Front Cover
 Table of Contents
 List of papers
 Systems descriptions
 Experimental programmes
 Networking and workshop assess...

Group Title: Networking workshops report
Title: Report of a networkshop on draught power and animal feeding in Eastern and Southern Africa
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00089579/00001
 Material Information
Title: Report of a networkshop on draught power and animal feeding in Eastern and Southern Africa Ezulwini, Swaziland, October 4-6, 1983
Series Title: Networking workshops report - International Maze and Wheat Improvement Center ; 2
Physical Description: 93 p. : ill. ; 30 cm.
Language: English
Creator: CIMMYT, Eastern and Southern African Economics Programme
Publisher: CIMMYT, Eastern and Southern Africa Economics Programme
Place of Publication: Mbabane, Swaziland
Publication Date: 1984
Copyright Date: 1984
Subject: Animal traction -- Congresses -- Africa, Eastern   ( lcsh )
Animal traction -- Congresses -- Africa, Southern   ( lcsh )
Animal feeding -- Congresses -- Africa, Eastern   ( lcsh )
Animal feeding -- Congresses -- Africa, Southern   ( lcsh )
Genre: non-fiction   ( marcgt )
conference publication   ( marcgt )
Spatial Coverage: Swaziland
General Note: Cover title.
General Note: "June 1984."
 Record Information
Bibliographic ID: UF00089579
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: oclc - 38751386

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Table of Contents
        Table of Contents 1
        Table of Contents 2
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
    List of papers
        Page 8
        Page 9
        Page 9a
    Systems descriptions
        Page 9b
        Page 9c
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 40a
    Experimental programmes
        Page 40b
        Page 40c
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
    Networking and workshop assessment
        Page 78a
        Page 78b
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
Full Text




EZULWINI, SWAZILAND October 4-6, 1983

Networking Workshops: Report No.2

CIMMYT Eastern and Southern Africa Economics Programme,
P.O. Box 1473,
MBABANE, Swaziland.

June 1984


Introduction and Summary 1
List of Papers 8

Section One: Systems Descriptions

Sudan (Western) 10
Botswana (East and South-East) 14
Swaziland (Middleveld) 18
Zambia (Central Province) 21
Zimbabwe 24
Kenya (High Potential) 28
(Medium Potential) 30
Malawi 32
Lesotho 35
-- Farmer Strategies in India 40

Section Two: Experimental Programmes

Animal Nutrition in Lesotho 42
Tillage in Botswana 45
Tillage in Zambia 47
Tillage in Zimbabwe 49
Forage Legumes in the Ethiopian Highlands 52
Forage Legumes in Botswana 56
Forage and Crop Residues in Western Kenya 58
Forage and Crop Residues in Eastern Kenya 61
Crop Residues in Kenya 69
Livestock Research by ILCA 75

Section Three: Networking and Workshop Assessment

Networking 79
Workshop Output 80
Workshop Assessment 81


1. List of Participants 83
2. Opening Speech 85
3. Guidelines for Presentations 90
4. Instruction for Group Discussions 92

4 6 OCTOBER 1983

I irt C. ci ti C_- t- 1 -n3 ri = r ci 8 tE. -i m L rn a r- y

Report Layout

This report presents the proceedings of the CIMMYT networkshop on
Draught Power and Animal Feeding held in Swaziland from 4 6
October, 198:3. The report is arranged in three sections. A
summary inventory of the cross-country experiences discussed is
given at the end of this introduction.

In Section One a summary is given of each country's presentation
of the local farming system based on wall charts prepared by
participants. An abbreviated version of the discussions on the
systems as depicted on the wall charts and presented the previous
day in submitted papers follows in each case.

The relevant submitted papers) referred to are not included in
this report. Copies of any of the listed papers can be provided
on request from: Allan Low, Regional Economist, CIMMYT, PO Box
1473, Mbabane, Swaziland.

Section Two deals with the experimental programmes on animal
nutrition, tillage, forage production and crop residue feeding
that were described by participants. For each presentation an
abstract of the submitted paper is provided, together with any
additional information not given in the papers. This is followed
again by an abbreviated version of the discussions on each
experimental programme.

Again, the papers referred to are not included in this report,
but can be obtained on request.

Section Three summarises discussion on the need for networking
and how to do it and indicates how the partic.:ipan ts ,assessed this
networkshop in terms of general approach and the specific format

A first draft of this report was distributed to all participants
in November 1983. We are grateful for the numerous
comments/ammendments received, most of which have been
incorporated in this final version

The Origins of the Networkshop

In 1982 USAID and CIMMYT agreed to co-operate in a project to
promote on-farm research with a farming systems perspective in
Eastern and Southern Africa. USAID's interest stemmed from its
commitments to .agricultural research and extension projects in
ten countries through the region from Sudan to Swaziland.
CIMMYT's interest stemmed from its world-wide experience in
on-farm research with a farming systems perspective and the
desire to expand the coverage of its on-farm research support
programme in the Eastern and Southern African region.

One of the major activities of the combined CIMMYT/USAID project
is networking. This activity is defined as the facilitation of
discussion and exchange among both national researchers and USAID
team members in various programmes so that the accumulating
experience can be shared. The first networking activity took the
form of a seminar for senior agricultural research administrators
which was held in Nairobi in April 1983. At this seminar the
research administrators endorsed the need for networking and
recommended that this include focusing on specific technology
issues, and the inventoring of cross-country problems and
solutions (CIMMYT, 1983).

The Need for Networking and the Role of Technical Workshops

There is no need to emphasise the value of exchanging ideas,
information and experience in any sphere of life. In a new field
of activity, such as on-farm research, the need for special
networking efforts arises because:

(a) practitioners tend to be few and scattered;

(b) formal links, e.g., through specialised journals, do not
ex i st;

(c) the development of sound methodologies will not speedily
occur in the absence of cumulative trial and error

On top of this we have the situation in Eastern and Southern
Africa where on-farm research is being promoted very largely
through international donor-aided projects in support of national
research and extension programmes. Each of these projects is
individually funded, contracted, implemented and assessed. Even
within USAID there is no provision for exchange of ideas and
experience among the various contractors. Regrettably,
co-operation and exchange of experiences between donors is even
less likely.

In these circumstances there is every chance that the lessons and
experience gained in one project in one country will not be taken
account of in the planning or implementation of the next project
in a neighboring country. Given the young stage of on-farm
research methodology and the way it is being rapidly promoted
through a large number of individual projects (over 30 USAID
projects in Africa have an on-farm research component),
networking is expected to pay substantial dividends.

The dividends from technical workshops such as this are expected
to come from generating exchange and widening the data and
experience base from which technical researchers can draw in
their local specific situations. We do not expect networkshops
such as this to come up with common solutions to a common
problem. Rather, we expect that knowledge of .how the same
problem manifests itself under different circumstances and a
knowledge of a range of potential solutions that have been tested
under different conditions will allow researchers to:

(a) better understand the specific nature of the problem in
their own localities;

(b) be in a better position to choose technologies for on-farm
experimentation that have a good chance of being successful
in their particular localities; and

(c) identify gaps in current component research efforts.

Networking Modes

Given the need for networking, the next question is what is the
best way to go about it. The widespread distribution of a
regular publication is one obvious way of keeping OFR/FSP
researchers in touch with what is going on outside their own
localities or countries. CIMMYT puts out a farming systems
newsletter from Nairobi. Experience suggests that this type of
networking tends to be somewhat passive in that most readers are
happy to receive information and ideas but only a few make
significant contributions. More active ways of seeking out
ideas, experiences and results and bringing them to the notice of
others can usefully complement the newsletter approach.

Mike Collinson recently circulated some proposals on how a more
active networking initiative might function. He suggested that
the first stage should be to identify specific areas of concern
that were common across country locations. The next stage would
involve the inventoring of two aspects of the area of concern:

(a) First, we would describe the characteristics of local
specific systems where the problem area was important. The
aim would be to see how the problem manifested itself under
different agro-climatic, economic and social conditions and
to inventory the ways in which farmers were managing the
problem under different circumstances.

(b) The second inventoring task would concentrate on listing
past and ongoing component, station-based, research and
generate a directory of current researchers working on the
problem area.

The aim of the two inventoring tasks being to give on-farm
researchers ready access to a body of information which they can
bring to bear on particular problems when they are diagnosed as
important in a local farming system.

This workshop was programmed to follow the same format, i.e.,
description of systems where access to draught power has been
diagnosed as important and reporting on experimental work that
addresses this problem through animal feeding or reduction in
draught requirements.

The advantage of workshops such as this one is that they maximise
potential for active participation among researchers. But not
all researchers involved in a particular problem can possibly be
represented at one time so coverage of systems and experimental

work cannot be as comprehensive as with the direct inventoring
approach. We think that both the inventoring and workshop modes
have their place and it is envisaged that both types will be
initiated under the CIMMYT/USAID networking activity.

A report of the first inventory exercise on the draught power and
animal feeding problem in Botswana, Lesotho and Swaziland is
being sent to conference participants and is available to others
on request from the CIMMYT Swaziland office.

Draught Power and Animal Feedina

The subject of this first technical networkshop is draught power
and animal feeding because timely access to draught for initial
cultivations has been observed to be a major problem in many
local farming systems from Sudan to Lesotho. Difficulty in
gaining access to adequate draught power appears to be limiting
the development of many farming systems. Because oxen are weak
at the start of the season, or because they are not readily
available at all, plantings are delayed, seedbeds are
inadequately prepared and weeding becomes a problem. This
problem of draught power presents us with a major research
opportunity where we could make a big impact on the productivity
of many farming systems in the region by improving the
availability and/or effectiveness of draught power.

Workshop Summary Inventory of Cross-Country Experiences

The following summary of workshop proceedings is presented in the
form of an inventory of interventions identified as being
appropriate in local systems and of research work described and

Table 1 presents a summary of the interventions thought to be
appropriate in the local farming systems described by
participants. The types of intervention mentioned have been
divided into two broad categories:

(a) interventions aimed at improving the capacity of animals to
generate more draught power;

(b) interventions aimed at reducing the demand for draught.

It would appear that the latter type of intervention was thought
to be more appropriate in Southern (Zambia, Zimbabwe, Botswana,
Lesotho and Swaziland) than Eastern Africa (Kenya, Sudan). This
may be related to better agronomic opportunities for increasing
animal feed production in Kenya especially through crop residues
and forage production. But it is probably also related to the
fact that in areas where the animal/grazing clash is more severe
(e.g. Kenya as opposed to Zambia), labour intensive methods of
producing and feeding crop residues or forages become more




(a) Feedi nq

Forage Crops
Fodder Shrubs
Forage Legume inter-cropping
Pasture improvement
Crop residues/by-products
Controlled grazing/stocking
Mineral Supplements
Other Supplements
Selective feeding
Education on nutrition requirements

Country in which
the intervention
was mentioned as
being appropriate
in a local system

Zi mb/KmKh/Bot
Bot/Les/KmKh /Su

(b) Animal

Water conservation
Improved conformation (breeding)
Cow management in draught
Stock management
education/extension (destocking)




Reduced/minimum tillage
Change tillage/planting sequence
Tractor use
Improved implements
Maize breeding/improvement
Cultivation planting techniques
Education on handling animals
and implements in draught

Zi mb
Za/Sw/Zi mb/Bot

Zi mb

* e.g. Winter ploughing/staggered planting with


Bot = Botswana

Kenya (medium potential)
Zambi a

Kh = Kenya (high





Apart from those generalisations no clear-cut divisions of
appropriate interventions by region or country emerged. Some
parallel work on forage legumes is being conducted across
countries. Data on the treatment and feeding of crop residues
came largely from Kenya, while reports of work on tillage methods
and planting sequences came from the southern countries.

In terms of the two broad categories of intervention, the
experimental programme work discussed in Section Two break down
as follows:



pen vs range feeding value of protein mineral lick


Babala and Dolichos Lablab production
direct and undersowing of grasses and legumes
range re-seeding and strip discing

Western Kenya

silage production
intercropping legumes
Napier grass for zero grazing
stall feeding
stripping and topping of maize plants
intercropping sorghum and early ratooning
pigeon pea interplanted
leucaena production

Eastern Kenya

improved productivity of natural pastures
physical and chemical treatment of crop residues
grown pasture and fodder crops
Zebu vs crossbred oxen in draught

Ethiopia (ILCA)

forage legume screening and germplasm bank
improving digestbility of crop residues
feed quality requirements for traction



plough planter vs broadcast and plough
tractor draught implements


winter ploughing
seed placement vs dribbling
planting implements
minimum tillage and herbicides


minimum tillage (ripper type and herbicides or direct
planting into winter ploughed fields)


- single ox plough
- use of cows in draught


CIMMYT (1983) "Report of a
Research Administrators from
Networking Workshops Report

Seminar for
Eastern and
#1, Nairobi,

Senior Agricultural
Southern Africa",
Kenya, April 18-20

4 6 OCTOBER 1983



G. Dlamini

Networking and the Draught Power/
Animal Feeding Problem

Draught Power in Botswana
I The Draught Power Problem
in Perspective

Draught Animal Feeding and Management
on Swazi Nation Land

Aspects of the Draught Power Constraint
on Maize Production in Recommendation
Domain No. 5, Kabwe Rural District,

The Crop Livestock Interrelationship
and Farmer Adaptation to Problems of
Reduced Cattle Numbers and Lack of
Dry Season Feed in Communal Areas of

The Draught Power Problem in Zimbabwe:
An Extension's View of the Causes,
Effects and Solution

A Description of a Local Farming
System (in Kenya) focusing on Draught
Power and Animal Feed Problems

Sources and Utilization of Power
and Equipment for Malawi's
Agricultural Sector

Animal Feeding and Management in India

A Realistic Look at Livestock
Research and Development in Lesotho

Value of a Winter Lick Fed to Cattle
under Lesotho Farmer Conditions

A. Low

T. Farrington
C. R. Riches

V. Watson,
S. Mamba

C. Chabala

E. M. Shumba

G.D. Mudimu

H.K. Muiruri

D.D. Singa

K.V. Ramanaiah./

E.W. Klosterman

M. Molapo,
L. Motjope
E.W. Klosterman


Opening Speech

Draught Power in Botswana
II The Design and Testing
of Solutions

Ground Preparation and Planting -
A Key Constraint for Ox Cultivators
in Zambia

On-Farm Minimum Tillage Experiment:
A Possible Solution to Draught Power
Problems gripping Zimbabwe's
Communal Areas

Forage Legumes in the Ethiopian

The Grazing Animal and Land
Pressures in Small Farm Systems

Feed Production Research for
Smallholder Agriculturalists in
Western Kenya

Animal Feeding in Small Farm Systems

ILCA's Approach to Livestock
Nutrition in the Farming Systems

T. Farrington
C. R. Riches

T.G. Maynard

E. M. Shumba

S. Jutzi

D.R. Chandler

Owi no

S. Tessema

A. K. Mosi
M. Butterworth


Improving the Nutrition of Grazing
Animals using Legumes, Fertilizer
and Mineral Supplements

Research on a No-till, Tropical
Legume-ley Farming Strategy

R.J. Jones

R.J. Jones,
R.L. McCown

Copies of the above papers are available on request from:

Allan Low,
Regional Economist,
PO Box 1473,


Systems Descri oti ons


(Western Sudan Agricultural Research Project)

Dr. Henson substituted for the invited Sudanese participants who
were unable to attend and gave the following presentation in lieu
of their submitted papers. Dr. Henson's presentation addressed
research being carried out by the Western Sudan Agricultural
Research Project (WSARP). The Nuba sedentary system specifically
was addressed.

