Red with Light
Black Patches Green
Midrib Upperside Color Light Green Light Green Medium Green with Medium Green Light Green with Red Medium Green with
Red Purple at the Purple and Medium Red Purple at Middle
middle Green

Midrib Underside Color Pink Purple with Light Half Yellow Green and Red Purple with Light Light Green Pink Purple with Light Red Purple with Light
Green half Pink Purple with Green Green Green
Light Green
Petiole Upperside Color Light Green Light Green Light Green Light Green Light Green with Purple Light Green
Brown

Petiole Underside Color Light Green with Biack Yellow Green Rusty Brown with Light Light Green Medium Green and Pink Purple with Light
Patches Green Brown with Black Green
Patches and Spots

Leaf Color Medium Green Green Green Green Green Green

Leaf Tip and Edge Color Medium Green Green Red Green Red Red
 33 M ikias Yeshitila and Zcrihun Yemataiv

Research gaps and Recommendation*

R esearch gaps
• The num ber of already collected and conserved accessions is far below the expected
number of vernaculars in the different growing regions of the country.
• There are also other species of enset distributed in the parts of the world mainly in
Africa and Asia which could have different merits for future breeding purpose.
However
• Currently, the already collected enset accessions are only conserved in situ in
farmers' fields and ex situ at on station field of Areka Agricultural Research Center,
without in vitro back up conservation techniques.
• Despite, the attempts made so far on the morphological characterization of the
clones, this technique is rudimentary and a well-established taxonomic classification
and descriptor list are still lacking.
• Molecular characterization works are also not conducted on the large number of
enset accessions collected and conserved at Areka on-station field to strengthen the
information obtained from the morphological characterization.
• Enset yield estimation models accounting the inter clonal, age group, agro-ecological,
and harvesting time differences to predict the different yield products (kocho, bulla,
amicho, and fiber) of an enset plant non-d.estructively are still lacking. The already
developed enset yield estimation model is having a limited use in that the sample
clones were considered at one location and only works for 'kocho' yield estimation;
models for 'bu lla', 'am icho' and fiber yield estimations are not yet developed.
• On the development of enset varieties, very limited attempts have been made so far
which is only concentrating on the selection of elite clones from the available
accessions through phenotypic evaluation and characterization techniques. Still now
no other breeding techniques like crossing, mutation, genetic engineering have been
tried to develop new enset varieties, despite the enormous significance of enset in the
country.

Recom m endations
• Collection should be conducted on both cultivated and wild types of enset clones in
the different parts of the country, which have not yet been carried out to the extent of
the need. Other species of the genus ensete found in the other parts of the world
should also be introduced and preserved.
• Accessions which were collected so far maintained in field gene bank are exposed to
all biotic and abiotic aggressions. Therefore In vitro conservation facility should be
strengthened to complement the in situ and ex situ conservation.
• Morphological characterization which is conducted so far should be further
strengthened and supplemented by molecular characterization techniques bv
encom passing a large number of accessions.
• Enset yield estim ation models accounting the inter clonal, age group, agro-ecological,
and harvesting time differences should be developed to predict the different yield
products (kocho, bulla, amicho, and fiber) of an enset plant non-destructively.
• Conventional breeding in combination with mutation and molecular breeding is
required to further broaden enset genetic base which could contribute towards
successful variety development. In general, knowledge of and facilities for
 34 Past Research Achievement and Existing Gaps on Enset Breeding

biotechnology research are of paramount importance in the short term.

As compared to the importance of enset as a food security crop for a large proportion
of people in the country, the attention given to strengthen the human power status
and facilities of research is very low which need to be re-vitalized by the national
developm ent seekers.

References
Alemu, K. & Sandford, S. 1991. Enset in North Omo Region. Farmer's Res. Project Technical
Pamphlet No. 1. Farm Africa, Addis Abeba. 49 pp.
Belehu, T. 1996. Enset Research in Ethiopia, 1985-1993. In, Tsedeke, A., Hiebisch, C, Brandt
SA, Seifu, G. (eds.) Enset based sustainable agriculture in Ethiopia. Proceedings from the
International Workshop on enset held in Addis Ababa, 13-20 December 1993, pp. 221 -
227.
Bezunhe, T. 1983. Evaluation of some Ensete ventricosum clones for food yield with emphasis
on the effect of length of fermentation on carbohydrate and calcium content. Tropical
Agriculture 61, 111.
Bezunhe, T. 1984. Evaluation of some Ensete ventricosum clones for food yield with emphasis
on the effect of length of fermentation on carbohydrate and calcium content. Tropical
Agriculture 61,111-116.
Bezuneh, T. 1996. An overview on enset research and future technological needs for
enhancing its production and utilization. In, Tsedeke, A., Hiebisch, C, Brandt SA, Seifu,
G. (eds.) Enset based sustainable agriculture in Ethiopia. Proceedings from the International
Workshop on enset held in Addis Ababa, 13-20 December 1993, pp. 3-12.
Birmeta, G. 2004. Genetic Variability and Biotechnological Studies for the Conservation and
Im provem ent of Ensete ventricosu. Doctoral thesis, Swedish University of Agricultural
Sciences, Alnarp. 91 pp.
Brandt, S.A., Spring, A., Hiebisch, C., McCabe, J.M., Tabogie, E., Diro, M., Wolde-Michael, G.,
Yantiso, G., Shigeta, M. & Tesfaye, S. 1997. The tree against hunger. Enset-based agricultural
systems in Ethiopia. American Association for the Advancement of Science with the
Awassa Agricultural Research Centre, Kyoto University for African studies and
L niversity o f Florida, Washington.
CSA, 1997. Area, production and yield of crop of private holdings in 1996/97 in Meher
season. Addis Ababa, Ethiopia.
CSA, 2009. Area, production and yield of crop of private holdings in 2008/2009 (2001 E.C) in
Meher season. Addis Ababa, Ethiopia.
Cheesman, E.E.1947. Classification of the bananas: The genus Enset Horan. Kewr Bulletin 2:
97-117.
Diro VI., Haile B., and Tabogie E. 1996. Enset propagation research review. P. 242-249. In:
Tsedeke-Abate, Hiebsch Clifton, Brandt Steven A. and Seifu Gebremariam (eds). Enset-
based Sustainable Agriculture in Ethiopia: Proceedings from the International Workshop
o i Enset. 13-20 Dec 1993. Institute of Agricultural Research, Addis Ababa
Diro, M. 1997. Effect of propagation method and corm type on number and growth of enset
(Enset ventricosum) suckers. MSc Thesis University of Alemaya, Ethiopia. 69 pp.
* Elias lya. 1998 Soil fertility management and nutrient balance in Kindo Koisha farms, 1996
Unpub.
 35 M ikias Yeshitila ami Zerilnm Yemataiv

Ethiopian Science and Technology Commission (ESTC) http://www.capitalothiopia.com

f 2003.
Gebremariam, S. 1996. Enset research in Ethiopia In, lsedeke, A., Hiebisch, C, Brandt SA,
Seifu, G. (eds.) Enset bused sustainable agriculture in Ethiopia. Proceedings from the
International W orkshop on enset held in Addis Ababa, 13-20 December 1993, pp.204-220.
Gomez,K.A, and Gomez, A.G. 1984. Statistical Analysis for Agric. Research. John Wiley and
Sons. New York
Haile, B., Diro, M., Tabogie, E. 1996. Agronomic research on enset. In, l sedeke, A., I liebisch,
C, Brandt SA, Seifu, G. (eds.) Enset based sustainable agriculture in Ethiopia. Proceedings
from the International Workshop on enset held in Addis Ababa, 13-20 December 1993,
pp. 235-241.
Negash, A. 2001. Diversity anti conservation of enset (Enset ventricosum (Welw.) Cheesman)
and its relation to household food and livelihood security in southwestern Ethiopia. PhD
thesis, Wageningen University and Research Centre.
Pijfs, T., J. Ainoid, A.M. timmer, Zewdie Wolde - Gwbriel, and C live E. West. 1994.
Cultivation, Preparation and consumption of enset in Ethiopia. /. o f Sci. Food and Agri.
Shank, R. 1994. The Enset Culture, Technical Report on Enset or the False Banana United
Nations-Emergencies Unit for Ethiopia.
Shigeta, M. 1991. Folk in situ conservation of enset (Enset ventricosum (Welw.) Cheesman):
Towards the interpretation of indigenous agricultural science of the Ari, southewestern
' Ethiopia. Asia and African Area Studies. 2, 1-25.
Simmons, F.J. 1956. The role of Ensete in Ethiopia The Geographical Review 46, 271-272.
Simmonds, N.W. 1962. The evolution of the bananas. Butler and Tanner Ltd. London.
Spring, A., Haile, B., Tesfaye, S., Abebe, Y., Amaldegn, A., W oldemichael, G., Tabogie, 1
I Surur, O., Tsegaye, A., Shimeles, S., Habte, T., Menjeye, T., Tadesse, 1 1996. Enset
farming system in southern region, Ethiopia: Report on a Rapid Rural Appraisal in
Guragie, Hadiya, and Sidama zones. Deutsche Gesellschaft fur Technische
Zusam m enarbeit (GTZ), Addis Ababa, Ethiopia, mimeo.
Tabogie, E., Diro, M., and Haile, B. 1996. Review of past and present enset improvement
activities. P228-234. In, lsedeke, A., Hiebisch, C, Brandt SA, Seifu, G. (eds.) Enset basal
sustainable agriculture in Ethiopia. Proceedings from the International Workshop on enset
held in Addis Ababa, 13-20 December 1993, pp.228-234.
Tabogie, E. 1997. Morphological characterization of enset (Ensete ventricosum (Welw )
Cheesm an) clones and the association of yield with different traits. MSc Thesis, Alemaya
University of Agriculture, Alemaya, Ethiopia. 89 pp.
Tsegaye, A. & Struik, P.C. 2000. Influence of repetitive transplanting and leaf pruning on dr\
matter and food production of enset (Enset ventricosum (Welw.) Cheesman). Field Crops
Research 68, 61-74 (Chapter 6).
Tsegaye, A.2002. On indigenous production, genetic diversity and crop ecologv of enset
(Enset ventricosum (Welw.) Cheesman). Doctoral thesis, Wageningen University, l'he
Netherlands. 105 pp.
Tsegaye, B. 1991. Community management of crop genetic resources in the enset complex
farming systems of southern Ethiopia: A case study from Sidamo region. MSc thesis,
Agricultural University of Norway.
Westphal, E. 1975. Agricultural system in Ethiopia. Centre for Agricultural Publishing and
Documentation, Agricultural University, Wageningen. The Netherlands.
Woldetensay, A. 1997. The ecology and production of Enset ventricosum in Ethiopia. Doctoral
thesis, Swedish University of Agricultural Sc iences, Uppsala. 129 pp.
 36 Past Research Achievement and Existing Gaps on Enset Breeding

Yemataw, Z. 2010. Variability study and indigenous classification methods of enset (Enset
ventricosum (Welvv.) Cheesman) clones in SNNPR. MSc thesis, Hawassa University,
Ethiopia.
Yeshitila M and Diro M. 2009a. Estimation of Kocho Yield from linear dimensions of enset
(Enset ventricosum (Welw.) Cheesman) plant. Proceedings of the 13th Annual Conference
of the Crop Science Society of Ethiopia. In Kebebew Assefa and Woldeyesus Sinebo (ed.)
Addis Ababa, Ethiopia. 123-134 pp.
Yes litila M and Diro M.2009b. Phenotype and genotype base variability study of enset (Enset
ventricosum (Welvv.) Cheesman) clones at Areka condition. Proceedings of the 13lh Annual
Conference of the Crop Science Society of Ethiopia. In Kebebew Assefa and Woldeyesus
Sinebo (ed.) Addis Ababa, Ethiopia.364p
Yeshitila M, Diro M, Handoro F, Yemataw Z, and Muzemil S. 2009. Improved Enset (Enset
ventricosum (Welw.) Cheesman) production technologie. National Production Manual.
2009, MoARD and EIAR, Ethiopia
Yeshitila M., Yemataw Z., Muzemil S., Ayalew A., Negash F., Michael K., Bekele A., Chindi
A., G/Tensay F., Melaku D., W/Michael G. 2011. Registration of enset (Ensete
ventricosum (Welw.) Cheesman) varieties Yanbule, Gewada, Endale, Kelisa, Zerita and
Mesena. Accepted paper fo r EJAS (Ethiopian Journal o f Agricultural Science)
 Research Experiences oil Enset
Biotechnology in Ethiopia: A Review

Tesfaye D isasa1and Muliigeta Diro*

1Ethiopian Institute o f Agricultural Research, H ole tta Agricultural Research Center, Plant
Biotechnology Research Project, P. O. Box 2003, Addis Ababa, Ethiopia.
■Southern Agricultural Research Institute, Crop Research Process, Hawassa, Ethiopia

Summitry
•

Enset is herbaceous perennial multipurpose crop which is one of the most important
food crops for approximately quarter of the population of the country. Though Enset
is one of the national commodity crops and grown by large number of population, its
production is highly constrained by a number of biotic and abiotic factors. Diseases
insects, lack of infection free and high yielding quality planting materials are among
the major problems limiting its productivity. Biotechnological tools are one of the
interferences in addressing these challenges in support of conventional technologies.
During the last few years, some promising methods especially on tissue culture
optimization have been developed by different researchers. However, most of those
protocols were not reproducible and inefficient that needs further optimization and
refining. Recently, a protocol that has enabled to multiply disease free planting
material on massive scale has been optimized at Holetta Agricultural Research
Center. The application of molecular tools also play an important role especially in
identifying and developing marker linked to bacterial wilt resistance or tolerance
genes. Similarly, effective management of bacterial diseases requires precise
diagnosis and accurate identification of the causal agents through modern
techniques like molecular diagnosis. Despite of these potentials, lack of bacterial
resistant clone and limited effort to diagnose bacterial pathogen in tissue culture
derived plantlets are still some of the gaps that need further investigation. Using the
currently optimized protocol, it is possible to produce and supply large number of
quality planting material within a short period of time.
 38 Research Experiences on Enset Biotechnology in Ethiopia

Introduction

Enset is a herbaceous perennial crop belonging to the family musaceae (Cheesman,

1947). Both Enset and banana plants are similar in morphology but Enset is larger
and thicker than banana plant. The plant composed of leaves, pseudostem,
underground corm and adventitious roots. It reaches up to 13 m long and a diameter
of m (Birmeta et al., 2002). Unlike banana, the edible part of Enset is prepared from
the pseudostem and the underground corm rather than fruits. Enset is diploid
(2n=18) species (Cheesman, 1947) including the wild relatives.

Although Enset is adapted to wider climate condition, it usually grows best between
altitudinal ranges of 1500 and 3100 meters above sea level (Tsegaye, 2002; Birmata et
al., 2004b). Most Enset growing regions have an annual precipitation of 1100 to 1500
mm and a mean temperature of 10-21°C.

Enset is a stable food and multipurpose crop for approximately quarter of the
Ethiopian population. It is one of the national commodity crop cultivated by
sm allhold er subsistence farmers m ainly in Southern and the central Oromiya regions
(CSA, 1997). It plays enormous roles in the socioeconomic of the country. Enset is not
only an important food crop but also serves as fibre and important source of feed
particularly during the dry season (Bezeneh and Feleke, 1966; Nurfeta et al., 2009).
The main edible parts of Enset are the corm and psedostem. Kocho, Bulla, and Amicho
are the major food types obtained from Enset. The fermented product of Enset,
Kocho, is rich in carbohydrate and mineral substances like calcium and Iron. It is the
bu k of the fermented starch obtained from the mixture of decorticated (scraped) leaf
sheathes and grated corm. Bulla which is considered as the best quality of Enset
food, which is obtained through a special processes mainly from fully matured Enset
plants. 'Amicho' is the boiled Enset corm which contains higher dry matter than any
other part of the crop (Nurfeta et al., 2008).

It'g tolerance to abiotic factors and adaptability to wider climatic conditions have
enabled Enset for its rapid expansion to different parts of the country. Despite its
tolerance to some climatic and environmental fluctuation, the production and
productivity of Enset is threatened by different biotic and some abiotic factors.
Diseases, insect pests, and lack of high quality planting materials which are free of
diseases are the major constraints of Enset production system.
% 39 Tesfni/e Disasa and Mulugeta Diro

Enset bacterial wilt is one of the major bottle-neck threatening Enset production and
* productivity. This disease is caused by Xanthonionas campestris pv musacearum,. The
disease causes a yield loss of up to 100% in major growing areas (Ashagrai, 1985).
Xanthonionas canipestris is believed to be originated from Ethiopia (Ndunga ct al.,
200(5). Plants may be affected at any growth stage, including full maturity. The
disease can be spread via infected plant material, contaminated farming and
processing tools, insects, wind and humans/animals.The incidence and spread of the
disease through out the Enset growing areas of the country is persistently increasing
and up to 100% infection have been reported in different areas (Mesfin et al., 2007).
These days bacterial wilt is becoming a regional threats in that significant yield losses
ha\ e been reported across East African countries.

Existing research outputs and reports indicated that there are some Enset clones
. suspected to be tolerant/resistant to bacterial wilt. However, artificial inoculation of
previously reported tolerant/resistance Enset clones with bacterial wilt indicated
that all clones showed wilt symptoms with different level of incidence (Welde
Michael et al., 2008). Furthermore, they indicated that some of the clones recover
after the initial symptom development. Still more investigation is needed to generate
reliable information using modern techniques like molecular characterization and
development of markers linked to bacterial wilt resistant gene.

liiotcchnological Approaches and Research

Achievements in Enset Improvement

Agricultural biotechnology might be considered as an important approach in

addressing challenges of various biotic (pests and diseases) and abiotic (inorganic
factors such as drought, salinity and soil fertility) in agriculture. Coupled with
conventional technologies, it effectively addresses the problems of food security and
increased farmer incomes challenges. Biotechnological tools range from tissue
culture to molecular approaches and genetic transformation. In vitro techniques and
molecular tools are thus one of the potent means of interference especially in
producing disease free planting material and to identify bacterial wilt
- resistant/tolerant Enset clones. In addition, In vitro techniques help.in conservation
of quality planting material while molecular tools play an important role to identify
, and develop markers linked to bacterial wilt resistance/ tolerance genes, and thereby
to use them for marker assisted selection and breeding for better disease control and
to enhance productivity of the crop. Molecular characterization of different Enset
 40 Research Experiences on Enset Biotechnology in Ethiopia 9

clones helps to identify the existence of tolerance/resistance genes to bacterial wilt.

This helps to develop molecular markers and primers linked to the responsible
genes. This paper deals with detail review of some of the practical application of in
vitro culture and molecular markers for the improvement Enset production and
productivity. Some of these activities are summarized in Table 1.

Table 1. Application of biotechnological tools in Enset improvement

Biotech
approach Techniques Purpose Reference
Shoot tip Disease free Mathew and Philip, 1996; Negash et al., 2000; Birmata
multiplication et at., 2004b; Diro and Stedan, 2005\HARC unpubl.
Somatic embryo Micro propagation Afza et at., 1996; Negash et at., 2000; Mathew and
Tissue culture Philip, 2003
Zygotic embryo Regeneration Diro and Stedan, 2004
Adventitious buds Micro propagation Afza et at., 1996
RAPD Genetic diversity Birmeta et at.. 2002; Birmeta et at.. 2004a
Molecjlar AFLP Genetic conservation Negash et at., 2002
markers
rDNA Pathogen detection Birmeta et at., 2004c
Molecular RAPD Pathogen diversity Odipio etat., 2009
Diagnosis study
Rep-PCR Pathogen Aritua et at., 2007; Aritua et al., 2008
characterization

EiiM't tissue culture

Plant tissue culture is an important step in plant biotechnology since it facilitates
rapid vegetative multiplication (micro-propagation) of valuable plant material for
agriculture, horticulture and forestry. Multiplication of single explants into several
thousands of plants in very short time allows producing required amount of new
quality planting material in time. In addition, through tissue culture techniques, it is
possible to produce disease free plants, disease and pest resistant plants and others.
Furthermore, plants grown from tissue culture that pass through callus phase are
subjected to soma-clonal variations which might also be of breeders' interest like -
tolerance to pests, diseases, etc. Through tissue culture, it is also possible safely
clone rare and endangered plants and conserve them in 'tissue banks' through cryo ^
preservation. This approach can also be applied to conserve explants from superior
plants plant cultured in media are easier to export than soil-grown plants, as they are
 41 Tcsfaye Disasa and Mulugeta Diro

pathogen free and take up little space (nowadays many companies use this system).
The techniques in tissue culture, especially haplodization, also help in increasing
selection efficiency for crop improvement. The ability to produce progeny all
year-round is also another advantage of tissue culture.

Various in vitro culture and in vitro conservation researches have been conducted on
Enset (Table 1). The success of Enset in vitro culture is limited by several factors.
Some of the major constraints include; the excess exudation of phonolic compounds
during culturing, apical dominance nature of the crop and the presence of
endogenous bacteria contamination. Attempts have been made to suppress apical
dominance through the application of meristem wounding (Birmata et al., 2004b;
Diro and Stedan, 2005). Initiation of several buds from a single explant was achieved
through meristem wounding on modified MS medium (Birmata et al., 2004b).
However, this protocol has its own drawback of reproducibility. The degree of
browning due to phonolic exudation varies form explant to explant (Diro and Stedan,
2005). According to the authors, explants that are collected from in vitro grown
seedlings showed no blacking while the majority of explants from field grown
suckers were lost as a result of extensive blackening when cultured. The use of
activated charcoal, supplementing the media with different additives and growing
cultures under dark condition can prevent or remove inhibitory phenolic compounds
from the media. Establishing and maintaining the mother plant in controlled green
house prior to in vitro culture reduces the loss of explants as a result of endogenous
bacterial contamination (Birmeta et al., 2004c). The current research result showed
that the degree of endogenous contamination depends on the size and age of the
explant in addition to mother plant management (unpublished data). The smaller
and the younger the explant is the little endogenous contamination to occur.
Recently, an efficient disease free in vitro multiplication protocol was optimized at
Holetta Agricultural Research Center of the Ethiopian Institute of Agricultural
Research. This has enabled to produce large number of disease free seedlings within
short period of time (Figure 1). The activity involves several stages that ranging from
selection and establishment of mother plants to acclimatization of the plantlets in
green house and field condition.

In vitro conservation of Enset plays an important role in efficient preservation of

germplasm. A protocol for in vitro conservation of Enset has been achieved through
the application of growth retardants in growth medium and keeping the cultures at
low temperature (Negash et al., 2001).
 42 Research Experiences on Enset Biotechnology in Ethiopia

Applications o f M olecular M arkers in E nset

With the discovery of DNA marker technology, several types of DNA markers and
molecular breeding strategies are now available to plant breeders and geneticists.
Some of these DNA markers include; Random Amplified Polymorphic DNA
(RAPD), Restriction Fragment Length Polymorphisms (RFLP), Amplified Fragment
Length Polymorphic DNA (AFLP), Single Nucleotide Polymorphism (SNP)
Inter-specific Simple Sequence Repeat (ISSR) and Micro satellites (nSSR & cSSR).
These are useful tools in various applications like DNA fingerprinting, germplasm
conservation and genetic improvements. Limited attempts have been done so far to
analyze the genetic diversity of Enset clones as well as their wild relatives using
molecular tools such as RAPDs (Birmeta et al., 2002; Birmeta et al., 2004a) and AFLPs
(Negash et al., 2002).

Diagnostics and Characterization o f Enset B acterial W ilt

Plant diseases caused by various pathogens have been recognized as one of the most
important factors limiting agricultural production. The effective management of
these diseases requires precise diagnosis and accurate identification of the causal
ag;ents. Traditional methods such as visual inspection of the disease or its symptom,
examination under a microscope, pathogen isolation/culturing, biochemical,
serology-based methods including ELISA and conducting pathogenecity tests are
sometimes inadequate to reveal the nature of the causal agent. Developments in
biotechnological tools like monoclonal antibodies and nucleic acid-based methods
particularly polymerase chain reaction (PCR) has added a new dimension in the
diagnosis of diseases.

The use of rDNA analysis to identify some bacteria and fungi from in vitro and
surface sterilized field samples of Enset plantlets was undertaken (Birmeta et al.,
2004c). Result showed that both bacteria and fungus were isolated from both
samples. Among the isolated bacteria most of them are found to be gram positive.
Effort to characterize Xanthomonas campestris pv. musncearum, the causal agent of a
wilt of Enset and banana using molecular techniques was undertaken by researchers
at different times (Aritua et a l , 2007; Aritua et al., 2008). Results showed that these
causal agents are homogenous and very similar to other strains isolated from sugar
cane and maize. From this genetic relatedness one can speculate that these
populations may be evolved from the same ancestor.
 V i

Fig.1. In vitro propagation of Enset. A, Mother plant establishment. B, Initiation of Multiple shoots. C, multiplication of shoots.
D, elongation and rooting. E, Acclimatization
 44 Research Experiences on Enset Biotechnology / in Ethiopia

Despite of the various conventional and biotechnological applications in Enset

improvement, still there is a gap in resistant variety development against bacterial
wilt. Furthermore, there is no appropriate means of bacterial wilt identification for
tissue culture derived seedlings in the country although morphologically the
seedlings appear to be free of any pathogens.

Discussions

The application of biotechnology particularly tissue culture plays an important role

in Enset improvement. The technology helps to multiply bacterial wilt tolerant and
high yielding clones on massive scale in short period of time. The use of meristem or
shoot tip culture has enabled one to produce disease free plantlets. The regeneration
of whole plant from zygotic embryos strongly assists breeding program of Enset
since germination of seeds is quite poor (Negash et a l, 2000; Diro and Stedan, 2004a).
Multiple shoot induction can be achieved with different media combination but the
major factor in multiple shoot induction is meristem wounding technique in order to
avoid apical dominance (Birmata et a l, 2004b; Diro and Stedan, 2004a). According to
Diro and Stedan (2004), increasing the concentration of cytokinin doesn't have
significant difference on multiple shoot induction. The current achievement of large
number of multiple shoots from an explant through mersitem wounding
(unpublished data) is persistent with the report of Birmata et al. (2004b) and Diro and
Stedan. (2004a). this further strengthens the role of meristem wounding over plant
growth regulator treatments in Enset micro propagation. Meristem wounding
should be done with great care otherwise extensive wounding result in callus
induction. The currently optimized technology can be applicable for large scale
commercial propagation of quality planting material of Enset clones.
Conservation and effective utilization of plant genetic resource contributes to food
and nutritional security, poverty alleviation and environmental protection, which are
major challenges faced by humankind in the century. Conservation of Enset
gem p lasm can be possible by ex situ conservation of clones. However, traditional
method of ex situ conservation of germplasm of vegetatively propagated species like
Enset through field maintenance requires very large area of land. In vitro approaches,
including tissue culture maintenance and cryopreservation, are recognized as useful
tools for medium- to long-term conservation of these clones. Molecular techniques
such as AFLP (Negash et al., 2002) and RAPD (Birmata et a l, 2002; Birmeta et a l,
2004a) analysis are being used for aiding these methods. This plays an important role
to study the genetic variations within and among, populations. These variations
 45 Tesfai/c Disnsa and Mulugeta Diro

might be due to several factors such as mutation, introgression, recombination,

adaptation to environments and continuous selection. Result from RAPD analysis o'
Enset clones showed that there is high genetic diversity ir. wild Er.set population;
which could be a potential source of useful gene for the improvemc nt of cultivated
Enset (Birmeta ct al., 2004a).

The lack of appropriate bacterial wilt identification method in tissue culture derived
seedlings hinders the effort of further multiplication of the clones on a massive scale.

C o n clu sio n s and F u tu r e P ro s p e c ts

The various research outputs indicated that application of biotechnology for Enset
improvement is feasible for rapid multiplication of disease free planting material, in
vitro conservation, molecular characterization and development of markers linked to
bacteria wilt resistance genes. Shifting of medium status from semi solid to liquid
medium during shoot multiplication and rooting would be a strategic step for
automation and cost efficiency in disease free multiplication using bioreactors. The
current achievement of protocol optimization promises a large scale disease free
propagation of Enset in the future. Once this protocol is verified to be successful, it
can further be adopted by regional and private tissue culture labs in order to
multiply and supply large number of quality planting material using the optimized
protocol. The existing in vitro multiplied plantlets are morphologically appeared to
be free of any pathogen; however, further evaluation of plantlets using appropriate
disease diagnostic techniques is mandatory. Methods for pathogen identification
should be developed prior to in vitro multiplication of the selected clones. Further
development or application of molecular tools including co-dominant markers is
necessary for the improvement of Enset as well as development of resistant variety
against bacterial wilt.

A ck n o w led g em en t

W 3 are grateful to Mr. Demissew Sertse for his critical comments on the manuscript.
We also extend our gratitude to colleagues from biotechnology research of HARC,
who assisted in providing the necessary information.
 46 Research Experiences on Enset Biotechnology in Ethiopia

References

Afza, R., Van Duren, M. and Morpurgo, R. (19% ) Regeneration of Ensete ventricosimi through
somatic em bryogenesis and adventitious buds. Plant Cell Reports 15: 445-48.
Aritua, V., Nanyonjo, A., Kumakech, F. and Tushemereirwe W (2007), Rep-PCR reveals a
high genetic homogeneity amongUgandan isolates of Xanthomonas cnmpestris pv
niusaceanim. African Journal o f Biotechnology 6 (3): 179-83
Aritua,V., Parkinson, N., Thwaites, R., Heenev . I. V., Jones, I). R., 1ushemereirwo, VV .
Crozier , }., Reeder, R„ Stead , I). E. and Smith, J. ( 2008). Characterization ol the
Xanthomonas sp. causing wilt of Enset anti banana and its proposed reclassification <is ,i
strain of X. vasicola. Plant Pathology 57: 170-177
Ashagrai D. (1985). Studies on bacterial wilt of Enset, £. ventricusiun and prospects for its
control. Ethiopian Journal of Agricultural Science 1:1-14.
Bezuneh T. and Feleke A. (19% ) The production and utilization of the genus Ensete in
Ethiopia. Econ Botany 20: 65-70.
Birmeta, G., Nybom, H. and Bekele, E. (2004) Distinction between wild and cultivated Enset
(Ensete ventricosum ) gene pools in Ethiopia using RAPD markers. Hereditas (Lund)
140:139-48.
Birirteta, G., and W elander, M. (2004) Efficient micropropagation of Ensete ventricosum
applying meristem wounding: a three-step protocol Plant Cell Reports 23: 277-83.
Birmeta, G., Passoth, V., Roos, S. and Welander, M. (2004) Identification of bacteria and yeasts
from in vitro and surface-sterilized field samples of Ensete ventricosum by rDNA
analysis. Biotechnology letters 26:1867-72.
Birmeta, G., Nybom, H. and Bekele, E. (2002). RAPD analysis of genetic diversity among
clones of the Ethiopian crop plant Ensete ventricosum. Euphyticn 124: 315-25.
Central Statistics Authority (1997). Area production and yield of crops of private holding in
19% /1997 in Meher season, Addis Ababa.
Cheesman, E.E. (1947). Classification of the bananas Kew bulletin 2:97-117
Diro, M. and van Staden, J. (2005). The type of explants plays a determining role in the
m icropropagation of Ensete ventricosum. South African Journal o f Botany 71 (2):154-59
Diro, M. and van Staden, J. (2004) Germination of zygotic embryos and in vitro growth of
seedlings of wild types of Ensete ventricosum. South African Journal o f Botany 70: 635-9.
Diro, M. and van Staden, J. (2004) Propagation of Ensete in vitro: a review. South African
journal o f Botany 70: 497-501.
Diro. M.and van Staden, J. (2003) In vitro regeneration'of Ensete ventricosum from zygotic
em bryos of stored seeds. South African Journal o f Botany 69: 364-9.
 47 Tesfaye Disasa and Mulugeta Diro

Fekadu, D. and Ledin, I. (1997) Weight and chemical composition of the plant parts of Enset
(Ensete ventricosum) and the intake and degradability of Enset by cattle. Livestock
Production Science 49, 249-57.
Mathew, M.M. and Philip, V.J. (2003). Somatic embryogenesis versus zygotic em bryogenesis
in Ensete superbum. Plant Cell, Tissue and Organ Culture 72: 267-75.
Mathew, M.M. and Philip, V.J. (1996). Clonal propagation of Enset (Ensete superbum (Roxb.)
Cheesman) through shoot tip cultures. Plant cell reports 16: 232-34
Mesfin T. Gizachew W. and Kidist 13. (2008). Enset disease management. In Root and Tuber
Crops, pp 209-25 (Gebremedhin W oldegiorgis, Endale Gebre and Berga Lemaga).
Ndungo, V., Eden-Green, S., Blomme, G., Crozier, J.and Smith JJ. (2006). Presence of banana
xanthonionas wilt (Xanthonionas campestris pv. musacearum) in the Democratic Republic
of Congo (DRC). Plant Pathology (Oxford) 55.
Negash A, Tsegaye A, van Treuren R, Visser B (2002) AFLP analysis of Enset clonal diversity
in south and southwestern Ethiopia for conservation. Crop Science 42: 1105-11.
Negash, A., Krens, F., Schaart, J. and Visser, B. (2001) In vitro conservation of Enset under
slow-growth conditions. Plant Cell Tissue and Organ Culture 66: 107-11.
Negash, A., Puite, K., Schaart, J., Visser, B. and Krens, F. (2000) In vitro regeneration and
micro-propagation of Enset from Southwestern Ethiopia. Plant Cell Tissue and Organ
Culture 62: 153-58.
Nur eta, A., Eik, L.O., Tolera, A. and Sundstol, F. (2009). Effect of Enset (Ensete ventricosum)
leal supplementation on feed intake, digestibility, nitrogen utilization and body weight
gain of sheep fed untreated or urea and calcium oxide-treated wheat straw. Livestock
Science 122:134-42
Nurfeta, A., Eik, L.O., Tolera, A. and Sundstol, F. (2008). Chemical composition and in sacco
dry matter degradability of different morphological fractions of 10 Enset (Ensete
ventricosum) varieties. Animal Feed Science and Technology 146: 55-73
Tsegaye, A. (2002). On Indigenous Production, Genetic Diversity and Crop Ecology of Enset
(Ensete ventricosum (Wehv.) cheesman). PhD thesis, Wageningen University and Research
Center, The Netherlands. 198pp.
Welde Michael, G., Bobosha, K., Blomme, G., Addis, T., Mengesha, T. and Mekonnen, S.
(2008). Evaluation of Enset clones against Enset bacterial wilt. African Crop Science
Journal 16(1): 89 - 95.
 Research and Development
Experiences on E n set Agronomy

Mulugeta D iro1and Aclmasu Tsegaye2

1 Southern A g ricu ltu ral R esearch In stitu te, P.O. B ox 6, H a w a ssa , E th io p ia
E -m ail: m d .ch im sa @ iia h o o.co m
2H azvassa U niversity, P.O. B ox 5, H a iv a ssa , E th io p ia

Abstract
This paper reviews enset agronomy that includes management o f enset in the nursery and
field. Enset propagation is usually carried out vegetatively using suckers from
underground stem, corm; while seed propagation is rarely used. In conventional vegetative
propagation, conns from two to three-year-old plants are used at a vegetative phase though
farmer use tioo to six-year-old plants. Apical buds should be removed from mother cornis to
induce growth o f adventitious buds. Planting spaces o f the mother conns in the nursery is
1.5x1.5m betiueen plants and rows. Applying farm yard manure o f 3kg per planting hole
half at planting and the rest after emergence o f suckers) is recommended. On an average,
50-60 suckers emerge from a parent conn in two to three months after corm planting. From
study conducted on enset propagation, halved conns gave earlier differentiation o f buds and
produced more vigorous suckers than whole and quartered cornis. Suckers remain with
parent cornis at least for one year and then they are transplanted from one to fo u r times
until they are transplanted to the main field. In here, spacing o f 3m between rozus and 1.5m
betiueen plants are used under research. A spacing area o f 4 to 4.5m2 per plant and 10-20t
h a 1farm yard manure can be used in the main field. Results from repetitive transplanting
studies showed that enset plants which were transplanted once, twice and thrice flow ered at
about 104, 234 and 260 weeks, respectively. Those plants that were transplanted once gave
more kocho dry matter per hectare and per year than those transplanted tiuice or thrice. The
kocho yield o f enset per unit space and time, in terms o f edible dry weight and energy, was
found to be much higher and follozved by the root and tuber crops, than the yields o f any
other crop cultivated in Ethiopia. Farmers are almost fu lly dependent on organic fertilizers
for enset cultivation. Improving organic matter content o f enset farm should be considered
in terms o f supplying nutrients and also in improving the moisture holding capacity o f the
soil. Analysis o f yield gap betiueen enset grown under optimal and those grown under
suboptinial conditions and the relationship o f livestock and enset cultivation need to be
assessed to address lim iting factors in their priorities.

Introduction
This paper deals with enset agronomy experiences in research and development.
Agronomy is a branch of agriculture dealing with physical and biological factors
including soil management, tillage, crop rotation, breeding, weed control, and
climate-related to crop production. It commonly refers to field crops while
horticulture (garden culture) is concerned with fruits, vegetables, flowers and
 49 Mulugeta Diro and Admasu Tsegaye

ornamental plants. However, it is customary to consider agronomy as an aspect of

horticultural crop management.

A production ecological approach disentangles growth- and yield-defining factors

(genetic potential and solar radiation), growth- and yield-limiting factors (water and
nucrients) and growth- and yield-reducing factors (weeds, pests and diseases) in
agricultural production systems (IAC, 2004). It was further explained that growth-
and yield-limiting and growth- and yield-reducing factors can be influenced by
agronomic practices under field conditions where measures include fertilization,
irrigation and crop protection. Genetic improvement can affect crop performance
under all production conditions. In Africa, where few farmers have access to either
irrigation or affordable chemical inputs, and where growth- and yield-reducing
factors contribute to large pre- and post-harvest losses, farmers' actual yields are
typically a fraction of tire genetic potential, even for the current varieties (De Jager et
al., 2001 cited in IAC, 2004). In this situation, research may be more efficiently
directed at closing the gap by focusing on growth- and yield-limiting and growth-
and yield-reducing factors.

Similarly, to improve enset productivity focusing on growth- and yield-limiting and

grcwth-and yield-reducing factors would be important. In this paper, enset
agronomic practices, as growth- and yield-reducing factors, with regard to planting
m aterial m ultiplication, nursery m anagem ent, transplanting, field establishment and
m anagement is presented and discussed.

Cultivated enset produces seeds but only after a long juvenile period and even if
seeds are produced their germination potential is very low because of seed
dormancy. However, wild species of Ensete and in a few cases cultivated ones are
also propagated from seed. Therefore, the plant is usually multiplied vegetativelv
using whole or split corms. In the vegetative propagation, the whole or split corms
are used after cutting with some portion of pseudostem and removing apical buds
but after uprooting or without uprooting the mother plant (Kefale and Sandford,
1991; Mulugeta and Endale, 1994). Different propagation methods and corm types
we:re compared for vigor and number of suckers (Terefe et al., 1994; Mulugeta et al.,
2002). Influence of age of parent plants, size and depth of planting hole for parent
corm, amount of farmyard manure used for sucker production were investigated at
Areka Agricultural Research Center (Mulugeta et al., 2001; Mulugeta et al., 2008) and
the achievements are presented.

C oim for sucker production is mostly planted at a homestead on fertile soil and
manure is applied. Suckers emerge in 2-3 mondis and remain with the parent corm,
undisturbed, for about a year and then, transplanted to a nursery or main field.
Farmers in Sidama and Gedeo zones, and surrounding areas transplant enset suckers
to the main field once, one year after propagation, and thin to regulate population
density. In areas like Kambata, Hadiya and Guragie zones young enset plants are
 50 Research and Development Experiences on Enset Agronomy

transplanted commonly 2-3 times within nurseries and then to the main field. In
- these areas, where repetitive transplanting in nursery is practiced, the suckers are
transplanted from nursery to nursery usually every year or every two years in some
cases and areas. Spacing varies with stages and ages of the transplants, increasing at
ead‘ successive transplanting stage. Achievements related to influence of repetitive
transplanting and leaf pruning on dry matter and food production of enset and
comparison with yields of other carbohydrate-rich food crops (Admasu and Struik,
2000; Admasu and Struik, 2001) are reviewed and presented.

Corm planting for sucker production in most of the growing areas is carried out
shortly before the rain starts or after the onset of rain, which is mainly from
December to March. Enset field is almost exclusively fertilized with farm yard
manure and use of commercial fertilizers is not common. Soil fertility gradients of
enset farms were investigated and found to be low in the outfields (Asnakech, 1997;
Evasu, 1998; Eyasu, 2000; Tilahun and Mulugeta, 2005). Enset grown in the outfield
* was also found to be drastically decreased compared to enset plants grown in
homestead (Tilahun and Mulugeta, 2005). This paper presents major achievements
on enset's horticultural practices, field management and identifies gaps for future
* research interventions.

Achievements
Agronomic research activities were started in the 1970s by different researchers of
different institutions. Some of the achievements were published in tire Proceedings
of the International Workshop on Enset (Tsedeke et al., 1996). This paper presents
research outputs, which were generated and/ or reported on agronomic practices
after the first workshop on enset.

Propagation
Cultivated enset produces seeds after a long juvenile period and the germination
potential of these seeds is very low because of mechanical seed dormancy imposed
by hard seed coat where an embryo is kept between hard micropylar collars (Fig. 1).
Embryo development is also sometimes poor. Moreover, enset utilizes its stored
carbohydrate during fruiting and eventually dies. It is harvested before or shortly
after flowering thus reducing viable seed production. However, wild species of Enset
and cultivated ones are propagated from seed in a few cases. Therefore, the plant is
usually multiplied vegetatively using whole or split corms and grown as clones.
* Nevertheless, zygotic embryo culture can be used to generate seedlings from
botanical seeds.
 51 Mulugeta Diro and Admasu Tsegaye

H IL U M M IC R O P Y L E

M IC R O P Y L A R
PLUG

M IC R O P Y L A R C O L L A R
N U C E L L A R PAD
EMBRYO

O UTER
IN T E G U M E N T
ENDOSPERM

IN N E R
IN T E G U M E N T

C HALAZAL
M ASS

E P ID E R M A L F R A G M E N T

Fig. I: Longitudinal cross section of the seed of Enset ventricosum, adapted from Mulugeta, 2003

Age o f parent plants for sucker multiplication

Age of parent plants used for sucker propagation is between two and six years but
usually varies from place to place. Under research, influence of age of parent corms
on number and vigor of suckers was investigated at Areka Agricultural Research
Center (1994 and 1995) using one to five year-old corms of Hal'a clone (Mulugeta et
a l., 2001). Their results showed that corms of all age group gave rise to suckers (Fig.
2). However, based on combined analysis of variance, one year-old parent corms
gave significantly less number of suckers than the other treatments. Corms from age
two to five produced closer number of suckers. However, considering long cycle of
propagation in e n s e t, two to three-year-old parent corms can be used for sucker
production. Three to four year-old plants are more commonly used in different enset
growing areas, which is consistent with findings of this report.
 Source: Mulugeta et al, 2001

Fig. 2: Sucker production from mother corm of different ages

Type o f m other corm

F a m e rs use whole or split corms to produce suckers. Parent corm is uprooted, apical
bud removed and planted. In some cases, based on production tradition or climatic
conditions, pseudostem is removed (leaving some portion with corm) without
uprooting the corm but apical bud is removed. Propagation methods and corm types
were compared for sucker production at Areka Agricultural Research Center
(Mulugeta et al., 2002). The result in Table 1 shows that halved corms resulted in
more number of suckers when parent corms were uprooted, apical bud removed and
planted (Method 1), and mother corms not uprooted but apical buds removed at
original site of parent corms (Method 2). Halved corms when used with Method 1 or
Method 2 resulted also in earlier emergence of suckers that led to more vigorous
suckers than whole or quartered corms.

Tab e 1: Number of suckers per plot under different propagation practices

Propagation Corm type

Method Whole Halved Quartered
Method 1 79 cde 110 abc 64 def
Method 2 113 ab 141 a 89 bed
Method 3 94 bed 52 ef 40 f
Means followed by the same letter are not significantly different from each other at a probability of 5%.

* Key
Method 1: replanting mother corms immediately after removal of apical buds;
Method 2: mother corms not uprooted but apical buds removed;
Method 3: mother corms transferred to new holes three months after removal of apical buds. Source Mulugeta et al., 2002
 53 Mulugeta Diro and Adniasu Tsegaye

Planting depth for halved conns

Depth of planting hole for corms varies based on sizes of the parent corms. To
compare depths of planting hole for sucker production, an experiment was
conducted using corms of three-year-old Hal'a clone, which were longitudinally split
into two halves and planted in holes of different depths. The results showed that in
the first year planting at 20 cm depth gave more number of suckers per corm than
the rest but the difference among the 20, 30 and 35 cm depths was not significant
while the 50 and 60 cm deep holes resulted in significantly less num ber of suckers
although the difference between the two was not significant (Table 2). In terms of
growth of suckers, deeper holes (50 and 60 cm) resulted in thinner and shorter plants
compared to 20, 30 and 40 cm depths. Therefore, 20-30 cm deep hole has been
recommended for planting of halved corms.

