1 Forage and Livestock Production in Ethiopia-1 PDF

FORAGE PRODUCTION IN ETHIOPIA: A CASE STUDY WITH IMPLICATIONS FOR LIVESTOCK PRODUCTION
FORAGE PRODUCTION IN ETHIOPIA:

A CASE STUDY WITH IMPLICATIONS FOR LIVESTOCK
PRODUCTION
ALEMAYEHU MENGISTU
Ethiopian Society of Animal Production

Addis Ababa, Ethiopia
Alemayehu Mengistu, 2002 I

Published by:
Ethiopian Society of Animal Production
PO Box 80019, Addis Ababa, Ethiopia
© Alemayehu Mengistu, August 2002
Author's Address
Alemayehu Mengistu
Pasture/Forage/Range Development Consultant
And
Visiting Assistant Professor at Department of Biology,
Addis Ababa University,
PO Box 62291
Addis Ababa
Ethiopia
Tel. + 251-1-184415
Fax + 251-1-624546
E-mail: alemayehumengistu@yahoo.com
Alemayehu Mengistu, 2002 II

TABLE OF CONTENTS
TABLE OF CONTENTS .......................................................................................................................... III
FOREWORD .......................................................................................................................................... V
ACKNOWLEDGEMENT ......................................................................................................................... VI
PREFACE ............................................................................................................................................ VII
ABBREVIATIONS .............................................................................................................................. VIII
I. OVERVIEW OF LIVESTOCK FEEDING IN ETHIOPIA ............................................................... 1

Introduction.......................................................................................................... 1
Agro-Ecological Zones of Ethiopia...................................................................... 2
Seasonal Constraints to Forage Production .......................................................... 3
Integration of Livestock and Cropping Systems .................................................. 3
Improved Forage Species ..................................................................................... 4
II. INTEGRATION OF LIVESTOCK AND CROPPING SYSTEMS IN ETHIOPIA ............................. 7
Dominant Cropping Systems in Livestock Producing Areas ............................... 7
Opportunities for Integrating Livestock and Cropping Systems .......................... 7
Key Principles ...................................................................................................... 8
Impact of Increased Crop Productivity on Livestock Feeding ............................. 9
Financial Impact of Integrated Farming Systems ............................................... 10
III. IMPROVED FORAGE EXPERIENCE IN OTHER COUNTRIES .............................................. 13
Australia ............................................................................................................. 13
New Zealand ...................................................................................................... 17
South-East Asia.................................................................................................. 17
Caribbean ........................................................................................................... 18
IV. IMPROVED FORAGE PRODUCTION STRATEGIES .............................................................. 19
Backyard Forage Production .............................................................................. 19

Undersowing and Interplanting .......................................................................... 21
Contour Forage Strips ........................................................................................ 23
Forage Crop Production ..................................................................................... 25
Agroforestry ....................................................................................................... 26
Oversowing Common Grazing Areas ................................................................ 28
Stock Exclusion Areas/Forage Banks ................................................................ 29
Permanent Pastures ............................................................................................ 32
Roadside Sowing ............................................................................................... 32
Aerial Sowing .................................................................................................... 33
Summary ............................................................................................................ 34
Alemayehu Mengistu, 2002 III

V. FORAGE SPECIES ....................................................................................................................... 37

Principles for Selection and Testing................................................................... 37
Species with Proven Capability ......................................................................... 38
VI. IMPROVED FORAGE UTILISATION STRATEGIES............................................................... 41
Dairy Production ................................................................................................ 45
Forage Conservation .......................................................................................... 46
VII. FORAGE SEED PRODUCTION AND DISTRIBUTION .......................................................... 49
Importation of Initial Seed ................................................................................. 49
Contract Seed Production................................................................................... 50
Seed Harvesting, Cleaning and Storage ............................................................. 52
Seed Distribution ............................................................................................... 55
VIII. REQUIREMENTS FOR SUCCESS .......................................................................................... 57
Farmer Participation at all Stages....................................................................... 57
Strong Institutional Support ............................................................................... 58
A Planning Approach to Implementation ........................................................... 59
Motivated Technicians Located throughout the Project Area ............................ 60
Integration with Other Activities........................................................................ 61
Initial Importation of Improved Forage Seed ..................................................... 61
Financing Mechanisms ...................................................................................... 62
Secure Land Tenure ........................................................................................... 62
Monitoring and Evaluation ................................................................................ 63
IX. FINANCIAL AND ECONOMIC IMPACT OF FORAGE DEVELOPMENT PROJECT ............ 67
Introduction........................................................................................................ 67
Public Sector Investment and Recurrent Costs .................................................. 68
Farm-Level Investment and Recurrent Costs ..................................................... 70
Selection of Forage Development Strategies ..................................................... 71
Benefits from Forage Development and Rates of Return ................................... 72
Forage Production and Utilization Models ........................................................ 73
Financial and Economic Rates of Return from Forage Projects ......................... 76
LIST OF ANNEXES .......................................................................................................................... 79
ANNEX 1: Species Description and Characteristics .......................................... 81
ANNEX 2: Animal Performance of Improved Forage ....................................... 95
ANNEX 3: Inoculation of Legumes ................................................................. 107
ANNEX 4: Model Contract for Supply of Pasture/Forage Seed ...................... 108
BIBLIOGRAPHY ............................................................................................................................ 109
Alemayehu Mengistu, 2002 IV

FOREWORD
As part of its mission, the Ethiopian Society of Animal Production (ESAP)

promotes the publication and dissemination of knowledge in research and development of
animal production in Ethiopia. This publication is the third of its kind that ESAP promoted
and produced.
The rising demand for high quality animal products both for the domestic and
export markets calls for more inputs into the production process, particularly in the
provision of improved level of feeding. This becomes even more important in view of the
need in Ethiopia for gradual transformation of the predominantly low-input and subsistent
agriculture towards one of market-oriented to increase the contribution of livestock
resources to the livelihood of their owners, and hence to the national economy. The
challenge is to identify and develop viable options for increased production and utilisation
of quality feeds in the major production systems. The Fourth Livestock Development
Project (FLDP) of the Ministry of Agriculture had demonstrated a series of alternative
strategies in improved forage production over large parts of Ethiopian highlands. FLDP
was the first of its kind in Ethiopia in implementing extensive forage development
interventions specifically to the smallholder mixed crop-livestock farmers of the
highlands, and in so doing it generated valuable lessons for similar livestock development
programmes. However, the experience in FLDP have not been comprehensively reviewed
and reported in a form suitable for widespread distribution and readership. This
publication contributes to meeting this objective.
The author of this publication has accumulated extensive experience in research,
training and development of forage production in Ethiopia. He had also been involved in
the planning, development and management of the FLDP at the national level. His
contribution of this publication is, therefore, highly commendable.
This book is designed to serve needs in basic research, training and extension of
improved forage production of mainly the highland mixed smallholder production systems
as well as high-potential mid-altitude agricultural areas with emphasis to livestock -crop
integration and natural resource management. Apart from discussions on suitable
strategies for improved forage production and utilisation, this publication provides review
of prevailing livestock feeding practices in Ethiopia, suggests opportunities for better
livestock-crop integration and presents brief descriptions of relevant experiences in other
countries.
ESAP appreciates the kind offer of the author for granting permission to publish
this study to help disseminate available knowledge in improved forage production, and
calls for similar contributions in other areas of animal production in Ethiopia from
experienced researchers, trainers and development practitioners.
The designations employed and the presentation of the material in this
publication are entirely those of the author and do not imply the expression of any opinion
whatsoever on the part of ESAP. Further queries can be directed to the author.
Workneh Ayalew, Ph.D.

President, ESAP
Alemayehu Mengistu, 2002 V

ACKNOWLEDGEMENT
The fieldwork and early on-farm trials, which provide much of the
information for this study, were conducted by staff from the Ethiopian Ministry
of Agriculture under the Animal Nutrition and Forage Production Program of the
Fourth Livestock Development Project. The study was also conducted in
collaboration with Alan Robertson and the financial analysis was made by Philip
Young. In addition, Berhanu Shiferaw, Hadera Gebru, Tegegnwork Haile, David
Young and Shane Colville-Stewart also contributed to the publication. Special
thanks also goes to Ato Gebre for his encouragement.
The author wishes to express his gratitude to those people who have
contributed and provided encouragement towards the realization of this
publication. Finally, the author extends its acknowledgement to the Ethiopian
Society of Animal Production for making the publication of this study possible.
Alemayehu Mengistu, 2002 VI

PREFACE
This study reviews the Animal Nutrition and Forage Production Program
of the Fourth Livestock Development Project (FLDP) implemented in Ethiopia
between 1987 and 1994 with finance from the World Bank, and the Government
of Ethiopia. The study discusses the main technical, institutional, economic, and
farmer centered issues, which were central to the success of the forage production
program. It outlines appropriate strategies and policies for the integration of
improved forage production into farming systems in Ethiopia. The objectives of
this program were to increase the quantity and quality of forage produced and to
reduce the incidence and impact of soil degradation.
Rapid population growth and land degradation is decreasing the area
available for livestock grazing in Ethiopia. This results in overgrazing of non-
arable areas, which exacerbates the land degradation cycle. This trend can be
productively reversed by developing and implementing sustainable farming
systems, which integrate livestock, and cropping systems to increase the quantity
and quality of ruminant forage.
Conservation-based improved forage production provides the means to
reduce the impact of livestock on increasingly limited and degraded resources.
The study analyses the principal biological, technical, economic and institutional
issues and summarises the opportunities and constraints for wider adoption of
improved forage production and use It outlines strategies for sustainable forage
and browse production from integrated farming systems.
The financial and economic benefits from increased forage production are
substantial. They are not limited to just increased animal production, because if
implemented correctly, the forage development strategies discussed in this paper
impact on many aspects of whole farm systems. For example, the use of multi-
propose tree legumes results in increased production of forage, fuel wood,
building materials, nitrogen for crop production, honey production, and even
materials for the construction of local agricultural implements. In addition, forage
production increases food crop production through reduced soil erosion and
improved soil structure. In this regard, increased dung production is important in
the overall fuel and nutrient balance. Improved forage production must be part of
an integrated agricultural production system if Ethiopia is to overcome declining
animal and food crop production.
Alemayehu Mengistu
Addis Ababa, Ethiopia
August, 2002
Alemayehu Mengistu, 2002 VII

ABBREVIATIONS
AEZ Agro-Ecological Zone

AFRDMD Animal and Fishery Resources Development Main Department
ARDU Arsi Rural Development Project
CADU Chilalo Agricultural Development Project
CP Crude Protein
DM Dry Matter
ESAP Ethiopian Society of Animal Production
FLDP Fourth Livestock Development Project
ILCA International Livestock Center for Africa
M&E Monitoring and Evaluation
TLU Tropical Livestock Unit
Alemayehu Mengistu, 2002 VIII

I. OVERVIEW OF LIVESTOCK FEEDING IN ETHIOPIA
Introduction
Livestock production contributes up to 80 percent of farmers' income in

Ethiopia and about 20 percent of agricultural GDP. Ethiopia has the largest
livestock population of any country in Africa. Nutritional factors are the binding
constraint to sustaining livestock production in Ethiopia. Uncontrolled grazing of
increasingly scarce common areas has contributed to the degradation of many
range and pasture lands. Degradation in the form of soil erosion, deforestation,
and declining soil structure and fertility has a social and economic cost which
nations and individuals cannot afford. Simple biological solutions to these
problems have been developed by the Animal Nutrition and Forage Production
Program of the Fourth Livestock Development Project (FLDP) in Ethiopia. The
solutions are readily adopted by farmers and pastoralists, and are economically
viable. Most importantly, improved forage production provides a source of
protein, which greatly increases the productivity from crop by-products, a
valuable energy source for ruminants.
Increasing populations and declining land productivity results in
increasing demand for arable land in much of Ethiopia. This increasing demand
for cropping land to produce food for humans reduces the amount of land
available for natural grazing and forage production. Livestock numbers have
increased to meet the demand for draught animals resulting from increased
cropping activity. These conflicting developments place an unsustainable demand
on land resources, which is compounded by the transport of nutrients away from
fields in the form of grain, crop residues and dung used for fuel. Soil fertility and
structure is declining resulting in decreased productivity, erosion and general
degradation of the natural resources upon productivity, erosion and general
degradation of the natural resources upon which most East African economies
depend. The positive aspect of increased intensity of sustainable cropping
systems is an increased supply of crop by-products. These present a valuable
source of energy, which, if supplemented with protein from improved forages,
provide an economic and productive ration for ruminant livestock. By using
multipurpose browse legumes in forage strategies, additional fuelwood is
produced to substitute for dung fuel, which can then be returned to the soil to
maintain crop and forage productivity. Thus integration of livestock and cropping
systems is essential for sustainable natural resource management improved
livestock productivity.
Sustainable livestock and crop production in Ethiopia is dependent on
dramatic changes in livestock management systems. The key components of
Alemayehu Mengistu, 2002 1

these changes are a shift towards more intensive feeding systems, with more
emphasis on cut-and-carry feeding, and a gradual shift away from uncontrolled
grazing, particularly on uplands and sloping areas. This may need to be combined
with decreasing livestock populations in some areas – perhaps associated with
small-scale mechanisation of cropping systems, which currently rely on animal
draught power for cultivation. The use of woody leguminous species in
agroforestry, alley cropping or browse coppice systems is one of the key
elements of sustainable agricultural systems in Ethiopia. Legumes are especially
emphasised because of their multipurpose utility, and their dual roles in animal
nutrition and the maintenance or improvement of soil fertility and hence crop
production.
The FLDP was a five-year program designed to address these issues in
Ethiopia. Its main objective was to improve livestock and agricultural production
in Ethiopia through increasing the efficiency of resource utilisation at farm level.
Increasing foreign exchange earnings and decreasing land degradation through
increasing live animal and hide and skins exports and through import substitution
of dairy products are also important objectives. The project commenced in 1988
and achieved significant success with its Animal Nutrition and Forage Production
Program. The objectives of this program were to:
• increase the supply of forage for ruminant livestock;
• conserve soil on arable land and catchment areas;
• increase meat and milk production by increasing the quantity and quality of livestock
feed;
• increase manure production;
• increase draught power for cropping; and
• increase fuelwood and other tree products.
FLDP successfully developed and implemented a number of forage

production and animal-feeding strategies, which were integrated with cropping
systems and in almost all cases, avoided displacement of arable crops. The key
strategies were complementary to arable cropping – something, which increased
their acceptance by farmers. The strategies were designed with farmers and
demonstrated on farms to increase the spontaneous adoption of key strategies.
Because of the diverse growing conditions and farming systems in Ethiopia, a
range of strategies and species mixes were developed and implemented for the
major agro-ecological zones.
Agro-Ecological Zones of Ethiopia
The agricultural and livestock production potential of Ethiopia is

determined by soils and agro-ecological zones. Most soils in Ethiopia can sustain
some form of improved forage production. Agro-ecological zones (AEZ)

determine what species can be used and what forage strategies can be used to
integrate livestock and cropping systems. The major factors considered in
determining AEZ are:
• length of growing period – a function of rainfall, evapotranspiration, soil water storing
capacity and meteorological hazards;
• thermal zone – a function of temperatures prevailing during the growing season and
closely related to altitude; and
• landscape – a function of aspect, soil type and slope.
There are about nine principal AEZs where livestock and cropping
systems are environmentally and economically suited to forage production
(AACM, 1987). These are combinations of length of growing period and thermal
zone classes, which can be summarised as follows:
Length of Growing Period Thermal Zone (Altitude)

LII 91 – 150 days T2 500 – 1300 meters
LIII 151 – 210 days T3 1300 – 2000 meters
LIV 211 – 270 days T4 2000 – 3000 meters
Seasonal Constraints to Forage Production
During the latter part of the dry season livestock feed is normally in short
supply and is also of poor quality. Residues from cereals (wheat, teff, millet and
sorghum for example) are the main source of forage but these are low in protein
and have poor digestibility. Removing them from the fields also reduce organic
matter content in the soil which degrades soil structure and increases the
erodibility of cropped land. The production of adequate quantities of good quality
dry season forages to supplement crop residues and pasture roughages is the only
way to economically overcome the dry season constraints affecting livestock
production in Ethiopia. The use of deep rooted perennials such as browse
legumes reduce the impact of the dry season because browse species have root
systems which better able to exploit soil water reserves than forage species
(Alemayehu M., 1988).
Integration of Livestock and Cropping Systems
Because of growing pressure on land resources from increasing

populations and greater cropping intensity, forage crops can only be produced in
sufficient quantities if livestock and cropping systems are integrated. Livestock
and cropping systems can be integrated in a number of ways, including:

Agroforestry Where trees and shrubs producing browse or forage pods are
integrated with cropping systems;
Intercropping Where crops for human consumption are undersown or intercropped
with forage legumes; and
Ley Farming Where crops for human consumption are relayed or rotated with
forage legumes.
Increasing the intensity of cropping through agroforestry, intercropping or

ley farming requires careful management of soil structure and soil fertility if
productivity is to be sustained. Animal waste and agricultural by-products are
important soil conditioners and sources of nutrients. Unless they are returned to
the cropped soil, nutrient transfers will rapidly decrease soil productivity and
increase the risks of soil degradation. Feeding tethered or kraaled animals in
fields returns nutrients to the soil ready for subsequent crops. The use of
multipurpose browse trees and shrubs increases fuelwood resources available to
farming households – decreasing the need to use dung as fuel and increasing the
availability of dung for use as fertiliser. The use of legume forages frequently
increases soil nitrogen available for food crops because of their ability to fix
nitrogen. Finally, improved legume forage and browse species provide a
sustainable source of protein which enhances the ruminant livestock productivity
from crop residues high in energy. There is thus a resource stabilising cycle of
integrated livestock and cropping systems which can be started with improved
forage and browse legumes or broken without them. Work in much of Africa
demonstrates that intercropping legumes and cereals increases the productivity
and sustainability of farming systems and improves the quantity and quality of
livestock feed available from such systems.
Improved Forage Species
Many indigenous forage species in Ethiopia have low productivity or low

digestibility, which reduces their usefulness for livestock nutrition. Leguminous
species selected for their productivity, palatability, and ability to withstand
managed grazing can significantly increase livestock production. There are many
leguminous forage and browse species suited to Ethiopia, and these are discussed
in Chapter V and detailed in Annex 1. Many of the tropical species originate
from the Caribbean and Central America, and many of the temperate species
originate from the Mediterranean and West Asia. Improved grasses, many of
African origin, have greater palatability and productivity than other indigenous
species and are therefore desirable additions to pastures and common grazing

areas. Because most forage production strategies involve integration of forage

and crop production systems, potentially rampant species are not generally
recommended except where they are to be used exclusively for non-arable
grazing areas.


II. INTEGRATION OF LIVESTOCK AND CROPPING SYSTEMS IN

ETHIOPIA
Dominant Cropping Systems in Livestock Producing Areas
More than 75 percent of Ethiopia's livestock population is found in the

mixed farming areas of the highland and middle altitude zones. The AEZs, which
support mixed livestock and cropping systems, have three dominant cropping
systems (AACM, 1987). In lowland areas, below 2000 m altitude, valley mixed
agriculture supports rainfed cereals (maize, sorghum, teff), and tree crop
production. Livestock provide meat and milk in these areas as well as draught
power. In middle altitudes, between 2000 and 2400m, horticulture-livestock (hoe
culture) and some plough culture complexes support enset (false banana – Ensete
vetricosum), root crops, tree crops, and rainfed crops (including sorghum, teff,
chickpea and maize). Livestock are important sources of meat, milk and
transport, and provide draught power. High altitudes, above 2400m, support a
highland mixed crop-livestock complex, which includes rainfed cereals and
pulses (including barely, maize, wheat, horsebean and vetch) and places
importance on cattle as a source of milk and draught power.
Opportunities for Integrating Livestock and Cropping Systems
Rainfed cereal and tree cropping systems in lowland areas present

significant opportunities for integrating forage and crop food production. Longer
growing periods and suitable thermal and soil conditions in much of this area
enable undersowing and intercropping strategies to be adopted successfully. A
broad range of suitable forage legume species exist and have been successfully
demonstrated and adopted throughout this zone in Ethiopia. Alley cropping with
browse legumes and contour forage strips are also appropriate in this zone –
combining conservation cropping with production of forage of browse as well as
other tree products (fuelwood, timber and honey for example). Intensification of
cropping systems through agroforestry and intercropping or undersowing offer
significant increases in productivity and sustainability in lowland areas.
Undersowing and intercropping strategies used in lowland farming
systems are also suited to middle altitude systems. Tree crops can also be
undersown with sprawling leguminous forages, which maintain soil structure and
fertility as well as producing forage. Permanent pastures and stock exclusion
areas developed for improved forage production reduce grazing pressure on
cropped areas and, where browse legumes are used, provide an alternative fuel
source which releases dung resources for fertilisation of cropped areas. Forage

strip and alley cropping strategies are also suited to middle altitude cropping
systems and have the advantage of being able to use a wider range of species than
the lowland or highland systems.
Highland cropping systems are less suited to undersowing or
intercropping but forage crops grown as relays or in rotation with cereal crops
offer opportunities for better integration of livestock and cropping systems.
Contour strips of browse or forage legumes combined with thick grasses increase
the sustainability and productivity of most soils whilst also providing high
quality forage to supplement low quality roughages and crop residues. Poorly
drained areas and uplands can be developed as permanent pastures and stock
exclusion areas which, although not directly integrated with cropping areas,
reduce grazing pressure on cropped land. Inclusion of woody browse legumes in
stock exclusion areas not only increases quality forage production but also
provides an alternative fuel source, which enables dung resources to be used on
cropping areas. In this way, well-managed permanent pastures and stock
exclusion areas provide an important resource, which is integral to sustainable
crop production. Increasing cropping intensities to support growing populations
demand more draught animals, which places an unsustainable burden on the
reduced areas available for grazing. A combination of small-scale mechanisation
and increased use of browse legumes on upland stock exclusion areas are
essential if highland agricultural systems are to be sustained.
Key Principles
Livestock development has frequently focused on animal health or

improved livestock breeds. Unless livestock producers first improve the quantity
and quality of forage available for their livestock, these investments will
normally be uneconomic and environmentally unsustainable. Livestock
development therefore needs to firstly focus on forage production. This is more
readily adopted by farmers and is most economically done by better integrating
livestock and cropping systems. The key principles for the successful integration
of livestock and coping systems include:
• The widespread linkage between improved forage and browse production and other
livestock development benefits such as artificial insemination (AI), animal health
services, provision of breeding stock, and agricultural credit.
• The use of forage and browse legumes to increase the quantity and quality of livestock
forage, to supplement crop residues and natural roughages to increase their intake and
utilisation by ruminants, and to maintain soil structure and fertility. As a general rule a
mixture of one part improved legume forage to two parts crop residue or natural
roughages will economically optimise nutrients available to ruminants.
• The use of cut and carry systems to control grazing of stock exclusion and cropped
areas and to preserve uplands, catchments and recharge areas essential for sustainable
water supplies. Cut and carry systems optimise the use of forage by controlling browse

and forage harvesting and reducing the amount energy used by animals for walking.
Animals should be tethered or temporarily kraaled near forage sources to reduce cut
and carry labour requirements and to improve nutrient cycling in livestock-cropping
systems. Tethering systems are particularly suited to cattle being fed from contour
forage strip, undersowing, and alley cropping strategies. Kraaling systems are
particularly suited to small ruminants being fed from stock exclusion areas, undersown
cereal or tree crops, and forage banks.
• The use of browse legumes in agroforestry, alley cropping and forage bank systems to
add a third dimension to the root and foliage resources of farming systems. This third
dimension - roots reaching deep into the soil resource and branches reaching upwards
– is especially important where lateral expansion of cropping areas is restricted
because of increasing population or land degradation. This offers the single most
important opportunity for increasing sustainable productivity in Ethiopian farming
systems. Appropriate species have been widely demonstrated and accepted by farmers
in much of the country.
• Conservation of soil and water resources by using more productive and sustainable
farming systems, which focus on utilising improved forage and browse resources to
increase household income from livestock fattening or increased milk production.
Impact of Increased Crop Productivity on Livestock Feeding
Cropping systems are expanding and intensifying to feed growing human

populations and overcome decreasing productivity due to soil degradation and
poor husbandry. By adopting strategies, which integrate livestock and cropping
systems, there is considerable potential to not only increase crop yields but to
also increase the quantity and quality of forage for ruminant livestock. The
positive impacts of increased sustainable cropping include more crop by-
products, more forage and browse legumes where forage production strategies
are integrated with sustainable cropping, and a better mix of nutrients from these
sources of forage. In most areas of Ethiopia sustainable cropping systems will
produce more dry matter of better nutritional value for ruminant forage than
common grazing areas. This factor overcomes the key negative impact of
increased cropping which is a reduction of the area available for livestock
grazing. Table 2.1 demonstrates the benefits of improved legumes – experience
which is widespread throughout the highlands of Ethiopia.