The Farming System

The area under consideration is approximately 250 miles south
west of Khartoum in the Nuba Mountain area of the Kardofan
region. Animal traction as a mechanism for crop production is
only in its infancy. The Nuba practise a sedentary system of
agriculture, but maintain livestock as well. The area is fairly
densely populated, but the amount of land available is not a
limiting factor. The Nuba cropping system includes livestock
with animals for traction now being introduced by a French group
co-operating with the project.

Western Sudan: Nuba Sedentary Production System

tattlJe I iveweighL

Quantity of Forage

Quality of Pasture

J F M A M J J A S 0 N D

Rainfall (800mm)

Land Preparation



Hlarvestinq _____

In the cropping system in the Nuba mountains, each household
maintains three different farming areas. First, a house garden
is located around the dwelling, usually on the foot slopes of the
hills. Vegetables are grown here, e.g. early maturing maize,
sweet sorghum, tomatoes and others. This is for family
consumption and provides an early food input at the end of the
dry season. Some of the produce may be sold in the market. The
second area of crop production takes place on the near farm,
which is generally on sandy or sandy-clay soils close to the
mountains. Crops grown are for family sustenance and marketing:
sorghum, groundnuts, okra and sesame are produced. The third
area is the far farm usually on cracking clay soils. Sorghum,
sesame, cotton and groundnuts may be grown primarily for market,
but used for subsistence as well.

The sequence of planting commences with the house garden in early
May followed by the near farm in May/June, and the far farm in

Rainfall is about 800mm, falling late May to September. Land
preparation takes place April/May. Weeding: June to August.
Harvesting: September to December. Tillage of the far farms is
difficult due to the heavy soils. There are difficulties of
transportation from far farms so crops stay in the fields for a
long time.

Families usually consist of a man, his wife or wives and
children, making up an independent production and consumption

In terms of livestock nutrition and production, cattle, goats,
sheep and pigs are maintained with cattle most numerous. The
cattle, sheep and goats depend primarily on natural range forage
for the bulk of their nutrition. Crop residues are also grazed
in the field, but play a more limited role than natural
vegetation in terms of importance. Livestock numbers are limited
since the Nuba are primarily sedentary crop producers.

During the dry season from November to May livestock lose weight
and produce decreasing quantities of milk due to the poor quality
and limited availability of forage. After the rains begin,
natural forage regenerates, but is dominated by annual species,
which mature early and subsequently decline rapidly in
nutritional value. After the onset of the rains, the livestock
regain their lost weight, only to repeat the weight loss cycle
again as the dry season begins. This is shown in the included
systems calendar.

The use of animal traction has only recently been introduced by a
French group co-operating with the WSARP. The Nuba are
interested in the use of bullock power, but it is too early to

Based upon field observations, the primary constraint to
livestock product and hence on the potential use of animal
traction is nutrition. Nutritional inadequacy results in slow
growth, poor reproductive performance, increased susceptibility

to disease and others. Females reach sexual maturity at 4-5
years of age and reproduce only about every 2 years. These
factors plus the advsere impact of nutrition on milk production
greatly decrease livestock production.

In an attempt to alleviate the nutritional constraints, the WSARP
is investigating the use of forage legumes and grasses on fallow
land. Lab-lab and clitoria are examples. The harvesting of
native grass for hay is also being attempted as is the use of
supplementation such as sesame cake. It is too early to evaluate
the results at this time.


Asked whether the annual species in the pasture types would
respond favourably to cutting and still maintain nutritional
value, Dr. Henson replied that annuals which have been evaluated
are of reasonably good quality if they can be cut at the right

Asked for information on farm sizes, Dr. Henson said that the
house garden is very small just the area round the house. Near
farms are usually less than 1 ha., and far farms approximately 2
- 4 ha. The distance from the house garden to the far farm can
be about 2 4 kms.

In some instances people are involved in mechanised co-operative
schemes and can hire tractors and therefore farm a larger area.

The house garden requires early maturing varieties of vegetables
to get food as quickly as possible. From a horticultural
standpoint, utilisation of species which have a long producing
time would be better. Early maturing varieties of sorghum could
give more flexibility but have not been as acceptable as the
traditional longer maturing varieties.

Asked about cutting grass on communal areas, Dr. Henson indicated
that people would be able to harvest large areas. Availability
of grass is not a problem, but labour is.

Pastoralists utilize the same areas as the Nuba for pasture, but
the pastoralists go north with the rains; therefore the
sedentary people have access to the land while pastoral people
are north.

Burning is probably the most important factor in changing the
forage species from primary perennial to annuals. It is done for
a number of reasons. Pastoral people feel that if they burn they
get more nutritious new growth. However, there is limited
re-growth before the next season. This may have evolved from
earlier times when the land was dominated by perennials. Farming
people burn to keep out pastoral people. Also grass fires are
started accidentally by charcoal burners, honey hunters and from
domestic cooking fires.

It was suggested that there was a different appearance in
perennials due to burning rather than grazing and that with a

combination of grazing and burning you still get perennnials
coming up.

In the Sudan situation in Southern Kordofan it was said that
grazing pressure (forage resources in terms of mass of forage) is
under-utilized. Examination of present circumstances suggest
that the major impact on shift of species from perennial to
annual is because of burning.

The major range degradation is in the northern areas and is less
evident as one moves south. Very little work has been done to
identify the causes. When the rains come in the south the mud
gets very deep because of the clay soils and insects and endemic
diseases become greater. Pastoralists move their livestock north
to the sandier areas and where most markets are located the
result is increased grazing pressure in the more ecologically
fragile north and underutilization of the southern areas where
the forage is optimal in terms of nutritional quality. As
indicated, when the rains come it is very difficult to maintain
cattle on heavy clay soils. The Nuba people move their small
numbers of livestock and put them on the mountains. They
recognize livestock as an important economic resource. They
invest in cattle and migrate those cattle with the nomads.
Well-to-do people have joined their cattle up with the nomads so
that they migrate. Young men take jobs with the nomads caring
for the cattle.

Asked whether they use manure in the home gardens, Dr. Henson
said that they bring livestock in and maintain them whenever the
house garden is not actively growing. Cattle feed on residues
and make manure.


Refer to paper # 3 by Farrington and Riches

The Aqroclimatic Environment (East and Southeast Botswana)

Rainfall is variable and averages 500mm per annum. Rains start in
October/November. A mid-season drought sometimes occurs around
the middle of December. Rainfall picks up again after Christmas
and eases off in February. In mid-July frost.is common.

Crops grown include sorghum, maize, millet and cowpeas. Minor
crops include jugobeans, mungbeans, melons, squashes, pumpkins,
groundnuts and occasionally sunflower.

t U I S W A N A


Range condition

J F M A M J J A S 0 N D

Rainfall (+500mm)

Land Preparation





- - -

- - a fl -


-O -a


The range exhibits a very rapid growth of fresh grass at the
onset of the rains. Liveweight gains are greatest in the first
two months of the rains and tail off towards the end of Febru'ary
as grasses flower. Where overgrazing is not severe some
liveweight gains occur in July/August. Cows calve in August and

The Nature of the Draught Power and Animal Feeding Problem

Cattle make up 0BX of the draught, donkeys 107. and tractors the
balance. No more than 457. of farmers have access to their own
source of draught power. The majority of farm-households will
have problems in getting a draught span together as they will
have to borrow or hire animals. Access t6 draught power is only
one of the factors causing farmers to plant late. Other reasons
are that earlier planted crops mature earlier and are more prone
to attack by birds and that cattle range freely on arable land
during the winter until the onset of the first rains.

Planting does not start until after the first rains and only a
few days following rains are suitable for planting. Because of
this time constraint, uncertain rainfall and shortage of labour
due to off-farm employment, seed is usually broadcast and then
ploughed in either by draught animals or tractors.

Compounding these problems is the large increase in cattle
numbers (over 47. per annum over the last decade) and overgrazing.
As a result cattle are undernourished by the end of the dry
season and ploughing is delayed while they gain strength once the
rains start. Where grazing is poor animals move further afield in
search of food and a farmer can spend three days locating his
animals when the rains start.

Farmer Strategies

In the high risk situation of Botswana farmers tend to adopt a
low input approach. Low plant populations are planted, little
fertilizer or insecticides are used and few farmers do any
weeding. The exception to these low investments in cropping is
the hire of tractors. The costs at 40 Pula per ha are barely
covered by the average value of production (45 Pula/ha). However,
tractor hire enables farmers to follow one of their main
strategies for combating labour scarcity and rainfall uncertainty
which is to plough and stagger plant as large an area of land as
possible. Land is generally not limiting and very few farmers
have resources to plough more than what they currently have
access to.


The discussions centred around four major themes:

a) Migrancy
b) The number of animals required for making up a plough team
c) Cattle numbers, control, and fencing
d) Tractorisation

Since some Batswana have as many as three rural homes, there was
a query on how they spread their resoruces between the three:
village, lands and cattle posts. The village and the lands are
generally not too far apart. Women and children from 10 years do
most of the work on the lands. Most households indicated that
given sufficient water and cattle feed, they would stay on the
lands the year round. The cattle post is generally far from the
village and lands and is where the cattle are herded. Boys and
the old men of the family are generally employed at the cattle

Many young men and women work in off-farm wage employment. The
question as to where this work existed was answered as being on
the mines in Botswana or South Africa or in the Civil Service.
These wage earners generally send back money to hire tractor or
oxen as a substitute to coming back to do this work themselves.

Size of Plough Teams

In answer to the question why farmers on small land holdings
require 6-8 animals to plough, it was stated that anyone who is
ploughing will use 6-8 animals whether the area to be ploughed is
1ha or 8ha. Animal condition dictates the need for teams of 6-8.
For row planting and cultivating a two animal team is used.

It was asked if oxen become stronger later in the season and
whether this enabled teams to be reduced. The answer was that
farmers do not reduce numbers of animals in their teams, but that
the area covered per day may be greater.

Cattle Control

The importance of fencing to control cattle movement was noted
but participants queried the expense of fencing and its economic
viability. There was said to be a government subsidy on fencing.
Subsidised fencing was being well taken up in order to keep
animals off the land at the beginning of the season rather than
to keep fodder in at the end of the season.

Emphasis was now being placed on identifying communities who see
the need for grazing control and could be educated into
conservation. This is becoming increasingly difficult with
increasing populations of cattle and humans. Whether traditional
controls through headmen saying when enough stock were on a
specific area could work was questioned in view of the increased
populations of today.

It was asked whether there is a possibility of increasing rural'
ground water availability to cope with increasing cattle numbers.
Ground water resources are being studied and new dams are being
built. But although ground water supplies seem adequate, it is
not known whether they are fossil or rechargable. However, the
supply of more water must also be considered in conjunction with


work on grazing. Under the Tribal Grazing Land Policy there is
now a freeze on boreholes being drilled by individuals, only
syndicates can now drill. No individual has the right to deny
watering facilities to his neighbour, but these rights can be


A question was raised as to whether the tractorisation strategy
being employed by farmers was a problem in terms of production.
It was agreed that the use of tractors increased the speed of
ploughing, but it was said that tractors often did not do a good
job and tended to shallow plough. Also, there was the problem
that all of a farmer's land could be ploughed quickly at one
time, but that farmers often preferred to stagger their planting
to minimise the risk of total crop failure.

SWAZILAND (Middleveld)

Refer to paper # 4 by V. Watson, Z. Mamba and S. Mamba

The Aaroclimatic Environment

Rains start in October and increase on through to December. In
January and Februuary 150 mm per month is received. The rains
begin to taper off in March and April.

A few farmers do winter ploughing from July to September. Most of
the ploughing is done in October/November, tapering off in
December and January. Planting starts towards the end of October
through to November with some late planting in December. Beans
are planted in December/January. Weeding takes place from the
beginning of December to February. Maize is grown on 807.% of the
land and is often intercropped with pumpkin and beans. Other
crops grown are jugo beans, groundnuts, grain sorghums, cow peas
and millet.


Middle Veld

%o Crude protein -

100 - -


J F M A M J J A S 0 N D

Land Preparation -- --

Planting --. -

Weeding --

.a r estingII "o


The Nature of the Problem

Most homesteads derive income from three highly related
activities: off-farm wage employment, cropping and livestock.
Very few homesteads earn a living totally from the land. About
827. of the Swazi Nation land farmers have off-farm income and all
engage in subsistence cropping as well. Often wages from off-farm
employment provide for recurrent cash needs: school fees,
supplementary food, farm inputs and other consumer demands.

Cattle start picking up in September and are in peak condition in
November/December. During land preparation time cattle are not in
good condition to do the job required, mainly because there is a
shortage of feed in relation to the number of cattle. Farmers
regard cattle as a form of security and the tendency is to keep
them in preference to selling. The result is that planting is
done late and soil preparation is hurried and poor.

Far-mirs make only broad distinctions between livestock in terms
of management along lines of species. Cattle are grazed by day
and returned to kraaal at night and the accumulation of manure is
used to fertilise the soil. In winter cattle are free to roam
crop land as well as the veld.

Possible Interventions

In a recent survey conducted in the Highveld region 607. of
farmers indicated that if oxen had improved feeding and were in
better condition, they could begin ploughing earlier or do winter
ploughing. On the stocking of maize the most frequent response
was that it was not done because farmers were not aware of the
benefits. One quarter of farmers indicated that they did store
maize stover for animals but not especially for oxen. Better
storage and selective feeding are research opportunities.

Supplementary feeds could be produced in connection with the
"green belts" developed in the RDAs in areas between the crop
lands and the veld.

The demand for tractors is high and outstrips demand. The Tinkabi
is a low cost machine which has been purchased by 400 small
farmers, but its lightness and 16 horsepower engine does not
permit adequate penetration of compacted soils.

Government has implemented a fencing programme in which veld
areas are separated from crop land and paddocks divide veld
areas. Paddock demonstrations have been very effective because
farmers can see the results. To enter a grazing scheme farmers
usually pay E2 per beast per year and 80 cents per month.