Table 2: Effect of planting depth of parent corm on multiplication of suckers

Depth of ____________First year___________________________________ Second year_________________________

planting hole No. of No. of
(cm] Suckers per Pseudo-stem suckers per Plant height Pseudo-stem Pseudo-stem
___________________ corm_________height (cm)_______corm__________ [cm]_________ height(cm) diameter (cm)
20 50 a 4.4 a 60 a 71.61 ab 8.77 a 2.47 ab
30 36 ab 4.1 ab 36 ab 85.29 a 14.75 a 3.42 a
40 35 ab 3.2 ab 26 ab 70.96 ab 11.60 a 2.50 ab
50 13b 2.7 ab 8b 33.73 be 5.06 ab 0.82 be
60 7 b 2.1 b 1b 15.08 c 1.44 b 0.19 c
CV(% ) 89.6 49.3 104.7 63.11 88.2 77.5
Means followed by the same letter within each column are not significantly different from each other at 5% level of probability

Transplanting

Single transplanting
Single transplanting and stand regulation in the main field is commonly practiced in
Sidama, Gedio Zones and in other bordering areas. In Sidama, about one year after
corm planting, suckers locally called Funta, are uprooted with a mother corm,
detached and sorted out into two size groups of weak and vigorous suckers. One or
two times transplanting are common (Spring, 1996). The vigorous suckers are
directly transplanted into the main field and planted at narrow spacing (50 to 75 cm
apart) while the weak ones are planted in a nursery in a group of 2-3 suckers per hole
for a year of further growth and development. Farmers give different names to a
plantation in the main field every year as plants age and size increase (Table 3).
Suckers stage (Funta) is transplanted into the main field at an area of about 0.3m2 per
plant: and takes a local name Kora after one vear. Kora is thus two-year-old enset
plantation, which is one year in the nursery plus one year in the main field. When
they are three-year-old, the plants are locally named Ketelo at the same place in the
main field. At this stage, weak plants (Dnkulo) are thinned to improve spacing and
reduce competition among the remaining plants. The remaining plants occupy 0.5-
 54 Research and Development Experiences on Enset Agronomy

lm 2 per plant. Dukulo plants are planted in a narrow spacing with younger plants of
the same size. After one-year growth, Ketelo is named Simancho. Thus, Simancho is a
four-year-old enset plantation. At this stage, plants are also thinned to regulate
stands and the thinned plants are mainly used as mother plants to produce suckers
while corms can also be cooked and consumed. Pseudostem and leaves are used as
feed for cattle. At Simancho stage each plant occupies an area of about lm 2. After a
year, the remaining plants take the name Mallancho, which is a five-year-old enset
plantation. At this stage too, thinning is carried out to optimize plant population
density. Thinned plants are processed into kocho with mature plants or their corms
are consumed. When the plantation attains an age of six, it is called Itancho. Plants in
this stage are mature enough for processing into kocho. Some of the plants may
flower and take the name Kalimo.

Repetitive transplanting
Repetitive transplanting is used in areas such as Kambata, Hadiya, Gurage and
South West Shewa. Here, the experiences of Gurageis considered describing multiple
transplanting in nursery. The mother corm is planted at a spacing of about 1.5 x 1.5
m. Two to three months after mother corm planting, suckers emerge and stay with
the mother corm for about a year. These suckers are called Fonfo. Suckers from Fonfo
stage are sorted out into size group of small and big. The smaller ones are called
Ariye and replanted in a group for further growth and development. The bigger ones
a e transplanted into Simmiia. Thus, Sinimua are suckers that have been transplanted
once and can be planted in rows with two to three suckers in group or in rows of
single plants (Spring et al., 1996). The same authors explained that the spacing
between grouped plants of Sinimua stage is about lm x lm and within the group
0.2m. Planting of this type of sucker is carried out on the onset of the small rainy
season, usually in February and March.

After one year in Sinimua stage, suckers are uprooted and transplanted into next
stage, Tiket. Thus, Tikets are suckers that have been transplanted twice. Spacing used
at this stage is about lm x lm , between rows and plants. At the end of one year,
either the suckers are transplanted to the next stage or the leaves are defoliated
mainly to intercrop maize. The defoliated plants in Tiket stage are called Girciem.
Suckers stay in Tiket stage for one or two years and transplanted for the third time to
Hiha stage. Plants stay in this stage for one or two years and transplanted to main
field, where it is left until maturity. Spacing used at this final stage is about 2.5m x
2m, between rows and plants, respectively. Plants in this stage are called enset and
need about three to four years to be harvested for kocho. The total time required
from propagation to harvesting is about seven to eight years in the mid-altitude
areas and even more than 12 years in cooler high altitude areas.

Single and repetitive transplanting and leaf pruning study was carried out to define
enset production systems (Admasu and Struik, 2000). From the results, it was
observed that enset plants transplanted once, twice and thrice flowered within 104,
 55 Mulugeta Diro and Admasu Tsegaye

234 and 260 weeks after transplanting, respectively. Leaf pruning did not affect the
rate of progress from planting to flowering. At 104 weeks after transplanting, fresh
kocho yield was higher for once transplanted corms than those transplanted twice or
thrice (Table 4). Leaf pruning significantly reduced kocho yield of once transplanted
plants. For transplanting twice and transplanting thrice, leaf pruning did not
significantly affect production rates of fermented kocho. At 130 days after first
transplanting, transplanting thrice resulted in significantly lower production rates of
fermented kocho, both with and widiout leaf pruning.

The maximum fresh weights of kocho after fermentation from enset plants
transplanted once, twice and thrice were 25.9, 54.1 and 37.1kg/plant, respectively
(Admasu and Struik, 2001). When yield was expressed per unit of space and time,
the maximum fresh yields of fermented kocho (70% moisture) were 19, 33 and 26 t
ha- y 1, respectively. Transplanting once, therefore, can be used to have early
harvesting while twice transplanting is recommended for final higher kocho yield.

Tab e 3: Local names and stages of enset in nursery and main field in different growing areas

Some enset growing areas

Sidama
(Thinning and
No. Stages of enset naming of stages) Wolaita Hadiya Gurage SW Shewa
1 Suckers on mother corm Funta Hata Dubbo Fonfo Nedo
2 Once transplanted/2nd stage Kora Beshasha Sima Sam'a Delge
3 Transplanted twice/3rd stage Ketelo Gardua Ogoja/Ero Tiket Mesiye
4 Transplanted thrice/4th stage Simancho - - Hiba Eyiba
5 Thinning stage for Sidama/5th stage Mallancho - - - -
6 Enset plants in the main field before Itancho Halla Belesa Eset Teqa
flowering
7 Flowered enset plants Kalimo Wosa Kalima Borena Shiya kan
___ ■ J
base

Table 4: The effect of number of transplanting and leaf pruning on fresh weight (kg/plant) of fermented kocho harvested and
processed at different weeks after removal from the mother corm

Weeks after removal from the mother corm

(A) pruning (B) 104 130 156 182 208 234 260
Once Not pruned 18.6 25.9 20.3
Pruned 14.6 16.9 18.0
Twice Not pruned 3.3 12.5 13.5 22.5 27.5 54.1 28.5
Pruned 4.2 12.0 16.5 16.6 24.8 53.9 30.5
Thrice Not pruned 1.7 3.3 3.9 9.5 15.5 37.1 31.0
Pruned 1.5 2.6 4.3 10.0 12.5 29.9 26.0
CV% 21 29 14 23 20.1 43.8 29.0
LSD (0.05) 2.3* 5.3* 2.7* NS® NS NS NS

.09, 'indicates significant difference between treatment at P< 0.05.NS = non-significant.

Source:- Admasu and Struik, 2001

Soil fertility management in enset field

Farmers apply alm ost exclusively organic waste to enset field year after year.
Amount of organic waste applied depends up on its availability, which in turn
 56 Research and Development Experiences on Enset Agronomy

depends up on number of animals owned by the household. Farmyard manure

(FYM) of 10-20t ha-1 y-] may lead to better growth and yield of enset (Personal
observation). Preliminary results show that the use of 250kg Urea and 100kg
DAP ha-1 y-1 for two years resulted in better growth and yield of enset (Abav, 2010,
unpublished). Research was conducted on farmers fields of resource rich, group I
(G l) and poor (G2) for four years (2001-2004) at Gununo, Southern Ethiopia (Tilahun
and Mulugeta, 2005). The objectives of the study were to evaluate the effect of soil
fertility gradients on enset growth, identify the major growth limiting nutrients and
identify farmers' decision making criteria in allocating organic resources to various
crops.

Enset transplants were planted in homestead and outfields and application of

fertilizers by farmers to different units over seasons and years was recorded and
enset growth and nutrient content were measured. The results showed that G l
farmers produced two times more organic waste than G2. About 80% of the organic
waste produced was allocated for maintaining soil fertility, while 20% being
allocated as a cooking fuel (Table 5). From organic waste allocated for soil fertility
maintenance, 65% was allocated for tine enset field in the homestead.

There was significantly higher N, P, K and Ca content in the homestead soils than in
the outfields, regardless of farmers' resource endowment. The P (0.5-1 ppm) content
of the outfield was the lowest, less than 25% of the P (2.5-10 ppm) content of the
homestead. Similarly, organic matter in the outfield (2-2.8%) was lower than that of
the homestead (5-6%). Enset plants grown in the outfields experienced about 90%
reduction in height and 50% reduction in pseudostem diameter, regardless of
resource categories. The NPK content of the plant tissues grown in the outfields was
significantly higher, in some cases up to 150%, than those planted in homestead. It
can thus be stated that growth reduction of enset in the outfield was not directly
related to NPK deficiency, but it could have been caused by off-season moisture
stress in the outfields, manifested by low soil organic matter.

Farmers were asked for five major reasons for applying most of the organic waste in
the homestead year after year, The following were their response: 1) there is no
enough organic matter to apply in sufficient amount all over the farm (100%), 2) the
most important crops are in the homestead (mostly enset, coffee and taro), 3) enset is
traditionally grown in the homestead and the use of organic matter follows it, 4)
there is lack of labor to carry the organic matter to the outfields (33%) and 5) soil
erosion will remove the organic matter (33%) if applied in die outfield (Tilahun and
Mulugeta, 2005). Therefore, shortage of farmyard manure and importance of
homestead crops are the major criteria to allocate organic waste to different fields.
__ 5 7______________________ Mulugeta Diro and Admasu Tsegaye

Table 5: Organic resource production by farmers and its distribution to different farm sub-units in Areka, 2002

F armers' Organic manure

category No. of animals (kg/week) Use (%) Distribution in the field (%)
G1 Cattle Sheep FYM Others Soil Home­ Mid- Mid- Out-field
(wet) fertility Fuel stead fieldl field2
A 4 2 101.5 15.0 75.0 25.0 70.0 30.0 0 Inorganic
B 3 0 72.5 12.3 90.0 10.0 50.0 30.0 20.0 Inorganic
C 4 1 116.0 11.5 80.0 20.0 70.0 20.0 10.0 Inorganic
Mean 3.7 1 96.7 12.9 81.7 18.3 63.3 26.7 5.8
SE 0.3 0.6 12.8 1.1 4.4 4.4 6.7 3.3 30.0
G2
A 2 0 72.5 8.0 70.0 30.0 40.0 30.0 0 .
B 0 2 29.0 7.5 100.0 0 85.0 15.0 10.0 .
C 2 0 58.0 6.2 60.0 40.0 65.0 25.0 13.3 .
Wean 1.3 0.7 53.2 7.2 76.7 23.3 63.3 23.3 8.8
SE 0.7 12.8 0.5 12.0 12.0 13.0 4.4
Source: Tilahun and Mulugeta, 2005

With regard to inorganic fertilizer, 250kg Urea and 100kg DAP per year for the first
two years were recommended for Areka and other similar areas to get higher yield
Abay, 2010, unpublished). However, soil fertility is location specific and fertilizer
trials may be needed for different locations along, with its profitability.

Enset productivity compared to other carbohydrate-rich crops

The kocho yield of enset, in terms of weight and energy, was investigated at Areka
Research Center, Southern Ethiopia, and compared with the yields of other main
starch crops grown in the country (Admasu and Struik, 2001). The results showed
that edible dry matter produced was higher from enset (2425 g per m2 per harvest)
than that from root and tuber crops (177 g per m2 per harvest) and cereals (104 g per
m:: per harvest). In terms of per day productivity, enset gave 1.72 g n r2 day-1 edible
dry matter followed by 1.01 g rrr2 d a y 1 from root and tubers while cereals produced
0.69 g n r2 day-1 (Table 6). Therefore, kocho vield of enset per unit space and time, in
tenns of edible dry weight and energy, was much higher than the yields of other
carbohydrate-rich crops and cereals cultivated in Ethiopia.
 58 Research and Development Experiences on Enset Agronomy

Table 6: Average yields and edible dry matter production rates of major crops grown in Ethiopia as compared with enset under
different crop establishment methods

Edible Dry Edible dry Growth Edible dry

portion matter matter (g m period matter (g
Crops Yield" (%) (%) •2 harvest’) (days) m -2 d a y 1)
Enset (Ensete ventricosum )
(Average) 10154 80 30 2425 1339 1.72
Transplanted once 3686 80 32 944 730 1.29
Transplanted twice 14958 80 30 3590 1643 2.19
Transplanted thrice 1817 80 29 2742 1643 1.67
Cereals (Average) 121 100 86 104 140 0.74
Teff (Eragrostis tef) 94 100 89 83 120
Bsrley (Hordeum vulgare) 104 100 87 90 150 0.69
Wheat (Triticum durum) 147 100 87 128 150 0.85
Maize (Zea mays) 159 100 80 127 150 0.85
Sorghum (Sorghum bicolor) 126 100 85 107 150 0.71
Finger millet (Eleusine coracana) 97 100 89 86 120 0.72
Root and tuber crops (Average) 781 85 29 177 193 1.01
Irish potato (Solanum tuberosum) 713 85 20 121 120 1.01
Sweet potato (Ipomoea batatas) 821 85 30 209 150 1.40
Cassava (Manihot esculenta) 688 83 40 228 270 0.85
Taro (Colocasia esculenta) 932 85 30 237 210 1.13
Yam (Dioscorea sp. 750 85 27 172 270 0.64
+g m based on average spacing

G aps

Yieid gap due to better agronomic practices in enset cultivation has not been
quantified and the most limiting agronomic practices have not been identified. Some
farmers produce very high yield per plant and per unit area of land, which can be
exemplified by very high productivity of enset in some farmers' field of Sheka zone,
Southwestern Ethiopia. Understanding extent of the gap and identifying the reasons
behind would be helpful to narrow the gap in order to improve the productivity of
enset. Animal manure is badly needed for enset production, but availability depends
on the number of animals owned by the farmers. Enset on the other hand plays an
important role to feed animals/livestock. Animals' byproducts (meat, milk, egg) are
also important to supplement protein, which is low in enset produce. Therefore, the
relationship between animal rearing and enset cultivation need further investigation
to develop both together, while looking for alternative organic and/or inorganic
sources at the same time. The influence of agronomic practices on growth- and yield-
reducing factors such as diseases and insect pests and identifying sound cropping
system need further study.
 59 Mulugeta Diro and Admasu Tsegaye

Discussion
Propagation
Seed propagation is not common and there is no need to promote it for enset
cultivation as vegetative propagation produce suckers with better growth and
uniformity. However, seed propagation would be important for varietal
improvement and conservation of germplasm. Under such condition in vitro
germination of zygotic embryo can be used to regenerate seedlings from botanical
seeds (Mulugeta and Van Staden, 2004). In vegetative propagation, halved corms
prcduced a large number of final suckers when planted immediately after removal of
apical buds or when left undisturbed for a year after removing apical buds.
However, less number of suckers was obtained from halved and quartered corms
when mother corms were transferred to new planting holes three months after
removal of apical buds, which could be due to the damage caused to suckers while
corm splitting.

Halved corms can be used to propagate enset in most of the enset growing areas for
better number and growth of suckers especially in light of the need to replace enset
plantation devastated by enset bacterial wilt. Nevertheless, farmers often use
different propagation methods considering the prevailing environmental conditions.
These include: whole corm is, for instance, used in areas where dry planting is
exercised, hoping rain will start some time after planting; apical bud removal
without uprooting the parent corm is preferred for low-moisture stressed areas
because root-to-soil relationship stay undisturbed. Experiments on propagation were
carried out at Areka Agricultural Research Center and would not represent different
environmental conditions of farmers.

With regard to age of parent corms, two to three year-old plants can be
recommended for sucker propagation to shorten cycle of propagation but the use of
three to four year-old plants (Sinianho in Sidama, Gardua in Wolaita, Ogoja/Ero in
Hadiya, Hiba in Gurage and Eyiba in SW Shewn, for example) at their vegetative phase
is currently used in enset growing areas and it is consistent with the findings and
thus no need to change it. However, when suckers are needed to establish new
plantation or re-establish plantation devastated by bacterial wilt, it is possible to
multiply suckers from younger (2-3 year-old) parent plants. Depth of hole for halved
corm planting has been recommended to be 20-30 cm but size of the hole depends on
age-size of the parent corms. In the future, influence of planting depth on growth-
and yield-reducing factors such as enset root mealy-bug infestation should be taken
into consideration.

Transplanting
Under field experiment, prolonged time from first transplanting to flowering
increased fresh and dry matter yield of fermented kocho per plant or per unit space
and time of once and twice transplanted plants. This could be because of prolonged
 60 Rcsenrcli and Development Experiences on Enset Agronomy

interception of photosynthetically active radiation by the crop and opportunity for

uptake of N and other nutrients especially under low input conditions. Plants
trarsplanted thrice, showed decreased fresh and dry matter of fermented kocho per
plant or per unit space and time at flowering due to the occurrence of corm rot which
might be aggravated by transplanting shock due to several times transplanting
(Admasu and Struik,2001).

In areas where single transplanting is used, farmers plant suckers first using narrow
spacing and subsequently regulate stands increasing spacing between plants from
stage to stage. Thinned out suckers are used for different purposes. Single
transplanting (transplanting once) was not a similar treatment with farmers practice,
as it was transplanted using wider spacing, and thus results are not comparable. In
areas where repetitive transplanting is used, farmers practice it with defined
objectives. They use nurseries to produce vigorous transplants that withstand
competition when planted as substitute of harvested plants in an already established
plantation and thrive in relatively poor soil as it is planted away from the homestead.
It is therefore difficult to include farmers practice as treatments in evaluating
repetitive transplanting. However, when new plantation is to be established or
plantation damaged by bacterial wilt is to be re-established, transplanting once can
be used to get reasonably good early harvest for kocho yield or transplanting twice
can be used to get better final yield, but only if coupled with better crop management
practice.

Soil fertility management in enset Held

The use of organic fertilizer (FYM) in enset cultivation is the major means for soil
fertility management. It will also continue to be the major means as enset is mainly
used as a food security crop than for a market. Assessment made in different enset
growing areas showed about 4-7 kg plant-1 y-1 fresh dung with beddings is used to
fe tilize enset in the main field but some use more depending on the number of
livestock they owned (UNDP, 1996). It can therefore be indicated that 10-20t h a 1 y-1
is being used. Estimation of amount of FYM used in enset field by farmers is difficult
because FYM is applied continuously based on availability of organic waste.

From organic waste allocated for soil fertility maintenance, 65% is allocated for the
enset field in the homestead and organic matter in the outfield was only about 40% of
the homestead (Tilahun and Mulugeta, 2005). Asnakech (1997) also reported that
fields where enset had been continuously cultivated for several decades had higher
organic matter contents and better nutritional status than non-enset fields. Soil
fertility is being maintained, and even increased, in farm components such as the
enset-garden, darkiia (area near to the homestead planted usually with maize) and
taro fields (Eyasu, 1998). It was also emphasized that erosion does not occur in these
fields, probably because of high organic matter and a more stable soil structure, the
presence of mulch material and greater care provided by the farmers.
 61 Mulugeta Diro and Admasu Tsegaye

The preferential application of organic waste to the homestead was partly because of
the limited manure available due to reduced number of animals and partly as a
result of decline in farm size which resulted in fewer opportunities to produce and
apply cattle manure and crop residues (Tilahun and Mulugeta, 2005). It was also
emphasized by Eyasu (2000) that households with no or few animals, lack access to
manure as it is becoming an increasingly valuable resource.

Productivity o f enset
Com parison of the edible dry matter and energy production rates of enset with the
production rates of main crops grown in Ethiopia is difficult because of the following
reasons: 1) The average edible dry matter and energy yields of main crops
considered were calculated using yield data reported by Ethiopian Central Statistics
Authority and Southern Nations, Nationalities and Peoples' Regional Government
Bureau of Agriculture; 2) The average edible dry matter and energy yields of kocho
were taken after a considerable loss of dry matter due to the complicated traditional
harvesting and fermentation processes; 3) The growth period within a crop may
show variation depending on the cultivar, altitude, cultural practices, etc. For
example, there are some sweet potato and maize cultivars that have shorter growth
period but there are also other cultivars with a much longer growth period (Admasu
and Struik, 2001).

The comparison, however, showed that the edible yield of enset (w'eight and energy)
is much higher than that of cereals or root and tuber crops, which could be due to the
longer growth period of enset. Therefore, the cultivation of enset in densely
populated areas under low input conditions can sustain the population better than
any other crop. Food products from enset are low in proteins and vitamins and
needs supplementing with legumes, vegetables and fruits. Enset has also other
advantages where its perennial canopy intercepts heavy rain and reduces soil
temperature and, thereby, protects the soil against erosion, decreases organic matter
decomposition and reduces leaching of plant nutrients and thus contributes to
sustainable production system.

There are several factors that make enset yield measurement difficult (Hiebsch et al,
1997). This is because enset: has harvested products with high and variable moisture
content; has several harvestable products with different uses; may be intercropped
with other crop species; is a multiple-vear crop; is transplanted from one to four
times at wider spacing; may be harvested at different ages or stages; may be
harvested at any time during the year; and often has stands with a mixture of clones
that have similar growth rates and uses.
 62 Research and Development Experiences on Enset Agronomy

References
Admasu Tsegaye and Struik, P.C. 2000. Influence of repetitive transplanting and leaf pruning
on dry matter and food production of enset (Ensete ventricosum (Welw.) Cheesman).
Field Crops Research 68: 61-74.
Admasu Tsegaye and Struik, P.C. 2001. Enset (Ensete ventricosum (Welw.) Cheesman) kocho
yield under different crop establishment methods as compared to yields of other
carbohydrate-rich food crops. Netherlands Journal o f Agricultural Science 49: 81-94.
Asnakech W oldetensaye. 1997. The Ecology and Production of Ensete ventricosum in Ethiopia.
PhD Thesis, Swedish University of Agricultural Sciences, Uppsala.
Available: (w vvw .cc.colum bia.edu/cu/cup/)
Columbia Electronic Encyclopaedia. 2007. ww w .cc.colu m bia.ed u /cu /cup/, accessed on
17/07/2010. Columbia University Press.
Encarta World English Dictionary. 2009. (hitp:,//encarta.msn.com/, accessed on 17/07/2010.
North American Edition, Bloomsbury Publishing Pic.
Eyasu Elias. 1998. Is Soil Fertility Declining? Perspectives on Environmental Change in
Southern Ethiopia: Managing Africa's Soils. No. 2, Russel Press, Nottingham.
Eyasu Elias. 2000. Soil enrichment and deletion in southern Ethiopia. Pp 65-82. In: Hillhorst,
T. and Muchena, F. (Eds.). Nutrients on the move: Soil fertility dynamics in African
farming systems.
Hiebsch, C., Endale Tabogie, Gizachew W oldemichael, Dougherty, Mulugeta Diro and
Shiferaw Tesfaye. 1997. Difficulties in Measuring enset yield and human carrying
capacity in indigenous systems. In: Ethiopia in Broader Perspective: Papers o f 13th
International Conference o f Ethiopian Studies, Vol I-III,Eds. Fukui, K., E. Kurimoto and M.
Shigeta. Shokado book sellers, Kyoto.
InterAcademy Council (IAC). 2004. Realizing the promise and potential of African
Agriculture: Science and Technology strategies for improving agricultural productivity
and food security in Africa. Inter Academy Council, Amsterdam, the Netherlands.
Kefale Alemu and Sandford, S. 1991. Enset in North Omo Region. Farmers' Research Project
Technical Pamphlet No. 1, FARM Africa, Addis Ababa.
Mulugeta Diro and Endale Tabogie. 1994. Preliminary investigation on enset culture. P. 120-
131. In: Edward IT, Lema Desalegne (Eds.). Horticulture Research and Development in
Ethiopia: Proceedings of the Second National Horticultural W orkshop of Ethiopia 1-3
Dec 1992. IAR, Addis Abeba.
Mulugeta Diro, Endale Tabogie and Atnafua Bekele. 2001. Effect of age of parent enset (Ensete
ventricosum) plants on production of suckers. P. 99-103. In: Bedada Girma, Lemma
Desalegne, Mandefro Negussie, Bulcha Weyessa, Berhane Lakew, Fasil Kibebew and
Geletu Bejiga (Eds). Crop Science Society of Ethiopia (CSSE). Sebil, Vol. 9. Proceedings of
the Ninth Biannual Conference, 22-23 June 1999, Addis Ababa, Ethiopia.
Mulugeta Diro, Mikias Yeshitila, Ermiyas Tesfaye and Endale Tabogie. 2008. Enset
Agronomy. P. 195-207. In: Gebremedhin W oldegiorgis, Endale Gebre and Berga Lemaga
(Eds.). Root and Tuber Crops: The Untapped Resources. Ethiopian Institute of
Agriculture, Addis Ababa, Ethiopia.
Mulugeta Diro, Seifu Gebremariam, Asfaw Zelleke and van Staden, J. 2002. Growth of enset
(Ensete ventricosum) suckers under different horticultural practices. South African Journal
o f Botany 68: 430-433.
 63 Mulugeta Diro and Admasu Tsegaye

Mulugeta Diro and Van Staden J. 2004. Germination of zygotic embryos and in vitro growth
of seedlings of wild types of Ensete ventricosum. South African Journal o f Botany 70:
635-639.
Spring, A. 996. Enset farming system in Southern Region, Ethiopia. Report on a rapid rural
appraisal in Gurage, Hadiya and Sidama Zones. Enset Need Assessment Project Phase I,
Awasa, Ethiopia
Terefe Belehu, Geleta Legese, Mulugeta Diro and Endale Tabogie. 1994. Sw eet Potato and
Enset Propagation Studies. P. 288-300. Edward H, Lema Desalegne (Eds.). Horticulture
Research and Development in Ethiopia. In: Proceedings of the Second National
Horticultural W orkshop of Ethiopia 1-3 Dec 1992. IAR, Addis Abeba.
Tilahun Amede and Mulugeta Diro. 2005. Optimizing soil fertility gradients in the enset
(Ensete ventricosum) systems of the Ethiopian highlands: Trade-offs and local
innovations. In: Bationo et al., Improving human welfare and environmental
conservation by empowering farmers to combat soil fertility degradation.
Tsedeke Abate, Hiebsch, C., Brandt, S.A. and Seifu Geberemariam (Eds.). 1996. Enset-Based
Sustainable Agriculture in Ethiopia. In: Proceedings on the International Workshop on
Enset held in Addis Ababa, Ethiopia, 13-20 December 1993.
UNDP/ECA (United Nations Development Program/Economic Commission for Africa).
1996. Manual on Production and Utilization of Enset (Ensete ventricosum) in South and
Southwestern Ethiopia. Joint Publication of the Southern Nations, Nationalities and
Peoples' Regional Governm ent and the Economic Com mission for Africa. (In Amharic
language with English summary). Vol. 2. Addis Ababa, Ethiopia.
 Research Achievements, Experiences and
Future Direction on Bacterial
Wilt of" E n set
F ik re IlandoroV rai'iku Ilundnina2 and Endale Mailii
]H awassa Agricultural Research Centre, P.O Box 06, Hawassa, Ethiopia
13 EIAR, A mho Agricultural Research Center, P.O.Box 37, Anwo, Ethiopia

Abstract
Enset (Ensete ventricosum ) is one o f the most important staple food crops in
Ethiopia. However, its production has been threatened by a devastating bacterial
disease caused by Xanthonionas campestris pv. musaceariim. This disease was
officially reported in Ethiopia for the first time in the 1960's. It is a vascular
disease resulting in yellowing o f the leaves, wilting and finally collapsing o f the
entire plant. The pathogen is known to sunnve in the soil, plant debris, on
surface o f contaminated knife, alternative host plants and spread by any object
that comes in contact with contaminated plant parts and domestic and wild
animals. Currently, cultural practices and effective sanitary control measures
are one o f the most principal control measures for the disease. This paper
presents a review o f works done on the disease, research gaps and future
research perspectives.

Keywords: Enset, bacterial wilt, research achievements, future research respective

I lit 1 *0 ( 1 net ion

Enset (Ensete ventricosum) is one of the most important staple food crops in Ethiopia.
The plant is drought tolerant and multi-purpose crop. All plant parts are utilized for
different purposes. Hence, kocho, bulla and amicho are used as human food. The by
products from the plant is used to make different household items. Fresh enset
leaves are used as bread and food wrappers, cattle feed, serving plates, and pit liners
to store kocho for fermentation (Brandt et al., 1997). Though, the plant has multi­
purpose uses, it is principally cultivated for its carbohydrate rich food products. It is
used as staple or co-staple food for about 20 % of Ethiopian population particularly
in the South and South-western part of the country (Brandt et al., 1997). Due to its
drought tolerance, enset is regarded as one of the priority crops in Ethiopia, as it
makes major contribution to the food security scheme of the country. Regions where
enset is used as staple or co-staple food are usually less affected by the recurrent
drought periods that occur in Ethiopia (Brandt et al. 1997).
 65 Fikre Handoro et. al.

However, notwithstanding the fact that the plant makes immense contribution to the
food security of the country, there are lots of biotic and abiotic problems facing enset
production. Among the biotic constraints, diseases caused by bacteria, nematodes,
fungi and viruses; mammalian pests such as porcupine, mole rat, and wild pig and
insixrt pests such as mealy bugs have been identified as serious problems, Of all the
b io ic constraints, bacterial wilt disease, which is caused by Xantlioinorias cnmpestris
pv. musacearum(Xcm), (Yirgou and Bradburv, 1968, 1974) has been the most
threatening one. Enset bacterial wilt, is known to cause severe damage, as it attacks
and kills the plants at any growth stages, including full maturity (ready for harvest).
Once the plants are attacked by the disease, especially at late maturity stage, it affects
whole systems, and usually causing a maximum yield loss. A serious outbreak of the
disease was reported by Ashagari (1985) with losses of up to 70%. The results
obtained from recent bacterial wilt disease assessment made in some enset fields of
die Southern Nations, Nationalities and Peoples Regional State (SNNPR), showed
losses of up to 100% under severe damage (Shank and Chernet, 1996, Anonymous,
2008). Many researchers (Anita et n/,,1996; Shiferaw, 1998; Tsegaye et al., 1998;
Million et al., 2003; and Endale et al., 2003) reported that both the area and
productivity of enset is declining continuously due to this disease. Therefore, this
paper is aimed at reviewing research achievements and experiences gathered
through the years on this deadly disease of enset and provide an indicative
approach to future research directions.

The Pathogen: E n s e t Bacterial M ilt

In Ethiopia, a disease symptoms similar to bacterial infection of banana, known as
Moko disease or bacterial wilt was reported on enset for the first time in the 1930's
(Florence, 1939). However, a bacterium causing a wilt disease of enset was officially
reported in Ethiopia for the first time in the 1960's (Yirgou and Bradbury/ 1968) and
named Xanthomonas musacearum. The same bacterium was later confirmed as causing
a similar disease on cultivated banana and other Musa spp. (Yirgou & Bradbury,
1974) and was subsequently reclassified as X. campestris pv. Musacearum (Young et
al., 1978).

Enset bacterial wilt is a vascular disease (Thwaites et al., 2000) that results in
yellqwing of the leaves, wilting and finally collapsing of the entire leaves/ plant
(Figs. 1A and B).
 66 Research Achievements, Experiences and Future Direction on Bacterial Wilt o f Enset

Fig.1 A Enset plant wilt caused by enset bacterial wilt disease (Xcm)

Fig.IB. Collapse of enset plant caused by enset bacterial wild disease (Xcm)

Symptoms: The symptoms, which include yellowing of leaves, wilting (often

associated with loss of turgor and collapse of the petiole) and excretion of a
yellowish bacterial ooze from cut tissues is characteristic of both banana and enset
bacterial wilt (Thwaites et al., 2000; Tushemereirwe et al., 2003, 2004). A cream or
yellow-colored ooze, typical of many bacterial infections, exudes within a few
minutes after cutting the tissue and abundant quantities may be produced over a
period of several hours.
 67 Fikre Handoro et. al.

Taxonomy and Nomenclature of the Pathogen

Biochemical and cluster analysis studies conducted in Ethiopia on enset and banana
X an isolates indicated that there are variations among isolates (Quimio, 1992;
Gizachew, 2000; Kidist, 2003). Phenotypic characterization and PCR-based studies
conducted on both enset and banana Xcm isolates from Ethiopia also revealed
possibility of variation among the isolates (Tsehay, 2009). However, studies done on
Ugandan isolates of Xcin using Random Amplification of Polymorphic DNA (RAPD)
analysis did not reveal any significant differences in clustering, with exception of a
single isolate that had unique fingerprints. All the isolates compared also showed no
significant difference with regard to incubation period for appearance of symptoms
and the severity of symptoms in pathogenicity test (Odipio et a l, 2009). Molecular
studies conducted by fatty acid methyl esters, genomic fingerprinting using rep-PCR
and partial nucleotide sequencing of the gyrase B gene on X an, indicated that
strains of X an have close homology to strains of Xanthonionas vasicola. Therefore, the
result of this study suggests for reclassification of X. campestris pv. musacearum as X.
vasicola pv. musacearutn (Aritua et al., 2007). But this has not been formally approved
as a new combination of names (Karamura et a l, 2008). However, some reports have
already reported Xanthonionas vasicola pv rnusacearuvi (Xvm) as causative agent of
Bacterial wilt disease of enset and banana in South and South-western part of
Ethiopia (Shimelash et a l, 2008) and in Eastern Africa ( Biruma et al., 2007). Karamura
et al. (2008) have also listed enset as a host plant of X. vasicola pv. musacearum.

Sources of Inoculums, Mode of Infection and Transmission of Xcm

Concerning sources of inoculum, mode of infection and transmission of Xcm
especially of enset in Ethiopia is not yet well studied to the level needed as compared
to the seriousness of the disease. However, a few studies have generally revealed
some facts on sources of inoculum, mode of infection, transmission and survival of
Xcm.

Studies done to understand survival of Xcm in the soil indicated that the pathogen
can survive in the soil for some specified time (Quimio and Mesfin, 1996; Mwebaze
et al.l 2006; Welde-Michael et al., 2008). The result of a preliminary work showed that
the bacteria can survive in chopped plant debris in the soil for over six months. It
was also indicated that, plant residues, contaminated soils and water (overflow of
water from infested to uninfested fields), infected mats and traded products
including fruits, leaves and planting materials (suckers and corms) are thought to be
the major sources of inoculum of Xcm (Mikias et a l, 2010; Eden-Green, 2004; Million
et al.. 2003). The findings of other studies also revealed that Xcm survived on the
surface of contaminated knife for up to 3 to 4 days (Ashagari, 1985) and found to
transmit the pathogen from infected to healthy plants (Welde-Michael et al., 2008).
Generally, it was also reported that the pathogen is easily spread by any object that
comes in contact with contaminated plant parts (Brandt et al., 1997). Xanthonionas
campzstris pv musacearum was observed to survive in pruned leaf petioles and leaf
sheaths for at least 3 months (Welde-Michael et al., 2008). Plant families of
 68 Research Achievements, Experiences and Future Direction on Bacterial Wilt o f Enset

Cannaceae (Cana family), Costaceae (Costus family), Heliconiaceae (Heliconia

family), Marantaceae (Prayer-plant family), Strelitziaceae (Birds of Paradise Flower
family) and Zingiberaceae (Ginger family) are considered as host plants (Karamura
et n l, 2008) and could be possible source of inoculum for the pathogen.

Works done on dissemination mechanism of the pathogen have indicated to be

mediated by several factors. Study on insect transmission of Xcm on enset suggested
the possibility of transmission of the pathogen by insect vectors (Wondimagegne,
1981). Similarly farmers also claim that the pathogen is transmitted by insect vector
(Tariku Hunduma, personal communication). However, this fact is not yet fully
clarified. Insect mediated transmission of Xcm on banana identified potential insect
vectors and the importance of these vectors in ABB bananas through the male buds
has been underlined by the fact that early de-budding readily arrests disease spread
(Karamura et n l, 2008).

Contaminated tools used in routine field activities (pruning, de-leafing, digging,

weeding and harvesting) are key mechanism by which the disease spread locally
within the same farm and between farms (Mikias et n l, 2010; Karamura, 2006; Million
et nl, 2003).

Vertebrate pests such as mole rates and porcupine are claimed to transmit the
disease (Mikias et nl., 2010; Brandt et nl., 1997; Shank, 1996). Domestic animals such as
cattle are also expected to spread the disease (Mikias et nl., 2010). Though evidences
from banana has demonstrated the potential of bats and birds to transmit the disease
(Karamura et n l, 2008), there is no evidence in the case of enset Xcm. These organisms
need to be taken into consideration and cannot be undermined as vectors of the
disease. Study conducted on banana on the entry of the pathogen suggested to be
facilitated by mechanical injuries or injuries caused by soil-borne organisms such as
nematodes and insects. However, this needs further investigation in the case of enset.

Management of Enset Bacterial Wilt Disease

Lack of adequate information on the biology and epidemiology of the pathogen and
the perennial nature of the plant have affected the development of effective control
measures as yet. Consequently, management options have focused on methods that
reduce the initial inoculum and subsequent spread of the pathogen.
Cultural practices and sanitary control measures are one of the most principal
control measures for enset bacterial wilt disease. Regular inspection of fields and
effective eradication of inoculum would contribute immensely to lower the level of
infection. Practices such as disinfecting farming and processing tools by washing
them in Sodium hypochlorite (Berekina) and with fire (Karamura et a l, 2008) have
been known to reduce disease transmission from infected plant to healthy one.
According to these authors, the solution can be prepared by mixing one cup of
berekina in five cups of water. As diseased enset plants and debris are the potential
sources of primary inoculum, uprooting and burying the infected enset plants
 69 Fikrc Handoro et. at.

(eradication and disposal of diseased enset plants/debris from the fields) has also
resulted good management options. Karamura et a l (2005) identified early detection
and destruction of the diseaseci plants as a key step in preventing disease spread.
Cultural and sanitary measures have been practiced bv farmers in different enset
growing zones of the country. According to Million et al. (2003) about 71% of the
farmers reported that careful application of sanitary control measures helps to
control enset bacterial wilt disease. They also suggested that cultural practices such as
deep tillage, exposing the soil during dry season prior to planting, proper spacing,
spot rotation of infested sites to reduce disease spread. Avoiding overflow of water
from infested to uninfested fields, controlling porcupine, mole rats and other
domestic animals that may transmit the pathogen within the fields and surrounding
areas (Mikias et al., 2010; Brandt et al., 1997; Shank, 1996) also help to minimize
pathogen spread.

Use of clean planting materials (suckers/ transplants, corms) and strict controls on the
movement of planting materials from one area to other (developing local quarantine
system) have also been well recommended for the management of enset bacterial
disease (Brandt et a l, 1997).

Otiier option for bacterial wilt management is the use of resistant or tolerant host
plans. Farmers in different enset growing areas of the country cultivate different enset
clones and claim that the clones show varying degree enset of susceptibility to
bacterial wilt. In spite of this fact, owing to the perennial nature of plant,
screening/breeding activities done on this plant in search of tolerant/resistant clone
against the pathogen is scanty as compared to work done on cereal crops. However,
Hawassa and Areka Agricultural Research Centers have been making concerted
efforts in this regards. Areka Agricultural Research Centre has collected over 600
different enset vernaculars from different enset growing areas and maintained them at
the Centre (Anonymous, 2009). Screening works done both at the Centre by artificial
inoculation and in different farmers' fields under natural disease conditions showed
that "M aziya" enset clone has better resistance/tolerance than "A rkia", which is
highly susceptible check (Fikre and Gizachew, 2007). Developing tolerant/resistant
em et clone needs further investigation of the genetic diversity of the plant in the
country and this needs particular attention from the government side in terms of the
human resource development and financial support.

No chemical trial has been conducted so far and there is no information regarding
the use of chemical as an option for the control of enset bacterial wilt. Though
biological control of bacterial diseases using microbial antagonists are known to be
effective (Priou et a l , 2006), this option has not yet been hied so far in the
management of enset Xcm.

In addition to what has been tried to manage/control enset bacterial wilt disease,
continuous and uninterrupted public awareness creation program about the disease
 70 Research Achievements, Experiences and Future Direction on Bacterial Will o f Enset

is mandatory. According to Karamura et nl., (2008), 'T h e battle against Xcm wilt
needs to be multi-faceted, requiring mobilization of all available human as well as
financial resources. Everyone must be aware and be committed to make sound
contribution towards the control of the disease. All stakeholders need to be given
clea -/concise and appropriate messages, stating what needs to be done by whom,
hov\, where and when so that they in turn could play their roles effectively". It is
therefore, unequivocal that concerted efforts be exerted in order to adjust the research
strategy with the provision of adequate human and financial resources to effectively
address the pathogen and with clear messages to the scientific and local communities
for their unduly participation in such noble tasks.

Research Gaps and Recommendation

Since the time of first identification and official report of the pathogen (Xcm) (Yirgou
and Bradbury, 1968), various research attempts have been made on some aspects of
the disease especially on cultural and sanitary control measures. However, there are
still some major research gaps that need to be addressed. The role of insect vectors,
vertebrate pests (mole rates and porcupine), bats and domestic animals such as cattle
in the transmission of the disease is not yet adequately investigated. The survival
nature of the pathogen during fermentation of enset mass into primary food products
and the role of the latter products in the transmission of the pathogen is not yet well
determined. One of the major challenges is the lack of adequate knowledge of the
epidemiology of the pathogen, which had been done only in such fragmented way. It
needs to be conducted in a uniform hierarchical manner as its understanding is
interlinked with the development of disease management options. Lack of sufficient
knowledge on variability and/or similarity of the pathogen from all over the country
are also another challenge. The taxonomy of the pathogen has not been resolved, it is
still under discussion. The clear understanding of the taxonomy and its variability
and or similarity would help in screening the local enset clones for developing
resistant/tolerant clones as one management option. Although it has been claimed
that huge enset vernaculars are known in the country, no sufficient work has been
corducted on its genetic diversity. Characterization and classification of the existing
local enset clones for their genetic diversity in relation to bacterial wilt disease and
productivity should be supplemented with modern agricultural biotechnology. Lack
of clean planting material has been considered as main source of disease spread over
localities and hence this needs further attention in the future. Modern tissue culture
tec uniques need to replace the cultural method of sucker propagation and provide
farmers with disease free planting materials that would contribute to the battle
against this deadly disease.

Owing to lack of effective control measures, farmers currently call this disease "enset
AIDS" just similar to human HIV/AIDS which is not yet under effective medical
treatment. Consequently, the management of this disease need to focus on
prevention of disease transmission and spread. Eden-Green (2004) emphasized the
use of control measures that reduce or prevent further spread of the disease to new
 71 Fikrc Hnndoro et. a I.

areas or areas that are not yet infected. Eden-Green indicated that the greater degree
of control of this disease depends on prompt removal of sources of inoculum and/or
retiucing or eliminating opportunities for further spread. It is therefore important
that the awareness of every stakeholder be raised at every level about this disease
through different but effective campaigns. Everyone must know about the disease.
Especial attention should be given to areas that are not yet affected by the disease.
The campaigns can be used to familiarize stakeholders with the disease symptoms
and undertake effective phytosanitory practice as the only available control measure.
Brandt et al. (1997) recommended for the use of healthy, disease-free suckers for
planting material and destruction or complete removal of diseased plants as the best
options for the control of enset bacterial wilt. To this effect, the establishment of
national and local quarantine system is verv essential. Quarantine measures can be
re-enforced with bye-laws and other policy interventions, which in turn require
appropriate institutional frameworks such as taskforce formation/ implementation
and both human and financial resource commitments to make the system work
(Karamura, 2008).

References
A rita, S. H. C., Endale, T. and Gizachew, W. 1996. Enset need assessment project.
Phase I Report. Awassa Ethiopia
Anonymous, 2008. Awassa Agricultural Research Centre, Progress Report, Plant
protection Research Division. Awassa, Ethiopia.
Anonymous, 2009. Areka Agricultural Research Centre, Progress report, crop
research case team.
Ar itua, V., Parkinson, N., Thwaites, R., Heeney, J.V., Jones, D.R. Tushemereirwe, W.,
Crozier, J., Reeder, R., Stead, D.E., and Smith, J. 2007. Characterization of the
Xanthonionas sp. causing wilt of enset and banana and its proposed
reclassification as a strain of X. vasicola. Plant pathology(2007).
Ashagari, D. 1985. Studies on the bacterial wilt of Enset (Enset ventricosum) and
prospects for its control. Ethiopian journal o f Agricultural Science 7(1):1-14.
Biruma, M., Pillay, M., Tripathi, L., Blomme, G., Abele, S., NTwangi, M.,
Bandyopadhyay, R., Muchunguzi, P., Kassim, S., Nyine, M., Turyagenda, L. &
Eden-Green, S. 2007. Banana Xanthonionas wilt: a review of the disease,
management strategies and future research directions. African journal o f
Biotechnology 6 (8), 953-962.
Brandt, S. A. Spring A., Hiebsch, C., McCabe, J. T., Endale T., Mulugeta, D.
Gizachew, W., Gebre, Y., Shigeta, M., and Shiferaw, T. 1997. The "Tree A gainst'
Hunger". Enset-Based Agricultural Systems in Ethiopia. American Association
for the Advancement of Science with Awassa Agricultural Research Center,
Kyoto University Center for African Area Studies and University of Florida,
Washington, DC, USA, 66 pp.
 72 Research Achievements, Experiences ami Future Direction on Bacterial Wilt o f Enset

Karamura, E.B., Turyagyenda, F.L., Tinzaara, W., Blomme, G., Molina, A. and
Markham, R. 2008. Xanthomonas wilt of banana in east and central Africa.
Diagnostic management guide. Bioversity international, Uganda
Eden-Green, S. 2004. How can the advance of banana Xanthomonas wilt be halted?
Info Musa 13(2): 38-41.
Endale, T., Million, T, and Gizachew, W. 2003. Enset-Based Farming System of Kucha
Woreda, Gamo-Gofa zone. Research Report No 52 EARO. SNNPRS, SARI, PP.
48.
Fikre, H. and Gizachew, W. 2007. Evaluation of enset (Ensete ventricosum) clone
Maziya against bacterial wilt caused by Xanthomonas campestris Pv. musacearum.
African crop science proceedings Vol. pp. 887-890 El. Mini Egypt
Florence, R. 1939. Istituto agronomico per L'Africa, April, 1939. About and ensete rot.
Gizachew, W. 2000. Variation in isolates of enset pathogen (Xanthomonas campestris
Pv. musacearum) and reaction on enset clones (Ensete ventricosum (Wele.)
Cheesman). Msc. Thesis. Alemaya University, Ethiopia.
Karamura, E. B., Kayobyo, G., Blomme, G, Benin, S., Eden-Green, S. J., Markham,
R.2006. Impacts of BXW epidemic on the livelihoods of rural communities in
Uganda. Proceedings of the 4th International Bacterial Wilt Symposium, 17-20
July 2006, Central Science Laboratory, York, UK. p. 57.
Karamura, E. B., Osiru, M., Blomme, G. 2005. Containing banana Xanthomonas wilt.
InfoMusa 14(1): 45-46.
Kidist, B. 2003. Characterization of Xanthomonas campestris pv. musacearum isolates:
causal agent of enset bacterial wilt disease. MSc Thesis. Addis Ababa
University, Ethiopia.
Mikias, Y., Fikre, H., Mulugeta, D., Zerihun, Y., Sadik, M. 2010. f'/'WrV
Pp. 24
Million, T., Kidist, B., Mulugeta, D., and Gizachew, W. 2003. Enset Bacterial wilt
Control in Gurage zone. Research Report No 53 EARO. SNNPRS, SARI. PP. 23.
Mwebaze, J.M ., Tushime, G., Tushemereirwe, W.K. and Kubiriwa, J. 2006. The
survival of Xanthomonas campestris pv. musacearum in soil and plant debris.
African Crop Science Journal. Vol. 14. No. 2. pp. 121-127.
Priou, S, Marquez, M, Gutarra, L. 2006. Biological control of bacterial wilt of potato
(Ralstonia solanacearum) using an ntagonistic endophyte strain of Pseudomonas
putida. Proceedings of the 4th International Bacterial wilt symposium, 17-20 July
2006, Central Science Laboratory, York UK, p.47.
Quimio, A.J. 1992. Annual report of the Enset Team Support Project of PADEP III;
Awass Research Centre. IAR
Quimio, A.J. and Mesfin, T. 1996. Diseases of enset. hi: Enset-based sustainable
agriculture in Ethiopia Pp. 188-203.
Shank, R. 1996. Mole rats: Another Possible Mode of Spreading Enset Bacterial Wilt.
Emergencies Unit for Ethiopia (UNDP-EUE).
Shiferaw, T. 1998. The role of enset (Ensete ventricosum) in sustainable
intensification of agriculture. In sustainable intensification of agriculture in
 73 Fikre Handoro et. al.