Table 2.1: Effects of Protein Supplements on Liveweight Gains of Zebu Bulls

Kept on a Basal Diet of Maize Stover in Cameroon
Liveweight Gain Stover Intake

Treatment (g/day) (kg/day/animal)
Basal Diet -8 4.16
Cottonseed Cake 154 4.12
Leucaena 236 325
Source: Wegad and Ndumbe, 1986
Table 2.2: Comparison of Browse and Concentrates on Dry Matter Intake

and Milk Yield of Barka x Boran Cows in Ethiopia
Concentrate Browse Mean Milk

(kg DM/ (kg DM/ Total DM Yield
Group d/cow) D/cow) (kg/d/cow) (kg/cow/d)
Control 5.27 - 5.27 7.62
0.3# Sesbania 3.44 1.69 5.13 8.25
0.6 Sesbania 2.05 1.92 3.97 7.70
0.3 Leucaena 3.26 1.41 4.67 8.25
0.6 Leucaena 1.86 3.11 4.97 8.65
# 0.3 and 0.6 show group concentrate allowance.
Source: Biru et al, 1988
Because uplands are generally unsuitable for sustainable cropping, they

are often used to graze livestock displaced from newly cropped land. This is
unsustainable and is best managed by excluding livestock from uplands and
developing them as forage and browse reserves to be harvested for cut and carry
feeding systems. Uplands can be quickly and cheaply enriched using oversowing
and planting strategies. Unless uplands are managed in this way, they will
normally erode. This quickly degrades the productivity of uplands and threatens
the productivity of lowlands and water systems influenced by the upland
catchment area.
Financial Impact of Integrated Farming Systems
Benefits from integrated crop/forage production systems are substantial,

prolonged and complementary. Agronomic benefits are well documented and
include increased crop yields, reduced soil erosion, improved livestock
production – higher weight gains, lower mortality rates and increased milk
production – and additional supplies of fuel (dung and wood). However, the
financial benefits from integrated livestock/cropping systems are not so well
known and are more difficult to quantify accurately. For example, the impact of
improved supplies of forage can be reduced by the presence of internal parasites,
and milk production can be limited by low genetic potential. Furthermore, it is

not necessary to include all benefits attributable to forage development programs
in the analysis of projects based on the integration of cropping and livestock.
This is because there are numerous primary, secondary and even tertiary benefits
associated with such projects, and it is usually possible to generate a satisfactory
rate of return without having to include all benefits. Forage development projects
are characterised by low levels of public and on-farm investment and recurrent
expenditure, and therefore benefits per farm need only reflect increased
productivity of the order of 20% for such projects to be financially and
economically viable.
Examples of integrated crop/livestock models used for project analysis are
given in Annex 2. These were prepared for analysis of the FLDP, and are
proving to be reasonably accurate, if not conservative. The models, based on a
spreadsheet model called STRATMOD, indicate financial returns per incremental
person day of about Birr 3.00 - 8.00. These are quite acceptable when compared
with an opportunity cost of time of about Birr 0.50 to Birr 1.50 per day,
depending on the food cropping cycle. Furthermore, the financial returns detailed
in the models are based on the inclusion of only two main streams of benefits –
the impact of improved nutrition on livestock production and the value of
increased supplies of two important by-products (wood from tree legumes and
dung for fuel). If other benefits such as increased milk yields, improved crop
yields (due to increased supplies of nitrogen), reduced soil erosion, and reduced
herding time were also included in the analysis, returns per incremental person
day would be substantially higher – probably of the order of Birr 15 per person
day. Person day rates of this magnitude indicate very high financial rates of
return and acceptable economic rates of return.
Examples of the livestock benefits from integrated crop/forage models are
given in Annex 2.


III. IMPROVED FORAGE EXPERIENCE IN OTHER COUNTRIES
Australia
Temperate Australia can be divided into three zones on the basis of

rainfall and grazing management systems, as shown in Table 3.1. Only zones 2
and 3 have a sufficient rainfall to support improved forage crops.
Table 3.1: Temperate Australia Agroecological Zones
Zone Rainfall
1. Shrub Rangeland < 300mm
2. Cereal Livestock 300-600 mm
3. High Rainfall > 600 mm
Pasture and forage production was of secondary importance to cereal

production during the early years of agricultural development in Australia and
consequently received little attention. Stock originally grazed native pastures.
The native tussock grasses and other perennial species were well adapted to long
dry periods, irregular rainfall patterns and poor soil fertility – specifically
phosphorous deficiency. Cultivation for crop production and increased impact
from treading due to the high grazing intensities of introduced hooved livestock
resulted in native pastures loosing productivity and becoming severely degraded
in some areas. Fig. 3.1 also shows the impact of unsustainable cropping practices
on wheat yields between 1860 and 1895 before the use of fertiliser and forage
legumes increased the sustainability and yield of wheat production in temperate
Australia.
In 1920, a variety of Trifolium subterraneum named Mount Baker was
identified which was suited to southern Australia's growing seasons and farming
systems. This started the development of what came to be known as the
Australian ley farming system of crop/pasture rotation. Medicago spp., which
favour alkaline to neutral soils and Trofolium spp., which favour acid to neutral
soils; increased the sustainability of temperate Australian farming systems by
enabling livestock to be integrated into cropping systems. These legumes provide
cereal root disease control, improve soil structure thorough added organic matter
and ground cover, and increase soil fertility through added nitrogen as well as
increasing the quantity and quality of livestock forage. The key to sustainable
farming systems in southern Australia is the widespread use of forage legumes.
The high rainfall zone of temperate Australia sustained low stocking rates
until phosphate fertilisers and selected trace elements were used to correct

nutrient deficiencies. This and the introduction of forage legumes such as Alfalfa
(Medicago sativa), White Clover (Trifolium repens), and Tagasaste/Tree Lucerne
(Chamaecytisus palmensis) increased sustainable carrying capacity from 0.1
TLU/ha to 1.5 TLU/ha. In 1990 more than 75 percent of the temperate cereal-
livestock and high rainfall zones used legumes in their farming systems –
covering an area of more than 20 million hectares.
Fig. 3.1: Trends in Wheat Yields in Australian Since 1870

Source: Donald, 1982
Tropical Australia has a limited range of natural forage species, which

produce the quantity and quality of forage required for efficient livestock
production. Since the 1950s there has been a pasture revolution in tropical
Australia with a large increase in the range of forage grasses and legumes grown.
Since 1945 more than 50 tropical and sub-tropical legumes have been introduced
from expeditions to the Caribbean, Southeast Asia, India and Africa. Only a third
of these have become widely adapted and naturalised. Fig. 3.2 shows a
cumulative introduction record of grasses and legumes into tropical Australia.
Over the past 25 years, the area sown to improved pasture and forage legumes in
tropical Australia has increased to nearly 5 million hectares of which nearly 2
million hectares contain legume varieties. This is illustrated in Fig. 3.3.

Fig. 3.2: Names and Cumulative Number of Forage Species Introduced Into
Queensland, Australia a/
The first letter of each species aligns with its date of first release or use. Species widely
planted, past or present, and those now naturalised are underlined.
Source: Gramshaw and Walker, 1988

Fig. 3.3: Area of Sown Patures and Forages in Queensland, Australia – Total
Area Sown and Proportion Sown Solely to Grasses
Source: Gramshaw and Walker 1988
Farmers in tropical Australia sow approximately 300,000 hectares of

improved pasture and forage species each year. Legumes used by farmers in
tropical Australia include centro (Centrosema pubescens) and Puero (Pueraria
phaseoloides) in the high rainfall coastal zone; and Siratro (Macroptilium
atropurpureum), Seca (Stylosanthes scabra) and Verano Caribbean Stylo (S.
hamata) for the drier coastal areas. Highland farmers use Greenleaf Desmodium
(Desmodium intortum), Glycine (Neonotonia withtii), Creeping Vigna (Vigna
luteola), Haifa White Clover (Trifolium repens), and Safari Clover (T.
semipilosum). The coastal wetlands and tablelands, which have rainfall above
1500 mm, are sown to improved grasses including Guinea Grass (Panicum
maximum), Panic (P. maximum var. trichoglume), Signal Grass (Brachiaria
decumbens) and Para grass (B. mutica). Highland areas are sown to Rhodes Grass
(Chloris gayana), Setaria (Setaria anceps), and Kikuyu (Pennisetum
clandestinum).
In the savannah areas of tropical Australia, with rainfall between 700 and
1500 mm, farmers replace natural grasses with improved forages to increase the
quantity and quality of feed for cattle. Legumes successfully used by farmers in
these areas include Seca Stylo (Stylosanthes scabra), Siratro (Macroptilium
atropureum), Wynn Cassia (Cassia rotundifolia), Verano Stylo (S. hamata),
Creeping Vigna (Vigna luteola), Glycine (Neonotonia wightii), Leucaena
(Leucaena leucocephala), Alfalfa (Medicago sativa), and Haifa White Clover
(Trifolium repens). Legumes well suited to soils in this zone with low to
moderate fertility are Oxley Fine Stem Stylo (S. guianensis), Miles Lotonis
(Lotonis bainesii) and Maku Lotus (Lotus sp.).

New Zealand
New Zealand has placed importance on improved forage production for

many years, with significant research work being conducted on temperate
species. Some of the temperate forage legumes developed in New Zealand are
relevant to highland areas of Ethiopia – especially Trifolium repens. Recognising
the need to increase productivity of uplands and to reduce soil loss and provide
shelter, New Zealand researchers have emphasised the use of browse species to
increase the quantity and quality of livestock forage. Particular emphasis has
been placed on Tree Lucerne or Tagasaste (Chamaecytisus palmensis) because of
its high productivity, excellent nutritional characteristics and good palatability,
especially for small ruminants. Significant areas of the uplands are now planted
to contour strips and forage banks of Tagasaste, with annual production of up to
900 kg dry matter (DM) per tree. Contour browse strips occupying 30 per cent of
the land area – leaving 70 per cent for pasture – gives up to 22 per cent more DM
per hectare than improved pasture alone (Townsend and Radcliffe, 1990). Small
ruminants in New Zealand show a marked preference for fresh Tagasaste but
cattle find dried or wilted Tagasaste more palatable than fresh leaves (Lambert,
1989). The New Zealand experience with Tagasaste is especially relevant to the
upland stock exclusion areas of the Ethiopian highlands.
South-East Asia
South East Asia is not a significant producer of ruminants but several

countries have cultures which value ruminant meat, especially Malaysia and
Indonesia. Forage production strategies for ruminants has focussed on the use of
shade tolerant, sprawling legumes undersown into tree crop plantations. This was
the most economic and sustainable means of producing high quality forage in
these countries. Undersowing coconut, young oil palm, and rubber plantations
with forage species reduces weed growth, protects the soil from erosive rains,
and leads to increased income during the establishment of tree crop plantations
and greater income security generally.
Experience in mature coconut plantations in Indonesia suggest that up to 5
TLU per hectare can be sustainably grazed to give long term production of 550
kg liveweight gain/ha/year (Humphreys, 1987). Key species used in South East
Asia under tree crop plantations are Macroptilium atropurpureum, Centrosema
pubescens, Stylosanthes guianensis, setaria anceps, and Brachiaria decumbens.
This experience is relevant to some lowland areas of Ethiopia, especially where
coffee, enset, and fruit crops are grown in higher rainfall areas.

Caribbean
The Caribbean is not a significant livestock producing region but

increasing demand for meat and milk products developed interest in improved
forage in the 1970s and early 1980s. Many forage legumes with desirable
productivity, palatability and grazing tolerance characteristics were indigenous to
the region or to the Yucatan Peninsula in Mexico. The predominantly acid and
infertile Oxisols and Ultisols supported a range of Stylosanthes, Desmodium and
Brachiaria species. Researchers from the Caribbean Agricultural Research and
Development Institute (CARDI) triailled selections of many grass and legume
species throughout the region. This work was coupled with livestock feeding
trials to ensure that species adapted to the soils and climate of the region were
also palatable and nutritious to ruminant livestock. All new legume introductions
were inoculated with appropriate rhizobia to ensure adequate growth and
nitrogen fixation.
This work has led to improved pastures and some agroforestry in Antigua,
Barbados, and Jamaica. Key species are Macroptilium atropurpureum,
Neonotonia withtii, Stylosanthes hamata, Centrosema pubescens, Leucaena
leucocephala, Setaria anceps, Panicum maximum, and Andropogon gayanus.
More than 25 per cent of the livestock pastures of the Caribbean now include
improved forage species (CARDI, 1986). Recent developments include extension
of alley cropping systems using Leucaena, Calliandra and Gliricidea species, and
the use of forage banks of Stylosanthes humata for dry season feeding.

IV. IMPROVED FORAGE PRODUCTION STRATEGIES
The strategies developed and successfully implemented by the FLDP in

Ethiopia evolved from experiences in other countries and an understanding of the
importance of matching forage systems to AEZs. The strategies are farmer
centered and were developed with farmers to maximize sustainable income
generation and food production at the household level. The key forage production
strategies are conservation based and promote the use of legumes as improved
forage (Alemayehu M., 1989; Robertson, 1990). The key strategies are divided
into two categories:
On Farm Strategies Common Land Strategies

• Backyard Forage Production • Oversowing Common Grazing Areas
• Undersowing and Interplanting • Stock Exclusion Areas/Forage Banks
• Countour Forage Strips • Permanent Pastures
• Agroforestry
Backyard Forage Production
Backyard forage production is based on small plots and hedges of

productive forage and browse planted within house compounds and around their
boundaries. This is the most important initial strategy since it is developed in the
farmer's household, and is very convenient for intensive feeding of dairy animals
or fattening of meat animals. The higher fertility levels typically found in and
around house compounds also helps with the successful establishment of
backyard forage. This strategy has a major impact in exposing farmers to the
management and productivity of new species and also provides a seed bank to
help establish new plantings for other forage strategies. Woody leguminous
browse species are particularly suited to this strategy because of their
multipurpose benefits and rapid growth rates. Tall growing tropical grasses are
also suited to backyard forage development. Tree legume hedges have been the
most widely adopted backyard forage strategy and need to be used as an
incentive for broad-scale forage development based on contour forage strip and
undersowing strategies. This strategy introduces farmers to the concept of
supplementing crop by-products and poor quality roughages with high quality
forage in a location, which facilitates close attention to management.
Backyard forage provides significant quantities of both forage and
fuelwood where they can be conveniently used. Other benefits perceived by
farmers include shelter, increased privacy, wood products construction and
implements, and bee products. The multipurpose benefits of backyard forages
provide a range of incentives for farmers to adopt this strategy. It should be one

of the first strategies to be promoted by extension agents since it is easily

established and managed, and provides the means to reduce grazing pressure on
common grazing areas. Backyard forage can be cut and carried to tethered or
housed animals, or cut and conserved for dry season use in mixes with crop
residues and natural pasture hay or roughages. Experience from Ethiopia testifies
to the utility of backyard forage species used by the FLDP and summarised in
Table 4.1. Experience in New Zealand suggests that Tree Lucerne will produce
up to 900 kg DM/tree each year (Townsend and Radcliffe, 1989) and has a forage
value similar to alfalfa. Goats fed maize husk and Leucaena in Zambia, mixed in
a ratio of 3:2 on a DM basis, gained 29 g/day compared with goats fed maize
husk and 1 per cent urea which gained 5 g/day. The digestibility of the ration
increased from 47 per cent for maize husks alone to 63 per cent for the 3:2 maize
husk:Leucaena ration (Phiri, 1992). Supplementation of Guinea Grass (Panicum
maximum) hay fed to goats with 100 g DM/day Sesbania sesban leaves resulted
in total DM intake of 626 g DM/day compared with 498 g DM/day without the
browse supplement (Ash, 1990).
The backyard forage strategy provides an opportunity to reach large
numbers of farmers very quickly and can therefore have a great impact
nationally, even in the short term. Demonstrations of about 100 browse legume
seedlings or grass sets should be established in the housing compounds of contact
farmers. This numbers is necessary to ensure sufficient high quality forage to
supplement conserved roughages and crop by-products fed to household
livestock. Forage seedlings or sets can be planted in any pattern to suit the needs
of the household but simple boundary hedges/shelter belts or forage blocks are
the most widely accepted designs for backyard forage plantations.
The extension emphasis should be on browse legumes and large grasses
and the production of bare rooted seedlings in backyard nurseries. This ensures
that farmers develop the capacity to grow their own seedlings or sets for
expansion of forage development using contour forage strips and other strategies.
In this way, farmers develop familiarity with the propagation, growth and
management of key species. Backyard nurseries are typically 4 to 5 square metres
in size and are initiated with small packets of seed containing 50 to 100 grams of
seed. It is feasible to distribute these seed packets to vast numbers of farmers
each year. Wide distribution of seed and promotion of hedges, backyard forage
banks, ensure the farmers' capacity to grow bare rooted seedlings for planting in
other areas.
Once these components of the backyard forage strategy have been
adopted, extension efforts can focus on the use of backyard forage to reduce
grazing pressure on common areas and increase livestock productivity from poor
quality roughages. Wide acceptance of the backyard forage strategy also provides
a sound foundation for farmers to establish grazing management groups or
pastoral associations to control grazing on common lands and cropped areas. This

then provides the basis for adoption of the contour forage strip and livestock
exclusion area strategies.
Table 4.1: Key Species for Backyard Forage
Altitude Browse Legumes Forage Legumes Grasses

<2000m Leucaena Greenleaf Rhodes Grass
Sesbania Silverleaf Elephant Grass
Pigeon Pea Alfalfa Panicum
2000-2400m Sesbania Alfalfa Phalaris
Pigon Pea Vetch Elephant Grass
Tree Lucerne Verano stylo
>2400m Tree Lucerne Alfalfa Phalaris
Vetch Oats
Undersowing and Interplanting
Undersowing and interplanting is the establishment of forage species in an

annual crop or perennial plantation. This strategy provides the most convenient
approach to rapidly increasing on-farm forage supplies over a large number of
farmers and should have a major impact in the short to medium term. The use of
legumes in this system will contribute to the improved fertility and structure of
cropping soils. Farmers seeing on-farm trials of undersowing and interplanting
accept the strategy readily and understand the benefits and techniques very
quickly. This is normally the second strategy to promote after backyard forage
has been adopted by farmers. Undersowing and intercropping are probably the
most important of the forage development strategies.
Undersowing works best with sprawling, low growing annual legumes but
can also work well with climbing legumes. The strategy is particularly suited to
the production of tall growing cereals such as maize, sorghum or millet but also
works with other cropping systems. Undersowing with legumes produces large
quantities of high quality forage for utilisation by either post harvest grazing or
cut and carry systems. The undersown forage protects the soil from erosive rains,
can contribute nitrogen for the food crop, and balances the forage value of crop
residues such as stover and straw to increase its intake and utilisation. The
strategy works well with sprawling and climbing legumes but is also effective
with other forage legumes and dual purpose legumes such as cow pea.
Tree crops and some vegetables can also be undersown or interplanted
with leguminous forages. The establishment of annual or perennial legumes
under tree crops is a reliable strategy, which is well accepted by farmers. It is
particularly appropriate to the more intensive horticultural and forestry systems
where the undersown legume is intensively managed with cut and carried
systems for livestock feed. The strategy primarily involves lower altitude systems
where fruit, coffee, coconuts, enset or chat are grown. There is also broad
application with eucalyptus and Acacia plantations grown for fuelwood.
Where crop weeding practices are very thorough, forages should be
undersown at the time of final weeding. This avoids any risk of the undersown
legume competing seriously with the cereal crop but often means that the
legumes have insufficient time to produce ripe seed prior to crop harvest. In areas
of poorer weeding practices, undersowing should coincide with an earlier
weeding. In this way sufficient legumes survive any subsequent weeding to
provide an adequate seeding capacity prior to crop harvest. Early maturing
cereals generally favour better forage production because they compete with the
undersown forage legume for a shorter period of the growing season. The
competitive balance between crop and undersown or intercropped forage legume
is very sensitive to sowing time. This will vary with soil and crop type, season,
and management practices, and is best determined using on-farm demonstrations.
Farmers understand the benefits of undersowing or intercropping and
adopt this strategy over a wide range of traditional cultivation and cropping
practices. Farmers are attracted by the simplicity of the program and by the high
yields of forage, which require no management input because the forage legume
is protected from grazing by the crop. Farmers acknowledge that undersowing
does not reduce crop yields but do not accept that the use of legumes helps
maintain soil fertility, even though this has been successfully demonstrated. The
incentive for adoption is large quantities of high quality forage in return for a
minimal investment. High adoption rates can only be maintained if supplies of
seed are available. Relatively large quantities of seed are required (typically 8 to
10 kg pre hectare for annual legumes) unless early undersowing practices are
used to ensure adequate seed set at the end of each season.
Good stands of undersown legumes produce 2,500 to 3,000 kg dry matter
per ha from one cut in farmers' fields (Robertson, 1990). Farmers advise that
grain yields are not depressed but that sprawling legumes such as the Desmodium
and vetch dramatically reduce weed infestations – effectively replacing weed
growth with high quality forage. Verano stylo (Stylosanthes hamata) undersown
into a three week old sorgum crop near Kaduna in Nigeria yielded 1.6 t/ha
sorghum grain, 3 t/ha DM stylo forage, and 6 t/ha sorghum residue (Saleem,
1982). The forage and by-product resulting from this undersowing is a balanced
growth diet for ruminants. Compare this total production of 1.6 t/ha grain and 9
t/ha forage with the production from the control crop without undersown forage –
2.0 t/ha grain and 7.5 t/ha sorghum residue. Middle altitude farmers in Ethiopia
undersowing maize with Desmodium uncinatum harvested an average of more
than 6 t DM/ha/year (Tadesse, 1990). This is enough quality forage to mix with
12 t DM natural hay or crop residue and fatten about 150 sheep or 15 oxen over a
120-day fattening cycle (see Chapter VI). Highland wheat crops in Ethiopia
undersown with a range of indigenous Trifolium species yielded significantly

more DM than control crops without undersown legumes. The most significant
outcome of this work was the successful intercropping of wheat with forage
legumes without any significant reduction in wheat yield. Trifolium quartinianum
was particularly efficient with broadcast undersowing at Holetta yielding 1.1 t/ha
wheat grain, 2.1 t/ha wheat straw and 3.1 t DM/ha clover hay (Kahurananga,
1988).
Table 4.2: Key Species for Undersowing and Intercropping