The discussion covered cattle feeding, stocking rates and
farmers' strategies and problems.

a) Cattle feeding
Asked whether there was a digestibility cut-off with
supplementary feeding so that no weight gains took place
thereafter, it was said that if supplementation was started
between February and March they gained weight for 4 months and
then continued to do so when the grass greened up. This was
not said to be a common practice, it only occurred in the
grazing demonstrations where 5 hectares are to be established
for fodder production.

On the question of whether small farmers going in for forage
production did this on small plots and whether neighbours
objected to this, it was stated that if a farmer indicates
that he wanted to raise a dairy cow he has to put some of his
crop land into forage and fence this area.

b) Stocking Rates
A question was asked on the stocking rate in Swaziland and the
carrying capacity of the range. Swaziland was divided into 4
regions where the carrying capacities differed. Grazing time
on the high veld is 5 months, on the middleveld 8 months and
on the lowveld 12 months. No supplementation is necessary in
the lowveld. The national herd has been stable for the last
few years, but the data indicates that the calving percentage
has dropped from 39% to 30% in the last decade or so.

c) Cropping Problems and Farmer Strategies
Because of delayed land preparation planting is late and
yields are reduced. If the rains are delayed and the people
cannot get back from working in town they end up doing
ploughing and planting as one operation which puts a burden on
them and the oxen. This is why in some areas 95% of the land
is planted with the safim planter. Part of the reason for
farmers doing staggered planting is that they have to borrow
oxen or implements, but they also stagger plant because of
hail risk, lack of inputs and to spread the weeding burden.
The situation of the homestead, the labour they have, the
equipment they own, the cash that is available for inputs
(including tractor hire) and the other activities that they
are involved in all affect their timing and method of
ploughing, planting, weeding etc.

Nutritional considerations also affect farmer practices. Maize
planting is staggered to produce a long period of green
mealies. Pumpkin has a double function: leaves are eaten as
spinach and the fruit itself is eaten.

Noting that 400 farmers had spent E5000 on a technology, it
was asked if they had an abundance of off-farm income. It was
stated that most of the men have jobs off the farm and that
Swaziland was a country where people do spend a lot of money
on farming. 80% of farmers hire tractors and some of them own
tractors. For the same reason herbicides represented a
potential intervention: people were not afraid of spending
money on farming.

ZAMBIA (Central Province)

Refer to paper # 5 by C. Chabala.

The Aqroclimatic Situation

The climate is typified by a long dry season from March to
October and a short rainy season between November and March.
Both sandy and clay soils occur. Dambos are common where the
soils in shallow basins drain slowly and the grass remains

Maize is the staple. Local maize is planted first for early food
supply then hybrids are planted for consumption and sale.
Cotton, sunflower and vegetables are also grown. During the dry
season dambos are used for planting vegetables. Cultivations are
all by oxen or tractor, only 2% hoe. 54% of farmers own oxen and
73% own ploughs. Seedbed preparation is done using oxen and

Livestock production is poor. Most farmers do not have any
deliberate livestock management policy.

The Drauaht Power Problem

The draught power problem is reflected in two ways. Cattle are
poorly managed and therefore very weak. Secondly farmers do not
train their oxen.

In this area we are talking about a very extensive system where
farmers are fully occupied with arable farming and are
cultivating up to 10 hectares by ox cultivation. There is no
land limitation. The decision is whether to plant more or
whether to go back and weed the first planting. Priority is
generally given to the establishment of crops.

A major problem is late crop establishment. The draught
constraint comes in when there is enough moisture available in
the soil for planting and at the beginning of the rains there are
a lot of days when an acceptable moisture threshold is not

Possible Interventions

a) Maize breeding for types that can be established in a low
moisture threshold

b) Improving soil moisture by winter ploughing or using planters
that place seed more optimally

c) Increase rates of planting so that larger areas can be planted
with reduced or zero tillage


Asked why animal food supply does not appear as an intervention
or priority, Mr. Maynard answered that the people who did the
survey were not livestock specialists. Also there are dambos
within the area which provide winter feed and it has not been
identified as a key problem. Cattle do not appear to be in bad
condition and there is a potential clash with vegetable gardens
competing with cattle feed production.

Dr. Henson pointed out that a survey was done in an area of the
Sudan which included economists agronomists and in which the
information in relation to livestock was not satisfactory. The
survey was done again with a team including livesock specialists
and different perspectives emerged. It is therefore important to
have a correct mix of specialists.


Burning dambos was said not to be a problem. Neither was
livestock numbers a problem at the moment. However with recent
immigration into the area which had resulted in a doubling of the
population in 5 years it is likely to become more of a problem.
Tsetse was being controlled at the moment, but with increasing
herd sizes and if control measures were not kept up, tsetse could
come into the area again.

Asked what interventions were seen in terms of training animals,
it was said that this required closer contact between extension
and veterinary staff who do not go into the area. The facility
exists but it is not being exploited.


Refer to papers #6 and #7 by E. Shumba and G. Mudimu

The Aqroclimatic Environment

As in many parts of Africa, rainfall in Zimbabwe is seasonal.
The rainfall season is November to March. This is followed by a
dry winter period, April/May to August and then a hot dry season,

Vigorous grass growth starts with the onset of rain. Peak growth
is in January. In May the grass begins to dry. Quality of grass
improves as from the onset of rains and it peaks at the time of
flowering of the grass. From then on the quality drops.


1. Supplementary
I editiny
2. Sale of cull

LiveweighL Gain

Protein ContLent


Land Preparation




J F M A M J J A S 0 N D


- -

SAw _

- ~ -

Problem Points:

1. Draught
2. PlanLing Labour
3. Weeding Labour

4. Poor grass quality
and quantity
5. Loss of body mass

Cattle ownership is associated with differences in the cropping
system. In a recent survey in Mangwende it was found that those.
households with cattle have larger arable holdings, better land
preparation, timely weed control, apply manure and achieve higher
crop yields than those without cattle. Cattle owners cultivated
larger areas of maize, groundnuts and vegetables and sold these
for cash, giving them a larger income than non-owners. While
only 10%/. of non-owners grew sunflower, 45%. of owners did so.

Cattle ownership and crop production performance also tended to
be closely related to other factors associated with cattle
ownership in the area. These included more members working
permanently on the farm, more farmers involved in farmer
organizations, fewer household heads involved in off-farm
employment, less female heads of households, more non-farm cash
income to supplement crop inputs purchased on credit and less
risk aversion.

Farmers' Responases to The Problem of Falling Cattle Numbers

In 1975, 72%. of sample farmers owned cattle in Mangwende compared
with 467. in 1982. Average herd size (cows and oxen only) fell
from 6.86 to 3.29 and the total pool of animals dropped by 52%.
over the same period.

Cattle owners have reacted to falling numbers by including cows
in their draught teams and supplementing veld grazing with crop
residues in winter. Use of cows in plough teams has contributed
to lengthening the calving interval (one calf every three years)
and slow calf growth rates (weaned after two years) and hence low
build up of herds.

To counteract the winter feed problem farmers continue to
supplement veld grazing with crop residues until September to
October. However these residues do not have much feed value.

Mangwende farmers, particularly the cattle owners, have tackled
the problem of a weak draught power pool at the beginning of the
growing season by winter ploughing fields after crop harvesting
when soil is still moist. A few farmers (117.) indicated that
they planted directly into winter ploughed land with or before
the first rains to solve the problem of insufficient and weak
draught power at the start of the growing season. However 65%. of
sample farmers indicated that the problem of early weeds
discouraged the practice.

To counter the combined problems of draught power and low
fertility sample farmers have intensified their crop production
by concentrating available inputs on smaller areas. These
farmers have indicated a willingness to hire tractors at higher
rates than they now pay for oxen hire (Z$24/ha).


Discussion focused on forage legumes, control of livestock
numbers and the use of cows for draught.


a) Eorage Lequmes

Asked why it had been decided to intercrop maize with forage
legumes, it was said that this had only been done in the lower
rainfall areas where the situation is most serious. In high
rainfall areas it is not going to make any improvement. We
are trying to improve quality of feed in winter by
intercropping crop residues with legumes since farmers are
already using crop residues. To prevent other animals from
eating the crop it is necessary to cut and take it to the
homstead and store it. In higher rainfall areas the quantity
of the residues would be sufficient. Planting forages in
fallow land is limited by land availability. Legumes had been
sown into the veld but the problem there was lack of control
of animal numbers on the improved range.

To questions on when and what legumes were planted the answers
given indicated that cowpea and three kinds of beans were
being used. Legumes were late planted into the early planted
maize crop. The effect on maize yields was being monitored.

Asked whether the amount of arable land that could be fenced
to produce forage would be enough to supply the whole herd or
just for the oxen over the winter, it was said that both
possibilities existed.

It was suggested that one of the principal requisites for
planting forages is a favourable land tenure system. A team
was said to be looking at ways in which local people can have
title to their land. Government policy was that if you are a
wage earner you are not entitled to land. In resettlement
areas if you have an urban source of income you are not
entitled to resettled land.

b) Cattle Control

It was said that one of the major problems in Zimbabwe with
regard to fodder was that there were too many cattle and that
people should be educated to sell excess animals (e.g. old
oxen and cows). But farmers want to keep large numbers of
animals and there is a resistance to sell even in times of
drought. Asked how one could meet this resistance and educate
farmers to sell excess stock, Mr. Mudimu indicated that they
were thinking in terms of fencing off the grazing areas.
Government is encouraging the formation of grazing committees
and hoping that these committees will come up with a solution.
The trouble was that farmers were not concerned with quality,
they were only interested in numbers.

Also there is very little incentive to manage the range
because it is communal. Overgrazing is occurring mainly
because the management aspect is missing. Extension
programmes have tended to neglect the livestock aspect.

It was asked how far people in Zimbabwe had got in thinking in
terms of ideal herd structures (number of breeding cows
related to the rest of the herd), and what a reasonable unit

herd would be acceptable. This had not happened in the
communal areas, but in the resettlement areas beef models had
been developed for each agro-ecological zone. Projections had
been made up to 30 years. Five livestock units per homestead
was being used as a standard in high rainfall areas and up to
25 in low rainfall areas.

c) Useof Cows for Drauht

Since the calving interval was 3 years could not cows be used
for draught in the interim, it was asked. The recommendation
was said to be that a farmer does not use a cow for draught
when it is in calf. This decreases its productivity as her
system is being disturbed.

Asked if there was any interest from animal production people
in terms of using cows for draught and how this might affect
the reproductive system, the answer was given that the animal
production people are not interested in this. All they look
for from females is a calf. If there is a shortage of draught
animals they suggest building up the herd.

.KENYA .LLh. otgentia.L)

Refer to paper #8 by H Muiruri

The Aqroclimatic Environment

The average annual rainfall is 990 mm and occurs in three peaks.
The soils are mainly sandy loams and slightly acidic. The
average altitude is 2,000 meters above sea level with a monthly
mean temperature of 18 degrees celsius.

Mono-Modal: Hiqh Potential

Liveweight change

DM production

Dry Matter

J I N I\ 11 J J A U N I)

Rainfall (700 2 100mm):

Land Preparation


Rain in every month



Problem Points

-. "0.m

-.4 "

Most farms are 1-2 hectares and owners have registered title.
The main crops are maize, inter-cropped with beans and peas and
potatoes and vegetables. Land preparation is done during the dry
period of January to March by tractors mostly. Generally no
harrowing is done. Planting occurs with the onset of rains,
which come at the end of March to mid-June. Most farmers plant
maize and grow only one crop per year but during the short rains
in October/November, they may grow a second crop of beans and
potatoes. Weeding is done by family labour in May and June.
This is a job for women and children and as children are back at
school at this time there is generally a labour shortage for this

Livestock are kept for dairy production not for draught. Dairy
cattle are grazed on vacant plots, on public land and along
roadsides. There is very little potential for animal traction as
farmers will not use their expensive animals for draught. They
will continue to rely on tractor hire for land preparation.


The main problem with cattle is poor nutrition during the dry
period. Farmers have not set aside any of their land area for
cattle. The vacant plot areas will reduce as more settlement
takes place and roadside grazing is to be banned. Farmers will
then have to set aside areas for fodder crops and will have to
practise supplementary feeding.

At the moment maize stovers are used as supplementary feeds in
the dry periods. These are of low nutritional value though.
Drought resistant fodder crops will have to be grown to provide
feed in the dry months.


1. High Potential Areas

Asked to elaborate on the use of crop by-products, Mr. Muiruri
said that the main by-product fed was maize stover. Some farmers
are using maize cobs and had rejected grain. Asked why bean hay
was not used as a feed, it was said that beans and peas are,
harvested around July/August and at this time there is a lot of
forage in residues and empty plots. There is no conservation of
these residues.

Asked whether anyone had tried making silage, it was said that
this had been tried but there had been problems. There was no
motorisation and the problem of the container to make it in had
not been overcome. Oil drums could be used but are costly and
difficult to obtain. Sugar bags are cheaper but do not hold very
much. Another problem was being able to chop the material finely

It was explained that after harvest the whole area becomes
communal and this is why farmers do not grow short maturing crops
because they would leave the land empty earlier.

Asked about the annual income of the average farmer in view of
the high prices paid for weeding labour, Mr. Muiruri said that
sources of income are both on-farm and off-farm. In most cases
the value of the crop harvested would be $300 per hectare. There
would also be sales of milk.

On the question of the potential use of donkeys for draught, it
was said that donkeys were very important for transport of
produce and of water. Most families will keep 1 or 2 donkeys,
but they are not used for cultivation.

It was said to be important that interventions to improve cattle
feed in the high potential areas did not reduce yields on arable
lands. Three possibilities were mentioned:

a) Intercrop with pigeon pea which can be cut off and will keep
on growing and maize will grow around them.

b) Try to plant forages along fenced roads. All animals are
bomaed at night so the fodder could be fed in the bomas, but
this depends on labour availbalility.

c) Utilise the by-products of large plantations nearby of sugar,
rice and pyrethrum. Pyrethrum is available cheaply and is a
high quality feed.

2. Medium Potential Areas

Discussion also covered problems in the medium potential areas
which were rather different to those in the high potential

In the medium potential areas feed is distributed more evenly
through the year on account of the sharper bimodality in
rainfall. Here cattle are kept for draught and there is no
arable land constraint since no more than 65% of farm areas are
put under cultivation. The major problem here is overstocking.
In some cases there is no alternative to destocking. Farmers do
not like to sell their animals because they are a form of
investment and security against crop failures.

Natural pastures are an important feed source. Bush clearing,
burning and re-seeding could increase the feed available from
this source. However if destocking is not possible there is a
limit to what can be done to improve natural pastures and more
emphasis should be put on using crop residues more effectively by
better collection methods and better storage. At the moment most
farmers graze their fodder crops in situ and lose 40% of their
nutrient value.