Ethiopia. Proceeding of Second Conference of Agricultural Economics of

Ethiopia, 3-4 October 1996. AESE. Addis Ababa, Ethiopia
Shm elash, D., Alemu, T. Addis, T., Turyagyenda, F.L., and Blomme, G. 2008.
Banana Xanthonionas wilt in Ethiopia: Occurrrence and insect vector
transmission. African Crop Science journal, Vol. 16, No. 1, pp. 75 - 87
Thwaites, R, Eden-Green, S. J, Black, R. 2000. Diseases caused by bacteria. In: Jones
DR (Eds). Diseases of banana, Abaca and Enset.
Tsegaye, B., Tesfaye, G., Kidist, B., Shiferaw, M , and Bola, F. 1998. Study on Coffee
and Enset production constraints in Kucha Woreda. Kucha KHC Integrated
Development Project.
Tsehay, M. 2009. Studies on Genetic and Phenotvpic Variation between Enset and
Banana Isolates of Bacterial Wilt pathogen (Xanthonionas campestris pv.
musacearum) in Ethhiopia. MSc Thesis. Ambo University, Ambo, Ethiopia.
Tushemereiwe , W-K., Kangire, A. Smith, J. Nakyanzi, M., Karyeija, R., Kataama, D.
and Musiitwa, C. 2001. An outbreak of banana bacterial wilt in Mukono
district: A new and devastating disease, NARO/ KARI.
Welde-Michael, G., Kidist, B., Blomme, G., Addis, T., Mengesha, T. and Mekonnen,
S. 2008. Evaluation of enset clones against bacterial wilt. African crop science
Journal, Vol. 16, No. 1, pp,89-95 P. Uganda
Wondimagegn, E. 1981. The role of Poecilocarda nigrinervis, Pentalonia nigeronervosa
and Planococus ficus in the Transmission of Ensete wilt Pathogen, Wolaita,
Ethiopia, MSc. thesis. Addis Ababa University.
Yirgou, D. and Bradbury, J. F. 1968. Bacterial wilt of Enset (Ensete ventricosa) incited
by Xanthonionas musacearum. Phytopathology 58:111-112.
Yirgu, D., and Bradbury, J.F. 1974. A note on wilt of banana caused by enset bacterial
wilt organism. Xanthonionas campestris. East Africa Agricultural and Forestry
Journal. 40
Young, J. M., Dye, D. W., Bradbury, J. F., Panagopoulos, C.G., Robbs, C. F., 1978. A
proposed nomenclature and classification for plant pathogenic bacteria. New
1 Zealand Journal o f Agricultural Research 21, 153-77.
 Experience on Bacterial Wilt Management in
Eastern Africa: The Case of Banana

Eshetu Ahmed1and Mohammed Yesuf*

Ethiopian Institute o f Agricultural Research, P.O. Box 2003, Addis Ababa, Ethiopia
2Melkassa Agricultural Research Center, P.O. Box 436, Nazareth, Ethiopia

Abstract
Banana and plantains are among the most important crops in the world both as staple
food and income generation crops. Enset, which is a banana like crop, is widely grown as a
food source especially to the South and Southwest part o f Ethiopia. A number o f disease
problems are recorded on banana, plantains and enset. Among others, Xanthomonas
cam pastris pv. musacearum is one o f the devastating bacterial pathogen causing
banana/enset wilt and lead to complete yield loss. In Uganda, research information that
include incidence, distribution, economic importance, development o f selective media,
survival o f the pathogen in the soil and banana debris, hosts range and management
practices to be follow ed were generated which play a significant role in tackling this
important disease o f banana and plantain. The bacterial wilt pathogen o f banana is also
the causal agent o f enset wilt. The purpose o f this review is therefore, to compile
experiences acquired/gained in banana bacterial wilt management in some other East
African countries. Thus, the findings would help to extrapolate and adopt the
management o f enset bacterial wilt under Ethiopian condition. Therefore, this paper
discusses the research and development experiences o f East African countries, Ugandan
experience in particular; on banana bacterial wilt management including available
research recommendations, technology dissemination, task force formation and
community mobilization through effective awareness creation and deployment o f
participatory development approach.

Introduction
Enset (Ensete ventricosum) is a relative of banana since both belongs to the family
Musacea. However, enset is cultivated only in Ethiopia. It is a staple food crop for
over 20 million people. Currently, it is dominantly grown in the southern and south
western part of the country. It is expanding and growing in the northern part of the
country. People in urban South of the country are well accustomed to the different
types of food products of enset such as kotcho and bulla, which are made from enset
through fermentation process are the two main forms of processed products used.

Banana and plantains (Musa spp.) (here after called bananas) is the fourth most
important global food commodity in tropical and sub-tropical zones of the world.
They are cultivated in over 100 countries covering about 10 million hectares, with
 75 Eshetu Ahmed and M ohammed Ycsuf

annual production of about 88 million tones of products (Sharrock and Frison, 1999).
According to these authors, bananas and plantains provide more than 25% of food
energy requirements for about 70 million people in Africa of w hom 20 million are
frcm East Africa alone.

In Uganda for example, banana supports over 70% of the population for food and
income generation. Uganda has the world's highest per capita banana consumption
at 220-460 kg per year (1.6 kg per person per day). It is second to India in production
with 9.8 million metric tons per year (FAO, 1998).

One of the strongest characteristics of the crop is its unprecedented tolerance to

drought. This makes e n set as the most suitable commodity where currently the world
is zhallenged with serious climate change and water deficit.

On the other hand, the most important production limiting factor of the crops is the
banana bacterial wilt (BBW) caused by Xanthonionas campastris pv. niusaceanini. Full
classical taxonomic description of the bacterium was given by Bradbury (1986). The
objective of this review is to compile the experience of other countries on the
management of Xanthonionas wilt, so that the findings or experiences of others could
be adopted to the enset production system in Ethiopia.

Research Experience

The disease was first reported on enset in the 1960s in Ethiopia (Yirgou and
Bradbury, 1968). It was later reported on bananas in the Keffa, Shoa, Sidamo,
Harerge and Gamogoffa regions of the country (Yirgou and Bradbury, 1974).

Lately in the year 2000, the disease was reported in Uganda affecting plantations of
banana fund plantains. Currently, the disease is reported from 36 districts (out of 58)
in Uganda. New outbreaks have also been reported in Rwanda (Reeder et al., 2007),
in the Democratic Republic of Congo (Ndungo et al., 2005), Tanzania (Mgenzi et al.,
2006a, 2006b), Kenya and Burundi (Smith et al., 2008).

Various survey reports conducted since 2001 in Uganda indicated that in all the
affected districts, all banana types [(Pisang awak ABB), Blugoe, Highland bananas
(AAA-EA), Gross Michael and Neypoovan] were affected. Consequently, in one year
70 - 80% incidence was reported for many affected plantations.

Since the occurrence of bacterial wilt in Uganda in 2000/1 intensive research

undertakings were considered and more than 15 research articles were published
about the various aspects of the disease in such short period of time.
 76 Experience on Bacterial Wilt Management in Eastern Africa

Basie research
The classical taxonomic description (Brandbury, 1986), complementary fatty acid and
molecular information of the pathogen were generated (Tushemereirwe et nl., 2004
Aritua et nl., 2008). Diagnostic molecular tools such as PCR primers developed for
iden ification and detection of the pathogen (Aritua et nl, 2007a & b).

The results of the studies conducted in Ethiopia indicate the occurrence of low levels
of genetic variation between strains of X. campestris pv. musncearum, collected from
the irst outbreaks and recent outbreaks were also observed in other countries
(Aritua ct nl., 2007). They also indicated that DNA sequencing studies have also
shown that the species are closely related to X. vasicola.

New semi-selective media was developed for artificial multiplication of the

pathogen. The semi-selective media designated as CCA contains lg yeast extract, lg
glucose, lg peptone, lg NH4C1, lg M g S 04.7H 20/ 3g K 2H P04, lg beef extract, lOg
peptone, 14g agar, 40mg cephalexin lOmg 5-fluorouracil and 120mg cycloheximeide).
With this media it was possible to recover 82% of the target bacteria from soil
compared to the standard media (yeast peptone glucose agar /YPGA/ containing 5
g yeast, 5g peptone, 4g glucose, 12g agar per liter), which gave only 42% (Mwebaze
et nl. 2006a).

Investigations were made to determine the survival of Xantliomonns campastris pv.

lnusacearimi in soil and infected banana debris. The pathogen persisted longer (two
times) in high moisture conditions (28%) than in low moisture soil conditions (14%).
The pathogen populations declined rapidly in non-sterile soil (only 15 days) than in
sterile soil (90 days). The pathogen could not persist in sterile moist soil for a period
longer than 90 days and 30 days in dry soils and its survival period in the field was
longer in the soil than in banana debris (Mwebaze et al., 2006b). However, the
population of the pathogen in banana debris in the field declined rapidly while the
decline was gradual in the laboratory. Viable cells of the pathogen could not be
recovered from plant debris incorporated in soil or on soil surface of banana debris
after 21 days. Therefore, according to the same authors, the disease causing
bacterium, X. campastris pv. musacearum survived in the absence of a host for only 20-
90 days depending on the environment.

Stud es were made to determine the survival of bacterial cells on metallic tools
through smeared onto blades and survival was monitored for over 72 hours by
checking growth of colonies on bacterial culture media. Xanthomonas remained viable
and infective for the entire 72h period. Therefore, the results proved that bacteria
rema ined viable for extended periods of time and provided evidence to support the
recommendation given to farmers to disinfect their tools regularly to avoid disease
transmission.
 77 Eshetu Ahmed and M ohammed Yesuf

Moreover, the reaction of 42 local banana genotypes to X. campastris pv. musacearum

was tested and tiiere was no significant difference among test materials. Among the
germplasms, 5 were natural diploids, 15 natural triploids, 10 hybrid diploids and 12
hybrid tetraploids. They were tested for their reaction to X. campastris pv. musacearum
at the age of three months in Uganda. All were found susceptible except M.
balbisiana (natural wild type), which showed some resistance reaction (Sskiwoko et
a l, 2006a). Although, the disease had been observed in Eastern Highland bananas
and exotic dessert and beer bananas, the clones varied in the incidence of the disease
(Tushemereirwe et a l , 2003). Disease incidence was highest on Kivuvu (73.5%),
followed by Kayinja (30.4%) tiien cooking bananas (11.7%) and the least (2.8%) was
in Sukari Ndizi (Tushemereirwe et a l, 2003).

The banana parts such as fresh leaves, dry leaves, fresh pseudo-stem, sheath, fruit
peelings and corms picked from diseased plants were used to inoculate plantlets.
Inoculations with fresh banana parts, in particular, were able to cause disease
incidence between 5 - 2 0 % to healthy potted banana plantlets only after wounding
the test plant roots. Bacteria were more abundant in fresh leaves and least in dry
leaves (Table 1). The number of bacteria cells isolated from banana corms, fresh
leaves, fruit peeling and pseudo-stem sheaths were similar. More than 75% of
bacterial isolates suspected to be the banana wilt bacterium that were isolated from
the- d ifferen t banana parts were p a th o g en ic to banana plantlets (Table 1). Therefore,
practices that involve movement of fresh banana parts should be discouraged to
avoid plant materials that could carry viable and pathogenic bacteria (Tumushable et
a l 2006).

Using dried leaves did not cause any wilt symptom, though isolation of the
pathogenic bacterium was possible. However, none of these infected dried parts
affected any plant whose roots had not been wounded indicating that wounding
seems to be a prerequisite for infection in this case (Tumushable et a l 2006).

Table 1. Mean number of isolated bacterial cells of banana, percentage of pathogenic isolates and incidence of pathogen on
inoculated plantlets.

Logio (bacteria
Pathogenic isolates
cells/g plant
(percentage)
Plant parts/ treatments* tissue +5) Incidence (%)
Corms 9.65±0.89ab 79 (n=8) 6.7 (n=15)
Dry leaves 7.59±0.46b 82 (n=20) 20 (n=15)
Fresh leaves 10.35±0.95a 100 (n=9) 0 (n=15)
Fresh pseudo-stem sheath 9.73±1.02ab 93 (n=7) 13.3 (n=15)
Fruit peelings___________ 9.62±0.89ab 90 (n=9) 13.3 (n=15)
Source: Tumushabe et al., 2006; * part of plants obtained from diseased plant

Means with the same letter within the column are not significantly different at 0.05
level of probability by pair-wise t-test of test of least square means.
 78 E x p erien ce on B a c te r ia l W ilt M an ag em en t in E astern A frica

Host range test was done using three categories of plants, i.e. l.W ild banana relatives
(Musa ornate and M. zebrine), 2. Ornamentals/weeds/ banana intercrops (Canna
indica, Heliconia rnetallica, Ageratum conyzoides, Conmielina sp., Bidens pilosa, Ananas
coniosus, Zingber oficinale, Ipotnia batatus, Lycopersicon esculentum, Datura stramonium,
Capsicum spp., Galinsoga peiviflora, Elettaria cardamonum) and 3. Species that are hosts
to Xanthomonas campastris group (Manihot esculentum, Penisetum piirpureum,
Saccharum officinale, Amaranthus dubious).

Typical symptoms were visible on plants species Musa ornate, M. zebrina and Canna
indica inoculated with X. campastris pv. musacearum. There were no symptoms
observed on other tested plants (Ssekiwok et a l, 2006b).

Tripathi and Tripathi (2009) evaluated 10 cultivars of banana by injecting bacterial

inoculum in pseudostem of in vitro plantlets and potted plants. Pisang Awak, Dwarf
Cavandish, Giant Cavandish and FHIA-17 were highly susceptible; Mpologoma,
Mbwazirume, Sukali Ndiizi and FHIA-25 were susceptible while Nakitembe was
least susceptible. Musa balbisinna was found resistant (Tripathi and Tripathi, 2009).
Apparently, all commercial cultivars are found susceptible to the pathogen.

Insects found on banana inflorescence are possible sources of inoculums in banana

plant. Insect species that carried the bacterium on their bodies were determined as
possible vectors of the disease (Tinzaara et al. 2006). The most abundant insects
visiting banana flowers are stingless bee (Plebeina denoiti (Vachal) (Aphidae), fruit
flies /Drosophilidae/ and grass flies /Chloropidae/. Female flowers had twice as
many insects as male flowers (Table 2). The bacterial cells have been isolated from
the stingless bee / P. d en oiti/, honey bees (Apis melifera), fruit flies and grass flies that
had been collected from male flowers of both asymptomatic and symptomatic plants.
The bacterial cells isolated from P. denoiti were more than two times as many as other
insect groups (Table 3). Moreover, insects generally visited during the entire day
with the peak visitation from about 12.00-2.00 (Tinzaara et al., 2006).

Table 2. Number of insects visiting female flowers /asymptomatic and symptomatic/ of infected and non infected banana fields

Infected banana fields

Non-infected banana
Symptom f'e^ s
Common Asymptomatic
-tic
tnsef/family name Female Male Male Female Male
Plebeina denoiti (Aphidae) Stingless bee 39.4±4.1 34.1 ±2.9 26.6±2.8 11.8±2.7 7.0±1.1
Undetermined species (Aphidae) Stingless bee 2.8±0.5 2.6±0.3 2.3±0.4 2.3±9.0 2.5±0.5
Meioonula sp. (Apidae) Stingless bee 1.0±0.5 1.7±0.3 2.0±2.0 0.0±0.0 0.0±0.0
Apis melifera (Apidae) Honey bee 3.6±0.6 3.3±0.5 3.3±1.2 2.0±0.6 2.8±1.2
Undetermined species Grass flies 2.8±0.4 3.7±0.6 11.8±2.5 7.5±4.5 2.7±0.9
(Chloropidae)
Undetermined species (Aphidae) Fruit flies 14.1 ±4.1 9.3±1.6 10.2+1.8 21 3±6.2 2.7±1.2
Source: Tinanrn cl nl., 2006
 79 Eshetu Ahmed and Mohammed Yesuf

Table 3. No. (±s.e) of Xanthomonas campastris pv. musacearum colonies isolated from insect
vectors collected from asymptomatic and symptomatic flowers of the cultivar Kayinja

Common Mean no. of cfu per insect

Enset/family
name Asymptomatic plants Symptomatic plants
Ptebo ina denoiti (Aphidae) Stingless bee 1645±1197(3) 6073±3274(25)
Undetermined species (Aphidae) Stingless bee 2637±977(5) 1368±3274(9)
Undetermined species (Chloropidae) Grass flies -b 2543±1963(7)
Undetermined species (Drosophilidae) Fruit flies 1647±1197 (3) 2398±1294 (11)
Apis melifera (Apidae) Honey bee - 5058±3275(6)
Souice: Tiimnrn et at., 2006;

A study was also carried out to estimate the economic value of the likely loss due to
banana bacterial wilt (BBW) if not controlled and the potential of implementing the
short term control. If BBW is not controlled, Uganda stands to lose an estimated 295
million dollars worth of banana output valued at farm gate prices. This translates
into an annual 200 USD (Table 4) of food and income per household at stake
(Kalyebara et cil., 2006).

Table 4. Estimated annual economic losses with and without intervention

Year Loss if BBW is not controlled (USD Loss if ABCC is adopted (USD
2005) 2005)
2000 0 0
2001 2506660 2732612
2002 4509878 4563088
2003 13512297 13600690
2004 30108824 30419187
2005 137605308 129510853
2006 204532511 159306130
2007 274388276 168336522
2008 365611044 199109127
2009 453769274 251619347
2010 689802425 366013547
2011 605826477 321455376
2012 532073689 282321678
2013 467299501 247952083
2014 410410866 217766612
2015 360447804 191255894
2016 316567202 167972568
2017 278028586 147523733
2018 244181627 129564322
2019 214455168 113791274
Total 5,605,637,416 3,144,814,644
Average 295,033,548 165,516,560
Source: Kalyebara et at., 2006

Means of disease transmissions are mainly through male flower buds by pollinating
insects, bats and sunbirds. Other means of transmission are use of contaminated
A 80 E xperien ce on B a c te r ia l W ilt M an ag em en t in E astern A frica

* planting materials and cutting tools, browsing animals and water when it moves
around infected soils.

Taxonomic similarity of X. campastris pv. musacearum to X. vasicola pv. musacearum

was mentioned. Some authors renamed the pathogen to X. vasicola pv. musacearum .
It th:s is to be true, then scientists need to establish the host ranges especially of
maize, sorghum and sugar cane.

Disease Management
There are two key pillars of actions in combating BBW. These are first promptly
removing sources of inoculums and reducing opportunities of spread. Cutting down
and heaping diseased plants as a way of eliminating inoculums. Since this is tedious
and cumbersome, systemic herbicides like glyphosate and 2,4-D were tested using
five concentrations to determine the one that is more effective in destroying infected
banana plants. All plants injected with 2, 4-D herbicide snapped and died within 30-
40 days while with glyphosate it took 90 to 120 days (Okurut et a l, 2006). Both
herbicides were able to kill at least 85% of the plants. Besides these, herbicides dose,
material size and nature of suckering was studied. Resuckering rates between both
herbicides did not significantly differ. The resuckerings for 2,4-D were 11.3% and
7.2% at station and farmers' field while for glyphosate 15% and 1.5% were the values
for on station and farm, respectively. Herbicides were able to destroy plants within
radius of 16.8±1.76 cm from injected plant/mat. Higher dose killed better and earlier,
but caused higher resuckering. In Uganda, based on environmental safety,
glyphosate is recommended for farmers use.

The disease causing bacterium, X. campastris pv. musacearum survives in the absence
ol a host for only 20-90 days depending on the environment (Mwebaze et al. 2006). It
is recommended to suspend pruning, corm removal, ploughing and leaf harvesting
and resume these activities only after the disease has been cleared (3-6 months after
the last diseased plant). The use of tools when removing infected plants or
harvesting and utmost care not to infect other plants (clean the tool every after a
plant harvested) are recommended routine practices.
Works are under way to develop resistant banana varieties through genetic
transformation (both Agrobacterium-media ted transformation and microprojectile
bombardment) (Tripathi et al., 2008).

Development and Implementation of Control

Strategy
* In Uganda, task force was established to develop a strategy and action plan to
eradicate the disease. The strategy emphasized massive creation of awareness,
cutting and burying infected banana stools, quarantine, decapitation of male buds,
 81 Eshctu Ahmed and Mohammed Yesuf

and disinfection of farm tools used in the affected fields. Such quick measure
reduced disease incidence below 10% per year, but could not be sustained due to
high implementation costs. Consequently, the disease was reported in more
districts necessitating a change of strategy and hence the National Agricultural
Research Organization (NARO) was directed to develop a comprehensive research
ard development strategy from eradication to contain and manage the disease.

Establishm ent aiul coordination of* the banana bacterial wilt control
initiative ( B B W C I)
Technical committee from Ministry of Agriculture, Animal Industry and Fisheries
(MAAIF), NARO, Mekerere University, local government and International
Collaborating Institutions was formed and charged with the task of guiding
implementation of the activities for control of the disease. A National Coordinator to
spearhead the efforts to control the disease w*as appointed and tasked to coordinate
all activities on BBW control in the country as one entity known as the National
Banana Bacterial Wilt Control Initiative (BBWCI).

The BBWCI was charged with responsibility to ensure that all institutions, that have
banana in their activity portfolio, integrate BBW control in their action plans. To
strengthen coordination and monitoring of BBW control activities, a system of task
forces was provided for at various local government levels. It w as envisaged that the
National Coordination System, with the National Steering Committee (National Task
Force) as its apex, would link to the farming communities in villages through district
task forces, sub-county task forces, parish task forces and finally village task forces.
Such arrangement would ensure participation of all stakeholders in control of the
disease. The activities of BBWCI at district, sub-countrv, parish and village levels
Would be coordinated by public extension staff with task forces at the different levels
playing the monitoring and overseeing roles at their respective levels.

With the BBWCI structure in place, teams simultaneously embarked on activities

aimed at containing and controlling the disease as well as monitoring the impact oi
the efforts.

rS ational sensitization campaign for control of*the disease

"h e BBWCI set priorities to enable most effective use of the insufficient financial
resources available to it.

• Highest priority was accorded to the disease free/threatened zone where plantations
needed protection from the disease. Pockets of disease outbreak were anticipated and
the goal of the initiative in these areas was to eradicate disease outbreaks.
• The second priority was halting further advance of the disease frontline zone (the
advancing edge of the disease endemics) - main producing region in the country. The
goal of the initiative was to push the endemic zone backwards by eradicating the
disease in affected plantations in the frontline.
 82 Experience on Bacterial Wilt Management in Eastern Africa

• The third priority was ranking zones where the disease was considered endemic. As
the short term objective of this initiative was to enable farmers to cope with the
disease with ultimate goal of eradicating the disease from farmers' fields.

The initiative deployed both conventional communication methods ideal to

reaching wide audiences and participatory communication methods and
initiating action at community level. The sensitization campaign involved raising
stakeholders' awareness about the disease and its control through trained
trainers and multiple communication channels. The conventional communication
approaches deployed are:

M ultiple communication channels: BBWI produced and distributed 40000

posters in 2004 as well as 67000 brochures, 100000 newspaper inserts and
6000calender posters in 2005. Radio spot messages and talk shows were done in
five different languages. The campaign was scaled up in the first half of 2006
with the production of 100000 refined posters, radio spot messages and talk
shows in ten languages on 18 local radio stations and 10000 brochures targeting
extension workers. The use of bill boards and school poster package were also
piloted.
® T ra in ed tra in e rs: Two hundred and ninety six extension workers all drawn
from the three zones were trained on identification and control of BBW in
preparation for deployment as trainers in 2004. Additionally, 148 service
providers drawn from the National Farmers' Foundation were trained in 2005 to
support the trained extension workers.
• S e m in ars and workshops: Several seminars and workshops at district and
grass root levels were conducted by trained trainers. These enabled the many
local leaders, opinion leaders, NGOs, and interested farmers to join the
expanding pool of trainers.
• P u b lic gatherings: Going at social gatherings such as religious occasions,
funerals gatherings, political rallies and markets etc was founded instrumental
in spreading information about BBW and tracking down its progress. The
technique was mostly used in the unaffected areas to track the disease
outbreaks.

The information from farmers and task forces in affected areas was channeled to
the coordination office through their local government extension workers. The
BBW control team made on-spot checks for each of the new areas where the
disease was reported and mobilized the local leaders to set up task forces for
containing and controlling the disease using a participatory approach.

T h e B B W Control Recommendation Packages

• Avoid introducing the disease into any area, garden or unaffected plants. Avoid
using tools in the plantation until all are disinfected, or sterilized before using them.
This prevents stool to stool spread of the disease through the tools; restricting entry
of infected bananas or banana parts into any area/garden.
 83 Eshehi Aliined and M ohammed Yesuf

• Break male buds from ail banana plants at about two weeks after flowering with a
forked stick. This will prevent disease transmission by insects which visit the flower.
Using a stick therefore, prevents transmission that would otherwise be transmitted
by cutting tools.
• Clear all disease plants by uprooting and heaping or using 'roundup' (glyphosate)
herbicide as recommended by the BBWCI at the rate of 1 ml product/banana mat).
This eliminates source of inoculums.
• Clean tools used to clear infected plants by briefly heating them with fire flame or
dipping them in a "Jik " (sodium hypochlorite solution) in 1- 5 water to Jik ratio.
Cows and goats should not be fed on diseased plants or be allowed to pass through
banana plantation, as these mechanically spread the disease from infected plants to
healthy plants (through feeding on the plants).
• If freshly diseased plants are not observed for 3 to 6 months, resume using the tools
in the banana plantation. However, it is advisable to remain vigilant as fresh infection
is highly likely to occur.

Community action, in the form of bylaws and working together, is important .to
mobilize errant farmers who do not want to implement recommended control
measures. The public stands to lose when these errant farmers are left alone, and
therefore the problem needs to be addressed by public action.
The conventional top-down approach was instrumental in swiftly raising awareness
of stakeholders about the disease across the whole country, but was not effective in
triggering actions aimed at controlling the disease. Therefore, participatory
development communication (PDC) approach was formulated and deployed which
was effective previously in Canada (Bessette, 2004). This approach centers on action
plans developed by communities to address specific problems facing them. A
development worker trained in this approach facilitates the community to develop
the action plan.

• The approach was introduced to community leaders to get buy-in as foundation for
full implementation and scaling out.
• Community members, their leaders, development workers and researchers were
brought together to share information on the BBW and other banana management
problems. It ended with the farmers identifying strategies they wished to try to solve
the BBW problem (prevention if farmer had unaffected plantation, or eradication
measures if plantation is affected).
•» Community members were grouped on the basis of the strategies applicable to their
BBW situation (protecting unaffected gardens or eradicating the disease)
«» Each group identified actions required to implement the solution to be tried,
resources required, technical needs, partnerships required and who would do what
«• Each group developed a communication plan to eventually enable them share lessons
learned with the wider community if their efforts paid off.
• Finally, the community developed a monitoring and evaluation plan, a schedule of
activities and agreed on how to source for required resources.
 84 Experience on Bacterial Wilt Management in Eastern Africa

k The plan was then implemented and the success story created at this site was shared
with wider community through various communication channels and released at a
massively attended open day presided over by MoAAIF
The community action plan strategy was scaled out to new areas through a training
of trainers approach as is done in conventional extension approach.

Gaps/Challenges to Fully Implement Cultural

Practices
Unfortunately, farmers are generally reluctant to cut off male buds especially on
Kayinja and Kivuvu. Farmers have advanced various reasons for not cutting off the
male buds. Some reported that they wanted to use the male buds to identify affected
bunches so that they do not harvest them for consumption. Some noted that it is in
* their tradition not to remove male buds from Kayinja because the practice reduces
the cuality of alcohol produced from such bunches. Others just found it tedious to go
through Kayinja plots to remove the male buds (Bagamba et a l , 2006).

References

Aritua V.: Parkinson N, Thwaites R, Heeney JV, Jones DR, Tushemereirwe W, Crozier J,
F eeder R, Stead DE, SmithJ. 2008. Characterization of the Xanthomonas sp. causing wilt of
enset and banana and its proposed reclassification as a strain of X. vasicola. Plant Pathology
57(1), 170-177.
Aritua, V.; Parkinson, N.; Thwaites, R.; Heeney, J.V.; Jones, D.J.; Tushemereirwe, W.; Crozier,
J.; Reeder, R.; Stead, D.E. and Smith, J. 2007a. Characterization of the Xanthomonas sp.
causing wilt of enset and banana reveals it is a strain of X. vasicola. Plant Pathology. 57(2),
157-164.
Aritua, V.; Parkinson, N.; Thwaites, R.; Weller, S.R.; Parkinson, N.; Tushemereirwe, W.;
Reeder, R. and Smith, J. 2007b. Conventional PCR and real-time TaqMan assays for
detection of the banana xanthomonas wilt pathogen and related X anthom onas vasicola
strains. Plant Pathology. Cited In:. Smith J], DR Jones, E Karamura, G Blomme and FL
Turyagyenda. 2008. An analysis of the risk from Xanthonionas campestris pv. musacearum
to banana cultivation in Eastern, Central and Southern Africa. Bioversity International,
Montpellier, France.
Bagamba, F.; Kikulwe, E.; Tushemereirwe, W.K.; Ngambeki, D.; Muhangi, J.; Kgezi, GH.;
Ragama, P.E.; and Eden-Green, S. 2006. Awareness of banana bacterial wilt control in
Uganda: 1. Farmers' perspective. African Crop Science Journal, Vol. 14(2): 157-164
Bessette, G. 2004 Involving the Community. A Participatory Guide to Development
' Communication. Southbound, Penang and International Development Research Centre
(IDRC), Ottawa.
Bradbury, J.F. 1986. Guide to plant pathogenic bacteria. CAB International, Wallingford.
United Kingdom.
 85 Eshetu Ahmed and M ohammed Yesuf

FAO 1998. Production Year book for 1998. Food and Agricultural Organization. Rome.Italy.
Kalyebara,M.R.; Ragama, P.E.; Kagezi, G.H.; Kubiriba, J; Bagamba; Nankinga, K.C. and
Tushem ereirwe, W. 2006. Economic importance of the banana bacterial wilt in Uganda.
African Crop Science Journal, Vol. 14(2):175-183
Maina, M. and Muthoni, S. 2008. Survival of Xanthomonas vasicola pv. musacearum on metallic
tools. J. Applied Biosciences Vol. 3: 92 - 94.
M;,*enzi, S.R.B.; Eden-Green, S.J. and Peacock, J. 2006a. Overview of banana xanthomonas wilt
in Tanzania. In: Saddler, G.: Elphinstone, J. and Smith J. (Eds.). Programme and Abstract
Book of the 4th International Bacterial Wilt Symposium, 17,h-20th July 2006. The Lakeland
Conference Center, Central Science Laboratory, York, United Kingdom. P. 107
M jjenzi, S.R.B.; Muchunguzi, D.; Mutagwaba, T.; Mkondo, F.; Mohamed, R. and Aritua, V.
2006b. An out-break of banana bacterial wilt disease in Muleba district, Kagera
regxanthomonas wilt in Tanzania. In: Saddler, G.: Elphinstone, J. and Smith J. (Eds.).
Programion, Tanzania. AfricanCrops. Net.
(http://w w w .africancrops.net)/neus/apnKK>/ bacterialvvilt.htm).
Muhangi, J.; Nankinga, C.; Tushemereirwe, W.K.; Rutherford, ML; Ragama, P.; Nowakunda,
K.; Abeyasekara, S. 2006. Impact of awareness campaigns for banana bacterial wilt
control in Uganda. African Crop Science Journal, Vol. 14(2):93-103
Mwebaze, J.M .; Tusiime, G.; Tushemereirwe, W\K. and Maina, M. 2006a. Development of a
semi-selective medium for Xanthomoas campastris pv. musacearum. African Crop Science
Journal, Vol. 14(2): 129-135.
Mwebaze, J.M .; Tusiime, G.; Tushemereirwe, W.K. and Kubiriba, J. 2006b. The survival of
Xanthomonas campastris pv. musacearum in soil and plant debris. African Crop Science
Journal, Vol. 14(2): 121-127.
Ndungo, V.; Eden-Green, S.; Blomme, G.; Crozier, J. and Smith, J. 2005. Presence of banana
Xanthomonas wilt (Xanthomonas campastris pv. musacearum) in the Democratic Republic of
Congo (DRC). Plant Pathology 55:294.
Naam beki, D.S.; Tushem ereirwe, W.K. and Okaasai, O. 2006. Awareness of banana bacterial
wilt control in Uganda: 2. Community leader' perspectives. African Crop Science Journal,
Vol. 14(2): 165-173
Okurut, A.W.; Tushemereirwe, W.K.; Aritura, V. and Ragama, P.E. 2006. Use of herbicides for
control of banana bacterial wilt in Uganda. African Crop Science Journal, Vol. 14(2):143-149.
Reeder, R.H.; Opolot, O.; Muhinyuza, J.B.; Aritua, V.; Crozier, J. and Smith, J. 2007. Presence
of banana bacterial wilt (Xanthomonas campestris pv. musacearum) in Ruwanda.
(http://www.bspp.org.uk/jan2007/2007-01.asp).
Sharrock, S.L and Frison, E.A. 1999. In: Networking banana and plantain. INIBAP Annual
Report 1999.
Smith, J.J.; Jones, D.R.; Karamura, E.; Blomme, G. and Tuyagyenda, F.L.. 2008. An analysis of
the risk from Xanthomonas campastris pv. musacearum to banana cultivation in Eastern,
Central and Southern Africa. ISBN: 978-2-910810-82-5.Bioversity International, 2008.
Rome, Italy/ Montpellier/France.
Smith, J. 2007. Summary of a workshop entitled Expert Consultation on progressing the road
map for the control of banana xanthomonas wilt in Uganda and across East Africa’;
hosted at Central Science Laboratory, York, UK, 24-27th July 2006.
Ssekiwoko, F.; Taligoola, H.K.; and Tushemereirwe, W.K. 2006b. Xanthomonas campastris pv.
musacearum host range in Uganda. African Crop Science Journal, Vol. 14(2): 111-120.
 86 E xperien ce on B a c te r ia l W ilt M an ag em en t in E astern A frica

Ssekiwoko, F; Tushem ereirwe, W.K.; Batte, M.; Ragama, P.E. and Kumakech, A. 2006a.
Reaction of banana germplasm to inoculation with Xanthonionas campastris pv.
musacearum. African Crop Science Journal, Vol. 14(2): 151-155.
Tinzaara, W.; Gold, C.S,; Ssekiwoko, F.; Tushemereirwe, VV; Bandyopadhyay, R.; Abera, A.;
a id Eden-Green, S.J. 2006. Role of insects in the transmission of banana bacterial wilt
African Crop Science Journal, Vol. 14(2):105-110, 2006.
Tripathi, L.; Tripathi, J.N. and Tushemereirwe, W.K. 2008. Rapid and efficient production of
transgenic East African Highland banana (Musa spp.) using intercalary meristematic
tissues. African /. Biotechnolog}/ Vol. 7(10):1438-1445.
Tumushable, A.; Ssekiwoko, F; Tushemereirwe, W.K. 2006. Potential of infected banana parts
to transmit Xanthomonas campastris pv. musacearum. African Crop Science Journal, Vol. 14(2):
137-142.
Tushem ereirwe, W.K.; Kangire, A.; Smith, J.; Ssekiwoko, I7.; Nakyanzi, M.; Kataama, D.;
Musiitwa, C.; and Karyeija, R. 2003. An outbreak of bacteria! wilt oil banana in Uganda.
Infomusa 12:6-8.
I ushemereirwe, W.K.; Kangire, A; Ssekiwoko, F.; Offord, L.C.; Crosier, J.; Boa, F. Ruthford,
M.; and Smith, J.J. 2004. First report of Xanthonionas campastris pv. musacearum on banana
in Uganda. Plant Pathology 53:802-802.
Tushem ereirwe, W.K.; Okaasaji, O.; Kubiriba, J.; Nankinga, C.; Muhangi, J.; Odoi, N. and
Opio, F. 2009. Status of banana bacterial wilt in Uganda. Banana Bacterial Wilt in Uganda.
African Crop Science Journal, Vol. 14(2): 73-82, 2006.
\ iigou, I), and Bradbury J.F. 1968. Bacterial w ill of enset (Ensete oentnco^iim ) incited In
Xiinthtwiouas m usacearum sp.n. P hytopathology 58, 111-112.
Yirgou, D. and Brudbury, J.F. 1974. A note on wilt of banana caused by the enset wilt
organism Xanthomonas musacearum. East African Agricultural ami Forestry Journal 40:111-
114.
Regional Strategy for Banaiia/Ensct Bacterial
Wilt Management in East Africa: Flic Road
Map for the National Action Plan
Tcsfahun F c n ta 1 and Elriad K aram ura2
1Agri-Sennce Ethiopia, P.O.Box 2460, Addis Ababa
1 Bioversity International, Kampala, Uganda

Abstract
Bacterial wilt (Xanthomonas campestris pv. musacearum) has continued to
threaten enset and banana production in East Africa, endangering the livelihoods
o f poor, small-holder farmers. Man y countries now regard it as a major priority
constraint to banana production, overtaking weevils, nematodes, and fungal and
viral diseases, fo r which farm ers applied cultural practices that affected the
differences in cultivars' resistance / tolerance on farm. These way farmers suffered
in reduced productivity, but maintained a reasonable level o f food and income
security. However, with the arrival o f banana bacterial wilt in the region, entire
crop holdings were wiped out in some areas, where highly susceptible genotypes
were dominating the farming systems. It is not yet established how the disease
spread out" o f the Ethiopian highlands into the mid-altitude regions o f East and
Central Africa, where the disease reportedly causes 80-100% crop loss especially in
ABB beer bananas in Uganda and Democratic Republic o f Congo (DRC). Previous
efforts to control the disease produced partial solutions and the disease has
continued to invade previously disease free areas and to resurge in areas where it
had been controlled. Current control strategies are facing major challenges
including the inability to detect latent infection (in planting material and other
enset/banana products), the user-unfriendliness o f some recommendations such as
JJanie and chemical decontamination o f implements; lack o f genotypes resistant to
the disease and lack o f coherent institutional fram eworks for organizing and
mobilizing stakeholder partnerships within and between countries, to exploit
stakeholders' synergies at local, national and regional levels.

Introduction
The communities of the East African plateau (Western Kenya, Uganda, North­
western Tanzania and Eastern D R Congo) have traditionally depended on a
perennial banana cropping system for food and income. This farming system is a
slow-changing and rural-based economy that lias ensured regional food security for
many centuries. In recent decades, this system has come under stress, with
decreasing farm size, decreasing fallow periods and production progressively,
unable to meet household needs, leading to increased inter-communal tensions. In
 88 R e g io n a l S tratcg}/ fo r B au an a/E n set B a c te r ia l W ilt M an a g em en t in E ast Africa

the coming decades, the sub-region may face additional stresses, including climate
change, characterized by increasing temperatures and greater rainfall variability,
plus changing social structures due to HIV-AIDS, rural-to-urban migration and
declining returns from agriculture. Until the beginning of the century, the main
biotic threats (weevils, nematodes, fungal and viral diseases) were managed via
cultural practices exploiting differences in cultivars' resistance / tolerance on farm.
In this way farmers suffered reduced productivity, but maintained reasonable levels
of food and income security. However, following the arrival of unset and banana
Xamhonionas wilt (BXW) (Xanthonionas cam pest ris pv. nuisacetinun) in the region, entire
crop holdings were wiped out in some areas, where highly susceptible genotypes
were dominating the farming svstems. Until the 1960s, the bacterial wilt had been
known only in Ethiopia, in both bananas and enset.

In 2001, the disease was reported simultaneously in Central Uganda and in North
Kivu province of D R Congo and a year later in North-western Rwanda. Between
2002 and 2006, the disease spread to the major banana growing regions of Uganda,
Western Kenya, and other countries in Central Africa, where it caused 80-100% crop
loss especially in beer bananas in Uganda and D R Congo. BXW continues to
seriously threaten enset and banana production in the region and sustainable wilt-
di sease control measures must be implemented, to avoid further devastation.

lU^ional Distribution o ffe ^ / IJa n a n a Bacterial Wilt

Bacterial wilt is currently restricted to Africa. The first report of X.campestris
pv.nnisncennuu causing a wilt on Musaceae is on enset in Ethiopia (Yirgou and
Bradbury, 1968) with the disease reported as being present in many iJ//sef-growing
localities in Central and Southern Ethiopia. However, the bacterial wilt of enscl with
symptoms consistent with X.cnmpestris pv musacearum, was described previously in
the late 1930s. Subsequently, X.cnmpestris pv musacearum as a pathogen of banana
w:is first described in Kaffa province and later in warm, moist areas of other
provinces in Ethiopia (Yirgou and Bradbury, 1968). Enset is banana like plant on
which approximately 20% of the Ethiopia's population depends for food, feed and
fiber (Tsedeke et til.,1996) The occurrence of the disease on banana was a less
common crop at that time. Surveys conducted on enset wilt incidence in 24 different
localities in 1997/98 and 1998/1999 crop seasons indicated that the percentage
incidence was highest at Gera and Suntu in 1997/1998, while in Waka, Gera and
Solemo the incidence was high only in 1998/99 seasons (Abraham, 2009).

Xanthonionas wilt was first reported in Uganda in Mukono district of Central

Uganda in 2001 (Tushemereirwe et til, 2004; Karamura et al., 2006). By 2006, it was
confirmed in 35 districts in all regions of the country. It has now been reported in 319
sub-counties out of a total of 986. It is said to be endemic' in 244 sub-counties with
 89 Tcsfahun Fcnta and Eldad Karamura

spread contained or controlled in 70 sub-counties. The experience in managing

banana bacterial wilt some African countries is presented in brief.

D em ocratic Republic o f Congo

Banana Xanthomonas wilt was first observed in 2001 at Bwere, in the North Kivu
Province bordering Uganda and Rwanda two years before it was officially recorded.
Since then, the disease has been spreading consistently in all directions from Bwere,
gaining about 5km per year. A second outbreak was discovered in 2003,10 km from
Bwere, with the same spread pattern.

Bananas represent the main staple food in North Kivu. Prior to the disease
infestation in the region, 70% of the cultivated land was occupied by bananas,
generating 60% of household income. A survey conducted in the Bwere region
indicated that the BXW outbreak has caused complete failure of banana harvests in
most farmers' fields. Yields have declined from 20 t/ha/year to almost zero with a
corresponding income loss of about 1600 $/ha/year (Karamura E et al, 2006).

Wilt enset is affected in Congo, supporting the idea that the disease is related to the
bacterial wilt in Ethiopia. Among the cultivated bananas the disease mainly affects
Kayinjas, Ndizi, plantains, EAHB and Cavendish in that order. The symptoms
observed usually involve yellow discoloration of the stem, yellowing and dying off
of the leaves, yellow ooze and discoloration of the fruit. Farmers believe that the
infection gets into the plant through the leaves or flowers and continues down to the
stem and corn.

Movement of the disease has been associated with the lakes and there is a suspicion
that migrating birds that regularly visit the lake shores may be involved with the
introduction of the disease to the region. It is also suspected that leaf-feeding insects
are involved in disease spread. In affected areas, the farmers are losing hope. Strict
sanitation measures are under adoption, including chopping the pseudo-stem into
pieces and leaving them to dry.

Kenya
Banana is a major crop in Kenya covering 74,000 hectares (2% arable land) and over 1
million t/year are produced. The Ugandan Phytosanitary Department informed
Kenya when BXW broke out in 2001. Scouting surveys were later conducted in
Western and Coastal regions of Kenya in September and December 2004. This was
followed by deploying interviews and transect walks, and collecting samples. BXW
was not encountered but a range of other pest and disease problems (Weevils,
Fusarium wilt, Sigatoka etc.) were observed. It was also noted that bananas are
imported from Uganda and Tanzania and marketed in major towns in Kenya. The
introduction of the disease into Kenya was likely through plant materials.
 90 Regional Strateg}/ fo r Banana/Enset Bacterial Wilt Management in East A frica

A number of actions have been taken since the scouting surveys, including stepping
up border inspections, instituting a system of import permits to control and monitor
cross-border movement of banana materials and products, issuing phytosanitary
certificates for bananas entering Kenya; banning sucker importation (except tissue
cultured materials under quarantine), intensifying information exchange with
Ugandan counterparts (two plant inspectors visited Uganda to familiarize
themselves with the disease), instituting regular communication with stakeholders
about the threat of the disease; and certifying nurseries distributing banana
seedlings.

T anzania
Xanthomonas wilt was first detected in September 2005 by farmers in Kabale village
of Izigo division within Muleba District, which is in the Kagarea region borderining
Lake Victoria, Uganda, Rwanda and Burundi. In January 2006, the outbreak was
confirmed as Xanthomonas wilt and symptoms were also seen in nearby Kabale B
sub-village. Plants in Izigo, Kikondo, Bumilo, and Magata villages in Maruku district
were also affected. The scattered outbreaks were separated from each other by tens
of kilometers. It was estimated that banana plants belonging to about 100 households
had the disease. Spread from Eastern Rwanda was suspected, although there were
no reported out breaks in this area at that time.