<2000 m Not Appropriate Cow Pea Not Appropriate
Verano Stylo
Greenleaf
Wynn Cassia
2000–2400m Not Appropriate Siratro Not Appropriate
Vetch
Greenleaf
>2400m Not Appropriate Vetch Not Appropriate
White Clover
Native Clovers
Contour Forage Strips
Forage strips are broad based mixtures of herbaceous and tree legumes,
and grasses planted on contour bunds or in narrow strips along the contour
without any physical structures. This is a multipurpose strategy providing forage,
shelter, soil stablisation, and fuelwood. Forage strips planted along the contour
contribute to soil conservation by directing ploughing along the contour and by
reducing run-off down the slope. This increases infiltration and reduces soil
erosion, especially where a thick sward of grass or herbaceous legumes is
included in the forage strip. Contour forage strips are particularly successful
when perennial, thick rooted grasses are mixed with woody leguminous species.
Because this strategy integrates forage production in cropping areas, potentially
weedy species such as stoloniferous grasses should not be used for forage strip
plantings.
Farmers perceive the principal benefits of forage strips to be the fuel and
forage products rather than the conservation attributes. It is these benefits, which
should be promoted as the incentives for adoption of forage strip strategies. They
key problem with forage strips is the difficulty some farmers have in establishing
them where livestock have free grazing access to fallow land or crop stubbles
after harvest. This problem is best overcome by involving shepherds in forage
strip establishment and promoting cut and carry feeding of animals tethered in
the field. Thus contour forage strips are more easily promoted once backyard

forage and undersowing is established to provide alternative forage sources to

stubble and fallow grazing. Animals can be kept away from planted forage strips
during their establishment if conspicuous species such as vetch are included in
the forage strip mix. In some areas, alley farming can be developed by using
long-lived browse species as part of the species mix for contour forage strips.
Alley farming requires careful location and marking of contour strips which
should be wide enough apart to allow ploughing and harvesting operations to
take place without disruption. Alley farming is best established with bare rooted
seedlings.
Typical on-farm demonstrations of contour forage strips would include up
to 1 hectare of forage strips at 4 to 10 metre horizontal intervals between strips.
Strips are up to 1 metre wide and should be continuous along the contour to
maximse their conservation function. Contours can be marked out using a simple
A frame and pendulum device. Alternatively, water levels made of two staffs
with a water-filled tube between them can be used. Where alley cropping is
developed using woody legumes along contour forage banks, seedlings or seeds
of woody species should be planted at 1 metre intervals along each contour strip.
Large bare rooted seedlings are most successful because they have a quick visual
impact and are more easily protected from grazing animals. The most reliable
species include those listed in Table 4.3. Pioneer species such as Pigeon Pea,
Phalaris, and Greenleaf Desmodium are particularly reliable understorey species
when planted with Leucaena or Tree Lucerne. Stoloniferous species such as
Rhodes Grass are not suited to contour forage strips because of their weed
potential in crop areas.
Demonstrations need to focus on the production benefits of contour forage
strips to overcome some farmers' fears that contour forage strips reduce their
arable area and so decrease their income or food security. In fact, because of
shelter, soil conservation and nitrogen benefits, well-designed contour forage
strips frequently increase the productivity of the area between strips in addition to
the products from the strip itself. This is especially true of alley cropping systems
where the third dimension provided by browse legumes increases the
productivity of the farming system. Contour forage strips produce between 2,000
and 5,000 kg dry matter per hectare of planted strip, or between 340 and 850 kg
dry matter per hectare assuming 6 metre intervals and one metre wide strips. In
addition to this benefit, there are yields of wood for fuel and construction, shelter
benefits, nitrogen fixation and bee products (honey and wax).

Forage Crop Production
Where farmers use a cropping rotation or have sufficient land, they can
grow a short-term forage crop. Short-term forage crops can be reliably introduced
over a wide range of sites but are most appropriate for farmers who rely on dairy
production for their income. Annual leguminous species mixed with cereals
provide the best quantity and quality of forage in highland areas but annual
legume forages optimise forage production in middle altitude and lowland areas.
Farmers in Ethiopia are shifting towards perennial forage production strategies
because annual forage crops do not integrate livestock and cropping systems
except where fallow is commonly used. In these areas fallow reduction strategies
based on leguminous forage crops are appropriate. Oats and vetch have
performed well over a wide range of AEZs, with oats showing good tolerance of
relatively low fertility and poor drainage. Lablab is very productive at lower
altitudes and competes well with weeds whereas alfalfa does not persist under
rainfed condition in Ethiopia.
Table 4.3: Key Species for Contour Forage Strips

<2000m Leucaena Siratro Panicum
Sesbania Axillaris Setaria
Pigeon Pea Silverleaf Vetiveria
Greenleaf
Vetch
Verano Stylo
2000-2400m Tree Lucerne Greenleaf Phalaris
Sesbania Axillaris Setaria
Pigeon Pea White Clover
Native Clovers
Vetch
Alfalfa
>2400m Tree Lucerne White Clover Phalaris
Native Clovers
Vetch
Maku Lotus
Alfalfa
Farmers accept oat/vetch and lablab strategies, especially where fattening

or dairy enterprises are viable. However, as demand for subsistence food crops
increases, forage strategies which can be integrated into cropping systems will be
adopted in preference to annual forage crop strategies.

Table 4.4: Key Species for Forage Crop Production

<2000m Not Applicable Siratro Rhodes Grass
Greenleaf Panicum
Silverleaf Setaria
Vernano Stylo
Lablab
2000-2400m Not Applicable Greenleaf Phalaris
Silverleaf Setaria
Vetch
White Clover
Alfalfa
Lablab
>2400m Not Applicable White Clover Phalaris
Alfalfa Oats
Vetch Cocksfoot
Agroforestry
Agroforestry is the combination of trees and agriculture in an integrated

and sustainable farming system. Many of the forage production strategies can be
developed as agroforestry systems. In particular contour forage banks and
undersowing of tree crops or forest plantations can be designed as agroforestry
systems where leguminous browse species provide an upper storey in a forage
system or undersown legumes and grasses provide an under storey in a forestry
or horticultural system. Agroforestry maximises the use of land by adding a third
dimension to the above and below ground areas of utilisation. This aspect is
particularly important for farmers with limited land resources. Because many
agroforestry strategies include leguminous species, they are also attractive to
farmers facing problems of declining soil productivity.
Experience in Ethiopia and elsewhere shows that the height and frequency
of cutting agroforestry browse species has a significant impact on their
productivity and forage value. For example, Leucaena produces more DM at
longer cutting intervals (>3 months) and moderate cutting height (75 to 100 cm)
than more severe defoliation. Table 4.5 shows that the leaf nitrogen from three
year old trees was also increased with longer cutting intervals (Karim et al,
1991).

Table 4.5: Effect of Cutting Height and Interval on DM Yield of Leucaena
Cutting DM Yield (g/tree) Cut Biomass N g/tree)

Height (cm) 1 month 3 months Mean 1 month 3 months Mean
25 20 60 40 0.65 1.50 1.08
50 22 71 46 0.70 1.80 1.25
75 28 126 77 0.92 3.15 2.03
100 50 96 69 1.42 2.38 1.90
Mean 30 88 59 0.92 2.21 1.56
Source: Karim et al., 1991
Similarly, total DM yield of Sesbania increased with increased cutting

interval with the highest yields recorded at 8 week cutting intervals. The 100 cm
cutting height gave maximum yields from 12 month old Sesbania, as shown in
Table 4.6 (Galang et al, 1990). Pigeon Pea also gives maximum yields with a
cutting frequency of about 8 weeks - yielding up to 50 t DM/ha each year
(Udedibie and Igwe, 1989). Table 4.7 shows that although DM yield was
maximised with a longer cutting interval, crude protein (CP) was maximised and
crude fiber (CF) minimised with shorter cutting frequencies.
Table 4.6: Effect of Cutting Height (cm) and Frequency (weeks) on the Total
Productivity (t/dm/ha) and Leaf Content (%) of Sesbania sesban cv nubica
Cut Height → 50cm 100cm 150cm 50cm 100cm 150cm

Cut Interval Total DM (t/ha/year) Leaf Content of DM (%)
4 weeks 2.4 3.3 3.3 87 88 91
6 weeks 3.2 4.2 4.2 69 70 77
8 weeks 4.1 4.9 4.4 54 60 65
Mean 3.2 4.1 4.0 67 71 77
Source: Galang et al, 1990
Table 4.7: DM Yield and Chemical Composition of Pigeon Pea Leaf Meal
Cut at Different Time Intervals
Cutting DM Composition of DM (%)

Interval Yield CP CF Ca P
(t/ha)
4 wks 2.3 24.3 24.8 1.39 0.31
6 wks 2.4 21.9 26.1 1.24 0.22
8 wks 2.7 20.1 27.1 1.09 0.23
Source: Udedibie and Igwe, 1989

Oversowing Common Grazing Areas
Oversowing is the simplest of the forage development strategies and can

be undertaken at very low cost depending on the seeding rates used. It involves
broadcasting or sowing improved forage species into common grazing lands,
native pastures and degraded areas without any cultivation or other inputs.
Typically there is no attempt to modify grazing management but existing
stocking rates should not be increased after oversowing. The strategy includes
sowing roadsides from vehicles and is suited to aerial seeding where very large
areas are to be developed. Aerial seeding is also another way of establishing
improved extensive grazing areas using oversowing techniques. This strategy is
most suited to pioneer legume species, which grow quickly and seed prolifically.
Because of the low input nature of this strategy, incremental forage yields are not
large but pioneer species with good grazing tolerance and natural seeding ability
gradually colonise common areas and improve the overall species composition
available for grazing. Natural spread of seed with water movement, grazing
animals and wind action can be rapid, enabling very large areas of land to be
developed so long as grazing management is possible to enable plants to become
established and set seed.
Farmers are more likely to gain long term advantages from oversowing
strategies if there is some of grazing management group or pastoral association,
which manages common grazing areas. This could be associated with dairy user
groups but requires the majority of farmers using common grazing land to
recognise that there is an overgrazing or low productivity problem, which can be
solved with oversowing. The provision of seed and technical support for
oversowing strategies is a sufficient incentive to encourage farmers to organise
grazing management groups or pastoral associations. Such organisations are only
successful if they are initiated by farmers in response to their perceived needs. If
suitable sites are chosen and effective grazing management of common lands
exists, oversowing of grazing areas is the most cost effective strategy for broad-
scale forage production.
Because this strategy is implemented on common grazing land by the
government, farmer acceptance is not an important issue for implementation if
the work is carried out by government staff. It is, however, a major issue for
management of oversown areas. For this reason grazing management groups or
pastoral associations, which are a prerequisite for successful long term
establishment of oeversown forage, should be closely involved in implementation
as well as management of oversowing strategies. These associations can
broadcast seed with hand cranked seed broadcasters. These inexpensive and
simple implements are easy to use and very robust. They are equally suited to

fertiliser spreading and so are an attractive implement for farmer groups and can
act as an additional incentive for organisation of grazing and pastoral groups.
Village groups should aim to oversow up to 10 ha each year in low and
medium altitudes and about 2 ha each year in the highlands. The most reliable
species for oversowing have been the stylos, which have established and begun
spreading on an extremely wide range of sites in Ethiopia. Wynn cassia and
climbing/sprawling legumes such as greenleaf and Siratro have also shown
promise. Experience in the sub-humid middle altitude areas of Ethiopia shows
that even after a short time oversown Stylosanthes guianensis (cv Schofield) and
Desmodium uncinatum can make up more than 15 per cent of pasture DM
composition and yield more than 3 t DM/ha (Tadesse A., 1988).
Table 4.8: Key Species for Oversowing Grazing Areas

<2000m Leucaena Siratro Rhodes Grass
Sesbania Axillaris Panicum
Greenleaf Setaria
Silverleaf Buffel grass
Seca Stylo
Verano Stylo
Wynn Cassia
2000–2400m Sesbania Siratro Phalaris
Axillaris Setaria
Seca Stylo
Verano Stylo
Greenleaf
Silverleaf
White Clover
Alfalfa
Alfalfa Cocksfoot
Maku Lotus
Stock Exclusion Areas/Forage Banks
Stock exclusion areas are an important means of protecting degraded

areas, key watersheds, and common land. They also provide an opportunity to
develop forage banks for use during droughts or periods of seasonal forage
shortage. Stock exclusion areas are particularly important for the conservation of
highlands but are only accepted by farmers where they see sufficient benefits to
organise grazing management groups or pastoral associations to control stock
exclusion areas and voluntarily keep stock out. The introduction of browse
species, productive legumes and improved grasses can rapidly increase the
productivity of exclusion areas. The strategy is suitable for aerial seeding

techniques which enable very large areas of land to be sown to forages quickly.
Rehabilitation of degraded areas using forage species normally provides a
good incentive for farmers and pastoralists to organise grazing management
groups or pastoral associations. Because degraded land has low value as a
common grazing resource farmers are usually willing to voluntarily exclude
livestock from these areas. Rehabilitation of degraded areas with forage species
provides an incentive for these initiatives, especially when farmers understand
the benefits of forage development. Without farmer initiated grazing
management groups or pastoral associations to control grazing, stock exclusion
areas and forage banks are unsustainable. The extension effort therefore need to
focus on the benefits of collaborative management of common lands and initially
focus on degraded areas where benefits will be maximised and the likelihood of
farmer resistance will be minimal.
Cultivation is not necessary to establish forage banks or rehabilitate stock
exclusion areas, especially on very bare sites, but broadcast sowing should take
place after commencement of the main rains to ensure that there is enough soil
moisture to sustain germination. Direct seeding with chisel tyned cultivators may
be necessary in degraded areas with scalded or hardpan surfaces. Leguminous
browse and tall grass species should always be included in stock exclusion areas
to maximise the production potential and drought resistance of the species mix.
Woody species can be planted by direct seeding but generally develop more
successfully where they are planted as bare rooted seedlings early in the main
rainy season. Annual cut and carry forage production from improved low and
medium altitude sites in Ethiopia is in excess of 6000 kg per hectare (Robertson,
1990). Other benefits include soil conservation, better recharge of shallow
aquifers, and production of fuelwood and bee products where browse species are
included in the stock exclusion area.

Table 4.9: Key Species for Stock Exclusion Areas

<2000m Leucaena Siratro Plicatulum
Sesbania Axillaris Buffel Grass
Seca Stylo Setaria
Verano Stylo
Cook Stylo
Wynn Cassia
Greenleaf
Silverleaf
2000 – 2400m Sesbania Siratro Phalaris
Tree Lucerne Axillaris Setaria
Leucaena Seca Stylo
Verano Stylo
Vetch
Greenleaf
Silverleaf
White Clover
Alfalfa
Maku Lotus
Forage banks should be established at the beginning of the wet season.

Stylosanthes hamata cv Verano and S. guianensis cv Cook are particularly
suitable for forage banks and should be established with 8 to 10 kg seed per
hectare. Forage banks are left ungrazed during the growing season to provide a
supply of high quality forage during the dry season. Once established, these
species can support up to 5 TLU/ha for up to 4 hours per day during the dry
season (Otsyina et al, 1987). Burning is not necessary for establishment but
kraaling animals on the area to be established as a forage bank prior to sowing
helps reduce weed competition and adds manure to the soil. Forage banks are
particularly important for maintaining priority animals in the household herd –
for example lactating animals and weaners. Browse legumes such as Leucaena
and tree lucerne also act as good forage banks if they are left uncut during the
growing season. Many farmers regard backyard browse plantings as forage banks
and this is an appropriate use for backyard forage strategies, which should be
promoted by extension agents.
Annual targets for stock exclusion areas of 2 to 10 hectares per village
area are possible but much larger areas have been rehabilitated where strong
village support has resulted in the formation of grazing management groups to
voluntarily exclude livestock from areas to be developed. The improvement of
stock exclusion areas is suited to cut and carry systems and is rapidly adopted by
farmers where there is a history of fattening livestock for local markets. The
location of intensive fattening or dairying enterprises adjacent to stock exclusion

areas facilitates the efficient use of cut forage and provides an additional
incentive for farmers to collectively manage their common grazing resources.
Once farmers have agreed to exclude livestock from an area, it should not be re-
opened for grazing. Forage produced from stock exclusion areas should always
be cut and carried to livestock to maintain the protected nature of the improved
forage resource.
Permanent Pastures
Permanent pastures comprise a broad range of annual and perennial

legumes and perennial grasses. Productive mixed pastures can be readily
established, particularly in the low and medium altitudes with warmer growing
conditions. Grazing management is a significant problem for sustainable pasture
production in some regions, which is best overcome with cut and carry systems.
Permanent pastures are most useful for dairy farmers who rely on optimal
productivity of their livestock investment for their livelihood. Permanent dairy
pastures should include a mix of legumes and grass species with high palatability
and productivity.
Table 4.10: Key Species for Permanent Pastures

<2000m Not Applicable Siratro Rhodes Grass
Greenleaf Panicum
Silverleaf Setaria
Seca Stylo
Verano Stylo
2000 – 2400m Not Applicable Verano Stylo Phalaris
Greenleaf Setaria
Silverleaf
Vetch
White Clover
>2400m Not Applicable White Clover Phalaris
Alfalfa Oats
Vetch Cocksfoot
Maku Lotus
Roadside Sowing
Roadside sowing is a successful means of implementing the oversowing

strategy. It is quick and effective and provides an impressive visual impact which
can be used to excite farmer interest and provide an incentive for the formation of
grazing management groups or pastoral associations. This strategy can be highly

cost-effective, particularly when using species with the ability to spread under
grazing. Sowing a broad grid of suitable roads provides a convenient mechanism
for introducing improved forage species to a large area since the rate of spread
from a very long narrow transects is high. 10 km of roadside sowing equates to
about one hectare of oversown grazing land. Seeding rates are typically 0.5 to 1.0
kg per kilometer of roadside. Mixed seed should be emptied from sacks or
buckets from the back of a reasonably fast moving vehicle. In this way the vortex
currents carry seed onto the roadside verge. Roadside sowing is most suitable for
quickly establishing and prolific seeding species, which tolerate grazing. The
stylos are the most successful species used in roadside sowing in Ethiopia.
Aerial Sowing
Aerial sowing enables very large areas to be oversown with improved

forage seeds. The success of establishment depends largely on the selection of
suitable sites. The most suitable sites have rough often gravelly surfaces. Sites
with compacted or hardpan surfaces do not enable good establishment of aerial
sown or broadcast seed. Stylosanthes are particularly successful for aerial
oversowing – being extremely resilient to grazing and a successful pioneer
species
Aerial sowing is particularly suited to the rehabilitation of large
catchments, which include relatively inaccessible areas. Where grazing is
restricted or there are protected niches because of thorn bushes or rocks,
leguminous browse species should also be included in aerial sowing mixes.
Leucaena is especially appropriate for this purpose. Seed is best dispersed from
fixed wing aircraft, which travel at sufficient speed to create air currents for seed
dispersal. If helicopters are used, spinners are normally required for efficient seed
distribution. Flag bearers on the ground or the use of prominent landmarks are
necessary to plan and manage aerial seeding operations.
In some other countries, including Australia and New Zealand, aerial
seeding has been used successfully to oversow pastures on millions of hectares of
uplands and range areas. Temperate, sub-tropical and tropical species, especially
herbaceous legumes, have been successfully established with aerial sowing.
Aerial sowing enables small quantities of seed to be uniformly and efficiently
spread very quickly and economically. The low seeding rates necessary for this
type of sowing are a major advantage for those species with the capacity to
rapidly increase density and spread. These include the stylos, the desmodiums,
some trifoliums, and Wynn cassia. Experience in Ethiopia shows that even on the
most degraded sites, Verano stylo will establish and seed within three months of
aerial sowing. Successful sowing was undertaken shortly after commencement of
the main rains.

Summary
Table 4.11 illustrates the perceived significance of the various forage

strategies in Ethiopia (FLDP, 1989). The potential economic impact, with a score
of five representing the maximum impact, is assessed on the basis of increasing
forage production and benefits in terms of fuelwood supply, erosion control and
contribution to the maintenance of soil structure and fertility. The technical
possibilities are assessed for low to medium altitudes and highlands, with an
indication of acceptance of the strategy amongst farmers in Ethiopia. Major
research needs to support further development of the strategies are listed. Fig. 4.1
gives a forage activity calendar for Ethiopia.
Table 4.11 Summary of Forage Production Strategies
Strategy Potential Technical Possibilities Farmer Research

Economic >2400m <2400m Adoption Needs
Impact
Backyard 3 Good Good Good Labour
Undersowing 2-4 Good Moderate Moderate Timing
Forage Strips 5 Good Good Variable Management
Forage Crops <1 Good Good Moderate
Agroforestry 4 Good Moderate Moderate Establishment
Oversowing 1 Moderate Poor Good Species
Forage Banks 4 Good Moderate Good Species
Pastures <1 Good Good Poor
Roadside 1 Moderate Moderate Good
Aerial 1 Moderate Moderate Good Seed Rate
Table 4.12 summarises further observations on the success of the various

strategies and their limiting factors.