Land is available for growing fodder crops in these areas.
Legumes and drought resistant fodder crops need to be evaluated.

The point was put that in the majority of situations in this part
of Africa people have a tradition of extensive animal management
systems in which low labour inputs are needed. At what point, it


was asked, do we anticipate that people will accept the concept.
of investing more labour time into animals and when does it
become economic to do so.

Dr. Tessema answered that if we really want to integrate
livestock into the production system we must look at the feed
resources available from the various elements of the system.
Fodder production and conservation is labour intensive but it can
also provide a high output. Cattle can utilise high quality feed
and give sufficient return to utilise the land more efficiently
than simply letting it into natural pasture. Some of the very
small farms will not be able to do this. But we have projected
that, for farms of 7-10 hectares, this will be possible. If you
work out the flow of labour use, there are only a few times when
there is competition between labour in terms of fodder. At these
times you can hire labour from the surrounding area. There is a
market for milk and production can be intensified to the extent
that the farmer becomes a commercial producer.

Bi-Modal: Medium Potential

Use of cultivated

Use of Crop /Use of crop
Residues/ Residues
Liveweight change .-- -. Residues

DM production

J F M A M J J A S 0 N D

Rainfall (500 700mm) -----_

Land Preparation -


Werv Ii []I-


Problem Points


Refer to paper #9 by D. Singa

The Agroclimatic Environment

In Malawi the inter-tropical convergence zone, the rain-bearing
equitorial low pressure zone, and the North-east Monsoon are
experienced during summer. Trade winds blowing into and from the
Indian Ocean, often bringing with them considerable moisture, are
experienced during winter months. The seasons commonly fall under
"wet" (November to March) and "dry" (April to October). Mean
annual rainfall ranges from 1000 to 3000 mm, depending on


Cattle Liveweiqht


\ ~ N

J F M A M J J A S 0 N D

Land Preparation





The principal food crop, maize, is grown all over the country and
covers 78%7. of the cultivated area. Of the cash .crops,
smallholders grow dark fire-cured, sun air-cured and oriental
tobacco and confectionary groundnuts (Chalimbana). Cotton is
grown in the lower areas of the country.

Almost all smallholder cultivation is done by hand, using a
tanged hoe set in a wooden handle. In addition to hoes,
smallholders use axes, pangas, slashers, ox-drawn implements
(ploughs and ridgers), water cans and sprayers. There has been a
tremendous increase in the use of machinery since 1969. The same
applies to work oxen, ox carts, ploughs and sprayers.

Problems and Research Work

One of the major constraints faced by the small farmer in Malawi
is labour shortage at peak periods. The country has a unimodal
rainfall and timeliness of operation is very important. Late o
inadequate land preparation with accompanying late planting and
weeding difficulties have usually resulted in low yields.

Research on reducing labour demand has included:

a) Animal power: Improvements are being made in breeding,
nutrition, management, harnessing and the use of oxen for the
majority of smallholder farmers.

b) Animal drawn equipment: Over the past year a multipurpose
animal drawn toolframe, which is capable of accepting a wide
range of attachments has been developed, tested, recommended
and is beinq manufactured locally. Attachments current
available are plough, ridger and groundnut lifter. Work o
planter, cultivating tines, weeding sweeps and clod-crusher i.

Oxcarts have been developed and are used more for transportin
farm produce to the market, fertilizer from the market t
farm, manure from farm to field, firewood from field to barn
and for other forms of transport.

c) Weedinq trials investigating alternative cultivation system
and subsequent weeding methods using ox-drawn equipment 0
maize and groundnuts are being conducted.

d) Hand tools are being developed for maize shelling, groundnu
shelling and rice threshing and wheat harvesting.


Discussion focused on types of animal draught.

There was said to be no experience of cows being used for draugh
in Malawi. Horses were not used for cultivation not so muc
because of their size but because of the shape of the horse. Th
angle between the horse and the ground is more .obtuse and mor,
force is therefore needed to plough. Implements have to b!
different. The advantages to the introduction of buffaloes wouli


be that they are more powerful than oxen, can work longer, have a
high resistance to disease and can live on material that oxen
would not eat.

Since an animal is said to be able to pull 10%7.-147. of its weight,
the heavier the animal the better. Because of its shape the horse
has certain advantages and it was asked whether anyone had tried
using Zebra which have the same shape as a horse. It was said
that Zebra are difficult to work with, maybe Zebra crossed with
horses would be possible, but not straight Zebra.

The advantage of Malawi Zebu over Zebu cross Fresians is that
they are more disease resistant. Trials are ongoing to
investigate the best way to feed the Zebu to get more work out of
it. Comparisons of different legumes and varieties of grasses are
being made. Reference is made to the paper, pages 15-17.


Mr. M. Matli gave the following presentation in lieu of
submitted paper.

Farming systems started in 1979 and focused on creating
productive agricultural enterprises, looking at mechanisms which
are acceptable to farmers and to their management skills
appropriate to their available resources and the communal 1an
tenure system. There are agronomists and specialists in rang
livestock, farm management, marketing, extension and sociology
operating in three prototype areas:

Mountains Molumong
Lowlands Siloe

L E S 0 T H O

Feed quality

Livestock weight


Land Preparation

I' I nL ing

Wpedi n()

J F M A M J J A S 0 N D

. MM

-NUM. doa

- Wo -



In 1980 farming systems conducted a survey and identified three

1. Rich farmers
2. Middle farmers
3. Poor farmers (see Table 1)


Characteristics of

Poor, Middle and Rich Households,

HOUSEHOLD TYPE BY Siloe2 Nyakasoba Molumong

Poor Middle Rich Poor Middle


Poor Middle Rich

Percent of Total Households
Average Number of Fields per
Average Number of Hectares per
Average Number of Adult Labourers
per Household
Average Number of Absentee Adult
Hale Labourers per Household
Average Number of Cattle managed
by Household
Percent of Household managing
at least 4 Cattle

45.0 35.0 18.3 60.7 32.1

7.1 17.2 51.7 31.0

2.0 2.4 3.3 1.9 2.3 2.5 2.0 2.1 3.7

2.7 3.2 4.7 1.1 2.0 1.6 1.8 2.1 3.8

1.6 2.5 2.7 1.4 2.0 2.5 1.2 1.9 3.2

0.6 0.9 0.4 0.6 0.7 0.7 0.6 0.7 0.4

0.6 3.8 2..0 1.7 4.9 8.5 1.2 11.1 27.1

3.7 47.6 81.8 11.8 66.7 100.0 20.0 73.3 100.0

'The households were typed on the basis of ownership of ox-drawn ploughs, planters and cultivators
as follows:

'Poor' households own none of the above equipment
"Middle' households own a plough only
and "Rich' households own a plough and a planter and/or cultivator

2One household in Siloe owned a cultivator, but not a plough or a planter and was excluded from this analysis

The number of adult labourers per household (as defined in Section II) also increases from the 'poor' to the
'middle' to the "rich' households. "Rich' households average at least one adult labourer per household more
than the 'poor' households. However, the number of hectares per adult labourer per household are not all
that different between the "poor" and the 'rich' households.

The Problem

The survey was based on the ownership of ox-drawn implements as
well as of cattle. In Lesotho all livestock (cattle, sheep,
goats for mohair) is dependent for a very long time on the
veld. Lesotho is also a horse-riding nation and donkeys and
asses are also kept. There are excess animals and much
over-grazing which, together with soil erosion, has led to less

food for livestock. This lack of food together with cold winters
results in very weak cattle in spring at the start of ploughing.

Research pp r coach

The farming systems section was developed to combat this lack of

(a) RaD9_e Improvement: The range section conducts trials and
looks at enclosures. Fodder and food have to be used for
more than a year. They are also conducting trials on the
range that is infested with weed. Four treatments have been
tried: hand pulling, spraying with herbicides, arresting
and burning. Results indicate that burning has been the
most effective treatment but as far as increase in feed is
concerned it has been the lowest. They have also been doing
work on perennial forages, looking at digitaria, trifolia
species. Because of drought, germination was not good but
what they got indicated that if things had gone well
production would have been good.

(b) Livestock Feedi nq: (Refer papers # 11 and 12) Here we have
included the growing of fodders and the use of concentrates.
Concentrates have not had much emphasis because concentrates
would mean spending money and not many farmers can afford
that. Selective feeding is practised. Farmers are
encouraged to feed draught animals in April or May so that
they can sustain weight in satisfactory conditions until the
grass starts coming up after the onset of the rains.

(c) Tillaqge: The agronomy section is looking at minimum
tillage, hoping that it will help conserve moisture and
reduce the need for draught power. They are also doing
winter ploughing and trying to encourage farmers to reduce
the number of animals that trample their fields during

(d) Fodder Production and Crop Residues: Winter fodders being
tested are: triticale, barley, vetch, turnips and kale. We
are aiming for maximum utilisation of all crop residues.
Farmers do harvest crop residues but what usually happens is
that due to labour shortage some farmers do not take all of
it home. We should try and encourage farmers to harvest all
crop residues and educate them to realise how much feed
animals require.

(e) Ma.r:et.inq section is investigating markets for excessive
livestock. It is also attempting to find out whether
livestock farmers can buy feed for their animals.

(f) Extension and sociology sections are trying to mobilise
farmers into accepting the Ministry's recommendations of
winter ploughing, rotational grazing and livestock
reducti ons.

(g) Livestock Control: The Ministry of Agriculture is to impose
charges on livestock imported into the country. RIO per

head for cattle and R5 for sheep and goats. Even so, people
still want to import cattle.

Di scussi on

Asked how minimum tillage was done, how it conserved moisture and
how animals could be prevented from trampling fields, it was said
that instead of a plough a cultivator only was used. This
conserved moisture because the soil was not being turned, but
only the surface scratched. To reduce the number of animals that
trample the land, you cut and stook stover and then winter plough
instead of allowing the animals to trample the land.

It was pointed out that there was a high differentiation between
the three geographical areas in terms of herd size. As the
altitude increases, one finds more livestock. Generally in the
low lands there are less cattle, more in the foothills and even
more in the mountains.

Asked if one wanted to start feeding in April/May what farmers
would have available as feed, Mr. Matli replied that crop
residues are often kept from the previous year. Japanese radish,
barley and oats can be grown as winter fodder. People rely
heavily on the veld. In answer to the question whether farmers
in the mountains were growing barley and oats, it was stated that
this was in the low lands and the foothills. Nothing was grown
in the mountains in winter. Trials have been conducted in the
mountains, but the crops are not frost resistant. Nothing
survives in the mountains in winter. In the low lands, both
winter and summer crops are planted. In the mountains only
summer crops are grown.

On ground preparation and planting it was explained that ground
preparation may be started in May because winter ploughing is
practised. Actual land preparation, if winter ploughing is not
done, starts late July, August and September but planting is not
done until after September because of the danger of frost.

Asked how the population reacted to winter ploughing, Mr. Matli
said that it was a question of educating the farmers. It was
pointed out that in Botswana the ground was too hard until after
the onset of the rains.

It was suggested that it would be better to sell younger animals
which would have a better value than to encourage the sale of old
stock to reduce numbers.

It was thought that it was very difficult to tell farmers to sell
their animals. They prefer to sell their animals when they are
old and believe in having a lot of animals for prestige. We
should try to improve the condition of the animals which the
farmer wants to sell. Farmers should be encouraged to stall feed
animals for two or three months. An animal will gain about 100
lbs as a result of stall feeding whereas without stall feeding it
would be an inferior animal.

Another participant agreed that farmers resist sell ing any

animal. We are trying to make them see that old animals are no
longer productive. At least you can get some money for an old
animal and relieve pressure on grazing. We must change the
situation where people are not livestock-oriented and they must
also breed *for the market so that they also sell young animals.

It was also noted that the cow will always be kept, regardless of
age, because of the possibility of another calf. They disregard
the fact that she is infertile once she has lost teeth.


Refer to paper #10 by K. Ramanaiah.


Dr. Ramanaiah described some of the elaborate strategies that had
been adopted by Indian farmers to improve animal nutrition and
increase efficiency and speed of crop cultivation using animal

In India 90%. of farm power comes from animal traction.
Preparatory cultivations, weeding, harvesting, threshing, lifting
water from wells and transport are all operations performed with
animal draught. In addition to being used for draught, buffaloes
and cattle produce large quantities of milk. Cattle are not kept
for beef production by the small farmer.

Forane Production

Various means of fodder production being used by the Indian
farmer were described:

1. Croq residues

In general crop residues play a very important role and
Indian farmers give as much importance to the crop residues
as they give to the grain.

a) The use of groundnut haulms is common and the method of
storing and feeding is elaborate. The haulms are stored
mixed with roughage in such a way that the animals get the
required proportions of pulse and roughage when they

b) Cereal straws are cut into small pieces to increase
efficiency of utilisation and generally a variety of mixed
roughages are fed. Quality of roughages are improved by
adding mineral salts, urea molasses etc.

2. Forage Crop Production

This is achieved either through sole cropping or by
intercropping with food crops. Elaborate rotations and
intercropping mixes were described.

3. Foraqe Trees

Leaves of trees and shrubs are an important source of fodder
especially during drought periods. Camels, goats and sheep
derive the major portion of their feed from the leaves and
pods which are either browsed or cut and fed.

81EETI "m -InLAI

Experimental Progirammes


4. Pasture Grass

Improvements to natural grasslands include soil and water
conservation measures, reseeding and legume introduction.

Improved Efficiency_ of Animal Traction

Increased efficiency of animal traction is brought about by (a)
better training of the animals and (b) by improving the
agricultural implements.

Farmers' implements are locally made and there has been
continuous improvement and innovation over time. Local village
blacksmiths do all the manufacture and maintainance and it is
rare for farmers to depend on factory made implements. Examples
of planters (using bamboo tubes), ploughs, cultivators, riders
were shown.


Paper # 12 entitled "Value of a Winter Lick Fed to Cattle Under
Lesotho Farmer Conditions" was presented by Ms. Motjope.


Description of Ex.periments

This paper described a series of experiments. that were conducted
to encourage and enable farmers to improve their livestock
feeding methods, especially those animals used for draught

The objectives of the experiments were:

a) To compare the performance of oxen fed in pens during the
winter period to those fed and managed in the usual manner
by village farmers.

b) To determine the value of a supplemental protein and
mineral lick when fed with each of the above two systems.

c) To enable farmers to improve their total oxen feeding and
management methods.

Two of each farmer's cattle were to be fed at the research pens
and two at home in his usual manner. Rumevite Cattle Block was
the supplemental lick selected for use in the experiments. It was
fed to half the cattle at the research pens and half of those at
the home farm.