Uganda
M aiianinii B a n a n a B a c te ria l W ilt
Banana is a very important food and cash crop supporting over 70% of the
population of Uganda. Production in all year round reaches a potential yield up to 60
t/ha/year. The crop, therefore, has high industrial potential. It has however, been
constrained by a number of factors:
In the implementation of the national action plan, a number of challenges have been
encountered. These include the shortage of resources to implement priority activities,
the mobilization of key stakeholders along the production-consumption chain (e.g.
central and local governments, development partners, community leaders, faith-
based organizations, NGOs and farmers) to support the control efforts and the weak
linkage between research and extension. At the farm level, the initiative has met
challenges posed by traditionally low levels of adoption of agronomic practices in
affected areas and negative attitudes towards implementing some of the
recommended practices such as sanitation based husbandry. These challenges
however are being addressed through participatory approach and awareness
campaigns.

Community Approaches in Managing B X W

P a rticip a to ry development com ninnication (P D C )
Farming communities were directly targeted using a participatory development
communication (PDC) approach. This approach involves members of the
 91 Tesfaliun Feuta ami Eldad Karamura

community to take part in problem identification and analysis, and enables the
community to analyze and explore alternative solutions to the problem and to
identify the best solutions which they are ready to implement. PDC helps to know
why some members in the community don't implement some control measures and
to identify the constraints that they face in implementing control measures. Using
community knowledge, researchers are able to develop technologies more
effectively. The model was tested in three BXW-affected zones. Stakeholder
planning workshops also took place involving agricultural extension staffs fro
district and sub-county levels, political leaders (Local Council levels 1 - 3);
educational institutions, schools, cultural and religious leaders, NGOs, CBOs and
farmers.

The workshop participants identified and prioritized banana-related community

needs and constraints and agreed on technologies to address them. Through PDC,
interactive discussions identified 1PM and BXW problems (pests, diseases, soil
fertility, banana cultivars, etc.) and agreed on solutions, including field
demonstrations to show BXW symptoms, transmission and control measures.
Subsequently, the information acquired was used to develop community BXW action
plans to be executed by task forces at various community levels.

In the execution of PDC, a number of challenges emerged. Participants had diverse

perceptions about the cause of BXW (including deliberate introduction by scientists,
exotic bananas, tissue culture plants, degraded soils, and Mulinga tree, use of
molasses in banana systems, etc.). They also discussed reasons why some people in
the communities did not implement the recommended control measures. The
reasons given included ineffective measures, labor intensive measures, laziness, lack
of e fective alternatives, costly disinfectants, miscommunication from neighbors,
researchers who wanted to eliminate traditional technologies, negative attitudes
towards technologies, contradictory messages, and fear that new cultivars might
degrade the soil.

The PDC's interactive approach enables farmers, who were ignorant of basic crop
management and IPM practices to take advantage of the ex te n siv e experience of
other farmers. Community leaders, having been made fully aware of the threat,
pledged to support the effort to fight the disease, and it was recognized that the
responsibility of the BXW problem has to be shared by the whole community.

Com m unity o rgan ization al fram ew ork

, In Uganda, the community task forces, consisting of four people, mobilized
individual communities in partnership with parish and village councils, trained
people on BXW, directed the development and implementation of community
* actions plans, and monitored and communicated progress to the task force at the
next community level. The task force was also responsible for the establishment of
 92 Regional Strateg]/ fo r Banana/Enset Bacterial Wilt Management in East Africa

demonstration plots and maintaining networks with other partners in control of

BXW (agric-extension, NGOs, NARO, MAAIF and others).
Training videos in different languages were used to disseminate information on
BXW symptoms, transmission and control, PDC methods, the sensitization process
arid success stories of BXW control in model districts. Fact sheets on BXW
symptoms, transmission and control; guidelines for formation of BXW task forces in
the community; posters and brochures on BXW; and banana production manuals
were also disseminated.

The experiences gained from the initial PDC model in the pioneer areas were used to
soile out activities to other areas. The original model was slightly modified in order
to gain more political backing and funding support, by targeting the Chief
Administrative Officer and the Local Council chair person at the highest level in the
district rather than the District Agricultural Extension (DAE).

T h e Regional Strategy fo r M aii;t“ iii“ /Jusct/W.xnunix lia c tc r ia l W ilt

The disease outbreak in Uganda was reported in 2001. Since then the Banana
Research Network for Eastern and Southern Africa (BARNESA) Steering Committee
had rated BXW as a top R4D priority and has taken steps to coordinate regional
efforts aimed at alleviating the impact of the disease on the livelihoods of the affected
communities. In this regard, INIBAP-BARNESA has coordinated a multi­
ins titutional, multi-stakeholder study (funded bv the DFID-UK and IDRC-Canada)
to assess the impact of BXW on the livelihoods of communities along the production
consumption pipeline, and to understand how the communities are coping with the
disease. In addition, INIBAP in collaboration with the FAO is convening a number
of regional and international meetings to develop plans for addressing the problem.
This strategy was developed from the recommendations of a regional workshop held
in February 2005. It is envisaged that the involvement of international, regional,
national and local level actors and interventions to coordinate and bring synergy into
research and extension efforts and to support farmers and communities in 'people-
focused' activities.

In the short-term, the strategy aims to generate and disseminate robust, diagnostic
tools that facilitate disease recognition, management and control. In addition it
envisions a program for spatial surveillance to monitor what is happening where in
the region and a regularly-updated portal for information sharing. At all levels of
the production-consumption chain, capacity must be strengthened. All activities will
need to be coordinated and monitored to ensure that corrective measures are taken
in time.

In the medium term, a regional impact tracking mechanism that regularly generates
and packages information products for policy makers will be put in place and
integrated into the overall regional strategy for integrated pest and disease
management. Regional policy dialogue should be strengthened to allow the
 93 Tesfaliun Fenta and Eldad K a ram u ra

coordination and management of trans-boundary epidemics to food security and

household incomes. In the long term, a systems approach should be adopted to
boost the health of farming systems, taking full account of genetic diversity, the
rescurce base and biotic stresses. Grass-roots ownership and sustainability should
be ensured by deploying a livelihoods approach to improve prospects for marketing
bananas and banana products. Indigenous germplasm threatened by the disease
must also be effectively conserved in perpetuity to ensure that farmers can replant
traditional genotypes once the effects of the epidemic have lessened.

This strategy regional approach to address bacterial wilt of enset/banana is based on

essential learning, from within and outside the region, gained from successfully
conirolling other bacterial wilt diseases, and itself a model of regional response to a
disease epidemic. Such a strategic and comprehensive approach has the potential to
bring the disease epidemic in the East and Central Africa under control and prevent
its further spread. However, its success depends on the ability of the diverse partners
in this effect to mobilize resource and government support to reach communities
throughout affected areas and beyond- and then to ensure com mitment to disease
management campaign and good practice in the longer term.

Ethiopia
Im p ortant issues to be eonsidered in designin'*; the road map lo r n ational aetion plan
In designing the road map, the strategy should underpin the importance of
integrating gender, poverty and environmental conservation considerations based on
the Millennium Development Goal. MDGs are therefore well placed in the national
development context of the country. In line with the objective of poverty eradication
and bringing about social development, the Government of Ethiopia has invested in
both physical and human capital formation which could be considered as best
practice to address the challenges of achieving the MDGs (MDGs Report, 2010).
Nonetheless, the following elements of activities need to be considered in order to
draft the action plan for integrated management of banana/en set bacterial wilt
control in Ethiopia.

• C olle ctio n o f the b a se lin e inform ation: This includes covering disease severity,
farmer perceptions and coping measures being implemented; the establishment of
disease status with respect to its distribution and determ ining the economic
importance of same and its likely impact on enset production. A database will be
continuously updated with this information at Ambo Plant Protection Research
Center, EIAR.

• • G eneration of inform ation on etiology and epidem iology: The evaluation of

inoculation methods for early screening of young plants, determination of pathogen
variability, the methods of pathogen penetration into the host/alternate host, and
* modes and rates of transmission to other hosts; determination of survival
 94 Regional Strateg}/ fo r Banana/Enset Bacterial Wilt Management in East Africa

mechanisms and duration of the pathogen under different environments should

receive due attention.

• D evelopm ent of appropriate technologies for disease m anagem ent: This could be
done using farmers' participatory approaches to evaluate and promote control
measures known to be effective against other bacterial wilt diseases elsewhere,
including removing male buds; rouging diseased plants; developing
resistant/tolerant banana clones through conventional and genetic engineering
methods and disseminating appropriate technologies for the containm ent and control
of the disease.

• Creation of awareness about the disease to reinforce control: Strategies that focus
on awareness creation through mobilization of farmers and their leaders to deploy
sanitation based agronomic practices to protect unaffected areas, blocking further
disease spread, eradicating the disease in the 'frontline' areas and coping with or
eradicating the disease in already-affected areas should be strengthened. The
awareness creation shall include preparation of illustrative posters, fliers, films in
various local languages. The farmers training centers plays a vital role in awareness
creation through short term trainings by developing appropriate modules.

•» Strengthening research and developm ent capacities at all levels: A strategy7 that
addresses research and development with participation of scientists in training in
specialized skills for handling BXW; training of trainers at district, sub-county and
community levels; strengthening capacity of farmers' grass-root institutions to handle
BXW in pilot sites; and developing infrastructure for handling BXW.

<* Establishm ent of partnership among stakeholders at national and regional level:
Establishm ent of effective partnership with sound development plan to elucidate the
vector-disease-host plant relationships in order to understand the survival strategies
of the disease under a range of environments; develop appropriate technologies to
curb long distance disease transmission and arrest intra-farm disease transmission.
Conduct multi-location evaluation of germ plasm linked to clean seed production
systems at grass-roots platforms. Establishment of environment effects of the control
measures on soil fauna and flora and soil conservation is very essential. Disease
surveillance approaches linked to GIS and feeding into strategies for raising public
awareness will strengthen the frameworks for mobilizing and em powering banana-
chain actors to own the bacterial wilt problem and adopt approaches that will deliver
quality bananas and enset production to the market.

• M onitoring the impact of research and developm ent: To make appropriate

adjustments to enset bacterial wilt strategies and accordingly to inform the
governm ent's policy makers with respect to the management and control of the
epidemic, it is important that the impact of research and developm ent is continuously
monitored. The implementation of the strategy need to be monitored and evaluated
as deemed essential. However, it can be done twice a year and shall be reported to
responsible body - the Agriculture and Rural Development Partners Linkage Council
(ARDPLC).
 95 Tesfahun Feuta 'and Eldari Karamura

Recommendation
At present focus shall be made to create wide awareness among the enset growers on
the importance of sanitation based agronomic practices, the nature of the disease as
any ;^reen enset plant can't be assumed free from the disease unless it is tested and
found free from same. Effective disposal methods of infected enset plants have to be
communicated to farmers.

Disease free rhizomes can be propagated with the help of tissue culture to start with
clean planting materials. The sanitation based agronomic practices need to be
supported with early wilt detection techniques. This will make the sanitation
practices more reliable when it is started with clean materials.

The awareness creation approach shall be more systematic using leaflets, posters and
films. Modules and communication booklets have to be developed in local languages
of the enset growers. Farmers training centers (FTCs) can serve as a good forum for
awareness creation.

Regional collaboration is under way to develop PCR based diagnostic method in

Kwanda Research Laboratory, Uganda in collaboration with IITA based in Tanzania.
McKnight Foundation is planning to support the enest bacterial wilt project in
Ethiopia at the same time rendering support to banana bacterial wilt project being
undertaken by Bioversity International in Uganda and Kenya. Thus, the bacterial
wilt project will have a regional program to strengthen the collaboration for
managing enset/banana bacterial wilt in East Africa.

The Ethiopian Institute of Agricultural Research (EIAR) in collaboration with

Regional Research Institutes shall be able to strengthen its ties with the regional
bacterial wilt program. It should also play active role in the Banana Research
network for Eastern and Southern Africa being coordinated by Bioversity
International based in Uganda.
 96 Regional Strategy/ fo r Banana/Enset Bacterial Wilt Management i)i East Africa

References
Abraham Tadesse (Ed.). 2009. Proceedings on increasing crop production through improved
plant. Plant Protection Society of Ethiopia (PPSE) - Volume II PPSE and IAR,
Addis Ababa, Ethiopia. 542pp
Dangnachew Y. and J.F Bradbury 1974. A note on banana caused by enset wilt
Xanthomonas campestris pv. Musacearum. East Africa Agriculture and
forest Journal 40: 111- 114.
Karam ura, E.B., Turyagyenda, F.L., Tinzaara, Blomme, W.G.,Molina, A. and
Markham, R. 2008. Xanthomonas wilt of banana in east and central Africa. Diagnostic
and Management Guide. Bioversity International, Uganda
Karamura E. B, M.Osiru, G.Blomme, C.Lusty and C.Picq (eds). 2006. Developing a regional
strategy to addess the outbreak of Banana Xanthamonas in East and central Africa.
Proceedings of the banana Xanthomonas wilt regional preparedness and strategy
developm ent workshop held in Kampala, Uganda 14-18 February 2005. International
Network for the Improvement of Banana and Plantain, Montpellier, France.
Millennium Development Goals in Ethiopia. 2010. The 2010 annual progress report of MDG.
http://www.et.undp.org/index
Smith J.J, DR Jones, E Karamura, G Blomme and FL Turyagyenda 2008. An analysis of the
risk from Xanthomonas campestris jrv. Musacearum to banana cultivation in Eastern,
Central and Southern Africa. Bioversity International, Montpellier, France
Tsedeke Abate, Clifford Hiebsch, Steven.A.Brandt, Siefu Gebremariam 1996. Enset based
sustainable agriculture in Ethiopia. Proceedings of the International Workshop on
Enset in Addis Ababa, 13-20 December 1993.
Tushem ereirw e W, A. Kangire, J. Smith, M. Nakyanzi, D. Katuuma, and C. Musiitiwa 2001.
An outbreak of banana bacterial wilt in Mukono district: a new and devastating
disease. KARI report, Ugan
 The Essence of Domestic Quarantine against
E n set Bacterial W ilt (luring Technology
Dissemination in Ethiopia
D ereje Gorfu
Holetta Agricultural Research Center. EIAR, P.O.Box 2003, Addis Ababa, Ethiopia

Abstract
Enset (Ensete ventricosum) is widely cultivated in Ethiopia especially in the
Southern part o f the country and bacterial wilt is a serious disease. At present,
enset is expanding to other areas through technology dissemination scheme by
governmental and non-government organizations or individuals, which poses the
risk o f inadvertent introduction o f this disease to new areas. For a country like
Ethiopia with large differences in ecological and physical features, there is sound
biological evidence that justifies domestic quarantine. An attempt was made to
discuss on possibilities o f effective domestic quarantine against enset bacterial wilt
(EBW) during technology dissemination by considering the practical implications
o f the (i) behavior o f the pathogen, (ii) ecology and distribution o f enset, (iii)
human activities, (iv) general quarantine principles, and (in) the development o f
agriculture to effect domestic quarantine. Establishment o f enset nursery center
(ENC) where clean planting materials are developed, inspected and certified by
local government officials and extension serznces are proposed. In addition, how
eradication programs at a country level could be launched with the participation
o f all concerned partners is mentioned. At a local level the possible development
arrangement and important regulatory and mass mobilization issues that deal
with ecological management o f EBW is also discussed.

Keywords: Enset, diseases, bacteria, wilt, EBW, domestic quarantine, pest, risk
analysis, Ethiopia

Introduction
About 25 species of Ensete are equally distributed in Asia and Africa (Mesfin and
Gebremedhin, 2008). Among these species, Ensete ventricosum is widely grown in
Ethiopia and is a staple food for over 20 million people in the Southern part of the
country. It is estimated that about 146 thousand hectares in SNNPR and 79 thousand
hectares in Oromiya are covered with enset (CSA, 2005). Presently, enset is
expanding to other inherently non-enset growing areas of the country, especially to
the north, through technology dissemination scheme by governmental and non­
government organizations or individuals. This obviously poses the risk of
 98 The Essence o f Domestic Quarantine against Enset Bacterial Wilt

inadvertent introduction of serious diseases like enset bacterial wilt (EBW) and
probably other pests to new areas of the country.

Therefore, quarantine precaution measures against inadvertent introduction of pests

into new areas during dissemination and exchange of technologies were important
topics at global level for many years. Likewise, in Ethiopia there was an extensive
review of plant quarantine information regarding import and export certification of
plant materials (Merid et a l, 2008) while no mention was made on domestic
quarantine. However, the plant quarantine guideline published by the Ethiopian
Institute of Agricultural Research (EIAR) for National Agricultural Research System
(Dereje, 2006) provides the possibility of effective precautions through production of
disease-free planting materials with effective phytosanitary measures as components
of domestic quarantine system. For a country like Ethiopia with large differences in
ecological and physical features, there is sound justification to consider domestic
quarantine when enset is introduced to new areas, particularly to the north part of
the country, where there are big natural barriers.

Importation of plant materials into a new areas necessitates serious consideration of

lecjent development in regulatory sciences based on sound biological, economical,
socUal and policy issues. Thus, the current regulatory trend demands to consider
guidelines and procedures outlined by the sanitary and phytosanitary standards
(Sip ) as per the m temational Plant Protection Convention rectified by the
Government. This paper attempts to provide some possible measures and
.<l .^roaches of local regulatory measures in order to be effective during enset
attrib u tio n within the country. This is because, introducing plant materials to new
area requires to conduct pest risk analysis (PRA) that include initiation (based on
specifics on import permits), risk assessment (determining the type and magnitude
of £he risk) and risk management (effecting phytosanitary measures to prevent the
risk). Accordingly, this paper presents summary of EBW information and discusses
the possibilities and procedures of domestic quarantine against enset bacterial wilt
(EBW) during enset technology dissemination in the country. The discussion aimed
co effect domestic quarantine action against EBW based on (i) behavior of the
pathogen, (ii) ecology and distribution of enset and wilt, (iii) human activities
relevant to enset production, (iv) mechanisms of quarantine, and (v) possible
development arrangement to effect quarantine in enset growing regions. It considers
EBW pathway analysis from major enset growing regions as a source of planting
materials and new areas as importers of enset for planting. It also provides
inspection and phytosanitary measures to safeguard the new enset areas. Finally, it
forwards possible development arrangements and important regulatory and mass
mobilization issues that deals with ecological management of EBW.
 99 Dereje Gorfii

* Enset bacterial wilt (E B W )

W ilt and the pathogen

Bacterial wilt of enset was first recognized by Dagnachew and Bradbury about four
decade back (Yirgou and Bradbury, 1968). They identified the causal agent as
Xanthomonas musacearum following sound biological evidences revealing a new
Xanthomonas species at that time although the name was later changed to
Xai thomonas campestris pv. musacearum (Jones et al., 2007). Very recently, however,
comparative analyses of Xanthomonas campestris pv. musacearum with fatty acid
methyl esters, genomic fingerprinting using rep-PCR and partial nucleotide
sequencing of the gyrase-B gene showed some uniqueness in identity of this
pathogen. Pathogenicity studies indicated that strains of X. vasicola pv. holcicola and
X. asicola from sugarcane induced no symptoms on banana, whereas X. campestris
, pv. musacearum produced severe disease (Aritua et al., 2007). This result supports a
future reclassification of X. campestris pv. musacearum as X. vasicola pv. musacearum
when more data are available. This pathogen, causative agent of enset wilt, is found
« in major enset growing regions of Ethiopia although it has not been yet reported on
other Ensete species in other countries.

Although no information was found on enset, the incubation period of wilt was
found to be at most 14 days on banana suggesting the possibility of fast detection of
the disease during quarantine investigation (Ssekiwok et al., 2006). Using
hypersensitive reaction, no variation was detected in 90 isolates of the pathogen
collected from enset grown in southern Ethiopian (Mesfin et al., 2008).

H ost range
Enset and banana are natural hosts of Xanthonionas campestris pv. musacearum (Xcm)
in Ethiopia (Yirgou and Bradbury, 1974). In Uganda Musa sehrina, M. ornate, and
Canna indica were also found to be alternative hosts of this pathogen (Ssekiwoko et
al., 2006). The pathogen was found to frequently devastate enset in Ethiopia and
severely damage banana in Rwanda (Reerer et al., 2007) and Uganda (Ssekiwoko et
al., 2006). DNA fingerprints were identical for Xcm isolates from enset in Ethiopia
and Xcm isolates studied from banana in Uganda, Ethiopia and Democratic Republic
Congo. Apart from these, no other hosts have been demonstrated to be similar so far.
Nevertheless, it is possible that other closely related plant species in family
Zingiberaceae, Marantaceae and Cannaceae could be infected by this pathogen
although no evidence was found so far.

This pathogen causes typical wilt on banana like that on enset (Yirgou and
Brandbury, 1974) and there was severe infection of banana in the country. In
Ethiopia, 40 banana and 114 enset clones were tested for resistance using artificial
inoculation by the pathogen. After 90 days, 98.5% incidence was recorded on banana
 100 The Essence o f Domestic Quarantine against Enset Bacterial Wilt

while only 68.8% incidence was recorded on enset with only 10% incidence in some
clones (Mesfin et a l, 2008). This indicates that Ethiopian banana industry is now at
risk with this disease and the need for establishment of sound phytosanitary
measures is highly indispensible.

Transm ission
Transmission of Xcm from diseased to healthy plants occurs through many ways and
contamination is the major mechanism. Contaminated farm tools and kitchen
utensils (knives, machetes, spades, and hoes) are major inoculants (Dereje, 1985;
Welde-Michael et al., 2008). Spread by animals browsing on infected leaves, use of
infected planting materials, repeated transplanting which damage corms and roots,
and possibly insects visiting bacterial oozes on enset foliage may also occur (Yirgou
and Bradbury, 1974). Since enset is mostly harvested for its pseudostem and corm, it
is not allowed to flower and thus insects infecting flowers in banana are not common
in e:iset, and hence, this transmission mechanism does not seem to be important in
enset in Ethiopia.

S u r v iv a l
The bacterium causing wilt of enset does not seem to survive long under natural
conditions. It survives only up to nine days in soil and up to 90 days in infected plant
residues (Welde-Michael et al., 2007) when the inocula reproduced in artificial media
was kept under different conditions to emulate natural situations. It did not survive
for two months in soil when infected tissues were kept in 6cm depth. This indicates
that the pathogen is short lived organism like many other bacteria which provides
opportunity for effective phytosanitary measures.

B asic Principles Relevant to Domestic Quarantine

M echanism o f Control Relevant to E nset D istribution

Mechanism of plant quarantine operates under five set of guiding principles and
procedures at domestic (local) or cross boundary (international) level. These are (i)
embargoes, (ii) inspection and certification, (iii) disinfection, (iv) special permits, and
(iv) unrestricted shipments. The scope of this paper is limited to only local
precciutions (domestic quarantine), embargoes and unrestricted shipments which are
less relevant and hence the other three principles operate very well for enset under
our conditions. In order to be effective, both pre- and post-entry quarantine
measures for new areas are very important and complementary.

To be specific for enset technology dissemination within Ethiopian, possible pre-

entry'quarantine may include (i) importing enset from pest-free enset nursery centers
(ENC), (ii) inspection and certification at the place of origin (ENC), (iii) production of
disease-free seedlings with special arrangement, (iv) examination and thereby
 101 Dereje Gorfu

sorting the planting materials (corms, suckers, seedlings, etc.) to salvage healthy
seeclings. These measures are dependent on the results of pest risk analyses based
on sound biological evidences. On the other hand, post-entry follow-up may include
growing enset at closed fields, inspection and cleaning by rouging and disposing
risk .' samples/ plants. In Ethiopia the movement of planting materials is free and
depends on the will of individuals. There is no policy related to internal quarantine
by the regional governments. Hence, some of these principles are difficult to apply.

Accordingly, enset movement in Ethiopia requires a special consideration of

establishing nursery fields at specific regions that serve as origin of clean planting
materials and able to facilitate the inspection and certification procedures at local
level. Healthy seedlings could be obtained by selecting, cleaning, sorting and
destruction of diseased ones. In addition appropriate treatments could be possible to
disinfect farm tools under local nursery arrangement. Individual farmers and/or
cooperatives could be major actors upon receiving appropriate training and
technical/regulatory backstopping by extension personnel. Under this arrangement,
special permits could be given by local officials with advisory services by extension.
These would enhance safe movement of enset in the country and enable the system
to salvage safe planting materials, whenever possible, by restricting and destroying
unsafe enset plants to avert the risk of EBW in the enset and banana industry of the
country.

Pest R isk Analysis (F R A ) Relevant to E B W

In actual practice, on a worldwide scale, issues of inadvertent importation of
potentially hazardous pests into new areas arise in relation to several dimensions
that include biological, economic, political and social scopes (Kahn, 1979). These are
factors determining the entry status of enset planting materials, and subsequent
follow-up in new areas. When only one of these factors, especially biological factor, is
in use to determine the movement status of an item, the activity ought to be based on
pest risk assessment (PRA). This is a thoughtful process whereby the movement
status of enset plant, plant product, package, vehicle, used farm tools, common
carrier, etc. is based on calculated risk of inadvertently introducing EBW with enset
as moved by man (Kahn, 1979; Merid et al., 2008). Consideration of such arrangement
could help to manage difficult diseases like EBW through selecting the most
appropriate step to limit the spread.

PRA has three phases including (i) initiation, (ii) risk assessment, and (iii) risk
management. Risk assessment considers two areas of information that eventually
determine the pest balance of the areas. Pest balance, is the list of pests that are
present in the area of origin minus the list of pests widely distributed in the
importing region. From this, two pest categories including those potentially
requiring phytosanitary measures and those pests excluded from the risk assessment
are determined. Diseases of enset in Ethiopia are listed by Quimio and Mesfin (1996).
 102 The Essence o f Domestic Quarantine against Enset Bacterial Wilt

This information enables us to differentiate pests of quarantine concern to the region.

Listing pests and determining the mode pf transmission and dissemination from
source to destination are important and useful tool to decide import permission or
on the type and level of post-distribution follow-ups. The behavior of EBW raises
concern as quarantine disease for new areas and all phytosanitary measures
described in the previous section for enset movement within Ethiopia are also part of
PRA.

Origin of E B W R isk and Possibility o f Phytosanitary

Measures

Behavior o f the pathogen

Xanthomonas campestris pv. musacearum causes wilting of enset plant in short time
and. completely kills with no chance of recovery. Once enset got inoculated with the
pathogen, no plant would remain from being infected or wilted. Contamination is
the major mechanism by which the pathogen is transmitted from diseased to healthy
plants and spread through regions. It is relatively a short-lived pathogen with
narrow host-range. These types of pathogens art known to be venerable to sanitation
practices and eradication campaigns. Thus, phytosanitary measures against EBW are
important and feasible practices when implemented at country or regional level
considering all hosts (Ensete spp., Musa sp. and Canrui sp.) that could harbor the
pathogen.

E B W Incidence and E n set Ecology

The mean EBW incidence in the different agroecologies of the country Table 1 shows
0.94% in H2, 0.66% in SH2, 1.10% in SH I and 0.42 in M2 agroecology in the country.
EBW was not found in PH I and SM I zones. This clearly shows that EBW is a disease
of mid to high altitude areas with humid and sub-humid climate. According to
Mesfin et al. (2008), incidence was not influenced by agro-ecology and year.
Evaluation of the data in Tablel does not support this generalization. Thus further
studies are necessary to verify about the ecology of the disease. Generally, there are
many uncertainties in the identification of enset diseases due to lack of sound
biological evidences. For instance, leaf streak virus and leaf narrowing, stunting and
chlorotic leaf streaks might be the same disease but at different developmental
stages. In addition, even if Addis et al. (2008) suggest that bacterial wilt spread from
enset to banana grown alongside in Ethiopia and spread to banana in east and
cenlral Africa, no sound biological evidence was shown regarding its pathway.
Therefore detailed studies are necessary to clarify the many doubts that hang around
enset diseases in Ethiopia. Generally, enset regions are distributed in the Southern
region of the country (Figure 1) and it is taken pushed to non-enset regions in the
norih. The wide ecological elasticity makes EBW management difficult without
 103 Dereje Gorfn

strong development arrangement that ensures the use of clean planting materials
and effects sound phytosanitary measures supported by community and extension
services.

Regarding the host conditions, there were some signs of tolerance in clones of enset
dur:ng screening trials for resistance and enset was found to be better tolerance than
banana. Although information on the influence of varietal, age, soil fertility,
moisture, temperature, season, etc. on EBW is lacking, it can easily be seen that enset
pseudostems hold a lot of water in its tissue which is suitable for bacteria growth and
multiplication. This situation may guarantee the pathogen to multiply and infect
enst t plant at wide ecology even if the weather becomes harsh. Any phyosanitary
measures planned against EBW should include all natural hosts of the disease
namely enset (Ensete spp.), banana (Musa spp.) and an ornamental plant (known as
Canna indica) because these plants could serve as primary sources of inoculums.

Table Incidence of enset bacterial wilt during the 1998 and 1999 crop seasons in the southern region of Ethiopia

AEZ Description based on climate Sites studied Incidence (%) Area*

Code (altitude) Mean/maximum (ha)
H2 Tepid to cool humid (mid to high Aletawendo, Bonga, Gazer, Gerese, 0.94/4.20 1844
altitude) Hagereselam, Jinka
SH2 Tepid to cool sub-humid (low to high Agena, Angacha, Gera, Jima, Leku, Limu 0.66/2.90 1688
altitude)
SH1 Hot to warm sub-humid (low to mid Areka, Dedo, Gunchere. Tocha, Waka 1.10/3.80 1684
altitude)
M2 Tepid to cool moist (mid to high Chelelektu, Feshagenet, Solemo, Yirgalem 0.42/1.30 712
altitude)
PH2 Tepid per-humid(low to high altitude) Shewa-Gimira 0.4 372
PH1 Hot to warm per-humid (low to mid Mizan-teferi 0 204
altitude)
SM1 Hot to warm sub-moist (lowland and Gedeo 0 128
________plateau)_________
Source: Mesfin et al. (2008)
 104 ____ The Essence o f Domestic Quarantine against Enset Bacterial Wilt

trm r t w avt u ’vcrt

. .i.- i___________________
Area Under Enset as a
TIG RAY
Percentage of Total Cropped Area
LEG EN D
A rea U n d er E n s e t a t %

None or No Data
A FAR 0 01 • 5 00%
5 01 • 15 00%
■ I 15 0 1 - 3 0 00%
■ 30 01 • 57 00%
■ Lakes
BENISHANGUL □ Regional boundary
GUMUZ

Data source IFPRI and CSA 2005

G A M B E L LA ,
O R O M IA

SOMALI

SNNP,

C o n f u t 'd b y
Dwtiafca N u i i s s w

» w t WflrOE turvn
Figure 1. Enset growing regions and distribution in Ethiopia

Human Activity

Seedlin g propagation
Enset is commenly propagated by vegetative means in all places and rarely from
seed. During vegetative propagation, apical buds are removed from corms of two to
six years-old enset by cutting the pseudostem, boring the apical meristem and
sc rapping/removing old roots. Sometimes wilted shoot may be removed and the
corms of these diseased plants are used as propagating material. These corms
(healthy or diseased) are planted in a hole within enset field whether it is clean or
infected soil. Fifty to 60, some times over 100, suckers emerge from this corm in two
months. These manipulations involving cutting, boring, tiring and wounding
somehow have direct or indirect influence in bringing the pathogen together with
the host so that infection could take place, which affects the quality and health of
enset seedlings.

A d u lt plant m anagem ent

Owners of enset attend their field from establishment to harvest almost every day for *
trcinsplanting, loosening the soil, placing cow-dung/trash from cleaning, cutting
 105 Dercje Goifu

leaves for animals, food preparation or marketing, and finally for processing and
fermenting Kocho. Enset fields are not uniform and hence there could not be a break
crop in all enset cultures. During enset development period a repetitive transplanting
takes place particularly in Gurage areas. At the early stages, big seedlings are planted
individually while weak ones are planted in group. These operations, like that of
propagation, involve many wounding and smearing conditions where several tools
and many enset plants are involved that may again increase the chance of bacterial
wilt to initiate. Borrowing of tools and exchange of planting materials may increase
the chance. In the same way, phytosanitary measures should consider these
conditions and apply good agricultural practices that are eventually supported by
code of conducts at local level and harmonized with "Kebele Rules".

t tilix a tio n o f enset parts in kocho preparation and packing

Pseudostems and corms of enset are chopped into small pieces and rolled by enset
leaves and put into kocho and bulla fermenting pits. The fermenting dough is cut
and mixed every 5-7 days for many times before getting ready for use. No evidence
is available whether infected enset are mixed with or could still carry pathogenic
4 bacteria after fermenting. If kocho dough carries the pathogen, all tools and utensil
used to process uncooked Kocho and Bulla would also become important in
spreading the disease. True stem (generative organ), bract, flower, fruits and seeds,
might not be important in the epidemiology of EBW as they are not used much in the
propagation of enset. Enset leaves are also used to pack many foods, dairy and
vegetable products for market. Enset leaves as packing material could again become
an important and critical source for the spread of EBW.

M ovem ent o f e n se t
Enset movement in the country (mainly to Amhara and Tigray regions, and probably
to other drought prone areas) takes place for several purposes that include:
technology transfer (local or selected clones by research), germplasm for research
and for aesthetical values as ornamental plants. Fresh enset parts (particularly
leaves) arc used as packing materials while processed Kock and Bulla move as food
materials for marketing. Any of these materials can come from diseased enset or
even could be contaminated by tools and handling and carry the pathogen from its
origin to their destinations and become sources of primary inoculums. Taking into
account the current experience and that of banana wilt caused by the same pathogen
in east Africa, once the pathogen enters an area it is difficult to eradicate. Therefore,
lim ting its distribution through strict local phytosanitary measures and strategic
enset movement scheme becomes univocally J essential and feasible when considering c/
» the experiences of domestic quarantine in other countries (PPS, 2008).
 106 The Essence o f Domestic Quarantine against Enset Bacterial Wilt 4

Possible Intervention Strategy

IOnset movement and dom estic q u aran tin e

Strategic approaches are necessary to combat FBW. These include performing sound
planning for movement of enset planting materials within the country, and to layout
and provide appropriate working procedures and protocols for the local
phytosanitory measures. This would be possible by implementing fully or partially
the following tactical interventions for strategic approaches.

G e n e ra l D evelopm ent Arrangem en t

• Define enset growing regions into distinct agro-ecology where the varieties (clones),
climate, production and processing methods are distinct.
• Establish enset-nursery-center (ENC) that house all enset clones found in that specific
agro-ecology to serve as a source of planting materials. Collect germplasm from
similar agro-ecologies to enrich the ENC with local clones.
• Use only these nurseries as sources of planting materials for future enset distribution 4
within the country.
• Develop good enset propagation and production practices (GEP) that is compatible to
local practices and tools. Impose the GF.P to locally endorsed code of conducts *
(Kebele Rules) for enset propagation, production and marketing. Use this Kebele
Rules strictly as parts of local administration and extension services.

Inspection and C e rtific a tio n at Knsct \ur.scr> C e n te r (ISN'C1)

• Inspect the planting materials and propagating practices followed at ENC in order to
provide healthy propagating materials to farmers
• Certify enset seedlings from ENC by local extension and administrative services. A
sim ple method to determine infection of enset by wilt pathogen is to uproot seedlings
and cover them with white plastic sheet for two days to enhance bacterial growth in
the plant. Then, cut the upper part of the seedling to observe bacterial presence. After
half an hour, yellow bacterial ooze comes from conducting tissues which confirm
infection. Infected plants are buried with all the cut tissues to keep the ENC clean
• If the destination is Northern Ethiopia, examine at quarantine check posts of "A bay"
and "Tarm aber". Unless certified, the enset seedlings should not pass the ch eck
posts.

ICradicatiou P ro g ra m s o f I0IJW
» EBW should be designated as a pest to be regulated by domestic quarantine that
requires eradication programs at national level
<» Mobilize enset farmers and development partners for general campaign against EBW
in enset region using simple decision and motivation tools (technical manuals,
leaflets, posters, mass-media, etc) »
«• Major activities may include training farmers, inspection of fields, destroying and
burying of diseased hosts (enset, banana and Canna ), and endorsing GEP for local
uses. Phytosanitary measures described in the next section provides adequate1 *
operation protocol that need to be in the technical manuals used in the campaign
 107 Derejc Gorfu

• Maintain clean enset and banana farms by ensuring GEP with the help of extension
services
« Monitor and evaluate results of eradication programs and document the success

P h jio s a n ita r v M easu res Relevant to Local Conditions

« Use clean propagating and planting materials
« Remove, destroy, compost or deeply bury infected enset plants at soonest possible
time. Enset like banana can be readily killed by 2,4D and Glyphosate herbicides
(Okurut et al., 2006). Discourage movement of fresh enset materials in the field or
between fields as it can effectively contaminate healthy enset plants or fields
• Disinfect farm tools (knives, machetes, spades, hoes, diggers, etc.) using flame or
chemicals (Clorox = berekina) after being used
• Isolate enset fields by fence to control entrance of domestic and wild animals
• Avoid repetitive transplanting of enset, in other words rotating enset from hole to
hole every season need to be avoided (this practice is common in Gurage zone)
• Distribute only very few (if possible only one) planting material. During corm
movement, especial precautions should be followed on sanitation and disinfection
« Use rotation for at least one season or keep the hole open during the dry season as
the life of the pathogen is short in soil
« Minimize exchange of farm tools with other farmers
« Enhance awareness and educate, especially women, on the transmission and control
of the disease.

GARS/ Challenges
« Lack of data for mapping the distribution of EBW that may facilitate the combat
against the disease
« Lack of quantitative data on occurrence (seasonal appearance of wilt), prevalence
(proportion of enset fields showing wilt) and incidence (proportion of wilted plants
in a field) that understanding the impact of wilt based on actual and potential loss
data is actually meager
• The relationships of wilt incidence to soil types and characters, altitude and
temperature variation, drought incidence, rainfall patter bacteria essentially require
abundant free water, age and variety of enset are lacking to formulate sound
management strategy against EBW
• Scientific evidence on whether enset is really drought resistant/tolerant or not is
lacking. The notion of enset as drought resistance/tolerant appears to be unlikely
because all enset growing areas have enough rain (w'ater) in the year and enset
carries a lot of water in its tissue. Further in Abiro (a moisture stress area in south
Gonder), only about 60 kg of enset was obtained in 6 years by sufficiently caring
generating the doubt on the drought tolerance nature of enset (personal observation).
• Contamination as key factor in EBW development was not wrell understood by
majority of enset growers, but the importance of the disease w'as well known. Thus,
operation at community level and implicating subsequent phytosanitary measures at
grass route level are challenges to the extension system as this demands strategic
arrangem ent to avail clean planting materials at zonal level.
• There is a lack of awareness among enset growing community on the importance of
contamination as primary factor of wilt in enset.
 108 The Essencc o f Domestic Quarantine against Enset Bacterial Wilt

Acknowledgements
The author would like to thank Ato Dereje Nigussie (EIAR) for generating enset
distribution map, Dr Eshetu Ahmed (EIAR) for discussions on EBW management
anil reference materials, and Mr. Endale Gudeta (HARC) for useful discussions on
enset propagation and production in Southern Ethiopia

References
Addis, T., L.F.Turyagyenda, T.Alemu, E.Karamura and G.Blomme. 2008. Transmission of
Xanthomonas vasicola pv. Musacearum to banana and enset through garden tools in
Ethiopia (available at http:/www.e-conference.elevva.org/agriculture on 05 July 2010)
Aritua, V., N.Parkinson, R.Thwaites, J.V.Keeney, D R.Jones, W .Tushem ereirwe and J.Crozier.
2010. Characterization of the Xanthonionas sp. causing wilt of enset and banana and its
proposed reclassification as a strain of X. vasicola (available at Correspondence to H-
mail: julian.j.sm ith@csl.gov.uk)
CSA (Central Statistical Authority, Ethiopia). lc>97. Sample \survey for 1996/97 seasons
CSA, Addis Ababa.
Dereje Ashagari. 1985. Studies on the bacterial wilt of enset (Ensete ventricosum) and prospects
for its control. Ethiopian Journal o f Agricultural Sciences 7: 1-14.
Dereje Gorfu. 2006. Plant Quarantine for Agricultural Research. Technical Manual. EIAR. Pp
20
Jones, D.R., A.C.Hayward and J.E.Thomas. 2007. Committee on comm on names of plant
diseases: banana diseases (England not published).
Kahn, R.P. 1979. A concept of pest risk analysis. EPPO Bulletin 9: 119-130.
Merid, K., Dereje G. and Adane A. 2008. Status and prospects of plant quarantine in Ethiopia.
Pages 563-580. In: Proceedings of the 14th Annual Conference of the PPSE. 19-22 Dec
2006, Addis Ababa, Ethiopia. PPSE and EIAR.
Mesfin T. and Gebrem edhin WG. 2008. Enset: Introduction. Pages 155 and 156. In: Root and
Tuber Crops: the untapped resource, Gebremedhin WG, Endale G and Berga L. (eds)
EIAR, Addis Ababa.
Mesfin T.,Gizachew W/M. and Kidist B. 2008. Enset Diseases management. Pages 209-224. In:
Root and tuber crops: the untapped resources. Gebremedhin WG, Endale G. and Berga
L. (eds), EIAR. Addis Ababa, Ethiopia
Okurut, A.W., W .K.Tushemereirwe, A.Aritua, and P.E.Ragama. 2006. Use of herbicide for
control of banana bacterial wilt in Uganda. Africa Crop Science journal 14: 143-149
Quimio A.J. and Mesfin T. 1996. Diseases of enset. Pages 188-203. In: Enset-based sustainable
agriculture in Ethiopia. Tsedeke A., Hiebsch, C., Brandt, S.A. and Seifu GM (eds), EIAR.
Addis Ababa, Ethiopia
PPS: 2008. (Available at Hppt//:www.pps.go.jp/English/jobs/index.html on 24Aug 2010)
Reeres, R.H., O.Opolot, J.B.Muhinyuza, A.Aritua, J.Crozier and J.Smith. 2007. Presence of
banana bacterial wilt (Xanthomonas campestris pv. musacearum) in Ruwanda. Neiv Disease
Reports 14: 52-54.
Ssekiw'oko, F., H.K.Taligoda, and W .K.Tushemereirwe. 2006. Xanthomonas campestris pv.
Musacearunv. host range in Uganda. African Crop Science Journal 14:111-120.
 109 Dcrcjc Gorfu

W elde-Michael, G., K.Bobosha, T.Addis, G.Blomme and T.Mengesha. 2008. Mechanical

« transmission and survival of bacterial wilt of enset. African Crop Science Journal 16: 97-
102.
Yirgou D. and J.F. Bradbury. 1968. Bacterial wilt of Enset (Ensete veiitricosnm) incited by
Xnnilhviionns musacearum sp. PJn/topathologi/ 58:111-112.
Yirgou D. and J.F. Bradbury. 1974. A note on wilt of banana caused by the linset wilt
organism Xanthomonas niusncenrimi. East African Agricultural ami Forestn/ Journal
40(1):111-114.

*
 Mole Rat and Porcupine as Production
Threat and their Management Options
LeiUseged Begasliaw1, W a»a M azcngia'and M ikias Yesliitela*
1 Food for the H ungry Ethiopia, Shnshogo, Hadya, SN N PR. E-m ail: lbegashnw 2005@ yahoo.com
Hazuassa A gricu ltural Research Center, Hazuassa, SN N PR, E-m ail: zuaga966tvyahoo.com
3 A reka A gricu ltural Research Center, A reka, W olaita, SN N PR, E-m ail: m ikiasyes& yalioo.com

A b s tra c t
A m on g vertebrate pests, porcupine an d m ole rat are the m ost im portant and dam age a great
deal o f crops. P orcupine and m ole rat eat m ainly sw eet potato, yam , enset, potato anti
cassava in C u n u n o w atershed, W olaita zone, SN N PRs. A frica H ighlands Initiative (AH l)
p articipatory problem identification and prioritization, fa rm ers in the w atershed put
fo n v a rd s porcupine and mole rat as the most im portant vertebrate pests and brings
trem endous crop losses. Collective action fo r the control o f porcupine determ ine effective
control m ethods, identify effective form o f collective action institutions m ake use o f local
policies and social negotiation were the main objectives o f the study. P articipatory m ole rat
control zoas aim ed at to identify the effective traditional m ole rat control m ethods fo r fu rth er
use in collective action. This paper explains that different steps w ere fo llo w ed before
m obilizing the collective action. D ifferent tier discussions and negotiation zoas m ade from
key-inform ant level to kebele level. The collective action utilize the traditional control
m ethods such as deep ditch digging (3-4m deep), circular ditch digging around graveyard
and zoire trap zoere used. Zinc phosphide (RA TO L™ ) as chem ical control zoas used in
com bination w ith the first tzoo traditional control methods. In order to control the
ruatershed-trial site, three additional adjacent Kebeles zoere added as a buffer zone, since
porcupine travel 14.8 k over night, fo r m ole rat control three W olaita traditional m ole rate
:on trol zoere com pared zoith D itta/chencha trapping system. D evelopm ental G roup (DG)
uas identified as the m ost effective fo rm o f co!lectiz>e action institutions that able to enforce
the by-law s for m obilizing the collective action against porcupine. The m odified deep ditch
diggin g zoith RA TO L zoas the m ost effective control m ethod in most porcupine inches
'xcept in graveyard, stony areas and underneath forest niches. C ircular ditch digging zoith
ZATOL zoas effective in graveyard rather than using the traditional control m ethod alone.
W ire trap zoas foun d effective in ston y and underneath fo rest niches. Frequent social
negotiation and support zoere m ade betiueen different interest group like least vs. most
iffected com m unities by porcupine, safety-N et vs. non-safety net fa n n ers that helped a
great deal for successful collective action at G ununo zuatershed. In participatory m ole rate
control, farm ers selected D itta/C huncha and W olaita type-II as V and 21"1, respectively,
and decided to use in the collective action at zuatershed level. The approach follozued in the
collective action against porcupine can be scaled out and replicated to other areas having
sim ilar problem . Institutionalizing the collectiz’e action zuill be decisive steps fo r sustainable
vertebrate pest control zuhen scaling out by any GO s and N G O s that zuill en gage the sam e
activities.
 Ill Leulscged Bcgashaw et. al.