Table 4.12 Key Observations on Forage Strategies
Strategy General Comments Limiting Factors

Backyard Widespread adoption; useful point of Many areas too small to have
Forage entry for new species. impact on production.
Undersowing Widespread adoption with annuals; Continued government support
shifting towards being self- slowing shift towards self-
sustaining; some species now locally sustainability.
traded, and consumed as human food.
Oversowing Success dependent on site selection; Availability of perennial
some excellent persistence/spread legume seed at very low cost.
over ten years; better results below Suitable delivery systems for
2,400m large inaccessible sites
Aerial seeding of 1988/99 successful
on suitable sites, including for tree
legume establishment.
Stock Exclusion Very successful for control of Availability of perennial
Areas degradation and fodder production; legume seed at low cost;
excellent performance of introduced appropriate policy on
legumes in many sites; need to utilisation; local adoption of cut
reinforce all with legumes; need to and carry management.
allow regular cutting from outset.
Intercropping Some success with Desmodium under Some of the most suitable
coffee and citrus, insufficient species not yet available in the
emphasis to date; need to concentrate field.
on areas with livestock.
Hedgerows of Some excellent results with Inadequate awareness of
Leguminous Leucaena, Sesbania, tree lucerne; but benefits, inadequate stock
Trees scale often too small. Need to control in most farming
expand to include crop areas, and to systems; most Leucaena not
link to intensive utilisation for inoculated.
fattening and dairying. Utilisation
good in only some systems, but
improving in most areas.
Contour Forage Most destroyed with collapse of Local availability of suitable
Strips Producer Cooperatives; beginning to grasses and companion
regain momentum in some areas; legumes.
potentially very important and needs
to be stressed; needs to be linked to
intensive livestock enterprises.
Source: Alemayehu M., 1989

Activity Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Expected Rainfall
Monomodal
Bimodal
Extension
Identification of new farmers
Initial extension for new farmers
Extension for existing farmers
Farmers field days
Seedlings
Establishment in nurseries
Planting out
Strategies
Undersowing, belg season
Undersowing, main rains
Oversowing grazing areas
Sowing in exclusion areas
Mixed pasture establishment
Backyard forage establishment
Forage strip establishment
Seed Program
Harvesting most legume seed
Harvesting most grass seed
Seed cleaning, treatment
Seed packaging, labelling
Distribution to extension teams
Fig. 4.1 Forage Activity Calendar for Ethiopia


V. FORAGE SPECIES
Principles for Selection and Testing
The key principles for selecting improved forage species and their
cultivars focus on their ability to persist under normal management conditions
and produce large quantities of high quality forage. This means that the species
should tolerate grazing, and be able to flower and set seed under normal grazing
conditions. Suitable species will be drought tolerant in order to maximise
production in an environment characterised by a dry season. A mixture of species
should be selected for each AEZ to ensure biodiversity, and thus minimise the
risks from pests and climatic extremes. Species with different plant forms and
modes of reproduction should also be selected for each AEZ to maximise the
opportunities for integrating improved forages into different farming systems and
ecological niches. For example, tall growing species such as Seca stylo are
suitable for cut and carry systems associated with strategies for oversowing
natural grasslands. Similarly, sprawling vigorous legumes such as Siratro and
Greenleaf desmodium are suitable for undersowing and intercropping. Maku
Lotus and Halifa white clover are suitable for wet bottomlands in highland areas
and so on. Browse and other multipurpose leguminous tree species should be
introduced into every AEZ. Seed bearing species (such as stylo and tree lucerne)
should be mixed with vegetatively reproducing species (such as Rhodes grass or
hybrid Phalaris) to optimise ecological stability of introduced forage mixes.
When assessing growth rates and productivity, it is important to
understand the life cycle and growth habits of each species and cultivar. Stylos,
for example, are slow to become established but after two or three years are
highly productive. Similarly, tree lucerne and Leucaena often require more than
12 months to become established but are highly productive after this time.
Because the ultimate objective of forage production is to increase the quality of
livestock forage as well as the quantity, qualitative aspects of forages should also
be considered during selection and assessment of new forages. Palatability,
digestibility and nutrient balance should be measured. The occurrence of toxic
substances – for example indospicine in Indigofera spicata or mimosine in
Leucaena – should also be considered. Much of this information is readily
available so that the initial screening should be restricted to species with a
reasonable probability of being useful in each AEZ.
Because there is a considerable body of knowledge appropriate to forage
production and development in Ethiopia, it is economically and technically more
efficient to implement a wide range of strategies at farmer level without prior
adaptive research. Not only is there sufficient knowledge to proceed with some

confidence, but many of the problems which would affect adoption of the
strategies are related to socio-economic and institutional factors which are
unlikely to be resolved by adaptive research. The cost of seed is small compared
to the cost of extensive research and the economic cost of delayed
implementation of forage development programs. In addition, the accumulation
of local knowledge of the various strategies is much more rapid through
observation by farmers (the ultimate users of the program) and technicians over a
wide range of development sites. Finally, this farmer-centered approach is more
flexible and thus better able to make rapid modifications in response to changing
market, socio-economic and environmental conditions. The farmer centered
approach to development and implementation of forage production strategies also
provides better "hands-on" training for technicians, extension agents and farmers.
Some forage species, especially some grasses, may become rampant under
favourable conditions and should either be restricted in their use or excluded
from forage production strategies. For example stoloniferous grasses such as
Rhodes grass are highly suitable for pastures or stock exclusion areas but should
not be used for undersowing or on contour forage strips because of their potential
to become serious weeds.
Species with Proven Capability
Tables 5.1, 5.2 and 5.3 detail improved forage and browse species, which
have been successfully used under FLDP. Many of them have also been used
successfully in other parts of Africa and on other continents. Detailed
information on the characteristics and requirements of these species is in Annex
1.


VI. IMPROVED FORAGE UTILISATION STRATEGIES
Improved forage can be used to maintain more animals, to provide

increased quantities of better quality forage to the same number of animals, or to
strategically feed selected groups of animals. If increased forage production is
used to maintain more animals the incremental production is relatively small and
will depend on the additional number of animals. For example, a 25 per cent
increase in forage increases production by 25 per cent if the number of animals is
also increased by 25 per cent. If the number of animals is not increased, this same
amount of extra feed more than doubles production. This is illustrated in Fig. 6.1.
Strategic feeding to selected groups of animals is better still. Strategic feeding
means controlling what animals eat so that high quality forage is used where it
will generate most production. In Ethiopia livestock fattening and dairy
production and the most productive strategic feeding strategies.
Fig. 6.1 Impact of Different Forage Utilization Strategies

Source: AACM, 1989

Livestock Fattening
Livestock fattening is one of the best ways for farmers to quickly realise
returns on improved forage production investments. It is a strategy, which
significantly increases household incomes and provides a quick and tangible
incentive for further adoption of improved forage production strategies. The other
attraction of livestock fattening strategies is that they generate income without
using significant amounts of arable cropland. The amount of land used to enclose
animals at night is small and most of the forage production strategies do not
compete with crop production. Livestock fattening programs must include
improved forage production strategies as a precondition if they are to be
sustainable and successful.
It is necessary to estimate how much forage is available before deciding
how many animals to fatten. Visual estimation skills can be developed by cutting
and weighing samples and careful observation of animals performance at
different feeding rates. Initially, a conservative approach should be adopted since
it is better to successfully fatten a small number of animals than to run short of
feed and fatten none at all. Fattening programs are most efficient if they are
planned to coincide with the period of maximum forage growth. If market
conditions or other factors make it more attractive to fatten animals at other times
of the year, improved forage should be conserved as forage or browse hay.
Fattening programs should not begin until forage and browse supplies sufficient
to complete a fattening cycle are assured – normally after the second year of the
forage program.
Fattening programs should only use cut and carry feeding strategies and
should provide as much forage as animals can eat for 24 hours per day.
Underfeeding is the main cause of failure in fattening programs. This can be
overcome by ensuring that animals have feed available at all times, including
through the night. If there is no feed left in the morning, animals should be given
more on the following evening. Fattening programs are more efficient and
require less labour if the animals are fed close to the source of forage. Carrying
feed long distances from the field to the village is very labour intensive and
usually results in underfeeding. It is much better to tether animals near the source
of forage and give then cut forage in the field. This practice also helps return
nutrients to the field. Forage only needs to be brought back to the village for
night feeding. As a general rule, if animals are tethered or kraaled where forage is
being cut, one person can feed 7 to 10 oxen (up to 80 sheep or goats) during the
day and cut enough forage to bring back on donkeys for night feeding.

Animals should be fattened for a fattening cycle of between 90 and 120

days. In most parts of Ethiopia the period of assured feed surplus is only
adequate for a single fattening cycle. Fattening programs should start towards the
end of the main wet season – in September or early October – to coincide with
the time of maximum forage growth and the availability of crop residues. There
are a number of activities, which precede the commencement of fattening, and
these are outlined in Fig. 6.2, which shows a fattening activity calendar for
Ethiopia. A late start can mean that forage will run out or decline in quality. In
Ethiopia a late start can also mean than animals do not reach peak condition for
the Christmas market.
Activity Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Expected Rainfall
Monomodal
Bimodal
Fattening Program
Buy Livestock
Feeding Period
Sell Livestock
Extension Program
Identify/Select New Fatteners
Training
Supervision
Analysis of Results
Tours and Workshops
Credit Program
Prepare/Submit Applications
Appraise Applications
Loan Disbursement
Loan Repayment
Annual Review
Fig. 6.2 Fattening Activity Calendar

Animals being fattened should be fed as much leguminous forage as is

available and then be topped up with other feed. This ensures that the protein
content and digestibility of the ration will not constrain growth. Feeding too
much protein rich legume forage is less risky than feeding insufficient protein.
No supplements or by-products should be mixed with the ration unless they are
cheaply and locally available. As a general rule, a dry matter mix of one part
leguminous forage mixed with two parts crop residue or unimproved hay will be
a balanced fattening ration for ruminants. Cattle should receive about 10 kg of
this DM per day and sheep or goats about 1 kg.
It is important that animals being fattened have the opportunity to drink as
much fresh water as they want at least once each day. Inadequate or dirty water
will reduce feed intake and affect the performance of animals. As a general rule,
cattle require about 10–14 litres of fresh water each day and sheep or goats
require about 4-7 litres of fresh water each day. The exact amount will vary with
animal liveweight, climate, type of forage and other factors.
Animals in the fattening program should receive adequate disease and
internal parasite protection. Unhealthy or unproductive animals will not give a
good return to the scarce quality forage resources. Animals should be vaccinated
against the major endemic diseases (for example Rinderpest, Foot and Mouth
Disease, and Contagious Bovin Pleuro-Pneumonia). Animals should be drenched
with a broad spectrum anthelmintic at the beginning of the fattening cycle. This
is especially important for sheep fattening. Animals which are performing poorly
after 30 days should be culled and sold – even if this incurs a loss, since this is
better than wasting good forage on unhealthy or unproductive animals which are
not growing.
Selecting animals for fattening will depend on the availability of stock and
the potential markets for finished stock. Breeds with good growth characteristics
should be selected. Most lowland cattle of the Boran type are suitable. Animals
should have a large skeletal frame, which is capable of producing a heavy
carcass. Stock should be docile and easy to handle and should not be too old –
their teeth should be sound. Steers are preferable to bulls because they grow
faster and are more docile. The most profitable animals to purchase for fattening
are those, which are lean with a faintly visible backbone and ribs, visible hip
bones, and a slightly recessed tail head. Thinner animals normally take too long
to recover and fatter animals usually have a higher price per kilo, which limits the
potential for increased values. If possible, determine why animals are lean.
Animals from an area where feed is short or which have just finished the
ploughing season are likely to gain weight rapidly.
Individual households generally only have enough forage to fatten one ox.
If there is not enough forage for this, or if the household considers the risk too
great, 2 to 5 sheep or goats should be considered. Groups of households may
decide to pool their forage resources to fatten a group of animals. This makes
more efficient use of labour but requires careful attention to animal health at the
beginning of the fattening cycle to avoid transmission of diseases or internal
parasites amongst the fattening group.
It is important that accurate records be kept of each fattening cycle to
determine whether each cycle is profitable and to provide information for
planning subsequent cycles. Records should be kept by farmers using self-
monitoring techniques, and extension agents. Local units of forage weight or
volume should be used. Animal performance is best measured with girth tapes
used to estimate weight gain and growth. Visual appraisal and hand testing along
the back and rear of the animals will also give a subjective appraisal of animal
performance.

Credit can be used to help farmers without cash resources to purchase

animals for fattening. Fattening loans should only be used where needed and
should not be forced on reluctant borrowers. Loan applications and approvals
should be processed in time to allow purchase of animals for fattening before the
end of the main wet season. The loans repayment period should be over one year.
This will avoid the need to reschedule if there are poor results in one year, and
will help maintain a positive cash flow throughout. Insurance of animals should
be a compulsory part of any fattening loan. Loans should not be made available
for construction of animal housing. This will help discourage over-investment in
unnecessary facilities and encourage the use of local building materials. For
forage-based fattening loans, money should not be made available for purchased
feedstuffs or concentrates. Rations should be 100 per cent forage except where
by-products are cheaply and locally available as and alternative to crop residues.
Dairy Production
Dairy products are an important source of human food and income for
many rural communities in Ethiopia. Where sedentary communities raise dairy
animals there are significant opportunities to increase production by increasing
the quantity and quality of forage available to lactating animals. Wherever
possible, breeding cycles should be timed to ensure that lactations begin at the
time of maximum forage availability – near the end of the main wet season.
Improved dairy breeding and artificial insemination programs should be
conditionally tied to improved forage production strategies. This is the best way
to ensure sustainable production from these investments as well as to maximise
the benefits from them. Most of the principals outlined for fattening programs
also apply to dairy production programs.
Nutrition is a more significant constraint to sustaining milk production in
Ethiopia than is the genetic potential of dairy animals. Natural pasture would
rarely support milk yields of more than 3 to 4 kg per cow per day but improved
forages, such as those outlined in the strategies in Chapter IV, can sustain yields
of up to 8 or 10 kg under good feeding and management conditions (Walshe et al,
1991). Because milk is a high energy, high protein product lactating animals
benefit from the addition of by-products or concentrates in their ration. Suitable
by-products or concentrates should have high digestibility, high energy
(carbohydrate) and proteins levels and low fiber content. Most high protein (for
example soybean or fish meal) and high energy (for example maize or barley)
concentrates are expensive, especially where dairy producers compete for them
with poultry producers. Low protein (for example noug or cottonseed cake) and
low energy (for example oats or bran) concentrates are suitable for dairy animals
and are more cost effective than the expensive concentrates. Note that leaf hay
(dried leaves) from browse species such as Leucaena, pigeon pea, Sesbania and

tree lucerne have high energy and high protein levels and are suitable
concentrates for dairy animals if there is sufficient available to mix with other
feeds. A backyard browse plot or hedge will normally produce enough leaf hay
for one dairy cow and thus has a significant impact on household dairy
production.
A large amount of feed is required to maintain dairy animals, but the share
of total feed needed for maintenance declines as the milk yield increases. For
example, a cow producing 500 kg of milk pear year needs about 2.5 feed units
per kg of milk, while the requirement drops to about 0.5 feed units per kg for a
cow producing 5,000 kg per year (Walshe et al, 1991). This emphasises the
importance of improved quantity and quality of forage for efficient dairy
prediction. Because dairy animals typically have a lactation period of more than
the 90 to 120 days required for fattening cycle, forage needs to be conserved to
provide high quality rations throughout the lactation period and the dry season
when forage is scarce.
Forage Conservation
Forage conservation helps to bridge the quantity gap between livestock

feed requirements and the production of forage. If good quality forage is
conserved, the nutritional gap between high quality (wet season) and low quality
(dry season) forage may also be bridged. Hay produced from natural grasses,
improved forage legumes and browse legumes is the most appropriate conserved
forage for small-scale fattening or dairy production in Ethiopia. Forage
conservation is especially important for dairy production because it ensures a
supply of balanced nutrients for dairy animals throughout their lactation.
A mixture of grass and legume forage should be harvested for quality hay.
The legumes increase the digestibility and intake of the conserved forage. If
mixed grass-legume forage is not available for hay making dried legume forage
can be mixed with grass hay in a dry matter ratio of about 1:2 to provide a
balanced ration for ruminants. For high quality hay, forage should be cut just
before grasses flower. This will produce good yields of hay with high
digestibility and protein content. Although cutting hay after plants have flowered
produces more hay, it is of lower quality than earlier cut hay. Quality hay is
typically cut 4 to 6 weeks after the pasture or forage crop has been closed to
grazing.
Quality hay is cut before flowering of the grasses and ideally between
rainy periods. Cut hay should be dried as quickly as possible and should be
turned regularly during drying to keep the hay mould free. Once the hay is dry, it
should be stacked using the "heap-up" methods. Haystacks should be constructed
so that they are tall and thin – with the smallest possible surface area at the top to

increase their resistance to rain. Flat topped and broad haystacks are not
weatherproof – the hay will be damaged by rain.
When conserving crop residues such as straw, layers of leguminous forage
or browse hay should be sandwiched between the layers of crop residue. This
increases the feeding value of the crop residues and provides a balanced livestock
diet. Alternatively, separate haystacks can be made from crop residues and forage
and browse legumes.


VII. FORAGE SEED PRODUCTION AND DISTRIBUTION
The production of forage and browse to overcome dry season livestock

feed shortages is dependent on the availability of reliable supplies of quality seed
at the time of planting. Forage development programs need to include local seed
production to ensure their long-term sustainability and economic viability.
Because of the wide range of agro-ecological zones and farming systems in
Ethiopia, many species are required, but local production of seed for the key
species can be initiated very early in the forage program. The principal objective
of a local seed program should be to meet the forage and browse seed needs of a
forage development program.
Importation of Initial Seed
The first years of an improved forage production program will normally

need to be established with imported seed. The cost of seed is small compared to
the benefits of rapid and widespread implementation of improved forage
production. Small quantities of a very wide range of species and cultivars should
be imported along with larger quantities of proven species or cultivars to initiate
forage programs. Regional trials should be conducted with untried species and
cultivars to assess their suitability for wider use. Wherever possible these trials
should be conducted in farmers' fields but careful planning and supervision of the
trials is important to protect them from grazing and to ensure than small
quantities of seed are effectively used.
International competitive bidding tenders normally result in the best value
for money from imported seed. Australia is the largest producer of tropical forage
seeds and has a well-organised and competitive seed exporting industry. Tenders
should be called in time to receive and distribute seed before the sowing season
and should also recognise seasonal aspects of supply and delivery times. Most
seed can be airfreighted at reasonable cost but major shipments would normally
be seafreighted. Delivery schedules normally require 3 to 6 months. Tenders
should be prepared and called for international competitive bidding at least 7
months before the sowing season. This allows 1 month for bidding, 1 month for
award of the tender and contract finalisation, 3 months for delivery of seed, and 1
month for distribution of farmers. This tight schedule assumes that there will be
no delays and that the forage program coordinator has good control of tender
award and contract approval, customs clearance, and distribution activities. If
delays are anticipated, additional time should be allowed.
Imported seed should be certified true to type and of minimum purity, and
be tested for germination prior to shipment. Table 7.1 gives some germination
and purity standards for a range of species. Imported seed should be supplied in

sealed, weatherproof sacks or small cloth bags with printed labels on each sack or
bag identifying the species, cultivar, source, date of harvest, and recommended
storage conditions. Seed treatments are not normally required for forage seed but
inoculants and gum adhesives should be included for leguminous species, which
are not naturalised in the forage program area. Inoculant requirements are listed
in Annex 3.
Importation of forage and browse seed should not be necessary after the
start-up phase of a forage program – typically three years. self-sufficiency in
major forage species and cultivars requires a planned seed production program
which is farmer based and market oriented. The basic objective of a successful
forage seed production program should be low cost seed production by farmers
based on contracts, which do not compete with subsistence food production. This
approach has proved successful in Ethiopia where forage seed production is now
treated as a cash crop by small farmers.
Contract Seed Production
The most successful method of producing forage and browse seeds in

Ethiopia has been to contract farmers to grow or collect seed. Contract seed
production involves establishing a contractual agreement between a farmer and
the seed purchaser – usually the Ministry of Agriculture, but sometimes a seed
trader. The seed contract is a legally binding agreement between the purchaser (a
project or Ministry or trader) and the farmer or a group of farmers. Both the
purchaser and the producer must make certain commitments under the seed
contract.
The Purchaser Must: The Producer Must:

• Provide seed for initial sowing. • Produce seed to an agreed quality.
• Provide close supervision and technical backup • Sow, manage and harvest the crop.
for the seed plots. • Clean the seed after harvest and
• Purchase the seed for cash at an agreed price for deliver it at a specified time.
certain quality at a specified time.
Contract prices are based on estimated yield, production costs, the market
for seed (or program requirements), and the cost of imported seed. Local seed
prices will normally be significantly less than imported seed prices because they
exclude shipping or airfreight costs. As an example, contract seed prices paid to
Ethiopian farmers for seed produced in 1990/91 were typically one third to one
half of the import price. Contracts must be arranged well before the crop is grown
and should detail: which species and cultivar is to be grown; who will supply the
parent seed and when it will be delivered; technical supervision during planting,

growing and harvesting; minimum quality standards for produced seed; and the
price paid for a given quantity and quality of seed.
Before contracting seed growers, the contracting agency (buyer) should
conduct group meetings with farmers to ensure that they fully understand the
technical and contractual aspects of seed production. Although it is desirable to
spread the program over a wide range of geographical locations and agro-
ecological zones, it is important to cluster production sites to facilitate
supervision and seed collection. The area of land needed to produce the quantities
of pasture seed required for most forage development schemes is very small
compared to the total cropping land available. Thus pasture seed production will
not compete with food production. Apart from contract seed production, farmers
can also opportunistically collect seed from browse species and other species
growing in stock exclusion areas, forage strips, and undersowing sites.
Opportunistic seed collection can also be contracted.
Site Selection: The most suitable regions for forage and browse seed
production have: an adequate growing season to support good seed set and
maturation; freedom from frost; even, sunny conditions during flowering to
promote flower opening, pollination and high rates of photosynthesis during seed
differentiation; access to labour for harvesting and seed cleaning; and access to
markets and seed storage infrastructure. Weed free areas or areas that have a
history of reasonably clean cropping are preferable to weedy areas to minimise
the problems of weed competition in the seed crop.
Seedbed Preparation: Seed crops need to be established in a clean, fine
and firm seedbed with sufficient seed to ensure a strong, dense plant population,
which will compete with weeds and maximise yields. Ethiopian experience
suggests that the maximum size of a seed crop for small farmers is about 0.25 ha
(FLDP, 1991). Legumes seeds need to be treated to soften hard seeds, which will
not germinate without treatment. Browse legumes and forage legumes with less
than 500,000 seeds per kg should be treated. The simplest say is to boil water in a
tin, remove the tin of boiling water from the fire, and immerse a cloth bag
containing the seed in the hot water for about 10 minutes. The treated seed
should then be rapidly cooled by spreading it out in a thin layer. Stylos are
sensitive to heat so they should only be immersed for 3 minutes. Where hot water
treatment of seed is impractical, scarification is suitable alternative. The simplest
way to scarify or scratch the seed coat is to combine some seeds with sand or
gravel and thoroughly mix them together so that the gravel or sand scratches the
seed. This will normally take at least 15 minutes.
Sowing: When using cultivars or species new to an area, legume seed
should be inoculated with appropriate rhizobia (see Annex 3) to ensure that they
fix nitrogen. Seed should be sown as soon as possible after the beginning of the
main rainy season. This reduces the risks of crop failure. Small forage seeds
(more than 20,000 seeds/kg) are broadcast onto the seedbed and raked in. Larger

seeds are either sown in rows or buried, with a light cultivation, no more than 3
cm deep after broadcasting. Sowing in rows reduces the quantity of seed required
and makes weed management and harvesting simpler. Row spacings should be
50 cm for small seeded forage legumes, vetch, stylos, desmodiums, and grasses.
1 m row spacing should be used for siratro, axillaris, and cowpea. Browse
legumes being grown for seed should be planted in rows 3 m apart. Sprawling
legumes such as axillaris, desmodiums, and siratro produce much higher seed
yields if seed crops are grown on trellises, browse trees, or along fencelines. Seed
crop sowing rates are suggested in Table 7.1. Some plants reproduce without
seed – from sets or cuttings. Several grasses are more effectively reproduced this
way – for example hybrid Phalaris and these may be grown by farmers and sold
on a contract basis of so much per rooted cutting or set.
Management: The overall aim of seed crop management is to
consistently succeed in producing a seed crop with not only a high yield of
quality seed, but also a crop, which allows efficient seed harvesting. This
essentially means having a crop of uniform age and is best achieved by:
establishing an adequate, uniform plant population; developing a dense cover to
exclude weeds, encouraging flowering at the same time; and ensuring that
flowers produce mature seeds. Regular crop inspections are important to control
weed and pest populations. Weeds should be hoed or pulled by hand. Weeds not
only compete with the seed crop but they also increase the risks of contaminating
forage seed with weed seed – something which increases the work required for
effective seed cleaning. Advice on pest management should be sought from local
Plant Protection Specialists or Extension Advisers.
Seed Harvesting, Cleaning and Storage
Harvesting: Most tropical legumes flower and set seed over a long
period and frequently shed seed quickly. This makes it very difficult to judge
when to harvest seed. Techniques used to judge ripeness include testing for ease
of seed removal; seed hardness; and field colour. When most seed can be easily
removed by gentle rubbing or shaking, then seed is normally close to shedding
and should be harvested. If seed rubbed in the palm of the hand is hard and dry
then it is mature and ready to harvest. Grass seed should be bitten to ensure that it
is full. The seed or pods of some species, for example Siratro, Rhodes Grass, tree
lucerne and Leucaena, change colour as they ripen. The optimum harvest time
usually occurs before maximum flower density occurs. Hand harvesting of
tropical pasture seed, particularly if labour is experienced and well supervised,
can lead to high yields of good quality seed. Hand harvested yields are generally
higher than yields from mechanical harvesting. As an alternative to hand picking,
mature seed of both grasses and legumes can be removed from the plant by
shaking it into a basket or bag. Hand picking and shaking two or three times per

week will maximise seed yields and farmer income. Small seeds can be collected
from threshing areas by sweeping. This technique is particularly suitable for
stylos, Wynn Cassia, and Axillaris.
Table 7.1: Seed Production Characteristics for Selected Species
Species Min % Purity Seeds /kg Sowing a/ Seed Yield

Germination % ‘000 Rate kg/ha kg/ha
Forage Legumes
Alfalfa 60 95 400 15 200-600
Axillaris 60 95 120 3-5 150-500
Wynn Cassia 60 90 250 2-4 150-500
White Clover 60 90 1,500 2-3 100-400
Cow Pea 75 98 10 10-20 500-800
Greenleaf 60 90 750 3-4 100-400
Silverleaf 60 90 210 4-5 100-400
Siratro 60 95 75 3-5 200-950
Seca Stylo 40 90 425 5-10 200-750
Verano Stylo 40 90 400 5 150-750
Cook Stylo 40 90 300 5 300
Maku Lotus 60 90 2,200 1 100-400
Vetch 75 98 14 15-20 700-2,500
Browse Legumes
Leucaena 60 98 24 4 5-10 a/
Pigeon Pea 60 98 20 5 1-5 a/
Sesbania 60 95 100 3-4 5-10 a/
Tree Lucerne 60 98 60 3-4 5-10 a/
a/ for seed crops b/ Seed yield range per tree
Source: FLDP, 1991 and O’Reilly, 1987
Further observations on seed production are summarised in Table 7.2.

Table 7.2: Observations on Forage Seed Production on Government Farms

and Smallholder Contract Schemes
Key Indicator Government Farms Smallholder Contract

Yield per area Variable High
Cost per unit of seed High Low
Predictability of annual Low Very high
production
Relaibility of supply Variable Very high once
established
Capital investment required Variable – high Negligible
Organisation required Continually high Initially high
Suitable species groups Grasses All annual and
perennial legumes
Requirements for success Continual funding, Technical supervision,
supervision, organisation of reliable market, timely
labour purchase
Cleaning and Drying: Legume seed should be dried as soon as possible

after harvest to achieve a seed moisture content of 8 to 10 per cent. This ensures
good seed viability. Seed can be sun dried without damage to the seed. Grass
seeds should be heaped immediately harvest so that they will "sweat" to assist
final maturation of the seed. Grass seed is more sensitive than legume seed and
should be dried slowly to maintain its viability. Sun drying is not recommended
because of this but grass seed can be dried in the shade. All drying seed should
be turned regularly – at least once per day – to ensure efficient drying. Dried seed
is then threshed using animals, a mortar and pestle, or beating with sticks or
flails. Regular inspection of the seed is essential to avoid damage to the seed.
Threshed seed is then cleaned to remove seeds of contaminant species, soil, chaff
and poor seeds. Winnowing and sieving are the normal means of cleaning seed.
Most farmers in Ethiopia are skilled at manual seed cleaning of both coarse a fine
seed, for example maize and teff.
Storage and Labeling: Once cleaned, harvested seed must be stored in a
cool, dry place. The length of life of a seed in storage depends on the
environment in the seed store. For storage beyond 5 years seed should be kept at
temperatures below 150C with relative humidity below 4 per cent. An indication
of seed store suitability can be calculated by adding the average temperature ( 0C)
and the average relative humidity (%) from the store. This is the storage index.
For example, a seed store with an average temperature of 200C and average
relative humidity of 45 per cent has a storage index of 20 + 45 = 65. Short term
storage (>6 months) requires a storage index of less than 80. Medium term
storage (6-18 months) requires a storage index of less than 70 and long term
storage (up to 5 years) requires a storage index of less than 50. Grass seeds
should be stored in a sealed container but cotton sacks or woven bags are suitable
for most legume seeds. Seed stores should be well ventilated, secure, and clean –
especially free from insects and rodents. Each seed container must be labelled
with information on: the species; the cultivar; the date of harvest; the location of
harvest; the weight of seed in the container; and any seed treatments (for example
scarification or insecticide dressing).
Seed Distribution
Distribution of seed must be planned in advance of the next cropping

season when improved forage will be planted according to one of the strategies
outlined in Chapter IV. This is best achieved with an annual forage plan, which is
developed by farmers and extension agents. This "bottom-up" participatory
element is essential for successful extension and management of forage programs
and is discussed further in Chapter VIII. The annual forage plan for the coming
season will ideally have contributed to determining how much forage seed was
contracted to be grown in the previous season. The coordination required for
successful forage program planning requires time and skill and is best done on a
Regional or National level. This coordination requires an accurate knowledge of
existing seed supplies so that any shortfalls can either be imported or taken into
account by modifying the forage program for the coming season. Once regional
forage program targets have been established, seed mixes and quantities are
prepared for each administrative region. This may involve moving seed from one
Region to another to ensure that each region has sufficient seed to meet farmer
demand or forage program targets.
At the regional or local level, seed is split and, where necessary, mixed
into seed packs for use by farmers. Seed packs typically have sufficient seed to
establish a small nursery of browse species for backyard strategies, enough seed
for undersowing 0.25 ha of crop, or enough seed to sow 250 m of contour forage
strips. Strategies such as oversowing or improvement of stock exclusion areas for
communal areas would have larger seed packs, which would be used by grazing
management groups or pastoral associations. Forage programs are most
successful where seed is inoculated and scarified at a Regional level before being
split into farmer seed packs. This helps ensure success during establishment.
Seed packs should be clearly labelled and include sowing and early management
instructions.


VIII. REQUIREMENTS FOR SUCCESS
Improved forage production program programs must be adapted to the

economic, social and environmental conditions in each region where they are to
be implemented. However, there are some fundamental requirements for success,
which should be included in programs. These include:
• active farmer participation at all stages;
• strong institutional support;
• a planning approach to implementation;
• motivated technicians located throughout the project area to extend forage strategies
and conduct adaptive research;
• initial importation of improved forage seed;
• financial mechanisms to enable contract seed growing;
• secure land tenure for farmers; and
• monitoring and evaluation
Farmer Participation at all Stages
Successful improved forage production programs involve farmers from

the very beginning. By conducting diagnostic survey or problem census meetings
during the identification and preparation phases of the program, it can be
designed to solve key problems identified by farmers. This ensures high adoption
rates and increases the likelihood of improved forage strategies developed for
each region being appropriate to local circumstances. For example, farmers
identifying fuelwood and shelter as priority problems are more likely to adopt
strategies involving browse legumes than undersowing strategies. Similarly, a
village or group of individuals, which collectively give priority to land
degradation of common lands are more likely to orgnise themselves into grazing
management groups or pastoral associations to implement oversowing and stock
exclusion strategies than farmers who place priority on food security issues.
Problem census and diagnostic survey techniques, especially where they are
farmer led rather than technician pushed, provide a clear message about which
strategies are likely to be most successful for each group farmers. Strategies
which solve priority problems provide the most successful starting point for an
improved forage production program which will gradually evolve to cover the
whole spectrum of strategies - backyard forage, integration of forage and crop
production, and enrichment of common lands.
Implementation of a program identified and prepared with farmers is
much more simple than initiating a program that is completely new to them.
Farmers must participate not only in the implementation of each strategy –
obviously on their own land and in their own backyards or compounds, but also
on common land – but they must also contribute to the planning and coordination
of implementation. This builds up improved forage management skills, which

will help ensure the program’s sustainability at the end of the initiation phase.
For example, farmers should be consulted about their forage seeding plans for the
forthcoming season and helped to convert this information into anticipated seed
requirements. This forms the basis of annual forage production plans discussed
below. Monitoring information should be shared widely with participating
farmers – to help them feel that they are part of a wider effort and to give them
technical information about the experiences of others which may be of use to
them. Information sharing can be achieved through simple newsletters or village
posters, or through regional demonstrations and meetings, which bring farmers
together. The latter approach is more expensive and difficult where transport
problems may exist.
Farmers should be integrally involved with adaptive research, especially
where it is conducted on-farm. In this way research ideas are constantly being
validated by the ultimate users of the technology being trilled or developed. On-
farm research is the most appropriate way to assess new forage cultivars and
develop the most appropriate forage production and utilisation techniques. Key
or contact farmers should be used for transferring improved forage production
technology to other farmers – either through farmer-to-farmer contact or through
formal field days and demonstrations. Finally, as detailed below, farmers should
conduct the monitoring and some of the evaluation tasks associated with
improved forage production programs.
Strong Institutional Support
Effective farmer participation is enhanced by strong institutional support

at all levels. Central government support is required for coordinating the flow of
funds for each forage production plan and to budget for future activities, as well
as overall coordination of resource allocation to the program. This includes
allocation of seeds, field technicians working with farmers, vehicles, adaptive
research activities, and training. Regional government support is required to
coordinate local activities, aggregate locally produced annual forage production
plans, and support local technicians with subject matter specialists in areas such
as forage production and forage utilisation. Local offices effectively act as the
liaison point between farmers and central government - they are the crossroads
between "bottom-up" activities focusing on farmer participation and "top-down"
activities, which coordinate and allocate resources made available by
government. Local offices should be staffed by motivated and active technicians
who work with farmers to extend each strategy, modify technical packages with
on-farm adaptive research, and facilitate the preparation and implementation of
annual forage production plans. Improved forage production programs are

normally staffed from existing resources rather than adding new staff at the
regional and local level.
Adequate financial and infrastructure support is important for the
successful implementation of forage programs. Transport is especially important
for participatory programs. Small 4WD vehicles or motorbikes, where dogs and
wildlife are not a problem, should be available to each local technician to
encourage them to work in the field and to motivate their activity. Regional
coordinators and subject matter specialists also need access to vehicles for
periodic visits to local areas, demonstrations and farmer group meetings. Access
to telephone or radio communications between regional and central offices is
essential for effective and timely coordination of project activities – especially
during the preparation of annual plans. Regional offices should have a modest
but up-to-date technical library to which local technicians and farmers have
access. This is most easily organised through a library of original journals and
books kept at central level from which appropriate papers and chapters are copied
to each region for inclusion in a series of subject matter file.
Central government staff coordinating improved forage programs need
access to word processing resources and printing or copying facilities for
preparation of posters, newsletters, extension bulletins and technical briefing
papers. A photocopier is invaluable for facilitating the flow of information
between subject matter specialists and local technicians, and between farmers.
Computers are needed at central and ideally regional level to coordinate budgets,
organise and keep track of seed distribution, and analyse monitoring and research
data. Training in the use of word processing, spreadsheet, and data analysis soft
ware needs to be included with all computer purchases.
Central and regional administrators need to organise and monitor farmer
and local technician training. This will sometimes involve subject matter
specialists but could also include international training courses. Because of the
many components required for successful production of improved forage, large
programs are normally most successful if the program coordinator is supported
by special coordinators for key areas – seed, forage production, forage utilisation
and training. This central program organisation can be duplicated at regional
level if the program is sufficiently large. Otherwise a single regional coordinator
can work with the special coordinators at central level and the local technicians at
village level.
A Planning Approach to Implementation
Successful implementation of forage production programs is enhanced by

basing implementation activities on annual plans, which are developed at a local
level by farmers with support from local technicians. Annual plans are used as an
integral part of farming management and provide a focus for liaison between

farmers and local technicians and regional subject matter specialists. The timing
of forward planning, seed requirement calculations, seed ordering and
distribution, and sowing are all developed with farmers during the annual
planning process. Annual plans developed by each farmer or household are
aggregated into village annual plans by self-organised farmer groups - for
example pastoral associations and grazing management groups. Village annual
plans are aggregated at regional level to make regional forage production plans,
and finally pooled together at central level to prepare a national forage
production plan for that particular year. Fig. 8.1 provides an indicative annual
planning cycle for a forage program.
Village contact Demonstrat- Village

and base data -ions and problem
collection options menu census
Village plans Annual village Problem solv-

aggregated into forage plans '-ing & house-
Regional plan within budget -hold plans
Regional plans
reviewed and Implementation Monitoring
approved
Fig. 8.1 Annual Planning Cycle for Forage Program
Motivated Technicians Located throughout the Project Area
Motivated technicians need to be locally based to extend forage strategies

and conduct adaptive research with farmers. The local technician ultimately
determines whether the role of the executing institution in the forage program is
successful. They must facilitate diagnostic surveys or problem census activities
with farmers, conduct initial on-farm demonstrations and promote a widespread
understanding of forage utilisation. Once the program is being implemented,
local technicians should initiate demonstrations, which will have an early visual
impact. This provides an important incentive for wider participation in the forage
production program by farmers. A lot of seed and seedlings should be made
available during the early years of he project to have a saturation effect on each
locality. Motivated technicians will make good use of this seed, and its
availability helps reinforce their position within the local community, which in
turn strengthens their motivation.

On-farm adaptive research is needed to assess new cultivars and modify

the technical packages to suit local social and environmental conditions.
Adaptive research can also be used to emphasise certain benefits which farmers
perceive with the program or to emphasise solutions to priority problems
identified by farmers. For example, many farmers appreciate the shelter and
privacy uses of browse species so demonstrations and adaptive research can be
oriented towards optimising these attributes – browse hedges can be
demonstrated and optimal in-row spacing, cutting height and cutting frequency
can be the subject of adaptive research. This approach to forage production helps
farmers develop an understanding of the wide range of direct and indirect
benefits of improved forage. By ensuring that farmers participate at all levels,
benefits, which they perceive but which are not apparent to technicians will not
be overlooked.
Local technicians should also work with farmers to develop their
understanding of farm feed budgeting and feeding priorities. This should be
included in forage utilisation activities. The value of improved forage needs to be
demonstrated to farmers in relation to the normally more expensive agro-
industrial by-products.
Integration with Other Activities
Improving the quantity and quality of forage is central to increased

livestock productivity in Ethiopia. Without adequate forage farmers cannot
realise the full benefits from animal health and breed improvement programs.
Improved forage production programs should therefore be fully integrated with
other livestock development activities. Most importantly, improved forage
production should be a pre-requisite for participation in other programs. This not
only acts as an incentive for wider adoption of forage production strategies but
also increases the benefits from other, usually more expensive, programs. For
examples, farmers seeking Government-provided animal health services should
first be required to have adequate and balanced forage available for their
livestock. Similarly, farmers wishing to participate in artificial insemination or
improved heifer programs should also demonstrate that they have a sustainable
supply of balanced forage for their animals. This integration also ensures that all
technicians and administrators involved with national, regional and local
livestock production include sustainable forage production in their strategies.
Initial Importation of Improved Forage Seed
As detailed in Chapter VII, improved forage programs must commence

with imported seed of cultivars for which there is expected to be strong demand
from farmers, as well as cultivars to be assessed in on-farm trials. Programs

identified and prepared with farmers will have a much clearer idea of the quantity
and cultivars of seed, which are needed to commence implementation. Because
imported seed is cheap relative to other project costs, adequate quantities of
forage seed should be imported as early as possible during project
implementation. This enables local technicians to saturate participating farmers
with seed and so establish an early visual impact. This is extremely important for
reinforcement of the forage program and its extension to a wider community of
participants.
Financing Mechanisms
Institutions executing improved forage production programs need to have

access to project or other funds for local seed purchases. Unless simple
administrative procedures exist to distribute and account for funds required for
seed purchases, it is very difficult and time consuming to enter into forage seed
production contracts with farmers. Without such seed production, forage
programs are unlikely to be sustainable, unless a private sector seed industry
exists. Where private seed traders and producers exist, it may be more
administratively simple for executing agencies to issue selected companies with a
head contract to produce certain quantities of certified seed. These companies can
then sub-contract production of this seed to farmers. Financial mechanisms for
other program procurement should be within the control of the program
coordinator at central government level. This reduces the delays in procurement
of inputs for forage programs and enables more accurate preparation of annual
plans and their budgets.
Secure Land Tenure
Security of land tenure is one of the issues normally raised by farmers

during problem census and diagnostic survey activities for improved forage
programs. Land tenure issues have wide political, economic and social
implications, many of which are outside the objectives of improved forage
programs. However, security of land tenure is a very powerful incentive for
adoption of improved forage strategies – especially where they involve perennial
browse and grass species. Freehold title provides the best incentive for individual
farmers to adopt improved forage strategies on their cropping land. This is
especially important for long term strategies such as contour forage strips and
agroforestry.
Where freehold title is not possible, leasehold title also provides and
incentive for forage production. Leasehold titles guarantee tenurial lease from the
Government for a given period of time – typically up to 45 years – but may have
management convenants attached. If the covenants are not observed the

leasehold is forfeited to the Government. Leasehold titles are especially useful

for common lands to be rehabilitated with oversowing strategies or to be
managed as stock exclusion areas. Secure leasehold tenure acts as an incentive
for the formation of grazing management groups or pastoral associations, and the
lease is normally established in the name of the group or association. This not
only give farmers or pastoralists a reason for forming these groups or
associations but provides a mechanism for peer group pressure to regulate
grazing management and implementation of improved forage strategies on
common land.
One strategy, which is normally very successful, is to use grazing land
rehabilitation contracts as a vehicle for implementing oversowing, agroforestry,
or stock exclusion strategies on common land. These contracts provide grazing
management groups or pastoral associations with leasehold title to certain
common land. The leasehold title has land rehabilitation, improved forage
production and grazing restrictions attached to it. These would normally include
oversowing and stock exclusion strategies, provision for cut and carry harvesting
of forage and browse at certain time of the year and a requirement for regular
monitoring of plant cover and soil erosion. The leasehold title should be for a
period of at least 10 years and should be strictly monitored by a regional
administrator with support from local technicians. This approach works
successfully in Asia and Turkey and should be included in problem solving
sessions with farmers during the identification and preparation of improved
forage programs.
Monitoring and Evaluation
The impact and sustainability of improved forage production programs

will depend on the existence of a database of information, which can be used to
modify technical packages and demonstrate the benefits of each strategy being
promoted by the program. The information required for this database can mostly
be collected by farmers and local technicians and should include:
• area planted to each species and cultivar;
• DM production from each cultivar under different management strategies;
• local climatic data – especially rainfall, maximum and minimum temperature and frost
incidence;
• livestock numbers by species and type (male, female, castrate etc.);
• purchase price and weight for animals to be fattened;
• daily rations fed to animals being fattened;
• selling price and weight for fattened animals;
• daily rations fed to dairy animals;
• daily and total milk yield for dairy animals;
• veterinary expenses; and
• labour inputs for improved forage production and utilisation.
This type of information can be conveniently collected by self-monitoring

or auto-recording strategies in which farmers enter information in simple daily
livestock record books designed for the purpose. As with other activities the
livestock monitoring record book should be designed with farmers. It must be in
their local language and should be based on their technical terms for the
parameters being measured. This local technical lexicon is translated into
international terms at the regional or central office responsible for recording and
analysing monitoring data. Thus in Ethiopia, for example, timad and quintal
might be used rather than hectare or kilogram, with the appropriate conversion
factors being applied by technicians analysing the data. Separate monitoring
books would normally be developed with farmers for forage production,
livestock fattening and dairy production. The use of auto-recording strategies for
livestock monitoring has been successfully used in several projects, including the
Second Livestock development Project in Mauritania. The principal benefits of
auto-recording monitoring are that data reflects socio-economic aspects of
farming systems, which cannot normally be record by outsiders. Such data not
only enables production to be monitored but it also is able to compare the
productivity of different management and socio-economic systems. This
information becomes very important for adapting technical packages to different
socio-economic and environmental circumstances.
Forage production monitoring can be adequately done with simple
sheets, which are completed on a weekly basis. This forms the basis for
evaluating the implementation of annual forage production plans and the
preparation and modification of subsequent annual forage production plans.
Harvesting data should be entered from the weekly summary sheets in
monitoring books used for livestock fattening or dairy production. The weekly
monitoring sheets are summarised onto monthly forage production sheets by the
local technician working with the farmer. Local technicians take copies of this
monthly summary for collation and analysis at regional or central level.
Analytical results are later returned to the local technician for discussion with
farmers. This helps farmers see how they are performing relative to their peers. It
also enables the local technician to not only target people in most need of help
but also identify the most successful management strategies for integration into
subsequent demonstration and extension activities. Auto-recording forage
production monitoring forms include:
• units(s) of production (area, length of strips, number of browse trees etc.);
• soil type;
• drainage score;
• farming system into which forage has been integrated;
• inputs (fertiliser, dung, other);
• labour inputs (weeding, harvesting, etc.);

• wood production by species;

• forage production by species; and
• seed production by species.
Units and terminology used will be those chosen by the farmers during
design of the forage production monitoring sheets. Local technicians should
review this data when they collect the monthly summary sheets of the above
information to which they would include:
• strategy name(s);
• success score relative to neighbouring farmers;
• primary forage use;
• secondary forage use; and
• local climatic data (rainfall, maximum and minimum temperature, frost).
Livestock fattening monitoring: by farmers is best developed as a herd
book which records performance of individual animals. Daily records should be
kept for each animal (identified by ear tag, name or distinguishing mark)
throughout the fattening cycle (see Chapter VI). Experience in Mauritania
suggests that were farmers design the livestock monitoring books they actively
collect data and take a keen interest in its analysis. Base data should include:
• animal identity and date of purchase;
• breed, sex, and age of animal;
• purchase price and weight (estimated by girth measurement);
• sale price and weight (estimated by girth measurement); and
• sale date.
Dairy production monitoring follows similar principles to monitoring of
livestock fattening but will take place throughout the year to take account of the
lactation period and the management during pregnancy and preparation for the
subsequent lactation. Base data should include the identity, breed, and age of
each animal being monitored. Most farmers only have one or two milking
animals so auto-recording monitoring is not an onerous task.
Daily livestock fattening and dairy production monitoring data is aimed at
identifying the production impact of different management strategies and should
include for both types of monitoring:
• time animal was housed;
• time animal was grazing outside;
• time animal was tethered and fed outside;
• quantity (local measure) of cut and carry forage provided;
• source of cut and carry forage;
• quantity (local measure) of crop residues provided;
• quantity (local measure) of by-products provided;
• quantity (local measure) of natural grass hay provided;
• time animal was watered (local measure);
• labour inputs (local measure) that day;

• principal source of labour (son, daughter, self, hired help etc.)

• veterinary inputs and costs; and
• other notable events that day.
These daily forms are normally filled out by children with their parents at
night – thus having an additional social benefit of enabling children to practically
use their literacy and numeracy skills, and become involved with improved
livestock management at an early age. The recording book should be designed
with a weekly summary sheet, which the farmer completes with the local
technician on a regular basis. This forms a focus for the relationship between
these partners in the improved forage development program, and enables the
local technician to question any inconsistencies. At the same time, analysis of
previous data can be discussed and any modifications to management practices
discussed. Auto-recording is well suited to monitoring fattening programs and
dairy production.
Assessment of forage legume performance is a more complex matter
because it requires a long term monitoring framework and an appreciation of the
morphology of each species and cultivar being monitored. For example, stylos
generally have poor productivity during the first year of establishment. After that
they are much more productive. Apart from forage production monitoring
mentioned above, assessment of new cultivars and different management
strategies should be done by local technicians and regional subject matter
specialists as part of the adaptive research activities supporting the improved
forage production program. Persistence of species and cultivars under normal
management conditions should be one of the major criteria monitored.

IX. FINANCIAL AND ECONOMIC IMPACT OF FORAGE

DEVELOPMENT PROJECT
Introduction
The forgoing chapters indicate that numerous forage development

strategies and species mixes have been identified and proven for a range of
ecological niches in Ethiopia. However, the impact of increased forage
production on animal performance and farm-level incomes is not as well
understood or documented as the agronomy of forage production. Internationally,
the effect of improved nutrition on animal production is well proven and
documented, and countries such as Australia now have very large areas of
improved forage developed for cattle and sheep production in the both the
temperate and tropical zones. The financial impact of improved forage
production is also well understood in many countries, but this is not the case in
Ethiopia, where past emphasis has been on animal health, rather than health and
nutrition.
The missing information relates to the animal response predictions once
improved forage is available and dietary levels of energy and protein increase,
and to the financial impact at the farm-level. As discussed previously, benefits
from increased forage production are numerous and diverse. Therefore it is
necessary to keep analysis of forage production and animal response models
simple and in formats which can be used by field extension staff as part of their
farmer training programs. If used properly, forage development and animal
response models can be powerful extension and training tools, particularly if they
are progressively upgraded as local information on yields and animals responses
– e.g. weight gains and survival rates – become available as project M&E
systems provide data.
During preparation of FLDP, numerous forage production/animal
response models were constructed as part of project analysis. These models were
completed before spreadsheet programs for computers were available, and
therefore were cumbersome to use and manipulate. However, use of computers
became widespread in Ethiopia during the project implementation phase, and this
resulted in considerable flexibility in the approach to financial and economic
analyses of project benefits and costs.
The remainder of this chapter discusses the likely financial and economic
impact of forage development projects, using models prepared during FLDP
implementation. The models could be used for analysis of forage production
projects in any part of the country, provided project agronomists have a clear

understanding of the local farming systems and the crops grown for food and
cash.
Public Sector Investment and Recurrent Costs
A feature of FLDP was the high financial and economic rates of return.
This was because development relied on the simple intervention of local seed
distribution and an extension service, which was essentially in place at the time
of project preparation. The strategy at the outset was to use existing resources –
government staff, farm-level resources (land, labour and limited capital), and
local farmers for seed production, after an initial period of importation.
The only major capital items required for a successful forage development
project are:
Simple seed storage sheds – preferably constructed on a regional basis -
with basic seed cleaning equipment. In most instances, seed can be cleaned to a
satisfactory standard by hand, packed in small bags and then distributed prior to
rainy seasons.
Adequate transport facilities for extension staff. Depending on the size of
the project area, this can vary from four-wheel drive vehicles and motor cycles,
to smaller fleets for smaller projects. Good transport systems are essential,
because timeliness of operations such as seed harvesting and planting is most
important. Preferably, transport should be dedicated to the forage project. In
addition, it is necessary to equip forage development projects with fleets of small
and medium sized trucks for transport of seed, seedlings and cuttings. Transport
for extension staff will vary with the type of service used. If Development Agents
are located in strategic villages, field-level extension staff only require a bicycle,
motor cycle, or horse/mule. If these staff are scattered widely through the project
area, they will need more expensive forms of transport. Experience indicates that
a satisfactory approach is to locate multi-disciplinary Development Agents in
selected villages, particularly if farm sizes are small and the farming community
lives in villages or extended family groups.
A central office for general project coordination. Space is usually available
in the Ministry of Agriculture's headquarters. Ready access to within-country
communication services is important, so that seed is delivered on time and funds
for local seed purchases are available when required. Farmers growing seed
under contract soon lose interest if they are not paid according to the agreed
schedule. The central office should have adequate computer and report
production facilities, and at least access to a local printer for production of
extension materials and training aids.
As with capital expenditure, forage development projects have low
requirements for recurrent expenditure. Apart from vehicle operation and
maintenance, there are no major items of plant and equipment, which need to be

maintained and/or replaced, and if designed carefully, it should be possible to

keep incremental staffing down to a minimum. The aim should be to increase the
effective utilisation of existing staff resources, rather than to plan on the basis of
appointing new staff. Experience indicates that staff educated in a wide range of
disciplines and with limited exposure to forage agronomy developed into
effective forage extension staff once field-level training courses were in place.
Training is an important recurrent cost, but need not be excessive,
provided the number of overseas higher degree courses are kept in perspective.
Astute use of technical assistance can result in the development of large numbers
of specifically trained forage technicians who can be more effective. A carefully
planned and controlled training and staff development program is an integral part
of a successful forage project. Timely study tours – in-country and overseas – are
often the most cost-effective means of introducing new forage production
technology. For example, the use of Leucaena for erosion control on steep land
in Indonesia was common practice long before the system was introduced into
Ethiopia. Today, the best way to convince extension staff that the system works
is to run brief study tours to the Eastern Islands of Indonesia where thousands of
farmers are now stable and productive because of the use of Leucaena in an alley
cropping system.
The cost of seed is also a major recurrent expenditure item. As detailed
above, the key to success is to import large quantities of seed for at least the first
three years of the project. This will ensure that the impact of the improved forage
is observed over a wide area. It also protects against the negative impact of early
failures of the contract seed production system. One of the main reasons for the
success of FLDP was management's willingness to import initial quantities of
seed. This was in addition to substantial in-country seed production. In addition,
it is necessary to allocate adequate local budget for the purchase of seed grown
under contract by farmers. These funds must be available on time if contractors
are to continue to participate in such schemes. Payment for local seed production
worked smoothly when fund allocations were planned well in advance. However,
there were examples of seed wastage when funds were inadequate or not
delivered to the regional offices on time. A well-organised seed procurement
system is essential for successful forage production projects, as eventually
imports become too expensive and governments rightly express the desire for
self-sufficiency in forage seed.
FLDP experience shows that some technical assistance can be useful,
particularly in the areas of seed production and forage agronomy. In some
instances, further assistance may be required in animal nutrition and small-scale
fattening, forage and animal production economics, monitoring and evaluation,
legume nodulation and other technical problems, and project management. Most
of the latter can be provided by short-term consultants. If funds are limiting, the
key disciplines are seed production, processing and storage; and forage

agronomy. These two key aspects of forage production projects must proceed in
unison – seed is required to support an expanding forage development program,
and to a considerable extent, large areas of forage are required for seed
production. This is because a high proportion of the seed required can be
produced from forage set aside when flowering occurs. Multi-purpose forage
trees are capable of producing seed and forage at the same time, provided that
farmers are aware of this opportunity.
Farm-Level Investment and Recurrent Costs
As with public sector costs, farm-level investment and recurrent costs

associated with forage development are usually low and affordable by peasant
farmers, provided that seed is locally distributed. By far the largest farm-level
input is family labour – for forage planting and utilisation. It is important to
ensure that labour demand for forage development and utilisation activities do
not overlap with the labour inputs required for food cropping activities. Managers
of forage production projects must be aware of the likely impact of forage
activities on the overall supply/demand situation for family labour. This means
that a M&E system should be established in the first year of the project so that
assumptions used in the forage production/animal response models can be
confirmed/rejected and the predictive models modified. This is particularly
important in situations where the opportunity cost of farm labour is high, e.g.
during the period when crop weeding is critical for good food crop yields.
Indicative labour supply/demand figures for forage production and utilisation are
scarce and difficult to collect accurately. More reliable figures are available from
projects in South East Asia. Attempts to collect data in Ethiopia were partially
successful, but FLDP did not succeed in obtaining good information on labour
use profiles for forage related-activities. Therefore the predictive models
discussed are based on observations in FLDP's project areas and measured labour
use figures from Asia. Given that labour is by far the largest input at the farm-
level, it is important that forage development projects focus on accurate
collection of data on this cost item.
Other farm-level inputs for forage development projects include:
• Simple animal feeding equipment – for small-scale fattening and dairying. This could
include a container for water cartage if animals are being fattened under the tethering
system.
• Basic shelter for dairy cattle – note that one of the main constraints to milk production
in Ethiopia is inadequate grazing hours. This problem is exacerbated by cattle
spending prolonged periods in poorly ventilated and inappropriate (European design)
dairy barns.
• Simple hand tools for forage harvesting – sickles are adequate for non-tree legumes
and all grasses, whereas small handsaws are best for harvesting tree legumes. Trees

such as Sesbania and tree lucerne can be damaged if branches are hacked off with
blunt instruments.
• Basic seed storage containers – these can be the same as those used to store cereal
seed.
• All livestock should be treated against the main "killing" diseases such as rinderpest
and anthrax, etc. In addition, good responses to the control of internal parasites are
achieved when nutrition is improved. This means that animals in a fattening program
should be drenched, provided they can be separated from other livestock to avoid re-
infection.
Selection of Forage Development Strategies
A key to successful forage projects is the selection of appropriate forage

development strategies, which complement existing farming systems. FLDP was
successful because the strategies discussed in Chapter IV did not compete for
land with regular food cropping activities. In some instances, additional labour
was required, but this was offset by increased returns from both crop and animal
production. It is essential to ensure that all elements of a forage development
program can be integrated into the prevailing whole farm production system. Any
conflict between forage production and food/cash cropping activities will result
in low forage production adoption rates.
This means that project staff must have a sound understanding of current
farming systems before forage development activities commence. Such an
understanding requires far more than simply a knowledge of the crops grown and
farm sizes. Experience from Ethiopia indicates that it may be necessary to
conduct some baseline surveys prior to the commencement of forage
development, so that information on the following key variables is available:
• Land use (by season and activity – cropping or grazing). This includes information on
the areas of fallow, availability of communal grazing, area of house/garden compound,
and cropping intensity.
• Crops grown, the area of each crop and crop by-product production. In some areas,
crop residues are the main source of animal forage, and this often forms the basis of
improved animal nutrition through the addition of leguminous protein and additional
energy from improved forages.
• The number of labour units available to average families – expressed as adult
equivalents.
• If the availability of draught power is a limiting factor, it is important to have
information on the number of oxen per farm and the number of oxen per hectare
cropped.
• The structure and make-up of an average farm's livestock herds and flocks. This will
enable assessment of the likely impact of increased quantities of improved forage on
animal production and farm incomes. This information, when combined with data on
farm area and land use, will also enable the existing stocking rate to be calculated,
preferably as Tropical Livestock Units (250 kg liveweight) per ha.

• Information on the sources and use of fuel for cooking and heating will provide a
guide as to the importance in incremental dung (for fuel) and wood (from tree
legumes).
• Current livestock feeding practices – use of supplementary straw or hay during the dry
season (or the short rains) will indicate an understanding of the importance of feeding
crop residues, and a familiarity with an animal production system which will benefit
substantially from the introduction of improved forages into the farming system.
Table A2.1 in Annex 2 lists profiles of selected Service Cooperatives in

highland Ethiopia. The information in the table was collected as part of a Service
Cooperative Development Planning exercise and was to form the basis of farm-
level development planning when Ethiopian agriculture was collectivised. The
concept proved too ambitious, but did provide useful information which
supported the strategy of integrating improved forage production into existing
farming systems. Once the area cropped and the types of crops are known, it is
possible to determine the area/km of each forage development strategy for
average farms in the survey area. For example, Table A2.1 shows that farms
associated with the Kolba Service Cooperative in could implement the following
forage development strategies:
• 0.1 ha of pasture development on communal grazing land, assuming that 30% of the
area was sown;
• 0.7 ha of undersown legumes (75% of the area cropped with cereals and tree crops
which are suitable for undersowing – sorghum, maize, millet, enset, coffee and chat);
• 0.6 km of alley planting – based on 85% of the area sown to annual crops and 0.4 km
of alley per ha;
• 0.4 km of strip/contour planting – based on 85% of the area sown to annual crops and
0.3 km of strip/contour per ha;
• 600 m of backyard forage in a plot; and
• 180 backyard forage trees.
Benefits from Forage Development and Rates of Return
Once the quantities of each forage development strategy have been

estimated for average farms, it is possible to estimate the level of benefits, which
flow from each strategy. The benefits from forage development are diverse and
wide-ranging, from improved animal production to increased fuel supplies.
Therefore it is important to keep the analysis of forage-based benefits simple and
realistic. It is not necessary to quantify all primary and secondary benefits,
because past analysis of forage projects indicates that high financial and
economic rates of return can be achieved even when secondary benefits (such as
increased food crop production resulting from improved soil fertility) are
ignored. The approach should be to only include those benefits, which can be
easily quantified up to the point where a satisfactory rate of return is obtained.

Thereafter, the analytical exercise becomes somewhat academic, because projects

with economic rates of return above 25% are often viewed with a degree of
suspicion by economists who are not familiar with forage and animal production
systems.
Analysis during FLDP project preparation was based on incremental
production of meat and milk. The effect of improved supplies of fodder on
draught power, and the impact of increased quantities of fuel (dung and wood)
were ignored. The project was estimated to generate an economic rate of return
of about 35%. This figure was based on a much more substantial management
infrastructure than the system eventually implemented by the government. The
mid-term review of FLDP estimated that the project's economic rate of return
would be approximately 25%, based on progress during the first three years and
realistic farmer acceptance rates. In conclusion, it is apparent from FLDP
experience that forage development projects generate high financial and
economic rates of return when only selected benefits are quantified and included
in the analyses.
Forage Production and Utilization Models
During the course of FLDP a series of forage production/animal response

models were developed. These were based on the following:
• Area/length of each strategy for an average farm in a given agro-ecological zone;
• Phased development of each forage development strategy over a five-year period;
• A figure for the maximum level of forage utilisation for each strategy – e.g. 70% of
the forage produced by backyard fodder tree would be available for animal production,
and the corresponding figure for alley planting would be 85% of the incremental
forage produced;
• The percentage legume content in the incremental forage produced by each strategy;
• The number of person and oxen-pair labour days required to establish a given
area/length of strategy;
• Increased forage yield (kg DM/ha) for each strategy, and the phasing of this increase
over a five-year period;
• Estimated labour use profile for each strategy as forage yields increase over time;
• Energy (MJ-ME/kg) and protein (crude protein per cent) of grasses, legumes and crop
stovers; and
• The value of incremental supplies of dung and fuelwood, on a per kg basis.
Once this information is fed into a spreadsheet model called
STRATMOD, it is possible to calculate the following for each nominated forage
development strategy:
• Labour use for forage development and utilisaion;
• Number of oxen pair days required for forage establishment;
• Incremental forage yields over a 10 year period;
• Total incremental dry matter available, expressed as kg DM, MJ-ME, and CP%;

• Incremental dung and wood production, and the total value of by-products.
An example of the STRATMOD model, based on forage development
strategies for the Chelo Service is given in Table A2.2 in Annex 2. The Chelo
model is based on the following forage development strategies.
• Pasture development – 0.01 ha;
• Undersowing – 0.74 ha;
• Alley planting – 0.34 km;
• Strip planting – 0.26 km;
• Backyard pasture – 570 m2; and
• Backyard trees – 170 (number).
Table A2.2 indicates that incremental forage production under these

conditions would range from 318 kg DM in the first year, to 2,393 kg DM seven
years after the program commenced. The corresponding figures for incremental
energy and protein would be 2,927 and 22,093 MJ-ME, and 54 and 419 kg CP.
Incremental dung production, based on 50% digestibility and 75% recovery
would range from 119 kg in the first year to 897 kg by year 7. The equivalent
figures for incremental fuelwood production would be 44 and 935 kg. The total
value of by-products would increase by Birr 458 by year 7. This indicates that
farm incomes would at least double, once the entire farm was "saturated" with
improved forage. This is before consideration of the impact of improved forage
on animal production.
Once incremental forage has been estimated using the STRATMOD
model, it is possible to "superimpose" STRATMOD on the profiles of whole
farms (refer Table A2.1) to determine the impact of increased supplies of
improved forage on animal production. The farm profiles in Table A2.1 include
estimation of energy and protein balances for existing livestock populations. For
example, for farms associated with the Kolba Service Cooperative, it appeared
that inadequate protein was limiting efficient use of surplus energy – there was a
surplus of 6,324 MJ-ME, but a deficit of 156 kg CP in the "without" project
situation. When the impact of the forage development strategies is considered,
the energy surplus increases to 35,154 MJ-ME, and the protein deficit becomes a
surplus of 400 kg CP.
In terms of impact analysis, the next step is the most difficult, because it is
based on the subjective assessment of how existing livestock on the target farm
respond to increased supplies of energy and protein. The approach taken for
analysis of FLDP, which is reflected in Table A2.1, was based on the estimated
rate of change in the supply of energy and protein. For example, if over a six-year
period, the levels of both energy and protein slowly changed from negative to
positive – as is the case for the Fonco Service Cooperative – it was assumed that
off-take rates for sheep/goats and cattle during the same period would increase by
a maximum of 20% and 10% respectively. In the case of Fonco Service

Cooperative, this would men that an extra 0.27 sheep/goats and extra 0.42 cattle
would be sold six years after the forage development program commenced. If
sheep/goats and cattle are valued at Birr 30 and 250 per head respectively, this
means that annual gross income would increase by Birr 105 by year 6. For
simplicity, it was assumed that incremental income in the first year would be
zero, and that thereafter income would increase from Birr 35 in year 2 to Birr 105
by year 6. Further details are given in Table A2.1.
For those farms with immediate surpluses of protein and energy, for
example farms associated with Dembessa Service Cooperative, it was assumed
that the impact of increased supplies of improved forage would be immediate,
and that a static level of increased income would be generated from year 2
onwards. For example, for the Dembessa farms it was assumed that annual
incomes would increase by Birr 64 per year, one year after forage development
commenced.
A further analytical step was necessary for those farms with large
surpluses of energy and protein, for example Kolba Service Cooperative. It was
assumed that these surpluses would be used to fatten culled oxen for specific
markets, such as Christmas and Easter when livestock prices are high. In the case
of Kolba, it was assumed that by year 3 there would be adequate incremental
forage to result in increased production from the existing herds and flocks and to
fatten 0.5 oxen or about four sheep/goats. This activity would generate a gross
margin of about Birr 350 per oxen, or approximately Birr 175 in year 3. The total
incremental income from livestock production activities in year 3 and 4 would be
Birr 290. By year 5, incremental income would have increased to about Birr 465.
When added to the increased income from dung and fuelwood (Birr 644 for
Kolba farms) increased annual income attributable to the forage development
activities would amount to Birr 1,110 – refer Table A2.1 for details.

Financial and Economic Rates of Return from Forage Projects
The STRATMOD model calculates total incremental labour days required

to develop and utilise improve forage supplies. For example, Table A2.1 shows
that farms associated with the Service Cooperative at Hulahubeta required an
additional 63 labour days per year once the forage program reached full
development. If this requirement is assessed against an annual incremental
income of Birr 235, this is equivalent to Birr 4 per additional day. A return per
person day of Birr 4 is high when compared with the opportunity cost of labour
of approximately Birr 1. Table A2.1 in Annex 2 shows that incremental returns
per person day would vary from Birr 1 (first year's operation in Hulahubeta
Service Cooperative) to Birr 8 per day in year 6 for forage development on farms
associated with the Cheki Service Cooperative.
The financial rates of return for farmers participating in forage
development projects are invariably high because cash expenditure is minimal –
the main input is labour which usually has a low opportunity cost. This means
that analysis of forage development programs at the farm level is best completed
by consideration of returns to incremental labour. It should be noted that the
returns specified for selected Service Cooperatives do not take into account the
following benefits, which were not quantified:
• Increased draught power and the impact on crop production due to improved oxen
liveweights;
• Increased crop production due to increased supplies of organic matter and nitrogen
from legumes grown on crop land, particularly through the undersowing strategy;
• Decreased herding time by children – this results in more time for education;
• Reduced soil erosion on crop and grazing land though the use of vegetative barriers
(alley cropping and forage strips) and oversowing on steep communal grazing land;
• Improved dry season stream flow due to increased filtration and reduced run-off;
• Improved human nutrition resulting from increased supplies of milk – from both
indigenous and improved dairy breeds;
• Improved human nutrition from increased consumption of meat – small ruminants and
cattle; and
• Improved standards of housing due to increased supplies of building materials from
multi-purpose tree legumes.
The economic rates of return from forage development projects are also
high. This is because public sector investment is low and incremental staffing can
usually be kept to a minimum by improving the skills and knowledge of existing
staff. In most instances, incremental production consists of non-traded goods –
milk, meat fuelwood, etc. This means that financial prices have to be adjusted by
standard conversion factors, which reflect the fact that non-traded goods (those
that do not earn or save foreign exchange) are worth less than traded goods
(which do earn or save foreign exchange).

At the time the analysis was undertaken there was a wide separation
between the fixed official exchange rate and the shadow or market exchange rate.
In these circumstances, the standard conversion factor was estimated to be
between 0.3 and 0.2. However, the impact on the economic rate of return of low
conversion factors was more than offset by low shadow prices for labour – this
figure is as low as Birr 0.50 per day in some areas when crop weeding is not an
important activity. In summary, forage development projects generate high
economic rates of return and are not unduly influenced by exogenous risk factors,
provided that seed is produced and distributed.


LIST OF ANNEXES
ANNEX 1 : Species Description and Characteristics.
ANNEX 2 : Animal Performance of Improved Forage.
ANNEX 3 : Inoculation of Legumes.
ANNEX 4 : Model Contract for Supply of Pasture/Forage Seed.


ANNEX 1: Species Description and Characteristics
The following information provides a general introduction to the key

species and cultivars, which have proved successful in the FLDP. The
information is general and the species and cultivars described by no means
exhaustive. These species and cultivars provide a useful starting point for
improved forage programs because of their adaptability and proven forage value.
However, forage programs should include wide screening of other species and
cultivars during their early years to identify the broadest genetic base possible for
improving the quantity and quality of forage available for ruminant livestock.
IMPROVED FORAGE LEGUMES
Forage legumes are herbaceous (not woody shrubs) and are used in
undersowing, intercropping, oversowing or grazing areas, improvement of stock
exclusion areas, and in mixed pasture establishment. Forage legumes with a
sprawling or climbing habit can also be used in backyard forage plots.
Establishment is always by direct seeding. Because of this, only those species
and cultivars with good seed production capacity and ease of seed establishment
should be selected. Ease of establishment and persistence are other important
factors to consider during assessment of cultivars for selection. Species and
cultivars with proven capabilities in Ethiopia are detailed below. Their attributes
are summarised in Table 5.1 in Chapter V and Table 7.1 in Chapter VII. Other
species assessed by FLDP included Neontonia wightii (Glycine), Arachis pintoi
(Forage Peanut cv Amarillo), Calopogonium mucunoides (Calopo). Vigna
parkeri (Creeping Vigna cv Shaw), Aeschynomene falcata (Joint Vetch cv
Bargoo), A. americana (Joint Vetch cv Glenn), Centrosema pascuorum (Centro
cv Cavalcade or Bundey), and Ornithopus compressus (Yellow Serradella).
Macroptilium atropurpureum (Siratro)
Siratro is a cultivar of M. atropurpureum, a perennial, sprawling/climbing

forage legume with an important role in undersowing and improving stock
exclusion areas. Its primary use is for forage, with secondary uses being for
erosion control and nitrogen fixation. Siratro grows below 2400 m altitude and
requires more than 600 mm annual rainfall. It is adapted to a wide range of well-
drained soils, including low fertility sandy soils, but performs poorly on wet
sites. Siratro requires careful grazing management for high productivity and is
better suited to cut and carry systems based on undersowing and stock exclusion
area strategies. Repeated low cutting kills plants but they respond well to
continuous light cutting. Siratro has moderate palatability, which increases with

age. Its low energy content makes it unsuitable as a quality dairy ration but it is
useful for fattening strategies. It establishes easily with sowing rates from 0.5 to
2 kg/ha and is a useful component of roadside and aerial seeding mixes. If newly
established plantings are allowed to seed in their first year plant density will
quickly increase because of heavy seed setting. This is one of Siratro's best
features and makes it highly suitable for oversowing strategies below 2000 m in
areas with annual rainfall exceeding 750 mm. Seed production is best from crops
grown on trellises, tall cereal crops or browse trees – with harvests of up to 1000
kg/ha possible. Siratro is an excellent understory species for plantation and
forestry crops if tree spacing is sufficiently wide – this is essentially an agro-
forestry strategy with good soil conservation characteristics because of Siratro's
ground cover characteristics. It climbs very vigorously and is useful for
controlling weedy grasses and other species.
Macrotyloma axillare (Axillaris)
Axillaris is a perennial, sprawling/climbing forage legume highly suited to

undersowing, intercropping and improving stock exclusion areas. Cultivar Archer
has been used most successfully in Ethiopia. It grows best in warm to hot areas
below 2400 m altitude and requires more than 600 mm annual rainfall, and
complements Siratro and Greenleaf Desmodium. Axillaris is adapted to a wide
range of soils, including low fertility sandy soils, and is moderately tolerant of
waterlogging. Axillaris needs careful grazing management for high productivity
and is most productive with cut and carry systems based on undersowing and
stock exclusion area strategies. It should not be cut below 20 cm since repeated
low cutting will kill plants, but they respond well to continuous light cutting.
Axillaris has low to moderate palatability, which increases with age and makes it
suitable for oversowing common grazing land and soil conservation areas.
Livestock may need a familiarisation period when they are first introduced to this
forage. It establishes easily with sowing rates from 0.5 to 1 kg/ha and is well
suited to roadside and aerial seeding strategies. Seed production is limited by a
short flowering period and is more successful when crops are grown on trellises,
tall cereal crops or browse trees. Harvests of up to 500 kg/ha possible. Axillaris
can be used as an understorey species for plantation and forestry crops if tree
spacing is sufficiently wide. It combines well with Elephant grass.
Desmodium uncinatum (Silverleaf)
Silverleaf is a perennial, sprawling forage legume suited to undersowing,

intercropping and improving stock exclusion areas. This species is less important
than the related Greenleaf Desmodium. It grows below 2200 m altitude, requires
more than 900 mm annual rainfall, and is tolerant of cool weather and light

frosts. Silverleaf should be grown on relatively fertile, well drained soils. It needs
careful grazing management for high productivity and is most productive with
cut and carry systems based on undersowing and stock exclusion area strategies.
Continuous moderate cutting or grazing is preferable to occasional very heavy
cutting or grazing. Silverleaf should be allowed to seed at least in the first season.
It is palatable and thus is better managed for cut and carry systems. Silverleaf
requires a moderate to fine seedbed with sowing rates from 1 to 2.5 kg/ha.
Because of this, it is not suited to oversowing strategies. Seed production is more
successful when crops are grown on trellises, tall cereal crops or browse trees
with harvests of up to 400 kg/ha possible.
Desmodium intortum (Greenleaf)
Greenleaf is similar in many respects to Silverleaf but it is less tolerant of

cool weather and light frosts. It is a perennial, sprawling forage legume suited to
undersowing, intercropping and improving stock exclusion areas as is one of the
most important forage legumes available for Ethiopia. Its primary use is in forage
production and its secondary uses are for nitrogen fixation and erosion control.
Greenleaf grows below 2400 m altitude and requires more than 700 mm annual
rainfall. Greenleaf should be grown on fertile, well-drained soils and needs
careful grazing management for high productivity. It is most productive with
regular cut and carry systems based on undersowing and stock exclusion area
strategies. Continuous moderate cutting or grazing is preferable to occasional
very heavy cutting or grazing management for high productivity. It is most
productive with regular cut and carry systems based on undersowing and stock
exclusion area strategies. Continuous moderate cutting or grazing is preferable to
occasional very heavy cutting or grazing, and it should be allowed to seed at least
in the first season. Greenleaf is palatable and thus is better managed for cut and
carry systems. It requires a moderate to fine seedbed with sowing rates from 1 to
2 kg/ha and so is not suited to oversowing strategies. Seed production is more
successful when crops are grown on trellises, tall cereal crops or browse trees
with harvests of up to 400 kg/ha possible. The simplest collection method is to
strip ripe pods from the stalk, between thumb and forefinger (Robertson, 1990).
Pods are thoroughly dried, and seed is threshed out carefully using traditional
mortar and pestle systems. It is normally possible to harvest and clean at least 1
kg per person day. Greenleaf is exceptionally successful under perennial tree
crops and forestry because of its shade tolerance.

Stylosanthes species (The Stylos)
The stylos are very useful because they are often extremely hardy, and
will grow on soils of very low phosphate status and on some quite acid soils.
Palatability is variable, but always increases during the during the dry season
when the plants "hay off", because of this, sytlos are less suited to cut and carry
systems and more suited to common grazing strategies such as oversowing or
improvement of stock exclusion areas. Most sytlos establish readily by surface
sowing even on rough seedbeds. Burying seed may prevent germination. Typical
seeding rates are 1 kg/ha from oversowing and 2 kg/ha for undersowing.
Establishment and early growth are assisted by controlling companion grasses
with grazing. Seed should be heat treated before sowing (see Chapter VII) but
generally does not need to be inoculated. Most stylos are tolerant of heavy
grazing pressure, and some will set seed and spread even under severe grazing. In
fact the key to good stylo management is to prevent excessive competition form
companion vegetation by regular cutting or grazing. Stylos typically take two to
three years to become fully established, after which their productivity increases
significantly. This aspect of their growth should be taken into account during
assessment of their capability in new areas. Stylosanthes fruticosa is native to
some parts of Ethiopia. The stylos with proven capabilities in Ethiopia are
detailed below:
Stylosanthes hamata (Caribbean Stylo cv Verano)
Verano is a short-lived perennial plant growing to about 30 cm in height.

It grows best at low altitudes and needs more than 500 mm annual rainfall. It
performs best in warm to hot environments with productivity and seed set
significantly reduced by low temperatures including low night temperatures.
Verano is adapted to a wide range of soil types but performs best with good
drainage. In the presence of vigorous grasses heavy grazing is usually necessary
to maintain a high Verano population in the sward. This cultivar tolerates the
fungal disease anthracnose, which affects most stylos. Verano seeds heavily (up
to 1500 kg/ha) over a long period and is best harvested by sweeping the ground
in the early dry season. For this reason, soils, which compact but do not crack in
the dry season are more suitable for seed crops than cracking or self mulching
soils. It should be possible to harvest and clean at least 5 kg clean seed per person
day from a good seed crop. Verano has a high proportion of hard seeds and quick
establishment will be improved by treatment of seed (see Chapter VII). Because
of its heavy seeding characteristics and ease of establishment, Verno is well
suited to roadside and aerial seeding strategies.

Stylosanthes scabra (Shrubby Stylos cv Seca)
Seca is a long lived perennial plant growing to 150 cm in height. It grows

best below 2000 m altitude and needs more than 500 mm annual rainfall but is
more cold tolerant than Verano. Seca is adapted to poor, sandy, acid soils and
requires good drainage. This cultivar tolerates the fungal disease anthracnose,
which affects most sytlos. It is the least palatable of the stylos and is very hardy,
making it ideal for stock exclusion areas and oversowing degraded common
grazing areas. Its great genetic diversity makes it highly adaptable with a good
capacity to spread. Seca seeds moderately (200-400 kg/ha) and is slow to flower
in its first season. However, it has a strong capacity to rapidly increase in density
after the first two seasons. Seed is best harvested in the early dry season by cut-
and-thresh techniques followed by sweeping the threshing ground. On oversown
areas, Seca is most successfully spread by allowing animals to graze plants in full
seed and then moving them to unsown areas where the seed passes through the
animals and spreads the plant to the new area.
Stylosanthes guianensis (Perennial Stylo cv Cook)
Cook is a long-lived perennial plant growing to 60 cm in height. It grows

best at low altitudes and need more than 700 mm annual rainfall. Cook is adapted
to poor, sandy, acid soils and requires good drainage. This cultivar is susceptible
to the fungal disease anthracnose, which affects most stylos. Cook seeds lightly
(less than 300 kg/ha) and has a restricted flowering period. Because of this it is
less tolerant of heavy grazing than other stylos. Seed is best harvested in the early
dry season by cut-and-thresh techniques followed by sweeping the threshing
ground.
Lablab purpureus (Lablab)
Lablab is a vigorous annual or short-lived perennial legume with very

vigorous seedlings, which is best promoted as a dual purpose species. It grows up
to 2400 m altitude and requires more than 400 mm annual rainfall. Lablab is
suited to a wide range of soils but will not tolerate salinity or water-logging.
Being large seeded, it establishes easily on a rough seedbed with seeding rates of
18-20 kg/ha for pure stands, 15 kg/ha for undersowing and 2 kg/ha for forage
strip sowings. Successful establishment requires the seed to be covered before
germination. Lablab should be cut or grazed regularly and lightly. It should not
be cut below 25-30 cm. It has moderate palatability and cattle may require
several days to become acquainted with it. Lablab leaf supplements other forages
well and its seed is an excellent human food. Lablab seeds heavily (500-1000
kg/ha) with maximum yields being produced from trellised crops or those
growing with browse trees.
Cassia rotundifolia (Wynn Cassia)
This is synonymous with Chamaecrista rotundifolia. Wynn Cassia is a

very hardy, low management species, which will persist and spread very rapidly
in areas below 2000 m altitude. It is sensitive to frost and prefers sandy soils of
low fertility. Wynn Cassia is extraordinarily drought tolerant and can produce
seed within two months of establishment, because drought stimulates early
flowering. It tolerates heavy grazing pressure and because of its low palatability
is well suited to oversowing strategies and improving stock exclusion areas.
Wynn Cassia establishes well with oversowing, even without any land
preparation and with seeding rates less than 1 kg/ha. Heavy grazing of
companion species assists with early establishment and normally encourages
persistence by reducing competition from grasses. Wynn Cassia is suited to
undersowing strategies and will normally set seed before harvesting the
companion crop, if sowing is undertaken at the time of final weeding. This
species has shown remarkable ability to increase its density and spread form
roadside and aerial seeding where there is little competition, making it ideal for
rehabilitation of degraded areas. Wynn Cassia seed is best harvested in the same
way as Verano stylo, with one person being able to sweep, thresh, and clean at
least 2 kg of seed per day.
Vigna unguiculata (Cow Pea)
Cow Pea is an annual dual purpose legume suited to a wide range of

environments. It has a wide range of genetic material. Cow Pea grows in
lowlands up to 2500 m and is drought tolerant – maturing with anything more
than 300 mm annual rainfall, depending on cultivar. This species grows on a
wide range of well-drained soils and will tolerate gentle cutting or grazing during
the growing season. Cow Pea will establish on a fairly rough seedbed and is
sown at 15 kg/ha for pure stands and about 12 kg/ha for undersowing. Seed does
not need to be inoculated in most parts of Ethiopia. Cow Pea is ideal for
undersowing or intercropping strategies and complements maize and sorghum
particularly well. In this case the cowpea is utilised after harvesting the
companion crop – either by cut and carry harvesting, or direct grazing. The
mixed cowpea hay and stover makes excellent conserved forage for dairy
animals. Seed should be collected for subsequent sowing before grazing. Cow
Pea seed is also a valuable human food. It yields 500-800 kg/ha seed with
undersowing or intercropping strategies.

Vicia dasycarpa (Vetch)
Vetch is a vigorous climbing/sprawling annual legume with a wide range

of adaptation and high level of farmer acceptability. It grows well between 1500
and 3000 m altitude and is suited to a wide range of rainfall – typically anything
above 400 mm per annum. Vetch grows on a wide range of soils but requires
good drainage for optimum productivity. It is ideally suited to undersowing,
mixed pasture and backyard forage plots and establishes readily, even on rough
seedbeds. The seed should be inoculated before sowing. Typical sowing rates are
20 kg/ha for pure stands, 12 kg/ha for undersowing, and 5-12 kg/ha as a pioneer
component of mixed pasture. When sown at 12-20 kg/ha with oats, vetch makes
excellent hay. On many sites natural regeneration from self-sown seed is
minimal, necessitating annual sowing. Vetch is most suited to undersowing and
is self-regenerating where it is allowed to mature and seed before harvest of the
companion crop. Seed yields between 400 and 1000 kg/ha are common but
shattering occurs. Because of this, vetch grown on trellises or tall companion
crops such as maize and sorghum are ideal for seed collection. One person can
harvest and clean up to 15-25 kg seed per day (Robertson, 1990). Languedoc
Vetch (Vicia sativa) is better suited to low rainfall areas because it matures
earlier than common vetch.
Medicago sativa (Alfalfa)
Alfalfa is a long-lived perennial legume producing large quantities of high

quality forage under good management. Common cultivars include Hunter River,
Hairy Peruvian, Siriver, Paravivo, and Sequel. Alfalfa (also known as lucerne)
grows over a wide range of altitudes and is tolerant of a wide range of
temperatures. It develops a deep taproot, which enables the plant to withstand
drought once established. Alfalfa produces good forage yields with more than
600 mm of rainfall and is ideally suited to irrigation. Alfalfa is suited to a wide
range of well-drained soils but best production is on neutral to slightly alkaline
soils. Inoculated seed should be sown into a well-prepared seedbed with seeding
rates of 8-10 kg/ha for pure stands, and 5-6 kg/ha for mixed forage plots. Young
seedlings require early weed control for best results and so row planting is
preferable. Seed should be sown to no more than 1 cm depth on heavy soils and
no more than 2 cm depth on light soils. Alfalfa produces the most forage when it
is cut with a rotational system with an interval of 30 to 45 days during the
growing season. As a rough guide, alfalfa is ready to harvest when about 10 per
cent of the crop is flowering. Alfalfa should be cut about 3 cm above the crown
to avoid damage to the growing points and weeds should be controlled after each
harvest. Alfalfa is best suited to cut and carry systems supervision. Alfalfa has

very high feed value and should be used as a supplement for crop residues and
natural hay in a mixture of 30 per cent alfalfa and 70 per cent other roughages.
Trifolium repens (White Clover)
White Clover is a versatile perennial clover, primarily for altitudes above

2400 m, but because it is drought and heat tolerant it has some uses in lower
altitudes. It has a much longer growing season than native trifoliums in Ethiopia
and will typically grow for some time into the dry season. White clover seeds
heavily and will spread rapidly, even under heavy grazing pressure. Native
trifoliums in Ethiopia include Trifolium decorum, T. burchellianum, T.
quartinianum, T. semipilosum and T. tembense.
Lotus pedunculatus (Maku Lotus)
Maku Lotus is a perennial legume with a very wide adaptation, primarily

above 2400 m altitude. It requires more than 800 mm rainfall, is frost tolerant,
and will grow on a wide range of sites from those that are well drained to those
that are waterlogged. Maku Lotus is tolerant of heavy grazing and, being shade
tolerant, can grow through a canopy of weeds and companion grasses. It is useful
for improvement of stock exclusion areas, cover cropping in agro-forestry
strategies, oversowing (especially in bottom lands subject to waterlogging) and
contour forage strips. Maku Lotus does not produce prolific quantities of seed but
has proven its ability to persist and spread in Ethiopia.
IMPROVED BROWSE LEGUMES
Tree legumes are extremely important elements in improved forage

production programs because of their productivity and multi-purpose uses. Being
perennial trees or shrubs they have a three dimensional root system and crown,
which greatly increases the productivity of systems where land area is limited.
Apart from large quantities of quality forage, browse legumes have deep rooting
systems to increase their productivity during the dry season, and they produce
other products such as fuelwood, construction timber, seeds, and bee products. In
addition, browse species provide shelter and privacy, which is valued by many
farmers. The following species have fast growing attributes, which make them
especially suited to improved forage programs. Their attributes are summarised
in Table 5.2 in Chapter V and Table 7.1 in Chapter VII.

Leucaena leucocephala (Leucaena)
Leucaena is a browse legume of great importance in Ethiopia. Its major

use is in the production of highly palatable, nutritious forage. Its secondary uses
include fuelwood production, nitrogen supply for companion crops, erosion
control, and shelter. Leucaena fuelwood burns well and the cut plant coppices
rapidly. It is suited to warm-hot regions below 2000 m altitude and requires more
than 400 mm rainfall. There are now many cultivars and hybrids of Leucaena
which will increase the range of this genus. For example, L. leucocephala x L.
diversifolia hybrid KX3 will tolerate cooler conditions and should be suitable to
2400 m altitude. Leucaena prefers well-drained, fertile soils and will not tolerate
highly acid or waterlogged soils. It is most productive on neutral to alkaline soils.
Leucaena is susceptible to the leaf sucking psyllid Heteropsylla cubana which,
although uncommon at the moment, gives a reason to discourage Leucaena
monocultures in favour of leguminous browse polycultures. Suitable mixing
species include Sesbania sesban, Cajanus cajan and local leguminous browse
species such as Acacia sp. and other exotic species, such as Calliandra
calothyrus and Flemingia.
Leucaena and its companion browse legumes are well suited to backyard
forage programs, contour forage strips, agro-forestry systems such as alley
cropping, oversowing on sites with self mulching soils, intercropping with
perennial tree crops, and for soil conservation on stock exclusion areas. It is best
established by direct seeding for all strategies except contour forage strips and
agro-forestry where seedlings are more appropriate because they reduce the risk
of young plants being killed by early grazing. Bare rooted seedlings are the best
strategy for nursery production of plants to be established in the field. When
transplanting bare rooted seedlings, leaves should be carefully stripped from the
seedling to reduce moisture stress. If bare rooted seedlings are planted the same
day that they are dug from the nursery, 90 per cent survival is possible. This is
why village nurseries are more practical than centralised nurseries. Contour
forage strips and alley cropping systems should have Leucaena or their browse
legumes spaced at about 50 cm between plants in the row. Offset double row
planting is useful in some areas, in which case between row spacings of 1 m are
normally successful. As detailed in Chapter IV, Leucaena produces best with cut
and carry systems but is well adapted to browsing also. In Ethiopia annual yields
of 2 kg DM forage per metre of contour forage hedge and up to 10,000 kg DM/ha
have been achieved on farmers fields. Pure stands will produce more than 2,000
kg seeds per ha with single trees in contour forage strips producing up to 2 kg
seeds per year. Seed is best collected by plucking ripe pods from the trees before
they shatter. This generally done on an opportunistic basis. Cut trees continue to
flower and set seed and make seed collection easier because they have less
height. Leucaena is a good seed crop for farmers new to this industry because of

its high production and because it seeds over a long time, thus spreading the
labour demand. It is possible to harvest and clean 12 –20 kg seed per person day
(Robertson, 1990).
Sesbania sesban (Sesbania)
Sesbania is an adaptable browse legume which will live for up to 7 years

and will grow in some environments not suited to either Leucaena or tree
lucerne. It is highly palatable and its major use is a forage with secondary uses
for shelter, and nitrogen fixation for companion crops. Sesbania wood is not
useful for fuel or construction but can be used for fencing. It produces best below
2000 m altitude, is very frost sensitive, and not very drought tolerant – requiring
more than 600 mm annual rainfall for survival. Sesbania grows on a wide range
of soils including very poor acid sands and waterlogged soils, but responds well
to improved fertility. It should not be cut or browsed before it has become
established, after which regular cutting produces the best results – as detailed in
Chapter IV. Cutting intervals of about 6 weeks result in good productivity.
Infrequent cutting or browsing results in very low leaf:stem ratios. Sesbania
produces less forage than Leucaena but what it does produce is of very high
quality. It should be grown with other browse species to reduce the impact of
insect attack, which can be severe in monocultures. Sesbania is highly suited to
contour forage banks, living fences in backyards, and alley farming. It establishes
rapidly from direct seeding and this is the establishment method of choice. When
sown into a well-prepared seedbed at the beginning of the major wet season and
kept weed free, Sesbania establishes rapidly. Cutting or browsing should not
begin until the trees are more than 1 m in height. It does not self-seed easily so
direct seeding is required for all establishment. Despite this, Sesbania seeds
heavily, especially from trees, which have not been cut for 12 months, and can
yield more than 1 kg seed per tree. Pods are harvested by hand and one person
can typically harvest and clean 3-5 kg seed per day (Robertson, 1990).
Cajanus cajan (Pigeon Pea)
Pigeon Pea is a short-lived dual purpose shrub legume providing forage,

grain for human consumption, and low quality fuelwood. It lives up to 4 years
and is an excellent crop to improve food security and integrate forage and
cropping systems below 2400 m altitude. Pigeon Pea establishes rapidly on a
wide range of well-drained soils and requires more than 350 mm annual rainfall
for good production. The plant tolerates only light cutting or browsing and is best
developed in contour forage strips, house compound hedges, and intercropping
strategies. It recovers well from light cutting and is highly palatable, as detailed
in Chapter IV. Because of its use for human food, Pigeon Pea is a useful plant to
introduce the concept of contour forage strips and alley cropping to lowland
farmers. Once Pigeon Pea is established, and farmers are used to the idea of a
productive third dimension to their cropping systems, longer lived browse
species such as Leucaena can be introduced to fill gaps as Pigeon Pea dies out.
The plant is best established by direct seeding using 10-50 kg/ha for pure stands
(not recommended for forage strategies) and about 1 kg/km of contour forage
strips. Seed should be sown into a well-prepared site and covered with no more
than 2 cm soil. Early growth is slow but once established plants require little
attention. Pigeon Pea intercrops very well with maize an sorghum. Pigeon Pea
productivity depends very much on the cultivar used. Annual seed yields of 1,500
to 3,000 kg/ha have been recorded but pod boring insects reduce yields. One
person can harvest up to 20 kg seed per day (Robertson, 1990).
Chamaecytisus palmensis (Tagasaste/Tree Lucerne)
Tree lucerne is a temperate, multipurpose browse legume of major

importance to highland areas of Ethiopia. It is one of the few highly productive
browse species for altitudes above 2000 m altitude. Its primary uses are forage
and fuelwood but farmers also value it for shelter, bee forage, nitrogen fixation,
and soil conservation purposes. The species has a remarkable level of genetic
diversity with major differences in canopy shape, branching habit, leafiness, and
flowering capacity. This diversity is a major advantage in view of the wide range
of AEZs over which the plant is grown. Tree lucerne tolerates mild frosts but will
not tolerate water logging at all. It is most suited to well-drained fertile soils but
is productive on infertile acid sands as long as they are well drained. It is drought
tolerant once established but requires more than 400 mm annual rainfall for
maximum productivity. Tree lucerne is well suited to backyard forage programs,
contour forage strips, agro-forestry systems such as alley cropping, oversowing
on sites with self mulching soils, and for soil conservation on stock exclusion
areas. It is best established by direct seeding for extensive strategies but intensive
strategies such as backyard forage, contour forage strips and agro-forestry should
use nursery-grown seedlings because early seedling growth is very slow. This
also reduces the risk of young plants being killed by early grazing. Bare rooted
seedlings are the best strategy for nursery production of plants to be established
in the field. Seedlings should be planted the same day that they are dug from the
nursery, making village nurseries more practical than centralised nurseries. Tree
lucerne should be planted at about 50 cm between plants in the row for contour
forage strips and alley cropping systems. Off-set double row planting is useful in
some areas, in which case between row spacings of 1 m are normally successful.
As detailed in Chapter IV, tree lucerne produces best with cut and carry systems
but is well adapted to browsing also. It should not be harvested until it has
reached more than 1 m height. In Ethiopia annual yields of 1 kg DM forage per
metre of contour forage hedge and up to 4,500 kg DM/ha have been achieved on

farmers fields. Single trees in contour forage strips produce up to 1 kg seed per
year after they are two years old. Seed is best collected by plucking ripe pods
from the tree before they shatter. This is generally done on an opportunistic basis.
Cut trees continue to flower and set seed and make seed collection easier because
they have less height and more synchronised flowering. Tree lucerne is a good
seed crop for highland farmers new to this industry because of its high
production. It is possible to harvest and clean up to 4 kg seed per person day
(Robertson, 1990).
IMPROVED GRASSES
Most ruminant livestock in Ethiopia rely on local grasses for their

roughage and much of their nutrition. Many of these species have low
palatability, poor productivity and inadequate nutrients to maintain animals,
especially during the dry season. Improved grasses, many of them selected from
other parts of Africa, have better productivity, palatability and nutrient
characteristics that make them desirable for inclusion in improved forage
production programs. Species and cultivars with proven capabilities are detailed
below. Their attributes are summarised in Table 5.3 in Chapter V. other species
showing promise include Andropogon gayanus (Gamba Grass), Panicum
coloratum (Bambatsi Panic).
Cenchrus ciliaris (Buffel Grass)
Buffel Grass is extremely drought tolerant and is a very robust grass for
areas below 2000 m with more than 250 mm annual rainfall. It is adapted to
heavy cutting or grazing but is less palatable than many other grasses. Buffel
Grass establishes well from seed and is well suited to improvement of stock
exclusion areas and rehabilitation of degraded areas.
Panicum maximum (Guinea Grass/Panic)
Panic is an erect grass, useful for strip planting or mixed pastures in areas
below 2400 m altitude with more than 500 mm annual rainfall. It grows on most
soils but requires high fertility for good productivity. Panic produces good
quality forage and is well adapted to cutting or grazing. Suitable cultivars include
Petrie and Gatton.
Setaria sphacelata (Setaria)
Setaria is a widely adaptable species for areas for areas below 2400 m
altitude with more than 700 mm annual rainfall. It grows on a wide range of soils
and tolerates waterlogging. Setaria is ideal for contour forage strips where it can
be established by direct seeding or from splits. It does not produce seed readily –
an advantage for contour forage strip strategies because it reduces the risk of this
plant becoming a weed. Cultivars include Narok, Solander, Nandi, and
Kazaungula, with Solander being the best adapted to highland areas.
Chloris gayana (Rhodes Grass)
Rhodes Grass is a stoloniferous grass suited to areas below 2400 m with

more than 600 mm annual rainfall. It grows on a wide range of soil types but
requires good fertility for high productivity. Rhodes Grass provides effective
erosion control but should not be used for contour forage strip strategies because
it can become a crop weed. It tolerates heavy grazing and cutting and so its
erosion control attributes are best used for stock exclusion areas and oversowing
strategies.
Pennisetum purpureum (Elephant Grass)
Elephant Grass is a very productive, sterile grass for cut and carry systems
and is well suited to contour forage strip strategies in areas below 2400 m and
with more than 700 mm annual rainfall. It responds dramatically to manuring and
is an excellent backyard forage – where it makes productive use of higher
nutrient loads. Elephant Grass must be cut regularly to maintain its forage
quality. It combines well with browse legumes and vigorous perennial forage
legumes such as Greenleaf Desmodium. Elephant Grass is planted as sets at the
beginning of the wet season.
Paspalum plicatulum (Plicatulum)
Plicatulum is a very versatile, low input grass for areas below 2000 m
altitude with more than 700 mm annual rainfall. It is drought tolerant and adapted
to very low fertility soils and to waterlogging. Plicatulum tolerates heavy grazing
once established and is well suited to stock exclusion areas and other degraded
sites. It produces good seed crops and is not recommended for contour forage
strips, although it has good soil conservation properties. Cultivar Bryan performs
well in Ethiopia.
Phalaris aquatica (Phalaris)
Phalaris is the most important grass species for forage development and
soil conservation in Ethiopia. It has good forage qualities and is best mixed with
other forages to optimise its value as a livestock feed. Phalaris performs well
between 1800 and 3000 m altitude, is frost and drought tolerant and is productive
with more than 400 mm annual rainfall. It requires fertile soils for strong growth
but will survive on poor soils, although its conservation value is diminished on
such soils because of weak growth. Phalaris establishes slowly but once
developed is well adapted to heavy grazing or cutting. It is suitable for contour
forage strips where its soil conservation properties are put to best use, but is also
suitable for backyard forage and mixed pasture strategies. In intensive programs,
Phalaris establishes well from splits either from nurseries or from breaking up
old clumps. It requires through seedbed preparation for direct sowing and does
not establish well with competition. Phalaris should be lightly grazed or cut
during early establishment to encourage the plant to stool. Phalaris does not set
fertile seed except where long growing seasons are possible, in which case annual
seed yields of 300-400 kg/ha are feasible with adequate nutrition. This lack of
fertility in many sites is an advantage for contour forage strip strategies because it
reduces the risk of Phalaris becoming a crop weed.
Dactylis glomerata (Cocksfoot)
Cocksfoot produces high quality forage in areas above 2400 m altitude

with more than 500 mm annual rainfall. It is less productive than Phalaris but
has a role in mixed pastures and contour forage strips, where it should be mixed
with leguminous species for best results.

ANNEX 2: Animal Performance of Improved Forage












ANNEX 3: Inoculation of Legumes
Legumes obtain some of their nitrogen from bacteria called rhizobia,

which invade their roots to form nodules. Different legumes require different
species of rhizobia. Some legumes are able to use rhizobia, which are already in
the soil, for example vetch, stylos, siratro, axilaris, lablab and cowpea all react to
rhizobia, which are commonly found in healthy soils in Ethiopia. Other legumes
need specific rhizobia to enable them to fix atmospheric nitrogen. Some of these
are listed in Table A3.1. Inoculation with the appropriate rhizobia is
recommended for most legume forage crops. It is cheap and ensures an early
supply of nitrogen to the developing plants. Where undersowing or intercropping
strategies are used, it also increases the opportunities for mutual benefits from
integrated forage and cereal production.
Seed inoculation is a very simple process. Peat-based inoculum is mixed
with seed either with water alone or with water plus an adhesive agent such as
gum arabic. For example, a packet of inoculum is mixed with 1 to 2 litres of
water and thoroughly mixed with 50 kg of legume seed. The seed is dried in the
shade because sunlight kills rhizobium. Since insecticide also kills inoculum,
seed treated with insecticide should be dried for several days before inoculation
and sown immediately into a very moist seedbed.
Table A3.1: Legume Rhizobium Requirements
Rhizobium Type Species

Group 1 (cowpea) a/ Siratro, Axillaris, Seca Stylo, Verano Stylo, Lablab, Glycine
Group A Centro
Group B Lucerne/Alfalfa
Group C White Clover
Group D Red Clover
Group E Lotus
Tree Lucerne Vetch, Tagasaste/Tree Lucerne
Leucaena Leucaena
Desmodium Greenleaf, Silverleaf
FSS Fine Stem Stylo
a/ Naturally Present in Ethiopian Soils

ANNEX 4: Model Contract for Supply of Pasture/Forage Seed
Contract of Sale for Pasture/Forage Seed
of ............................................................................................ (Species, Cultivar) (For seed

produced in the growing season ....................... to ....................................)
between The Fourth Livestock Development Project (FLDP), as purchaser and
.................................................................................................. (as producer)
(Address of Producer: Zone........................Awraja ....................................
Wereda....................................Service Co-operative ...................................)
The Producer agrees to:-
• Establish an area of ..... ha for the primary purpose of seed production of ..............
• Prepare a seedbed to a level deemed acceptable by an FLDP representative.
• Control the access of stock to the site.
• Control weeds as required by the FLDP representative.
• Use fertiliser, pesticides and herbicides as recommended by the FLDP representative.
• Harvest seed at times and with techniques as nominated by the FLDP representative,
with seed being provided for collection within 4 weeks of harvest.
• Clean seed to approximately .............. % purity.
• Follow the general technical advice of the FLDP representative on all matters of seed
crop husbandry.
• Sell seed to FLDP.
• Not declare the termination of the contract, except under the conditions expressly
provided by the law.
Failure to comply with the above conditions will result in disqualification form further
participation in the seed production programs.
The Fourth Livestock Development Project agrees to:-

• Provide initial seed requirement.
• Provide adequate supervision of all aspects of crop management and seed production.
• Purchase all seed collected during the production year ......................., from the
nominated area, at a price of Birr ......................... /kg, in cash payable at the time of
collection or as agreed with the producer.
• Collect/purchase seed at times and places to be determined by mutual agreement.
Signed
...................................... ....................................... ................................................
Producer's Name Producer's Signature FLDP Co-coordinating Unit
...................................... ....................................... ................................................
Service Co-op. Exec. Representative's For Zonal Office
Committee Representative Signature AFRDMD
Date .........................

BIBLIOGRAPHY
AACM. 1987. A Simple System for Defining Agro-Ecological Zones in

Ethiopia. Final Report – a National Summary. AACM Advisory Team,
Working Agriculture, Addis Ababa, Ethiopia.
AACM. 1989. Fattening Manual. Fourth Livestock Development Project

Ministry of agriculture, Addis Ababa, Ethiopia.
Alemayehu M., 1987. Feed Resource in Ethiopia. In: Animal Feed Resources for
Small-scale Livestock Producers. IDRC, Nairobi, Kenya.
Alemayehu M., 1989. Recommended Forage for Ethiopian Environments and

Current Strategies for Utilization. Proceedings of the XVI
International Grassland Congress, Nice, France.
Ash, A.J. 1990. Effect of Supplementation with Leaves from the Leguminous
Trees Sesbania grandifolia, Bizza chinensis, and Glericidea sepium on
the Intake and Digestibility of Guinea Grass Hay by Goats. An. Feed
Sc. And Tech. 28 (3-4): 225-232.
Biru, L., Lambourne, L.J., and Fana, T. 1988. Feeding Value Sesbania and
Leucaena Browse. In: Proceedings of Joint Workshop on Utilisation of
Research Results on Forage and Agricultural By-Product Materials as
Animal Feed Resources in Africa. ILCA, Ethiopia.
CARDI. 1986. Pasture Research and Development in the Eastern Caribbean.

Caribbean Agricultural Research and Development Institute,
Barbados.
Carter, E.D. 1975. The Potential Role of Integrated Cereal Livestock Systems
from Southern Australia in Increasing Food Production in the Near
East and North Africa Region. FAO/UNDP Regional Project on
Improvement and Production of Field Food Crops.
Donald, C.M. 1982. Innovation in Australian Agriculture. Sydney University

Press.
FLDP. 1989. Forage Extension Manual. Fourth Livestock Development Project.

Ministry of Agriculture, Addis Ababa, Ethiopia.

FLDP. 1991. Forage Seed Production Extension Manual. Fourth Livestock

Development Project. Ministry of Agriculture, Addis Ababa, Ethiopia.
Galang, M.C. et al. 1990. The Effect of Cutting Height and Frequency on the
Productivity of Sesbania seban var. nubica in a Sub-Tropical
Environment. Nitrogen Fixing Tree Research Reports Volume 8.
Gramshaw, D., and Walker, B. 1988. Sown Pasture Development in Queensland.

Old. Ag. J. 144 (2): 93-101.
Humphreys, L.R. 1987. Tropical Pastures and Fodder Crops, Second Edition.
Longman, UK.
Karim, A.B. et al. 1991. Effect of Cutting Height and Cutting Interval on Dry
Matter Yield of Leucaena leucocephala. Agroforestry Systems 16 (2):
129-138.
Kahurananga, J.C. 1988. Intercropping Trifolium spp. In Wheat and its

Suitability for Smallholder Farmer Conditions of the Ethiopian
Highlands. In: proceedings of Joint Workshop on Utilisation of
Research Results on Forage and Agricultural By-Product Materials as
Animal Feed Resources in Africa. ILCA, Ethiopia.
Lambert, M.G. et al. 1989. Forage Shrubs in North Island Hill Country. 2 Sheep
and Goat Preferences. NZ J. Ag. Res. 32 (4): 485-490.
O'Reilly, M.V. 1987. Better Pastures for the Tropics. Yates, Australia.
Otsyina, R.M. et al. 1987. Manual of Forage Bank Establishment and

Management. ILCA. Addis Ababa, Ethiopia.
Phiri, D.M. 1992. The Effect of Browse Supplementation on Maize Husk

Utilisation by Goats. Agroforestry Systems. 17: 153-158.
Robertson, A.D. 1990. Final Report on Forage Development Activities and

Proposals. FLDP, Ministry of Agriculture, Ethiopia.
Saleem, M.A.M. 1982. Effect of Time of Planting Stylosanthes hamata cv.

verano on Sorghum Grain and Legume Yield. ILCA Annual Report,
Ethiopia.

Tadesse, A. 1988. The Underexploited Potential of Improved Forages in the Mid-

Altitude and Lowland Areas of Ethiopia. In: Proceedings of Joint
Workshop on Utilisaiton of Research Results on Forage and
Agricultural By-Product Materials as Animal Feed Resources in
Africa. ILCA, Ethiopia.
Townsend, R.J., and J.E. Radcliffe. 1990. Tagasaste Forage Production Systems.
NZ J. Ag. Res. 33 (4): 627-634.
Udedibie, A.B.I., and F.O. Igwe. 1989. Dry Matter Yield and Chemical
Compositor of Pigeon Pea Leaf Meal. An. Feed Sc. And Tech. 24 (1-
2): 111-119.
Walshe, M.J. et al. 1991. Dairy Development in Sub-Saharan Africa - A Study of

Issues and Options. World Bank Technical Paper No. 135.
Washington, D.C.
Wagad, D., and Ndumbe, R.D. 1986. The Effect of Different Protein
Supplements on Weight Gain and Voluntary Intake of Maize Stover
by Cattle. In: Proceedings of a Workshop on the Utilisation of
Agricultural By-Products as Livestock Feeds in Africa. ILCA, 1986.

1 Forage and Livestock Production in Ethiopia-1 PDF

Hochgeladen von

Dokumentinformationen

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

1 Forage and Livestock Production in Ethiopia-1 PDF

Hochgeladen von

Copyright:

Verfügbare Formate

FORAGE PRODUCTION IN ETHIOPIA: A CASE STUDY WITH IMPLICATIONS FOR LIVESTOCK PRODUCTION

FORAGE PRODUCTION IN ETHIOPIA:

Ethiopian Society of Animal Production

Alemayehu Mengistu, 2002 I

© Alemayehu Mengistu, August 2002

Alemayehu Mengistu, 2002 II

TABLE OF CONTENTS .......................................................................................................................... III

PREFACE ............................................................................................................................................ VII

ABBREVIATIONS .............................................................................................................................. VIII

I. OVERVIEW OF LIVESTOCK FEEDING IN ETHIOPIA ............................................................... 1

Backyard Forage Production .............................................................................. 19

Alemayehu Mengistu, 2002 III

V. FORAGE SPECIES ....................................................................................................................... 37

Alemayehu Mengistu, 2002 IV

As part of its mission, the Ethiopian Society of Animal Production (ESAP)

Workneh Ayalew, Ph.D.

Alemayehu Mengistu, 2002 V

Alemayehu Mengistu, 2002 VI

Alemayehu Mengistu, 2002 VII

AEZ Agro-Ecological Zone

Alemayehu Mengistu, 2002 VIII

I. OVERVIEW OF LIVESTOCK FEEDING IN ETHIOPIA

Livestock production contributes up to 80 percent of farmers' income in

Alemayehu Mengistu, 2002 1

FLDP successfully developed and implemented a number of forage

Agro-Ecological Zones of Ethiopia

The agricultural and livestock production potential of Ethiopia is

Alemayehu Mengistu, 2002 2

Length of Growing Period Thermal Zone (Altitude)

Seasonal Constraints to Forage Production

Integration of Livestock and Cropping Systems

Because of growing pressure on land resources from increasing

Alemayehu Mengistu, 2002 3

Increasing the intensity of cropping through agroforestry, intercropping or

Improved Forage Species

Many indigenous forage species in Ethiopia have low productivity or low

Alemayehu Mengistu, 2002 4

areas. Because most forage production strategies involve integration of forage

Alemayehu Mengistu, 2002 5

Alemayehu Mengistu, 2002 6

II. INTEGRATION OF LIVESTOCK AND CROPPING SYSTEMS IN

Dominant Cropping Systems in Livestock Producing Areas

More than 75 percent of Ethiopia's livestock population is found in the

Opportunities for Integrating Livestock and Cropping Systems

Rainfed cereal and tree cropping systems in lowland areas present

Alemayehu Mengistu, 2002 7

Livestock development has frequently focused on animal health or

Alemayehu Mengistu, 2002 8

Impact of Increased Crop Productivity on Livestock Feeding

Cropping systems are expanding and intensifying to feed growing human

Alemayehu Mengistu, 2002 9

Table 2.1: Effects of Protein Supplements on Liveweight Gains of Zebu Bulls

Liveweight Gain Stover Intake

Table 2.2: Comparison of Browse and Concentrates on Dry Matter Intake

Concentrate Browse Mean Milk

Source: Biru et al, 1988

Because uplands are generally unsuitable for sustainable cropping, they

Financial Impact of Integrated Farming Systems

Benefits from integrated crop/forage production systems are substantial,

and milk production can be limited by low genetic potential. Furthermore, it is

Alemayehu Mengistu, 2002 11

Alemayehu Mengistu, 2002 12

III. IMPROVED FORAGE EXPERIENCE IN OTHER COUNTRIES