Due to an inadequate supply of fodder, farmers were erratic in
taking their animals to the research pens every night. Thus only
the Rumevite comparison under home conditions was considered
valid. The results obtained indicate no consistent differences
among replicates in response to either method of feeding with or
without access to Rumevite.

Supplementary Information Presented

Data on further feeding experiment using hominy chop as an
energy supplement were presented and the results are summarised
in the following Table.

gxPected Versus Observed
Performance of Hominy Chop


1.5 kg Hominy Chop fed daily = 2.325 Mcal NEg
= 0.55 kg daily weight gain
Average Gain (k.-g)
Fed Control Difference
11/7 to 28/4 (44 days)
Observed +2.5 -6.9 9.4
Expected (44 x 0.55) 24.2

24/8 to 21/9 (28 days)
Observed -3.7 -2.3 1.4
Expected (28 x 0.55) 15.4

The large discrepancies between the expected and observed weight
gains were said to show that farmers had not been feeding the
hominy supplement as instructed. Also it was stated that the
animals that had been fed the hominy supplement were worked
harder and this may be another reason why observed differences in
weight gain were so much lower than expected.


The discussion focused on:

a) Other experimental work on animal nutrition in Lesotho
b) Protein content of the supplementary feed
c) Clarifications on the hominy chop treatment

a) Other Nutritional Work

The rumevite supplementation experiments showed clearly that
since there is a shortage of energy in Lesotho, there is no
point in feeding protein to cattle. Thus hominy chop is now
being purchased. Experiments are now being conducted on
feeding hominy chop

a) to sheep at critical times such as breeding and lambing
b) to cows in milk production
c) to selected young oxen

Other nutrition interventions being examined include beef
cattle improvement through performance testing and controlled
grazing for sheep and goats.

b) Protein Content of Supplementary Feed

The supplementary feeding experiment seemed to be using high
levels of protein. Figures of 27.57. protein and 77. urea were

quoted. It was explained that the supplement was only a lick
and animals only ate about 25 gm of the lick per day.

c) Hominy Chop Treatment

While there was a problem with farmers taking fodder to feed
cattle in pens, the hominy chop was supplied free to farmers
where they needed it so it was more readily given to the
cattle. It was said to be not so much the time involved in
cutting and carting fodder that resulted in small amounts
being fed, but that farmers would bring only one bag and
expect it to last a week.

It was explained that hominy chop was maize bran, similar to
wheat bran. It was the hull part of the maize and included
the germ. About 25% of the harvested maize was said to end up
as hominy chop. It contained about 7-8% protein and was 65%
digestible. It currently costs 17c/kg and comes in 50-100kg


Paper # 13 entitled "Draught Power in Botswana: the design and
testing of solutions" was presented by Mr. C. R. Riches


Types of Trials and Some Results

Three types of trials have been conducted;

a) Increasing Efficiency of Animal Drauqht

The possibility of reducing the draught team from six to four
cattle for primary tillage was investigated by the
introduction of two multi-purpose wheeled toolcarriers the
Versatool and the Makgonatsotlhe. Moulboard ploughing was
replaced by chisel tines or Texas Sweeps. However the size
and condition of the cattle did not enable the draught team
size to be reduced. In addition the need for more than one
pass with tines and weed problems led participating farmers
to revert to the plough.

The Sebele single row planter used with a Mahon cultivator
required one extra planting operation, but needed less labour
and draught than broadcasting on a whole farm basis because
of the smaller areas needed to be planted for subsistence.

For those not wishing to or being able to make more than one
pass the plough planter was developed and tested. With this
equipment seed is dropped into the furrow slice to be covered
by the subsequent bout of ploughing. Trials indicate that
fewer animal hours at equivalent team sizes are required to
establish sufficient area for subsistence by plough planting
than by traditional plough and broadcast methods.

b) Substitution of Animal Drauqht by Tractor Power

The plough planter was modified to fit a tractor plough. More
on-farm testing is needed to establish reliability.

Partial substitution by tractor power may be achieved with
the use of a single row ox planter and cultivator. Sorghum
stands planted by single row planters have given mean
production levels of 700 kg/ha in on-farm trials over 4
years. The farmer would need to plant 3.8 hectares at this
yeild level to ensure family subsistence and cover all cost.
Since 700 kg/ha may be difficult to maintain on a whole farm
basis the system may only become applicable on larger areas.

c) ImoYing fraught Animal Ni.trition

From the limited data available from trials to date, it would
appear that even 1/2 hectare of late sown Babala and 1/2
hectare early planted Dolichos could each produce 600 kg dry
matter, sufficient for supplementary feeding to six 350 kg
animals for 50 days on a 50:50 cereal/legume ration. This is

applicable to farmers with relatively high resource levels.
For other farmers emphasis would be placed on use of crop
residues rather than grain fodders.


Questions on the working of the planter and plough planter were
answered by saying that the planter was a single row machine and
the land did not have to be harrowed before using it. The planter
was designed for sorghum, millet and cow peas. It works less well
with large seeded crops such as maize but a drop planter for
maize is now being developed. The two furrow plough planter has
been mounted with a fertilizer applicator, with the single furrow
version addition of a fertilizer applicator makes it difficult to
balance. On the question of the availability of rural blacksmiths
for maintaining the ploughing and planting machinery, it was said
that there was a shortage of these skills but that people were
being trained.

Questions on fodder related to the possible difficulty of getting
the roots out of the ground with sorghum fodder and whether
animals were fed in a closed place in winter. No problem was
reported with disposing of the sorghum fodder. It was either cut
and carried or grazed off. One of the advantages of supplementary
feeding was said to be that it made the animals more manageable
because they willingly came to the place where they were being
fed. Feeding took place in the open.


Paper # 14 entitled "Ground Preparation and Planting: a key
constraint for ox cultivators in Zambia" was presented by Mr. T.


The Constraint

Maize is both the staple food and the main cash crop for the area
and there are acute difficulties in getting sufficient areas
planted near the beginning of the rainy season, before the rapid
drop in yield potential which occurs at a rate of 27. per day
through December. Maize planting often continues well into

Technologies to Reduce the Constraint

a) Breeding varieties with shorter maturation periods

b) Improving the ability of the soil to provide sufficient
water by improved soil management (e.g. winter ploughing),
or by changes in the planting technique (e.g. use of a

c) Increasing the rate of planting


Winter ploughing has been widely promoted over a number of years,
but is rarely practised by farmers in the area. Even in systems
where more land is left fallow and which could be winter
ploughed, it is the exception rather than the rule, even though
most farmers have tried it on previous occasions and are aware of
any benefits.

Several planting methods were tested against current farmer
practice during the last growing season. The very poor
performance of all planters was further compounded by the very
dry planting conditions. A trial in the 1981/82 season showed an
average of 243 manhours if the recommended practice of hand
planting with a rope following ox cultivation was used, compared
with 118 manhours using farmer practice.

Further reduction in tillage is likely to involve the use of
herbicides. On-farm trials over four seasons tested herbicides in
conjunction with reduced tillage methods. The package was shown
to work technically where the pre-emergent herbicide was
synchronised with rainfall and prior to weed emergence.
Non-adoption by target farmers was perhaps due to this timing
conflicting with the farmers' desire to plant only after a
certain probability of sustained soil moisture had been reached,
which usually occurs after significant number of weeds are beyond
the susceptible stage, and conventional ground preparation is
necessary. Thus the key constraint is no longer addressed.


Under good conditions, the farmer's method of third furrow
planting achieves an adequate stand and is rapid, simple and
cheap. The advantage of alternative methods will depend on the
number of extra days he can plant (or the increased area per
day), and how much sooner this allows him to plant his required
area. This is difficult to determine, even when a lot of data is

There were no questions on this presentation


Paper # 15 entitled "On-Farm Minimum Tillage Experimentsl a
possible solution to draught power problems gripping Zimbabwe's
communal areas" was presented by Mr. E. Shumba.


Ongoing Trials

This paper discusses on-farm trials aimed at improving the
performance of the available draught power pool through minimum
tillage techniques. In Mangwende the use of a ripper tine in
conjunction with chemical weed control is being investigated,
while in Chibi South planting directly into winter ploughed
fields in conjunction with varied hand weeding treatments is
being tried.

Results in Manqwende

Minimum tillage work using the ripper tine was initiated in
Mangwende in the 1982/83 season. The tine treatment outyielded
the conventional late plough treatment across sites irrespective
of whether the field had been winter ploughed or not. To combat
the problem of early weed infestation associated with minimum
tillage, a herbicide treatment was included in this trial.
Although the use of the herbicide gave an overall 11.77. yield
advantage over hand weeding, it resulted in lower yields at 2 out
of 5 sites.

Supplementary Information

In addition to the information presented in his paper, Mr. Shumba
elaborated on some of the pros and cons of minimum tillage using

Type of Herbicide

Preplanting posed a problem of incorporation which needed to be
in wet soil or followed by rain. Pre-emergence gave the problem
of spray direction if the plants had already emerged.
Post-emergence seemed best but there was the question of whether
Gardomol or Atrozine was best. It depended on the moisture,
method of application and tillage method.

Weed Spectrum and Persistence

A good knowledge of the weed spectrum in the area was needed.
Persistence may pose a problem in relation to a
maize-groundnuts-millet rotation. For example Atrozine on heavy
soils can kill groundnuts after 18 months. Faster leaching of
such chemicals is one advantage of sandy soils on communal areas
in Zimbabwe. Use of herbicides rules out intercropping of
forages for animal nutrition improvements.

Appropriateness of Herbicide re Economic Circumstances

The following comparison between Chibi South and Mangwende was

Chibi South Manqwende
with without with without
cattle cattle cattle cattle

Yield (t/ha) 1.4 0.8 3.2 2.1
Income ($) 241 105 752 449
%. of Income
off farm 50 60 20 30

It was said that herbicides would not be so appropriate for Chibi
South as Mangwende because:

a) Yields were lower in Chibi South so other factors were more

b) Incomes were lower in Chibi South, so the ability to purchase
herbicides was less

c) The proportion of income earned off the farm was greater in
Chibi South so there was less incentive for farmers here to
purchase herbicides.

Labour Saving Advantages

While there may be no yield advantage from using herbicides, it
may allow labour to be released for other activities, especially
gardening which provides the second greatest cash income
throughout the year.

Mangwende ($)
with without
cattle cattle
Maize Sales 347 168
Vegetable sales 130 84
Groundnut sales 40 26
Sunflower sales 11 3

Other Problems with Minimum Tillage

a) Manure Incorporation

Chibi South Mangwende
with without with without
cattle cattle cattle cattle
7. applying
manure 74 12 52 12

As the table above indicates, many farmers with cattle
incorporate manure. However, this has only a 40-50% yield
advantage in the first year and minimum tillage should be
followed by conventional tillage every third year.

b) Pests and diseases

These may be more of a problem as a result of overwintering
in crop residues. Possible solutions are (i) removal and
storage, (ii) grazing in situ and (iii) burn ungrazed

c) Soil Erosion

This may be aggravated if residues are removed. Minimum
tillage should therefore be practiced in conjunction with
soil conservation measures, e.g. mulching, inter-row
cultivations and contour ridging.


Discussion focused on the tine tillage method. It was suggested
that if the tine could be bolted onto the plough this would save
having to supply separate implements for farmers. For the
experiments only 5 tines were taken into the area but demand is
growing and more are being provided.

Asked why the use of the tine seemed to be popular in Zimbabwe
but had been rejected in Botswana it was said that different
implements were used in the two cases. Also soil differences
could be important. It was stated that there are advantages of
deep ploughing and it was confirmed that the ripper tine did tend
to go too deep at first. Adjustments were made to keep this from
happening. It was again stressed that conventional ploughing
should be done every three years.

It was confirmed that the tine was used to rip the planting row
only. Asked how weed control was effected without ploughing, it
was restated that this was why herbicides were being investigated
to be used in conjunction with the tine.


The paper entitled "Forage Legumes in the ILCA Highlands
Programme" was presented by Dr. S. Jutzi.


The Setting and Problems

Due to the high population pressure in the highlands, forage
production from permanent pasture is confined to shallow, stony
and steep upland soils and to the seasonally waterlogged and/or
flooded valley bottoms. Forage production from the intermediate
arable land is currently restricted to the provision of crop
residues (barley, wheat, teff, oat, pulses) and in some areas to
a rather scarce spontaneous fallow land vegetation. Crop
residues account for approximately 15% of the total annual feed

Conserved forage is mainly fed to draught oxen and lactating
cows. The permanent pasture areas are heavily and permanently
overgrazed by cattle.


The constraints to increasing animal productivity through
improving the feed supply are:

total feed supply (relative to livestock numbers)
seasonal feed supply
low feed quality
feeding management (no splitting of herd)
malnutrition, lack of nutrition knowledge
lack of information on feed sources

Guidelines for Forage Research

Based on the present situation of the smallholder and on the
production constraints detected a number of guidelines for forage
research activities were discussed. These included the

1) On-station research is the appropriate way to identify the
basic technical problems. On-farm research is the best way
to commit researchers and extensionalists to the generation
of appropriate and accessible technologies.

2) Forage legumes are to play a key role not only in forage
production but also in food crop systems

3) Exotic pasture germplasm is to be tested against best
native germplasm

4) Any attempt to extend pasture growth into the dry season
will alleviate feed shortage in this critical time.

5) N-fertiliser is not a justified input in pasture production
of the traditional smallholder.

6) No pasture development without powerful back-up by herbage
seed production.

7) Pasture legumes are the basis for high quality forage
supply and for soil fertility improvement strategies.
Pasture grasses are the basis for the realisation of the
quantitative needs in animal feed utilizablee biomass).


Asked what recommendations of ILCA have been adopted by Ethiopian
farmers, Dr. Jutzi replied that only oat rich mixtures for dairy
cows had been adopted on any scale.

Screening for legumes was done on the basis of efficiency of
nodulation of native species. Application of phosphorus
benefited both leaves and nodules. Nodule weight decreased
slightly by the application of nitrogen. (Refer Figures 1 and

The problem of supply of hybrid seed was raised. At the moment
all legume seed needed in Botswana has to be imported. Asked
whether it may be possible for certain countries to be bulk
providers of certain species, Dr. Jutzi answered that support
services and training could be organised on a regional basis but
that commercial seed production needed to be done by companies or
farmers within specific countries.

The intersowing of legumes and the planting of hazard resistant
species in communal pastures was not a major problem in terms of
community involvemnet because the Ethiopian farmers were
organised into co-operatives and one dealt with the co-operatives
not individual farmers.



a) Leaves per plant and nodules
per plant.


0 -i


1. Control
2. 9 kg/ha N + 23 kg/ha P205
3. 27 kg/ha N + 69 kg/ha P205
4. 23 kg/ha P205
5. 69 kg/ha P205

(sample size: 60 plts / treatment)





Dry Matter in


b) Dry matter in nodules,
roots and leaves/steins.