Introduction

Every year, crop and animal pests deprive farmers off significant parts of their
production. Some estimates suggest that 10-40 percent of the world's gross
agricultural production is destroyed by agricultural pests. These pests include a huge
variety of different organisms that include rodents, insects, mites, worms, birds,
fungi, bacteria, viruses and virus-like organisms and weeds. Crested porcupine
(Hystrix cristatn L.) is a rodent, which causes a serious problem to farmers in many
parts of East Africa including Ethiopia. The crested porcupine is the largest and
heaviest of African rodent. It is one of the three species of East African porcupines. It
is the most widespread one but followed by South African porcupine (Hystrix
amcaenustralis). The smaller, more slender African brush-tailed porcupine (Atherunis
afiicam is) is found in isolated populations in Northwestern Kenya (Begashaw et al.,
2007). In Southern Ethiopia including Gununo watershed, however, crested
porcupine is the commonest resident. It belongs to phylum cordata, class mamalia,
order rodentia and to the family Hystricide. It is easily recognized by its most
notable feature-its quill, the stiff, thick spines banded wit black, brown, pale yellow
or white covering, side and tail, and they are mix in with soft hairs.
Porcupines inhabit holes made by other animals but also dig their own. They seem to
prefer moving along tracks or roads, travel up to 14.5km at night in search of food.
Three adjust Peasant Associations (PAs) of Gununo watershed were considered to
serve as a buffer zone for this study. Porcupines eat primarily maize cobs, followed
by roots of sweet potato, leaves of cabbage, roots and tubers of yams, potato, cassava,
and seeds of field pea. Graveyards, underneath trees and grassland are the most
common types of habitat of porcupine in Gununo watershed though they prefer to
livc» mostly in graveyards. This is because porcupines chew on bones for calcium and
other minerals albeit they are vegetarians. Farmers' traditional knowledge
assessment indicates that there are three different porcupine control measures. These
are wire-body trap, fumigation of caves with farm straws and pepper mix; deep
ditch digging (3-4m deep) at outlet of the porcupine cave, circular ditch digging
around local graveyards, digging and destructing the tunnels to the cave, sound
frightening material called " g ir a f, Amharic during patrolling of porcupine at night..
These different traditional control methods are used in different niches of porcupine
under different situations. In Gununo watershed, graveyard, grassland, underneath
forest area and following riverside are the major porcupine niches identified in the
study. Rodenticide like zinc-phosphide (RATOL™) is a registered chemical in
Ethiopia to control rodents. It was used in combination with the fore mentioned
traditional control measure pasted with maize cob or sweet potato roots.

In Gununo watershed, farmers have been claiming that mole rat (Spalnx spp.) is also
an important constraint to enset, maize, elephant grass, and cassava and coffee
seedlings (AHI, survey on needs and problems assessments,, unpublished, 2004).
 112 M o le R a t a n d P orcu p in e a s P ro d u ction T h rea t a n d th eii M a n a g em en t O p tion s

Mole rats eat the roots of enset, sweet potato, wheat and barley, reducing the
production of these crops significantly. They also eat grass and the soft part of any
plar t if there is no crop in the field. Mole rat enters a new field, start burrowing and
throw up soil forming molehills. It continues forming tunnels while living there and
traveling through them in search of food. According to the farmers, the living
quarters of mole rats consist of four compartments linked by tunnels. These are: a
sleeping area, a food storage area, a resting area, and a west accumulation area. The
tunnels are zigzag, narrow and deep. Some farmers attempted to control mole rats
through fumigating their holes, pouring water into the holes and using traps. There
appeared to be a fair amount of indigenous knowledge and experience offering room
for urther research on the trap. It operates on the principle of a spring mechanism. It
varies in the type of material used in making the noose or hooks. Some farmers use
potato or ginger, to attract the rat toward the trap. Farmers at Gununo watershed,
util ze a metal hook trap with different attractants and food baits.

Border of farms covered with grasses, and non-arable areas located between farms
are described as potential habitable areas for mole rates. Informal assessment of
farmers' copping strategies of these vertebrate pests showed that most farmers are e
lacking the techniques and alternatives for addressing the problem. However, some
farmers who were equipped with skills and techniques used for working and
manipulating of the trap. Digging and destroying the mole tunnels following the
feeding runways were also investigated to be an alternative mole management
option, particularly by those farmers who are not familiar with local trap.

Acquiring of trap shyness by the mole rates was raised and discussed among farmers
as a drawback for using of the local trap. Some farmers worked out the reason and
their experience with foreign odder induced from human being would be left on the
trap, while working with the trap on the field. They speculated that such foreign
odder could be the cause for orienting the mole rate to divert its route away from the
direction where the trap is placed. Furthermore, farmers' indigenous innovation of
rubbing hands with the soil obtained at mole mounds and applied during setting up
of the trap was used to quite acquiring of trap shyness. Food baits prepared with
sweet potato tuber or leaf, banana, or maize attached with aromatic plants (’Besobila1
/' Koseret'), which induced attractive flavor for attracting the mole rate to the
targeted trap.

In mole rate control, farmers' indigenous knowledge and innovations play significant
roie in the process of technology development and transfer. First, the best traditional
control methods should be determined in selected experimental farmers in Gununo
watershed. Hence, with the view to confirm and enhance their technical feasibility
and efficiency to test and assess their performances in collective actions at
community level for further use and utilization in mole rat management at
watershed level.
 113 Leulsefted Begashaiv et. al.

Porcupines and mole rat do not respect farm boundaries. Therefore, farmers who
control porcupine and mole rat niches on their own fields might still face damage on
their crops caused by porcupine coming from neighbors fields or other villages
where no control measure are taken. In that sense, crop damage caused by porcupine
constitutes a trans-boundary natural resource management problem, which, in
addition to technical interventions, requires organizational interventions to ensure a
coordinated effort among farmers. Collective action for natural resource
management that include joint investment for labor in pest management and
introducing technologies; setting and implementing rules to exploit a resource is
unequivocal for such pests. Trans-boundaries pest such as porcupine and mole rate
needs an effective collective action. Collective action is often viewed rather
uncritically as synonymous with social structures or formal organizations (Knox and
Meinzen-Dick, 2000). There are many definitions for collective action processes that
depend on its action or function in diverse biophysical socio-economic issues. In this
study a collective action is defined as direct actions carried out by group of people
working towards common goals (Lubell et al., 2002; Swallow et nl., 2001; Tanner,
1995). This may range from two neighboring resource user managing a common
boundary to a widespread social movement. Collective action tends to link property
right and common resource governances. Hence, to conduct an effective collective
action, there is a need to develop by-laws, bv the community, which involved in the
collective action, that govern the agreed upon action. Porcupine control at G ununo
watershed should utilize collective action to make use of by-laws (local policies),
effective forms of collective action institution (Developmental Group) and social
negotiation among many interested groups.

Achievement
Methodology
T ra d itio n a l knowledge assessm ent in the con trol o f mole ra t
In different localities, there exists diverse indigenous technical knowledge (ITK) for
the control of mole rat. Farmers at Gununo watershed, use commonly a stick-spring
with iron hooks, made by local artisans. The hook is tied with Eucalyptus sticks (that
serve as a spring for the trap system) with enset fiber rope (Fig. 1). Different types of
food bits (sweet potato roots, banana, etc) and attractants like "k osert" or " besobela"
are used to trap mole rat. Other traditional mole rat controls methods used at the
watershed include dig and destruct the tunnel following the tunnels or adding water
into the hole. Based on FARM Africa's experience on indigenous control methods in
participatory experiment was carried out on two traditional trap methods viz.,
Gessa/Boreda and Ditta/Chencha. Their findings indicated that Ditta/Chencha
were most preferred by the farmers and hence used in this experiment. Therefore, in
this study, the following traditional control methods were utilized and demonstrated
to farmers in Gununo watershed (Table 1). The experiment was carried out at four
4 114 M o le R a t a n d P orcu p in e a s P rod u ction T h rea t a n d th eir M an a g em en t O p tio n s

4 villages' watershed (Gegecho, Lay-busha, Tach-Busha and Offa) as a model and each
one of them utilizing single traditional control method. A visit across the villages
was made to share experiences and to decide on the most effective control methods
for a collective action that is preferred by the farmers.

Fig.1. Wolaita traditional mole rat trapping system (both type-l & type-ll that the hook is pasted with soil from mole rat hill)

In each village where infestation was high and of similar level, a farmer field was
selected to test the control measures. After a month, farmers visited the
demonstration field and discussed with the farmers that conduct the experiment. The
farmers evaluated the four traditional mole rat control methods based on the criteria
set up by the farmers themselves. Farmers' criteria were: time taken to set up the trap
and its simplicity, availability of material for making the trap (cheep source and took
less material), sensitivity (ability to trap instantly) and effectiveness (ability to
kill /catch the rat). For each ITK, farmers' evaluation was based on a 1-4 scale scoring
methods (1 = bad, 2= enough, 3 = good and 4 = excellent). The farmers in a group
evaluated the methods and based on final consensus ranked the four mole rat control
f methods accordingly. Farmers were given 25 field pea seeds and distributed tire
seeds based on the different values/weighs they gave for each criterion using PR A
tools (Table 2).
l
_ 115 L eu lseg ed B eg a sh a iv et. al.
»

Table 1. Description of traditional control methods used in the study

«
Traditional control methods Description of the methods/treatments
Wolaita trap type-l The trapping system consists of iron hook (that catch the mole rat), Eucalyptus stick (as
spring), fiber rope (made of enset fiber), attractants (“koserf or “besobela"), and food bite
(sweet potato leaves/tubers or banana fruit).
Wolaita trap type -II This trapping system is same as Wolaita -I in terms of the iron hooks, attractants and food
bait. The modification made on iron hooks are rubbed or pasted with soil from mole nill
itself. This was based on some elite farmers’ experiences at the watershed. Farmers
reported that strange orders from the iron hook and the person hand that set the trap cause
the mole rat to escape in Type -I.
Dita/Chencha trap system The trap is made of bamboo stick and string (“Katcha") made from enset fiber. The string
system has bamboo sticks (2-4 year old bamboo tree), loop and additional string system
(made of one year old bamboo tree). The trap is set up on very recently made mole rat soil
hill. The soil hill is removed, the tunnel outlet is closed with grasses material until the
trapping set up is finished. The outlet then opened after finishing the set up so that mole
rate sense the opening and bring soil by moving backwards towards the outlet. This is the
time that mole rate will be trapped. The mole rate is trapped here without attractants and
food bait like the Wolaita type. This system is based on the biology of the mole rat that it
does not allow an opening along the tunnels and hence it is trapped while trying to fill the 4
opening with soil.
Digging and destructing The digging follows the tunnels/pathways of the mole rat that lead to its resident area.
following mole rat tunnel Skilled person knows where to start the digging and the recent pathways that lead to mole
rat area in the tunnel. This is labor intensive and is done only during wet and off season ^
when there is no crop stand. Sometime digging takes much time and long distance, so it
depends upon the knowledge of the farmers using this system. However, if it is dug by
skilled person, then it is very effective._____________________________________________

Tabe 2. Farmers’ criterion value based on PRA tools

Farmers criteria Value/weigh (n=25)

Timu taken to set up the trap 8
Availability of materials and take less material to make the trap 4
Sen sitivity - trap instantaneously 3
Effectiveness - ability to kill______________________________ 10

After participatory evaluation of the different indigenous knowledge, one or two

effective traditional control methods will be further utilized for collective action at
watershed level.

Porcupine Control
The major steps to engaging on mobilizing the collective action are as follow:

Ivocul knowledge assessm ent

First, the team discuss with Farmers Research Group (FRG) leaders to identify key
informants who should be invited to elicit the views of men, women and diverse
groups. Invite also influential people to set the agenda together with the informants
for awareness creation. Then, this was followed by interviewing key knowledgeable
informants about different control methods. Second, community consensus was
reached on porcupine eradication. Hence, as per same agenda, further assessment of
 116 M o le R a t an d P orcu p in e a s P rod u ction T h reat an d th eir M an ag em en t O p tion s

traditional control methods in addition to the knowledge from key informants and
the need for formulation of by-laws, niche identification at village level and finally at
kebek level was discussed.

Id en tification o f effective form s o f collective action

The different local institutions such as local administration team (.niengistnwe budcn),
debo, edir, developmental group (DU), etc. assessed the effective forms of collective
action. Then, consensus was reached at different tier of discussions with the farming
community based on their ability to mobilize the collective action and enforce the
formulated by-laws. The selected local institutions were also given responsibilities to
engage in mobilizing the community and collect data using data collecting sheet
(established reporting system) on number of porcupine killed/caught, methods of
control, type of niches, etc using data collecting sheets.

♦ T ra in in g s
Capacity building on traditional porcupine wire-body-trapping system that was
concealed before (since skilled farmers charges each farmers with 25 birr/a
4 . porcupine catch) will be given by skilled farmers to the community at four rural
kebeles. The research team also will give orientation on the selected porcupine
control method to be used and its suitability to a specific porcupine niche so that
different types of porcupine control methods will be allocated for each type of niche.

Develop ancl endorse the by-laws.

The by-laws was formulated and endorsed by the community to govern the
experiment so that all farmers participate and conduct periodic PM&E to monitor: a)
implementation of agreed-upon action; b) the research process being implemented; c)
the effectiveness of each treatment through a meeting with the community where
research results are shared back.

M obilizing th e collective action and im plem entation o f the research

st rat cases/plan
Mega-phone and traditional horn instrument will be utilized for the announcement
of the suitable type of porcupine control methods for each specific niche and the
effective dates to start collective action for same. Mega-phone was used across all
villages singing local songs that encourage and motivate farmers to endorse the
collective action so that the selected effective porcupine control methods be effected

f O u tcom e A ssessm ent

Farmers presented many traditional methods for the control of porcupine.
Accordingly three methods were considered to be most effective: a) deeply dug pits
i at the outlet of a porcupine cave; b) circular ditches around graveyards, and c) a wire
trap system. A forth chemical treatment, Zinc phosphide (RATOL™), was also used
in combination with the first two methods as two additional treatments. Farmers
 117 L eu lseg ed B eg a sh a iv et. a I.

modified the first method of deep digging (3-4 m depth) to more shallow pit (1 to
1.5m deep) in combination with Zinc phosphide. Different porcupine control
methods were selected based on their suitability to different niches within each
Developmental Group's site. Mobilizing the collective action for the aforementioned
control methods were applied during the season when porcupines are most harmful
to crops.

The research involved collective action across all sub-Kebeles under each
Developmental Group (DG). Farmers selected DG units of collective action because
they have the ability to enforce local by-laws in support of collective action and
where only 25 to 30 households-may easily manage it and monitor activities during
implementation. During the campaign, each DG assigned one to two days per week
for collective action only to control porcupine in the watershed. It was further
decided that the Rural Kebele magistrate court and local leaders to follow up the by­
laws enforcement during the collective action.

i
 118 M ol ’ R at a n d P orcu pin e a s P rod u ctio n T h reat a n d th eir M an ag em en t O p tion s

Social negotiations were then supported among farmers whose crops are frequently
affected and the least affected households, as well as those farmers who participated
and not participated in the Safety Net Program1! .

Ther the by-laws were formulated through full participation of farmers and
distributed to all Rural Kebele and Village/DG leaders.

The campaign was launched following the establishment of control method,

adm nistrative units and by-laws for the operation of the collective action and
training the relevant individuals on the control methods and data collection
procedures. Farmers travelled on foot and vehicles with mega-phones and local
music thus publicizing the campaign across all DGs, villages and Rural Kebeles.
Following the campaign, records were taken by DG leaders on the number of
porcupine caught or killed by different farmers, villages, niches and the control
methods.

The collective action for the control of porcupine was conducted at Gununo
watershed as trial site and other three adjacent rural kebeles that served as buffer
zones and the surrounding watersheds so that re-contamination from the nearby
villages porcupine is protected (Fig. 2).

However, the impact/outcome study was conducted only for the Gununo watershed
trial site. Since the collective action covered all villages of the watershed, amongst
other methods viz., with and without, before and after, etc,. Before and after impact
assessment method was best suited and used for further impact assessment study.
A fte' mobilizing the collective action, then impact assessment was conducted using a
structured questionnaire,'which includes information such as number of porcupine
niches and visited farmers' plot, crop yield losses encountered per visit, frequency of
porcupine visit per week, etc. To study the overall impacts at watershed level,
focused-group discussion was made in four villages. Whereas household survey was
used for least and most affected villages by porcupine. From least and most affected
villages, Gegecho and Offa were selected, respectively for the former and latter
villages based on their level of porcupine infestation. The data that were generated,
were subjected for analysis using simple descriptive statistics.

1 The Safety Net Program is a governmental program designed to help low income farmers by paying them to
carry out, developmental activities (construction o f schools, offices, health centers , road maintenances, and so on)
for rural kebeles. Some non-participating farmers are uncooperative in the collective activities, arguing that Safety'
Net farmers must collaborate since they are paid to do so by the government. However, negotiations lead to the
joint conclusion that porcupine are a problem for both parties and affect each group equally, requiring joint efforts
by both groups
 119 L e u h e g e d B eg a sh a w et. a l.

Results

Mole rate control

The result of farmers' evaluations on on-farm experiment conducted with indigenous
mole rat control methods at Gununo watershed showed that Ditta/Chencha trap
was preferred than the rest with a total score X weight values of 84. It was also
preferred because of its high sensitivity and high effectiveness in capturing the mole
ra: compared to the other three methods. Both Wolaita-II (Fig. 3) and Ditta/Chencha
were equally valued (24) for simplicity to set up the trap. However, Ditta/Chuncha
traping system was valued the least by the discussant farmers in terms of
available/cheep source of material to make the trap. In general, the evaluation
conducted and the judgment made were subjective, the Chencha/Ditta trap was
preferred because of its sensitivity and efficiency in capturing the mole rat (Table 3).

Figure 3. Mole rat trapped by Wolaita-Type-ll trapping system

 120 1 M ole R a t a n d P orcu p in e a s P ro d u ction T h reat a n d th eir M a n a g em en t O p tio n s

Table 3. Farmers evaluation of different traditional mole rat control methods

Value (score x weight) of traditional mole rat control
No Farmers evaluation criteria Digging and
Wolaita-I Wolaita-I I Ditta/Chuncha
destructing
Time to set up the
1 2x8=16 3 x 8= 24 3 x 8= 24 1 x 8= 8
trap/simplicity
2 Availability of material 3 x 4 = 12 3x4=12 2x4=8 3x4=12
3 Sensitivity 2x3=6 2x3=6 4x3=12 1 x 3= 3
4 Effectiveness 2 x 10= 20 3 x 1 0= 30 4 x 1 0= 40 1 x 10=10
Sum total value 54 72 84 33
Rank 3rd 2nd 1st 4th
Deperding on the choice of farmers, either the 1st or 2nd ranked traditional mole rat control methods can be used for collective
action at watershed level or any type of administrative boundaries.

Figure 4. Porcupine trapped with 1.5m deep ditch digging + RATOL™

Porcupine control
After mobilizing the collective action, then impact assessment was conducted using
the before-after methods and structured questionnaire was prepared for focused-
group discussion and household-household survey for each village and for
least/most affected village, respectively. Collective action for the control of
porcupine resulted in a significant impacts in reducing crops losses, reducing the
time spent and labor required in guarding porcupine at night, improve the health of
the community and change social attitude and behaviors. Gununo watershed, more
than one thousand porcupines were controlled through collective action, social
negotiations and by-laws implementation.
 121 M ole R a t a n d P o rcu p in e as P ro d u ctio n T h rea t a n d th eir M a n a g em en t O p tion s

Table 4. Number of porcupine killed/caught using different control methods in Gununo watershed

No. of porcupine Total No. porcupine

Village Niches Control methods killed/caught controlled
Grassland Wire trap 2
Gegecho Under Eucalyptus trees Digging (1.5m) deep at the outlet of porcupine hole + RATOL™ 15
25
Grassland
Under Eucalyptus trees Digging (1.5 m deep) at the outlet of porcupine hole + RATOL™
Lay Busha River basin 25 25
Grassland
Graveyard RATOL™
18
Tach Busha Under Eucalyptus trees 39
Grassland Digging (1.5 m deep) at the outlet of porcupine hole + RATOL™ 21
Under Eucalyptus trees Digging (1.5 m deep) at the outlet of porcupine hole + RATOL™ 23
Chere 42
Graveyard Circular ditch digging + RATOL™ 19
Under Eucalyptus trees Digging 91.5 m deep) at the outlet of porcupine hole + RATOL™
Under forest
373
Under bamboo tree
Near river side
Graveyard Circular ditch digging + RATOL™ 107
Grassland Deep digging at the outlet of porcupine hole
Offa 88 831
Under Eucalyptus trees
Graveyards Wire trap
Under Eucalyptus trees
90
Under forest
Under bamboo tree
Graveyard RATOL™ alone 173
Graveyard Digging (1.5 m deep) at the outlet of porcupine hole + RATOL™
Under Eucalyptus trees Deep digging at the outlet of porcupine hole
Dembezamine 26 26
Grassland Circular ditch digging + RATOL™
RATOL™ alone
Graveyard RATOL™ alone
The three kebele- Under Eucalyptus trees
6 6
Buffer zones Under forest Grassland

Total number of porcupine controlled trough campaign 984

r
__ 122 I M o le R a t a n d P orcu p in e a s P ro d u ctio n T h rea t an d th eir M a n a g em en t O p tion s

Table 5. Methods applied by niche and their effectiveness

No. of porcupine
Method of control Niche where applied
killed/trapped
Method 1-RATOL™ alone In graveyard 197
Method 2-Circular ditch + RATOL™ Porcupine burrow located near graveyards 126
Mett od 3- 3-4m deep hole at the outlet of All porcupine burrow located away from
88
porcjpine burrow graveyards
Metf od 4-1.5. hole at the outlet of porcupine All porcupine burrows located away from
455
burrow + RATOL™ graveyard
Met! od 5. Wire-body trap At outlet of porcupine burrow 92
Number of porcupine controlled at Gununo Watershed 958

Observed Im pacts from C ollective Action in Porcupine Control

Sweet potato damage

(kg)
Yam damage (Kg) Time spent policing
(hrs/night)
No. Fam ily members
involved/night
4
No. V isit to health
center/year

Indicators □ Before ■ A fter

Fig. 5. Effect of collective action on observed impact of crop yield losses and some social aspect

FATOL™ was tested as a means to reduce the labor associated with deep digging
and to avoid digging in graveyard (Table 5). Different control methods were found
to be effective for different reasons. Shallow holes dug at the outlet of the porcupine
burrow combined with RATOL™ were most effective control method in terms of
r umber of porcupine caught or killed (Table 5 and Fig. 4) and measured changes in
local performance indicator. However, farmers are generally reluctant to use
chemical control methods due to cost. In the absence of RATOL™, deep digging and
wire traps may be used. The most livelihood impacts for farmers were due to
reduced crop damage (80%), improved health and labor savings (Fig. 5). Levels of
crop damages reduce by 80 % following intervention, and frequency of visits to
health clinic from weather-related illness also declined. Yet, one of most important
 123 Leulsegcd B eg ash azv et. a I.

successes in the mind of farmers was the reduction in effort required to guard the
field at night.

Table 6 . Impact of collective action for the control of porcupine on crops losses, level (%) of

Least affected village Most affected village

Indicators (Gegecho) (Offa)
Before After Improved Before After Improved
Swe 3t potato (kg) 7.80 3.75 51.92 50.33 9.50 81.13

Yam (kg) 5.10 2.15 57.84 17.33 8.83 49.04

Potato (kg) N/A N/A N/A 8.67 3.33 61.54

Haricot Bean (kg) 5.50 2.83 48.48 N/A N/A N/A

No. of porcupine niches 4.0 2.50 37.50 10.16 4.67 54.10

Frq. Of visit/week 4.2 1.30 69.05 5.67 2.67 52.94
No. guarding days /week 5.0 0.80 84.00 4.17 0.50 88.0 0
Time spent in guarding per night 3.7 1.05 71.62 4.45 1.00 77.36

In focused-group discussion, a total of 41 farmers whereas in household survey a

total of 20 farmers in least most affected v illages were interviewed using standard
questionnaire. Crops like maize, sweet potato, yam, cabbage, tannia, haricot bean,
cassava, field pea and pumpkin are the major liable ones to porcupine attack.
Whereas taro, enset, wheat, barley and teff are rarely attacked by porcupine. Most
(78.05 %) farmers responded that amongst major crops grown in the watershed,
maize is the most preferred by the pest followed by sweet potato, yam and haricot
bean.
The level of porcupine damage to crops was high (100% all farmers) in most affected-
Offa village before the collective action was put into effect. However, after the
campaign all the farmers' response were put within medium- and low-level ranges
(Fig. 6). In least affected -Gegecho village, before the campaign, farmers' report was
not high and after the campaign, most farmers (70%) report was within the low-level
range (Fig. 6).

Illustration 1. Farm ers witness on porcupine control

The high level o f fa rm er enthusiasm fo r theses outcom es may be sum m arized by
farm ers' testimonies. While eating b reakfast with one fam ily, our host stated, "If there
haa been porcupine infestation in our village, we could have not shared this breakfast
witn you. The porcupine would have finished it." As stated by an other farm er,
"Observing our problem , God brought you to us."
 124 M ole R at and Porcupine as Production Threat and their Management Options

120

Before ; After
I
Least affected M ost affected

G e g e c h o v illa g e O ffa v illa g e

Fig 6. Percentage levels of crops attacked by porcupine before and after collective action in least affected (Gegecho) and
most affected (Offa) villages

Fig. 7. Percentage yield of sweet potato and yam achieve in most and least affected villages after, before collective action in
the watershed.
 125 Lenlscged Begashazv et. a I.

As i result of mobilizing the collective action against porcupine, sweet potato yield
was increased by 81.13 % in Offa -most affected village. Whereas in the least affected-
Gegecho village, yam yield increased by 57.84% (Fig. 7). As shown in Table 7, on
average in the watershed, 75% of porcupine niches were controlled by collective
action. The frequencies of porcupine visit to each farmer's field per week, number of
day? spent for guarding porcupine by each farmer per week and time spent at night
in guarding porcupine by each farmer were significantly reduced by more than 50,
80 and 70 percent, respectively. As a result of theses most farmers were assured that
labor demand in cold night and health condition were improved (Table 8). Farmers
reported that, while guarding during cold nights they were exposed to different
ailments such as pneumonia, TB and malaria fever. Farmers witnessed the overall
impact as stated above in illustration 1.

Table 7. Porcupine niches before and after mobilizing collective action at Gununo watershed

■AWatershed Before After Coverage of collective action on

Village No. niches No. niches No. niches not porcupine niches (%)
controlled controlled
Lay B jsha 5 5 0 100.00
Tachl3usha 9 6 3 66.67
Offa 130 100 30 79.92
Cheru 8 3 5 37.50
Mear 38 28.5 9.5 75.00

Table 8. Impact of collective action against porcupine on frequency of visit per week, number of days and time spent
guarding, respectively per week and per night

Least affected Village Most affected

Reduction Reduction
Indicators Before After (% ) Before After (%)
Frequency of visit/week 4.2 1.3 69.05 5.67 2.67 52.94
No. of days guarding/week 5.0 0.8 84 00 4.17 0.5 88.00
Time spent for guarding /night 3.7 1.1 71.62 4.41 1.00 77.36

Ctaps
Farmers in many places try to control pest by themselves. Individual endeavors can
only work for some pest that occur in certain locality and have not trans-boundary
nature. However, pest that does not respect farm boundaries, in this case porcupines
and mole rat require concerted efforts to work together in a form of collective action.
Successful collective action that can enable mobilizing the farming community in a
given watershed or administrative boundaries seeks effective participation of
institutions, social negotiation and establishment of by-laws that govern the agreed
upon action. Hence, for effective porcupine control, several steps were undertaken.
 126 M ole Rat and Porcupine as Production Threat and tlicir Management Options

These include: social negotiation, awareness creation, training and different Lire of
discussions at key-informant level, village to kebele level to reach common
understanding for subsequent community mobilization for the collective action.
Befcre enforcing any forms of collective action. It is important to start with cheap
and easily adoptable pest control measure in a participatory manner that is tested in
the target community or watershed. In the control of mole rate, farmers at Gununo
watershed, preferred Ditta/Chencha and Wolaita type-I which were ranked 1st and
2iui , respectively at the respective watershed. These effective control methods were
demonstrated and tested at Gununo watershed indicating that aporcupine approach
collective action can be utilized in a mole rat control collective action at watershed
level. Thus, the approach used in porcupine control could be used for any type of
vertebrate pest in any other localities where the pest pose threat for major crops
grown in the area. Therefore, the approach need a scaling up effort particularly in
enset based farming community, where mole rat and porcupine are among the major
threat for enset production in the region . For sustainable control of vertebrate pests,
it is important that collective action be institutionalized so that the existing GOs and
NGOs together work actively in the region. Any development oriented institution
should be unduly responsible to initiate and make institutionalized arrangement for
mobilizing the collective action.

Discussion

Mole rat and porcupine bring tremendous crops damage and have been a threat to
the production of crops particularly in root crops and enset based farming
community. In the control of mole rat farmers preferred that Ditta/Chencha trapping
system is best in terms of sensitivity and effectiveness. The finding of FARM Africa
revealed that Ditta/Chencha trapping system preferred by farmers in the area
because of its simplicity and takes short time to set up the trap than the
Geresse/Boreda trap (FARM Africa, 2004). In general, the trail farmers and other
farmers participating in cross visit demonstration, pointed out that Ditta/Chencha
and Wolaita Type-II are more preferable to use for collective action at Gununo
watershed. The Ditta/Chencha trapping system is so effective because it is based on
the biology of mole rat. The trap is set in the open end of the tunnel. Mole rat does
not allow the tunnel opened. It brings mass of soil moving backward to fill the
opening. Then the mole rat gets trapped and eventually killed while trying to close
the open end of the tunnel. Whereas, the Wolaita Type-II system, which is based on
attractants and food baits, trap the rat while it is in search for food. Type- II was
effective next to Ditta/Chuncha because it was modified by avoiding strange odder
from the iron hooks and persons hand that set up the trap. However, most farmers in
Gununo watershed, complain because using the ordinary Type-I trapping system,
 127 Leulseged Begashaw et. al.

mostly allows mole rat to escape or unable to catch it, since mole rat can easily
identify the odder of the iron and the person who set up the trap.

The control of porcupine makes use of collective action that employs different
traditional and chemical control depending on the different niches of porcupine.
Based on assessment of traditional knowledge and literature reviewed, three most
effective traditional control methods were selected. The modified deep ditch (1.5m
depth with RATOL™ bait) was the most effective and reduced labor spent on
digging up to 4 meter deep furrow. Most porcupines consume the RATOL™ bait set
at :he outlet of the cave because the shallow ditch discourage them to dig and forced
them to eat the food bait pasted with chemical. In graveyard, circular ditch digging
wi±i RATOL™ was effective since porcupines are unable to dig deep ditch in a
graveyard. Wire trap is used in a niche that is difficult to dig or stony area and
underneath forest area. Hence, in the collective action for porcupine control wire
trap was also one of the most effective in the aforementioned niches.

The effective control measure alone can control pests such as porcupine and mole rat
unless it is embarked at larger area coverage like watershed or administrative
boundaries that make use of collective action, social negotiation and local policies or
otherwise. Pests like porcupine and mole rat do not respect farm boundaries that
demand the community work together in a collective action (German et al., 2008). It
is now widely recognized that collective action is a fundamental pillar of landscape
or watershed natural resource management. Different from farm-level management,
collective action require to regulate the rights and responsibilities of common
property-resources and public goods (Gebremedhin et al., 2002; Munk Ravnbourg
and Ashby 1996; Ostrom 1990; Scott et al., 2001), to manage biophysical processes
tha: do not respect farm boundaries (Munk Ravnborg et al., 2000) to negotiate joint
investments and technological innovations for enhanced productivity and to regulate
benefit capture (Meinzen-Dick et al., 2002). Porcupine and mole rat control calls for
individual farmers and resource users to come together to identify how agricultural
productivity and livelihood more generally might benefited from collective action
over individual action and to negotiate rules and regulations to govern such
innovation.

In enset based farming system, these two vertebrate pests are threats for the
production and productivities of the crop. At Gununo watershed, enset is less
affected by porcupine because other root crops such as sweet potato, yam, taro,
tanriia, etc are available throughout the year. The approaches utilized to control
porcupine in a collective action can be duplicable to other enset based farming area.
Empowering institution to enforce the by-laws and institutionalize the collective
action for sustainable porcupine and mole rat management the task remained need
to be scaling up for other enset growing region.
 128 M ole Rat and Porcupine as Production Threat and their Management Options

* R e co m m e n d a tio n

A number of important outcomes and impacts can be achieved through collective

approach to control porcupine and mal rat. It is of utmost importance that
motivation for working together towards common problems/goals be increased
among the watershed farmers. By doing so the time and energy spent in guarding
the crops at night would be decreased considerably. Consequently, it will lead to
substantial improvement in health quality and also increase household incomes and
in food security from reduced crop losses by these pests It is also essential to evaluate
effectively farmers' traditional control methods of the pests. Proper identification of
control approaches and effective utilization of the methods for collective action at
any region should receive due consideration.

The need for collective action in developing and maintaining resources and resource
# management technologies even at a micro-watershed level is of paramount
importance. Robust collective management depends on the level of existing
community organization and capital resource, which is the strength of the norms and
# social relations that enable people to work together to achieve their ultimate goals.
Some technologies also need to be adopted over a wide area to be effective, so
farmers who wish to adopt them must cooperate with their neighbors. It is essential
tha farmers jointly recognize the trans-boundary nature of pest management
problems, in order to legitimize the otherwise socially unacceptable interference with
the farming practices of others. The finding of this studies revealed that mobilizing
the collective action for common NRM problem would significantly change the life of
farmers in reducing yield losses and other social merits. Political leaders play a
significant role in collective action for helping and enforcing the binding rules or
norms that govern the agreed up on action. To make the collective action sustainable,
it is advisable to involve different stakeholders with full commitments and must in
place in their yearly routine work calendar.

R e fe r e n c e s
Begashaw, L., Waga Mazengia, and Laura German. 2007. Mobilizing the collective
action for vertebrate pest control: The case of porcupine in Areka. AHI brief,
February 2007, No E3, Africa Highland Initiative, Empowering community to
regenerate livelihood and landscapes, Uganda, Kampala.
# FARM Africa, Farmers' Research Project. 1999. Review of experiences with
participatory on-farm trials in the Southern Region of Ethiopia: Proceedings of
a workshop held in Sodo,17th - 22lui May 1998. FARM Africa and Institute of
, Sustainable Development, Addis Ababa, Ethiopia. 221-227pp.
 129 Lculscgcd Bcgaslnnv et. al.

Gaspert, F.,M. Jabbar, C. Melard and J.P. Platteau. 1998. Participation in construction
of local public goods with individualities: an application to watershed
development in Ethiopia. Journal o f African Economics 7(2):157-184.
Gebremedhin, B., J. Pender and G. Tesfaye. 2002. Collective action for grazing land
management in mixed crop-livestock systems in the highlands of northern
Ethiopia. Socio-Economics and Policy Research Working paper No. 42. Nairobi,
Kenya: International Livestock Research Institute.
Gereman, L., Waga M.; Wilberforce T,; Shenkut A.; Joseph T.; Simon N.; Leulseged
B.; Hailemichaiel T.; Zenebe A.; Mesfin T.; Francis A., Ashenafi M.; Kassahun
A.; Awadh C.; William C.; Tessema T.; Zewdie J.; and Kiflu B. 2008. Enabling
Equitable Collective action and Policv change for poverty reduction and
improved Natural Resource Management in Eastern African Highland. Africa
Highland Initiative; CAPRi Working paper 86.
Knox, A. and R. Mienzen-Dick. 2000. Collective action, property right and devolution
of natural resource management: Exchange of knowledge and implication for
policy-a workshop summery paper. CAPRi Working paper 11. Washington,
DC: International Food Policy Research Institute.
Lubell, M., Schneider, J.T. Scholz and M. Mete. 2002. Watershed partnerships and
emergence of collective action institutions. American journal o f Political Science
46(1): 148-163.
Meinzen-Dick, R., A. Knox, B. Swallow and F. Place. 2002. Introduction. In
Innovation in Natural resource Management: the role of property rights and
collective action in developing counties, Meinzen_Dick, R., A. Knox, F. place
and B.Swallow (Eds). Baltimore, MD: Jons and Hopkin University Press.
Munk Ravnbourg, H. and J.A. Ashby. 1996. Organizing for local-level watershed
management: Lessons from Rio Cabuyal watershed, Colombia. AgREV
Network paper 65:1-14.
Munk Ravnbourg, H., A.M.de la Cruz, M.del pilar Guerrero and O. Westermann.
2000. Collective action in ant control. CAPRi Working paper No. 7.
Washington, DC: International Food Policy Research Institute.
Ostrom, E. 1990. Governing the commons: The evolution of institutions for collective
action. Cambridge, UK: Cambridge University Press.
Scott, C.A. and P. Silva-Ochoa. 2001. Collective action for water harvesting irrigation
in the Lerma-Chapala basin, Mexico. Water Policy 3:555-572.
Swallow. B. M., P. Garrity and M. van Noordwijk. 2001. The effect of scales, flowers
and filters on property rights and Collective action in watershed management.
CAPRi working paper 16. Washington, DC: International Food Policy Research
Institute.
Tanner, C.L. 1995. Class, caste and gender in collective action: Agricultural labor
union in two Indian village. The Journal o f Peasant studies 22(4):672-698.
 Experiences oil IJiisef Technology
Dissemination and Value Chain Analysis
Getaliun l)e"ii
S A R I/H iv a ssa A gricu ltural R esearch Center,
P.O .Box-6, H w assa, E th iop ia. E -m a il:getahu n degu @ gin ail.com

A b s tra c t
This paper highlights the status o f enset technologies dissemination with respect
to promotion o f improved enset clones, enset processing devices, demonstration
o f improved enset decorticator ami squeezer, extension advice to farmers
concerning enset diseases, adoption o f improved processing devises, case study
in Southern ami Oromia regions, enset marketing and value chain analysis, the
case o f Amaro special woreda. The gaps on enset technologies disseminations are
identified. Enset production is highly affected by diseases, pests anil vertebrates.
Bacterial w ilt , mole rat, porcupines and enset mealy bug are the major problems
that influence the production system. The use o f backward anil inefficient
traditional methods and equipment in production, processing and marketing
activities, protecting and promoting the endogenous wealth o f crops such as enset
are major challenges. Most diseases are transmitted by farm ers when they cut
the affected and healthy enset plant with the same farm implements. Moreover,
mixing o f Bulla is cheap in the local markets as compared to prices in the other
urban areas and central market but mixing it with foreign material would affect
the exchange and contract enforcement iu the product. The challenges and
opportunities are discussed to draw appropriate recommendations in the area o f
interventions that are technically feasible, socially acceptable and economically
viable. Scaling up cost effective and efficient enset processing devices for small
scale and commercial farmers are mentioned. Promoting value addition for enset
grooving farm households from enset products such as bulla for starch extraction
and fiber for industrial raw materials are recommended.

In tro d u c tio n
The inadequate growth in production of food crops has led to increasing food
insecurity in many parts of the country. The principal causes of insufficient growth
in food production and for subsequent increase food insecurity are inadequate and
unreliable rain fall as a major cause of drought, soil degradation, land tenure,
geographical diversity, and population growth.

Enset has many different uses and contributes in securing food for smallholder
farmers even at times of food shortage period and drought. Over 20 million people of
 131 Gctahun Dcgu

th? Ethiopian population consume the crop as staple and co-staple food. The two
major enset growing regions are SNNPR and Oromia. Farm households in enset-
based agricultural system use the crop to feed their families and livestock. Other
ethnic groups in the regions of Amhara and Tigray also use enset for feed and enset
leaves used as wrapping material for butter, hulln and while baking bread.

The major food products of enset plant are koclw, bulla and amicho. Kocho is a bulk of
fermented starch made from a mixture of the decorticated leaf-sheaths and grated
conn. It can be stored for long periods of time* without spoiling. The combination of
kocho and kitfo is now virtually a common menu at restaurants. The best quality ensel
food, bulla, is obtained mainly from fully matured enset plants, ll can be prepared as
a pancake, porridge and dumpling. Amicho is the boiled enset corm, which consumed
in a manner similar to that of other root and tuber crops. Certain clones are selected
for their amicho production. Fiber is the non-edible product of enset left after
decorticate leaf sheath at harvesting time. The fiber has an excellent structure, and its
strength is equivalent to world fiber crop abaca (Godfrey, 1985). The fiber is also
used to make bags, ropes, twines, cordage, mats and as wrapping material. Average
yields of kocho range 31 to 60 t/ha/vear, bulla 1.6-3.0 t/ha/vear and fiber 201-400
kg/ha/year with great variability of clone types and its maturity period (Atnafua el
al., 1980). The yield of enset exceeds other root and tuber crops. Hence, re search,
extension and development organizations should support enset production so that its
role in the national productivity food security and poverty alleviation program
would be strengthened. The major objectives of this review was to identify
constraints and opportunities of production, processing and marketing, to assess the
major value additive activities of enset along the value chain, and to evaluate the
extent of enset processing devices dissemination.

M ethod ology

Secondary data are collected from published and unpublished sources such as the
adoption of improved processing devises, case study in SNNPR and Oromia regions
as well as marketing and value chain analysis the case of Amaro special woreila.
Marketing margin and costs are discussed in relation to the services and functions
provided. The Total gross marketing margin (TGMM) is the final price of the
produce paid by the end consumer.

TGMM = Consumers' price - Farmers' price X 100................................... (1)

Consumers' prices

Producers' share of the consumers price (PSCP) is the portion of the price paid by the
consumer that belongs to the farmer (producer). The efficiency of any marketing
system usually depends on the share of producers' to the consumers' price. The
 132 Experiences on Enset Technology Dissemination and Value Chain Analysis

better the efficiency of marketing, the better will be farmers' share of the of
consumers' price.

PSCP = Price paid by the consumer - Marketing gross margin X 1 0 0 ..... (2)
Price Paid by the Consumer

NMM = Gross margin - Marketing and transactions cost X 100..........(3)

Price paid by the consumer

Where: TGMM = Total Gross Marketing Margin

PSCP = Producers' share of the consumers price
NMM = Net marketing margin

A ch ie v e m e n ts

• Enset Technologies Dissemination

The majority of extension, development, and research on Ethiopian agriculture have
focused upon the cereal-based systems of the highlands of northern, central, and
eastern Ethiopia, and to a lesser extent upon the shifting cultivation economies of
sub-tropical and lowland western Ethiopia (Steven et nl., 1997).

The attention given for enset extension or technology popularization in the country as
a whole was low as compared to other crops. Steven and his coworkers revealed
that enset agriculture has received little attention on research, development and
extension. The reasons may include: because: I) the majority ol enset farmers live in
one of the least developed regions where making access and logistics are difficult; 2)
the system is unique when compared to cereal farming; 3) production process is
complex; and 4) there is the perception that it is eminently successful, sustainable,
and trouble less crop.

* SVomotion o f Enset Processing Devices

Ethiopian Institute Agricultural Research (EIAR) and Rural Technology Promotion
Center (RTPC) have generated different enset processing devices (scraper and
pressing device). These were developed and distributed in major enset growing areas
of the country. To popularize the improved enset processing devices first contacts
were made with communal institutions like farmers associations and service
cooperatives. Through different institutions these devices or equipment were
distributed to these communities free of charge for demonstration purposes. Deribe
(1996) posit that in almost demonstrated areas farmers accepted the improved enset
scraper and processing devices with few amendments for modification. From 1987 -
1992 a total of 180 scrapers and 32 processing devices were distributed in different
enset growing regions in Southern Ethiopia.
 133 Getahun Pegu

• D em onstration o f Improved Enset D eeortieator ami Squeezer

Attempts have been also made by Hawassa Agricultural Research Centre to
demonstrate and popularize the existing available enset processing technologies in
majDr enset growing regions of Southern Ethiopia. The improved enset decorticator
and squeezer were introduced in three major enset growing zones of Kambata
Tembaro, Dauro and Sidama. Demonstration was carried out in selected PAs. The
result indicated that in all three zones the decorticator and squeezer were accepted
by the participated farmers. Farmers noted several advantages of the improved
processing devices. These include: it produces quality cord; easy to operate; can be
operated either standing or sitting positions; separates the "bulla" and "kocho"
completely, thus produce quality bulla; saves labor and it is more hygienic.
However, further improvements of the processing devices were suggested by the
farmers as follows.
a. Improving the flat decorticator or clapper;
b. The squeezer is too narrow and heavy to move from place to place;
The opening and closing part of the squeezer should be flexible and lighter in
weight,
i. The squeezer should be large enough to produce more bulla per unit time.

• Improvement o f Enset Processing Device

The traditional harvesting and post harvesting procedures are cumbersome; labor
intensive, unhygienic, impose a lot of inconvenience to the working women, and
associated with great yield loss. These methods of processing may reduce the
quality and quantity of enset food and fiber. However, research has been conducted
in several institutions to develop improved processing devices. The most prominent
ones were the Institute of Agricultural Research at Melkassa and at Hawassa, the
M inishy of Agriculture, and Hawassa College of Agriculture. Efforts have also been
made to modify: 1) the decorticator that separates the leaf-sheath and pulp from the
fiber; 2) the pulverizer that grates the corm into fine pieces; 3) the kneader that
squeezes out unwanted water from fermented kocho; and 4) the shredder that chops
the fiber from the fermented kocho. However, such devices are primarily
experimental and have had little testing results and hence farmers' acceptance did
not materialize because of cost and inaccessibility (Metshen and Abate, 1994).

• Adoption o f improved processing devices

In order to improve the processing of enset products, some devices such as kocho
scraper and bulla extractor were introduced and disseminated to farmers in Wolayita,
Kembata-tembaro and Gurage zones between 1987 and 1992 (Deribe, 1996).
Assessment made on the adoption, indicated that nearly half of the sample-farmers
used the improved devices. Among the adopters, 88% suggested the need for further
improvement (Table 1). Half of the non- adopters also indicated that the devices need
further improvement to satisfv their needs under the circumstance. The other half of
the non-adopters reported that the devices are convenient but could not adopt them
due to financial problems (Million, 2008).
 134 Experiences on Enset Technology Dissemination and Value Chain Analysis

Table 1 Adopters' and non-adopters' reaction towards improved enset processing devices in Gurage Zone, 2001

Non-adopters Adopters
Devices’ conditions/ X2 statistics
Farmers reasons N % N %
The devices need further improvement 24 48 38 88 16.98***
The devices are complex 34 68 39 91 5.18***
The devices are expensive 13 28 34 79 20.6***
Extens on information 39 78 39 91 2.76*
Availability of credit 8 16 18 42 7.67***
Improved devices preserve the usual taste 19 62 31 72 10.81***
Thin hole in bulla extractor prevents easy 8 16 11 26 2.67(NS)
passaceof bulla
Note: ***=significant 0.01%, *=significant 0.05% and NS=None significant.
Source: Survey report, Million (2008)

Farmers7 decision to adopt or reject improved enset processing devices was

influenced by the simultaneous interaction of technical, socio-economic, physical,
and institutional factors. Out of 16 explanatory variables hypothesized to affect
farmers' adoption of improved enset processing devices, 9 were found to be
statistically significant. The significant variables include: farm size, number of enset
plarts in a field, frequency of processing per year, extension information, farmers
perception of the devices, taste of processed kocho or bulln, thin hole of bulla
extractor, training and expensiveness of the devices.

• Enset m arketing and value chain analysis

An agricultural value chain is usually defined by a particular finished product or
closely related products. This includes all firms and their activities engaged in input
supply, production, transport, processing and marketing (or distribution) of the
product or products.

The value chain concept entails the addition of value as the product progresses from
input suppliers to producers and to consumers. A value chain, therefore,
incorporates productive transformation and value addition at each stage of the value
chain where the product changes hands through chain actors, transaction costs are
incurred, and generally some form of value is added. Value addition results from
diverse activities including bulking, cleaning, grading, and packing, transporting,
storing and processing (Fig. 1).

In agricultural value chain analysis, a stage of production can be referred to as any

operating stage that is capable of producing a saleable product serving as an input to
the next stage in the chain or for final consumption or use. Typical value chain
linkage include input supply, production, assembly, transport, storage, processing,
wholesaling, retailing, and utilization, with exportation included as a major stage for
products destined for international markets. A stage of production in a value chain
performs a function that makes significant contribution to the effective operation of
the value chain and in the process adds value.
 135 Getahun Pegu

Coordination refers to the harmonization of the functions of a value chain-its

conduct. The result of good coordination between the stages of a value chain may be
reJlected in a good match between buyer preferences and seller supplies. That is,
better coordination in a value chain results in better matching of demand and supply
between the chain stages, resulting in efficient and low-cost exchange, quality
maintenance and value addition.

Agricultural value chain analysis can be viewed as a heuristic device or analytical

tool (Kaplinisky and Morris, 2001). The research can be descriptive, perspective and
designed to provide operational guidelines to improve efficiency of vertical
coordination. Agricultural value chain analysis systematically maps chain actors and
thedr functions in production, processing, transporting and distribution and sales of a
product or products. ____________________________________

HI ' ■ 1H H lI— S l: n d K

• Characteristics of bulla traders and marketing margin (Amaro special

woreda, 2008);
Traders who appear in bulla transaction are rural retailers, assemblers, wholesalers
anc urban retailers. They rarely add value to bulla except place utility. They buy bulla
from places of less demand to places where it is demanded more such as urban areas.
Most traders complained that adulteration of bulla with kotcho and cassava with the
motive to derive unfair profit is a threat to their endeavour and affected their
bargaining power and contract enforcement.

Rural retailers buy bulla from farmers at farm gate or at the same market place for
retail. They sell bulla to consumers or assemblers and use the premium for
 136 Experiences on Enset Technology Dissemination and Value Chain Analysis

purchasing the immediate and priority household needs. They are small in terms of
resources and use personal or family labour for all activities.

Assemblers buy bulln from farmers or retailers mostly from Jijola and Kelle markets
for resale to wholesalers in the neighbouring urban areas (Yirgacheffe, Dilla,
Shashemene and Awassa). They regularly move between markets and face higher
costs of transactions, accommodation and transport They incur the cost of transport
of goods, loading and unloading and absorb the risk inflicted by defects refused by
wholesalers, urban retailers and consumers. They use motorized transport, ISUZU,
in moving products from local to destination markets. Currently, about six
assemblers are working as bulk buyers.

No strong boundary between wholesale and retail trader is observed. All

wholesalers sell to consumers as retailers do. Many wholesalers have three to five
hard core client retailers down stream and more than one supplier up stream.
Wholesalers at urban area buy on credit from suppliers (assemblers) and sell on cash
to retailers and consumers. Mostly they work in their own premises and sometimes it
is rental. They are formal traders and pay between 400 and 500 Birr in renewing
license annually. They frequently complain about the existing unfair level field
between traders and the competition with unlicensed traders. The supply of bulla is
declining both in quality and quantity. In such a situation it is increasingly difficult
to stay competitive. Consequently, they have less incentive to stay in the business
and prefer to switch to others up on study.

Gross marketing margins are highest at assemblers and wholesalers stage because of
transactions and marketing costs arising from increased handling of products (Table
2). Assemblers and wholesalers also absorb costs of spoilage in the form of defects
arising when dealing with poor quality bulla. Producers' share of consumers' price
increases when sold locally by farmer traders and at distant markets by urban
retailers. However, long distance sell to urban areas have a lot of transactions and
marketing costs which is unbearable and expensive to small and scattered farmers.
Further, trading in bulla has been risky and has a problem of contract enforcement.
Given higher costs of handling and risks associated with marketing of bulla
assemblers and wholesalers are receiving better return in the form of net marketing
margin.
 137 Getalnui Pegu

Table 2: Costs (Birr) and margins per average locally packed bulla (five kg) for different market operators

Costs and margin Rural

retailers Assemblers Wholesalers Urban retailer
1. Marketing costs
1.1. Purchase cost 24.00 25.00 31.00 34.00
1.2. Processing cost - - - -

1.3. "ransport cost - 1.00 - -

1.4. loading and unloading - 0.12 - -

1.5. Tax and licensing fees - 1.50 1.25 -

1.6. Tax - 1.00 - -

1.7. Spoilage 0.10 0.20 0.20 0.10

1.8. Other costs 0.20 0.80 0.65 0.10
2. Transaction costs
2.1 Opportunity cost of labor 0.25 0.30 0.55 0.40
2.2. Opportunity cost of capital - 0.40 - 0.15
3. To:al cost 24.55 30.32 33.65 34.75
4. Sales 25.00 31.0 35.00 35.00
5. Gross margins 0.45 0.68 1.35 0.25
6 . To:al gross margin 4% 19.35% 11.43% 2 .86%
7. Producers share of consumers price 98.2% 97.81% 96.14% 99.29%
8. Net marketing margin 1.8% 2.19% 3.85% 0.71%

Source: Source: Survey results, 2008 (Agri-Service, Amaro Special Woreda )

The Z’filu e chain describes the full range of activities which are required to bring a
product or service from conception, through the different phases of production
(involving a combination of physical transformation and the input of various
producer services), delivery to final consumers, and final disposal after use.
Considered in its general form, it takes die shape as described in Figure 2. The
environment, sequence of major events, detail activities and principal actors were
identified along the value chain. As can be seen from this, production per se is onlv
onejof a number of value added links. Moreover, there are ranges of activities within
each link of the chain. Although often depicted as a vertical chain, intra-chain
linkages are most often of a two-way nature - for example, specialized design
agencies not only influence the nature of the production process and marketing, but
are in turn influenced by the constraints in these downstream links in the chain.

In addition to the manifold links in a value chain, typically intermediary producers

in a particular value chain may feed into a number of different value chains. In some
cases, these alternative value chains may absorb only a small share of their output; in
other cases, there may be an equal spread of customers. But the share of sales at a
particular point in time may not capture the full story - the dynamics of a particular
market or technology.
 138 Experiences on Enset Technology Dissemination and Value Chain Analysis

Inadequate Licensing &

extension by tax
Gov’t & collection
E n v iro n m en t
ASE by Gov’t

S eq u e n ce o f

m a jo r events

Seedling Identify mature Storing Preparing in

Preparation enset various
Land Transport
Cutting leaves forms
Planting D etailed
Labor and dry leaf Exchange
a c tiv itie s
Weeding sheaths
Compost
Cleaning Decorticating

Compost Squeezing

supply Dehvdration

P rin cip a l A ctors

Fig. 2. Bulla value chain map

 139 Getaliuii Pegu

• M arketing channels o f bulla

The survey data at Amaro, revealed that 41.3% of the marketable surplus of bulln
passes directly from the producer or rural retailers to the consumers in the district.
The remaining is taken outside of the district by assemblers and retailers (Fig. 3).
Generally, bulla reaches to the consumer in a variety of ways. These include: direct
sales to rural and urban consumers; direct sales to farmer traders or retailers and
assemblers at the market bv farmers. More often, smallholder farmers transport bulla
to the rural markets, mostly carrying themselves or by family members, and
sometimes sell directly to farmer trader (retailers) at the farm gate or in the market,
or directly to assemblers. In the major neighbouring urban centres, fore example, in
Yirga Cheffe and Dilla bulla originating from Amaro dominates the transaction. It
appears that bulla from Amaro is of high quality than those products coming from
other localities which are least demanded.

Bulln originating from rural areas of Amaro is transported as far as Hawassa,

Shcshem ene and Addis Ababa. High quality and accordingly high unit value have
improved its wider spatial transaction. High transaction cost in far-away market
places is obvious as the product moves and changes through the chain. Costs in
terms of transfer, attendants, risks (cost of physical deterioration and prices), loading
and unloading, and opportunity cost of capital tied and labour is progressively
increasing. This makes the product unaffordable to the majority of consumers. All
through the chain most respondents disclosed that bulln coming from Amaro is
quality but expensive. It is highly demanded by high income groups in the local as
well as high income urban areas out side the district where many households are
willing to pay.
 140 Experiences on Enset Technologi/ Dissemination and Value Chain Analysis

Small Scale Farmers (bulla)

Figure 3. Bulla marketing channel

G ap s
• Enset production is highly affected by diseases insect pests and vertebrates. Bacterial
wilt disease, mole rat, porcupines and enset mealy bug are the major problems that
influence the production system.
• The use of backward and inefficient traditional methods and equipment in
production, processing and marketing activities,
• The existing extension service component doesn't give enough emphasis on
protecting and promoting the endogenous wealth of crops such as enset,
• Most diseases are transmitted by farmers when they cut the affected and healthy enset
plant with the same farm implements,
® Mixing of bulln with foreign material is affecting the exchange and contract
enforcement in the product,
 141 Getahun Pegu

• Bulln is cheap in the local markets as compared to prices in the other urban areas
and central market,
• Enset products such as kocho, bulln and fiber were not adequately promoted for
industrial inputs. Value addition for enset growing farmers could be possible if bulla
extracted for starch and the fiber as raw material for various industrial products
(sacks, different bags, mattes etc).

Discussion
The diagnostic survey (Million, 2008) was conducted in the Southern region (Masha
district of Sheka Zone) and Oromiya region (Wenchi and Tikur Enchinii districts).
The results depicted that the three most important factors responsible for the decline
in enset acreage were bacterial wilt (47%), mole rate (13%) and moisture stress (16%).
Various research experiments were attempted to screen enset clones against bacterial
wilt under artificial inoculation. In addition the mechanism of pathogen
dissemination and its survival was investigated. Hence, no chemical control measure
was so far effective to control the disease. Therefore, several studies recommended
sanitary control measures as a short-term solution to the disease (Dagnachew and
Bradbury 1968; Dereje, 1985; Archido and Mwadin, 1993 and Million, 2008). Some of
these include:
• Digging pit/hole 3 m deep and 2 m wide outside of the enset field
• Uprooting infected plant
• If the infected plant is big, chopping into small pieces without contaminating
other plants
• Burying the uprooted plant in the prepared hole
• Finally disinfecting by flaming all farm tools used

The mature enset is usually allocated for process and household members participate
in selecting the specific plant. Household head or male members of the family took
part in removing all the leaves and uprooting from the plant. The subsequent tedious
operation became the responsibility of women who perform either in group or
individually based on the volume of work and urgency. A woman sits in front of a
decorticating board made of wood inclined at certain degree and supported by
strong mature standing enset plant. She lifts one leg and with it holds the leaf sheath
to be decorticated on the wood. Bamboo splits that have sharp cutting edges on both
sides are held by hand and the leaf sheath is scrapped step by step until white and
strong fiber is remaining. The scrapped leaf sheath piled together and the corn is
chopped with special wood and mixed with the leaf sheath. The mixed part is called
kotcl’o thereafter. In some places the wood has circular head on which there are
smaller and pen like projections that are used for chopping and scrapping of the
corn.

Bulla is the derivative that results during the processing of kotcho. The mass of kotcho
then squeezed by applying enough pressure to release semi liquid substance. This is
 142 Experiences on Enset Technology Dissemination and Value Chain Analysis

allowed to stand for some time and after precipitation and dehydration it forms
* bulla Despite the traditional bulla processing is an arduous task all the hand tools are
locallv made, cheap and easily available, but are out dated and mechanically
inefficient.

Therefore, enhancing the adoption of improved enset processing devices among the
enset growing households could bring about a significant impact for the farming
community regarding the quantity and quality of enset bi-products.

Farmers have limited access to information regarding enset production

technologies and marketing decisions. They predominantly produce these products
traditionally for consumption and markets. Currently, new market opportunities are
emerging based on the specific demand of consumers and/or end-use processors.
These expanding niche markets, both domestic and international, provide
opportunities for family-size farmers to exploit their comparative advantage in
* producing some types of value-enhanced products to specification.

To take advantage of these new opportunities, farmers may need to become more
* active participants in the value chain. First, they need to know what products are
needed, including specific quality traits or attributes. Second, they need to know
whether they have an inherent comparative advantage in producing these products
in relation to other potential suppliers.

Development of grades and standards in bulla, training of farmers is of paramount

importance. Providing genuine advice and other market operators not to adulterate
the product, and also sanctioning those who exercise trade malpractice deserve
seiious measure. Furthermore, establishment of sustainable marketing groups or
cooperatives and linking them with market in high income urban areas and central
markets would unequivocal be beneficial.

R e co m m e n d a tio n s
This review paper highlights enset production constraints and less adoption of
improved enset production technologies, the challenges and opportunities to enhance
its production, productivity and marketing of the products.
Research institutions and universities in collaboration should support enset
producers by developing/ generating appropriate improved enset production
, technologies based on farmer's priority problems, to enhance enset production and
productivity. The recent effort of other GOs and NGOs in promoting enset
production technologies should be encouraged and strengthen to ease adoption of
* enset production technologies, which are compatible to enset-based farming system.
The relevancy of enset production technologies and the extent and rate of adoption
should be assessed and accordingly establish sound strategy to enhance productivity
 143 Getahun Pegu

by providing sustainable training. Demonstration and scaling up of technologies

that have considerable impact to improve the livelihood of enset culture -based
farming communities in the country should be promoted.

The adoption of improved technologies for processing enset products by farmers is

hindered by socio-economic, technical and many other challenges and problems.
Therefore, efforts to facilitate adoption of the technologies should be strengthened.
For instance, it will be economically feasible to form small groups of farmers so that
they can collectively purchase or borrow the devices. The introduction of improved
en:iet processing should also consider the food habit of the society in a given area.

Assist in the development and dissemination of relevant and adequate market

information to small scale producers. This helps farmers to become more active
participants in the value chain. First, they need to know what products are needed,
including specific quality traits or attributes. Second, they need to know whether
they have an inherent comparative advantage in producing these products in
relation to other potential suppliers.

More emphasis should also be given to enset processing devices to alleviate the
tedious work and intensive labor requirement. It is also important to consider other
productive activities/options to the members to increase income of the household
and to improve the quantity and quality of kocho, bulla and other products of enset.
The promotion of the devices could be on credit and through partner organizations
participation in rural development GOs, NGOs and international organizations by
establishing kind of revolving fund raising mechanism.

Farmers indigenous knowledge and/or innovation on bacterial wilt control, mealv

bug and rodents should receive due attention along with the recommended sanitarv
measures to alleviate the problems and improve production and productivity until
resistant/tolerant varieties are developed in the long term.

It is advisable to assess the marketing opportunity for enset products mainly for bulln
and fiber in Ethiopia. Diversifying enset production in potential areas and/or niches
to overcome the food insecurity at times of draught for household food and livestock
feeds is indisputable. Meanwhile, the supply of bulln for starch extraction and fiber as
raw material for small processing and larger industries could be achieved through
value addition so that farmers and industries could also produce at their capacity to
maximize profit. The raw materials supply from enset producing farmers should also
be complemented by organizing small groups in commercial farming of enset and
large scale plantation by the respective industries themselves.
 144 Experiences on Enset Technology/ Dissemination and Value Chain Analysis

Reference
Agric-service Ethiopia.2008. Enset value chain, Consultancy report, Amaro special woreda.
Alenm, D. and Demese C. 2005. The national Extension Intervention Program (NIEP) and
Sustainable Agricultural Development: An exploratory study to steer the debate on
ADLI. In: Negatu W., Dadi L., Hailegebriel A., Belete S. and Gebremedhin B. (Eds.),
Revising vulnerability of rural livelihood in Ethiopia. Options for sustainable
development: Proceedings of the 7lh Annual Conference of AESE (Agricultural
Economics Society of Ethiopia), Addis Ababa, Ethiopia. AESE, Addis ababa
Anandajayaskerem P. and Birhanu G. 2009. Integrating innovation system perspective and
value chain analysis in agricultural research for development: implications and
challenges. In: Improving productivity and market success (IPMS) of Ethiopian
Farmers project working paper 16. ILRI (International Livestock Research Institute,
Nairobi, Kenya. 67 pp.
Atnafua Bekele, Medhin Zewedu, Mikias Yeshitiela and Ermiyas Tesfaye.2008.Post-harvest
management of enset products: In: Woldegiorgis, Endale Gebre and Berga Lemaga,
Eds). Root and Tuber Crops: The Untapped Resources, Addis Ababa, Ethiopia.pp 235-
251.
.......— Awassa Agricultural Research Center (ARC) Progress Report (1997/1999).
Berl anu, G. , Hoekstra D. and Azage T. 2006. Commercialization of Ethiopian agriculture:
Extension service from input supplier to knowledge broker and facilitator. IPMS
(Improving Productivity and Market Success) of Ethiopian Farmers Project Working
Paper 1. ILRI (International Livestock Research Institute), Nirobi, Kenya. 36 pp.
Deribe K. 1996. Development and Promotion of enset processing equipment. In: Tsedeke
Abate, Clifton Hiebsch, Steven, A. Brandt and Seifu Gebremariam, Proceeding of Enset
- Based Sustainable Agriculture in Ethiopia, Addis Ababa, Ethiopia, 1 3 - 2 0 December,
1996. pp 328 - 330. Eds
Million T .2008. Socioeconomics and Technology Transfer. Gebremedhin Woldegiorgis,
Endale Gebre and Berga Lemaga, (Eds). In: Root and Tuber Crops, The Untapped
Resources, Addis Ababa, Ethiopia, pp.241-251
SG 2000. Agricultural Project in Ethiopia. Project 2002. Activities and Outputs: An assessment
SG-2000 Agricultural Project in Ethiopia, pp 1-52. August 2002 Sasakawa - Global 2000,
Addis Ababa, Ethiopia.
Steven A. B. , Anita S., Clifton IT, Terrence J.M ., Endale T., Mulugeta D., Gizachew Wolde-
M„ Gebre Y, Masayoshi S, and Shiferaw T. 1997. THE "TREE AGAINST HUNGER"
Enset-based Agricultural systems in Ethiopia. Pp. 56. American Association for the
Advancement of Science with Awassa Agricultural Research Center, Kyoto University
Center for African Area Studies and University of Florida.
 Research Experience and Achievements on
Traditional Enset (Ensete ventricosum)
Fermentation Processes
Tariku IIiiiHitiina
Ethiopian Institute o f Agricultural Research, Ambo Agricultural
Research Center Email: tarikith2002@yahoo.com

A b s tra c t
The paper reviews the available literatures on traditional enset processing
techniques from enset maturity stage through processing to the final enset
primary food products. Enset growers in different localities use different locally
known maturity signs. These signs were not scientifically investigated whether
they are appropriate maturity signs or not in relation to nutritional contents.
Traditional enset processing fo r fermentation is tedious, unhygienic and
generally done by women using locally made equipment. Attempts made
hitherto 011 the modification o f the process, in general, are not to the level needed
as compared to the indigenousness o f the process. Microbiological studies done
so fa r on traditional enset fermentation process were concentrated on isolation
and identification o f the microbial population, in the ferm enting mass.
Traditional enset fermentation was considered lactic acid fermentation. Here,
attempts should be done to undertake controlled fermentation studies with
selected culture starters and optimize the process into modern food processing
technology. During post-fermentation time, it has been known that about 33 %
o f the products were lost due to spoilage and lots o f microbial contaminants
were isolated from the products. Therefore, modification o f storage facilities and
scientific investigation o f any related public health problems needs attention.
Hence, understanding, improving and standardizing traditional enset
fermentation process needs attention so that the burden on women can be
minimized, quality and quantity o f the food products get improved, the crop can
wisely be used and contribute to the food security o f the country.

Keywords: Ensete ventricosum, traditional fermentation

 146 Research Experience and Achievements on Traditional Enset Fermentation

Introduction
Enset (Ensete ventricosuni (Welw) Chessman) is a banana-like herbaceous and
monocotyledonous plant. The plant is robust, ranging from 4 to 11 m in height and
its pseudo-stem, which dilates at the base has a circumference of 1.5 to 3.0 m. Leaves
are borne on the pseudostem almost from the same point and on short petioles, and
are cbout 5 m long and 0.75 to 1.5 m wide (Taye, 1993).

Enset plant is particularly cultivated for its carbohydrate rich food products. The
plant is used as staple or co-staple food for about 20 % of Ethiopian population
particularly in the South and South-western part of the country (Brandt et al., 1997).
According to Shigeta (1993), the utilization method of enset for food is classified into
two groups: (1) the fermentation method of the pseudostem pulp and (2) the steam
boiling method of the underground corm. Amicho, kocho, and bulla (names vary from
one area to another) are the most commonly known primary food products
identified and described from enset. Amicho is made from enset corm and is
immediately consumed after it is boiled (similar to other root and tuber crops)
without being fermented. Kocho is obtained from the pseudo-stem and corm after
fermentation. Bulla (pure starch) is a purified product of kocho and is highly valued.
It is the precipitate that formed when kocho is drained through a fine mesh
(Liyulvork , 1993; Shigeta, 1993; Sandford and Helen, 1993). Annual production of
these primary food products is estimated to be remarkable. In 1994 the production in
Southern Nations, Nationalities and Peoples Regional State (SNNPRS) alone (Shank
and Chernet,1996) were estimated to be about 4,653,075 and 355,439 metric tons of
kocho and bulla, respectively. These primary food products have been produced for
generations by an age-old traditional enset processing techniques. Hence, the
objective of this paper was to review studies made by various researchers on die
traditional enset processing techniques with special attention on traditional enset
fermentation processes. These processes are discussed.

Enset maturity and fermentation process

Enset plant, particularly for fermentation in earthen pit is normally harvested at full
maturity (3 to 9 years) when it is thought to have high storage of food in the
pseudostem and corm. Farmers use different indicators of maturity. According to
farmers in West Shewa Zone, when enset matures, the inflorescence becomes shorten
and the junction of the pseudostem and corm becomes clearly visible above the
ground (Dereje and Endale, 2003). Farmers in Masha Woreda use drying of outer
leaf sheaths and inflorescence as an indicator of enset maturity (Million et al., 2003).
Other reports indicated that starch content of the plant is influenced by age of
harvesting. Enset harvested too young and too old has reduced starch content
(Kelbessa et al., 1993; Liyuwork, 1993). It is very imperative to scientifically identify
appropriate maturity indicators which coincide with time of high nutrient content.
 147 Tariku Hunduma

Identification of matured enset plant is a very crucial step to begin harvesting and
processing. Identification of mature enset plant, harvesting and processing is
generally the responsibility of women and the job is usually done in group (Belay et
a l, 2008). Though, basic traditional enset processing steps are the same, some steps
differ from one locality to another (Seifu, 1993). During enset processing for
fermentation, various multiple steps are incorporated, though pseudostem
decortications and corm pulverization are the most common steps in all enset culture
areas of the country. During processing, most of the works are done side by side by
sharing responsibility amongst the women in the group. The generalized traditional
enset processing for fermentation among the Wolayita and Sidama people and in
Western Shewa Zone was described by Mehtzun and Yewelsew (1994) and Tariku
and Mogessie (2011), respectively (Figs. 1 & 2).

After selection of mature enset plant and harvesting, pseudostem decortications and
corm pulverization continues using traditional equipment to yield mass of enset
which is going to be fermented in an earthen pit. In this process bulln is extracted
from fresh decorticated pseudostem mass. Bulla extraction method also varies from
one locality to another. Most of the time, the decorticated mass is put into a
perforated sack and squeezed allowing the extract to run down into a pit lined with a
plastic sheet. The filtrate is allowed to stand for some time to allow the thick paste to
sett'e leaving the watery supernatant above the sediment. The watery supernatant is
discarded and the white settled thick paste that remain is thinly spread over a clean
leaf sheathes or sheet of clothes to dehydrate (Mogessie, 2006). The dehydrated,
white powder is called bulla. There is a controversy on whether bulla is fermented or
unfermented primary food product. In some localities, bulla is considered as
unfermented primary food product (Tariku, 2008) and is also reported as fermented
product in other localities (Kelbessa et al. 1997).

The decorticated mass and the pulverized corm are mixed thoroughly and put into a
pit dug inside the enset plantation. The pits' inner side is lined with fresh enset leaves.
The mass is tightly covered and layers of more fresh leaves are put on the surface.
Finally, fermentation is allowed to proceed after putting heavy logs and stones as
weight over the tightly wrapped and sealed mass, possibly, to ensure the creation of
airtight anaerobic conditions in the pit (Mogessie, 2006).

Initiating the fermentation process

Some enset growing localities use starter to initiate the fermentation process.
Sandford and Helen (1993), reported the use of a mixture of herbs, orange, banana,
onion and garlic, which are rubbed into the sides of the corm as starter in Morocho
Wallano. Another study indicated the use of herbs and a small amount of fermented
kocho (left from the previous fermented kocho) as a starter (Liyuwork, 1993).
According to Kelbessa et al. (1993), small separately fermented corm (corm rubbed
with a black mass just picked up from rotten enset leaves, shigido) is used as a starter.
Study in W est Shewa Zone also indicated the use of a starter called gamma. Gamma,
 148 Research Experience and Achievements on Traditional Enset Fermentation

processed corm and rubbed with herbs, Koricha gamma is used as a starter culture
after it is changed to liquid form (Tariku and Mogessie, 2011; Dereje and Endale,
2003). According to these authors, gamma is used to speed up enset fermentation as
bikil (malt), which is used in beer preparation. There is general believe that the
pulverized mass will not ferment by itself unless mixed with the starter (Kelbessa et
al.,1993). However, it is known by some authors (Brandt et a l , 1997) that some enset
growers process enset without addition of any starter. When no starter is used, the
pulverized mass was laid on the ground in ambient temperature for about two to
three days to begin the fermentation process (Brandt et a l, 1997). In some localities,
the mixed mass is covered with enset leaves and left at ambient temperature for two
to five day to initiate fermentation (Berhanu, 1987). In the mid altitude of Western
Shewa, the mass is wrapped tightly with enset leaves and left at ambient
temperature for about two weeks to initiate the fermentation process (Tariku and
Mogessie, 2011). The role of these homemade starters and their microbial population
has not been investigated to date.

To ferm ent the pulverized mass, it is kept in the pit from a few weeks to several
months or years of incubation period depending on the ambient temperatures
(Mogessie, 2006). Different authors have reported different length of incubation time
for different localities (Sandford and Helen, 1993; Liyuwork, 1993; Kelbessa et a l,
1993). Study done on two different altitudinal ranges showed that the warmer
temperature of the fermenting mass at the mid altitude site contributed to the rapid
proliferation of microorganisms and hence rapid fermentation process and shorter
fermentation time than at the high altitude site (Tariku and Mogessiei, 2011).
Berhanu (1987) also reported that the fermentation process terminated within three
to six months of time in wormer regions. Though, time length of fermentation varies
from one enset zone to another, there is a general consensus that the longer the
fermentation time in the pit, the more tasty the food products are. According to
Kel bessa et a l (1993), enset cultivators use the long time fermented enset primary
food products as a prestige food for guests and ceremonial occasions.

During this fermentation time length, various processing steps such as reopening,
remixing and changing leaves lining the pit is done at varying interval of time for
different localities. For instance, study by Sandford and Helen (1993), in Morocho
Wallano (Northwest Wolayita) explained that the fermenting mass is moved to
freshly lined pits every fifteen days. Study in Western Shewa highland and mid
altitude showed that remixing, checking-up and changing of leaves lining of the pit
is done only when undesirable odor was detected and at every 15 day interval,
respectively, until the fermentation process is considered completed (Tariku and
Moggessiei, 2011). At the completion of the fermentation process, a moist fibrous
mas:> called kocho is obtained.
 149 Tariku Hutiduma

*MRU»

1. EBCOIl'nCATION Z PULVSROWiaK 3 .M IW 0 4 FBRMENnMlON

9.SQUEBZ1MU tfLKMEASHWG f.S M tE W M ) iSJFW N fl

Fig.1. Steps of enset processing for fermentation in Wolayita and Sidama (source: Mehtzun and Yewelsew, 199
 150 Research Experience and Achievements on Traditional Enset Fermentation

Mature enset plant

= practices at highland
= practices at mid
altitiirip.
Preparing the plant for
decortications, smashing
Bulla
and pulverization
(Extracted primary
food product)
Processing area
preparation

Decortications, smashing
Amicho (corm
and pulverization
consumed
after boiling)

Processed enset mass

Fiber
(by

Pit
preparati
Put into a pit Wrap tightly and leave at
ambient temperature to ferment
(about 2 weeks)

Wrap tightly
with enset
-----------J Pit preparation
I

Fermentation proceeds Put into a pit

(about a month)

Add
'jamma,
mix &
____t......
Wrap tightly
Wrap tightly with with enset
enset leaves
Remixing, Remixing,
checking up
checking
& changing Fermentation
eaves proceed H- up &
(at some
time interval) changing

Fit}. 2 Flow chart of traditional enset fermentation process at high and mid altitude sites of
Western Shewa Zone (Source: Tariku and Mogessie 2011)
 151 Tariku Hunduma

Microbiology o f traditional en set fermentation

Microbiological studies on traditional enset fermentation process particularly at its
natural pit are relatively inadequate as compared to other traditional fermented
foods of Ethiopia. Berhanu (1987) studied and described the microbiology of
traditional enset fermentation and also indicated the similarity of the
microorganisms to those involved in the fermentation of other vegetables (Table 1). It
has been believed that enset fermentation is a microbial disintegration of the grated
corm and pseudostem and imparts flavour and textural qualities to the fermenting
primary food product (Kelbessa et al., 1997). A microbiological study conducted on
traditional enset fermentation process in its actual pit (Berhanu, 1987) reported
Leuconostoc spp and Lactobacillus spp as groups of lactic acid bacteria (LAB) that bring
the desirable changes in the fermented primary food product. Meaza and Berhanu
(1985) identified Lactobacillus, Pediococcus, Leuconostoc and Streptococcus from kocho
and bulla purchased from different markets in Addis Ababa and suspected these
microorganisms to be involved in the fermentation process. Another similar study
(Ayele and Berhanu, 1993) also isolated these microorganisms from fermented kocho
purchased from markets in Addis Ababa. Kelbessa et al. (1997) revealed that
traditional enset fermentation is primarily lactic acid fermentation. Steinkraus (2002)
classified traditional enset fermentation as lactic acid fermentation. Tariku (2008)
isolated Lactobacillus, Streptococcus and Leuconostoc spp from traditionally fermenting
enset mass over the fermentation time. Berhanu (1987) reported Leuconostoc
inesenteroides as an initiator of traditional enset fermentation process. But, in another
studv (Tariku and Mogessie, 2011), this bacterium was not detected in a considerable
amount; rather coliforms and amylolytic Bacillus spp were found to be dominant at
the beginning of traditional enset fermentation process. The Bacillus spp. was
anticipated to contribute to the breakdown of starch to simple sugars favoring the
proliferation of LAB that can utilize simple fermentable sugars. The coliforms were
also anticipated to produce acids (lowering the pH) and created favorable condition
for the LAB (Tariku and Mogessie, 2011). These differences might have been resulted
as reports were from different fermentation conditions, the fermentation process may
vary from place to place depending on various factors. According to Tariku and
Mogessie (2011), after initiation of the fermentation process, the fermentative
microbial population was gradually dominated by homofermentative Lactobacilli,
resulting decrease and increase of the pH and titratable acidity (TA) of the
fermenting mass, respectively (Table 2). The lowest pH reported during traditional
enset fermentation was about 4.30 and 4.20 (Tables 1 & 2).

The heterolactics are more important than the homolactics in producing flavor and
aroma components such as acetylaldehyde and diacetyl (Jay, 2000) and were not
usually detected during active fermentation stage of traditional enset fermentation
process. The usually sour-tasting product of kocho could also be due to the activity of
predominant homofermentative Lactobacilli, which produced more acid than the
heterolactics. During traditional enset fermentation process, yeasts were also isolated
from fermenting enset mass over the fermentation period. Tariku and Mogessie
 152 Research Experience and Achievements on Traditional Enset Fermentation

(2011), isolated yeasts to the level of 6.62 x 104 cfu g-> at the highland and 1.12 x 105
cfu g-1 at the mid altitude and reported that their fermentative activities were less
likely due to their low number. The low number of yeasts could be due to some
faci:ors that can prohibit yeast proliferation, which requires an abundant continuous
supply of oxygen. This is not the case during traditional enset fermentation process.
Berhanu (1987) also isolated, identified and reported four yeast genera; Tricliosporoti,
Torulopsis, Rhodotondla and Candida from traditionally fermenting enset mass. In this
report, it was also not sure whether these yeasts were participating in the
fermentation process or simply as spoilage organisms. However, some report
revealed the fermentative activities of yeast along LAB (Kelbessa et al., 1997; Amoa-
Awua, et al., 1997; Amoa-Awua et al., 1996; Ovewole, 1992). Such discrepancies could
also be because the reports were from different fermentation conditions and indicate
that the process may vary from locality to locality.

Aerobic and anaerobic spore forming bacteria were also isolated from traditionally
fermenting enset mass, market kocho and India (Meaza nd Berhanu, 1985; Berhanu,
1987). The spore former were dominated by Clostridium spp. Kocho is known to have
a characteristic butyrous smell (Kelbessa et ui, 1997; Berhanu, 1987) and this was
suggested to be as a result of activities of clostridial species (Berhanuhe, 1987).

Effect o f fermentation on nutritional contcnt

Kelbessa et al. (1997), conducted a study on biochemical changes of traditionally
fermenting enset mass (Tables 3, 4 & 5).They showed a reduction of total protein,
ash. and total carbohydrate by 15 %, 16 %, and 34 %, respectively. In the same
manner, a progressive decline in carbohydrate content was observed and this was
assumed to occur due to activity of microbes, which derive their energy from
metabolism of carbohydrate. The level ol soluble and reducing sugars increased
during the first three weeks of fermentation and was assumed to happen as a result
of activity of native microbial amylases. At the end of the fermentation period,
sugars decreased significantly which is also assumed to occur as a result of
metabolism of these sugars
 153 Tariku Hunduma

Table 1; Estimated number of organisms/g dry weight of fermented kocho

Fermentation pH Moisture Temp, of Spore Leuconostoc Streptococcus Pediococcus Lactobacillus Yeast Total
(days) (%) Kocho (°C) Formers Spp faecalis cerenisiae _ spp/ count*
0 6.5 84 15 - - - - 1x 10' 1x 101
2 6.3 75 17 - 3 x10 1 7 x10 1 - - 3x102 4x102
4 6.0 72 16 - 1 x 104 2x 102 - 1.5x102 6x 101 1.04x104
6 5.8 69 15 3 x 102 4 x 1G6 4x10 3 - 9x102 2x 102 4.01x10®
8 5.6 66 15 2x 103 3 x 107 9x102 1x 101 3.3x10s 4x10 2 3.33x107
15 4.9 63 17 1x 103 3.6 x 1 0 9 9x10 2 4x10 3 5.4x109 9x10 2 9x109
22 4.5 60 18 2x 102 1.3 x 107 - 6x 103 1.67x101° 2x 103 1.67x101Q
29 4.4 60 18 8 x 102 4 x 106 - 4x103 2 .1x 109 3x10 3 2.1 x 109
36 4.4 59 15 3 x10 2 2 x 10s - - 7.4x107 4x102 7.6x107
43 4.3 51 18 1 X101 5 x 10“' - - 1.1 x 107 9x10 3 1. 11x 107
50 4.2 60 16 8 x 10' 1 x 103 - - 6x 105 8x 101 6x 10s
64 4.2 60 17 1 x 103 6 x 102 - - 1.2x 105 9x10 1 1.21x 10s
79 4.2 60 18 6 xIO 1 1x 102 - - 1.25x104 1x 101 1.27x104
-Population less than 10/g dry weight of fermenting kocho
* Values represent the sum of Lact. coryneformis sub sp. coryneformis and Lact. plantarum populations.
# Count represents the sum of the microorganisms present at any one of the fermentation periods
(Source: Berhanu, 1987)
 _____________ 154______ Research Experience and Achievements on Traditional Enset rennenlnlion

Table 2 Changes in means pH TA. temperature and moisture during traditional enset fermentation at two different altitudes

Fermentation _.pH TA (%) Temperature (°C) Moisture (%)

time (day) Mid-alt. Highland Mid-alt. Highland Mid-alt. Highland Mid-alt. Highland
0 6.05a 6 . 11a ND ND 17.0h 16.2cd ND 86.3a
5 5.87b 5.89b ND ND 27.0a 14.0e 86.0a 84.4ab
15 4.72c 4.89c ND ND 24.0b 15.5d 85.8a 82.2bc
36 4.63d 4.86c 0.07d 0.07f 190g 13.4e 83.0b 80.9c
52 4.52e 4.77d 0.18bc 0.13e 19.6f 14.1e 73.5c 78.0d
66 4.38f 4.62e 0.18c 0.14cd 20.1ef 14.0e 67.5d 70.2e
82 4.36f 4.31f 0.18c 0.15c 21.Od 17.0b 61.4e 64.1f
97 4.34g 4.30f 0 .20b 0 .20b 20.5de 16.4bc 61 8e 62.1 fg
112 4.31h 4.32f 0.23a 0 .22 b 22.0c 16.2dc 61.1e 61.0g
127 Final 4.33f Final 0.30a Final 17.0b final 60.7g
142 4.33f 0.30a 18.9a 60.3q

Mid-alt= Mid-altitude = 2252 masl Highland=2908 masl ND= not determined: Values are means of three determinations
Means in a column with the same letter are not significantly different from each other at 5% level of significance (DMRT).

Source: Tariku, 2008

 155 T ariku H undum a

to glycolytic end products such as lactic acid and other volatile

organic acids.

The amount/level of iron, phosphorous and calcium also decreased

significantly. Perhaps, this could be attributed to the possible
leaching out of the nutrients through the permeable fermentation pit.
In the contrary, the amount of free amino acids and non-protein
nitrogen (NPN) were found increasing and this is assumed to occur
as a result of proteolytic fermentation of enset. However, the study on
enzymatic activities of bacterial isolates from traditionally
fermenting enset mass showed more dominantly amylolytic than
proteolytic and lipolytic (Tariku and Mogessi, 2011). Such nutritional
content loss deserves attention for modification of the traditional
fermentation pit. The existing general consensus that disclosed the
longer fermentation period improves taste of the enset food products
also deserves further clarification in contrast to this declining
situation of most of the nutritional contents.

Keeping quality and post-fermentation loss

Traditionally fermenting enset mass in an earthen pit and enset

primary food products during storage has showed spoilage
problems. Fermenting enset mass in an earthen pit especially for an
extended period of time has shown spoilage problem; and the
disease manifest itself by bad smell and it can easily be transmitted
from infected to uninfected enset products. The spoilage is suspected
to be caused by fungal species (Brandt et al., 1997). Fungi such as
Petiicilliuni sp., Aspergillus sp., Cladosporium sp., Scopulariopsis sp.,
Rhizopns sp. and Mucor were isolated from enset primary food
products bought from market and were suspected to be contaminant
(Meaza and Berhanu, 1985).
 156 Research Experience and Achievements on Traditional Enset Fermentation

Table 3: Effect of fermentation on proximate composition of kocho (dry wt basis)

Free
Moistu Total amino
Pericd re protein Fat Crude fiber Ash NPN acids
(wee<)l (%) (%) (%) (%) (%) (%) (mg/g)
0 84±3 4.07±0.02 0.43±0.01 3.43±0.03 0.75±0.01 0 .01±0.01 8.1 ± 1.2
1 66±4 3.97±0.01 0.43±0.00 4.07±0.02 0.76+0.03 0.01 ±0.00 9.0±1.0
2 63±2 3.87±0.03 0.42±0.00 4.13±0.03 0.78±0.04 0.03±0.00 10.7±0.8
3 60±5 3.75±0.01 0.41 ±0.01 4.21 ±0.05 0.75±0.03 0.05±0.02 12.4+0.9
4 60±3 3.63±0.00 0.53±0.04 4.33±0.02 0.67±0.00 0.06±0.01 16.9±1.0
5 60±7 3.53±0.04 0.43±0.05 4.37±0.03 0.65±0.02 0.06±0.00 18.8±1.2
6 60±2 3.47±0.03 0.43±0.02 4.50±0.05 0.64±0.01 0.07±0.01 20.0 ±0.8
7 60 ±4 3.47±0.01 0.43±0.05 5.00±0.04 0.63±0.04 0.07±0.01 20.7±0.7
Values are means ± S.D. of three determinations

Another study associated appearance of molds, with poor wrapping of fermenting

mass, increased acidity and decreased moisture content (Tariku, 2008). Study
conducted in the laboratory on primary food products bought from market isolated
aerobic mesophilic flora dominated by Micrococcus and Bacillus spp. (Mogessie
andYewelsew, 1996). This study also suspected psychrophilic bacteria and molds as
an important cause of spoilage during cold storage. Some Gram-negative rods and
non-lactic Gram-positive cocci were occasionally picked from the aerobic mesophilic
counts due to occasional contamination and unhygienic processing environment
(Tariku, 2008).

Table 4: Effect of fermentation on carbohydrate components of kocho (dry weight)

Reducing Non­ Available

Period Starch Total sugar sugar(%) reducing carbohydrate
(week; (%) (%) sugar (%) (%)
0 68.8±2.5 4.3±0.1 1.9±0.3 2.4±0.4 72.1 ±0.4
1 68.5±1.4 4.0±0.4 1.5±0.1 2.5±0.3 72.1 ±0.5
2 65.9±3.7 3.5±0.3 1.2±0.2 2.3±0.0 72.0±0.7
3 57.3±5.1 6.2±0.7 2.6±0.4 3.6±0.2 61.5±0.6
4 55.3±0.7 4.5±0.8 0 .8±0.0 3.7±0.4 57.8±0.8
5 53.3±1.2 1.2±0.1 0.5±0.0 0.7±0.1 54.1 ±0.5
6 53.1 ±0.5 1.1 ±0.2 0.3±0.1 0 .8±0.2 54.0±0.9
7 53.1 ±1.2 0.3±0.0 0.3±0.0 - 54.0±0.7
Value', are means±S.D. of three determinations
 157 Tariku Hunduma

Table 5: Effect of fermentation on minerals and antinutritional factors of kocho (dry weight basis)

Iron P Tannin TIA Oxalic acid

Period Ca (mg/100g) (mg/1 OOg) (mg/1 OOg) (TIU/g) (mg/1 OOg)
(week) (mg/1 OOg)
0 288.0±23 3.4±0.1 154.3±11 676±41 8356±11 230±32
1 283.7±23 2.8±0.3 149.0±10 644+34 8137±120 228±23
2 271,7±17 2.5±0.4 140.7±12 440±23 7954±177 225±24
3 210.0±13 2.4±0.0 133.3±9 430+31 6520±123 224±22
4 186.3±21 2.4±0.0 131.7+11 422±30 5184±130 220±21
5 157.3±20 2 .2+0.3 131.3±7 390±32 4288±137 224±19
6 149.0±11 1.7±0.2 129.0±10 331 ±18 4079±112 235±32
7 142.3±10 1.6±0.2 128.0±21 333±21 3924±110 220+34
Values are means ± S.D. of three determinations. (Source of Tables 3, 4 &5: Kelbessa etal., 1997)

About 33 % of the products were known to be lost due to spoilage during storage
(Mogessie and Yewelsew, 1996). Enset cultivators store enset primary food products
for some extended time by using traditional techniques that are believed to improve
the keeping quality of the products. Wrapping the products with enset leaves and
burying them (most frequently used), changing wrapping leaves weekly and
changing pits consequently aerating, and changing leaves used for lining pits every
month were the methods used (Mogessie and Yewelsew, 996; Brandt et al., 1997).
About 57 % and 70 % of enset cultivators around Hawassa managed to keep bulla
and kocho respectively, two to three months without spoilage by using such
traditional techniques (Mogessie and Yewelsew, 1996). In the Gamo highlands
fermented enset primary food products were stored for five to six years (Olmstead,
1974). According to farmers in Wolaita, fermented enset primary food products can
stav up to eight years (Alemu & Sandfortd 1991). It is also reported that kocho can be
stored for up to 10 years or more by adding and thoroughly mixing fresh kocho
(Livuwork, 1993). It has been suspected that the long stay of the products could be
due to LAB dominantly found in the products and reduced pH of the products that
is inhibitory for most microbes that do not tolerate acid (Meaza and Berhanu, 1985;
Berhanu, 1987; Ayele and Berhanu, 1993). The occurrence of enterobacteriaceae to
the level of 6.95 x 106 cfu g-1 at early stage of the fermentation process and total
inhibition of them towards the final stage of the process. The inhibition was assumed
to be resulted possibly due to the antimicrobial substances produced by lactic acid
bacteria (Tariku and Mogessie, 2011). However, roles of LAB in the preservation of
enset primary food products and as to why those traditional methods improve the
keeping quality of the products need further investigation.

Loss of enset primary food products at this stage, after long years of investment on
the plant through cultivation and processing is extremely serious. In addition to loss
of the products, such spoilage problem may cause health risks, as most molds can
produce heat resistant mycotoxins. Works done on post-fermentation loss of enset
primary food products are very minimal and no scientific attempts have been made
to improve storage facilities for the products. Though all enset foods are well heat
 158 Research Experience and Achievements on Traditional Enset Fermentation

treated, some related public health problems such as existence of mycotoxins really
needs scientific analysis.