1. Control
2. 9 kg/ha N+23 kg/ha P205
3. 27 kg/ha N+69 kg/ha P205
4. 23 kg/ha P205
5. 69 kg/ha P205.

Leaves +






Paper # 17 entitled "The Grazing Animal and Land Pressures in
Small Farm Systemss must ruminant nutrition create inevitably a
confrontation between arable land and grazing requirements?" was
presented by Mr. D. Chandler.



In Botswana yields from crops are very low (average 210 kg/ha)
and risks are high. Yet 84,000 farm-households cultivate 300,000
hectares. Mixed farming is carried out by 68% of these
households, 18% have livestock only and 14% raise crops only.
There is said to be a biomass limitation to support acceptable
levels of production from livestock, but the extent to which this
is due to poor nutrition, bad management or overstocking is

Research Orientation

Herd structures need to be improved in order to provide better
draught. The only way to do this is to improve the productive
performance of the breeding female and subsequently improve the
growth rates of the young animal, whilst at the same time
reducing mortality to a minumum. Comparisons of productivity on
the Animal Production Research Unit ranches and communal areas
gives some indication of the improvements that can be made.

Productivity Levels
A.P.R.U Communal

Calving % 75 46-27
Mortality % 6-8 12-20
Weaning wt (kg) 175 125-80
8 month wt (kg) 275 200-140
Productivity per cow
at weaning (kg) 120 50
at 18 months (kg) 188 85

The only criterion that will make sense under the current
circumstances is the intensification of fodder production from a
unit area that will be of the magnitude to provide not only very
significant increases in total bulk of dry matter available to
the grazing animal, but also of such a quality that digestibility
and available protein levels can be expected to affect productive

Improving Biomass Supply

Various methods of improving biomass supply were discussed.
These included range regeneration, planted grasses, improved crop

production and forage legume production (intercropped, direct
planted, mixed sowing, undersowing, annuals or perennials).
Details were given of experiments on a number of these approaches
conducted at the Integrated Farming Pilot Project between 1979
and 1982. These included

a) direct and undersowing of a variety of grasses and legumes
in the 1979/80 and 1980/81 seasons

b) range reseeding in 1980/81

c) Dolichos Lablab trials, direct and undersowing of fodders
and range strip discing in 1981/82


It was said to be very difficult to get the Batswana farmer to
think of the problem in terms of draught power. The basic
problems were carrying capacity, range resource capacity,
stocking rate, grazing index and herbage availability.

At IFPP, researchers have never yet attempted to grow fodder
without basal phosphate dressing.

On the question of perennials it was said that Siratro provided
yields of 4000 to 5000 kg/ha of dry matter per year with never
less than 9% protein. It was best planted with E. Teff. There
was said to be no point in trying to establish a grass without a


Paper # 18 entitled "Feed Production Research for Smallholder
Agriculturalists in Western Kenya" was presented by Dr. S.

The Problem

Due to high population pressures availability of communal pasture
land is decreasing and the major constraint to livestock
productivity is inadequate nutrition and large ruminants are
going out of the system. There also exists a severe shortage of
family labour time, since most of the males migrate to off-farm
employment and the women left to take charge of the farm have
many non-farm responsibilities to see to.

Appropriate Feed Resource Production Options

Given the nature of the problem the feed resource production
options examined centered around the integration of dual purpose
goat production into the current system. Goats are more suitable
than cows in the circumstances due to their greater versatility
and are able to provide meat and milk products. Research was
aimed at providing sufficient quality and quantity of feed at
appropriate times during the production cycle so that goats could
produce milk and meat.

Interventions Tested

Non-successful interventions included:

a) Silage making
b) Intercropping forage legumes into maize and beans
c) Napier grass as a fence-row cut-and-carry crop
d) Stall feeding

Silage making was too time intensive, legume intercrops were not
successful because not enough biomass was produced from the low
growing plants to justify harvesting for animal feed. Napier
grass was not cut-and-carried frequently enough and was labour
intensive. Stall feeding was unacceptable because of the high
refusal rate on the part of the goats and the labour intensity.

More successful interventions included:

a) Stripping and topping maize plants
b) Intercropping sorghum and early ratooning
c) Pigeon pea interplanted and planted in fence rows

Stripping and topping reduced yields by 25 and 15 percent
respectively. Stripping from 6 weeks after planting to tasseling

and from the milk stage on may not reduce yields. Pigeon pea
interplanted in the 1982 dry season did well if seed beds were
well prepared and kept weeded.


Discussion focused on the topics of pigeon pea, leaf stripping
and Leucaena.

Pigqeon Pea

Asked how pigeon pea could be used a fodder crop, it was said
that it could be used as supplement mix for livestock.

On the question whether pigeon pea would be suitable for Botswana
and whether varieties were available for trial, it was stated
that in Kenya there was great hope for pigeon pea. It normally
takes 7-8 months to mature but work is being done on a single
season crop that can be cut off after the short rains, but the
plants remain in the field, re-grow (ratoon) and can be harvested
a second or even third time. Pigeon pea will still be standing
green at the height of the dry season. The ability of pigeon pea
to withstand dry conditions was confirmed by experience from IFPP
in Botswana.

It was suggested that pigeon pea could be used successfully in
Botswana but that in frost zones it would not be so productive.
Early establishment before the frosts, was suggested as a way of
overcoming this problem.

On the availability of seed we heard that it is available in
Kenya and could be bought from farmers in Zimbabwe.

As an intercrop in Western Kenya pigeon pea was planted with
maize during the short rains and left when the maize was
harvested. In the long rains maize was planted in between.

Stalk borer was said to be a problem with pigeon pea, but the
variety used in Western Kenya (Kioko) was resistant.

Leaf Stripping

The answer given to the question of how many leaves were stripped
per plant was 5 leaves per plant over 100 metre row.


Leucaena was said to have the problem of tainting milk. Goats
were said to have been particularly susceptible to this problem in
the Phillipines. Some animals have material in their rumens that
will break down the toxic substance. This is related to
environment and genetics. In Malawi fairly high levels were said
to have been fed without toxic effects. Management and feeding
regimes were said to determine the toxicity. Leucaena would be
used as an appetiser allowing increased utilisation of crop
residues. It takes about 60 days for the rumen to adjust to the

The biggest problem in Botswana was said to be in establishing
the crop. It proved difficult to establish on the range.
Digitaria is more easily established. Siratra requires two
discings. Next season earlier establishment will be tried.


Paper # 19 entitled "Animal Feeding in Small Farm Systems" was
presented by Dr. Tessema.

Abstract and Supplementary Information

The Croppinqg System

Farm sizes range from 2.0 to 24.0 hectares. Even where farm
sizes are less than 5 hectares, no more than 56'/ of the land is
used for crop production. A considerable proportion of the farms
are managed by women (26-477.). The men are away in town and
cities for extended periods earning wages.

The main crops grown are maize, sorghum, millet, pigeon peas, cow
peas, and beans. Although manure is produced in bomas, only a
limited number of farmers use it on their fields; mainly because
of the difficulty in carrying and spreading it on fields.

Natural pastures are the primary source of feed for livestock.
Most farmers use maize stover (grazed in situ or cut and carried)
and pigeon pea threshings as supplementary feeds. The
cultivation of fodder is practised only to a very limited degree:
only 87. of farmers reported growing fodder.

Objectives of On-farm Trials

The aim of the on-farm trials in the livestock sub-system was to
bring together and test improved technologies in a range of
feeding methods that will strengthen the role of livestock in
small farm systems and increase production and profitability.
Description of trial work conducted out of the National Dryland
Farming Research Station at Katumani included the following.

a) Improved Productivity of Natural Pastures


(Katuman 1981/82)

1st Cut 1st Regrowth 2nd Regrowth 3rd Regrowth TOTAL
Date Kg/ha Date Kg/ha Date Kg/ha Date Kg/ha Kg/ha



670 7.3.82

149 3.6.82 1410 31.8.82 203

750 29.3.82 208 25.6.82 814 21.9.82 178

672 12.582 1030 8.8.82



326 3.11.81 155



189 2147

Kq/ ha

2,57. 85 64 102 24


FOR A PERIOD OF 323 DAYS (13.8.81 2.7.82)


Stocking Rate*
No. of Animals
Initial Weight (kg)
Final Weight (kg)





20 28
30 29

Average Daily
Gain (gms) 189 161 30 31 25 36

* Livestock Units per hectare: 1LU = 250kg






Stocking Average Daily Animal Days Liveweight
Rates gains/animal per hectare gain per
(Kg) hectare (kg)

.35 LU/ha

(a) Steers
(b) Sheep
(c) Goats


.54 LU/ha

(a) Steers
(b) Sheep
(c) Goats



x (.35 x 323)
>x (.35 x 323)
x (.35 x 323)

*x (.54 x 323)
x (.54 x, 323)
x (.54 x 323)

TOTAL 38.76





(2.8.82 4.7.83)


GRAZING MANAGEMENT* Continuous Rotational Continuous Rotational

Number of Animals 18 18 18 18
Initial Weight (kg) 17.2 17.2 15.7 15.7
Final Weight (kg) 32,7 32,2 30,6 30,0
Average Daily
Gain (gms) 46,2 44,7 44,1 42,8

* The stocking rate for both continuous and rotational
grazing was similar and was adjusted to .54 LU/ha

The results presented in the tables above were obtained on a
natural grazing land that has been bush cleared. Our
results show that, grazing a properly managed (bush cleared
only) natural pasture, livestock could maintain their weight
and make modest gains at a stocking rate of between 2 and 3
hectares per livestock unit. However farmers were very
unresponsive to advice given on bush clearing. Many were
unwilling to carry out the work because they say it is a
hard and difficult task even though it does not conflict
with other (farming) operations as it can be done in the dry
season when there is little other (farm) activity.

b) Improved Utilisation of Crop Residues by Physical
Treatment, Chemical Treatment and Supplementation




NDF% 85,8 71,2
ADF% 51,0 49,4
ADL% 5,6 5,6
CP% 3, 19 3,86
INDMD% 35,9 44,5


(July/October 1981)

Average DM intakes (gms/animal/day)

Chopped Maize Stover Chopped Maize Stover plus
ANIMALS as the only source of feed natural pasture grazing

Untreated Treated Untreated Treated

and 650a 674a 470a 715b

a,b means with different subscripts were different at PO.05 level


(July/October 1982)

Untreated Treated
Grazing Untreated Treated Stover Stover
Alone Stover Stover + +

No. of Animals 16 16 16 16 16
No. of Days on Trial 90 90 90 90 90
Initial Av. Wt. (kg) 24.0 24.0 23.9 23.4 24.2
Final Av. Wt. (kg) 18 25.7 25.5 25.3 26.2
Av. Daily Gain (gms) -22a 19ab 18ab 21ab 22ab
Av. DM intake/day
(gms) 271 420 516 730
Daily Feed Cost
(K.Sh.) .51 .36 .92
Feed Cost/kg gain
(K.Sh.) 28.33 17.14 41.82

*MUM = Molasses, urea and mineral mixture fed at 200gm/animal/day
NaOH = K.Sh.13 per kg
MUM = K.Sh. 1.80 per kg
a,b means with different subscripts were different at P0.05 level



Grazing Treated Treated Treated Untreated Maize
Alone Stover + + + Silage
MUM Leucaena* Leucaena

Number of
Animals 8 8 8 8 8 8
No. of days 112 112 112 112 112 112
Initial Av.
Weight (kg) 35.9 35.0 35.2 35.6 34.6 33.9
Final Av.
Weight (kg) 32.1 39.4 40.4 40.4 38.8 40.8
Av. Daily
Gain (gins) -34a 39b 46b 43b 38b 62ab
Av. DM
(gms) 210 290 250 350 580

* Leucaena leucocephala was fed as green chop and mixed with chopped
stover at 20% of dry matter intake

a,b means having different subscripts were different at PO.05 level



30% Leucaena 50% Leucaena
Sheep Goats Sheep Goats

Number of Animals 6 6 6 6
Number of Days 48 48 48 48
Initial Av. Weight (kg) 15.7 17.0 15.6 16.7
Final Av. Weight (kg) 16.0 14.3 17.0 14.6
Av. Daily Gain/Loss (gms) 11.07 -95* 46.8 -76.4
Av. DM intake/day (gms) 453 474 456 416
Leucaena 126 132 212 196
Bana 327 242 235 218

One goat in this group had severe diarrhoea for 8 days

Farmers have particularly shown interest in the use of leucaena
to supplement stovers and many of them have already accepted and
planted seedlings in their farms.

One major factor upon which the technology of improved
utilisation of crop resudues would hinge is that of the use of
some kind of a chopping implement that will reduce particle size
of the crop residues. A locally fabricated machine costs K.Sh.
1700.00, which is a very big investment for a small farmer.

c) Growing Pasture and Fodder Crops

Two of the major problems with getting the farmer to cultivate
pasture and fodder crops are land and labour scarcity.
Because land is scarce a farmer usually finds it difficult to
set part of his cultivated land, his labour and oxen time for
growing fodder and/or pasture crops until he is assured that
he can get two or three-fold increase over the crop production
that he gets from the same plot of land. Furthermore the
growing of forage and pasture grasses and their utilization
for animal feeding is a labour intensive technology that
creates peaked labour demands at food crop planting and
harvesting time, or creates peaked labour demands by
increasing regularity of operations.

d) Ieeping Dairy Cows to Utilise Cultivated Fodder Crops

Trials conducted at the station showed that, when fed at ad
libitum level, Bana grass can support up to 6.83 kg of milk
per day without recourse to concentrates. An increased milk
production of 31% was achieved with 1.49 kg dairy meal per day
which, at current prices, was not economical.

e) Efficiency of Draught Breeds

Our results show that, given the present cultivation
implements, there would be no advantage in using a cross-bred
oxen over indigenous (Zebu) oxen. Properly and adequately
supplemented indigenous oxen do perform just as well as the
bigger and hungrier cross-bred oxen.

Rate of Adoption of Innovations

The rate of adoption of the above innovations was disappointingly
very poor. It seemed that farmers valued their leisure more then
the gains they could get from clearing bush to encourage good
forage growth. While most farmers are grazing their crop
residues in situ and realise that they are wasting about 40% of
production in so doing, they still go ahead and practise it. It
seemed that, in terms of labour use, farmers choose the least
burdensome way of doing a job, even if they are aware that an
increased input will give a higher return. The growing of fodder
crops created greater demand for labour and oxen time, which the
farmer cannot cope with if he has to carry out operations of
ploughing, planting and weeding for food crop production. Only a
handful of farmers were able to be persuaded to include fodder
crops in their cropping system.