Research Gaps and Recommendations

Enset plant has been traditionally fermented for generations in an earthen pit into
carbohydrate rich primary food products. This age-old traditional method has been
used for generations without any scientific modification. Studies done so far on
traditional enset fermentation and its primary food products mainly focused on
evaluation of effect of length of fermentation on carbohydrate and calcium content
(Taye, 1984), isolation of microorganisms from fermenting mass (Berhanu, 1987),
isolation of microorganisms from enset primary food products from market (Meazaa
and Berhanu, 1985; Mogessie and Yewelsew, 1996) and biochemical changes during
fermentation (Kelbessa et nl.r 1997).

Okafor (1992) reviewed the extent to which some important fermented foods of sub-
Saharan Africa had progressed toward commercialization. He measured the stage
tha: each food had attained on a scale of eight. Very few Ethiopian fermented foods
included in his review were categorized at a stage of scale one and two, which means
tha: the works done focused mostly on isolation and determination of role(s) of
microorganisms involved in tire food, respectively.

The recommended research priorities on traditional fermented foods include; (1)

improving and understanding of the fermentation processes; (2) refining of the
processes; (3) increasing the utilization of the processes; and (4) developing local
capabilities (Anon, 1992). Therefore, improving the understanding of traditional onset
fermentation process is very' crucial to scientifically promote the process forward.
According to Yong (1992), without the knowledge of the art of traditional food
fermentation, scientist cannot provide scientific explanation for the process and
proceed to provide further assistance in improvement of the process.

Though, traditional enset fermentation has been practiced for generations here in
Ethiopia, there are lots of challenging issues yet to be considered. Farmers in
different regions use different maturity signs. However, identification of appropriate
maturity indicators which coincide with time of high nutrient content is important.
Traditional enset processing for fermentation is tedious, unhygienic and generally
done; by women using locally made equipment. Modification of the processing
equipment can minimize energy and time consumption and can also improve
sanitary conditions during processing for fermentation. Attempts should be done to
undertake controlled fermentation studies with selected culture starters and
optimize the process into modern food processing technology. Enset is a multi­
purpose crop and it is everything for the farmers (Brandt et al., 1997). According to
the authors of, " Lost Crops o f Africa", no plant on earth can match enset for the
 159 Tariku Hunduma

number of products it provides (National Academy Press, 2006) and hence can
certainly contribute to food security of the country.

References
A len u , K. and Sandford, S. 1991. Enset in North Omo Region. Farmers' Research Project
Technical Pamphlet No. 1. Farm Africa, Addis Ababa
Anioa-Awua, W.K.A., Frisvad, J.C., Sefa-Dedeh, S. and Jakobsen, M. 1997. The contribution
of molds and yeasts to the fermentation of 'agbelima' cassava dough. Journal o f A pplied
M icrobiology 83:288-296.
Amoa-Awua, W.K.A, Appoh, F. and Jakobsen, M. 1996. Lactic acid fermentation of cassava
into agbelima. International Journal o f Food M icrobiology 31: 87-98
Anon.1992. Research Priorities in Traditional Fermented Foods. In: Application of
Biotechnology to Traditional Fermented Foods. Report of an Ad Hoc Panel of the Board
on Science and Technology for International Development, National Research Council.
National Academy Press. Washington D.C.
Ayele, N. and Berhanu, A. G. 1993. Inhibitory activity of fermented kocho on spoilage and
pathogenic bacteria. In: Tsedeke Abate, Hiebsch C, Brandt S. A. and Seifu Gebremariam
(Eds). Enset-Based Sustainable Agriculture in Ethiopia. Proceedings from the
International Workshop on Enset held in Addis Ababa, Ethiopia, 13-20 December 1993.
Bela':, A., Hunduma, T. Fekadu, E., Negisho, K. and Ali, A. 2008. Gender-based analysis of
production system in Ambo. In: Chiche, Y. and Kelemu, K. (Eds.), Proceedings of the
workshop on gender analysis in agricultural research (pp. 27-29) Addis Ababa, Ethiopia
Berhanu, A. G. 1987. Kocho fermentation. /. Appl. Bacterial. 71:514-518
Brandt, S. A. Spring A., Hiebsch, C , McCabe, J. T., Endale Tabogie, Mulugeta Diro, Gizachew
Wolde-Michael, Gebre Yntiso, Shigeta M., and Shiferaw Tesfaye. 1997. The "Tree Against
Hunger". Enset-Based Agricultural Systems in Ethiopia. American Association for the
Advancement of Science with Awassa Agricultural Research Center, Kyoto University
Center for African Area Studies and University of Florida, Washington, DC, USA, 66 pp.
Dereje, B. and Endale, T. 2003. Enset Production in West Shewa Zone. Research Report No.
49. Ethiopian Agricultural Research Organization (EARO),Addis Ababa, Ethiopia. 45 pp.
Jay, M. 2000. M odern Food M icrobiology 6th ed. Xii + 635pp. Aspen Publishers, Inc.
Gaithersburg, Maryland.
Kelbessa, U., Alemu, F. and Eskinder, B. 1997. Natural Fermentation of Enset (Ensete
\ientricosmu) for production of Kocho. Ethiop. /. Health Dev. 11: 75-81.
Kelbessa, U., Ayele, N. and Melaku, U. 1993. Traditional enset-based foods: survey of
processing techniques in Sidama. In: Tsedeke Abate, Hiebsch C, Brandt S. A. and Seifu
Gebremariam (Eds). Enset-Based Sustainable Agriculture in Ethiopia. Proceedings from
the International W orkshop on Enset held in Addis Ababa, Ethiopia, 13-20 December
1993.
Liyuwork, Z. 1993. Kocho processing in southern and southwestern Ethiopia. In: Tsedeke
Abate, Hiebsch C, Brandt S. A. and Seifu Gebremariam (Eds). Enset-Based Sustainable
Agriculture in Ethiopia. Proceedings from the International Workshop on Enset held in
Addis Ababa, Ethiopia, 13-20 December 1993.
Meaza, G. and Berhanu, A. G. 1985. Studies on the microbial flora of Kocho and bulla
purchased from markets in Addis Ababa. SIN ET: Ethiop. Sci J. 8:29-36
 160 Research Experience mid Achievements on Traditional Enset Fermentation

Mehtzun, T. and Yewelsew, A. 1994. Study of "Enset" Processing and Development of

'Enset" Processing Tools in the Southern Regions of Ethiopia. Aca/Noragric Research
Collaboration Project.
Million, T., Eshetu, A. and Endriase, G. 2003. Enset-Based Farming Systems of Masha
Woreda Sheka Zone. Report No. 51. Ethiopian Agricultural Research Organization
(EARO), SNNPRS Agricultural Research Institute, Addis Ababa, Ethiopia. 25 pp.
Mogessie, A. 2006. A review on the microbiology of indigenous fermented foods and
beverages of Ethiopia. Ethiop. J. Biol. Sci. 5(2): 189-245.
Mogessie, A. and Yewelsew, A. 1996. Microbial spoilage of market bulla and kocho, traditional
Ethiopian processed food products from enset ( Eusct ventricosum ). Ethiop. /. Agric. Sci. 15:
121-130.
National Academies Press. 2006. Lost Crops o f Africa. Volume II. Vegetables.
http: / /w w w. nap.ed u /ca ta log /11763. h tin
Okafor, N. 1992. Commercialization of Fermented Foods in Sub-Saharan Africa. In:
Application of Biotechnology to Traditional Fermented Foods. Report of an Ad Hoc Panel
of the Board on Science and Technology for International Development, National
Research Council. National Academy Press. Washington D.C.
Olinstead, J. 1974. The versatile enset plant: its use in the Gamu highland. /. Ethiop. Stud
12(2)147-153.
Oyewole, O.B. 1992. Cassava Processing in Africa. In: Application of Biotechnology to
Traditional Fermented Foods. Report of an Ad Hoc Panel of the Board on Science ant
Technology for International Development, National Research Council. National
Academy Press. Washington D.C.
Sandford, J. and Helen, K. 1993. The effect of gender on resource contribution, decision
making and influence: A comparison between enset, tef and maize. In: Tsedeke Abate.
Hiebsch C, Brandt S. A. and Seifu Gebremariam (Eds). Enset-Based Sustainable
Agriculture in Ethiopia. Proceedings from the International Workshop on Hnset held in
Addis Ababa, Ethiopia, 13-20 December 1993.
Seifu, G. 1993. Enset research in Ethiopia: 1976-1984. In Tsedeke Abate, Hiebsch C , Brandt S
A. and Seifu Gebremariam (Eds). Enset-Based Sustainable Agriculture in Ethiopia
Proceedings from the International Workshop on Hnset held in Addis Ababa, Ethiopia,
13-20 December 1993.
Shank, R. and Chernet, E. 1996. United Nations Development Programme Emergencies Unit
for Ethiopia. A Linear Model for Predicting Hnset Plant Yield and Assessment of Kocho
Production in Ethiopia, http://wwvv.arrica.upenn.edu/eue web/e nset96.htm . Accessed
on 6 July 2007.
Shigeta, M. 1993. Multipurpose utilization of enset among the Ari in south-western Ethiopia.
In : Tsedeke Abate, Hiebsch C, Brandt S. A. and Seifu Gebremariam (Eds). Enset-Based
Sustainable Agriculture in Ethiopia. Proceedings from the International Workshop on
Enset held in Addis Ababa, Ethiopia, 13-20 December 1993.
Steinkraus, K.H. 2002. Fermentations in World Food Processing. Com prehensive Reviews in
Food Science and Food Safety. Inaugural Issue.
Tariku, H. 2008. Effect of altitude on microbial succession during traditional Enset (Ensete
juen tricosum ) fermentation. MSc thesis. Addis Ababa University, Ethiopia.
Tariku, H. and Mogessie, A. 2011. Effect of altitude on microbial succession during traditional
enset (Ensete ventricosum ) fermentation. International Journal o f Food, N utrition and Public
Health, 4(1): 39-51
 161 T ariku H undum a

Taye, B. 1984. Evaluation of some Enstete ventricosum clones for food yield with emphasis
or the effect of length of fermentation on carbohydrate and calcium content. Trap. Agric.
(Trinidad) vol. 61 No.2.
Taye, B. 1993. An overview on enset research and future technological needs for enhancing its
production and utilization. In: Tsedeke Abate, Hiebsch C, Brandt S. A. and Seifu
Gebremariam (Eds). Enset-Based Sustainable Agriculture in Ethiopia. Proceedings from
the International Workshop on Enset held in Addis Ababa, Ethiopia, 13-20 December
1993.
Yong, L. F.M. 1992. Future Directions. In: Application of Biotechnology to Traditional
Fermented Foods. Report of an Ad Hoc Panel of the Board on Science and Technology for
International Development, National Research Council. National Academy Press.
Washington D.C.
 Research Experiences on the Development
of E n set Processing Equipment
Friew Keleinu
Melkassa Agricultural Research Center
P.O.Box 436, Nazareth, friewkelemu@yahoo.coni

Abstract
The processing o f enset, both the root ami the pseudo-stem has been a hack
breaking exercise imparting physiological stress oil the operators, who arc usually
women. To overcome these problems, concerted efforts have been made by
individuals and few organizations including the Ethiopian Institute o f
Agricultural Research (EIAR) and Debub University. The Agricultural
Mechanization Research Division o f EIAR at Melkassa made major
improvements on the clamping component o f the pseudo stem decorticating device
and amicho pulverizer. Different generations o f devices were developed in the
process. The first generation teas tested at Weliso and Eudeber, the second
generation at Chencha, and Areka. Based on the feed-back gathered from the above
places, the required mollifications were made and tested at Tikur Enchine and
Buea. The clamping device on the pseudo-stem decorticator, avoided the
stretching o f the leg above the iraist position for holding in place the pseudo-stem.
The newly improved amicho pulverizer has dramatically cut down the time o f
pulverizing to less than an hour for the whole corm compared to three hours
taken in squatting position in the traditional way. Four manufacturers from the
Kofele and Butajira were trained for one week in a ;w orkshop organized by the
Melkassa Agricultural Mechanization Research Center at Melkassa. Periodic
improvements are being made on the device. The device operated smoothly on
smaller diameter and softer varieties, while harder varieties were better operated
by splitting the whole into two halves as obseived at Tikur Encluue. The aright
o f the corm pulverizer is still higher than the required level and there is no
provision for mobility to move it from one village to the other. The research is
required to work hard and overcome these challenges. Despite these, the improved
devices will lessen the drudgery on the rural women, who are responsible for the
processing o f enset. It will also help to create manufacturing jobs and or private
mobile enset processing operators in the rural areas similar to tractor hiring and
combine service providers.

Introduction
Enset is a banana like plant, which is a major source of food for a substantial number
of the population. It has three major parts: the leaves, pseudo-stem and corm. I he
decorticated pseudo- stem and the pulverized corm are buried to ferment in a pit
 163 Fricxv Kcleinii

lined with the lea v es for few days. It is the source of food and fiber, where the
fermented product from enset is processed into various types of food and its fiber is
used in the manufacturing of ropes. It is widely grown in the highlands of South-'
Central and Southern Ethiopia supporting more than ten million people. The Gurage
/one is one of the important administrative zones where enset is widely grown. The
two main enset products utilized as food are locally known as kocho and bulla. Kocho
is the fermented product obtained from the corm and pseudo stem of enset. Bulla is
made by dehydrating the juice collected during the decorticating of the pseudo-stem
and grating of the corm. It is readily available and is a very suitable crop for human
consumption. It can be made into a number of traditional products either in the form
of moist or dried product by using simple technology.

Traditional Method of Processing

The harvesting ol etiset in most parts of the country begins at the end of October
and continues till January. For processing the whole plant is uprooted with the help
ol a special knife and brought to the processing site - an open place within the
plantation (Fig 1). A wooden plank is put at an angle of about 40 0 against a pole. A (
woman sits in front of the plank keeping a pseudo- stem piece with its concave side
against the plank securing it in position bv putting her foot on it as high as possible
(Fig. 2) (Kelbesa Urga cl.al., 1996).

The corm after removing the rootlets is grated (Fig. 3) and thoroughly mixed. The
mass is called amicho. Graters are traditionally made from animal shoulder bones or
pieces of wood. Amicho is squeezed by trampling before being placed in the
fermenting pit. It is believed that amicho will not ferment by itself unless mixed with
separately fermented enset, kocho. As a starter the amicho is rubbed with shigiclo - a
bl«ick mass picked from the rotten enset leav es. The shigiiio wrapped amicho is left for
five to seven days until it becomes soft and pulpy. This softened amicho is grated into
finer mesh and mixed thoroughly with the fresh amicho, pressed down by trampling.
It is covered with enset leaves and is left to ferment. The enset product is considered
ready for consumption after 90 days from the initial processing day, but can also be
kept for one or more year (Kelbesa Urga et.al., 1996). Generally the processing of
enset is a back braking exercise, which is a burden on the rural women in particular.
Earlier report revealed (Tave Bzuneh, 1996) that only women carried out enset
processing. Extracting food and fiber from enset is highly labor intensive. It was
found that it took about 1.5-2 days to fully decorticate, pulverize and knead a
matured enset. Taye Bzuneh indicated that the orientation of research should be to
improve the efficiency of harvesting, extracting food and fiber of enset in order to*
enhance the development of cottage industries at village level.
 164 Research Experiences on the Development o f Etiset Processing Equipment

Fig. 1 Preparation of enset

Achievements
In the early 1970s efforts were made by different individuals and organizations to
come up with different processing devices. However, the device, which was
originally developed to extract fiber from Abaca in the Philippines, did not save the
food element (Taye B. 1996). Atoro's device, which included chains and sprockets
were found difficult to maintain and repair and Milton's device was not tested (IAR
Annual Report, 1977/78)

Institute of Agricultural Research (IA R )

The Engineering Department of the IAR was established in 1977, comprising of Farm
Power and Machinery and three other sections. The Farm Power and Machinery
 165 Frieze Keletnu

Section started work on two projects namely Water Lifting Devices and enset. The
work on enset started in May 1977. During the 1984/85 research seasons, efforts
were made to improve the traditional enset decorticating devices. The project came
up with a number of enset processing devices (Fig. 4), which included an inclined
plank/clamp pseudo-stem decorticator, corm graters and bulln squeezer. All these
were meant to improve the ergonomics, with simple clamping mechanisms at a
normal operating position and efficiency. After a number of trials both were
achieved. Most of these are currently taken up by the Sodo Rural Technology Center
arid are distributed to users in the Sidama Region. According to Liyuwork Zewde,
1996, these devices were very much accepted by the women in her study area,
because they increased efficiency, reduced wastages and were found comfortable for
operation.

The second device of corm grater has some resemblance to that of the carpenter's
jack plane and comprised a set of thin plates in front of the main blade. The efficiency
of this device was three times that of the traditional tool, but the sliced material
varied in size, which delayed fermentation. Further effort was made to refine the
operation of the device and finally an adjustable rotary type blade pulverizer was
developed, which was much better than the previous one.

Regarding the bulln extraction devices, two flat board squeezer and bucket type
squeezer were developed. The first one was developed from two flat boards hinged
at the back, but lacked the strength for the intended job.

The second device was made out of metal frame wooden box with sliding piston in
it. Many fine holes were drilled on all sides of the wooden box to let out the
squeezed bulln. This device was found efficient and satisfactory.

T h e Fo rm er A nassa Agricultural College

The Awassa Agricultural College, Departments of Agricultural Mechanization and
Food Science conducted a study on enset processing tools. The study was conducted
in a 200 km radius covering the enset culture of the Sidama and Wollita zone. The
study group finally managed to develop a handy equipment made from local wood
without any metallic attaching material (nail). The equipment has three major fitting
pasts, which are assembled for the operation. It has a decorticating board with a
clarnp at the upper end, a pulverization chamber at the lower end and a sit with a
prc vision for placing a kneading tray on the other side, fitted at 45 degree angle in
slo :s made on both sides of the main stand. The device was simple and thought to be
within the economic reach of the target group. The dimension was within the
preferred measurement range, so that once the equipment was set decortications,
pu verization squeezing, shredding operations could be done comfortably at ease
(Fig. 5). The group report indicated that the women operators at the test sites were
veity happy with the equipment. Finally, the researchers have proposed to further
pursue the work so that the technology will be widely disseminated among the enset
 166 Research Experiences on the D evelopm ent o f Enset Processing Equipment

farming communities. The research was conducted under a research collaboration

, project financed byy ACA - NORAGRK ( Mehtzun et al., 1994)

Fig. 5. The Awassa College processing device

Current efforts

The research work on the improvement of enset processing devices at the

Agr cultural Implements Research and Improvement Center (AIRIC) at Melkassa
was reinitiated in the late 1990s. A project proposal was developed, reviewed at the
different forums and accepted. The proposal had had three components that
include: survey, prototype development and evaluation. The work started in
Sep ember 2000 with a survey. Each component is discussed.

4
 167 Friew Keleuut

Survey
A survey was conducted in 2000 in two districts to gather information and learn
about the enset culture with a major focus on the processing technique. The study
started with a structured questionnaire where discussion was held with farmers and
the district agricultural bureaus officials. While driving from Welkite to Endeber one
hardly sees any crops but only enset, Eucalyptus or open grazing land. The cattle that
one observes have small body confirmation and hardly observe any oxen. The area is
dominated by hoe culture which limits cultivation to perennial crop like enset. The
few cattle observed could be partly attributed to the existence of tsetse fly. The
Endeber District/Wereda population is mainly dependent on enset cultivation for
food and as a means of income for buying necessary goods. At Endeber,
Yeferenzeige Keble, it was witnessed that in one house hold, the owner had hired
about six ladies to decorticate the pseudo-stem and the amicho at a rate of two birr
per pseudo-stem and two birr per three corm with provision for the necessary daily
maintenance i.e. food and coffee for the daily operation on top of the mentioned rate
per head. At the time of the study, this practice was the accepted norm in the
community. The operators complained that it was indeed a back breaking exercise
as the women had to squeeze in a cramped position for the whole day and the need
for improvement on the decorticating devices is indisputable. The pseudo-stem-
decorticating device is a split bamboo beveled at one end, though looked inefficient
There was some reservation on the user's side to replace this device with a metallic
pait as there was suspicion that it may not enable them, to recover the fiber, which
fetched them substantial amount of money. The amicho- grating device, the gebangeba
w a s reported to be tiresome and inefficient. The ladies suggested that any
improvement on the devices should include a provision for sitting arrangement
during operation.

In other community called Bucha Mender, at Endebir the ladies had the chance of
using improved decorticating devices, pseudo-stem decorticator and bulla squeezer
tha: were introduced from the Sodo Rural Technology Center. They commented that
the clamping arrangement on the decorticator was time consuming and at times it
cannot be done from sitting position and because of the flat nature of the plank it was
not efficient in fiber recovery. There was mixed filling on the bulla squeezer. Some
said "it produces finer quality bulla, which could be for luxury use, but reduces the
volume and the money from selling the product". Others commented that" it did not
totally extract the whole bulla and did not produce comparable volume" to their
traditional practice. Finally, they opted for the improvement to be done on the above
short comings and a corm pulverizer be developed as so far they had not seen any
device in the area. Similar comments were received from the areas visited at Welkite
and Weliso.

Taking the above comments into account the Agricultural Mechanization Research
Program of the Melkassa Research Centre proposed to revisit the previous activities
 168 ____ Research Experiences on the Development o f Enset Processing Equipment

on enset and develop a better pseudostem decorticator and corm pulverizer

technologies.

T h e F ir s t G en eratio n D eco rtica to r

The major modifications were planned mainly on the enset decorticator and corm
pulverizer . The modified decorticator was provided with a grating board, with a
better provision for clamping and better operational position and an efficient
scraping tool. The decorticator was fabricated in such away that it could be handled
in the local wood workshop.

The first pulverizer model was designed and fabricated based on the IITA design of
cassava grater (Fig. 7). The device had a feeding mechanism and a rotating grater
fabricated from galvanized steel. The rotating part was originally made from sheet
metal, but later on was replaced with animal horn. The basic material for this specific
design was wood and mild steel with the understanding that it will be replaced with
corrosion free material in the final prototype. The grating chamber was fabricated
from galvanized sheet and the grater was fabricated from animal horn. This device
uses a feed chopped in a rectangular shape for ease of handling. A wheel sprocket
drive was added to overcome the drive force requirement (MARC, Progress Report,
2000 ).

Testing
First, enset corm was brought from Weliso and sliced to manageable size and was
tested in the laboratory. Later, the testing material was planted in the Center. It was
ready for use in two years after planting. Then, it was cut and removed from the
field. It was further cut in rectangular pieces to fit the grating chamber. Data on time
taken, output/ efficiency, rate/degree of pulverization and down time were recorded
(Tables 1 and 2).

Table 1. Performance of pedal operated corm pulverizer (Nov 12.2002)

Starling Stoppage time Finishing Total time Output Rate Unpulverized

Weiqnt (g) time (min) time (hr) (min.sec) (^ • ) (kg/min) (kg)
2267 2:05:03 - 2:06:04 1:06 1.314 1.28 0.790
3235 2:30:47 4.78 2:36:47 1:13 1.951 1.59 0.719

Table 2. Performance of hand operated corm pulverizer

Starting Stoppage time Finishing Total time Output Rate Unpulverized

Weight (g) time (min) time (hr) (min:sec) (kg.) (kg/min) (kg)
3953 9:46:45 9:48:45 2:00 2.470 1.25 0.572
3216 1:47:28 1:49:17 1:49 2.399 1.32 0.628

The first design required relatively higher force, which gave a lower output. A
second generation corm pulverizer had a higher force requirement observed in the
first design. This was more efficient compared to the first generation. With the new
 169 Frieiv Kelemu

design it was possible to pulverize at a rate of 0.6kg/ min compared to 0.1 kg/min
with the previous generation. It was found efficient and comfortable for the operator.
Still the work was further refined and a better result was obtained as indicated in the
above tables.
The refinement was mainly on the driving mechanism. A pedal operated pulverizer
with a sprocket chain driving mechanism with a saddle like arrangement for the
operator as in a bicycle. This provision enabled to exert more force using the legs.
This resulted in a higher rate of work (Table 1) compared with the hand operated
pulverizer (Table 2).

During testing on farmers' field the farmers complained on the wastage because the
unpulverized corm was not acceptable. Thus, there was a suggestion to come up
wiiih a device that could accommodate the whole corm in one go (MARC, Progress
Reports, 2001/ 2002). The second model was fabricated by adjusting the cutting
clearance and using galvanized steel as the cutting material. A wheel sprocket drive
was added to overcome the force requirement observed in the previous design. To
reduce cost and overcome corrosion problem, the third generation was fabricated
using animal horn as the grating material

The improved decorticator was fabricated in two versions one suitable for standing
position and the other one suited for sitting position. Both were tested in Yem and
Girnbo with the collaboration of Winrock International. Favorable response was
received from the users which encouraged the use of the popularization technology
(MARC Progress Report, 2002/2003). It was not easy to get quantified data except
qualitative information on the stress it imparts and the rate of decortications and
pulverizing by an average person. The pseudo-stem-decorticating device is a split
bamboo beveled at one end. The operators here again did not like this device with
metallic part, which might not enable them to recover the fiber, which is a good
source of income,
 170 Research Experiences on the Development o f Enset Processing Equipment

Despite the different designs forwarded to the users, farmers opted for a machine
that takes up the whole corm in one piece. Therefore, they argued that such a device
very well fits to their system and there will be minimal loss. Taking this into account,
a fourth generation of decorticator was fabricated using the late Araya Kebede's
design (1984) with a major modification on the feeding mechanism. The results of the
studies conducted to date, the major problem has been on the feeding rate during
grating, which is usually too much, burdens the operator and lowers the durability
of the cutting part. A thorough study was conducted to set the optimum feeding rate,
which lessens the load on the operator. A feeding mechanism, which delivers a
feeding rate of 2 mm per revolution of the cutting disc was designed. The equipment
was fabricated in the workshop and was tested on station arid at Areka Research
Center in collaboration with the Enset Research team. It was also demonstrated at
Chencha in collaboration with World Vision Ethiopia where it was found operational
and acceptable. The equipment handled a corm up to 45 cm in diameter and of 60
kg weight in less than an hour. At Chencha the corm was found to be as large as
80cm in diameter and 100cm in length. It required to be split into two parts as the
diameter was larger, but was found acceptable for the above mentioned diameter
and weight classes of corms. Though the pulverizer was found efficient in all the
tested areas , farmers still complained on the weight of the equipment and operation
technique, as it was different from the way they were used to. They even suggested
it to be motorized. Other feedbacks on some left over corm due to the various size of
the corm were reported at different sites.
 171 Friew Kelemu

The original prototype was modified to accommodate the whole corm. This was
based on field data. The feeding section was made large enough to accommodate
corms as large as 40cm in diameter and 60cm in length. A feeding mechanism which
delivers at a rate of 2mm per revolution of the cutting disc was designed. The
equipment was fabricated in the workshop and was tested on station and on farmer's
field. At Koefele (Fig. 7) and Butajira favorable responses were received from the
farmers. A sample weight of 20.798 kgs was pulverized in 11.5 minutes. Efforts were
also made to reduce the weight, and the size was increased to accommodate bigger
sizes of corm as well.

Fig. 7 Testing of corm pulverizer at Kofele

M anufactiiring
Four local manufacturers generally involved in general metal works were selected
from Butajera and Kofele area in May 2006 and were trained at AIRIC workshop on
the manufacturing of the corm pulverizer. As part of the training program they were
able to manufacture one prototype in a group (Fig. 8 ).
 172 Research Experiences on the Development o f Enset Processing Equipment

Fig. 8 . Trainees at work

Field D em onstration and Feed B a c k s

The /vork on the corm pulverizer did not stop at the training of Kofele and Butajira
wouid be manufacturers, but effort was made to introduce the technology to other
etisei growing areas, in the Ambo Weliso and Weskit areas. Effort was also made to
introduce the technology and collect feed back and make the technology more
amenable to the communities. In the mean time the improvement work has
continued on station.

To reduce the load on the operator, the handle for driving the operating wheel was
fabricated with a provision of three different radii of 26 cm (the original), 35cm and
40cm, with the assumption that the bigger radius will require less effort from the
operator to handle the load. From our observation, on tests made on 5 - 6 May 2009
(Table 3) on corms brought from Kulumsa, a higher rate of pulverization was
achieved using the 35mm radius arm than the 40cm radius arm, which the users
found also, a bit awkward requiring the operator to bend more as the handle tips
dow n in relation to the center.
 173 Friezv Keleimi

Table 3. Test result of pulverizing using variable arm length corm pulveizer on Enset Henawa Variety

A) Test results , 5 May 2009

Operator's
Nc Arm radius (cm) pulverized weight(Kg) Time(min: sec) weight (kg) Rate(kg/min’
1 40 1.346 1:44 80 0.774
2 40 1.242 1:13 63 1.014
3 40 1.589 1:33 56 1.019
Surn 4.177 4:30 0.923
1 35 3.127 3:.0.4 78 1.040
2 35 1.477 1:21 60 1.096
3 35 2.049 1:38 60 1.261
4 35 1.539 1:17 56 1.201
5 35 1.728 2:59 60 0.581
Surn 9.920 10:15
Rate' 0.969
■Pulveried wt(kg)/Time (min:sec)

Experience at T ik u r Enchiene
Based on the recommendation of the Ambo Research Center, the team visited and
demonstrated the enset processing devices to farmers at Tikur Enchetie on 22 May
2008, where both the decorticator and pulverizer were demonstrated and farmers
had shared their experiences.

The corm pulverizer was operated by different groups of men, women and young
people of both sexes. They have managed to operate the machine and still
complained about the effort required to operate it. There was a big difference in the
ease of operation compared to the experience at other sites. Finally, there was a
recommendation to split it into two parts and pulverize each piece separately. This
was done as recommended and the operators found it very easy to handle. This
comment was given by one of the women operators who tried the equipment under
the two different conditions. Generally a favorable response was given and the
pulverizer operated much better on corms of<40cm diameter. A split corm was
found easier to handle even in diameter class of >40cm.

Decorticator
The decorticator was tested by the women. They found it comfortable for the
operation. Comment was given on the clearance of the fixing mechanism, which they
recommended to have a low clearance or a tighter fit. They also recommended that
the fixing pedal should be placed higher so a tighter grip is attained.
 174 Research Experiences on the Development o f Enset Processing Equipment

llu i. Sodo D istrict Ju ly 3 0 4 1 , 2 0 0 9

Demonstration was conducted at Bui, Sodo district in Gurage zone from July 30-31 in
collaboration with Self Help NGO group (Table 4 and (Fig. 9). The enset decorticator,
corn’ pulverizer and bulln squeezer were demonstrated to a women group
comprising of sixteen women. Their responses were positive for the decorticator and
pulverizer. They commented that the decorticator takes more time and the scrapping
blade needs to protrude out so that it will be effective in scrapping in any position.
Comments were given to modify it so that it will have provision for scrapping in the
standing position as well. They commented that a simple rope hole arrangement
could be good enough to hold the enset sheath in place. The Self Help group asked
for some standing type decorticators, which were made ready at the workshop and
extended to them on payment basis.

The corm pulverizer was demonstrated to the group using a 45 cm- length and a
diameter of 20-30cm corm. Most of the ladies had shared their experiences and did
not have much complaint on the ease of operation. They commented that the corm
was pulverized to a fine degree and would have liked it to be less fine or coarse.
They were told that this could be adjusted by changing the size of the blade serration
to achieve a fine or coarse degree of pulverized corm. They also commented that it
should have a better provision for mobility, so that they can easily transport it from
one place to the other.

Table 4. Field demonstration results of corm pulverizer at Bui town

Diameter Pulverized Arm radius Time Rate

No___________ (cm) weight (kg)________ [cm]__________(min:sec)_______ (kg/min)
1 20-30 1.2 35 1:37 0.743
2 0.8 35 1 :5 0.738
3 1.2 35 1:8 1.058
4 1.0 35 1:10 0.857
5 1.6 35 1:20 0.959
6 2.0 35 1:30 1.333
Mean 1.046
 175 Frieiv Keletnu

Fig. 9 Bui ladies' group pulverizing corm using the newly developed equipment, Bui, July 2009

C urrent and Fu tu re D irections

The Agricultural Mechanization Research Process is currently working on
simplifying the operation and mobility of corm pulverizer devices as recommended
by farmers. The cutting disk is now designed with its major mass concentrating at its
outer rim (Figure 10), thus drastically reducing its weight. Similar works are being
done on the mobility aspect as well.

Fig. 10. The newly developed pulverizing wheel

Dissem inated Technology

Some decorticators are delivered to users in recent years.. Thirty enset decorticators,
one corm pulverizer and 16 decorticating blade with the holder were sold to
 176 Research Experiences on the Development o f Eusct Processing Equipment

individuals and groups through Self Help Development International in 2008/09

seasons.

(la p s and G ap Analysis

Generally the processing of enset has been a back breaking exercise and a very
difficult task for the rural women. In spite of the fact that efforts have made by
different groups, still there was the urgent need to come up with better processing
equipment. The women involved in the operation recommended for improvement of
the over all kocho processing methods in order to save energy, time and to make it
more hygienic.

Though substantial amount of work has been done to simplify and ease the
operation, it is still difficult to say that it is amenable to all working age and weight
groups. The corm pulverizer is not versatile enough to be moved from one village to
another. The bulla squeezer is not big enough to handle larger charge. All the
processing gadgets do not have any provision for automatic operation Farmers have
com mented that the degree of pulverization need to be as required by the user and
therefore, a mechanism is needed to be in place to control the outcome.

Though enset is an important crop, the demonstration and extension work has been
minimal to effect the dissemination of the technology among the user communities.
The participation of the private sector especially manufacturers is negligible, which
otherwise would have been instrumental in the multiplication and dissemination of
the technology. The insignificant level of involvement of financial institutions such as
unions is a major gap which needs serious attention.

The emphasis given to capacity building, staff training and strengthening the
research on the post harvest engineering aspect is far from adequate.
A number of steps ought to be taken to further alleviate the problem of processing
and hence making a healthy product so that people involved in the production,
processing and marketing will all be beneficiaries.

Discussion and Kcconinicndation

Enset is a very important life saving crop for over 20 million people in the country.
However, its processing aspect has not been given due consideration that allocation
of fund, training of personnel and strengthening of the engineering wing, which is
instrumental in the delivery of the processing technology are negligible.

Despite these, the research system needs to build momentum and address the
current issues at hand. The issues raised on efficiency of gripping, the enset
decorticator, the weight and mobility questions on the pulverizer need to be
 177 Frieiv Kelenut

addressed as soon as possible as priority issue. The demonstration work on the

available technologies should be aggressively' pursued to create awareness among
the enset growing communities at large. Private manufacturers such as micro
enterprises need to be trained and be involved in the manufacturing of the newly
improved devices. The research system need to be strengthened, staff need to be
trained in post-harvest engineering science and adequate financial support be
provided to scale out and deliver the required technology to make a difference in the
livelihood of the enset farmers and the population at large.

Inferences
Anonymous. 1978. Agricultural Engineering Department Research Program. Melkassa
Agricultural Research Center, Progress Report, IAR, 1977/78. p p !3
Ar.iya Kebede. 1984. Development of E nset processing devices. In Agricultural Implements
Research and Improvement team progress report 1984/1985. pp50-55
Kelbesa Urga, Ayele Negatu and Melaku Umeta. 1996. Traditional E nset based foods. Survey
of processing techniques in Sidama. In: Proceedings of the International Workshop on
E nset, held in Addis Ababa, Ethiopia,13-20 December, 1993. pp 305-313
Liyuwork Zewde. 1996. Kocho processing in southern and southwestern Ethiopia:
preliminary results. In: Proceedings of the International Workshop on Enset, held in
Addis Ababa, Ethiopia,13-20 December, 1993. pp315-320
Mentzun Tedla and Yewelsew Abebe. 1994. Study of enset processing and Development of
E nset Processing Tools in the Southern Region of Ethiopia. ACA/NORAGRIC research
collaboration project.
Taye Bzuneh. 1996. An overview on Enset research and future technological needs for
enhancing its production and utilization. In: Proceedings of the International
Workshop on enset held in Addis Ababa, E thiopia,!3-20 December, 1993.pp 1-12
 Research Achievements and Experiences
011 Enset Food Products

Mulugeta Team ir and Adane Tilahun

Melkassa Agricultural Research Center, P.O.Box 436,
E-mail: mutens@uuhoo.com, ailtilnhuii@gmnil.com

Abstract

Different researches were conducted by different scholars on enset crop processing and
utilization. In this paper major findings are reviewed. In Ethiopia more than 20 percent of
the population depends upon enset as a source o f food, means o f cash earning, insurance
against hunger, feed and fiber fo r industrial use. Enset is a starchy crop rich in
carbohydrate and some minerals (Fe, Ca and Zn). It is however very poor in fat, protein,
and vitamin A. The two popular foods prepared from enset by fermentation process are
kocho and bulla. The fermentation process is initiated by Leuconostoc mesentroides bacteria.
Koclw become easily contaminated by microorganisms when removed from the fermenting
pits. Spoiled kocho and bulla had high counts o f aerobic mesophilic bacteria, Micrococcus
and Bacillus species. The protein density and vitamin A o f kocho and bulla can be increased
by addition o f legumes and pumpkin pulp, respectively. The yield o f enset in terms o f dry
matter production and energy production per unit space and time is much higher than
cereal and other root and tuber crops. Starch, in small amount, from enset is currently
produced industrially by a local factory and its chemical composition, physicochemical
properties, and amylase content have been analyzed and documented. Modernization o f
fermentation process o f enset and other traditional foods must be given due attention. Food
fortification and supplementation o f enset food with nutrient or protein rich foods must be
developed. The bio-availability, bio-efficacy, other nutritional and functional properties of
the enriched product should be studied. The already achieved research outputs particularly
in the area o f food supplementation at the enset growing areas fo r vulnerable groups should
be sensitized. There should be an extensive research and development work on local
production o f starch from indigenous crops

Keywords: Enset, Kocho, Bulln, Fermentation, processing and utilization, starch

 179 Mulugeta Tcamir and Adane Tilahitn

In tro d u c tio n

Root and tuber crops are widely cultivated in Southern Ethiopia that supports a
considerable portion of the country's population as source of food. Prominent among
these are: potato (Solatium tuberosum L.), sweet potato (Ipomoea batatas L.), eusct (Ensete
veil ricosum (Welw.), Chessman), Godere (Coloeasia esculenta /..), yams (Dioscorea spp.),
Ethiopian dinch (Coleus pannflonis), koteharrie (Diaspora bulbiferous) and anchote
(Cor.cinia abyssinica). enset, anchote and some yams are endemic to Ethiopia (Addis,
2005).

Enset belongs to the order Scitaminae, family iVIusaceae. The genus Ensete comprises
of 5-7 species, half of which are African, the other half are Asian in origin (Taye,
1984). Enset is related to and resembles the banana plant and is produced primarily
for the large quantity of carbohydrate-rich food. Both enset and banana have an
underground corm, a bundle of leaf sheaths that form the pseudo-stem, and large
leaves. Enset, however, is usually larger than banana. The leaves are more erect than
those of a banana plant, have the shape of a lance head and about five meters long
and nearly one meter wide. Its pseudo-stem widens at the base to a circumference of
1.5 to 3.0 m. Depending on the variety and ecological condition of its cultivation, the
pseudo-stem length ranges from 2 to 5 m (Taye, 1984; Brandt et al., 1997). About 20
percent of the population of the country, concentrated in the highlands of Southern
and Southeastern Ethiopia, depends upon enset for human food, fiber, animal feed,
construction materials, medicines, means of earning cash income and insurance
against hunger.

Enset is one of the potential indigenous crops for food production (Taye, 1984;
Endale, 1991) and can be grown everywhere in Ethiopia. According to several
authors (Smeds, 1955 and Taye et al., 1967), the enset cultivation system is
economically viable and is one of the few successful indigenous crops that have
become sustainable in the agricultural systems. It is sustainable and has been
providing food for humans for generation from the same plot and maintains the
quality of life of the people. It grows in a wide range of environmental conditions.
Even though it is grown in many administrative regions, the dwellers of the Central
and Southwestern parts of Ethiopia (Sidamo, Shoa, Keffa, Ganio Goffa and Illubabor
administrative regions) are the onlv people that use enset as a staple and co-staple
crop (Simmonds, 1958; Addis, 2005).

The!plant is most productive if harvested at maturity, but can be harvested at anv

time after about 2 years of age (Kefale and Sandford, 1991). The same authors
expressed the yield potential of enset at maturity in Welavta to yield 1.3 - 3.5 more
edible energy per hectare per year than peasant-grown maize, which yields about
300() kg grain per hectare. The edible product is high in energy (190 Calories/gram)
but low in crude protein (2-5% by weight).
 180 Research Achievements and Experiences on Eusct Food Products

Enset provides fiber as a byproduct of decorticated leaf sheaths. The fiber has
excellent structure, and its strength is equivalent to the fiber of abaca, a world-class
fiber crop (Taye and Abraham 1967). About 600 tons of enset fibers per year are sent
to local factories. In rural areas the fiber is used to make sacks, bags, ropes, cordage,
mats, construction materials (such as tying materials that can be used in place of
nails), and sieves (Brandt et nl., 1997) . Fresh enset leaves are used as bread and food
wrappers, serving plates, and pit liners to store kocho for fermentation and future use.
During enset harvesting the leaves are used to line the ground where processing and
fermentation take place. The dried petioles and midribs are used as fuel, and to make
mats and tying materials for house construction. The pulp from the dried leaf
sheaths, petioles, and midribs is used as cleaning rags and brushes, baby
cushions/diapers, and cooking pot stands. Dried leaf sheaths are used as wrappers
for butter, kocho, and other items to transport to local markets. Enset leaves are an
important cattle feed, especially in the dry season when grasses are scarce. Leaves
are carried into the house for stall feeding of cattle during the nighttime.

Particular clones (or varieties) and parts of enset plants are used medicinally for both
humans and livestock to cure bone fractures, broken bones, childbirth problems (i.e.,
to assist in the discharging of the placenta), diarrhea, and birth control (as an
abortifacient) (Brandt et nl., 1997).

Achievements Made oil E n set Processing, Utilization and

Improving its \utritionaI Quality

processing and utilization o f enset products

The optimal harvest times for enset plant depend on the specific cultivars. Enset if
harvested too young has a low starch content and if harvested too late (after
fructification) it has a reduced amount of starch. The maximum output is obtained
when the plant is fully mature and ripe for harvest which is indicated by the
emergence of the inflorescence. Another indication is the emergence of fresh roots
from the base above the soil surface. The period of maturity depends on variety,
climate and soil fertility. Enset matures from 4-6 years after planting, whereas, for
amicho it would be ready any time at one year after planting in the field.

The preparation of enset for processing is a very time-consuming and tedious work.
Bv social custom, almost all the operations connected with enset processing are the
exclusive responsibility of women in the familv.

The major foods obtained from enset are kocho, bulla and amicho. Kocho is the bulk of
the fermented starch obtained from the mixture of the decorticated (scraped) leaf
sheaths and grated corm (underground stem base). Kocho needs a lengthy period of
processing and preparation. The first stage involves removing the leaf stalks and
grating of the corm. Then, the fibers are separated out and the pulp is crushed to
extract the starch. This is put in a pit about 1.5 in deep and 1 m diameter, wrapped
airtight with enset leaves before being packed down with stones. It is then allowed to
 181 Mulugeta Teamir and A d an e T ilah iin

ferm ent - a process, which may last anytime from 4 months to three years. The pit is
opened at intervals to allow aeration, and the enset leaves are replaced. This is
repeated until the desired fermentation quality is reached or when it is needed. After
fermentation the product is chopped to reduce the fiber size and then sieved. Finally,
the fermented starch is dried and treated as flour. Kocho can be stored for long
periods of time without spoiling. The quality of kocho depends on the age of the
harvested plant, the type of clone (variety) and the harvesting season. Moreover,
within one plant, the quality is influenced bv the part of leaf sheath and corm
processed. The preferred type is white in color and is obtained from the innermost
leal sheaths and inner part of the corm, while the lowest grade is blackish and is
obtained from the outer leaf sheath and corm. Although many different dishes are
prepared from kocho, a pancake like bread is the most common, which is eaten with
milk and cabbage. It is also extremely popular at restaurants that served kitfo - a
traditional food (finely minced uncooked or rarely cooked meat mixed with spicy
bu|ter). The other quality determinants of kocho are the skill of the processor
(woman) and the care taken in processing and time of processing, where drv season
is considered the best.

Appropriate mix of varieties and proper covering and protection of the fermentation
pit are also factors that determine the quality of kocho.

Bulla is obtained through passing the following processing stages: scraping the leaf
sheath, peduncle, and grated corm into a pulp; squeezing liquid containing a starch
from the pulp; allowing the starch to concentrate into a white powder and
rehydrating with water. It is considered as the1best quality enset food and is obtained
mainly from fully matured enset plants. Bulla can be prepared as a pancake, porridge,
or dumpling.

Amicho is the boiled enset corm, usually of a younger plant. Enset plants may be
uprooted for preparing meals quickly if the amount o f enset harvested is insufficient,
or for special occasion. The corm is boiled and consumed in a manner similar to the
preparation method for other root and tuber crops. Certain varieties are selected for
their amicho production.

Among the main processing steps of making kocho and bulla, fermentation is the key
operation. Therefore some scholars conducted research activities relating to
understanding the type of microorganisms employed to necessitate the fermentation
process and during the fermentation period; identifying microorganisms responsible
for the spoilage and deterioration of processed enset products, investigating the effect
of fermentation on the chemical composition (nutritional content/value) and the
effect of fermentation on anti-nutritional content of enset.

Gashe (1987) is one of the scholars who extensively studied and described the
microbiology of kocho fermentation. In his report, a microorganism named
 182 Research Achievements and Experiences on Enset Food Products

Leuconostoc mesentroides initiated the fermentation process and dominated the lactic
flora with counts of 107cfu/g on day 8. The pH of the fermentation mass dropped
from 6.5 to 5.8 in 8 days. Lactobacillus conjneforms and Lactobacillus plantarum
dominated thereafter and further reduced the pH to 4.2 after 50 days. Spore formers
were present at levels of : 103cfu/g during the first 15 days. Generally, the
population of Clostridium spp. was two to five times more abundant than Bacillus
spp.

C. stridium butyricum, C. beijerinckii, C. sticklandi, B. subtilis, B. megaterium, B.

liclieniforiinis and B. ccreus were among spore formers which appeared to show active
growth in fermenting Kocho. Yeast reached their highest counts of 103 cfu/g between
22 and 43 days and the yeast flora consisted of the Trichosporon, Torulopsis Rhodotoruln
and Candida.

Gashe also reported that kocho become easily contaminated by microorganisms when
removed from the fermenting pits and the major spoilage fungi belonged to
Penicilliuni, Tricliodernia and Chaetoniiuni species. In addition, bacterial species
belonging to Leuconostoc, Pseudomonas Bacillus and Envinia were isolated from slimv
kocho. The microbial spoilage was manifested in the form of discoloration.

Mogessie and Yewelsew (1996), have also studied the microbial load of kocho and
bulla sold at Hawassa open market. They observed that the products did not
undergo appropriate fermentation and had pH values around neutral. Kocho and
bulla had high counts of aerobic mesophilic bacteria and yeasts ( 10hcfu/g).
Coliform counts were markedly higher in bulla (105cfu/g) than in kocho (103 cfu/g).
Counts of enterococci, in both products, ranged between 104 and 105 cfu/g.
Micrococci and Bacillus spp. dominated the aerobic bacterial flora. Among the yeast
species, Rhodotorula glutinis, Kveromyces marxianus and Picliia mcmbranefaciens were
isolated from most samples. As kocho and bulla appeared to be processed in
unhygienic conditions, unfermented products are likely to spoil easily.

They also observed that when these products were stored at room temperature in a
loosely wrapped condition, both products had undesirable odor, slimy surface and
dark discoloration after eight days. Spoiled kocho and bulla had high counts of
aerobic mesophilic bacteria (aboutlO10 cfu/g) with Micrococcus and Bacillus species
dominating the spoilage flora. Psychrophilic microorganisms consisting of bacteria
and molds were isolated at levels of >104 cfu/g and mold spores caused dark
discoloration. Microorganisms active in starch hydrolysis, proteolysis and lipolysis
were encountered in the products at varying frequencies. Tightly wrapped samples,
which served as control, did not show any detectable spoilage in terms of odor,
consistency or color. The fermentation of kocho not onlv modifies and improves the
quality, but also aids in its preservation. The presences of antibiotic-like substances
in kocho and their inhibitory activity on Gram-negative bacteria such as Salmonella,
Klebsiella may have far reaching significance as in gastrointestinal tract microflora
 183 Mulugeta fcnmir and Adane Tilalum

and infection. Hence, detailed studies on the chemicals produced by Lactic Acid
Bactria during fermentation should be carried out (Nigatu and Gashe, 1994).

The presence of appreciable levels of oxalic acid, trypsin inhibitors (TI) and tannins
in enset that contribute the major component of the diet of the South and South
Western people of Ethiopia is likely limit its utilization. However, natural
fermentation of enset markedly reduced the content of trypsin inhibitor and tannins
thus enhance the digestibility of kocho protein and bioavailability of minerals
(Kelbessa et at., 1997).

T ablel. Effect of fermentation on anti-nutritional factors of kocho (on dry weight basis)

Period Tannin TIA Oxalic acid

(week) mg/1 OOg TIU/g mg/1 OOg
0 676 -41 8356: 11 230 ; 32
1 644 _34 8137 120 228 23
2 440 23 7956 177 225 24
3 430 -31 6870*123 2 2 4 -2 2
4 422 -3 0 5184 -1 3 0 220-21
5 390 l 32 4288 • 137 224 _ 19
6 331 18 4079- 112 235 32
7 333 -21 3924- 110 2 2 0 -3 4
Source Kelbessa Urga et al., (1997)

The bases and corms of enset show enzymatic browning reaction when cut and
exposed to air. This can occur with the corm in the fresh state, after storage or when
physiologically damaged. The cause of this process could be associated with the
reaction of phenolic substances (Kelbessa et at., 1997).

The other research areas conducted by the scholars was whether the kocho yield of
enset, in terms of weight and energy, under different crop establishment methods has
more advantages than other crops. Most of the previous reports published were on
fresh yield of fermented kocho and were based on survey works. In addition, they
lacked information on the age of plants harvested, stages of transplanting and the
duration of the fermentation periods of the products (Bezuneh, 1984; Makiso, 1976,
CSA, 1997). The yield report shows 23.5 (Bezuneh, 1984), 30.6 (Makiso, 1976) and
30.1!) (CSA, 1997) kg of kocho per plant. The yield is much lower than the yield
reported by Admasu and Struik (2001), who compared the performance of enset yield
with the yield of other major starch crops grown in the country. In this studv
maximum fresh yield of kocho after 88 davs of fermentation from enset plants
transplanted once (at 130 weeks after first transplanting), from enset transplanted
twice? and trice (at 234 weeks after first transplanting) were 25.9, 54.1 and 37.1
kg/plant, respectively (Table 2).
 184 Research Achievements and Experiences on Enset Food Products
Table 2 Fresh weight (kg/plant) of fermented kocho harvested at different weeks after removal from the mother
corm for different times of transplanting and leaf pruning treatments.

Transplanting Leaf pruning Weeks after removal from mother corm

(A) (B) 104 130 156 182 208 234 260
Once Without 18.6 25.9 20.3
With 14.6 16.9 18.0
Mean 16.6a 21.4a 19.1a

Twice Without 3.3 12.5 13.5 22.5 27.5 54.1 28.5

With 4.2 12.0 16.5 16.6 24.8 53.9 30.5
Mean 3.8b 12.3b 15.0b 19.6a 26.1a 54.0a 29.5

Thrice Without 1.7 3.3 3.9 9.5 15.5 37.1 31.0

With 1.5 2.6 4.3 10.0 12.5 29.9 26.0
Mean 1.6c 2.9c 4.1c 9.8 14.0b 33.5b 28.5

Without 7.9 13.9a 12.5 16.0 21.4 45.6 29.8

With 6.7 10.5b 12.9 13.3 18.6 41.9 28.3
Grand mean 7.3 12.2 12.7 14.7 20.1 43.8 29.0
CV (%) 21 29 14 23 17 23 22
P or LSD (0.05) A 1.6*** 3.8*** 1.9*** 3.8** 3.9** 11.4** Ns
P or LSD (0.05) B Ns 3.1* ns Ns ns Ns Ns
P o r LSD (0.05) AxB 2.3* 5.3* 2.7* P< 0.09 ns Ns Ns
Note: ns, *, **, and*** indicate non-significant, significant at p<0.05, 0.01, and 0.001, respectively (F test). Different letters
within a column indicate significant difference at p <0.05, according to Duncan's Multiple Rang Test
Source: Admasu and Struik, (2001)

Enset transplanted once, twice and thrice produced much more edible dry matter per
unit space and time compared to the other main crops (Table 3). The edible dry
matter production rate of enset plants transplanted twice and trice was much more
compared to all other crops, whereas the difference between enset plants
transplanted once and sweet potato or taro in terms of edible dry matter production
rate was very similar. The average edible dry matter production rate of enset from the
three crop establishment methods was about 133 and 69 % higher than the average
values for cereals and root and tuber crops, respectively. The dry matter percentage
of fermented kocho was almost comparable to that of taro, yam and sweet potato.
 185 Mulugeta Teamir and Adane Tilahun

Table 3. Average yield and edible dry matter production rates of major crops grown in Ethiopia as compared with
enset under different crop establishment methods

Crop Yield(g/m2>
Based on Edible dry
average Edible Dry matter matter (g/m2 Growth Edible dry
spacing portion (%) (%) pre harvest) period (days) matter(g/m2/day)
Enset
Transplanted once 3886 80 32 944 730 1.29
Transplanted twice 14958 80 30 3590 1643 2.19
Transplanted thrice 11817 80 29 2742 1643 1.67
Average enset 10154 80 30 2425 1339 1.72

Teff 94 100 89 83 120 0.69

Barlsy 104 100 87 90 150 0.60
Wheat 147 100 87 128 150 0.85
Maize 159 100 80 127 150 0.85
Sorghum 126 100 85 107 150 0.71
Finger millet 97 100 89 86 120 0.72
Average of cereals 121 100 86 104 140 0.74

Irish potato 713 85 20 121 120 1.01

Sweet potato 821 85 30 209 150 1.40
Cassava 688 85 40 228 270 0.85
Taro 932 85 30 237 210 1 13
Yam j 750 85 27 172 270 0.64
Average of root and 781 85 29 177 193 1.01
tuber crops

Source: Admasu T. and P.C.Struik,(2001)

The average energy production rates of enset under different crop establishment
methods and main crops (Table 3) are calculated using food composition tables.
Twice transplanted enset plants increased energy production rate by 80 and 27 %
compared to once and trice transplanted, respectively. Enset plants transplanted
one?, twice and thrice produced much more energy per unit space and time
compared to other high energy producing crops (Table 4). The average energy
production rate from the three crop establishment methods was about 286 and 172 %
higher than that of cereals, root and tuber crops, respectively.
 186 Research Achievements and Experiences on Enset Food Products
Table 4. Energy production rates of major crops as compared with the energy production rates of enset
under different crop establishment methods.

Energy Energy production

Energy yield KJ/100 g of production rate
Crop (g lm 2) edible yield (KJ/ m2) (KJ m2/day)
Enset
Transp anted once 294S 883 26040 35.67
Transpanted twice 11966 883 105660 64.31
Transplanted thrice 9454 883 83479 50.81
Average enset 8123 883 71726 50.26

Teff 94 1485 1396 11.63

Barley 104 1552 1614 10.76
Wheat 147 1494 2196 14.64
Maize 159 1569 2495 16.63
Sorghum 126 1502 1192 7.95
Finger millet 97 1469 1425 11.88
Average of cereals 121 1512 1720 12.25

Irish potato 606 431 2612 21.76

Swee' potato 698 569 3972 26.48
Cassava 571 640 3654 13.53
Taro 792 519 4110 19.57
Yam 638 464 2960 10.96
Average of root and tuber 661 525 3462 18.46
.crops
Source: Admasu T. and P.C.Struik, (2001)

\ u trition al composition o f enset

The main feature o f enset foods is their high energy value (1410-1950 kj per 100 g dry
matter of kocho and 1580-1850 kj per 100 g dry matter of bulln), derived almost
entirely from carbohydrate. Fresh kocho contains 47-62 g moisture per 100 g fresh
weight. The approximate composition of kocho per 100 dry matters is: 1.1-2.8 g
protein, 0.2-0.5 g fat, 95-98 g carbohydrates, 2.3-6.2 fiber g, 1.7 g ash, 60 mg calcium,
68 mg phosphorus, 7 mg iron, 0.06 mg thiamine, 0.08 mg riboflavin, 0.6 mg niacin.
The moisture content of bulla ranges from 44-55 g per 100 g fresh material. Table 6
shows that the approximate composition of bulla per 100 g dry matter is: 0.4-0.8 g
protein, 0.2-0.4 g fat, 93-98 g carbohydrates, 0.6-0.8 g fiber, 0.2 g ash, 91 mg calcium,
44 mg phosphorus, 5.8 mg iron, thiamine 0.02 mg thiamin, 0.2 mg niacin (Agren et
a l, 1968, EHNRI, 1997;.

Minaleshewa and Chandravanshi (2008) studied commercially available enset

products (kocho and bulla) for major, minor and trace minerals and they showed that,
the concentration of K was highest followed by Na, Ca, and Mg in both foodstuffs.
From trace elements analyzed, Zn was found to be highest next to Fe. Generally,
kocho contained higher concentration of the mineral nutrients compared to bulla.
Thus, kocho and bulla are rich in calcium and zinc compared to other similar
foodstuffs and contains comparable concentration of calcium, iron and manganese.
The toxic metals cadmium (Cd) and lead (Pb) were not detected in both types of
foodstuffs.
 187 Mulugeta Teainir and Adane Tilahun

Wolde-Gebriel et al. (2006) reported that bulla is more energy rich (204 Cal per lOOg)
than kocho (156 Cal per 100 g) on dry weight basis. All foods have low protein
content (4-22 g/kg) and no vitamin A. If correspondingly the energy values for food
grain like tef, wheat, barley and corn are estimated at 350 Cal perlOO g dry weights.
The food energy per unit weight of bulla and kocho are only 57% and 45 % of the food
grains. The protein quality of kocho is poor due to low content of essential amino
acids such as methionine and isoleucine (Besrat et a l, 1979). It is obvious that such
low levels of protein, if not supplemented from other sources, are inadequate to meet
the minimal protein requirement of human.

Yewelsew et al. (2007) analyzed the zinc, iron, calcium and phytate content (mg per
100 g fresh weight) and the phytate: zinc and phytate: iron molar ratios of the
perishable foods prepared from the major plant-based staples of Sidama region: enset
(£. ventroicosum), tef (£. tef) and maize (Zea mays L.). There was a wide range in the
mineral content of the three starchy fermented foods prepared from enset. Zinc
content ranged from 0.07 mg/100 g for bulla 1.33 nig/100 g fresh weight for amicho.
Amicho, followed by bulla, had a much lower content of calcium than fermented
baked kocho. Indeed, fermented, baked kocho had the highest level of calcium of all
the perishable, prepared foods analyzed (Table 5).

Of the three types of fermented tef injera samples analyzed, only one had a very high
iron content (71 mg/100 g) compared to 7.0 m g/100 g for die other two injera
samples and 3.4 mg/100 g for unleavened corn bread. The zinc content of all the
prepared cereal products was very similar, ranging from 0.93 mg/100 g for injera
prepared from a mixture of red and white tef to 1.75 mg/100 g for injera prepared
from red te f The calcium content of the unleavened corn bread was lower them that
of fermented tef injera. All the starchy fermented foods prepared from enset and
fermented tef injera had lower levels of IP5+IP6 than the unleavened corn bread.
Only unleavened corn bread had a calculated phytate: zinc molar ratio above 15,
whereas corn bread, boiled kidney beans and injera prepared from white tef had
phytate: iron ratios above 1.0.
 IS8 R esearch Achievements and Experiences Enset F ood Products

Table 5. Median (1st, 3rd quartile) zinc, iron, calcium and phytate contents (mg/100 g fresh weight) of perishable prepared foods from Sidama, Ethiopia

Food type (n) M oisture Zinc (mg/100 g) Iron (mg/100 g) Calcium Phytate Phy:Zn Phy:Fe
{9/100 g) (mg/100 g) (mg/100 g)
Enset starchy foods
Kocho, fresh: pulp (n=3) 85 (82,86) 0.09(0.08, 0.15) 1.1 (0.8, 1.6) 52 (46,60) 7 (6 ,9 ) 8.4 0.9
Kocho, fermented: pulp, baked (n=16) 36 (26, 44) 0.52 (0.32, 0.72) 6.2 (3.6,10.1) 162 (140,226) n.d.
Bulla: desiccated juice from enset pulp (n=4) 57 (54, 58) 0.07(0.05,0.09) 4.8 (3.6, 6.5) 44 (40,47) n.d.
Amicho: decorticated tuber before fermentation (n=4) 76 (71,77) 1.33 (1.03,1.61} 0.7 (0.6, 1.1) 25 (24, 29) 0.9 (0.2,13.0 0.1 0.6
Cereals
Tef dough unfermented (n=3) 57 (35, 66) 1.35(1.19,1.96) 6.8 (6.7. 26.4) 65 (51.91) 139(136,144) 8.4 0.6
Injera: from fermented white tef dough (n=1) 71.6 1.11 6.8 36 102 9.1 1.3
Injera: from fermented white and red tef dough (n=1) 67.8 0.93 7 35 36 3.8 0.4
Injera: from fermented red tef dough (n=1) 56.5 1.75 70.9 76 117 6.6 0.1

Com bread, unleavened (n=16) 35 (29, 49) 1.66(1.36.1.84) 3.4 (3.0. 5.5) 14 (12, 20) 394 (366, 442) 21.6 5.4
Legumes
Kidney beans, red: whole, dried, boiled (n=7) 38 (34, 59) 1.52(1.23,1,94) 2.7 (2.4,4.3) 69 (55, 96) 219 (176, 253) 14.6 6.0
N= number of food samples in each food type
Source: Yewelsew, A. etal., 2007
 189 Mulugeta Icam ir and Admic Tilalimi

Table 6. Composition of various enset products in terms of 100 grams of edible portion

2?

Protein (g]

Niacin (pg)
Thiamine
Fiber (g)
CHO(g)

Ca (mg)
Ash (g)
TO

Fe (mg)
O?

P (mg)
TO 1

(M9)
U- ix
Enset powder 196.00 49.70 0.30 0.90 0.20 47.70 1.20 1.60 77.00 60.00 10.10 0.04 0.05 0.00
Bulla powder 180.50 54.90 0.03 0.20 0.10 44 70 0.30 0.10 41.00 20.00 0.01
2.60 0.00 0.10
Bulla bread 186.10 53.30 0.03 0.30 0.10 46.00 0.30 0.30 45.00 18.00 0.01
4.60 0.00 0.10
Bulla porridge 80.30 81.00 0.03 0.20 1.10 17.40 0.50 0.30 30.00 10.00 0.01
2.60 0.00 0.10
Kocho powder 211.10 46.70 0.20 0.60 0.30 51.50 1.20 0.90 32.00 36.00 3.70 0.03 0.04 0.30
Kocho bread 219.40 43.70 0.20 1.00 0.20 5340 1.30 1.70 93.00 43.00 2.40 - 0.10 0.20
Kocho porridge 90.60 78.90 0.11 0.70 2.20 1700 1.30 1.20 50.00 21.00 0.01
4.90 0.03 0.10
Source: EHNRI (1997)
 190 Research Achievements and Experiences on Enset Food Products

Yewelsew et al.(2006) conducted a study to increase the energy density and vitamin
A content of corn and kocho based infant foods by supplementing with legumes
(kidney beans) and pumpkin pulps ( Table 7).

The study had showed that, protein density (energy from protein as percentage of
total energy) for legume-supplemented kocho, protein density was 14.9 % while the
un-supplemented kocho was only 1.5%. All amino acids were increased markedly by
adding beans to kocho because kocho alone contained only 1.5 % protein. The retinol
activity equivalent (RAE) is based on the assumption that 12 pg of dietary a\\-trans-\>-
carotene from food would be converted to the equivalent of 1 pg retinol while 24 pg
of other dietary pro-vitamin A carotenoids would be required to form 1 pg retinol.
Assuming that 50 % of the vitamin A activity contributed bv pumpkin came from a 11-
f/77/zs-l>-carotene, and 50 % from other dietary pro-vitamin A carotenoids, the
supplemented complementary foods both provided 54 pg RAE/100 kcal as fed.

Table 7 Nutrient density of the traditional and supplemented complementary foods as served1

Legume supplemented Traditional

Nutrient density Kocho (KBP) Kocho( K)
Energy density (Kcal/g fresh 0.46 0.49
Weighti
Protein density (%of energy) 14.9 1.5
Vitamii A density2( ,;«g RAE/
100Kcal) fresh wet basis 54.4 0.3
calcu ated values are based on chemical analysis for protein and vitamin A values and food table values for energy
Converted from total carotenoids (assumed to be 50% all-trans- carotene where 12 g= 1 g Retinol Activity Equivalent
(RAE) and 50% other carotenoids where 24 g = 1 g RAE)
Source: Yewelsew et al. (2006)

The protein concentration of KBP (kocho: kidney bean: pumpkin supplement) was
increased nine-fold over traditional kocho. The use of pumpkin improved the vitamin
A density in KBP, compared with the un-supplemented kocho diet; the improvement
in the vitamin A value was more than 180-fold in KBP. The addition of kidney beans
enhanced the protein content of kodzo-based complementary foods from 1.5 g/100 g
(dry weight) 13.8 g/100 g. The pumpkin in KBP provided 54 pg RAE per 100 kcal,
increasing the Vitamin A value of the mixes by 180-fold.

The addition of defatted sov flour at 5, 10, 15, 20 % to bulla flour gave acceptable
product and significantly enhanced the quality and quantity of protein (Mulugeta I .
personal communication).

I sin2, enset sis indu strial crop

One of the main products of enset is bulla, which is in a very small quantity. Starch
from enset is currently produced in a local factory owned by Yitbark and family
business for textile and corrugated box factories. It can also be used in food
processing.
 191 Mulugeta Temnir ami Adane Tilahun

Enset starch (Ensete ventricosum, Musaceae) has been examined for its chemical
composition, amylase content and physico-chemical properties bv Gebre-mariam
ard Schmidt (1996). The proximate composition of enset starch on dry weight basis
was found to be 0.16 % ash, 0.25 % fat, 0.35 % protein, and 99.24% starch. The
amylase content was 29%. Scanning electron microscopy (SEM) of enset starch
granules showed characteristic morphology that was somewhat angular and
elliptical. The starch has normal granule size distribution with a mean particle size of
46*1m. Its moisture sorption pattern was similar to that of potato starch but much
higher than maize starch. DSC parameters obtained from starch-water mixtures (1:2),
namely, the enthalpy of gelatinization (AH: 21.6mJ/mg), the onset temperature (T°:
61.8°C), the peak temperature (Tp: 65.2°C) and the end-set temperature (Tc : 71.7°C)
were higher than those obtained for potato starch. Brabender viscosity curves of 6%
starch paste showed lower peak viscosity (884 BU) than potato starch (1668 BU) but
greater than maize starch (302 BU). The breakdown was also lower than potato
starch but higher than maize starch. Retro-gradation of enset starch was substantially
greater than potato starch but less than maize starch.

( 1liaHcni>cs, ('constraints and Gaps in:

E nset processing and utilization

Research activities have been conducted related to processing and nutritional
improvements. However these research activities have not been enough to solve all
the problems existed in processing and nutrition improvement aspects.

Fermentation, by virtue of microbial metabolic activities, contributes to the

development of characteristic properties such as visual appearance, aroma, taste,
texture, safety and shelf-life of a specific food product (Holzapfel, 1997). However,
thej outcome of uncontrolled spontaneous fermentation is not predictable and
therefore variations in the quality and stability of the products could be of major
importance. In a rather crude and empirical approach, inoculation of raw materials
with a portion from previous batch ('back-slopping') of acceptable quality was
experienced as a mixed starter culture to accelerate the traditional fermentation and
improve predictability of the fermented product. Back-slopping is still used in mam
traditional food fermentations to control desirable changes that enhance
predictability of the product. Thus, priority should be given to the development of
appropriate starter cultures for use in kocho fermentation. The quality of starter
culture used is one of the factors that determines the final quality and safety of
fermented foods (Motarjemi, 2002). Therefore, searching of suitable microorganisms
for development of starter cultures has become an issue to scale up the production of
traditional fermented foods with enhanced predictability. Starter cultures are
selected and used specifically for a substrate such as milk, meat, cereals, legumes,
roots and tubers. Most commercial starter cultures originate from those food
 192 Research Achievements and Experiences on Enset Food Products

substrates to which they are applied today. Utilization of starter cultures enables to
exercise strict control of a fermentation process that results in a predictable product.

There is no well-developed and commercially available starter culture for

fermentation of enset to produce kocho and bulln. Moreover, no protocols or
methodologies are available to shorten the longer enset fermentation period thereby
practice at industrial level. Therefore, modernization of fermentation process of enset
and ether Ethiopian traditional foods must be given due attention.

Enset products such as bulla and kocho are inferior on macro- and some micro­
nutrient contents compared to other cereals and pulses. Moreover, micro-nutrient
deficiency level in the enset growing areas is high compared to the other regions.
Enrichment of these products with macro- and-micro-nutrients are inevitable. Some
research activities have been done to tackle these problems. However, research effort
should be strengthen in number and quality thereby employ the research outputs to
reduce and further eliminate the problems among the vulnerable group.

I sin a; enset as industrial crop

Since the nineteenth century there is an enormous expansion in the starch industry,
due largely to demands of the textile, color printing and paper industries, and to the
discovery that starch can be readily converted into dextrin, which are gum-like
product and other products. Starch is made mainly from maize, sorghum, wheat,
potato, tapioca, rice, rye, oat, barley and others.

Currently, banana starch is certain to join the group of industrial starch source,
because it can be obtained from culled bananas discarded by large banana
plantations. Banana bunches are cut from trees in plantations and sent to a central
processing station, where culls consisting of small or damaged fruit are removed.
Such culls represent 25% of the banana crop and 25% of the green banana is starch.
The starch can be readily recovered from banana pulp in a four-hour steep at an
appropriate pH. Banana starch consists of large (20 p m) granules with properties
suitable for a variety of applications. The production costs, essentially of cartage plus
that of starch extraction, are expected to give a market price that approaches or
equals that of corn starch.

However starch production particularly from enset in Ethiopia is meager. There is

only one company that produces a certain amount starch from bulln. However, the
company is refraining from producing starch from enset because of the ever
increasing cost of bulla as raw material coupled with the expensive production cost
of starch from enset. Users are importing from India and China, which is much
cheaper than the locally produced product.
 193 M u l u g e t a Tcamir a n d A d a n e T i l a h u n

Recommendation

A though enset feeds over 20 percent of the population and is indigenous to the
country, the attention given particularly on utilization, processing and opportunity
as raw material for industry is very insignificant. In this case there are only few
fragmented research achievements. Therefore, it is recommend that the following
research and development areas be addressed to strengthen the few fragmented
research achievements made so far
• There is no well-developed and commercially available starter culture for fermentation of
| enset to produce kocho and bulla. Protocols or methodologies that shorten the longer enset
fermentation period thereby practicing at industrial level must be strengthened
M odernization of the fermentation process of enset and other traditional foods must be
given due attention.
• Micro-nutrient deficiency level in the enset growing areas is high compared to the other
regions. Food fortification and supplementation of enset food with nutrient rich foods
must be developed. The bio-availability, bio-efficacy, other nutritional and functional
properties of the enriched product should be studied.
• The already achieved research outputs, particularly in the area of food supplementation
at the enset growing areas for vulnerable groups should be sensitized.
• Enset is a potential crop to produce starch for different industrial purposes. However,
there is very little information available on characterization and processing of starch from
bulla. As textile and other industries are growing fast, the need for starch as raw material
is unprecedented. Therefore, there should an extensive research and development
activities on the production of starch from locally and indigenous crops to facilitate and
enhance import substitution for same.
• The presences of antibiotic-like substances in kocho and their inhibitory activity on gram-
negative bacteria such as Salmonella, Klebsiella may have far reaching significance as in
gastrointestinal tract microflora and infection. Hence, detailed studies on the chemicals
produced by Lactic Acid Bactria during fermentation should be carried out.

Acknowledgments
We would like to acknowledge Dr. Mohammed Yusuf, Ato Kelbessa Urga, Dr
Gulelat Dessie and Ato Yitbarek Alemu for their kind cooperation to provide us
information in form of printing material, electronic and through brainstorming.

-I
 194 Research Achievements and Experiences on Enset Food Products

Rreferences
Abraham Besrat, Haile Mehansho and Taye Bezuneh 1979. Effect of fermentation on protein
quality and quality of enset, Nutrition Report International (20), 2
Add s, T 2005. Biology of Enset Root Mealybug (Cataenococcus Ensete) Williams and Matile-
ferrero (Homoptera: Pseudococcidae) and Its Geographical Distribution in Southern
Ethiopia. MSc. Thesis, Alemaya University, Alemaya (Ethiopia), ppl-96.
Admassu Tsegaye and P.C. Struik 2001. Enset (Ensete ventricosuni(Wc\\\'.)Cheesman) Kocho
yield under different crop establishment methods as compared to yields of other
carbohydrate-rich food crops. Netherlands journal o f Agriculutural Science p. 81-94
Agren, G.; Gibbson, R.1968. Food Composition Table fo r Use in Ethiopia I, Children's Nutrition,
Addis Ababa, Ethiopia and Institute of Medical Chemistry, University of Uppsala,
Sweden.
Bezuneh. T., 1997. Evaluation of some Enset ventricosum clones for food use with emphasis
on the effect of fermentation on carbohydrate and calcium content. Tropical Agriculture
(Trinidad) 61:111-116
Brandt, S.A.; Spring, A.; Hiebsch, C.; Yntiso, G.; Tabogie, E.; Diro, M.; W oldemichael, G.;
Tesfaye, S.; McCabe, J.T.; Shigeta, M. 1997. The Tree Against Hunger: Enset Based
Agricultural System in Ethiopia. American Association for the Advancement of Science
with Awassa Agricultural Research Center, Kyoto University Center of African Area
Studies and University of Florida, Directorate for Intentstional Programs, Washington,
D.C., pp 1-150.
EHNRI, 1997. Food com position table for use in Ethiopia. Part III. Ethiopian Health and
Nutrition Research Institute. Addis Ababa, pp 34.
F.ndale Tabogie, (1991). Enset in Ethiopia agriculture. Newsletters Agric. Research, Ethiopia 5
(1): 2-4.
Gashe, B.A. 1987. Kocho fermentation. Journal o f Applied Bacteriology 62, 473-477
Gebre-mariam T. and Schm idt PC, 1996. Isolation and physicochemical properties of enset
starch. Starch/ Starke 48 Nr. 6.S20S-214.
Holzapfel, W„ 1997. Use of starter cultures in fermentation on a household scale. Food Control 8:
241-258.
Kefa e Alemu and Sandford S., 1991. Enset in North Omo Region. FPR Technical Pamphlet
No. 1. Farm Africa. Addis, Ababa Ethiopia. P.33
Kelbessa Urga, Alemu Fite and Eskinder Biratu, 1997. Natural fermentation of Enset (Ensete
ventricosum), Ethiopian Journal o f Health Development 11(1):7 5-81
Loek T J Pijls, Arnold AM Timmer, Zewdie Wolde-gebriel, and Clive E West. 1995.
Cultivation, Preparation and Consumption of Enset (Enset ventricosum) in Ethiopia.
Journal o f the Science o f Food and Agriculture, 67: 1-11
Makiso, T., 1976. Study on the yield of enset. Sixth Annual Research Seminar. Institute o f
Agricultural Research, Holetta Research Station, Ethiopia, pp.246-254.
Minaleshewa Atlabachew, Bhagwan Singh Chandravanshi, 2008. Levels of major, minor and
trace elements in commercially available enset (Ensete ventricosum (Welw.),
Cheesman) food products (Kocho and Bulla) in Ethiopia. Journal o f Food Composition and
Analysis 21: 545- 55
 195 Mulugeta Tcamir and Adane Tilalnin

Mogessie Ashenafi and Yewelsew Abebe, 1996. Microbial load and incidence of
Staphylococcus aureus in market Bulla and Koclw, traditional Ethiopian processed fo
products from Enset (Ensete ventricosum), Ethiopian Journal o f Health Dcoelopment
10(2):U 7-U 2
Nigatu, A. and Gashe, B.A. 1994a. Survival and Growth of selected pathogens in fermented
Kocho (ensete ventricosum). East African Medical Journal, (71), (8).
Simmonds N. W., 1958. Ensete cultivation in the southern highlands of Ethiopia: A Review.
Tropical Agriculture 35: 302-307.
Smeds H., 1955. The Ensete planting culture of Eastern Sidam o, Ethiopia. ACTA
Geographica. 13 (4): 2-39.
Taye B., Feleke A. and Regassa B„ 1967. The cultivation of the genus Ensete in Ethiopia. Soil
and Crop Sci. Soc. Florida, 27: 133-141.
Taye Bezuneh, 1984. Evaluation of some Ensete ventricosum clones for food yield with
emphasis on the effect of length of fermentation on carbohydrate and calcium content.
Tropical Agriculture 61:111-116.
Tsegaye, A.; Struik, P.C. 2001. Enset (Ensete ventricosum (Welw.) Cheesman) Kocho Yield Under
Different Crop Establishment Methods as Compared to Yields of Other Carbohydrate-
Rich Food Crops, Netherlands Journal o f Agricultural Science, (49), 81-94.
Wolde-Gebriel, Z.; Pijls, L.T.; Timmer, A.A.M.; West, C.E. 2006. Review on Cultivation,
preparation and consumption of ensete (Ensete ventricosum) in Ethiopia, /. o f the Science
o f Food and Agriculture, 67:1-11.
Yewelsew Abebe, Barbara J. Stoecker, Margaret J. Minds, Gail E. Gates, (1996). Nutritive
value and Sensory acceptability of Corn- and Koc/io-Based foods supplemented With
Legumes for infant feeding in Southern Ethiopia, African Journal o f Food, Agriculture,
Nutrition and Development,6(1).
Yewelsew, A.; Bogale, A.; Hambidge, M.K.; Stoecker, J. B.; Bailey, K.; Gibson, R. 2007.
Phytate, Zinc, Iron and Calcium Content of Selected Raw and Prepared Foods
Consumed in Rural Sidama, Southern Ethiopia, and Implications for Bioavailability, J.
Food Com position and Analysis, 20: 161-168.
 Gender Differentials in Enset
Fanning' and Processing
Ycslii Cliiclic
Ethiopian Institute o f Agricultural Research (EIAR)

Abstract
This paper is a review o f gender related information on participation o f women
and men in Enset production and processing at different parts o f Ethiopia. The
purpose o f the paper is to reflect on the significance o f understanding gender gaps
in Enset research and utilization for addressing labor intensive problem and
quality o f Enset products. Literature reveals evidences on the close’ relation and
responsibilities o f women especially in post harvest management. Enset cloning,
planting and replanting are done by men while harvesting and processing is done
by women. M ajor processing methods include decortications, pulverization,
fermentation, kneading and shredding. Enset is usually processed annually in
most parts and usually by group o f women through organizing shared or hired
labor. Wooden and bamboo materials are used fo r kneading and shredding while
large knife is used to cut leaves and the Enset plant. Culturally men are not
allowed to enter to the processing area in most parts. Decision making in
processing and marketing is done by women and income from sale o f Enset is
controlled by women. Women have sole responsibility o f the haivesting and
processing o f Enset. Although attempts have been made for improvenient o f the
post harvest management practices, the conventional methods are still
dominantly used in the major Enset growing communities. Therefore, effective
gender responsive research is further required on availing better decorticators,
squeezers, processing devices and improved storage facilities that will reduce
labor and improve quality o f the produce. Strong multidisciplinary research
projects should be designed to address gender gaps in reducing labor requirement
and ensuring quality o f the products.

In tr o d u c tio n
Th.s paper constitutes findings reviewed from literature on the role of women and
men in Enset production with special emphasis to the improvement of the post
harvest managem ent techniques that involves substantial time, labor, knowledge
and special need. Observations related to post harvest management are highlighted
to illustrate the existing challenges for further investigation and detailed research
interventions.
 197 Yeshi Cliiche

The main objective is to indicate the relevance of considering gender gaps as

development component and amend Enset research strategies towards solving
problems related to productivity, quality and workload. Enset is one of the
indigenous crops of Ethiopia and major staple food for nearly 20% of the population.
Enjjet is widely grown in the Southern, Southwestern and central parts of the
country. It is also a crop used for multiple purposes. It is used for food, feed, fiber,
medicine and also considered as a sign of prestige (Endale,1997). Literature indicates
that Enset is known for its higher calcium content, carbohydrate, and iron content as
compared to other root and tuber crops and even some cereals. (Kefale and
Sardford, 1991). Enset food is prepared in various forms where m any traditional
dishes are prepared from it. It is a crop that can help to escape time of risk in food
and feed deficit because of its good capacity to tolerate irregular rainfall patterns that
may occur after the Enset plant is established. Currently, Enset starch is being used
for textile, paper and adhesive industries (Ethiopian Science and Technology
Commission, 2003).

Though Enset has enormous contribution to the food security, the progress in its
improvement, particularly in post harvest and processing management is negligible
and was not considered as high priority in technology generation.

Enset processing is constrained by cutting, grating, squeezing, chopping and storage.

Attempts have been made on improvement of post harvest management through
generating processing technologies/tools that can reduce labor, and improve quality
of products. However, the conventional methods are still dominantly used in the
major Enset growing farming communities.

The effort on improving processing tools was made by different stakeholders. The
major once include Agricultural Mechanization Department of the Melkasa
Agricultural Research Center, Home science Department of the then Awasa College
of Agriculture, now Hawasa University, the Sodo Rural Technology Development,
and the Ministry of Agriculture. However, there was limited coordination effort in
challenging the problem. Therefore, there is a need for implementing sound and
effective coordination effort to look deep into what has been done in order to
understand the system and the gaps and then to design better intervention
mechanism.
 198 Gender Differentials in Enset Fanning and Processing

Conventional processing methods of Enset

In Ethiopia, rural women play an important role in the agriculture sector both in
farming and livestock production. Study conducted at different parts of the country
indicates that women spend about 1 3 - 1 7 hours of a day in productive and domestic
obligations. In some parts of the country like the pastoral and agro-pastoral areas
women are also responsible for constructing mobile houses through collecting and
preparing proper materials. Therefore, women have multiple roles and
responsibilities in maintaining the wellbeing of their family through producing,
processing and providing food for their family.

The tasks and responsibilities of women and men in various field of activity vary
from location to location based on the cultural diversity, wealth status, education,
and other related factors such as values and norms. Although there are variation in
roles and responsibilities, women in Ethiopia are involved in seed selection and
maintenance, land preparation, weeding, fertilizer and pesticide application
harvesting, threshing and storing. The degree of involvement of women and men in
each activity vary from location to location as mentioned above. Furthermore, like in
many other developing countries, women are mainly responsible for food
preparation and provisions such as pounding, grinding, fetching water and
collecting fire wood. Child care and cleaning is also the responsibility of women and
gir s. In the farming activity, land preparation using animal power is done by men
except in rare cases while other activities are shared among household members.
With regard to livestock management, smaller animals are managed by women
while men are responsible for larger ones. Utilization and or processing of milk and
milk products (in small scale) is done by women.

In the case of Enset, cloning, planting and replanting are done by men while
harvesting and processing is done by women. Culturally men are not allowed to
enter harvesting area (Anita Spring, 1996; Judith and Helen, 1996). Enset is usually
processed annually in most parts and usually done by group of women through
organizing shared or hired labor. Decision making in processing, marketing, is done
b\ women and income from sale of Enset is controlled by women. Liyuwork (1996)
highlighted that Enset is processed by women farmers using wooden and bamboo
materials and bigger size knife is used to cut leaves and the Enset plant. According to
Liyuwork, the major processing methods include decortications, pulverization,
fermentation, kneading and shredding. She also noted that the conventional Enset
processing method is unsanitary and labor intensive and no adequate attempts were
made to improve the situation.

Mehtzun and Yewlsew (1994) made detailed assessment of Enset processing

methods and tools focusing on its improvement using gender perspective. They
have indicated that Enset is widely cultivated and utilized in the Southern, Western
and some central parts of the country. Lack of labor and time saving devices is one of
 199 Yeshi Chiche

priority problems particularly for women farmers of Enset who depend on this crop
as a source of food. Furthermore, Mehtzun and Yewelsew (1994) proposed to
generate multipurpose technology that can be used for decorticating, pulverizing,
squeezing, kneading and shredding. They looked at different attempts made by
different institutes particularly the agricultural research centers in generating sound
technology that provides simple adjustment to be easily accessed and become
operated. This feedback was utilized by the research group but still there is a need
for well organized and structured multidisciplinary effort for effective
implementation. The conventional Enset processing devices are illustrated in (Figure
1, 2 & 3,) below. Enhancing productivity of the crop and such device will make
effective contribution to reduce the problems related to food scarcity, productivity,
quality, and generating better income.

Figure 1: Selecting and cutting matured Enset plant and making ready for processing
 Figure 3: Digging the pit, lining and storing the Enset

Policies and strategies oil enset research in E IA R

The Government of Federal Democratic Republic of Ethiopia shows strong
commitment to support and strengthen gender-related activities at the national level.
Public ministries, institutions and organizations are encouraged to have women's
affairs offices and to increase the representation of women in different forums. The
 202 Gender Differentials in Enset Funning and Processing

Government has given special focus to women's contributions in social, economic,

cultural and political affairs (Bogalech, 2000). The national constitution allows equal
rights for women and men in all areas of social, political and economic development,
Ethiopia has also accepted the global and African regional platforms for actions for
the development of women. As part of the execution measures national endeavor,
EIAR has also established gender research coordination office at head quarter level to
coordinate the gender mainstreaming process in the National Agriculture Research
System.

The current agricultural research system in Ethiopia, evolved through different

course of actions that involve academic interest as well as participatory, problem-
focused and policy-oriented. Different approaches have been utilized to make
agricultural research more efficient and effective. These included package testing,
farming systems approach, and participatory rural appraisal. Each route has played
a significant role in sharpening the research focus towards being problem-oriented
and demand-driven. Experiences indicate that farmers adopt improved technologies
gradually depending on their priorities that are determined by their social, cultural
and economic situation. Furthermore, the local problems and needs of farmers vary
from location to location because of different local circumstances. Thus the farm
survey results have helped researchers to understand the complexity of the system,
to give due respect to farmers' knowledge and to define research topics accordingly
are important to consider. However, there was little room to consider gender aspect
in the research process, particularly in the research planning process.

Several studies indicate that the role of women in agriculture has been overlooked
and that this has contributed to the delay in the adoption of agricultural technologies
(Chiche, 2002). Many global efforts are considering gender as an important
component of development undertakings. In order to facilitate the consideration of
gender in the research systems, it has to be internalized in the system and the
organization's staff needs to apply it in their daily activities. This could also be
achieved by strengthening gender-responsive research in such a way that it can
adequately addresses the interests of all women and men farmers who are
responsible for maintaining livelihoods through farming. In this regard, The
Ethiopian Institute of Agricultural Research has developed gender mainstreaming
strategy that will enable researchers to consider gender factors in different research
activities. Gender considerations can facilitate the understanding of allocation and
utilization of available resources and benefits, identifying vulnerable groups and
their coping mechanisms and hence develop policies that would enhance
interventions for specific projects. Therefore, rigorous gender research is essential to
supplement development policies for effective and productive action.
 203 Yeshi Chiche

Conclusion
• Enset is widely grown in Southern and Western parts of Ethiopia, and is used for
wide range of purposes, food, feed, medicine and source of income.
• Clear division of labor by women and men are observed in Enset production.
• In almost all Enset growing areas, women are responsible for harvesting and post
harvest m anagement activities while men are more involved in the planting and
maintenance.
• Culture and norms have restricted men not to assist in processing
• Conventional processing methods of harvesting and planting are widely practiced by
women farmers.
• Facilities of the conventional methods the* pits and processing are not hygienic and it
is time consum ing and labor intensive.
• The product still needs further post harvest management research to improve the
productivity and nutritional quality.
• The national and institutional policies are conducive to conduct gender focused
research activities. However, there seems to be still a gap on how to harmonize the
long lived experience with the new approach of using gender perspective in
technology planning
• Capacity development for research technicians is important to enable them to use
gender perspective and thereby gender responsive projects could be developed and
implemented towards Enset processing and utilization
• Demonstration of improved processing devices have been developed but not widely
disseminated
• Efforts of Interventions made so far should be replicated, widely disseminated and
scaled out.
• Strong multidisciplinary research projects should be developed to generate better
processing and preservation methods for the subsequent and sustainable
implementation that improves the quality and food values of the products. Hence,
research should give a special attention and focus for larger scale of impact through
establishing task forces that would gain from similar activities.

Recommendations
Enset has been a preferred staple food for almost 20% of the Ethiopian population.
Yet, Enset has multipurpose uses than many other different crops. Despite its
demanding nature for harvesting and processing, Enset can easily be used as crop of
food security where women have sole responsibility of the harvesting and post
harvest management activities. However, research in the post harvest management
didn't make significant impact in terms of labor and time saving as well in enhancing
the quality of the product. One of the approaches in improving food quality is
through application of better pre and post harvest management techniques which are
essential to addressing problems related to food insecurity. Hence, improving food
availability, utilization and nutrition through better pre and post harvest
technologies is indispensable. Therefore, support from the research institutes in
 204 Gender Differentials in Enset Farming and Processing

consideration the pre and post harvest management practice/activities is vital in

addressing the problems associated with food insecurity and poverty.

Therefore, research intervention is required to consider the challenges of both

women and men to address major constraints of the community at large and also to
target the disadvantaged group of society (women) in particular. Introduction of
decorticators, squeezers, processing devices in particular and storage facilities that
will reduce labor and improve quality of the produce would have positive
implication on sustainability of the production. Processing devices should be
considered as one of the parameters in Enset improvement related activities.

In conclusion, detailed study of market oriented Enset food production methods that
focus on food quality and nutritional values need to be further developed. Finally,
this paper attempts to bring on board the relevance of gender element and its
consideration in the research design focusing on major challenges and constraints to
availing Enset food products. Previous studies on post harvest management of Enset
had identified factors that are important in Enset processing and management.
Future research endeavor should be based on past achievements as springboard for
post harvest problem identification and looking for alternative mechanism for
improvement.

References
Anita Spring (1996). Gender issues and farming systems research and extension in Enset
Agriculture in Ethiopia. In: the proceedings of Enset based sustainable agriculture in
Ethiopia. Institute of Agricultural Research. Ethiopia, 13 - 20 December 1993
Bogalech A (2000). Policy approaches to mainstream gender in Ethiopia. In: Proceedings of
the National Workshop on Institutionalizing Gender Planning in Agricultural
Technology Generation and Transfer Process. EARO, Addis Ababa. Pp. 126-144.
Endale Tabogie (1997). Morphological characterization of Enset (Enset ventricosuni (Welw.)
Cheesman) clones and the association of yield with different traits. Msc thesis, Alemaya
University of Agriculture, School of Graduate Studies, Alemaya, Ethiopia.
Ethiopian Science and Technology Commission (ESTC) http://www.capitalethiopia.com
2003.
Kefale Alemu and S. Sandford 1(991). Enset in North Omo Region. Farmer's Research. Project
Technical pamphlet No. 1. FARM Africa, Addis Abeba.
Judith Sandford and Helen Kassa. (1996). The effect of gender on resource contribution,
decision making and influence: A comparison between Enset, tef and maize. In: the
proceedings of Enset based sustainable agriculture in Ethiopia. Institute of Agricultural
Research. Ethiopia, 13 - 20 December 1993
 205 Yes!11 Cliiclic

Liyuwork Zewdie (1996). Kocho processing in Southern and Southwestern Ethiopia:

Preliminary results. In: the Proceedings of Enset based sustainable agriculture in
Ethiopia. Institute of Agricultural Research. Ethiopia, 13 - 20 December 1993
Mehtzun Tedla and Yewelsew Abebe (1994). Study of Enset Processing and Development of
Enset Processing tools in the Southern Region of Ethiopia. Awassa (Ethiopia), Awassa
College of Agriculture
Yeshi Chiche (2002). Comparative analysis of gender related farm households in the Arsi-
Negele farming zone in Ethiopia, MSc dissertation, University of Pretoria, Pretoria South
Africa