It is assumed that failure to develop integrated recommendations
for the whole farm system may perhaps have been the cause of low
adoption rates of the recommended practices. It has therefore

now been decided to carry out the pre-extension trials on a fewer
number of farms based on the whole farm approach. A well
balanced mixed farming farm plan is to be made specific to each
farm. It is intended that the station shall provide the basic
inputs necessary for implementing the recommended innovations.


Zebu vs Crossbred Oxen

Asked to elaborate on the preference for Zebu oxen, Dr. Tessema
said that with feed resources based on crop residues and minimum
supplementation of 700g of MUM, the Zebu were able to perform all
duties that were required for reasonable planting. That is, they
were able to go onto the land and plough at a reasonable depth
(12-13 cm) and within a time period that was acceptable (2.5 ha
in 8 days), and that there was no difference between the Zebu and
the cross-bred in these respects. However if you were to start
using implements that went deeper and required faster work, then
you would need to feed better and might even need to go to a
larger animal.

Dr Mosi noted that in Ethiopia their experiments had indicated
that under normal feeding conditions the bigger animal barely
performed at the same rate as the local animal, but when feeding
became better you could harness bigger implements to the bigger
animals. Mr. Singa noted that in Malawi implements had to be
modified a little to fit larger animals.

Asked whether any cattle were not supplemented with MUM in the
experiments comparing Zebu and cross-breds, Dr. Tessema answered
that all the oxen received the same supplementation and the
object of the experiment was to see whether they could all do a
reasonably good job of land preparation.

Measuring Draught Efficiency

The question was raised as to how one could measure cost
effectiveness of supplementing feed of draught animals under
communal conditions. It was stated that if you are late in
planting you lose 6% production per day. It was suggested that
the first thing to do was to ensure that the animal maintained
its weight. Under communal conditions it was thought that it was
impossible to record statistical results. Farm management
research was said to be missing; one needs to assess how
efficiently the animals are being used.


Dry Season Feed Resource Budgetary Problems
The Kenyan Experience and Efforts

Dr. Kayongo-Male gave the following presentation in lieu of a
submitted paper.

The Problem

Kenya is about 225 sq. miles in area, but only 15% of this area
lies within the high potential area. The human population is
approximately 17 million and the population growth is 4% per
annum. Kenya is therefore facing a very acute shortage of good
arable land. Within the high potential areas, the population
density is 500 people per hectare and because of this we estimate
plot holdings to run between 1 and 2 ha. per family of 8 12
people. Because of pressure on land agricultural production
increase is estimated to be below 17.% annually.

Kenya has about 10 million livestock units, mainly cattle, sheep
and goats. A big proportion of these animals live in high
potential areas and because of unfavourable land-to-man ratio
there is great competition between crop and livestock production.
It was shown that 75% of the land is given to crops and 25% is
given to livestock production. What has happened in Kenya as a
result of those unfavourable ratios is that a number of systems
have evolved such as zero grazing where forage is cut and taken
to the animal. This brings problems of variable quality and
quantity of feed given to the animal. A common figure quoted in
Kenya is that animals pick up about 80% of feed requirements and
this figure is much lower during the dry season.

Research Approach

We have taken three approaches to try and solve the feed shortage
problem. We have looked at conventional interventions, i.e.,
pasture improvement programme; crop by-products; and, of late,
we have moved into unconventional areas like maize stripping.

1. "Traditional" systems

For any animal production system to be successful a balanced
increase of feed resources and animals is very important.
The conventional systems of pasture improvement in terms of
grasses mixed with legumes; and in terms of fertilizer
application have been tried with the idea of increasing
pasture productivity. Fodder crops have been tried to be fed
as silage or green chop. We think this kind of system is
untenable to the small-scale farmer. With very small plots,
grazing will become a luxury. The technology involved in
making silage is not easily grasped by the small-scale
farmer. As inflation goes up we are going to find ourselves
producing grass at a very high cost. The cost of buying
seed, fertilizer and the cost of power are all becoming
prohibitively high. Grass is no longer the cheapest feed.

2. Crop Residue Feeding

The quality of agro-industrial by-products is very poor and
they are not very useful in terms of animal production as
shown in Table 1.

Animal Productivity on Untreated Crop Residues

Crop Residue Quantity Quality Animal Performance
tonness) %CP %ADF DMI (g) ADG (g)

Maize stovers 5.0 million 4.3 31.0 1 000 (+79 to +35.0)
Maize cobs 4.0 million 2.0 29.0 500 (+62 to -11.0)
Sugarcane tops 0.4 million 5.0 38.0 800 (-10 to 1.0)
Wheat straw 0.2 million 4.3 44.0 -
Rice straw 0.05 million 4.0 46.0
Barley straw 0.05 million 6.0 48.0

CP = crude protein ADF = acid detergent fibre
DM = dry matter intake ADG = average daily gain

We are trying to improve the feeding value of crop
by-products and the methods used are shown in Table 2.
Broadly, they are:

1. Physical treatment
2. Chemical treatment
3. Feed residues with something else (supplementation)

Common Methods of Improving Utilization
of Crop Residues in Kenya
Physical Chemical Supplementation

Grind Alkalis Concentrate mixed ration
NaOH starch
Ca(OH)= protein

Chop Bicarbonates
NaH CO, Minerals
Magadi soda

Urine Fodder Napier
Ammonia potato vines
Chlorine Molasses-Urea-Minerals

In all three methods, the idea is to try and increase
digestibility of the residue. This increases intake and
nutrient availability. The farmer can physically grind or
chop the residue but this is not feasible because of the cost
factor. As far as chemical treatments in Kenya are
concerned, we feel that the only thing which has a chance is
Magadi soda. This is not expensive and is easily available
because it is produced locally. There are dangers involved
in the use of chemicals by small-scale farmers and the
improvement to animal production does not warrant the risk.
Supplementation or feeding the crop residue plus small
amounts of common fodders such as napier and leucaena or
commercial and home-made concentrate seems to hold more
promise, at least from the data obtained so far (Table 3).

Animal Performance on "Treated" Crop Residues

Materials Treated with Species % Improvement

Maize stover NH, Sheep 65
Maize cobs NH, Sheep 44
Maize cobs NaOH Cattle 34
Maize cobs Magadi soda Cattle 24
Maize cobs Ground (10mm) Cattle 12
Dry maize leaves Fed with 200g conc. Sheep 500*

* The animals rose from a loss of 11.Og daily to a gain of 46.0g
per day

3. New Integrated Systems of Crop and
Animal Production (NISCAP)

This is the new idea that is being proposed after some
research by the scientists at the Department of Animal
Production, University of Nairobi. Since the common crop
residues cannot improve animal performance without the added
costs of treatment, the idea to be explored was can the
farmer feed crop residues before they deteriorate in quality
to warrant treatment? In maize we looked at 4 broad areas in
which this could be achieved, namely:

(i) Harvest early and dry the maize grain artificially
(ii) Top the maize plant
(iii) Strip the maize plant
(iv) Plant excessive numbers and thin

We decided on systematic stripping by picking one leaf per
plant once a week. We found, among other things, that by
defoliating 1 ha. of maize, we could sustain three sheep per
year (Table 4).


Established Quantity, Quality and Utilization by Sheep
of Defoliated Maize Leaves (H5123 H511)

Parameter Range Values

Dry matter yield 1.0-1.2 tons/ha./season
Crude protein content 10.5 14.6%
Fibre (ADF) content 37.7 48.2%
Fibre (ADL) content 5.1 9.8%
Digestibility 51.0 62.6%
Dry matter intake 770 950 g/sheep/day
Average daily gain 95 136 g/sheep
Feed conversion ratio 6.5 12.2
Carrying capacity 3 sheep/ha./year

With 1 million ha. of maize grown in Kenya per year, this
would increase feed resources. One of the advantages of this
method is that the labour involved in collecting materials
(stripping the leaves) is minimal about 2 kg of dry matter
can be collected in 1 minute. Secondly, data in Table 5
clearly shows that fresh maize leaves are superior to dry
maize leaves and other common dry season feeds such as Napier
grass and Rhodes hay.


Comparative Performance of Sh eep_and Cattle fed on
Maize Leaves and other Common Dry Season Feeds

Roughage Fresh Dry Napier Fresh Rhodes
Leaves Leaves Leaves Hay

Initial weight 20.2 25.0 21.0 68.3 77.0
Final weight 29.8 24.5 24.9 87.2 85.8
DMI, g/day 890 730 690 2 660 1 793
ADG, g 136 11 48 336 147
FCE 6.5 14.0 7.9 12.2

DMI Dry Matter Intake ADG Average Daily Gain
FCE Feed Conversion Efficiency

The major problems involved in maize defoliation as a
technique to increase feed resources are:

(a) Defoliation causes a grain loss. However, as the maize
matures you get a small boost in maize production when
one leaf is removed weekly (Table 6).


Grain Trade-Off with Age of Maize and Intensity of Defoliation

OFF/WEEK 75 100 115

1 10. 0 5.0 + 3.0
2 34.0 26.0 18.0
3 42.0 38.0 23.0

(b) Maize lodging increased lodging was related to
defoliation (Table 7) but the cause of increased lodging
is not very clear yet.


Effect of Defoliation on Maize Lodging

Lodging 7. Defoliated Undefoliated

Total 23.8 14.4
Stem 16.4 7.9
Root 7.4 6.5

Lodged Stjnd i Lodged Standing

Stem height, cm 324 315 315 328
Root weight, g 25 48 19 42
Lignin in stem, 7. 9.9 10.2 12.0 12.3

Four theories have been advanced so far:

(i) Stems grow taller with defoliation
(ii) Insufficient root development
(iii) Insufficient lignification of the stems
(iv) The act of defoliation itself

Theories (i) to (iii) have been investigated (Table 7)
with inconclusive results.

(c) Preservation the concern is dry season feeding. The
fresh highly nutritious maize leaves are available only
during the wet or growing season. During this period,
there is also plenty of grazing. Table 8 shows the
three methods that have been tried in a pilot study,
namely, sun drying; air/shed drying; and blanching
coupled with sun drying. The appearance, texture and
smell of the preserved material indicated that fresh
maize leaves can be preserved for animal feeding. The
adaptation of any of these methods will largely depend
on specific local conditions.


Possible Methods of Maize Foliaae Preservation

Start End (feel) (odour) (days)

Sun dry green grey-green soft Sweet/hay 4 5
Air/shade dry green deep green very soft Sweet/hay 6 7
Blanching + Sun
(Heat Treatment) yellow brownish-green soft Molasses 3 4


Discussion centred on leaf stripping and touched on pasture
improvement and chemical treatment of residues.

Leaf stripping.

On the question of storage of leaves it was said that this was a
pilot project so storage was not significant. However, it could
be a problem on a bigger scale. Leaves were fed fresh and
chopped. It was observed that in maize there is considerable
variation in leaf numbers and size. On the question of testing
the effect of stripping on different varieties it was stated that
only three varieties were tested H511 and H512 and 600 series.
The 500 series has about 14- 15 leaves per plant and the 600
series about 20 per plant. When plants were defoliated three
leaves were left on top plus the two leaves that subtend the
cobs, thus leaving a total of 5 leaves per plant.

Dr. Collinson observed that farmers in that area are actually
planting the 600 series to get fodder but as the 600 series has
just been introduced into that area it had not been used in the
experiments that much.

Asked whether the findings that starting defoliation before 115
days seriously penalised grain yield was disappointing, Dr. Male
said that grain yield was reduced but with gradual stripping this
was minimized and furthermore material was not accumulated so
storage problems were minimised.

The possibilities of carrying out grass weeds as a fodder while
stripping and/or thinning were discussed.

Use of Maqadi Soda

Asked whether people may prefer to use Magadi soda for purposes
other than chemical treatment of crop residues, Dr. Male replied
that at the moment Kenya has large deposits of soda and it is
hoped that more will be extracted and there will be no conflict
between home use and the export market. In Kenya sodium
hydroxide costs $4 per kg and Magadi soda 10 cents per kg.


Paper #20 entitled "ILCA's Approach to Livestock Nutrition in the
Farming Systems Context" was presented by Dr. A. Mosi.


ILCA's Nutritional Proqramme

In each of the zones of Africa the most severe husbandry problem
has been the provision of good quality feed year round for meat,
milk and for traction. The current nutritional programmes focus
on the farming system practised in each zone and how best the
system can be manipulated to provide more feed for increased
animal nutrition. This approach emphasises the following

(a) the use of cereal and legume crop residues for ruminant
nutri ti on

(b) the introduction of good quality forages into the arable
system and agropastoralist settlements

(c) a rational use of existing rangelands

(d) animal traction

(e) water provision during the dry season

Some Preliminary Findings

(a) Cereal Crop Residues: Cereal crop residues are of poor
nutritive value but it has been found that the addition of a
good quality legume forage can improve digestibility and
intake to a point where reasonable levels of productivity
are possible. Mixed farmers in densely populated areas
could use cereal crop residues and farm-produced legumes to
feed the livestock component.

The African Research Network on Agricultural By-Products
(ARNAB) has been formed to stimulate research in this area
throughout Africa. ARNAB headquartered at ILCA, publishes a
quarterly newsletter.

(b) Introduction of Good Quality Forages: A germplasm
collection has been established and a computerized indexing
system based on 120 descriptors has been devised which
allows the rapid identification of those accessions believed
to be best suited to the climatic edaphic and livestock
production conditions of various African countries.

Strategic feeding of forage legumes with crop residues would
enable agropastoralists to feed in-calf and lactating cows
after natural pastures have been exhausted. This target
group has the greatest protein deficit and is where improved
nutrition will generate the greatest benefits: increased
milk offtake for human consumption, higher calving rates and
faster calf growth.

The benefits of rationing scarce feed resources to selected
animals was demonstrated at the ILCA study areas in Kurmin,
Biri and Zonkwa in northern Nigeria. Agropastoralists were
encouraged to feed cotton seed cake to in-calf and lactating
cows only, instead of feeding it sporadically and
unsystematically to the whole herd, which was the general

The undersowing of stylosanthes into a stand of sorghum in
the vicinity of Kaduna is technically attractive, but the
desired date of undersowing coincides with a period of high
labour demand for weeding. This technique may thus prove
not too attractive for agropastoralists.

The ILCA team has demonstrated that under controlled
conditions DM yield of 6.8 tons/ha. containing 56%
stylosanthes and a CP range of 14% 6% (rainy season dry
season) is possible. Under pastoralist conditions, DM yield
is about 5 tons/ha. with up to 68%7. stylosanthes and a CP
range of 13% 7%..

(c) Animal Traction: The performance of work by local oxen may
involve a small increase in protein requirement but
considerable increments in energy demand. Appetite is
probably increased and marginal quality roughages may be
utilised. There are important nutritional implications for
feed utilisation strategies with draught oxen and research
is being carried out to determine the exact maintenance
requirements of Zebu and cross-bred animals as well as the
breed responses to feeding high and low energy levels on
work output and general body condition.


(a) Animal Feeding;: While there is a problem in producing
enough food to feed animals well the animal scientists are
producing animals with more body mass. It was asked whether
this aspect was being considered. No real answer was given.
Dr. Mosi said that the efficiency of feed conversion with
good quality feeds is better with fast-growing ruminants.
This is not true when poor quality feeds are offered to
genetically synthesized heavy breeds.

Asked how frequently animals need water, Dr. Mosi replied
that it depended on the moisture content of the feed the
animal was getting. With fresh forage you get a lot of
moisture. If it is dried food you need more water.

Ruminants need water at least every 3 days. To improve
performance access to water is needed.

On the question of the use of cassava and cotton seed cake
as feed for animals it was suggested that some varieties of
cassava contain some cyanide but the effects on ruminants
depended on quantities ingested. Usually scrapings from
daily food preparation are used as supplementary feed in
West Africa.

Asked to expand on what he thought were the nutritive
requirements for draught animals in terms of energy and
protein digestibility, Dr. Mosi said that current trials are
ongoing. Except in the early work of Brody, nowhere are
actual nutritive requirements of draught animals
established. We are measuring this on an input/output
basis. We can measure how much energy the animal is
receiving. The equilibrium stage of energy input and energy
output comes when the animal is neither gaining nor losing
weight but still offering traction. When you get to this
point you can then assess amounts of energy needed.

It was observed from some of the graphs (refer systems
descriptions) that cattle condition improves quite rapidly
at onset of rains under natural conditions. Mr. Maynard
asked whether much work has been done on just a period of
two or three weeks feeding and how this compares with
maintaining liveweight over the whole of the dry season in
order to improve draught power at the beginning of the
season. Weight attained prior to work is rapidly lost if
that condition is not maintained through balanced rationing.
Body reserves (especially fat) are mobilized to provide
work. These reserves must be replenished simultaneously
with depletion. Dr. Tessema said that in Kenya they fed
for one month prior to the start of ploughing with maize
stover plus MUM at a minimum amount. The cattle were worked
for a total of 8 days to finish 2.5 ha. Transport and
weeding work was still to be done. To keep them in
condition feeding continued over the dry season. Within 110
days with this kind of supplement as well as natural
pasture, both Zebu and cross-bred animals gained weight with
the work. It is necessary to keep animals in good condition
over a minimum period of about 6 weeks.

(b) Plouqh teams: Asked whether ILCA was looking at the
question of female animals ploughing and the effect on their
reproductive cycle of the stress period of ploughing, Dr.
Mosi replied that they were even looking at a single ox to
plough. Indications are that where land availability does
not exceed 2.5 ha. per family, one ox could prepare the land
satisfactorily. Dry female animals can do the work. What
is not known is how far into the pregnancy situation one can
use them. If they are dry animals will it lengthen the dry
interval? Trained dry cows can do as good a job as oxen.
It is not known how work will affect the incidence and
maintenance of pregnancy. These issues will have to be

(c) Ran.geland Improvement: Mr. Riches noted that even though
ILCA, as a matter of policy, said that we should not be
involved in re-seeding the rangeland we have heard that work
is going on by a number of people already involved in
experimental programmes of reinforcing the rangeland with
legumes. Dr.Mosi replied that ILCA is not against
re-seeding the rangeland. The cost factor and inputs such
as seeds do not exist in most African countries. If you are
going to conscientiously re-seed you must protect the
seedlings, so what do you do with the cattle? Dr. Henson
observed that experience in the US, where re-seeding of the
range has been tried for many years, has shown that, from an
economic standpoint, any substantive effort to re-seed
ranges has to be limited to specific localities. One finds
in African and US experience that if one could manage the
range and control livestock the range would regenerate with
natural growth.


Networking and Workshop Assessment



The objectives and context of the draught power and animal
feeding networkshop was outlined in paper #2 by A. Low.

The need for networking in on-for research was especially evident

a) Practitioners tended to be few and scattered

b) Formal links through established journals do not exist

c) Development of sound new methodologies must rely heavily on
cumulative trial and error experience.

The forms of networking suggested were:

a) Newsletters: experience with the farming systems newsletter
put out by CIMMYT from Nairobi suggests that this type of
networking tends to be somewhat passive in that many
readers are happy to receive information and ideas but only
a few of them make significant contributions.

b) Studies to inventory aspects of special concern in terms of
systems where the aspect is of importance as well as past
and ongoing related research work. The advantage with this
approach is that it can actively seek out experiences and
provide a comprehensive picture of ongoing work.

c) Networkshops with the same objectives as in (b) but being
less comprehensive though more interactive.


The discussion revolved around three aspects:

context and form of networking in general
percieved output of this networkshop
assessment of this networkshop in terms of:
a) approach
b) format

Conte!;t of Networks

It was agreed that there was a need for some co-ordination of
effort and for keeping people working in more remote areas
informed of what was going on elsewhere. However it was felt
that there may be a duplication of effort by different
International Research Centres. For example ILCA was operating a
By-Products and Residues network (ARNAB) and was going to start a
draught power network. IDRC had made some provision for a
Minimum Tillage network. SADCC was inventoring past research and

talking about developing an Agricultural Research network. It
was felt that there was a need for some organisation to take an"
overview of the situation and that the appropriate institution
may be IDRC.

It was agreed that newsletters on their own were not totally
satisfactory because the response was often poor. ILCA said that
they had tried to circumvent this by arranging annual visits to
members of the network and hold discussions of general interest
in situ. As the membership gets bigger ILCA intends to hold
annual get-togethers.

CIMMYT made it clear that its interests were very specific, to
try to bring together systems and component oriented researchers
focusing on the same problem. It did not propose to perpetuate
these groupings. Each workshop would have a different focus.
CIMMYT would however be happy to facilitate any networking which
emerged from these groupings, through travel and meetings.
CIMMYT would also endeavour to keep participants informed of
other activities in the region relevant to this focus.

Output from this Workshop

Participants were asked to comment on work they had heard about
outside their countries or areas that might be appropriate in
their own local specific situations.

a) The Botswana plough planter and the work in Zimbabwe on
direct planting into winter ploughed land with herbicides
and tines were said to have potential value in Swaziland as
time-saving devices.

b) The Kenya work on crop residues was thought to be relevant
to Western Sudan

c) The minimum tillage and herbicide work in Zimbabwe and
elsewhere was thought to have some relevance in Botswana,
but posed problems in Kenya where mixed cropping of cereals
and legumes was being introduced into the system.

d) ILCA's work on improved efficiency of harnessing animals to
implements was seen as being valuable in Botswana.

e) Selective feeding of animals in Kenya and Lesotho was of
interest in Malawi, where performance tested bulls given to
farmers had died because they had not been fed adequately.

f) Stall feeding work in Kenya and Lesotho could tie in well
with areas where work was being done on collecting and
feeding crop residues.

g) One beneficial output of the workshop was said to be that
participants had written down what would not have been put
on paper except in an annual report. Even though some
papers came late information had been recorded which may
not have been otherwise.

Workshop Assessment

a) General Aproach

It was felt that the divergence between systems was too great
and that areas of concern needed to be more closely defined.
There was acceptance that it had proved difficult for
presenters to get down to detail. People tended to preface
their presentations with general statements of the problem of
which everyone was already aware.

Disappointment was expressed that the presentations were so
abbreviated that much useful information was lost.

It was thought that the idea of using the systems calendars
(wall charts) was good and that time could have been saved if
these had been prepared in advance so that those areas with
similar systems could have got together.

It would have been advantageous for participants to have
received the papers in advance. This would have enabled
groups with similar systems or research work to get together
for presentations and discussions.

b) Workshop Format

The for-mat of the workshop was designed to expose the
component worker to the different system situations and the
implied differences which appropriate solutions would have to
handle. Participants were given guidelines for presentations
(appendix 3) and instructions for discussions (appnedix 4)
after arrival.

The prescribed formats were felt to be good but several
suggestions for improvement were made:

a) Guidelines for presentation must be given out well before
the meeting. Many presenters could not react adequately to
guidelines given to them on arrival.

b) The two group sessions (one on discussion of systems and
one on discussion of interventions in relation to those
systems) should be reorganised.

(i) Groups should have systems, problems and therefore
intervention possibilities in common.

(ii) Group discussions should wait until both systems and
interventions have been covered in plenary session.

c) Less time should be spent in general descriptions of
farming systems by systems researchers (this is what the
guidelines were intended to ensure). More time should be
spent on actual experimental results by component
researchers and this should include information on the
costs of the interventions.

onk F",11=:mnD I x

Appendix 1


Mr F Buckham
Malkerns Research Station
PO Box 4
Mal kerns

Mr C Chabala
Adaptive Research Planning Team
(Central Province)
PO Box 80908
Kabwe Regional Research Station
Kab we

Mr D Chandler
Integrated Farming Pilot Project
PO Box 69

Dr M Collinson
PO Box 25171
Nai robi

Ms T Faria
Faculty of Agronomy
PO Box 257

Mr T Farrington
Agricultural Research Station
P/Bag 0033

Dr J Henson
Office of International
Development Programme
Washington State University
Pul 1 man
Washinqton 99164

Dr S Jutzi
Ethiopian Highlands Programme
PO Box 5689
Addis Ababa,

Dr J Kategile
PO Box 62084

Dr E Klosterman
Agricultural Research Division
PO Box 829

Dr B A Koch
c/o D Norman ATIP
PO Box 90

Dr H Kayongo-Male
Faculty of Veterinary Medicine
PO Box 29053

Dr A D Malithano
PO Box 257

Ms Z Mamba
Malkerns Research Station
PO Box 4

Appendix 1

Mr S Mamba
Ministry of Agriculture
PO Box 162

Mr M 'Matli
Agricultural Research Division
PO Box 829

Mr T Maynard
Adaptive Research Planning Team
(Lusaka Province)
Mount Makulu Research Station
P/Bag 7
C hi 1. anLa

Mr C Riches
Agricultural Research Station
P/Bag 0033

Dr S Russo
c/o American Embassy
PO Box 596


Mr E Shumba
Agronomy Institute
Research & Specialist Services
PO Box 8100 Causeway

Ms L Mokjope
PO Box 829

Dr A K Mosi
PO Box 5689
Addis Ababa

Mr G Mudimu
PO Box 8117

Research Division

Mr H Muiruri
Animal Science Department
Egerton College
PO Njoro

Dr K V Ramanaiah
Department of Agronomy
Ft- Box 257


Dr J Siebert
c/o D Norman ATIP
PO Box 90

Mr D Singa
Chitedze Research Station
PO Box 158
Li 1onqwe

Dr S Tessema
FAO Drylands Farming Research
and Development Project
PO Box 340

Dr V Watson
Malkerns Research Station
PO Box 4
Mal kerns

Mr A Zimba
Chitedze Research Station
PO Box 158

Appendix 2

Systems Interactions

When it comes to the topic of this meeting, i.e., draught power,
the need to be thinking in systems terms becomes obvious. In the
system of farming we have in Swaziland, livestock husbandry
cannot be divorced from crop husbandry. Cropping depends on
cattle for draught power, for manure and for the transport of
inputs and harvest, while crop residues provide essential winter
season sustenance for livestock. By the same token, here in
Swaziland, we cannot think of farming activities being separate
from and unconnected with other farm household activities. Just
as few Swazi farmers are either croppers or herders few are
either housewives or farmers, few are either producers of crops
or producers of handicrafts, few are either farm workers or wage
earners, few either grow crops or brew beer. Most farm
households combine a few or more of these activities and more
time spent on one of them implies giving up time spent on

Sometimes it is possible to satisfactorily combine more of one
activity with another. Thus, for example, many male wage workers
send money home for their wives to hire tractors for ploughing.
By providing Government tractor hire pools in the Rural
Development Areas, we have enabled households to better combine
wage earning with family farming. Farming systems research
should be looking for similar ways to increase farm production,
not at the expense of other activities, but in conjunction with
them. Technologies that are complementary with other activities
will be adopted much more readily than ones that are competitive.
If you ask a Swazi rural housewife to spend so much time tending
her crops that she does not have time to fetch the family's daily
water requirements, she would be foolish to listen to you.

At least we should be considering how the competitiveness between
increasing farm production and reducing the output of other
activities can be minimised. This is the challenge for farming
systems research in Swaziland. In our Rural Development Areas
programme we have gone some way to reducing such conflicts
through, for example, our programme of rural piped water
supplies. Unfortunately, the production technologies that have
become available to us from conventional research have tended to
be competitive rather than complementary with other activities.
This is why they have often not been appropriate for the majority
ot Swazi rural families.

Here in Swaziland the draught power problem provides a good
e: be considered when examining ways of increasing crop production
on Swazi Nation Land. One of the factors causing low crop yields
is that draught cattle are weak at the beginning of the season so
that ploughing and planting is delayed and seedbeds are not
adequately prepared. At the same time more families need draught
poJoer at the beginning of each season.

Appendix 2

But we are often told that there are already too many cattle in
Swaziland. Not only do more people need draught teams each year,
but at the same time more households are requiring their land to
be cultivated more quickly than before, since men in wage
employment who return to their farms to plough the lands have
limited holiday periods in which they can operate. Given the
increased demand for draught in an overstock situation and the
need for speedier operations, we have looked at alternative
methods of draught power in Swaziland. I have already mentioned
the Government tractor pools. We have also developed the small
Tinkhabi tractor here in Swaziland and are doing work on
improving implements such as the Safim planter. I am pleased to
see from the agenda that you will have an opportunity of seeing
something of these areas of research and development while you
are here.

Alternative possibilities for better meeting the draught power
needs are also being examined. For example, we have an
inter-cropping research programme here which is looking at the
economics of inter-cropping forage legumes with maize. With all
these options, though, we need to think carefully about their
wider implications on the rural way of life. Can ever-increasing
numbers of cattle be supported to meet the future increase in
demand for draught power? What does more time and effort spent
on feeding and managing cattle mean in terms of other household

I would urge delegates to not only think in terms of a farming
system, but also to apply the systems approach to the broader
rural and household production systems, of which farming is an
integral and not an isolated part.


There are obviously a whole range of ways in which the draught
power problem can be tackled. I have mentioned some of these
al ready:

(a) improved winter feeding with forage legumes, conservation of
crop residues or supplementary feeding;

(b) improved draught implements to speed operations or reduce
draught requirements; and

(c) alternative power sources such as the Tinkhabi or
conventional tractors.

Other options include:

(d) zero or minimum tillage;
(e) chemical weed control; and
(f) selective oxen feeding and management.

Some of these options will be more feasible in one area than
another, but in order to judge whether these options are likely

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs