African Forage Plant Genetic Resources,
Evaluation of Forage Germplasm and
Extensive Livestock Production Systems
PROCEEDINGS OF THE THIRD WORKSHOP
HELD AT THE INTERNATIONAL CONFERENCE CENTRE
ARUSHA, TANZANIA, 27-30 APRIL 1987
by the
PASTURE NETWORK FOR EASTERN
AND SOUTHERN AFRICA (PANESA)
DECEMBER 1988
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia



African Forage Plant Genetic Resources,
Evaluation of Forage Germplasm and
Extensive Livestock Production Systems
PROCEEDINGS OF THE THIRD WORKSHOP
HELD AT THE INTERNATIONAL CONFERENCE CENTRE
ARUSHA, TANZANIA, 27-30 APRIL 1987
by the
PASTURE NETWORK FOR EASTERN
AND SOUTHERN AFRICA (PANESA)
Edited by B.H. Dzowela
 
DECEMBER 1988
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
Thi- Oni
  
1WUC-5B1-U5L6
Correct citation:
Pasture Network for Eastern and
Southern Africa (PANESA) . 1988.
African forage plant genetic resources,
evaluation of forage germplasm and
extensive livestock production systems.
Proceedings of the Third Workshop held
at the International Conference Centre,
Arusha, Tanzania, 27-30 April 1987.
ILCA, Addis Ababa.
PREFACE
These proceedings include five opening and keynote addresses
which provide introductory remarks on the activities of the
Pasture Network for Eastern and Southern Africa (PANESA) in its
regional attempts to enhance research and development of ruminant
livestock feed resources. These introductory remarks are
followed by papers on the themes of the workshop "African Forage
Plant Genetic Resources" on which there are four papers
presented. On the second theme "Forage Germplasm Evaluation"
there are 16 papers presented and finally on the theme "Extensive
Livestock Production Systems" there are 21 papers.
As a workshop organiser and editor of the proceedings, I
would like to pay tribute to all authors and especially those who
presented the papers at Arusha, Tanzania. From their
contribution it was possible to record success in this workshop
which fostered information exchange and cooperation among the
many individuals of diverse interests who attended. Many
individuals from Tanzania made significant contributions to the
success of the workshop. I would like to thank particularly
Prof. A.B. Lwoga in his role as Chairman of the Tanzania Forage
Resources Network for giving the local support and the publicity
the workshop received. All members of the PANESA Network
Steering Committee carried out their support-role in a most
helpful and professional manner. The administrative machinery at
ILCA Nairobi merit special mention for their work with hotel and
travel arrangements.
Preparation of this document involved several individuals
who made significant contributions and all these deserve
acknowledgement. Dr. Peter de Leeuw gave a critical view to some
papers, Mrs. Eva Ndavu and K.S. Buigutt helped with proofreading
and technical editing. Secretaries, Mrs. Angela Kiura and
Josphine G. Njuki patiently put in so many wordprocessing/typing
hours. Finally, the assistance of the Information Section of
ILCA HQ in publishing the proceedings is gratefully acknowledged.
Ben H. Dzowela
PANESA COORDINATOR

TABLE OF CONTENTS
PREFACE iii
OFENINO CEREMONIES
WELCOME ADDRESS - A.B. Lwoga 3
OPENING STATEMENT - Jackson A. Kategile 6
OPENING ADDRESS - J.C. Tothill 7
OPENING ADDRESS - Hon. Sumaiye 8
KEYNOTE ADDRESS - J.K. Kyambwa 12
SESSION I: AFRICAN FORAGE PLANT GENETIC RESOURCES
Some thoughts on the collection, introduction and
evaluation of pasture legumes in Zimbabwe -
J.N. Clatworthy and C. Madakadze 21
Review of range and pasture plants in Botswana -
G. Tacheba and A. Moyo 37
Distribution of potential forage plants in Malawi -
H.D.C. Msiska, J.T. Munthali and B.H. Dzowela 46
The distribution of Sesbania species in
the PANESA region -
J.F.M. Onim and B.H. Dzowela 54
SESSION II: FORAGE GERMPLASM EVALUATION
Forage germplasm at the International Livestock Centre
for Africa
J. Hanson and J.R. Lazier 69
A review of the potential of Brachiaria
species as forage crop for livestock in Tanzania -
N.A. Urio, S.V. Sarwatt and E.J. Mtengeti 74
Evaluation of the potential of some common
forage germplasms in Mauritius - D. Jotee 81
A review of forage germplasm introduction and
evaluation in Uganda - G.S. Byenkya 91
Forage dry-matter productivity variation of
Panicum and Cynodon ecotypes in Malawi -
H.D.C. Msiska, B.H. Dzowela and J.T. Munthali 104
Evaluation of Desmodium adscendens through
vegetative propagation in natural and planted pastures -
E.N. Sabiiti, P.R. Henderlong and J.S. Mugerwa 112
The screening of perennial Trifolium species
mainly from the Ethiopian highlands and their
potential for use in pasture - J. Kahurananga 120
Yield and quality of Lablab purpureus ecotypes
in high altitude semi-arid lands in Kenya -
M. Karachi 127
Adaptation of forage and browse legumes to the semi-arid
mid-altitude regions of Kenya - an interim report -
L.K. Menin, D.M. Njarui, W.M.Beattie,
B.A. Keating and R.K. Jones 134
Characterisation and preliminary evaluation of
accessions of Zornia species from the ILCA
collection -
J.J. Hakiza, J.R. Lazier and A.R. Sayers 149
Characterisation and evaluation of forage legumes
in Ethiopia: Preliminary examination of variation between
accessions of Stylosanthes fruticosa (Retz.) Alston -
J.J. Hakiza, J.R. Lazier and A.R. Sayers 174
Fodder production potential for zero-grazing systems
in the bimodal West Usambara Mountains of Tanzania -
R. Pfeiffer 192
The current status of knowledge on the feed value of
Crotalaria species -
S.V. Sarwatt and F.E.J. Mkiwa 196
A review of pasture and fodder crops germplasm
evaluation at Uyole, Mbeya in Tanzania -
T.J. Myoya and N.J. Mukurasi 204
A review on evalaution and agronomy of promising
lupine varieties in Kenya - F.N. Muyekho 222
The effect of lupine/maize meal supplementation
on the performance of lactating Friesian cows -
E.A. Mukisira 230
SESSION III: EXTENSIVE LIVESTOCK PRODUCTION SYSTEMS
Livestock production systems in Chokwe,
southern Mozambique - J. Timberlake 237
Production of the Malagasy extensive livestock -
J.H. Rasambainarivo and R. Razafindratsita 251
The prediction of rangeland production from
rainfall data in arid and semi-arid Eastern Africa -
P.N. de Leeuw and R. Nyambaka 260
Productivity of beef steers under rangeland
conditions in a semi-arid area of Kenya -
h.l. Potter 269
Classification of Kenya rangeland - M.O. Olang 286
Identification and ranking of production constraints
in agropastoral systems of Machakos district in Kenya -
A.W. Mukhebi and A.M. Gitunu 298
Relationship between phytomass productivity, animal
forage preference and environmental factors on
semi-arid grazed and ungrazed key range plants
in Tanzania -
M.M.S. Lugenja, R.N. Mero and A.R. Kajuni 310
Grazing behaviour of Maasai cattle - P.P. Semenye 325
The role of body weight changes and other
factors in the control of fertility of beef
cattle at Athi River Ranch, Kenya - E.K. Odhuba 331
The role of water pulses on the productivity
of savanna ecosystems in Kenya - E.K. Maranga 365
Nutritional values of the diets of Maasai cattle -
P.P. Semenye 373
Livestock feeding potentials, constraints
and possible improvements in the semi-arid
areas of Central Tanzania -
A.J. Kitalyi and M.A. Kabatange 383
The potential of pasture legumes and their
role in improving farming practices under
extensive livestock production systems
in Zambia - J. Kulich 394
Land-use changes in ranches which were set
up in marginal areas of Kenya -
M.O. Olang and T. Njoka 401
A strategy of livestock research adapted to
semi-arid small-scale mixed farming system -
the Katumani experience - S. Tessema,
E.E. Emojong, M. Maluti and P.N. de Leeuw 405
Legume-based cropping: A possible remedy to land
tenure constraint to ruminant production in
the subhumid zone of Central Nigeria -
G. Tarawali, M.A. Mohamed-Saleem and
R. von Kaufmann 417
Role of crop residues as livestock feed
in Ethiopian highlands - Daniel Keftasa 430
Nutrient quality of forages in Ethiopia with
particular reference to mineral elements -
E. Kabaija and D.A. Little 440
Effect of pasture mineral levels on extensive cattle
production in Kenya - A. Abate 449
Effect of season, fertilizer application and
age of regrowth on mineral content of guinea
grass (Panicum maximum Schum) and Giant
Star grass (Cynodon nlemfuensis Chedda) -
E. Kabaija and D.B. smith 458
Research needs for forage development in
Ethiopia - the Fourth Livestock Development
Project strategies -
Alemayehu Mengistu and A. Robertson 468
List of Participants 483
viii
OPENING CEREMONIES

WELCOME ADDRESS - A.B. Lwoga
Chairman of PANESA's Working Committee and
Chairman of the Executive Committee of the
Tanzania Forage Resources Network
On behalf of the PANESA Working Committee, on behalf of the
Executive Committee of the Tanzania Forage Resources Network and
on my own behalf, I should like to take this opportunity to
warmly welcome you all to this ceremony which marks the beginning
of the PANESA Workshop and Annual General Meeting for the year
1987. As a Tanzanian citizen, it is my great pleasure and
privilege to welcome all the foreign workshop participants to
Tanzania, and to Arusha in particular. I am not a resident of
this part of Tanzania, but I know that Arusha town is very famous
for its hospitality and, also, that the surrounding countryside
boasts of some of the rarest and most wonderful game sanctuaries
in the world. You are, therefore, most welcome to utilise this
opportunity, over and above the workshop business, to learn a bit
about Tanzania by exploring Arusha and the surrounding
countryside.
A word of special welcome is due to our Guest of Honour
today, the Honourable Sumaiye, Deputy Minister for Agriculture
and Livestock Development. I am most delighted that he has been
able to accommodate this event in his very busy schedule of
activities, and also that he has been able to come, to officiate
at this ceremony. I should, therefore, like to place on record
my deep appreciation to the Deputy Minister for the honour he has
accorded us by agreeing to perform this important function in the
history of PANESA.
Mr. Guest of Honour, it is pertinent on this occasion to
take a brief look at PANESA's origin and tasks. The Pasture
Network for Eastern and Southern Africa has its roots in a
workshop which was held at Harare in Zimbabwe in September, 1984
and attended by pasture and livestock experts from eleven eastern
and southern African countries. Having realised that inadequate
nutrition was the single most important constraint to livestock
production in the countries represented, the workshop resolved to
form a network which would strive to find solutions to the
problems affecting pasture production and, consequently, the
livestock industry in the Eastern and southern African region.
PANESA was launched in November 1984 with the overall aim of
improving the effectiveness of pasture research in participating
countries and expediting the application of improved technology
by farmers and graziers at all levels of management.
What is the nature of the problems that PANESA has committed
itself to solve? A glance at the past and present performance of
the ruminant livestock industry in eastern and southern Africa
throws some light on the urgency and immensity of PANESA's tasks.
Eastern and southern Africa as a region holds nearly 86 million
head of cattle or 51% of the total cattle population on the
African continent. The region, however, produces a mere 38% of
Africa's bovine meat and only 33% of Africa's cow milk. The
annual herd offtake is estimated to be 10%, representing about
143 kg of slaughter weight per head, compared with an offtake of
34% and a slaughter weight of 218 kg per head for the developed
countries. The average milk yield is estimated at 425 kg per
year per cow, compared with 3,217 kg per year per cow in
developed countries. Overall livestock production is so low that
it hardly meets the human requirements for livestock products in
the region. One consequence of this situation has been an
increase in imports of livestock and livestock products over the
years.
We in PANESA believe that inadequate nutrition for the
animals is one of the main factors accounting for the low
productivity of livestock in the region. The animals subsist
largely on natural pastures whose vegetation is not only scanty,
but also of low quality. In the eastern and southern African
region, natural pastures occupy 34% of the total land area while
cultivated pastures occupy a mere 6%. In the developed countries
corresponding proportions are 23% for natural pastures and 12%
for cultivated pastures. The tasks of PANESA, urgent and immense
as they are, involve removing the nutrition bottleneck through
the generation and/or adaptation and application of appropriate
technology.
Mr. Guest of Honour, since its formation in November 1984,
PANESA has carried out the following activities in pursuit of its
objectives:
1. It has started a newsletter, the "PANESA Newsletter", which
is a medium for the exchange of ideas and experiences among
all those interested in pasture and livestock research and
development.
2. It organised a workshop in Nairobi, Kenya in November, 1985
on the theme "Feed Resources for Small Scale Livestock
Producers" .
3 . It organised a training course for young early career
scientists and pasture technicians in "Forage Plant
Introduction and Initial Evaluation" in Ethiopia in October,
1986.
4 . It has started a collaborative pasture introduction and
evaluation programme in which national research institutes
participate in experiments that are coordinated on a
regional or sub-regional basis.
Mr. Guest of Honour, the workshop you will be opening in due
course is yet another milestone in PANESA 's efforts in the long
march towards its objectives. The theme of the workshop, which
arose from the Nairobi workshop in November, 1985, consists of
three topics, namely:
1. African forage plant genetic resources. This aims at taking
an inventory of important forage plant genetic resources in
the region.
2. Germplasm evaluation. This seeks to take stock of national
research efforts in forage germplasm evaluation in the
region.
3. Extensive livestock production. This aims at reviewing the
state of the art in the diverse extensive livestock
production systems in the region with the objective of
identifying common problems and generating strategies to
improve the productivity of the systems.
The workshop has drawn more than 60 participants from 14
member countries, namely: Botswana, Kenya, Lesotho, Madagascar,
Malawi, Mauritius, Mozambique, Rwanda, Somalia, Swaziland,
Uganda, Zambia, Zimbabwe and the host, Tanzania. The workshop
has also drawn participants from a number of international
research agencies including the International Development
Research Centre (Canada) and the International Livestock Centre
for Africa (based in Addis Ababa) .
Mr. Guest of Honour, this workshop would not have
materialised and, indeed, PANESA would not have come into being
in 1984 had it not been for the generous financial, technical and
logistical support of international research agencies sympathetic
to the cause of PANESA. The International Development Research
Centre organised and financed the Harare Workshop in 1984 and
further financed PANESA' s first budget, whose activities include
this workshop. The International Livestock Centre for Africa has
provided the headquarters as well as back-up facilities and
services for PANESA.
I take this opportunity of paying tribute to the fundamental
and far-reaching role these two organisations have been playing
in establishing PANESA and, thereby, helping to build up national
capabilities in pasture and livestock research and development in
PANESA member countries. I wish to express my sincere hope that
their commitment will endure, and also that other international
agencies will join hands with PANESA in improving livestock
productivity through better livestock nutrition in Eastern and
southern Africa.
Mr. Guest of Honour, it is with greater pleasure, honour and
privilege that I now invite you to officially open this workshop.
OPENING STATEMENT - Jackson A. Kategile
idrc, Nairobi
Honourable Sumaiye, Deputy Minister for Agriculture and Livestock
Development, Ladies and Gentlemen:
It is certainly a great pleasure to be with you in Arusha
for the third Workshop on Pastures in Eastern and Southern
Africa. For some of us, this is also our third time to be
together in a similar forum. But to those who are not familiar
with IDRC, Mr. Chairman, I request a minute to enable me to
introduce the organisation.
IDRC is a corporation created by the Parliament of Canada in
197 0 to stimulate and support research in developing countries
for the countries' own benefit. IDRC is funded entirely by the
Canadian Parliament to which it reports annually. Its operations
are guided by an international 21-member Board of Governors.
IDRC has its headquarters in Ottawa with regional offices in
Singapore (Southeast Asia, East Asia and the Pacific) , New Delhi
(South Asia) , Nairobi (Eastern and Southern Africa) , Cairo (West
and Central Africa) and Bogota (Latin America and Caribbean) .
IDRC gives support to developing countries to carry out research
in areas of agriculture, health sciences, social sciences,
communication and information. I am in the division of
agriculture, food and nutrition sciences within which animal
sciences is encompassed. The Animal Science Sub-program in
Eastern and Southern Africa gives priority to research aimed at
improving the traditional livestock production systems, including
nutrition, pastures and socio-economics. It is therefore, not
surprising that IDRC has been associated with your interests and
efforts in pasture improvement through the Pasture Network for
Eastern and Southern Africa (PANESA) . The network has grown from
a dream to reality. In 1984, an idea was floated, today ILCA is
the coordinating agency and an active committee is guiding the
activities of PANESA. A full-time coordinator is now in place.
The noble aims within the umbrella of PANESA include:
1. Strengthening pasture research capabilities.
2. Procuring, disseminating and exchanging forage germplasm.
3. Evaluating the germplasm procured.
4. Developing appropriate technologies for traditional systems.
5. Collecting, disseminating and exchanging information.
PANESA has attracted the attention of several agencies
because of the important goals that you have set for yourselves,
and because the modality of organisation is bottom-up thus
facilitating the inputs of scientists from the region into an
ILCA-coordinated network. PANESA also encourages national
networks.
I do hope that this workshop will enable you to exchange
information and experiences. Further, I do hope that you will
take time to discuss the regionally coordinated research
activities. In drawing up your plans for regional research
activities you may wish to bear in mind that there are some funds
(though limited) which can be made available for national
activities, and ILCA can assist you in:
- research protocols
- data processing
- supply of germplasm
- training of scientists and technicians
- supply of some literature.
It is certainly advantageous to draw upon ILCA's resources.
As it is not my intention to take too much of your time, I
would like to reiterate that IDRC is proud to be associated with
PANESA and looks forward to a fruitful co-operation.
OPENING ADDRESS - J.C. Tothill
Head, Plant Sciences Division
International Livestock Centre for Africa
Mr. Deputy Minister for Agriculture and Livestock Development,
the Commissioner for Livestock Development, Chairman Professor
Lwoga, Ladies and Gentlemen:
It is a pleasure for me to be here today: firstly to pay my
respects to you, Mr. Deputy Minister, on behalf of the
International Livestock Centre for Africa, and secondly to
compliment our hosts Professor Lwoga and conference organiser Dr.
Dzowela on the organisation of this meeting, which is my first.
ILCA has been honoured and challenged by the PANESA
membership, through its elected Working Committee, by being
invited to be the technical partner complementing IDRC's
financial partnership in this network. We have welcomed this
offer because we sincerely believe we have something to offer
PANESA and its membership, but we also have something to take and
that is the wisdom of the membership in helping us to more
effectively shape our own programme. That is, we see the
interaction as two-way.
As some of you may know ILCA is presently going through a
process of self examination and programme redefinition. This is
calling for much greater interaction with national scientists,
with much of this coming through the links that networks like
this provide. The effectiveness of technology transfer can be
magnified many-fold through the networking approach.
We have a strong commitment to the dual role that forages
can play in both improved animal production through nutrition and
crop production through better soil fertility. The two are
linked through the animals that most African farms maintain.
We therefore have great pleasure in coming together to
discuss these issues here in this beautiful, and I am sure
bountiful, region of Tanzania, and we thank you, Sir, for this
opportunity.
OPENING ADDRESS - Hon. Sumaiye
Deputy Minister for Agriculture
and Livestock Development
Mr. Chairman, Distinguished Delegates, Ladies and Gentlemen:
I feel greatly honoured to have been invited to open this
workshop on behalf of Hon. Jackson Makweta, the Minister for
Agriculture and Livestock Development, which will be dealing with
three themes namely:
1. African forage plant genetic resources
2 . Germplasm evaluation
3. Extensive livestock production.
The workshop bears special significance for Tanzania since
livestock production plays an important role in the socio-economy
of the country.
Thus, on behalf of the United Republic of Tanzania and, on
my own behalf, I would like to thank PANESA for deciding to hold
this workshop in Tanzania and the organisers for the successful
preparation of the workshop. I am glad to note that the workshop
has drawn participants from the Eastern and southern African
countries and from outside Africa. This, I believe, will provide
a unique opportunity for sharing experiences for the common goal
of happiness and progress of mankind in Eastern and southern
Africa. I would like, in this connection, to take this
opportunity again to warmly welcome all the foreign colleagues to
Tanzania. It is my sincere hope that you will find your stay
comfortable and also that your programme will allow you some time
to see and learn more about Tanzania.
Mr. Chairman, I believe I do not need to remind this
audience of the seriousness of the food crisis facing Africa.
Suffice it to say that our governments are concerned about the
consistent decline in food production throughout the continent
over the past two decades, a period in which there have been
marked increases in food production in the other continents,
including the Green Revolution in Asia. Our governments are also
worried about the upward trend in population growth in Africa, as
contrasted with the downward population growth trends in all the
other continents over the same period. The combination of these
two scenarios has meant that our people, with each passing year,
are becoming poorer, hungrier and less well nourished. It has
also meant that our governments have, over the years, had to
import increasing amounts of cereal grains and livestock products
to make up for deficits in domestic food production. In short,
the resulting situation has posed a real threat to the
continent's political and socio-economic survival.
Can this threat be averted? Mr. Chairman, my answer is yes,
even though, I know, the task will not be easy. For in the words
of Dr. Carl Eicher, a prominent agricultural economist (Professor
of Agricultural Economics at Michigan State University, USA) ,
"Africa's agricultural production must be doubled within the next
15 to 20 years to keep up with population growth rates of 3 to
4%H. Mr. Chairman, it is my strong belief that solutions for
Africa's agrarian crisis will have to be found within Africa, and
 
not from outside Africa. Effective crop and livestock production
strategies for the diverse ecological and farming systems in
Africa cannot be prepared in Europe or America. Such strategies
must emerge from Africa, prepared from results of intensive,
long-term research on agricultural production constraints. For,
while it is possible and useful to transfer or adapt some
existing technology from elsewhere, much of the knowledge needed
for agricultural production will be locale-specific, and will
have to be generated locally. It is in this context that I think
highly of the objectives and activities of your network which
are:
1. To strengthen national pasture research capabilities
primarily through training and data analytical facilities.
2. To procure, disseminate and exchange pasture plant germplasm
in the region and from international sources.
3. To evaluate promising pasture plant species/strains for
adaptation and production in different ecological zones and
agricultural production systems.
4. To develop appropriate pasture production technologies that
can be integrated into the prevailing livestock production
systems .
5. To collect, disseminate and exchange information.
Mr. Chairman, PANESA is a new network aiming, to use your
own words, at improving effectiveness in pasture research in
participating countries and expediting the application of
improved technology by farmers and graziers at all levels of
management. I hope you will keep the network new in its approach
towards tackling Eastern and southern Africa's livestock and
pasture problems. I hope you will learn from the experience of
older professional networks and associations in Africa and avoid
their mistakes. For example, due to financial constraints
affecting most of our countries, the choice of research projects
undertaken by many national research institutions and
professional associations and networks has been inspired by
donors. Such research does not, if at all, address the true and
felt needs and constraints of the host countries and cannot
therefore, generate technology that will help transform our
agriculture.
Allow me to suggest, therefore, that PANESA should consider,
as a matter of urgency, establishing livestock and pasture
research priorities and plans within member countries and for the
region as a whole. This will not prevent individual researchers
from pursuing their own desires but it will, at least, provide
guidelines for the allocation of scarce national resources and
also for donors to respond to.
Another common weakness of our research systems has been
that the results of research are often not accessible both to the
end-users (livestock-keepers) and other researchers and scholars
outside the boundaries of our countries or regions simply because
they are reported in working papers and published in limited
numbers of copies. It is my hope that PANESA will concern itself
not only with the generation of improved livestock and pasture
innovations but also with the adoption of such innovations by the
livestock keeper in his environment.
Mr. Chairman, your network is meeting in Tanzania at a time
when we are intensifying our efforts in agricultural development.
One of our major concerns is to transform the livestock industry
so that it can play a greater role in the socio-economy of the
country. In particular, we would like the livestock industry to:
1. increase the per capita income of individuals engaged in
livestock production, processing and marketing, with
emphasis being placed on producers in the traditional
sector;
2. increase the volume and quality of livestock products to
service local industries using livestock products as raw
materials as well as generate exportable surpluses and
reduce import requirements;
3. increase the production, processing, and marketing of animal
protein to meet national nutritional requirements; and
4. increase the use of drought animals, thereby reducing both
the use of imported energy on the farm and the human burden.
Ruminant livestock are an important component of the
livestock industry in Tanzania. They number over 12 million head
of cattle and 10 million head of sheep and goats, distributed
over approximately 45,900 sq km of mainly natural grassland.
Over 99% of all the cattle, sheep and goats are owned by
livestock keepers in the traditional, largely subsistence,
sector. Despite the diversity of production systems, this sector
has several characteristic features:
1. Grazing resources are utilised communally.
2. There is increasing encroachment on grazing land by arable
farming which, combined with increasing livestock numbers
over the years and periodic droughts, has led to serious
land degradation.
3. Productivity of the livestock is low; with calving rates of
less than 40%, calving intervals of about 2 years, calf
mortality rates in the range of 25-40% and an offtake of
less than 10%.
When we talk of transforming the livestock industry in
Tanzania, we therefore, essentially mean a multiplicity of
transformations in the traditional livestock sector:
It means transforming the outlook of the livestock keeper
from being a subsistence to a commercial producer.
It means serious attempts to preserve the natural rangeland
and improve its water resources, and prevent bush fires
which destroy the ecology, fauna and flora.
10
It means evolving production systems that not only utilize
but also conserve grazing resources.
It means evolving production systems that integrate
livestock production with crop production.
It means developing better adapted, higher yielding and more
nutritious forage species.
It means developing a more productive animal than the one
available at present.
It means establishing more effective production - marketing
linkages in the pastoral areas.
It means evolving land tenure systems that are consistent
with improved production systems.
All I am trying to say, Mr. Chairman, is that transformation
of the livestock industry in Tanzania and, I believe in most of
the PANESA member countries, requires the successful application
of a set of technologies appropriate to the millions of small
livestock keepers in the traditional sector. The generation and
application of these technologies is, I believe, a major need
that PANESA must try to meet.
Obviously, this will require execution of long-term, multi-
disciplinary research projects in which participants include not
only research scientists, but also extension workers and
livestock keepers as partners in a common endeavour. I am
confident that this workshop, which has drawn livestock and
pasture experts from within and outside Africa, will go a long
way in formulating an agenda aimed at providing effective
solutions to livestock and pastures in Eastern and southern
Africa.
All that remains for me is to say how grateful I am to have
been given this priceless opportunity to be with you this
morning, and to wish the workshop all the success that it so
rightly deserves.
I am very happy to declare the workshop open.
11
KEYNOTE ADDRESS - J.K. Kyambwa
Commissioner for Livestock Development
Ministry of Agriculture and Livestock Development
P.O. Box 9152, Dar es Salaam, Tanzania
Mr. Chairman, Distinguished PANESA members, Ladies and Gentlemen:
It is indeed a great pleasure for me to have this opportunity
through which I can share with you our ideas on pasture work and
animal nutrition and the direction of such a discipline in
Tanzania.
Secondly this meeting has additional importance not only in
exchanging ideas but also in getting acquainted with scientists
involved in this vital discipline in different sub-Saharan
countries.
It is obvious that people coming from different parts of the
Eastern and southern African region would have different
experiences which can be shared with us. As you are all aware,
present demand for food, grazing, fibre and fuel has increased
beyond the limits that nature can provide unaided. Future demand
for the land to produce these products will be even greater. The
potential of the land to produce these products is set by the
soil and climatic conditions and by the level of inputs and
management applied to the land. Any overuse beyond these limits
results in degradation beyond economic feasibility and
subsequently yields decline, threatening the lives of the people
in the countries, as it is happening now in sub-Saharan African
countries.
Mr. Chairman, let me highlight the forage resources in
Africa as a whole with examples of the constraints prevailing in
Tanzania.
THE FORAGE RESOURCE
The forage plant resources of Africa are composed of forests,
such as the Congo forests, woodlands/brushland and savannas which
cut across the continent. In these communities, there exist
numerous grasses and herbaceous plants which form the bulk of the
livestock diets. These communities have been used to enrich the
genetic pool of pasture plants, particularly grasses for other
tropical countries.
In its natural state the African biome is quite delicate;
its management requires the application of proper ecological
principles to maintain the system. With an exception of the
Congo forest, the woodland/brushland and grassland formations are
typical rangelands of Africa and are very important to us. These
rangelands are usually semi-arid and in most cases receive annual
precipitation below 700 mm. However, their potential in
livestock production is enormous. Much of this range resource
remains under exploited.
In Tanzania there are 60 m ha of rangeland potentially
suitable for livestock production but only 6m ha are currently
utilised. Underutilisation of this resource is due to high
tsetse fly infestation, poor management practices, lack of land
12
tenure policy, inadequate water supplies, overgrazing and lack of
markets .
Tsetse Fly Infestations
About 530 m ha are infested with tsetse flies. These areas
cover the bulk of western and eastern Tanzania. Some pockets of
tsetse also occur in northern Tanzania in the national parks and
game reserves. Consequently livestock expansion has been limited
by the presence of the flies and the lack of national policy for
tsetse control.
Poor Management Practices
In the traditional livestock sector, which accounts for 99% of
the total livestock numbers in the country, animal husbandry is
poor, resulting in low calving, weaning and offtake rates, and
high pre-weaning and herd mortalities.
Lack of Land Tenure
Most of the traditional grazing lands are owned publicly and
grazed communally. There are no restraining measures on the
number of animals an individual can keep. Under this communal
tenure system, there is not enough motivation for an individual
or group of individuals to invest in pastureland improvement
programmes, or to maintain the quality of existing
infrastructure .
Inadequate Water Supplies
Unreliable rainfall coupled with seasonal rivers, has
necessitated conditioning the livestock to drink water
irregularly. Prolonged dry seasons have made it impossible for
the livestock to maintain their weights resulting in excessive
unthriftiness of the animals. In severe cases prolonged dry
seasons followed by severe forage shortages result in livestock
deaths.
Overgrazing
There is a general inability on the part of the farmer to match
the animals to the available forage. This has resulted in
overstocking and subsequently overgrazing. This trend in turn
is characterised by soil degradation. The consequences have been
low herd fertility and high mortality rates.
Lack of Markets
Although the notorious killer diseases such as rinderpest have
been under control, the quality of livestock accrued from these
efforts are still below the standard required by the meat
importing countries. In addition to this, there have been
insufficient incentives locally to force the stock owners to sell
the right proportion of their livestock. Consequently, the
offtake rates have remained below 10%.
13
Further, a general lack of basic infrastructure such as
dips, watering points, livestock development centres, markets,
cattle handling facilities, dispensaries, etc. and pursuance of
conflicting objectives on the use of land, all combined limit the
widespread distribution of livestock.
The exploitation of the African forage plant resource will
therefore depend on how fast our local scientists can realise
appropriate interventions to mitigate the above constraints. The
testing and adaptation of simple techniques to improve the
productivity of the forage resource should provide a challenge to
research scientists.
The overgrazing trend characteristic of the heavy livestock
areas in Tanzania imply that the present pasture plants cannot
withstand grazing under a communal land-use system. Research
scientists in pastures are therefore confronted with the task of
providing plants that are going to establish easily, persist
under drought conditions, provide nutrients to livestock as well
as having a soil stabilising characteristic.
In Tanzania, considerable efforts have been given to pasture
research since the early 1930s, as animal nutrition became
evident as a limiting factor to livestock production. Up to 1970
findings on cases of range productivity, manipulation of species
introduction and evaluation and pasture agronomy, had been
realised through zonal research centres. These findings were
carried out by non-African scientists on short contracts; as a
consequence, they have been of little practical application.
The bulk of our range is still under communal land use
system. Past research also suffered from lack of coordination,
direction, continuity and a responsible institution for pasture
research.
Mr. Chairman, my country has embarked on rehabilitating the
present research institutions, such as the Tanzania Livestock
Research Organization (TALIRO) , with more support in terms of
animal diseases, tsetse, trypanosomiasis and pasture research.
The Pasture Research Institute, to be established at Kilosa,
will be charged with the responsibilities of carrying out
research in better methods of range resource use to farther
animal productivity in the country. The immense responsibility
will be the focus of discussion in the three-day workshop.
Mr. Chairman, having discussed briefly the status of pasture
research in my country, you will bear with me and look briefly at
the food and livestock situations. Food and agricultural output
has not kept pace with the increasing population due to many
factors which may not be different from those operating in other
PANESA member countries. These include: lack of infrastructures
(particularly feeder roads) and low level technology in
agriculture, livestock and fisheries. These problems have become
acute as a result of long-standing input supply, ineffective
extension services, lack of skilled manpower and inability of
past research efforts to make a break in key areas.
14
Agriculture is still the backbone of our country's economy,
contributing well over 50% of the foreign exchange earning.
However, the livestock sector's contribution to the economy as
compared to arable farming, has for a long time remained
insignificant.
Food crop production during 1972-1980 increased at 5% per
annum compared with 3.5% decline in cash crop production. The
increase in food crops was due to increase in the amount of
acreage brought into production while yield per unit area
actually declined.
Tanzania's present livestock population stands at 13 m
cattle, 5.8 m goats, 3.8 m sheep, 130,000 pigs and approximately
25 m poultry. Its distribution follows the pattern of human
population and is inversely correlated with the areas infested
with tsetse flies. The main areas of cattle raising are the dry
open grasslands or wooded grasslands where precipitation is
marginal for crop cultivation. Large concentrations are in the
northern and western parts of the country and the population is
expanding in the southern highlands.
In terms of livestock per head of the human population,
Tanzania is one of the most significant livestock rearing
countries in Africa. Yet the average consumption level of animal
protein per person is only 22% to 33% as estimated by the Food
and Agriculture Organization. There is a marked difference in
consumption level between rural and urban populations, as well as
between the regions.
In order to increase the livestock sector's contribution to
the national economy and increase consumption levels of animal
protein in the short and long term, my Government plans to
develop the traditional, commercial beef and dairy herds.
DEVELOPMENT OF THE HERDS
Traditional
Some 85 to 90% of the national cattle herd is under stock owner
cultivators. Livestock development strategies under stock-owning
cultivators will emphasise both livestock and crop production.
Experience has shown that as crop production declines, livestock
population also declines and vice versa. The remaining 10 to 15%
of the national herd is owned by pastoral ists who occupy the
grasslands of Serengeti, the woodland and wooded grassland of the
Maasai Steppe. This production system is characterised by
ownership of large numbers of stock, lack of livestock movement
control, communal grazing system, traditional husbandry
practices, absence of livestock records, rudimentary
identification methods, estimation of cattle numbers and
continuous breeding. Selection and culling is seldom done.
There is no routine disease control programme, except for the
nationally coordinated campaigns, like rinderpest. Traditional
pasture and range management methods and non-specialised grazing
systems are practised. When developed, it will be able to play
its part in raising the animal protein consumption levels to 4.3
kg protein per capita.
15
The development would require proper land-use plan and
provision of livestock improvement packages. This would be
achieved by settling farmers in villages and development of
village ranches followed closely by the application of principles
of range management. The overall development of the traditional
herd require that it increases by 69% in order to meet the year
1990 national beef targets, and a further 50% increase would be
needed to meet the year 2000 demand. To meet these, annual
offtake rate targets would have to increase from 10 to 12% by
1990 and 13% by year 2000. Carcass weights would have to
increase from 100 to 120 kg.
Commercial Beef Herd
Mr. Chairman, Ladies and Gentlemen, the commercial beef herd is
only a small portion of the total cattle population run
independently by the NARCO, TSC and ASC. However, the Government
has been charged with the responsibility of:
1. Building up an improved quality beef herd which will provide
a source of stock for the traditional sector;
2. Producing improved quality beef for home consumption and for
export ;
3 . Acting as a catalyst for improved ranching techniques for
the traditional sector;
4 . Fattening feeder steers purchased from the traditional
sector, thus increasing the overall national meat
production; and
5. Providing certain technical and training services where
possible.
The development strategy followed includes expansion of the
ranches and intensification of some of the existing ranches. The
expansion strategy has realised over 15 ranches. Many of the
ranches under NARCO are low-cost extensive ranches. Some of the
mature ranches have been intensified by investments in the
development of the environment and water. Consequently, they
have reached their production targets as
average production coefficients;
manifested in their
calving rate 56%,
weaning rate 50%,
pre-weaning mortality rate 9%,
- herd mortality 7%,
- offtake rate 15.4%.
The herd offtake rate will have to increase by 15 to 20% and
carcass weight from 150 kg to 165 kg in the long term (1990-2000)
to meet the projected beef demand as required by the National
Food Strategy of 1982.
Dairy Herd
Mr. Chairman, Ladies and Gentlemen, in the analysis of the
livestock sector, the role played by the dairy herd in the
16
nutrition of our people and the economy of the country has not
been forgotten. Development of the herd which comprises
commercial dairy farms, traditional cattle owners and the
smallholder dairy farms, aims at stimulating dairy production to
substitute for imported milk products in the short term and to
develop dairy units throughout the country in the long term to
provide milk for the rural population.
The commercial dairy farms are mainly run by the parastatal
organisations including Tanzania Dairy Farming Company Limited
(DAFCO) , Tanzania Sisal Corporation (TSCO) , National Agricultural
and Food Corporation (NAFCO) and co-operative unions. The
projected milk production targets of 10 m litres per annum have
not been realized due to a lot of problems including inadequate
herd size on various farms. This implies that imported milk
products have continued to increase to meet the market demand for
milk.
The traditional sector which produces about 365 m litres of
milk per annum plays an important role in supplying milk to the
rural population. Strategies to develop this sector should
therefore focus on up-grading and selection coupled with improved
management and feeding.
The smallholder dairy farmers increase the supply of milk in
rural areas and nearby urban areas. They are very efficient when
compared with commercial producers. This system is being
encouraged by ensuring that dairy heifers are available for them
and providing essential services such as artificial insemination.
Having discussed briefly various herds of the livestock
sector and the role they have in the economy and the nutrition of
the people in the country, a trend which may resemble that in
some of your countries, I have to point out that the
contributions made by the sector have been due to efforts and
progress made in the control of animal diseases, and to some
extent to the genetic improvement of local breeds of livestock.
Mr. Chairman, the livestock sector would have been in a
position to contribute more to the economy and nutrition of the
people, if sociological, economic and animal nutrition
constraints were solved. PANESA is therefore charged with the
responsibility of generating scientific knowledge in the short
and long term to utilise the range resources for the benefit of
current and future generations. The knowledge sought should be
able to allow for multiple range resource use for sustained range
productivity. Part of this task has been initiated by the Range
Management and Tsetse Control Section in my Ministry. It is
hoped that combined efforts of this section and those of the
Pasture Research Institute to be established at Kilosa and sister
pasture research institutes in PANESA member countries will
realise findings to exploit range resource. I wish you the best
of success in this workshop.
17

SESSION I: AFRICAN FORAGE PLANT GENETIC RESOURCES
19

SOME THOUGHTS ON THE COLLECTION, INTRODUCTION AND
EVALUATION OF PASTURE LEGUMES IN ZIMBABWE
J.N. Clatworthy
Grasslands Research Station, P/B, Marondera, Zimbabwe
C. Madakadze
Henderson Research Station, Zimbabwe
Abstract
Results of screening trials with pasture legumes in Zimbabwe are
reviewed. The Stylosanthes spp. proved especially suitable for
reinforcement of native grazing on sandy soils because of their
ability to establish under adverse conditions and to persist and
spread under grazing. On heavier soils or on ploughed land a
wider range of legumes showed promise, with Desmodium spp.
outstanding in the higher rainfall areas. On an infertile sand in
a lower rainfall region only Macrotyloma axillare cv Archer and
Macroptilium atropurpureum cv Siratro were productive.
The importance of defining the likely role of legumes in
local livestock production systems and relating the evaluation
procedure to this use is stressed. This is illustrated in the
trials described by an increasing trend towards including grazing
in the evaluation process.
INTRODUCTION
The introduction and evaluation of potential pasture grasses and
legumes occupy a large proportion of the time of most pasture
research teams and it is obvious that this is a field in which
the PANESA network will be deeply involved. The pasture section
at Grasslands Research Station has, since the mid-1950' s but
particularly since 1970, been involved in the introduction,
evaluation and incorporation into farming systems of forage and
pasture legumes. During this period the section made many
mistakes, learnt many lessons and changed its ideas and methods
several times. In the hope that these experiences may be of
interest to other workers in this field in Eastern and southern
Africa, an attempt has been made to summarise the experiences in
this paper. Most of the results presented and discussed here have
been published elsewhere, and for that reason the methods
employed are not described in detail. Emphasis is placed on
comparisons of several species or strains under common conditions
rather than on comparisons of techniques involving only one or a
few strains.
METHODOLOGY
Introduction Nursery
The procedures used in the introduction nursery (Clatworthy,
1975) are similar to those used in almost all introduction
nurseries worldwide. Seeds are inoculated with Rhizobium strains
known or thought likely to be effective on that species and are
21
sown in a single row on fertilized soil. The nursery beds may be
watered to ensure establishment but not thereafter. Records are
kept of the habit, vigour, persistence, flowering date and flower
characteristics, seed set, disease and pest attack and other
features of interest for each introduction. The beds are kept
weed-free for the first two seasons but not after that. Only in
exceptional cases are plants harvested from the nursery.
Seed is reaped from all lines initially to ensure a reserve
of seed, but once this has been achieved, more seed is bulked
only from those lines showing promise.
Where feasible, cattle are allowed access to the
introduction nursery from the fourth year onwards so that the
acceptability of the legumes to stock can be observed. The
introduction nursery is intended only as a very coarse screen to
eliminate those plants which are obviously unsuited to conditions
in the area. A large proportion of the introductions will
warrant further testing to determine their agricultural
potential, and it is with the techniques which should be used for
this testing that this paper is primarily concerned.
The introduction nursery is on Grasslands Research Station
(18° ll'S, 31° 30'E, 1630 m.a.s.l., 873 mm a.a.r., granite-
derived sandy loam soils) . Other trials to be discussed were
carried out on Henderson Research Station (17° 35'S, 30° 58'E,
1290 m.a.s.l., 869 mm a.a.r., mainly silty clay loam soils
derived from metasediments) and Makoholi Experiment Station (19°
50'S, 30° 47'E, 1200 m.a.s.l., 650 mm a.a.r., granite derived
sandy soils) . In all cases fertilizers were applied, but, to
save space, details of fertilizer treatments are not given here.
Establishment
Included in this section on establishment are trials in which
seedling density or establishment percentages were determined,
even though the work was primarily intended for some other
purpose. For this reason the same experiment may be discussed
under two or more headings.
Veld Reinforcement
With the anticipation that the main benefit from the use of
pasture legumes would come from their introduction into large
areas of native grazing or veld, early trials therefore
concentrated on this aspect of the work.
Experiment 1. (Clatworthy and Thomas, 1972). Seeded at Grasslands
in December, 1964 on native grassland on an infertile coarse
sand.
Treatments comprised:
unsprayed or sprayed with 1.4 1/ha of Gramoxone
disced to 25 or 100 mm depth
left bare or mulched with 500 kg/ha of hay after
seeding
Desmodium intortum, Lotononis bainesii, L. listii,
Stylosanthes guianensis and Trifolium semipilosum.
22
The seed was broadcast on the soil surface and was not
covered, except by mulch on the appropriate plots. Surviving
plants were counted on three occasions in the ten months
following seeding.
Experiment 2. (Clatworthy, 1980) . At Grasslands and Henderson,
mainly to test the effects of grazing management on the
persistence of twelve legume strains seeded into disced veld. The
legume seeds were broadcast on the soil surface in December, 1970
and plants were counted in February and July and early in the
next rains.
Experiment 7. Seeds of 13 strains of Stylosanthes spp. and of
Siratro were sown on fertilised disced strips in native grassland
at Chibero Agricultural College (18° 06'S, 30° 40'E, 1335
m.a.s.l., 805 mm a.a.r., clay soil derived from dolerite) and at
Matopos Research Station (20° 24'S, 28° 28'E, 1340 m.a.s.l., 585
mm a.a.r., coarse sand derived from granite) in December, 1981.
Seeding rates were calculated to allow 400 viable seeds of the
Stylosanthes spp. or 150 of Siratro. The seed was dropped in a
line on the soil surface and then lightly raked. The paddocks
concerned were included in the normal grazing rotations and
seedlings were counted at the end of the establishment season.
Sown Pastures
Quite early in the work it became obvious that farmers were
mainly sowing pasture legumes on ploughed land, usually after a
period of cropping, and accordingly, increased emphasis was
placed on this aspect.
Experiment 3a. (Grant, 1976). Approximately 250 viable seeds of
Desmodium intortum. Oxley fine-stem stylo and Siratro were sown
on three dates in the 1973-74 season on the surface or in drills
10 mm deep on soil which had been fumigated with methyl bromide.
Experiment 3b. (Grant, 1976) . The same three legumes were sown
in 1974-75 late in the dry season, just before the rains or after
rain had fallen and the first crop of weeds been destroyed. Half
the area was fumigated with methyl bromide, all seeds were sown
in drills 10 mm deep and half the plots were mulched with broken
maize cobs after seeding. In both of these experiments seedings
were counted frequently, with the final count during the
following rains.
Experiment 4. (Clatworthy, in press, a). Twelve strains of
Stylosanthes spp. covering S. fruticosa, S. guianensis, S.
humilis, S. scabra, S. subsericea and S. viscosa were sown on
ploughed and disced land at Grasslands and Henderson (1973-74)
and at Makoholi (1974-75) at a constant seed rate of 10.7 kg/ha.
Five strains were common to all three sites; others were chosen
on the basis of likely adaptation to the particular site. The
seed was broadcast and lightly covered and seedlings were counted
approximately two months after sowing.
Experiment 5. (Clatworthy, in press, b) . Twelve strains of
legumes which it was thought might prove suitable for use in ley
pastures were sown on ploughed and disced land at Grasslands and
Henderson (1973-74) and at Makoholi (1974-75) . Six strains were
23
common to all sites and the others were chosen on the basis of
their likely adaptation to that area. Seeds were sown at rates
applicable to commercial use in shallow drills 0.5 m apart and
were kept weed-free during the season of establishment.
Seedlings were counted at Grasslands and Makoholi early in the
first dry season after sowing.
Evaluation Trials
Included in this section are trials in which the comparative
performance of a number of legume strains, in terms of yield,
persistence, acceptability to stock or other desirable features,
is determined over several years.
Undercutting Only
Experiment 4. (Clatworthy, in press, a) . Twelve strains of
Stylosanthes spp. were sown at Grasslands, Henderson and
Makoholi. Yields were determined by cutting to a height of 75 mm
once or twice per season.
Cutting Trials Involving Grazing
Combined with cutting
Experiment 5. (Clatworthy, in press, b) . Twelve legume strains
with potential for use in ley pastures were sown at Grasslands,
Henderson and Makoholi. Yields were determined by harvesting
quadrats from each plot twice or three times per season. After
each sampling a group of cattle were put into the experiment and
remained there until the most selected legumes had been grazed to
approximately the sampling height. After the cattle had been
removed, the apparent acceptability of each legume was scored and
the whole experiment was then slashed to sampling height.
Alternating with cutting
Experiment 6. Thirty lines of Neonotonia wlghtii and four of
perennial Glycine spp. , with Silverleaf desmodium and Siratro as
controls, were sown in replicated rows at Henderson. At the end
of the establishment season the rows were sampled for measurement
of dry-matter yield and of seed production. In the following
season half of each row was harvested four times, and half once,
and yields were again measured. The trial area was then fenced
and grazed hard by cattle for two seasons; grazing was
sufficiently severe that topping after grazing was not necessary.
In the fifth season the plants were again harvested to determine
how well they had survived the grazing.
Grazing Trials
Experiment 2. (Clatworthy, 1980). Twelve legumes were sown in
replicated blocks on disced veld at Grasslands and at Henderson.
Each trial was fenced into six paddocks which were grazed at 250
and 50 cdh in the growing season only, in the dry season only or
year-round using a simulated rotational grazing system. The
percentage frequency of occurrence of the legumes, and of other
species, was determined early in the dry season each year. The
trial at Henderson was burnt by a very hot escaped fire in July,
24
1976. In the late dry season of 1978 the herbage in both trials
was mown and removed and in February, 1979 quadrats were
harvested from each plot for yield determination.
Experiment 7. Plants of 13 Stylosanthes spp. and of Siratro which
had been sown on disced strips through grazed veld were counted
annually in the late growing season in order to measure their
persistence and spread. Notes were also made of such features as
plant vigour, degree of grazing and any obvious signs of disease
or pest attack.
RESULTS
Establishment
Veld Reinforcement
Experiment 1. The seed was sown onto moist soil, but seeding was
followed by a period of hot, dry weather. Both the herbicide
spray and deep discing improved establishment, but the effect was
not additive. Mulching had little effect although it caused an
initial increase in seedling numbers on the shallowly-disced
treatment. Table 1 shows that the legumes fell into three
distinct categories: Stylosanthes guianensis had good initial
establishment and good survival; Lotononis bainesii and L. listii
had poor initial establishment but excellent survival, while
Desmodium intortum and Trifolium semipilosum had moderate to poor
initial establishment and poor survival. As a result of this
experiment, most of the establishment trials done between 1965
and 1970 were done with S. guianensis only.
Table 1. Total surviving plants of five legumes
seeded onto disced sandveld at
Grasslands Research Station, on 24
December, 19 64.
Date of count
Desmodium intortum
Lotononis bainesii
L. listii
Stylosanthes guianensis
Trifolium semipilosum
Experiment 2 . The twelve legumes were sown on disced veld in
December, 1970. At Grasslands seeding was followed by a hot, dry
period, but the weather at Henderson was considerably more
favourable for establishment. Stands of all legumes were
therefore denser at Henderson. At Grasslands the Stylosanthes
spp. had the largest establishment percentages, especially at the
count after the dry season. At Henderson the differences in
establishment percentages were much smaller and all legumes
except the two Lotononis spp. produced good stands.
Feb 1965 May 1965 Oct 1965
37 13 5
32 32 34
10 11 9
379 305 280
66 26 12
25
Table 2. Establishment as percentage of viable seed of 12
legumes sown on disced veld at Grasslands and Henderson
Research Stations in December, 1970.
Grasslands Henderson
Leguiles Feb' 71 Julyj 71 Nov 71 Feb 71 July' 71 Feb 72
D. intortum 6.43 0.26 0 04 27 92 15.07 7 78
D. sandwicense 10.78 0.45 0 09 39 89 25.81 16 72
D. uncinatum 14.70 0.20 0 03 61 46 43.65 33 91
L. bainesii 0.10 0.00 0 01 0 58 0.39 0 39
L. listii 0.72 0.01 0 01 1 92 0.80 0 61
M. atropurpureum 25.70 6.40 6 53 34 18 28.15 23 37
N. wightii 46.46 5.07 3 58 96 13 76.13 50 08
S. fru ti cosa 14.25 15.55 15 00 18 82 15.99 15 76
S. guianensis
Scho 30.56 24.55 12 13 45 84 44.32 43 72
s. guianensis
Oxley 30.61 30.60 20 45 24 36 19.41 16 30
s. guianensis
erect 55.90 53.43 18 78 95 39 88.18 72 69
s. humilis 17.90 16.66 9 96 17 89 14.39 15 78
Sown Pastures
Experiment 3a. The three legumes were sown on three dates, either
in shallow drills or on the surface of moist soil which had been
fumigated with methyl bromide. The numbers of seedlings which
had emerged three weeks after seeding are shown in Table 3.
There was little difference in stand densities between the first
two sowings, but the third date produced the greatest emergence.
Sowing in drills had a marked beneficial effect, which was least
with Siratro and greatest with Desmodium intortum.
Table 3. Effects of date and of depth of sowing on the numbers
of seedlings of three pasture legumes three weeks after
seeding.
Date Pasture Surface Drilled 10 mm
sown legume sown deep
Oxley stylo 20 82
1 Novi 73 Siratro 52 85
Desmodium intortum 6 131
Mean 26 99
Oxley stylo 8 90
23 Nov' 73 Siratro 56 104
Desmodium intortum 9 106
Mean 24 100
Oxley stylo 26 161
4 Jan' 74 Siratro 84 161
Desmodium intortum 24 208
Mean 4 177
26
Experiment 3b. The same three legumes were sown on three dates
in shallow drills on soil which had or had not been fumigated
with methyl bromide and which were or were not covered with a
mulch after sowing. Table 4 shows the number of seedlings which
established in the various treatments. Mulching proved
beneficial, and the effect was consistent for the three legumes.
Soil fumigation also increased establishment and the effect was
most marked with Desmodium intortum, which virtually failed on
the unfumigated plots.
Table 4. Effects of soil fumigation with methyl bromide and of
mulching with 10 t/ha of broken maize cob on mean
seedling numbers of three pasture legumes.
32
Not fumigated
Unmulched Mulched
Oxley stylo 38 62
siratro 35 43
Desmodium in tortum 3 9
Mean 25 38
Fumigated
Oxley stylo 64 72
Siratro 42 56
Desmodium in tortum 45 68
Mean 50 65 58
The results of these two experiments confirm those already
presented for the veld reinforcement trials which showed that
Desmodium intortum is considerably more susceptible to stress
during seedling establishment than are Siratro or Oxley fine-stem
stylo.
Experiment 4. Twelve strains of Stylosanthes spp. (but not the
same 12 at each site) were sown on ploughed and disced land at
Grasslands, Henderson and Makoholi. Seedlings were counted
approximately two months after sowing, and at Grasslands and
Henderson differences between the strains in seedling density
were highly significant. Seedling density was not affected by
seed size but was positively correlated with germination
percentage.
When establishment was expressed as percentage of viable
seeds sown, there were significant differences between the
strains at all three sites. There was a consistent negative
correlation between establishment percentage and the germination
percentage of the seed sample. This must have resulted from the
germination in the field of some hard seed which did not
germinate in the incubator. Seed size had no effect on
establishment percentage.
Experiment 5. Twelve strains of legumes with possible potential
for use in ley pastures were sown in shallow drills on ploughed
and disced land at Grasslands, Henderson and Makoholi. At
Grasslands and Makoholi seedlings were counted approximately six
months after seeding, and at both sites differences in seedling
27
density were highly significant. Seed size had no effect on
seedling density (showing that the different seeding rates were
effective in reducing variation) , but seedling density was again
positively correlated with germination percentage. Establishment
percentage of viable seeds sown was negatively correlated with
the number of seeds per gram, showing that larger seeds had a
greater probability of successful establishment.
These counts were done in June, by which time in Experiment
2 the Stylosanthes spp. had shown superior survival at Grasslands
over the other genera under the rigorous conditions of veld
reinforcement in a season of below-average rainfall. This
advantage was not demonstrated in the present trial in the easier
conditions on ploughed land on which weeds were controlled and in
a very wet season.
Evaluation Trials
Undercutting Only
Experiment 4. Twelve strains of Stylosanthes spp. covering a
total of seven species, with five strains common to all sites,
were sown at Grasslands, Henderson and Makoholi and were
harvested once or twice per season for a minimum of three years.
The results presented in Table 5 show that in the season after
sowing, maximum herbage yields of stylo at all sites were close
to or greater than 5 t/ha of dry matter. Strains of Stylosanthes
guianensis were the highest-yielding at all sites, the solitary
exception being in the establishment season at Makoholi where
greater yields were produced by the annual species S. humilis and
S. subsericea. But the persistence of the stylos was poor, and
few lines were productive by the fourth year of the trial. Even
strains such as Oxley fine-stem stylo, which was known to persist
well under grazing at Grasslands, had declined markedly by the
fourth harvest season. From observations in grazed plots, there
appeared to be a rough correspondence between the persistence of
the strains under cutting and under grazing. Nonetheless the
results provided only very poor indication of the possible value
of the Stylosanthes spp. under grazing conditions, especially for
the annual species for which the cutting treatments effectively
prevented seeding and regeneration.
Cutting Trials Involving Grazing
a) Combined with cutting
Experiment 5. Twelve strains of legumes of potential value in
ley pastures were sown in shallow drills on ploughed and disced
land and were kept reasonably weed-free in the establishment
season. From the second season onwards, yields were determined
by harvesting quadrats from each plot. After sampling, cattle
grazed the trial until the most palatable legumes had been grazed
to approximately sampling height. The cattle were then removed,
the extent to which the different legumes had been grazed was
noted and the trial mown.
28
--qia-■T30-po.»q-UTq e-d=on'S
O"STL'ZiE-8O-Z0 EE'-9MT
i-OST-O
S-LL OiO
-L O
O-O
-i
t-
-Z -T
E-E
-O
t
O-o■xbO
L-L
L--
OO Ot
-Efr
-O
-ZZ
-
-
0
O"E
O
t
9-O
O-frO
OTO
-OT
O-9T
OfrZ9
L---fr
-Ofr -O
CT6E
-S
S-,
-S
O99
E-Z6 LOE
O-T
OTOT
-O>ST-
-8EZ
O"TiT. 0-9
1
■--w--aH
i-i
-TO
L'L6
-ZO
0"OT
L'iZZ LOT E-fr9 E-ii"O i-ES
1
9-L
t
-Ti-
OZT-T
-eEE -88
L'Z-
--
i"O O"Z9i-6
-ZE
L- 0' V E' 0' 90' 0
»
O6O' EO
■-s-
--E
OO
-8O
O- "EST-frZT O-
L-L
O O O
L-LT
-6L
O
OTOE i-- i'frT. E-E
0
O"fr
O
---ex-iao
9-ii
OSOOZ
EOXT
O-T
-9LT
L'ZT A-89O"LOE L-98»
OSEZ
E-O9 L■-
1
EOHL'T L'OSE-STZ O-OOE O"frSTE-OO-
0"ZS
L'S8b O-ET-- OO8
OE
--ZL
---Tqru-js
OT9'-
O-frT-f-TSUa-ET-b"
-L8»89
OSTE OZTT L
-LT-
E-LT9SOO"EESO
-OEOT
-SZ8ETT
Eq-lUiU.-S -■TfumnO
E--T-a-q--s
O"fEHT
--ODST-
OT--
ON-
»
T8L i»n
--
9LOT
LLOT
6A-
-awqequiTZ-aqT-e q}E-T -q-in- -oi-qs30(ui-b-/-)sq -u A as ej
At both Grasslands and Henderson Desmodium intortum was the
outstanding legume, with an average annual yield of more than 8
t/ha at Grasslands and 6 t/ha at Henderson. The other two
Desmodium strains were also highly productive, as was Archer but
Archer proved considerably less acceptable to cattle than the
other legumes.
At Henderson, fire burnt the whole trial after the second
sampling season and affected all the legumes, although none was
completely eliminated. Yields in the final season were very low
compared with previous years, perhaps because the poorly drained
site became waterlogged in a year of exceptionally heavy rain.
Work in Australia has shown that few of the pasture legumes can
tolerate prolonged waterlogging (Mclvor, 1976) .
At Makoholi only Archer (1.7 t/ha) and Siratro (1.6 t/ha)
were productive. Results at this site were very different from
those at Grasslands and Henderson. Makoholi would normally be
regarded as having submarginal rainfall for sown pastures but
this trial coincided with a period of above-average rainfall, and
a more likely reason for the failure of the legumes would be the
inherently low soil fertility.
The effect of grazing in these trials was very small as the
plots were topped immediately after grazing, and for all
practical purposes these could be regarded as pure cutting
trials. In ley pastures the main selection criterion is to
produce the maximum yield of herbage over a relatively short life
of the pasture, and that can be measured sufficiently accurately
in cutting trials. The main advantage from the use of animals in
this trial was that it demonstrated clearly the very low
acceptability of Archer.
(b) Alternating with cutting
Experiment 6. Thirty lines of Neonotonia wightii, four of
perennial Glycine spp. and Siratro and Silverleaf desmodium were
sown in replicated rows at Henderson. When sampled at the end of
the establishment season, the highest yielding legumes were
Silverleaf desmodium and three strains of N. wightii cv Cooper,
CPI 25702 (from which Cooper was selected) and an introduction
from Angola which was almost certainly also Cooper. In the
second season, half of each row was harvested four times and the
other half at the end of the growing season only. The same four
legumes were again the highest-yielding. The trial was then
fenced and grazed by cattle for two seasons. In either season
there was a detectable difference in the acceptability of the
various strains, and in both seasons grazing was sufficiently
severe that slashing was unnecessary.
In the final year the legumes were harvested twice to
measure their persistence after the period of grazing. At this
harvest the three Cooper lines and Silverleaf desmodium produced
yields below average, and the three highest-yielding lines were
all Neonotonia wightii strains of Zimbabwean origin. These were
very similar in their visual characteristics and in flowering
time, and it is intended to combine them and release the mixture
as a pasture cultivar.
30
This trial illustrates the need to continue screening for a
reasonable period before conclusions are reached. It is tempting
to ascribe the failure of the Cooper lines to a lesser ability to
withstand grazing, but, in the absence of a cutting-only control,
this is not justified and may have been merely an effect of time.
At least, though, we can be confident that our selected lines
will persist for the normal life of a pasture, even under severe
grazing pressure.
Grazing Trials
Experiment 2. Twelve legumes were sown in six replicated blocks
on disced land at Grasslands and Henderson. Each legume was sown
in two plots in each block. The blocks were fenced separately and
were grazed at two stocking rates in the dry season only, the wet
season only, and year-round using a simulated rotational grazing
system.
At Grasslands only the five stylo strains and Siratro
established in reasonable density, and of these Oxley fine-stem
stylo proved outstanding. It persisted well and spread into
plots in which it had not been sown and appeared particularly
favoured by grazing during the growing season.
At Henderson a much wider range of legumes established and
persisted throughout the trial. Schofield stylo initially was
outstanding but it appeared short-lived and Silverleaf desmodium
was the most impressive legume in the middle stages. The fire in
1976 affected all the legumes, although none was completely
eliminated, but Lotononis bainesii and L. listil recovered
rapidly and were the most prominent legumes at the end of the
trial. Silverleaf desmodium was especially favoured by grazing
in the dry season only while the Lotononis spp. grew best in
plots which were grazed in the growing season.
The large differences between the results from the two sites
in this trial emphasise the need for multi-site testing covering
a range of soil and climatic conditions if results are to have a
widespread applicability. The results also emphasise the need for
the long continuation of trials and also for recurrent natural
hazards, such as fire, to be incorporated in the design. The
season of grazing treatments proved useful; doubling the stocking
rate did not. This type of trial is of value in determining
which strains justify large-scale testing in animal production
experiments.
Experiment 7. Seeds of a range of Stylosanthes spp. and of
Siratro were sown on disced strips in veld and grazed in normal
farm practice. Surviving plants were counted at intervals,
normally in the late growing season of each year. The results
are a combination of the ability of each strain to establish and
to persist. At nearly all sites the greatest number of seedlings
were from S. guianensis strains, but S. scabra proved hardier and
had a greater percentage survival. At Grasslands in 1984 there
was evidence of increases in stand density of S. fruticosa, S.
viscosa and all S. scabra lines, and this trend continued into
1986. All of the stylos were grazed and even the notoriously-
unpalatable S. viscosa appeared to be well eaten by the late dry
season.
31
U H U
32
I H H
CI N >■ 00
33
DISCUSSION
In Zimbabwe, as in most parts of the tropical and subtropical
world, the main legumes used in pastures are of Central or South
American origin. Few locally collected legumes were included in
the trials described here: of these, Stylosanthes fruticosa could
prove of importance in the medium-rainfall parts of Zimbabwe and
Neonotonia wightii on heavier soils in the better-watered areas.
Indigenous strains of Alysicarpus rugosus, Lotononis listii,
Macrotyloma axillare and Teramnus labialis (all of which
performed well in one or more of the trials described in this
paper) also occur in Zimbabwe. However, diseases and pests have
posed problems with indigenous strains, especially when grown in
pure stands for seed production and may limit their usefulness
compared with introductions.
For national institutions with limited resources the most
logical first stage is to introduce strains of pasture plants of
proven or suggested value under similar conditions elsewhere and
to test those. Inevitably research and extension workers and
farmers will collect, on a casual basis, seeds of grasses or
legumes which catch their attention, usually because of
productivity, growth habit or acceptability to stock. This is
healthy and can pay big dividends, but it is doubtful if a heavy
commitment to pasture plant collection is a wise use of national
resources and it is disturbing that emphasis has been given to
plant collection at the IDRC/PANESA/ILCA training course in
"Forage Plant Introduction and Evaluation" in October, 1986.
There are excellent chapters on pasture plant evaluation in
several books (Shaw et al . , 1976; 't Mannetje et al . , 1976;
Cameron and Mclvor, 1980; Jones and Walker, 1983). The stages
which have evolved in this paper largely echo those which are
outlined by those authors, particularly in regard to the need to
consider the likely use of the plant from an early stage in the
evaluation process. This is especially relevant in Africa, with
its multiplicity of livestock production systems, ranging from
intensive cut-and-carry milk production to extensive and largely
uncontrolled grazing of rangeland. For cut-and-carry system the
yield of the forage is the most important criterion, and this can
be adequately assessed in simple cutting trials, ideally combined
with some assessment of the nutritional value of the herbage.
For rangeland improvement, the criteria of success depend more on
persistence and spread under grazing, which can really only be
measured in situ. The likely use of the forage plants within the
local farming systems will need to be borne in mind in planning
the evaluation trials within the PANESA network. Except in the
most basic stage, that of the observation nursery, the type of
accession chosen for testing, and the form of treatments imposed,
should be defined to a great extent by the role the plant is to
play in the livestock system. In many cases in Africa the
farming systems, the roles of livestock within them and the
potential benefits which can be derived by the growing of forage
plants, are not yet clearly defined and this makes the task of
the researcher dealing with small-scale farming systems more
difficult than that of the worker with well-defined commercial
livestock operations. The authors believe, though, it should be
possible to postulate suitable roles and types of forage plants
for each area and to derive likely "short lists" of promising
34
lines for more detailed evaluation. They query whether time will
be best spent testing a complete and common spectrum of pasture
grasses and legumes under the full range of conditions covered by
the PANESA region as that seems a complete negation of the
region's collective experience built up over so many years.
A major deficiency of the experiments on ploughed land
described in this paper is that in all cases the legumes were
sown in pure stands. Except when intended as protein banks,
either grazed in situ or cut and conserved, there are
considerable advantages to combining a grass with the legume, and
the selection of productive and compatible mixtures is a complex
and demanding task. Again, though, a great deal can be learnt
from experience under similar conditions elsewhere. In
particular we in PANESA need to study the methods used, and the
lessons learnt, by the extensive and growing Tropical Pastures
Evaluation Network (RIEPT) in South America and the neighbouring
regions. Unfortunately the manuals of procedure produced by
RIEPT are all in Spanish, which limits their usefulness in
Africa, but it would be of great value if a senior member of that
network (and also possibly of the South East Asia/Pacific Forage
Research and Development Program) attend future PANESA workshops.
REFERENCES
Cameron, D.G. and Mclvor, J.G. 1980. Evaluation. In: R J
Clements and D G Cameron (eds) , Collection and testing of
tropical forage plants. CSIRO, Australia. pp. 71-87.
Clatworthy, J.N. 1975. Introduction and preliminary screening of
pasture legumes at Marandellas, Rhodesia, 1967-73.
Proceedings of the Grasslands Society of Southern Africa
10:57-63.
Clatworthy, J.N. 1980. Screening legumes for veld reinforcement.
Proceedings of the Grasslands Society of Southern Africa
15:111-116.
Clatworthy, J.N. (in press, a). Establishment and yield of
pasture legumes under cutting in Zimbabwe. I. Stylosanthes
spp. Zimbabwe Journal of Agricultural Research.
Clatworthy, J.N. (in press, b) . Establishment and yield of pasture
legumes under cutting in Zimbabwe II. Legumes for ley
pastures. Zimbabwe Journal of Agricultural Research.
Clatworthy, J.N. and Thomas, P.I. 1972. Establishment of
Stylosanthes guianensis in Marandellas sandveld.
Proceedings of the Grassland Society of Southern Africa.
7:76-83.
Grant, P.J. 1976. Some factors affecting the field establishment
of sub-tropical pasture legumes in Rhodesia. Proceedings of
the Grasslands Society of Southern Africa 11:97-102.
Jones, R.J. and Walker, B. 1983. Strategies for evaluating
forage plants. In: J G Mclvor and R A Bray (eds) , Genetic
resources of forage plants. CSIRO, Australia. pp. 185-201.
35
't Mannetje, L. , Jones, R.J. and Stobbs, T.H. 1976. Pasture
evaluation by grazing experiments. In: N H Shaw and W W
Bryan (eds) , Tropical pasture research: Principles and
Methods. Commonwealth Agricultural Bureau, Bulletin 51:194-
234.
Mclvor, J.G. 1976. The effect of waterlogging on the growth of
Stylosanthes guianensis . Tropical Grasslands 10:173-178.
Shaw, N.H., Jones, R.M. , Edye, L.A. and Bryan, W.W. 1976.
Developing and testing new pastures. In: N H Shaw and W W
Bran (eds) , Tropical pasture research: Principles and
methods. Commonwealth Agricultural Bureau, Bulletin 51.
36
REVIEW OF RANGE AND PASTURE PLANTS IN BOTSWANA
Gubungano Tacheba and Arabia Moyo
Animal Production Research Unit, Botswana
Abstract
Botswana vegetation is arbitrarily classified into the hardveld,
sandveld and Okavango deltaveld. In spite of this arbitrary
subdivision some species are specific to a particular vegetation
type. The species vary generally in their nutritive value; woody
plants (12.63% CP; 47.60% DMD) are relatively higher than grasses
(5.49% CP and 40.61% DMD). Within a species there is a seasonal
variation in nutritive value. For instance Panicum maximum is
higher in the wet season (10.4% CP; 60% DMD) than in the dry
season (8.7% CP; 36% DMD).
Agronomic trials were initiated in the 1970/71 season, but
these were discontinued and resumed in 1976 with a broader
objective of screening a wide range of germplasm suitable for
pasture production. These trials were terminated in 1982 only to
resume in 1986. It was found that Macroptilium atropurpureum,
Leucaena leucocephala, Lablab purpureus and Cenchrus ciliaris
were the most promising species. Current collaborative trials
with ICRISAT and ILCA involve the screening of varieties of
millet and sorghum, some forage grasses and legumes. In future
focus will be directed to indigenous species as well.
Where fertilizer application tests to range were conducted a
yield increase was recorded, but economic feasibility indicated
that it would be of no practical value in the immediate future.
INTRODUCTION
Botswana, extending from 18°S to 27°S and 20°E and 29°E is
situated on the Southern African Plateau with a mean altitude of
1,000 m.a.s.l. It covers 582,000 sq. km with marked
topographical features mainly in the east on the hardveld, where
the soil is capable of supporting arable agriculture. The
hardveld is a narrow strip covering about 20% of the country. In
the northwest is the Okavango Delta. The rest of the country is
covered by the Kalahari sand.
Rainfall averages 450 mm per annum. There is great
variation in amount and distribution from location to location
and from year to year. The rainfall is highest in the northeast
at 650-700 mm decreasing to a minimum of 100-200 mm in the
southwest. Rainfall occurs in October-April and over 90% of rain
occurs in November to March. Beginning and end of rains are
equally unreliable. Early planting might suffer from drought, or
late planting may never get established.
Evaporation often exceeds 2 m per annum, and humidity is
low. There is quite a large difference between morning humidity
and afternoon humidity. Temperatures reach 31°C in December and
January, and daily variation averages 13°C. Frost is common in
winter in the south.
37
Livestock form an important industry accounting for 2 0% of
the foreign exchange economy. In 1982 the cattle population was
3 million; this figure dropped to 2.5 million by 1986 due to
drought. However, the cattle population has been increasing as
drought years are often followed by good rain years. Cattle
density is high in the hardveld; however, in the past few years
the Kalahari has been intruded upon by large numbers due to
opening of more ground waters. The Okavango delta is still
lightly used due to the trypanosomiasis-transmitting tsetse fly,
and in other areas range degradation is widespread, especially
around watering points.
There are a number of reports that categorise the vegetation
of Botswana, notably those by Wear and Yalala (1971), DVH (1979)
and Timberlake (1980) . There are three major vegetation types in
Botswana. The first of these, the hardveld type, is
characterised by grasses Panicum coloratum, Themeda triandra
Cenchrus ciliaris, Enneapogon cenchroides , Eragrostis superba,
Chloris gayana, Heteropogon contortus, Brachiaria deflexa and
Bothriochloa insculpta. It is associated with such woody plants
as Acacia karroo, Albizzia anthelmintica. Balanites aegyptiaca,
Colophospermum mopane, Combretum apiculatum, Grewia bicolor, G.
flavescens, G. monticola, Lonchocarpus capassa and Terminalia
prunioides. The second type which is specific to the Kalahari
sandveld consists of grasses, Eragrostis pallens, Stipagrostis
uniplumis , Anthephora pubescens, Schmidtia pappophoreides and
Megaloprotachne albescens associated with woody plants Grewia
avellana, G. retinervis, Lonchocarpus nelsii and Terminalia
sericea.
The last type is specific to the Okavango Delta, and it
includes the grasses Phragmites communis, Dichanthium papillusum,
Panicum repens, Andropogon encomus, Echinochloa pyramidelis,
Imperata cylindrica, and woody plants Ficus verruculosa and
Hyphaene crinata.
The purpose of this paper is to discuss the forage values
and review some of these forage species that have been tried for
pasture in Botswana.
RANGE AGRONOMY METHODOLOGY
The history of range agronomy in Botswana goes back to 197 0 for
only a short period (FAO, 1971) . The work resumed in 1976 and
continued up to 1982 when there was a cessation of fodder plant
screening from 1983 to 1985. Finally the work resumed again in
1986. The range agronomy section has tested promising species
into farming systems programmes with local farmers. These were
primarily the introduced species Stylosanthes humilis, Neonotonia
wightii, Macroptilium atropurpureum, Stylosanthes guianensis and
Rhynchosia sublobata. It has also determined the nutritive value
(CP% and DMD%) of a variety of range grass and woody plants in
some vegetation types.
At some stage in the screening programme (1976/77) , the
increase in rangeland yield by applying fertilizer was tested at
Morapedi in an Acacia erioloba tree savanna on the Kalahari
sandveld. A combination of nitrogen and phosphorus was applied
38
in the form of urea and superphosphate respectively. Urea was
applied at four levels (0, 50, 100, 150 kg/ha) and superphosphate
at two levels (0, 33.2 kg/ha). The fertilizer work was extended
to Marale in 1979 with a vegetation cover composed of Acacia
nigrescens/Combretum apiculatum tree savanna in the hardveld.
Three fertilizers, nitrogen, phosphorus and potassium in the form
of ammonium sulphate, superphosphate and potassium chloride
respectively were applied in combinations of two. Levels of
ammonium sulphate were 0, 250, 500 and 750 kg/ha; those for
superphosphate were 0, 400 and 800 kg/ha, while those for
potassium chloride were 0 and 50 kg/ha.
RESULTS
The nutritive values in terms of crude protein and dry-matter
digestibility of some range grasses in the hardveld and the
Kalahari sandveld are shown in Tables 1 and 2 respectively. The
hardveld grasses are generally of a higher nutritive value than
those of the Kalahari sandveld. These data are means over a 29-
month period. However, within this period it was observed that
protein values were quite high during the wet season, resulting
in good livestock performance on them. During the dry seasons
the range grasses were characterised by low protein values, lower
than maintenance requirement, and thus animals lost weight.
Table 1. Crude protein (CP) and dry-matter
digestibility (DMD) of some range
grasses in the hardveld, values in %.
Species cp DMD Source
Aristida congesta
Panicum maximum
Eragrostis rigidior
Enneapogon cenchroides
Heteropogon contortus
Brachiaria nigropedata
Urochloa trichopus
Bothriochloa insculpta
Digitaria eriantha
4.9 3 7 APE
8.8 49 "
3.6 33 "
5.3 40 "
4.5 39 "
6.8 43 "
8.0 46 "
6.4 39 "
6.3 47 "
1977
Means 6.1 41
Note: APRU 1977 values are averages of 29 months.
Tables 3 and 4 show the nutritive values of some wood range
plants in the hardveld and the Kalahari sandveld, respectively.
There are no major discernible differences between the nutritive
values of hardveld and sandveld species. However, the woody
species are generally higher in nutritive value than grasses in
both vegetation types. This means that browsing is an important
supplementation feed resource for these animals on range.
Results of the exotic plants screening work indicated that
Stylosanthes spp. and Macroptilium atropiirpureum made good
39
Table 2. Crude protein (CF) and dry-matter
digestibility (DMD) of some range
grasses in the Kalahari sandveld,
values in %.
Species CP DMD Source
Aristida congesta
Eragrostis lehmanniana
Eragrostis pallens
Stipagrostis uniplumis
Anthephora pubescens
Schmidtia pappophoroides
Brachiaria nigropedata
Pogonathria squarrossa
Digitaria eriantha
4 4 37
4 0 33
3 9 32
4 2 32
6 0 49
5 5 48
6 5 42
3 9 38
5 8 47
APRU, 1977
Note: APRU 1977 values are averages of 29 months.
Table 3. Crude protein (CP) and dry-matter
digestibility (DMD) of some range woody
plant leaves in the hardveld, values
in %.
Species CP DMD Source
Acacia tortilis
Acacia karroo
Balanites aegyptiaca
Boscia albitrunca
Boscia foetida
Colophospermum mopane
Combretum apiculatum
Combretum hereoense
Grewia bicolor
Grewia flava
Lonchocarpus capassa
Tarchonanthus camphoratus
Terminalia sericea
Zizipbus mucronata
14 6 43 APRU 1985
12 8 40 M
9 7 54 11
16 7 na FAO, 1971
8 0 na n
8 1 na ii
9 7 na it
10 7 50 APRU 1985
12 4 28 ii
12 5 49 n
12 4 na FAO, 1971
10 9 53 APRU 1985
7 7 na FAO, 1971
12 0 61 APRU, 1985
Means 11.3 47*
Note: APRU 1985 values are averages of 12 months
and FAO 1971 are samples of April/May 1970.
OMeans of eight species only.
establishment and persisted for years. Amongst the five Cenchrus
ciliaris cultivars tested (Table 5) , Western Australia did poorly
in all the years. Mahalapye yields were better for Cenchrus sp.
than those at Sebele. Mahalapye is in the central hardveld with
average rainfall of 450 mm per annum, but Sebele is in the
southern hardveld with a rainfall of 500 mm. At Sebele the
40
Table 4. Crude protein (CP) dry-matter digestibility (DMD) of
some range woody plant leaves in the Kalahari sandveld,
values in %.
Species CP DMD Source
Boscia albitrunca
Grewia flava
Grewia retinervis
Lonchocarpus nelsii
Tarchonanthus camphoratus
Terminalia sericea
Ziziphus mucronata
17.9 57
15.2 40
16.3 55
21.2 45
10.0 33
7.0 37
19.4 69
Skarpe
Means 13.1 48
Note: Values are averages of 2 months, February and June
1977-1979.
Skarpe, C. Gaborone, Botswana, unpublished data.
Table 5. Yield of some pasture species at Mahalapye and Sebele
in kg per ha in 1978, 1979 and 1980.
Mahalapye Sebele
Species 1978 1979 1980 1978 1979 1980
Cenchrus ciliaris
cv Molopo 1192 4749 3948 4700
cv Biloela 2803 4153 3189
cv Gayndah 1395 3623 1161
cv U.S.A. 2790 4115 2698 5300
cv Western Aust. Trace Trace Trace 0
Cenchrus sentigerus 670 Trace* Trace
Cynodon dactylon
cv Giant Bermuda 799 1915*
Eragrostis curvula
cv Ermelo 4287 2509* 2709
Urochloa mosambicensis 2855 2015*
Panicum coloratum
cv Bambatsi 2513* 1900
Siratro 1704 2802* 3528 1500
Phasey Bean 1093 2086* 7130
Leichardt Dolichos 4165 3500
Desmodium intortum
D. uncinatum
Dolichos lablab 5168 5000
Leucaena leucocephala
cv Peru 550 2100
cv Cunningham 420 2266
cv Hawaian Cross 410 3233
cv Hawaian Giant 2000
OProduction from plots planted the previous year.
Source: APRU (1978, 1979, 1980, 1981)
41
Cenchrus sp. established poorly due to heavy weed competition.
The other grasses Cynodon dactylon, Eragrostis culvula, Urochloa
mosambicensis and Panicum coloratura showed lower forage yields
due to weed ingestation. There was however reasonable yields
from Eragrostis culvula, Urochloa mosambicensis and Panicum
coloratum especially at Mahalapye.
Amongst the legumes, Macroptilium atropurpureum cv Siratro,
Lablab purpureus cv Leichardt and Macroptilium lathyroides
established well and gave reasonable yields. Based on this work
Lablab purpureus is now being grown in dairy farming systems
project areas.
Leucaena leucocephala was established at Morale (near
Mahalapye) . The plant is susceptible to frost although it has
shown the ability to regrow once weather conditions warm up. The
major problem interfering with its successful establishment is
white termite attack and wildlife and goats that hammer it hard,
hence affecting its maintenance under range.
Results of the fertilizer trials indicated no significant
increase in range productivity to levels of urea application
beyond 50 kg/ha when phosphorus was held at zero. But when
phosphorus was increased to 3 32 kg/ha the yield was increased at
the other levels of urea application (Figure 1). Figures 2, 3
and 4 show yields of three grasses (Digitaria, Eragrostis and
Urochloa spp.) in response to fertilizer. Nitrogen reduced yield
at 500 kg/ha and beyond when superphosphate was held at zero
(Figure 2). This response was similar to that of ammonium
sulphate and potassium chloride, when ammonium sulphate was
increased beyond 500 kg/ha and potassium chloride was held at 50
Figure 1. Response of rangeland yield to nitrogen and phosphorus.
Wal«M(tg)
5500r
3000
2000 ■
U2
2 3
Suptrph■tphatt
42
kg/ha yield was reduced (Figure 3) . Figure 4 shows that an
increase in superphosphate increased yield more when potassium
chloride was held at zero.
Figure 2. Response of rangeland yield to nitrogen and phosphorus.
Wtljht (kg)
3500,-
30 00
S o Suptrphoiphott
2 00 0
 
* 3
Ammooium tulphaft
Figure 3. Response of rangeland yield to nitrogen and phosphorus.
 
l 300-
i000
2 3
A*Monium ivlphata
PC2
Figure &• Response of rangeland yield to nitrogen and phosphorus.
Weight (Kg)
SSOOr
5000
PC = Polonium cMoridl
2500
2000
i000
 
Super photphota
DISCUSSION AND CONCLUSION
For Botswana, a country that has suffered from a succession of
droughts, there appears to be limited scope for planted perennial
pasture plants. Annuals render themselves better to easy
incorporation in arable farming systems than perennials. The
latter might pose problems of eradication in cropping systems.
Interest in fertilizer trials ought to be reduced as the
economic feasibility has indicated that fertilizer application to
the range is unlikely to be of practical value in the immediate
future. The current trends in research are geared to
collaboration with ICRISAT and ILCA to come up with various
cultivars of fodder millets, fodder sorghums and some specialised
grasses and fodder legumes that would easily fit in small-scale
crop-livestock farming systems. Indigenous fodders ought to be
an integral part of this work.
REFERENCES
APRU (Animal Production Research Unit) . An integrated programme
of beef cattle and range research in Botswana. Ministry of
Agriculture, Gaborone, Botswana.
APRU. 1978. Livestock and range research in Botswana.
of Agriculture, Gaborone, Botswana.
Ministry
44
APRU. 1979. Livestock and range research in Botswana. Ministry
of Agriculture, Gaborone, Botswana.
APRU. 1982. Livestock and range research in Botswana. Ministry
of Agriculture, Gaborone, Botswana.
APRU. 1985. Livestock and range research in Botswana. Ministry
of Agriculture, Gaborone, Botswana.
DHV. 1971. Countrywide Animal and Range Assessment Project.
Vol. 3. Ministry of Commerce and Industry, Gaborone,
Botswana.
FAO. 1971. Range ecology in Botswana. Technical document No. 2.
FAO , Rome .
Timberlake, J. 1980. Vegetation map of South East Botswana.
Ministry of Agriculture, Gaborone, Botswana.
Wear, P.R. and Yalala A. 1971. Provisional vegetation map of
Botswana. Government Printer, Gaborone, Botswana.
45
DISTRIBUTION OF POTENTIAL FORAGE PLANTS IN MALAWI
R.D.C. Msiska
Lunyangwa Research Station
P.O. Box 59, Mzuzu, Malawi
J.T. Munthali
Chitedze Research Station
P.O. Box 158, Lilongwe, Malawi
and
B.H. Dzowela
ILCA, P.O. Box 46847, Nairobi, Kenya
Abstract
The occurrence of grass species of Panicum, Pennisetum, Chloris,
Urochloa, Setaria and Brachiaria and legume species of
Neonotonia, Macrotyloma, Dolichos, Sesbania and Aeschynomene in
different physiographic regions is described. Emphasis has been
put on the climatic variables which tend to influence species
distribution and the physiological maturity among legume species.
INTRODUCTION
Malawi, which is 94,396 sq. km, is dominated by the final section
of the East African rift valley system which fractured the high
south Central African plateau to form a deep trough occupied by
Lake Malawi. The lake which is 570 km long and 16-80 km wide is
drained by the Shire River following the line of rift into the
Zambezi River (Agnew and Stobbs, 1972) . The faulting and
subsequent erosions created diversified topographies which have
been modified by the influence of the lake, intertropical
convergence zone and the southeast Trade Winds to give rise to
tropical and sub-tropical environments. These environments
contain a variety of grasses and legumes which could be utilised
in our farming systems.
So far there has been no extensive plant collections in
Malawi, and as a result detailed distribution of forage plants is
rather sketchy. However, the few vegetation documents available,
the 1981 collection strip made by the senior author to some parts
of the central and northern regions and the knowledge of the
authors has formed the bases of this paper. Consequently, some
documents could have been overlooked out of ignorance.
The physiographic regions delineated by Pike and Rimmington
(1965) (Appendices 1 and 2) have been used to describe the
distribution of the forage plants. The climatic variables in
these regions differ a lot and warrant a separate description to
give a prospective collector all the necessary information.
46
INFLUENCE OF PHYSIOGEOGRAPHIC REGION
Lower Shire Valley
This area is a narrow extension of the Mozambique Coastal Plain
into Malawi with altitudes ranging from about 120 m in the north
to 40 m in the south. The area lies in the rainshadows of the
Thyolo escarpment in the north-west, the Nsanje Hills in the
south and the watershed between the Zambezi and the Shire rivers.
The average temperature and rainfall are 30°C and 630 mm
respectively.
Urochloa mosambicensis, Panicum spp. Clitoria ternatea,
Rhynchosia spp. and Sesbania spp. are found in this area.
Urochloa mosambicensis is widespread in the valley especially on
grassy areas occurring on sandy soil. This ecotype is superior
to U. pullulans , a perennial found elsewhere in the country at
altitude below 1500 m which tends to scorch whenever there is dry
spell during the rainy season. Setaria and Panicum, Sesbania and
Aeschynomene spp. grow along streams and dambos. Clitoria
ternatea is found around Ngabu on gravelly soils.
Mulanje Mountains - Shire Highlands - Zomba Mountain
This region whose average altitude ranges from 610 m to 3000 m
has an extended rainfall season as a result of the moisture
carried by the south-east Trade Winds from the Indian Ocean.
Average rainfall is about 3000 mm in Mulanje and Zomba and
decreases to about 1000 mm in some parts of the Shire Highlands.
Panicum maximum occurs along roads, streams and on scattered
hills of the Shire Highlands which contain Acacia woodii and
Albizzia gummifera vegetation. The commonest perennial wood type
has broad dark green leaves and is found growing in the heavily
disturbed areas. The giant types are found mostly on the
windward side whereas the medium types tend to occur on the
leeward side of these highlands. Pennisetum, Setaria, Cynodon
and Neonotonia spp. also occur in this area.
Central Region Lakeshore Plain
This area extends from Salima in the south to Nkhota-kota in the
north. It is a relatively dry area due to the prevalence of the
southeast Trade Winds which lose their moisture on the Mulanje,
Shire and Zomba Highlands. The altitude is about 500 m, and the
average temperature and rainfall are 23°C and 1270 mm
respectively. The area is characterised by seasonal swamps and
Acacia-Adansonia-Cordyla-Sterculia vegetation - interspersed with
alluvial fans along the Linthipe, Lingadzi, Bua, Kaombe, Dwangwa
and Dwambazi rivers.
Panicum, Pennisetum, Cynodon, Brachiaria, Setaria,
Neonotonia and Aeschynomene spp. occur in this area. Panicum
maximum is found along river and swamp fringes. The P. maximum
growing along the Dwangwa river could even be planted from setts.
The distribution of P. maximum seems to indicate that seed
originally came from higher areas of the Dowa-Ntchisi and Viphya
highlands. This hypothesis is supported by the absence of P.
maximum along rivers and dambos from the Chitala river going
southwards, a section which does not have corresponding highlands
further inland. The Cynodon spp. have head smut which limits
their chance of being suitable for collections. Neonotonia
wightii along the entire area shows a visual gradation from
diploid to tetraploid types from south to north corresponding to
the amount of rainfall received.
Dowa-Ntchisi Highlands
The area has some evergreen forests at altitudes of 17 00 m due to
the influence of orographic rains brought by the southeast Trade
Winds. Panicum maximum, Setaria spp. Pennisetum purpureum and
Neonotonia wightii are found in this area.
Past collections from this area yielded Panicum maximum cv
Ntchisi. This giant guinea grass has broad bluish green leaves
which makes it suitable for cut-and-carry, silage and grazing
(Addy and Thomas, 1976) . Ntchisi panic has found its way into
Kenya (Bogdan, 1977) and Zambia (Anon., 1979). Ntchisi panic
like most guinea grasses is mostly propagated by splits because
seed harvesting is hampered by variability in seed maturation on
the same panicle which quickly sheds or are eaten by birds.
Central Region Plateau
The central region plateau is the largest continuous surface in
Malawi extending from the margin of the rift valley in the east
to the Zambian border in the west. The altitude ranges from 970
to 1300 m with an average temperature and rainfall of 20°C and
800 mm respectively.
The vegetation consists of Acacia-Piliostigma-Combretum and
Brachystegia-Julbernardia savannah woodlands on the Lilongwe and
Kasungu plains respectively. Panicum maximum occurs along major
rivers of Lilongwe, Lingadzi and Bua whereas Brachiaria spp.
Setaria sphacelata, S. longiseta and S. splendida are found both
on higher ground and dambos. Setaria palustris is found on
dambos only. Most of these Setaria spp. are attacked by head
smut (Tiletia echnosperma) .
Macrotyloma spp. occurring on the Lilongwe plain consist of
both annual grabrescent and pubescent types . The grabrescent
type climbs onto the hyparrhenia and maize plants. This climbing
attribute onto the latter could be utilised to improve the
quality of maize stover for stall-feeding. Macrotyloma axillare
occurs on the Kasungu plain but its small leaves would limit its
usefulness as a forage legume. Neonotonia spp. which matures
towards the end of the rainy season occurs throughout this region
especially on termite mounds.
viphya Lakeshore and Highlands
The altitude ranges from 500 m on the lakeshore to about 1300 m
on the highlands. This region, apart from the Mulanje-Shire-
Zomba highlands, receives the next highest annual rainfall of
over 1500 mm. The southeast Trade Winds, after blowing over the
lake for a long distance, get recharged with moisture which falls
as orographic rain. As a result of this influence, this region
has an extended rainy season which supports growth of low
48
altitude evergreen forests on the lakeshore and on the
highlands. The Luweya catchment area, the area between
Chintheche and Nkhata-Bay and extending westwards to Mzuzu, has
the largest variety of Panicum and Brachiaria spp. compared to
any other area in Malawi. In contrast, the northern Viphya
Highlands contain relatively few of these grasses probably due to
the rainshadows created by Choma and Chimaliro hills within the
highlands. Setaria and Pennisetum spp. also occur in this area.
Most of these species are found along streams and on the
extensive Limphasa dambo. The heavily smutted Cynodon spp. are
found on the lakeshore only.
Neonotonia, Dolichos and Macrotyloma spp. occur in this
region. Among Neonotonia spp. are found both early and late
maturing ecotypes. The latter flower as late as August at the
onset of the dry hot season when wild fires are very common. The
slender pubescent Macrotyloma sp. is found on the lakeshore along
gullies whereas Macrotyloma axillare and the shrub type
Macrotyloma sp. occur on the highlands. M. axillare distribution
seems to indicate that it thrives well on well drained sandy loam
soil.
The Henga - Lower Kasitu Valley
The region lies between the Viphya to the west and Nyika plateau
and Njakwa Hills to the east. The altitude is over 1000 m with
an average temperature and rainfall of 21°C and 760 mm
respectively. The vegetation consists of the Brachystegia-
Julbernardia and Combretum-Acacia woodlands. Chloris gayana,
Cynodon spp., Panicum spp. and Neonotonia spp. occur in this
area. Most Cynodon spp. have head smut. Neonotonia spp. mature
early in May-June at the onset of the cool dry season.
The Mzimba Plain
Most of the area lies in the rainshadows of the Viphya highlands,
the Nyika plateau and the Njakwa Hills. The vegetation consists
of the Brachystegia-Julbernardia woodland. The altitude is about
1200 m and the average temperature and rainfall is 20°C and 760
mm respectively.
Chloris gayana, Panicum spp., Cynodon spp., Pennisetum spp.,
Neonotonia spp. and Clitoria ternatea occur in this area. No
collections of Chloris gayana have been made despite its wide
acceptance for grazing and hay making. However, Vwaza Game
National Park will remain the major area for future collections
whereas in settled areas termite mounds will be the best spots.
Large patches of Cynodon spp. in the Nkhamanga and Hewe areas
occur along rivers and low lying areas. Panicum maximum is
mostly found along the Rumphi and Runyina rivers. Clitoria
ternatea is found at Rumphi boma.
Nchenachena-Livingstonia Rills
The area lies on the eastern slopes of the Nyika at an average
altitude of 1400 m and is characterised by cool weather and an
extended rainy season due to the influence of the southeast Trade
Winds. The average annual temperature and rainfall is 20°C and
1700 mm respectively.
49
Neonotonia and Dolichos spp. occurring in this area mature
late in September-October at the onset of the short hot dry
season. The largest concentrations of Neonotonia spp. occur
between Junju and Kaziwiziwi where it is possible to find plants
2 0 m apart. Medium and giant Panicum maximum are concentrated
along rivers and valley bottoms. Cynodon spp. occurs at Rumphi
North bridge and around the Livingstonia plateau.
The Nyika Plateau
The plateau lies between 2,000-3,500 m and receives about 1270 mm
rainfall. The average temperature is about 14°C. Setarias and
Macrotyloma spp. are the only important forage species found on
the plateau. The Setaria because of the altitude have very small
leaves and may not be suitable for collections. Two types of
Macrotyloma spp. occur in the plateau - the grabrescent type
found in the open grasslands and the slender pubescent type is
found among the dense thickets along valleys. Kikuyu grass
(Pennisetum clandestinum) introduced from East Africa has almost
been naturalised on the plateau.
The Karonga Lakeshore Plain
The lakeshore plain average altitude is about 550 m. The average
temperature and rainfall are 25°C and 920 mm respectively.
Setaria, Panicum and Brachiaria spp. occur along channels in the
Kara, Wovwe and Lufilya rice schemes, the dambos through which
some major rivers from the Nyika plateau pass. Cynodon spp. on
this plain are severely attacked by head smut. Calopogonium sp.,
growing around the Lufilya rice scheme, is a recent introduction
by the Chinese agricultural technicians for green manuring.
The Chitipa-Lufilya Plain
The plain lies at an altitude of 1500 m. The average temperature
and rainfall are 20°C and 1000 mm respectively. Panicum, Cynodon
and Neonotonia spp. occur in many areas especially along the
drainage systems. Large concentrations of Panicum maximum are
found at Chisenga, at the foothills of the Mafingi Hills, and a
few plants are scattered along the Kaseye and Lufilya rivers.
Neonotonia wightii is found along the Kaseye and Songwe rivers,
and most of these mature in July. Cynodon spp. are widely
distributed on the Chitipa plain especially on sandy soils where
they form large patches of natural pastures. No head smut was
observed on these Cynodon spp.
The Misuku Hills
The hills rise to an altitude of 2000 m. Due to the influence of
the south-east Trade Winds, the area is covered by moist
Brachystegia woodland, semi-evergreen forests and montane
grassland forests. The average temperature and rainfall are 21°C
and 100 mm respectively. Neonotonia wightii , with a lot of
hairs, occurs in this area and matures late in September. A
Calopogonium sp. introduced by earlier agriculturalists grows
around Misuku Agriculture Station. There are few Panicum and
Brachiaria spp. in this area probably due to extensive
cultivation in this densely populated area. Cynodon spp. occurs
in scattered areas.
50
Summary and Conclusions
Panicum species tend to occur more in wet and fertile areas with
high frequencies than in dry areas especially on the leeward side
of high grounds. Neonotonia wightii also tends to follow the
same pattern except that occurrence is also high in wet fertile
areas. Chloris gayana , Urochloa mosambicensis, Brachiaria spp.
Dolichos spp. and Macrotyloma spp. have special habitats.
Therefore, with the present knowledge it is possible to make
trips to targeted areas to collect specific species.
REFERENCES
Addy, B.L. and Thomas, D. 1976. The utilization of Ntchisi panic
and Katherine pearl millet for summer stall feeding in
Lilongwe district. Ministry of Agriculture and Natural
Resources Bulletin No. 1/76, Lilongwe, Malawi.
Agnew, S. and Stobbs, M. 1972. Malawi in maps. Oxford University
Press, London.
Bogdan, A.V. 1977. Tropical pasture and fodder plants. Longmans,
London .
FAO. 1979. A Zambian handbook of pasture and fodder crops. Food
and Agriculture Organization of the United Nations, Rome.
Pike, J.G. and Rimmington, G.T. 1965. Malawi, a geographical
study. Oxford University Press, London.
51
Appendix 1
Figure 1. Physiographic Regions of Malawi.
 
52
Legend of Physiographic Regions as shown in
Figure 1.
la - The Upper Shire Valley
lb - The Middle Shire Valley
lo - The Lower Shire Valley
2a - The South-eastern Lakeshore
2b - The Southern Lakeshore
2c - The South-west Lakeshore
2d - The Nkhota-kota Lakeshore
2e - The Viphya Lakeshore
2f - The Karonga Lakeshore
3a - The Shire Highlands
3b - The Lake Chilwa
3c - The Namwera Hills
3d - The Chitipa Plain
3e - The Central Region Plateau
3f - The Mzimba Plateau
3g - The Henga-Kasitu Valley
3h - The Nchenachena
3i - The Chitipa-Lufilya Plain
4a - Mulanje Mountains
4b - Zomba Mountain
4c - The Dedza-Kirk Range Highlands
4d - The Dowa Highlands
4e - The Viphya Highlands
4f - The Nyika Plateau
4g - The Mafingi Mountains
4h - The Misuku Highlands
53
THE DISTRIBUTION OF SESBANIA 8FECIES IN
THE PANESA REGION
J.F. Moses Onim
Winrock International Institute for
Agricultural Development
SR-CRSP, P.O. Box 252, Maseno, Kenya
B.H. Dzowela
ILCA, P.O. Box 46847, Nairobi, Kenya
INTRODUCTION
The genus Sesbania belongs to the family Leguminosae and its
subfamily is Papilionoideae. There are four subgenera of which
Sesbania and Agati are of agricultural value. The subgenus
Sesbania is much more important in Africa, and it has a large
number of species. Some of the most important members of this
subgenus Agati is mainly found in southern Asia and its members
are more of perennial and tree types as compared to the
relatively more annual and shrub types found in the subgenus
Sesbania.
Members of the genus Sesbania are known for exceptionally
fast growth rates as well as a very high affinity for association
with several nitrogen-fixing Rhizobia in the soil that cause
formation of numerous and large nodules in the plant roots.
Members of this genus also have several potential uses including
forage, poles for light construction, fuelwood, pulpwood, live
fences, medicines, shade trees for other crops and gums. This
paper reviews these potential uses and makes a strong case for
germplasm collection of members of the genus Sesbania in the
PANESA region. It is envisaged that such a collection from
several African countries will yield many useful selections when
it will be subjected to screening for various economic and
agricultural traits.
SPECIES OF SESBANIA
Fifty species of Sesbania have been described in tropical and
subtropical regions of the world. Thirty-three of these species
are found in Africa (Table 1) . The remaining 17 species are
found in Australia (10) and Hawaii (7) species (Gillet, 1963;
Burbidge, 19 65) . The number of Asian species is not known (Char,
1983) .
In Africa the most dominant species is S. sesban (S.
egyptica) . However, there are several species of Sesbania in
Africa with unknown agricultural value and these have not been
evaluated in agronomic trials. Sesbania has been reported in
virtually all African countries from the tropical rainfall
forests in Zaire to the Nile valley in the Egyptian desert.
54
Table 1. Species distribution of Sesbania in various regions
of the world.
Region No. of species Source
Africa 33 Gillet, 1963
Australia 10 Burbidge, 1965
Hawaii 7 Char, 198 3
Asia Unknown Char, 1983
Objectives of the proposed Sesbania collection include:
1. To obtain Sesbania species from countries in the PANESA
region. The seeds will be divided into two portions, one
portion to be taken to a PANESA germplasm collection, and
the other portion to be given to each country where the
collection was made.
2. The collections to the PANESA germplasm bank will not be an
idle gene bank, but rather an active working collection.
The collection will be screened for several potential,
agricultural, forestry and industrial and medicinal uses.
3. The information will be catalogued by the co-ordinator of
PANESA.
4. Arrangements will be made to multiply seeds of the promising
collections and make these available on request to PANESA
member countries.
AGRICULTURAL USES OF SESBANIA
Livestock Feed
Several Sesbania species are an excellent livestock feed, both as
fodder and hay (Onim et al . , 1985). Dry-matter (DM) yields of
Sesbania forage is quite high when compared to other forage
legumes. Onim (1986) compared forage productivity of Leucaena
leucocephala (cv Cunningham) , pigeon pea (Cajanus cajan) and S.
sesban at Maseno Research Station in western Kenya. The forage
DM forage yields of the legumes after six months from planting
were 8000, 5500 and 3000 kg/ha for Sesbania, Leucaena and pigeon
pea respectively. Similarly nitrogen yields of these legumes
were 250, 175 and 120 kg/ha for Sesbania, Leucaena and pigeon pea
respectively (Figure 1) .
Nutritive values of forages of some Sesbania species are
presented in Table 2. The range of CP in Table 2 is from 18.8%
to 32.0% with a mean of 27.1%. The mean crude fibre content is
low (13.0%) and the mean calcium to phosphorus ratio is high
(3:8). It is therefore clear that Sesbania species have forages
of very high quality. Results of Katiyar and Ranjham (1969) ,
Chinnaswami et al. (1978) and Singh et al (1980) showed that
digestibility coefficients of three Sesbania species were quite
consistent when these were fed to sheep and goats (Table 3) . DM
and CP digestibility coefficients had means of 69.7 and 80.6
respectively. These results compare well with the DM
55
Figure 1. Cumulative DM, crude protein and nitrogen yields of
maize, pigeon pea, leucaena and sesbania after three
cuttings.
DM yields/cutting
frequency
8000
6000
4000
2000
0
2000
1500
1000
500
0
300
250
200
150
100
50
0
 
a) DM yields Kg /ho
tLSD = 3677 kg/ha
005
Cutting
frequencies
r777i. 3
2
rrrrm i
b) Crude protein kg /ha
 
c) Nitrogen yields kg/ha
 
Maize PP LEU
Crops
56
digestibility coefficient of 74.3 that was reported on S. sesban
from Maseno, Kenya by Sidahmed et al (1984) . The high nutritive
value of Sesbania forage was confirmed in a supplementation study
at the International Livestock Centre for Africa (ILCA) in
Ethiopia (Anon. , 1986) . In this study, teff (Eragrostis tef)
straw was supplemented with leaf hays of three browse tree
legumes - Acacia cyanophylla, A. seyal, S. sesban and Vicia
dasycarpa on a sheep feeding trial. Sheep body growth rates
Table 2. Nutritive values of forages of some Sesbania species
(% DM) .
Species DM CP CF Ca:P ratio
S. exaltata 26.3 18.8 12.4 4.9
S. bispinosa 23.8 30.0 11.4 2.4
S. sesban 25.8 26.0 16.2 6.8
S. macrocarpa 24.3 31.2 14.2 2.9
S. cinerascens 26.1 23.6 12.6 3.1
S. mosambicensis 25.2 32.9 10.9 1.9
S. grandiflora 18.1 26.9 12.3 4.5
Source: Rotar and Evans (1985) .
Table 3. Digestibility coefficients of Sesbania
as reported in several studies.
materials
A B C
S. bispinosa S. sesban S. grandiflora
15,7 45,7 10,7
adult rams kid goats adult rams
Dry matter 74.0 69.53 65.65
Crude protein 82.1 80.80 79.03
Ether extract 33.9 13.58 18.24
Crude fibre 63.7 56.13 57.30
N-free extract 75.9 67.82 68.60
Sources: Katiyara and Ranjhan (1969) ; Singh et al.
(1980); Chinnaswami et al. (1978).
showed that S. sesban hay was by far superior to all the other
legumes: in fact sheep on A. cyanophylla supplementation lost
condition (Figure 2a) . Nitrogen balance in sheep showed that
apart from cyanophylla, the other legumes were ranked for the
amount of nitrogen retained as Vicia dasycarpa, Sesbania sesban
and Acacia seyal (Figure 2b) .
These excellent nutritive qualities, especially the high CP
content, makes Sesbania a very useful supplement as a source of
CP to livestock. This is particularly crucial in small-scale
farming where feeding of crop residues or cut-and-carry systems
require a readily available source of CP. Sesbania hay lends
itself to this system very well because it is easily stored until
it is needed. Dairy meal (CP = 15%) can easily be substituted
with Sesbania hay as a source of CP (CP = 26%) to lactating dairy
animals.
57
Figure 2a. Growth rates and nitrogen balance of sheep fed
foliage from three different fodder trees or retch
hay as supplements to teff straw.
Body weight
(kg)
Acacia Acacia Sesbania Vicia
cyonophyllo se/ol sesban dasycarpa
20 -j (phyllodes) (Ijovss) (leaves) (aerial parts)
18 -
14
6-^
 
-r—
28
—I-
42
-1—
56
—1—
70
—I—
84
Source:
Day of experiment
ILCA Annual Report (1985/86)
Improvement of Soil Fertility
Sesbania species are known to fix between 500 to 600 kg/ha of
nitrogen per year. In a green manure experiment in Maseno, Kenya
(Onim, 1986) reported that S. sesban fixed up to 250 kg N/ha in
six months. Since urea fertilizer contains 46%N, this fixation
is equivalent to approximately eleven 50-kg bags of urea
fertilizer in six months. The use of Sesbania as a green manure
(GM) crop is a common practice in southeast Asia. Several
studies have shown that Sesbania can return into the soil as GM
between 80 and 120 kg of N within 90 days (Dargan et al . , 1975;
Bhardwaj et al., 1981).
Sesbania species are also often used in land reclamation
especially in salty (saline) and sodic soils as well as in mining
and excavation sites (Srivastava et al., 1973; Malik and Haider,
1977) . Both S. bispinosa and S. sesban have been used for these
purposes.
S. rostrata is native to tropical West Africa. This species
is unique because it fixes nitrogen not only in its roots in the
soil, but also in its aerial parts including stems and branches.
58
Figure 2b. Nitrogen balance in sheep.
N (grams per day) 111111 indigestible N VZWA Urinary N
Psa&fiSW Microbial N H;u.:..:.:::He_;.j Retained N
 
Acacia
soyol
Sssbanio
sesban
Vicia
dasycarpa
Research on this symbiosis in S. rostrata and its agronomic
implications has been conducted by scientists at the office of
Scientific Research Techniques, Outre-Mar, at the Scientific
National Research Centre in Senegal, Sudan, Zaire, Malawi,
Zimbabwe, Botswana, Caprivi Strip and Malagasy Republic (Gillet,
1971) . Whether this species can fix more nitrogen than the other
species is not well documented. More research is required in
this area.
Increased Milk Secretion
As far back as the 193 0s, lactating mothers in Java were
encouraged to eat young pods and flower buds of S. sesban and S.
bispinosa since it was believed that these stimulate milk
secretion (Ochse, 1931) . This galactagogue effect has also been
reported in cattle in Kenya where in the 1950s, S. sesban was
referred to as "milk shrub" (Brown, 1954) . Farmers were
encouraged to feed Sesbania fodder to lactating cows to enhance
milk secretion.
General Uses
Apart from the above-mentioned agricultural uses, there are
several other general ones. These include cases where Sesbania
are used as shade trees in coffee, cacao and other crop fields;
as live fences and for general agroforestry uses.
INDUSTRIAL AND DOMESTIC USES
Fuelwood
Perennial Sesbania species have been used for many years as a
source of fuelwood. Bulk density of Sesbania varies according to
species, rate of growth and age. However, values ranging between
240 to 616 kg/m3 have been reported (NAS, 1980 and 1983).
Fuelwood yields also vary depending on plant population used,
species and bulk density. For example, S. grandiflora in a
replicated trial in four sites in Hawaii at a spacing of 90 ocm x
90 cm (12,346 plants/ha), the trees had a mean stem diameter of 9
cm at breast height, and after three years the trees averaged 8 m
in height (MacDicken, 1983). In another trial, at one year of
age S. grandiflora had a height of 3.3 m, mean basal area of 24.5
cm and mean estimated wood volume of 24.6 m /ha.
S. sesban grows to a height of approximately 4.0 m a year in
western Kenya. Fuelwood potential of perennial Sesbania species
among rural peoples of Africa is very large. This genus has many
beneficial contributions to the environments where its forests
are established for fuelwood purposes. These include its ability
for rapid growth and thereby giving ground cover against soil
erosion, leaf fodder for livestock and improvement of soil
fertility through biological N-fixation and leaf litter. It is
therefore imperative that this genus be strongly encouraged in
soil reclamation, fuelwood and soil fertility improvement
programmes in African developing countries.
Fibre and Pulpwood
Sesbania species have also been used for fibre and pulpwood
extraction. The S. exaltata of the Americas has been used as a
source of fibre for fishing nets and lines by the Yuma Indians of
Arizona for many centuries (Parker, 1972) . In South East Asia,
S. bispinosa has been used for the same purpose (Sircar, 1948) .
Sesbania fibre has been found suitable for fish nets because it
resists decay. This quality qualifies Sesbania fibres for
several other marine applications like sail lashings. Sesbania
fibre is stronger than jute fibre (Townsend, 1973) .
Sesbania has also been used as an important source of
pulpwood. The best species for this role are S. sesban, S.
grandiflora and S. bispinosa. The same bulk densities referred
to before under fuelwood also apply here. When planted at a
spacing of 20 cm x 40 cm, up to 125,000 stems/ha can be
60
harvested. A paper mill exclusively operating on S. bispinosa
did well in west Pakistan (Hussain and Ahmed, 1965) . In India,
plantations of S. sesban have been established at Dandeli,
Karuataka and at Jajahmundry to be sources of pulpwood for the
West Coast Paper Mills and Andra Pradesh Paper Mill respectively.
Yields of between 50 to 55 tonnes of green wood/ha/year have been
reported in these plantations (Dutt et al . , 1983).
Most of the African countries rely on pine forests for their
pulpwood. Pines take up to 10 years to mature, and logs are
often hauled into paper mills from as far as 400 km away (e.g. in
Kenya) . This makes operational costs extremely high. The
potential of rapidly growing species like Sesbania and Leucaena
offer attractive alternatives. Both of these genera are also N-
fixing trees as well as an excellent high protein fodder and hay
for livestock feeding.
Human Food
S. grandiflora is known on both the continent and islands of
Southeast Asia for its large and edible flowers. Raw or lightly
steamed after removing calyx and pistil, they are used as an
ingredient of soups, salads and vegetable dishes. The white
flowers are preferred in the Philippines since red ones are said
to be bitter. Leaves are also cooked as a vegetable. In Sri
Lanka, one method of preparation is to cook chopped leaflets of
S. sesban with chopped onions in coconut milk, creating a
vegetable component of a traditional rice-based meal. S.
grandiflora leaves have been found to be bitter in trials in
India (Bai and Devadas, 1973) . S. grandiflora leaves had
therefore to be prepared in mixture with other leafy vegetables
(Gopaldas et al . , 1973).
Sesbania leaves and pods have been eaten by lactating
mothers to stimulate or increase milk secretion (galactagogue) .
Ochse (1931) reported that in Java, leaves and young pods of S.
grandiflora are eaten, especially by nursing mothers. According
to Brown (1954) , feeding S. sesban leaves to cattle is believed
to increase their milk production. Hurov (1961) calls S. sesban
the "Kenya milk shrub". The Haya people of western Tanzania use
this species for the same purpose.
Sources of Gums
Many Sesbania species contain gums which may be of potential
value for industrial uses. Natural gums, or mucilages, are
complex polysaccharides which have a wide range of uses. Their
varying physical properties are attributed to differences in the
degree of branching of and polymerisation of the sugars. Gums
are used in such wide-ranging products as ice cream, candy, soft
drinks, beer, pastries and heat-and-serve convenience foods.
Gums are also used in the manufacture of paper, textiles, paints,
in well drilling and in mineral assay.
Burkill (1935) reported that gums obtained from S. sesban,
S. formosa and S. grandiflora are very similar to gum arabic. A
United States National Academy of Sciences Report (NAS, 1979)
stated that Sesbania bark gums have been used as a substitute for
gum arabic, and suggested that with increasing scarcity of gum
61
arabic from Acacia senegal , exploitation of Sesbania bark gums
should be encouraged.
MEDICINAL USES
Sesbania species have a long list of medicinal uses among
different peoples of the world, especially in Africa and Asia.
The three most important species as sources of traditional
medicines are S. sesban, S. grandiflora and S. bispinosa.
Astringent
S. grandiflora juices and extracts have astringent quality. This
is the ability to contract body tissues and blood vessels. This
property is used for reducing fevers and for promoting fluid
discharge and subsequent drying of mucous membranes, hence
leading to healing of, for example, wounds. For systemic
disorders e.g. smallpox, decoctions are taken internally. Local
applications are claimed to bring relief to nasal congestion, and
rhinitis and associated headache (Watt and Breyer-Brandwijk,
1962) . The bark extracts are used in Java for thrush and stomach
trouble in infants (Burkill, 1935). Astringent quality in S.
sesban is even stronger and unique in many ways. Fresh root and
poultices of leaves have been used for scorpion stings, boils,
abscesses, rheumatic swelling and hydrocele (a collection of
watery fluid in a cavity of the body, especially in the scrotum
or along the spermatic cord) . Diarrhoea and excessive menstrual
flow are said to be relieved by a concoction of seeds.
Antihelminthic
Doses of up to 2 oz. of concoction of S. sesban leaf are used as
an antihelminthic against tapeworms and roundworms in humans. A
mixture of ground seed and flour made into a paste is used for
treating ringworms (Watt, 1983) .
Antibiotic
The Haya people of western Tanzania make a concoction of leaves,
barks and roots called "Mubimba" from S. sesban that they use to
treat a wide range of diseases including sore throat, gonorrhoea,
syphilis, yaws, fits and jaundice (Watt and Breyer-Brandwijk,
1962) .
Contraceptive and Arborifacient
It is believed in India that eating flowers for three days during
menstruation inhibits conception. To prove this belief, Pakrashi
et al. (1975) fed 50 gm of flower extract of S. sesban per kg of
body weight to breeding and pregnant mice. Those that were
breeding did not conceive while those that were pregnant aborted
ranging from 54-77%, depending on the type of extract.
Antitumoric
Antitumor activity has been reported for the North American
species (S. drummondii) and brief reviews of this work have
62
recently been published in popular scientific press (Gannon,
1983). This characteristic was revealed when Powell et al .
(1976) reported that ethanolic seed extracts of S. vesicaria, S.
punicea and S. drummondii were cytotoxic in the KB cell culture
and were active against lymphocytic leukemia P-388 in mice. The
active ingredient in this extract was named "sesbanine" by Powell
et al (1979) . A further fractionation revealed a second active
ingredient called "drummondol" (Powell and Smith, 1981) . This
antitumor medicinal value is of great interest in medicine as a
possible source of drugs for cancers. The major limitation,
however, is the very low yields of these compounds from Sesbania
seeds. Approximately 450 kg of seeds yield a mere 50 mg of
sesbanine (Powell et al., 1979). However, work is in progress to
produce synthetic sesbanine.
CONCLUSION
Several potential uses of Sesbania species have been reviewed in
this paper. Most of these potential uses are urgently needed by
the developing nations of Africa and Asia where populations are
increasing at a rate that outstrips supplies from available
natural resources. Since Africa has the largest diversity of
Sesbania species, it is envisaged that the proposed germplasm
collection when done, screened and catalogued will prove to be a
great resource for the poor and developing nations in the PANESA
region and beyond.
REFERENCES
Bai, G. and Devadas, R.P. 1973. Ability of two rice diets to
support reproductive and lactation performance in rats.
Indian Journal of Nutrition and Diabetics 10 (6) : 275-281.
Bhardwaj , S.P., Prasad, S.N., Singh, G. 1981. Economizing
nitrogen by green manures in rice-wheat rotation. Indian
Journal of Agricultural Sciences 51(2):86-90.
Brown, W.H. 1954. Useful plants of the Philippines. Republic of
Philippines Department of Agriculture and Natural Resources
Technical Bulletin No. 10, Vol.2.
Burbidge, N.T. 1965. Australian species of Sesbania Scop.
(Leguminosae) . Australian Journal of Botany 13:103-141.
Burkill, I.H. 1935. Dictionary of the economic products of the
Malay Peninsula. Governments of Straits Settlements and
Federated Malay States. London.
Char, w.p. 1983. A revision of the Hawaiian species of Sesbania
(Leguminosae) . M.Sc. thesis, Univ. of Hawaii, Hawaii, USA.
Chinnaswami, C.N., Venkatakrishnana, R. , Rama, R. 1978.
Investigation on the nutritive value of Agati (Sesbania
grandiflora) as green fodder for sheep. Cheiron (Tamil Nadu
Journal of Veterinary Science and Animal Husbandry)
7(1/2) :28-33.
63
Dargan, K.S., Chillar, R.K., Bhardwaj , K.K.R. 1975. Green
manuring for more paddy in alkali soils. Indian Farming
25:13-14.
Dutt, A.K., Pathania, U. , Kumar, V. 1983. Growth of Sesbania
sesban. Nitrogen Fixing Tree Research Reports 1, pp. 5-6.
Gannon, L. 1983. Back-to-nature chemistry. Science News
123(15) :235.
Gillet, J.B. 1963. Sesbania in Africa. Kew Bulletin 17:91-159.
Gopaldes, T., Ramakrishnan, I., Grewal, T. , Rajalakshmi, R. ,
Devadas, R.P. 1973. Use of legumes and green leafy
vegetables for infant and young child feeding: summary of
results of studies in three different parts of India. PAG
Bulletin 3(2) :51-53.
Hussain, A. and Ahmed, M. 1965. Sesbania aculeata, a promising
new crop in West Pakistan. World Crops 17(3):28-31.
ILCA (International Livestock Centre for Africa). 1986. ILCA
Annual Report 1985/86. ILCA, Addis Ababa. p. 50.
Katiyar, R.C. and Ranjhan, S.K. 1969. Yield and chemical
composition of dhaincha (Sesbania aculeata) - its nutritive
value for sheep. Indian Journal of Dairy Science 22:33-36.
MacDicken, K.G. 1983. Studies on the early growth rates of
selected nitrogen-fixing trees. M.Sc. thesis, Dept. of
Agronomy and Soil Science, University of Hawaii, USA.
Malik, K.A. and Haider, K. 1977. Decomposition of carbon-14-
labelled plant material in saline-sodic soils. In: Soil
Organic Matter Studies. International Atomic Energy Agency,
Vienna. pp. 215-225.
NAS (national Academy of Science). 1979. Tropical legumes,
resources for the future. National Academy of Sciences,
Washington, D.C.
NAS. 1980. Firewood crops, shrub and tree species for energy
production. National Academy of Sciences, Washington, D.C.
NAS. 1983. Firewood crops, shrub and tree species for energy
production, Volume 2. NAS, Washington, D.C.
Ochse, J.J. 1931. Vegetables of the Dutch East Indies.
Department of Agriculture, Industry and Commerce,
Netherlands East Indies Buitenzorg, Java. pp. 429-430.
Onim, J.F.M. 1986. Multiple use of pigeonpea. In: Proceedings
of ICRISAT Consultative Group Meeting for Eastern and
Central African Regional Research on Grain Legumes held at
ILCA, Addis Ababa, Ethiopia, 8-10 December 1986. pp. 115-
120.
Onim, J.F.M. , Semenye, P.P. and Fitzhugh, H.A. 1985. Small
ruminants in African agriculture. Proceedings of a
64
conference held at ILCA, Addis Ababa, Ethiopia, 30 September
- 4 October 1985. ILCA, Addis Ababa.
Pakrashi, A., Basak, B. , Mookerji, N. 1975. Search for
antifertility agents from indigenous medicinal plants.
Indian Journal of Medical Research 63 (3) : 378-381.
Parker, K.F. 1972. An illustrated guide to Arizona weeds.
University of Arizona Press, Arizona, USA.
Powell, R.G., Smith, C.R. Jr., Madvigal, R.V. 1976. Antitumor
activity of Sesbania versicaria, S. punicea and S.
drummondii seed extracts. Planta Medica 30(1) :l-8.
Powell, R.G., Smith, C.R. Jr., Weisleder, D. , Muthard, D.A. ,
Clardy, J. 1979. Sesbanine, a novel cytotoxic alkaloid from
Sesbania drummondii. Journal of the American Chemical
Society 101 (10) : 2784-2785.
Powell, R.G. and Smith, C.R. Jr. 1981. An investigation of the
antitumor activity of Sesbania drummondii. Journal of
Natural Products 44(l):86-90.
Rotar, P.P. and Evans, D.O. 1985. Sesbania as an LDC agro-
forestry resource. (Final Report) . Program of Science and
Technology Project, Department of Agronomy and Soil Science,
University of Hawaii, Hawaii, USA. p. 252.
Sidahmed A, Reed J, Butterworth M, Otieno K and Fitzhugh H A.
1984. Chemical composition and digestibility values for
some important forage resources consumed by dual-purpose
goats in western Kenya. Proc. of 3rd SR-CRSP Workshop,
Nairobi, Kenya.
Singh, C. , Kumar, P., Rekib, A. 1980. Note on some aspects of
the feeding value of Sesbania aegyptiaca fodder in goats.
Indian Journal of Animal Science 50 (11) : 1017-1020.
Sircar, J.K. 1948. Memorandum on fibres other than cotton and
jute. Indian Council of Agricultural Research (ICAR)
Miscellaneous Bulletin No. 66, pp.59.
Srivastava, G.P., Bhoi, R.L., Khan, M.R. 1973. Reclamation of
saline soil by a mixture of dhaincha and superphosphate.
Allahabad Farmer 47(1): 43-47.
Townsend, C.C. 1973. Papilionaceae. ln:C.C. Townsend and E.
Guest (eds) , Flora of Iraq. Vol.3: Leguminales . Ministry of
Agriculture of the Republic of Iraq, Baghdad. pp. 219-224.
Watt, G. 1893. A dictionary of the economic products of India.
Superintendent of Government Printing, Calcutta, 6 vols.
1889-1893.
Watt, J.M. and Breyer-Brandwikj , M.G. 1962. The medicinal and
poisonous plants of southern and Eastern Africa. E & S
Livingstone Ltd., London.
65

8ESSION II: FORAGE GERMPLASM EVALUATION
67

FORAGE GERMPLASM AT THE
INTERNATIONAL LIVESTOCK CENTRE FOR AFRICA
Jean Hanson and John R. Lazier
Forage Agronomy Group
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
Abstract
The International Livestock Centre for Africa (ILCA) holds a
large collection of forage germplasm of legumes, grasses and
browse in controlled storage conditions at its headquarters in
Addis Ababa. The germplasm collection is a major responsibility
of the forage agronomy programme at ILCA. The forage germplasm
available and its storage is described. The types of data
recorded about each accession and how to use these data to select
suitable germplasm for screening activities is explained.
INTRODUCTION
The success of a forage research or development and improvement
programme is usually dependent on the availability of adapted,
productive and appropriate forage germplasm for selection of
promising lines. In most African countries such germplasm has
been difficult to obtain in the past because the sources were in
institutions with no particular interest in or responsibility for
forage development in Africa.
It was in recognition of this serious impediment to forage
development in Africa that the Forage Agronomy Group (FLAG) at
ILCA Headquarters in Addis Ababa established a forage genetic
resources collection as a service to collaborators and other
research workers in national programmes in Africa and for forage
development work in ILCA programmes. This collection was begun
in 1982 and has grown over the past five years until it now
contains 9166 accessions of forage species, including legumes,
grasses and browse.
The maintenance of such a large and varied collection is now
a major responsibility of FLAG, which has accepted the
international responsibility to store this germplasm and make it
freely available for research and development. All original
material is being duplicated in other forage genebanks for
security under the IBPGR global network of base collections
(Hanson et al., 1984).
THE GERMPLASM COLLECTION
The germplasm collection is composed of about 75% legumes and 10%
each of grasses and browse with 111 genera and 341 species
represented in the collection. The major part of the collection
is made up of experimental lines of germplasm which were either
acquired from other institutions or original collections of
forages from various countries in Africa, including Ethiopia,
Kenya, Rwanda, Niger, Burundi, Eastern Zaire, Mali and Tanzania.
69
Due to FLAG being located in Ethiopia and the resulting ease and
economy of collection there, the largest number of ILCA collected
accessions are Ethiopian, and they constitute 30% of the total
collection. FLAG collecting missions in the other countries have
contributed a further 10% of the accessions. Commercial lines of
legumes and grasses make up about 5% of the collection and are
held as a service to users who wish to test commercial lines in
small quantities for preliminary evaluation.
Due to the great range of environments in ILCA's mandate
region and due to the centre of diversity of many important
legumes lying outside Africa, primarily in South and Central
America and Asia, FLAG has made considerable efforts to acquire
representative sets of germplasm of potential value for
development in Africa from institutes with collections from these
regions. Almost 20% of the collections were obtained from the
Centro International de Agricultura Tropical (CIAT) based at Cali
in Colombia which has a large collection of legume germplasm from
Central and South America, the main centre of diversity of forage
legumes. FLAG has strong co-operative links with the staff of
the tropical pastures programme at CIAT, and in the past CIAT and
ILCA have held joint collection missions for forages in Africa.
Another major donor of germplasm has been the Division of
Tropical Crops and Pastures (CSIRO) based in Queensland,
Australia from whom ILCA has received almost 10% of its
germplasm. The remaining accessions have been acquired in
smaller numbers from other research organizations, from
individual scientists or as duplicates of materials held by
ILCA's field programmes in Nigeria and Mali.
IMPROVING THE COLLECTION
Despite the number of accessions already in the genebank, there
is much variation in both exotic and native African species of
recognised forage value which is not represented in the
collection. In addition, there are more than 50 of the 111
genera held by the genebank which are as yet only represented by
very few accessions. These genera cannot be usefully tested with
such a narrow range of material.
African germplasm of forage potential has been little
explored, and legumes, grasses and browse can be profitably
collected. Concentration of collection activities in individual
species and genera has captured tremendous genetic variation, for
example in the East African highlands, Trifolium species
(collected by ILCA) , Brachiaria decumbens (collected by ILCA-
CIAT) and Panicum maximum (collected by ORSTOM) .
Genera worthy of attention in Africa due to their diversity,
include among others the legumes Macrotyloma, Lablab,
Aeschynomene , Crotalaria, the browse Sesbania, Erythrina and
Acacia and many species of grasses, including Digitaria,
Pennisetum, Cynodon and Chloris.
Forage species are also needed for development for specific
African environments. For example there are few truly promising
legumes for semi-arid and acid soil conditions. Browse germplasm
is very limited for acid soils and high altitudes. Such
70
germplasm may be obtainable from other genebanks, or collections
may have to be undertaken within or outside of Africa.
There is an urgent need to collect and preserve genetic
variation in species with genetic potential from areas subject to
rapidly increasing population pressures before they are lost due
to the resulting drastic environmental changes. Genetic erosion
in forages and other wild species is occurring at a rapid rate in
many areas, and timely action to collect is essential if valuable
germplasm is not to be lost.
Due to the small amount of forage germplasm collection done
to date in Africa, almost any well planned collection is bound to
produce new and substantial variation. Collecting missions
require considerable forward planning to be successful. The
target species, area of collection and time of year must be
defined in order to optimise the number of accessions obtained.
In 1987 FLAG has planned two collecting missions. One is genus-
specific to collect mainly Sesbania species in Tanzania during
July to August, although other genera of interest will also be
collected if time allows. The other is a more general Ethiopian
browse collection mission, which will probably be split into
several short missions beginning in May to cover the seeding
period of many species.
STORAGE OF MATERIAL
The material collected must be conserved in an appropriate manner
so that it remains available for distribution and is not lost or
merely stored without being utilised. The easiest storage method
is in a seed genebank. ILCA has recently installed new drying
and cold rooms to process and store seeds under conditions in
line with the IBPGR recommendations for the storage of seeds in
genebanks (IBPGR, 1985). Seeds are first dried to between 3-7%
moisture content in a dehumidified drying room, which operates at
15°C and 20% relative humidity. Under these conditions small
seeds reach the required moisture content within one week while
larger seeds take up to two weeks. When the seeds are dry they
are packed in moisture-proof laminated aluminium foil bags which
are carefully labelled and placed into store. ILCA has two types
of germplasm storage. The active collection, composed of larger
seed samples for distribution, is stored in a cold room at 5 C.
The base collection is for security storage only and is stored in
deep freezers at -20°C.
Another method of conservation is to maintain plants in
field genebanks. ILCA has a field genebank in the Rift Valley
where a large collection of Brachiaria species and some other
grasses are planted out. Many grass species rarely produce
viable seeds and at ILCA maintenance in plots is the only
feasible way to conserve these species. In vitro genebanks,
where meristem cultures are maintained in test tubes under
controlled environmental conditions, are an alternative method of
storage of species which do not readily produce seeds. FLAG is
already culturing meristems of Brachiaria species for
distribution, and work will begin later this year on methods of
culturing other grass genera and the development of in vitro
storage procedures.
71
GERMPLASM DATA
In order that germplasm can be efficiently utilised, basic
information should be available about each accession, thus
allowing selection of potentially adapted materials for specific
ecological zones. At collection, as much useful field data as
possible is recorded about each accession and the site where it
was collected. In genetic resources terms this data is known as
passport data. A standard collecting form is used to record the
data (Lazier, 1985) which is then entered into the computerised
database. More data is obtained from references and analyses of
soil samples. Further identification of herbarium specimens may
be necessary. The database which ILCA uses has been made as
comprehensive as is practical in order that a maximal amount of
data is available to researchers.
Another set of data known as preliminary characterisation
data is recorded when the material is first grown in the field.
This is, in general, observation data on morphology and
adaptation obtained from a small number of plants in unreplicated
plots. Further evaluation in replicated strip trials provides
agronomic data which allows selection of promising material for
multi-site evaluation trials in national programmes.
To date only the passport data has been entered into the
germplasm database, but it is envisaged that a supplementary
database of agronomic characters will also be established in the
development of the genebank documentation system. The computer
is a valuable tool in germplasm work because it allows rapid
selection of subsets of accessions by query on specific fields,
preparation of germplasm catalogues and lists of passport data
and manipulation of data.
HOW TO SELECT GERMPLASM FROM THE GENEBANK
With such a wide array of germplasm available, the user may find
it very difficult to select accessions for evaluation. The
forage agronomists in FLAG can provide advice on promising
materials to test, but selection is also possible directly by the
user. FLAG publishes a germplasm catalogue which is updated at
regular intervals (ILCA, 1985) . This lists all accessions
available in the genebank in three ecologically organised
volumes: tropical lowland, temperate/Mediterranean and tropical
highland. The accessions in each volume are divided into annual
and perennial legumes and grasses and presented in alphabetic and
accession number order. A section on browse species is included
in the tropical lowland germplasm catalogue, since the majority
of the multi-purpose tree species are of tropical origin. As our
collection expands with the addition of more temperate materials,
sections on browse may be added to other catalogues.
Since the catalogue cannot contain all useful data on an
accession because of the amount of data available and large
number of accessions, only the fields most useful for selection
of appropriate germplasm for evaluation were included in the
catalogue. This includes collection site data such as geographic
location, rainfall, soil type and pH and other identification
numbers of each accession.
72
The catalogue can be used to select annual or perennial
accessions of species of interest which were collected from areas
of similar environment to the trial site. Users can then fill
the ILCA numbers of selected lines onto an ILCA seed request form
and submit this to FLAG.
Germplasm is freely available from the FLAG genebank in
small experimental quantities of a few seeds to a few grams. The
recipients must multiply seeds of any promising material for use
in subsequent experiments themselves. In return for the seeds,
recipients are requested to provide the genebank with information
on the performance of the lines tested, in order that the
genebank can provide adapted germplasm for similar environments
with greater assurance, thus providing a better service and
considerable savings in time and labour.
WHAT GERMPLASM TO SELECT
It is obvious that with the wide range of environments
encountered in sub-Saharan Africa, that sets of adapted germplasm
must be developed for different ecological situations. FLAG
currently evaluates its germplasm in four major environments.
Three are in Ethiopia: tropical subhumid, mid-altitude on acid
soils, tropical semihumid highland. More recently evaluation was
commenced on tropical sub-humid lowland acid soils in Nigeria.
Over the last four years a series of trials have been carried out
in the Ethiopian sites and promising experimental materials have
been selected for each of these environments. These are then
planted in larger plots to see if they maintain their comparative
advantage compared to commercial or previously available lines
and to obtain sufficient seeds for larger scale trials. FLAG is
building up its stocks of seeds of promising accessions so that
they can go into wider testing in national programmes in 1988.
Collaboration with national research centres through PANESA is
the next important step in the development of promising forage
germplasm for utilisation in animal production systems in the
region.
REFERENCES
Hanson, J., Williams, J.T. and Freund, R. 1984. Institutes
conserving crop germplasm: the IBPGR global network of
genebanks. IBPGR, Rome.
IBPGR. 1985. IBPGR Advisory Committee on seed storage. Report of
the third meeting. IBPGR, Rome.
ILCA. 1985. ILCA forage germplasm catalogue 1985. Vols 1-3.
ILCA, Addis Ababa, Ethiopia.
Lazier, J.R. 1985. Theory and practice in forage germplasm
collection. In: J. A. Kategile (ed.), Pasture improvement and
research in Eastern and southern Africa. IDRC-237c.
73
A REVIEW OF THE POTENTIAL OF BRACHIARIA SPECIES AS
FORAGE CROP FOR LIVESTOCK IN TANZANIA
N.A. Urio, S.V. Sarwatt and E.J. Mtengeti
Department of Animal Science and Production
Faculty of Agriculture
Sokoine University of Agriculture
P.O. Box 3004, Morogoro, Tanzania
Abstract
Brachiaria spp. is a perennial grass native to East and Central
Africa. It has been introduced into humid tropical regions of
Latin America, southeast Asia, and northern Australia where it
has revolutionised grassland farming and animal production. The
potential of Brachiaria spp. in its native land on the other
hand, remains largely unexploited.
About 99 percent of Tanzania's ruminant livestock population
(22.5 million head of cattle, sheep and goats) derive all their
feed from pastures. Hence, there is need to explore productive
native pasture species and to provide high management practices
so as to increase their yield and quality. The impressive
results recorded on Brachiaria in foreign lands, together with a
few experimental results recorded in East Africa, indicate that
this species could have a significant impact on animal production
in East Africa.
This paper therefore reviews research work carried out in
Tanzania on Brachiaria spp. in order to stimulate more research
on this species.
INTRODUCTION
A number of Brachiaria spp. have been recorded in East Africa.
These are Brachiaria brizantha, B. ruziziensis, B. serrifolia, B.
mutica, B. dictyoneura, B. nigropedata, B. soluta, B. humidicola,
B. radicans, B. serrata, B. jubata, B. leucocrantha, B. platynota
and B. bavonei, (Bogdan, 1955; Van Rensburg, 1960; Mengistu,
1985) . The most common of these species are B. brizantha (Signal
grass), B. mutica (Para grass), B. ruziziensis (Congo grass).
Brachiaria brizantha occurs throughout tropical Africa from
sea level up to 2400 m under an annual rainfall of over 800 mm
(Bogdan, 1977) . In East Africa, Bogdan (1955) reported that B.
brizantha is very variable and several varieties show striking
differences in habit morphology and seed setting capacity. It is
possible that the different varieties will perform differently in
different ecological zones, but information of this kind is
lacking. Low seed viability is probably one of the chief
limitations to the wide use of Brachiaria spp. as a forage crop.
Identification and development of varieties with good seed
setting capacity is one area that ought to receive attention from
researchers. B. brizantha has also been noted to be resistant to
drought and aggressive, competing effectively with other species
and quickly covering the ground (Stomayor-Rios et al., 1960). At
Morogoro, Tanzania, a field of Brachiaria brizantha has been
74
maintained for over 20 years without replanting. In Sri Lanka,
Brachiaria brizantha has also been reported to perform well under
shade of coconut trees (Anken-Lagefoged, 1955) . This observation
is relevant to Tanzania because a number of dairy and beef farms
have been developed in coastal regions where inclusion of this
species under coconut trees would greatly improve feed quality
and quantity.
Brachiaria brizantha has also been noted to grow and give
higher yield when grown on acid rather than alkaline soils
(Stomayor-Rios et al . , 1960). This characteristic makes it
suitable to grow in fallow land after a continuous use of soil
acidifying fertilizers such as ammonium sulphate. According to
Bogdan (1977) , this grass also recovers well from close cutting.
This is also supported by observations on fields grown at
Morogoro. These attributes make Brachiaria brizantha potentially
suitable for the cut-and-carry feeding systems in highland areas
of Tanzania. Due to its poor seed-setting characteristic
Brachiaria brizantha is normally propagated by use of root
stocks. This method of propagation is rather laborious for
large-scale establishment but is not a major limitation for
smallholder plots.
Brachiaria ruziziensis (Congo grass) is a leafy semi-
prostrate, rhizomatous species adapted to humid tropics
(Whiteman, 1980) . It is less vigorous than Para and Signal
grasses. It appears to be more shade-tolerant than Signal grass
and thus is more suitable for inclusion under coconut trees. The
grass can be propagated both from root stock as well as from
seeds.
Brachiaria mutica (Para grass) is a perennial stoloniferous
grass widely grown in the humid tropics and subtropics with
rainfall of 1250 mm p. a. This grass has become the backbone of
the beef industry in South and Central America, Australia,
Philippines and Cuba (Bogdan, 1977) . This species is, however,
attacked by spittle bug in these areas, and attempts have been
made by CIAT to collect germplasm resistant to spittle bug from
East Africa. B. mutica is also said to perform well in
waterlogged and flooded conditions as compared to B. brizantha
and B. ruziziensis . This makes the species particularly suitable
for swampy areas. B. mutica forms an open sward and therefore
can combine well with legumes particularly Centrosema pubescens.
PRODUCTIVITY OF THE SPECIES
Herbage Dry-Matter Yield
There is considerable literature showing that Brachiaria spp. can
give high yields of forage under good climate and management
(Table 1) . Yields range between 5 and 36 t DM/ha/year depending
on soil fertility, moisture and fertilizer application (Bogdan,
1977) . There is very little information on the productivity of
Brachiaria spp. in Tanzania despite the fact that the grass is
indigenous to grasslands there. The little information available
is based mainly on Brachiaria brizantha.
75
ui0
■a
-h
<u
-H
§
0)
£1
+J
C
o
.H
10
111
c
-H
■p
■p
-
T3
C
io
0)
■p
10
1-4
c
ai
g1.
m a
4-1 a.
-h u
S5
10
iH -H
O H
10
i» -H
O X)
c o
lU 10
3 H
H CQ
iM
C iH
H O
J2
Eh
3 if
1 ~ 5
" S ut 5 ■
-Id J3 r- u
M -J 4 U » -iv. ifl J ■ H «
3 w g * i-
S« -1 G ■■- U
w I •* M
-- <t -h u •
a u u - .- 3
< < ui > ■ £
3 j s■s J s
M i■ 1
U. H H si
n m CD
PI 1N iN
76
Without fertilizer application, Frederiksen and Kategile
(1980) , reported a yield of about 3 t DM/ha on a 10-year-old
field (Table 2) . On fertilizer application yields were raised to
about 10 t DM/ha. Kidunda (unpublished data) reported yields of
9.5 t DM/ha on unfertilized newly established plots. Brachiaria
spp. generally respond very well to fertilizer application
(Tables 2 and 3) , and the age of the sward seems to influence
both the response and yield. Observations on the 2 0-year-old
swards at Morogoro indicate that very few grass species can
compete with Brachiaria spp. on persistence; this makes the
species especially valuable in grazing lands. Where irrigation
facilities are available, yields per annum could also be raised
severalfold.
Table 2. The effect of nitrogen application on the yield,
nitrogen utilization, CP% and nitrogen recovery rate
on B. brizantha at Morogoro in Tanzania.
Response Nitrogen
N level Yield tons kg recovery
kg/ha DM/hia DM/kg N IVOMD CP % Of DM rate
1973 •74 •73 •74 '73 <74 '73 •74 '73 '74
0 2.8 2.9 48 55 5.9 5.5
62.5 5.8 6.2 48 53 54 54 4.6 5.4 38 50
125 8.9 7.6 49 37 54 59 5.4 7.4 42 55
187.5 9.8 8.2 37 28 53 57 7.6 6.1 48 55
*IVOMD
Source:
In vitro organic-matter digestibility.
Fredricksen and Kategile (1980).
Table 3. Effect of nitrogen fertilizer application on the annual
dry-matter production (t/ha) ; CP% of DM and yield of CP
(t/ha) of Brachiaria ruziziensis and Chloris gayana.
Rate of N application (kg/ha/year)
0 224 440 896 1568 2240
Brachiaria
ruziziensis
DM yield (t/ha)
CP % of DM
CP % yield (t/ha)
6.1
6.7
0.6
13.9
7.7
1.2
21.8 26.5 25.9 23.5
10.1 13.9 16.3
4.1
16.8
Chloris gayana
DM yield (t/ha)
CP % of DM
CP yield (t/ha)
2.4 3.6 3.9
11.2
8.5
1.1
20.7
9.3
1.9
24.5 27.8 26.0 25.4
11.3 14.3 15.7 15.5
2.8 3.9 3.9 3.9
Source: Olsen (1982) in Uganda.
Nutritive Value
Although no systematic studies have been carried out to evaluate
the nutritive value of Brachiaria spp. a number of studies has
77
shown that the species is of high nutritive value. Frederiksen
and Kategile (1980) reported about 800 kg CP/ha in the fifth week
of regrowth (Table 2) and Nnko, (unpublished data) working on the
same plots, which were now 20 years old, reported 400 kg CP/ha at
a fertilizer application rate of 105 kg N/ha at eighth week of
regrowth. Elsewhere higher yields of CP have been reported with
higher N fertilization. In Puerto Rico, Vicente-Chandler et al . ,
(1972), reported yields of 3.5 t of CP/ha at a fertilizer rate of
896 kg N/ha and 60 days cutting intervals for B. ruziziensis. The
rate of N fertilizer will certainly be dictated by the economics
in a particular area, but if comparative studies are to be
conducted with other grass species such as Chloris gayana,
Brachiaria spp. is likely to outperform them in this respect.
Brachiaria spp. has been reported to have a fairly high
mineral content. Mtengeti (unpublished data) , studying mineral
status of some pasture species at Morogoro in Tanzania, reported
high magnesium content of B. brizantha as compared to other
pasture species studied. The species was reported to have fairly
reasonable contents of other minerals as well (Table 4) . Similar
trends of mineral contents were reported by Vicente-Chandler
(1959) with B. ruziziensis in Puerto Rico.
Table 4. Mineral composition of B. brizantha at preflowering and
flowering stage at Morogoro, Tanzania.
Stage
of
growth
Ash Ca P
%
Mg K Ca : P Cu
ratio
Zn
ppm
Mn
Pre-
flowering 11.20 0.33 0.27
10.57 0.33 0.21
0.24
0.25
2.76
2.88
1.22 6.9
1.55 3.90
20.30
20.44
63.90
81.57Flowering
Source: Mtengeti (unpublished data)
Although no systematic studies have been carried out at
Morogoro to study intake and animal production from Brachiaria
spp., observations have indicated preference and higher intakes
by cattle for Brachiaria compared to Chloris gayana. In studies
carried out in western Tanzania, Kapinga (1986) ueported that
Brachiaria spp. was the second most palatable species (after
Leucaena spp.) out of eleven species studied.
In countries where Brachiaria spp. has been managed well as
a forage crop, standard agronomic practices have resulted in very
impressive animal performance. In Tanzania, there is a need to
carry out identification, selection and improvement particularly
on those varieties that set viable seeds and if possible identify
varieties for the different ecological conditions. Combinations
of Brachiaria spp. with different legume species have given
impressive results elsewhere (Table 5) and are likely to give the
same results in Tanzania. This is worth studying and exploiting
for optimal production. The fact that Brachiaria spp. has
performed extremely well in countries where it has been
introduced should act as a catalyst to improve it in its native
land and utilise this useful animal feed resource otherwise lying
idle.
78
Table 5. Average stocking rate and cattle liveweight gains on
improved pasture.
stocking rate
Average Liveweight gains
Pasture (beast/ha) Per day Per year
Rhodes + Stylo
Para + Centro
Guinea + Centro
Guinea + Stylo
3.7
4.0
4.0
3.6
1.5 543
1.8 662
1.7 614
1.6 577
Source: Allen and Cowdry (1961) .
REFERENCES
Allen, G.H. and Cowdry, W.A.R. 1961. Beef gain from irrigated
pastures in the Burdekin delta. Queensland Agricultural
Journal 87:175-179.
Anken-Lagefoged, A.V. 1955. The role of grasslands in Ceylon's
agriculture. Tropical Agriculture 3:257-266.
Appandurai, R.R. and Arassaratnam, R. 1969. The effect of large
applications of urea nitrogen on the growth and yield of an
established pasture of Brachiaria brizantha (Stapf ) .
Tropical Agriculture 46:153-158.
Bogdan, A.V. 1955. Herbage plants at the grassland research
station, Kitale, Kenya. East African Agriculture and
Forestry Journal 20: 151-165.
Bogdan, A.V. 1977. Tropical pastures and fodder plants.
Longman , London .
Car-Costas, R. , Vicente-Chandler, J. and Abrunia F. 1976.
Comparison of heavily fertilized Congo, Star and Pangola
grass pastures in the Humid Mountain region of Puerto-Rico.
Journal of Agriculture of the University of Puerto-Rico
40:179-185.
Fredericksen, J.H. and Kategile, J. A. 1980. The effect of
nitrogen fertilization and time of cutting in first growth
in Brachiaria brizantha on yield, crude protein content and
in vitro digestibility. Tropical Animal Production 5:136-
143.
Funes, F. , Perez, L. , and Randa, A. 1980. Growth and development
of grasses in Cuba. 2. Effect of three cutting intervals
on the yield of eight grasses. CuJban Journal of
Agricultural Science 14:181-188.
Kapinga, P.X. 1986. Range development project for western
Tanzania. PANESA Newsletter No. 3. ILCA, Addis Ababa.
Mengistu, S. 1985. Survey and collection of forage germplasm in
Tanzania. ILCA Germplasm Newsletter No. 10: 16-17. ILCA,
Addis Ababa.
79
Olsen, F.J. 1972. Effect of large application of nitrogen
fertilizer on the production and protein contents of four
tropical grasses in Uganda. Tropical Agriculture 49:251-
260.
Sivalingam, T. 1964. A study of the effect of nitrogen
fertilizer and frequency of defoliation on yield. Chemical
composition and nutritive value of three tropical grasses.
Tropical Agriculturist 120: 159-180.
Stomayor-Rios, A., Ve'lez-Forturia, J., Woodburg, R. , Schertz,
K.F. and Sierra-Biacera, A. 1960. Description and cytology
of a form of Signal grass. Behaviour compared to Guinea
grass (Panicum maximum IACIC) . Journal of Agriculture of the
University of Puerto-Rico 44:208-220.
Van Rensburg, H.J. 19 60. Ecological aspects of the major
grassland types in Tanganyika. Proceedings of the 8th
International Grassland Congress, Reading, England,
pp. 307-393.
Vincente-Chandler, J., Servado Silva, S. and Fisarello, J. 1959.
Effect of nitrogen fertilization and frequency of cutting on
the yield and composition of Para grass in Puerto Rico.
Journal of Agriculture of the University of Puerto-Rico
43:215-48.
Vincente-Chandler, J., Silva, S., Albrunia, F. and Rodriguez,
J. A. 1972. Effect of two cutting heights, four harvest
intervals and five nitrogen rates on yield and composition
of Congo grass under humid tropical conditions. Puerto Rico
Journal of Agriculture 56:80-291.
Whiteman, P.C. 1980. Tropical Pasture Science. Oxford
University Press, New York.
80
EVALUATION OF THE POTENTIAL OF SOME COMMON FORAGE
GERMPLASMS IN MAURITIUS
D. Jotee
University of Mauritius, Reduit, Mauritius
Abstract
Background information on the climate, livestock production
systems, constraints to increased animal production and the role
and scope of pasture and cultivated fodders in Mauritius to
enable a proper understanding of the problem are highlighted.
The results of some recent experiments on the evaluation of
selected species of grasses for the different ecological zones
are described. Particular emphasis is laid on the need for
proper screening methods of pasture species and fodder crops and
appropriate management techniques to improve productivity for the
benefit of the livestock industry. Proposals for future research
orientation are also outlined.
LIVESTOCK AND FODDER DEVELOPMENT
Mauritius spends a considerable amount of money on the import of
ruminants. However, during the past years the Government has
given priority to development programmes in animal production,
with the objectives, inter alia, to improve human nutrition, to
create employment and to conserve foreign exchange.
Significant constraints to improved animal production in
Mauritius are the scarcity of land for fodder production and the
low productivity of existing low potential pasture land.
Historically, the establishment and management of improved
pastures has not been given priority by ruminant owners for
various reasons. In recent years, however, the need to
consolidate the livestock industries has been appreciated, and
this is reflected in the increased interest in pasture
improvement. Moreover, the Government is currently setting up
fodder units in various regions to provide incentives to small-
scale cattle keepers, as well as stabilising the sugar production
to about 600,000 tonnes/year so as to release land for food and
feed production. A better utilisation of the Crown Lands, inter
alia, through agroforestry is also being considered.
BASIC FORAGE RESOURCES
The basic feed resources can be classified into two main
categories:
1. Grasses, legumes and tree fodder.
2. Crop residues, wastes from animal production and residues
from the processing of food for human consumption.
The forage/fodder species currently grown depend, inter
alia, on the ecological conditions and the system of livestock
management adopted. According to Naidu (1973) , the most commonly
used species in order of importance for the cut-and-carry system
81
from sources other than pasture in the
regions of the island are shown in Table 1.
three agro-climatic
Table 1. Fodder species of importance in the
various ecological zones.
Ecological
zone
Fodder species in order
of importance
Subhumid Stenotaphrum dimidiatum
Cynodon dactylon
Panicum maximum
Leucaena leucocephala
Humid Pennisetum purpureum
Stenotaphrum dimidiatum
Paspalum paniculatum
Leucaena leucocephala
Superhumid Ischaemum aristatum
Pennisetum purpureum
Passiflora suberosa
On the other hand, in the drier regions, where forage
exploitation is to a large extent through grazing, the natural
pastures exist mainly under the plantations of Casuarina
equisetifolia. The pastures grazed consist essentially of
Stenotaphrum dimidiatum, Cynodon dactylon and Panicum maximum.
New species have been introduced of which Cynodon plectostachyus
and Leucaena leucocephala have been the most promising.
CULTIVATED FORAGES
In order to increase animal production in Mauritius, the
government is presently experimenting with the setting up of
fodder plantations in selected villages. Forages are also
increasingly being cultivated on government stations and to a
lesser extent on private land; the main species involved are
Pennisetum purpureum, Leucaena leucocephala, Setaria sphacelata
and Cynodon plectostachyus.
FORAGE EVALUATION STUDIES
Primarily due to the marginal role of pastures in meeting the
feed requirements of animals, the evaluation of forage in
Mauritius has been sporadic with little coordination and lack of
continuity. However, more systematic studies in the evaluation
of selected forages are presently being carried out. The main
research activities in respect of past evaluation of common
forages are highlighted to indicate the scope of past and present
activities. Unfortunately, the evaluation has not been thorough;
most of the data relate to agronomic and to a lesser extent some
nutritive value criteria.
82
Grasses
The need to exploit the potential of grasslands as an important
resource for extensive low-cost feeding (grazing) as well as for
fodder cultivation used in the cut-and-carry system is
increasingly being felt. Serious efforts to identify high
yielding and high quality grasses for the different ecological
zones, to assess the effects of fertilizer application and to
assess the effect of cutting interval and cutting height have
been made since the early 70s. These studies have been
undertaken mainly by the FAO Milk and Meat Project and the
Ministry of Agriculture and have concentrated mainly on the
collection of agronomic data.
The productivity of eight selected grasses have been
evaluated in two ecological zones (subhumid and superhumid) and
the results are presented in Tables 2 and 3.
Data for the other two species of grasses namely Cenchrus
ciliaris and Chloris gayana were not included in Table 2 due to
abnormally low productivity and poor persistency. The parameters
given for 1978 were consistently superior to those 1974; and
although Setaria sphacelata outyielded the other species in some
respects, the crude protein yield was not superior. Other
studies conducted in the superhumid zone confirms the superiority
of Setaria sphacelata (Jhundoo, 1986) in terms of fresh and dry
matter production. However, bearing in mind the fact that
protein is probably the most critical nutrient in Mauritian
forages, this does not necessarily imply that Setaria sphacelata
is the most suitable feedstuff.
Table 2. Production parameters of 6 species of grasses in super
humid zone for 1974 and 1978 (2 cuts) .
Grass species Fresh matter Dry matter Crude Crude fibre
ProteiVlt.ha-1 yr-1 t.ha--1 yr-l t.ha- t.ha- 1 yr-l
Setaria
1974* 1978** 1974 1978 1974 1978 1974 1978
sphacelata 96.2 120.2 27.4 34.2 1.4 1.8 6.3 7.9
Tripsacum
lexum 77.5 96.9 14.8 18.5 1.4 1.8 4.8 6.0
Brachiaria
ruziziensis 73.8 92.2 17.2 21.5 1.5 1.9 5.5 6.9
Brachiaria
brizantha 72.2 90.2 15.0 18.7 1.4 1.8 4.6 5.7
Pennisetum
purpureum 64.5 80.6 12.2 15.3 1.2 1.5 3.9 4.9
Panicum
maximum 57.2 71.5 13.8 17.3 1.4 1.7 4.5 5.6
S.E. of means ±3.5 ±3.5 +3.3 ±3.4 +0.4 +0.4 + 1.2 + 1.2
* 1974 : 5 months cutting interval.
**1978 : 4 months cutting interval.
Source: Naidu (1980, 1982).
83
The results of experiments carried out in the subhumid zone
are given in Table 3. Generally it can be concluded that
Pennisetum purpureum is the most vigorous, persistent and highly
adapted species, and thus can be safely recommended for the
subhumid regions especially for zero-grazing. Productivity data
for the five species for 1976 have been omitted due to their poor
performance probably resulting from unfavourable climatic
conditions.
Table 3. Production parameters of 8 species of grasses in
subhumid zone for 1976 and 1978 (3 cuts) .
Grass Fresh matter Dry matter Crude Crude fibre
species Prote
t.ha-
in -i
t.ha-1 yr t.ha"-1 yr-l * yr 1 t.ha- i yr-l
Pennisetum
1976* 1978** 1976 1978 1976 1978 1976 1978
purpureum 129.8 162.3 28.0 35.0 8.1 10.1 3.5 4.4
Centrus
ciliaris - 60.9 - 24.0 - 8.4 - 4.2
Chloris
gayana - 60.1 - 21.8 - 6.4 - 4.3
Pan i cum
maximum - 79.8 - 26.9 - 8.7 - 4.1
Brachiaria
brizantha 54.1 72.2 21.4 29.2 6.8 9.3 3.6 3.8
Brachiaria
ruziziensis 43.3 77.3 17.5 30.6 5.6 9.7 2.3 5.2
Digitaria
decumbens - 59.1 — 20.3 - 7.4 - 2.7
Setaria
sphacelata - 86.7 - 22.2 - 7.2 - 3.7
S.E. of
means ±3.5 +3.3 ±3.2 + 3.35 +0.4 ±0.41 ±1.2 _+1.2
Source: Naidu (1980, 1982).
On the other hand, assessment of the potential of
Stenotaphrum dimidiatum, the dominant species in the shaded and
open pasture of the drier regions and Ischaemum aristatum, a
common species of the superhumid zones have been virtually
neglected. Published data of Blair and Peerun (1970) on the
potential of these two forages, based on five cuts over a ten-
month period on the subhumid site at Wolmar show the superiority
of Stenotaphrum dimidiatum, especially in respect of the crude
protein content and yield (Table 4) .
84
Table 4. Dry-matter and protein yield of
Stenotaphrum dimidiatum and Ischaemum
aristatum.
Species Dry matter
t.ha-1
Crude protein
% dry matter
Crude
protein
t.ha-1
Ischaemum
aristatum 17.56 10.6 1.64
Stenotaphrum
dimidiatum 20.18 18.5 2.99
The production from Stenotaphrum dimidiatum is generally
much lower in the wet areas than Ischaemum aristatum. An
advantage of Stenotaphrum dimidiatum is its persistently high
crude protein content at various cutting intervals (Blair and
Peerun, 1970) (Table 5) . Increasing the cutting interval from
two to twelve weeks also increased productivity.
Table 5. Production and cutting interval of
Stenotaphrum dimidiatum.
Cutting
interval
Fresh
weight
(weeks) t.ha
Dry
weight
t.ha-1
Crude Crude
protein protein
% of DM t.ha-1
2 12.50 1.87 11.6 0.22
4 21.25 2.76 17.6 0.49
6 30.25 4.39 13.3 0.48
8 38.75 5.43 15.3 0.83
10 40.00 5.48 15.7 0.86
12 47.50 6.41 14.8 0.95
Data on fertilizer responses of the grasses are only
fragmentary. The need for properly fertilizing fodder
plantations and pastures not only with nitrogen, but also with
potassium, phosphorus and sulphur is not fully realised. As a
consequence there is a decrease in production in subsequent
cropping of these fodders.
The effects of different levels of nitrogen fertilizer on
the yield of three species of grass has been reported by
Heerasing (1986) . His experiments showed that increasing levels
of nitrogen application positively influenced yield in Pennisetum
clandestinum and Brachiaria brizantha only (Table 6) . However,
the results of an identical trial conducted earlier by the same
author did not reveal any improvement in yield of the three
grasses.
85
Table 6. Effect of fertilizer response of three
species of grass.
Rate of N
Yield (k/ha, 3 cuts)
application
(kg/ha)
Setaria
sphacelata
Pennisetum
clandestinum
Brachiaria
brizantha
0 135
121
130
130
65 70
92
87
94
10
15
20
69
63
82
Legumes
Besides evaluation studies on grasses, the potential of legumes,
mainly Leucaena leucocephala , has been assessed since the early
70s. It is naturalised legume and therefore offers considerable
potential for exploitation not only as fodder but also in
agroforestry programmes, alley farming, hedges and erosion
control. Its major advantage is the high crude protein level, a
nutrient which is deficient in tropical forages.
A series of agronomic trials on Leucaena leucocephala have
been conducted by Osman (1979), Rojoa (1982) and Heerasing (1984)
to assess the role of various management practices on the
productivity of this legume. Notably inter-row spacing of the
plants have been shown to exert an effect on yield (Osman, 1979;
Naidu, 1980; Rojoa, 1982; Heerasing, 1984,). Most of the authors
have come to the conclusion that an inter-row spacing of between
90-100 cm yielded the best results. Only Osman (1979) found the
optimum spacing to be 180 cm. Experiments on the effect of
cutting height on yields were also undertaken (Osman, 1979) who
in a 40-month period noted that the best cutting height was 75 to
150 cm, while Heerasing reported the optimum cutting height to be
37.5 cm.
Other conclusions reached by Osman (1979) are:
1. Cutting frequency has an important bearing on productivity.
A frequency of cutting of 90 to 150 days (depending on
season) was the most appropriate.
2. When properly inoculated with suitable Rhizobium strains,
fertilization is not needed.
3. Overgrazing by animals, especially the goat and the deer is
detrimental to Leucaena productivity.
4. A productivity of 13 tonnes/ha /yr dry matter and a
protein yield of 3.5 t/ha /yr have been obtained without
fertilization and irrigation in the humid zone.
Various nutritive value criteria have been assessed. Dry-
matter digestibility and various feeding trials have been
undertaken during the past decade. Generally, the recorded data
demonstrate that Leucaena leucocephala exerts positive effects on
animal performance when used as a supplementary source of protein
due to its ability to provide appreciable by-pass protein which
is important especially for high performance animals feeding on
low-quality fibrous feeds. Providing high levels of Leucaena
leucocephala in the diet is not only economically unsound but
influences certain parameters, inter alia, a reduction of milk fat
in lactating animals (Mardamootoo, personal communication) ,
depression in animal performance and the possibility of
intoxication.
The problem of leucaena toxicity, fortunately is a regional
one. In some countries, including Mauritius, it is well
established that there exist rumen micro-organisms capable of
degrading the toxic goitrogenic metabolite of mimosine (Jones,
1981) . Immunity against leucaena toxicity can easily be imparted
to animals via infusion of rumen fluid from immune animals or
suitable inoculum. This no doubt offers scope for increased
utilisation of this highly productive and persistent legume.
Other legumes which warrant further investigation to improve
grassland quality (Osman, 1979) are:
1. For the very wet regions:
Stylosanthes guianensis
Macroptilium atropurpureum
Neonotonia wightii
Vigna marina
2. For dry regions: Stylosanthes humilis.
The establishment method advocated is that of oversowing the
existing grass cover with the seed of these legumes.
Sugarcane
Another fodder plant of potential economic importance is
sugarcane (Saccharum officinarum L.) . The interest in the
sugarcane plant lies primarily in its high potential to trap the
sun's energy, which is in general superior to other commonly
grown crops, because of its perennial ity and its value as a dry
season feed. Biomass production of up to 64 tonnes dry matter
per hectare per annum has been recorded (Bachofen et al , 1981).
The use of sugarcane as a feed for ruminants is particularly
appropriate in times of low sugar prices, which makes the
utilisation of land through production of sugarcane for
conversion through animals an interesting proposition. The
potential of the whole sugarcane plant for feeding animals has
been given considerable research inputs in various sugar-
producing countries since the early 1970s (Jotee, 1984) .
An experiment was carried out in Mauritius to study the
effect of monthly cutting intervals from four to twelve months
(Rajkomar, 1977) . The results reproduced in Table 7 show a
slight increase in dry-matter yield but an appreciable decrease
in protein yield with time. Further studies on the effect of
87
Table 7. Effect of cutting interval on yield of sugarcane.
Cutting Fresh Dry % Crude Crude
interval matter
t/ha-1
matter
t/ha-1
protein fibre
(months) on DM basis t/ha-1
4 149.3 35.8 3.8 8.51
5 135.2 27.4 4.8 7.46
6 145.8 37.5 4.2 8.85
7 129.2 25.2 5.8 7.74
8 141.5 34.5 4.1 8.06
9 180.3 32.5 3.9 8.23
10 177.1 40.6 3.0 9.61
11 164.4 40.9 2.7 7.88
12 140.3 36.0 2.1 6.98
fertilizer application of 100, 200 and 300 kg nitrogen per
hectare (Rajkomar, 1977) did not reveal any positive response to
increased nitrogen fertilization on the overall yield.
Unlike most green forages in the tropics, sugarcane is
characterised by an increase in organic-matter digestibility from
six months to 24 months after planting (Kung and Stanley, 1982) ,
and therefore can serve as a useful fodderbank during the dry
season in certain countries. However, due to a decline in
protein content with time, proper supplementation with by-pass
protein in particular is desirable for satisfactory performance.
CONCLUSIONS
The success of any livestock development programme in Mauritius
will partly depend on the research input in pasture and fodder
crop production. Hitherto the resource persons involved in the
agronomic studies, chemical analyses and in the animal production
sector have always worked in isolation. This has resulted in a
lack of coordination which has hampered development; therefore, a
multi-disciplinary approach is essential, as well as the need for
standardisation of methodology and the use of maximum criteria
for assessment. However, before any new fodder and pasture
improvement plans are made, it is of utmost importance to make a
thorough assessment of existing data on available feedstuffs and
future requirements.
The integration of fodder production into forest plantations
has a good potential for the further development of the livestock
sector. Therefore in the drier areas, where the stomoxys fly is
not a problem, grazing by cattle/deer needs to be promoted. In
the other areas, emphasis must be laid on the zero-grazing system
mainly for the smallholders.
Future research programmes to increase forage production and
efficiency of animal production should, inter alia, include:
1. More in-depth studies to assess the various management
techniques used, viz. levels of cutting with respect to
frequency and ageing, and fertilizer responses profiles for
the main agro-climatic zones.
88
2. Further investigation of naturalised/indigenous and exotic
genotypes in relation to their persistency in long-term
experiments.
3 . Concurrent experiments to assess the nutritive value of
various forage and mixtures through digestibility studies
and feeding/grazing trials.
4. Further evaluation of other sources of feeds, namely the
sugarcane tops, maize stover and cobs and other crop
residues to enhance forage availability and hence
production.
5. Establishment of on-farm demonstrations to highlight the
benefits of improved fodder production techniques to
smallholders.
6. Provision of more training facilities, research inputs and
other logistic support.
7. Economic analyses of the various production systems.
Implementation of these proposals, no doubt, will contribute
to improving the scope and efficiency of animal production in
Mauritius.
REFERENCES
Bachofen, R., Snozzi, M. and Zurrer, H. 1981. Biomasse. Udo
Pfriemer Verlag, Munchen.
Blair, I. and Peerun, Z. 1970. Preliminary report on pasture.
Milk and Meat Project, FAO, Rome.
Heerasing, J.M. 1984. Annual Report. Ministry of Agriculture,
Natural Resources and Fisheries, Port Louis, Mauritius.
Heerasing, J.M. 1986. Annual Report. Ministry of Agriculture,
Natural Resources and Environment, Port Louis, Mauritius.
Jhundoo, C. 1986. Experimental work in progress on the
development of fodder crops and pasture lands. In: Report
of the Food and Agricultural Research Committee on Fodder
Crops and Pasture Lands, Port Louis, Mauritius.
Jones, R.J. 1981. Does ruminant metabolism explain the absence
of Leucaena toxicity in Hawaii? Australian Veterinary
Journal 57(1) : 55-56.
Jotee, D. 1984. Investigations on the feeding of sugarcane
products and their mode of presentation with Leucaena in two
breeds of goats in Mauritius. Dissertation, University of
Giessen, Federal Republic of Germany.
Kung, L. and Stanley, R.W. 1982. Effect of stage of maturity on
the nutritive value of whole plant sugarcane preserved as
silage. Journal of Animal Science 54(4):89-95.
89
Naidu, S. 1973. Fodder species used by small cowkeepers for
cattle feeding in Mauritius. Dissertation, University of
Mauritius, Reduit, Mauritius.
Naidu, S. 1980. Annual Report. Ministry of Agriculture, Natural
Resources and Fisheries, Port Louis, Mauritius.
Naidu, S. 1982. Annual Report. Ministry of Agriculture, Natural
Resources and Fisheries, Port Louis, Mauritius.
Osman, A.M. 1979. Protein from crops for cattle production in
Mauritius with emphasis on Acacia, Leucaena leucocephala
(LAM) de Wit. Proceedings of the National Agricultural
Conference. University of Mauritius, Reduit, Mauritius,
pp. 387-406.
Rajkomar, I. 1977. Annual Report. Ministry of Agriculture,
Natural Resources and Fisheries, Port Louis, Mauritius.
Rojoa, H. 1982. Annual Report. Ministry of Agriculture, Natural
Resources and Fisheries, Port Louis, Mauritius.
90
A REVIEW OF FORAGE GERMPLASM INTRODUCTION
AND EVALUATION IN UGANDA
g.s. Byenkya
Animal Health Research Centre
P.O. Box 24, Entebbe, Uganda
ABSTRACT
Forage gerntplasm introduction and evaluation work in Uganda has
been reviewed. Many species have been evaluated by different
workers in different parts of the country. This has been in
terms of their establishment, persistence and production
characteristics, fertilization and yields, defoliation, chemical
composition and nutritive values, animal productivity and their
seeding abilities. From these studies, the Ministry of
Agriculture has demarcated Uganda into six zones with
recommendations of various pasture species for each zone. Most
of the grass species evaluated and recommended are indegenous to
Uganda, while nearly all the legume species are exotic.
Availability of planting seed has been observed as the major
hindrance to pasture development programmes in the country.
INTRODUCTION
Uganda is one of the East African countries crossed by the
Equator, and lying between latitude 4°12'N and 1°29'S and
longitude 29°34'E and 35°E. It is bordered by the Sudan in the
north, Kenya in the east, Tanzania and Rwanda in the south and
Zaire in the west. It covers a total area of 235,886 sq. km.
The climate is good, with a mean annual rainfall varying
from 510 mm in the North-Eastern parts of Karamoja to 2160 mm or
more in the Sesse Islands in Lake Victoria. The rainfall is bi-
modal with the first peak being in March-May (first rains) and
the second in September-November (second rains) with dry periods
in between the peaks. Further north the two seasons tend to
coalesce into one rainy season of about five months centred
around July and one long dry hot season. A mean monthly maximum
temperature exceeding 37.8 C has been recorded in some parts of
the country with a seasonal variation barely above 6°C. A figure
of 10°C for mean monthly minimum temperatures has also been
obtained with a seasonal variation of just about 3°C.
Generally most soils are able to support some form of
agriculture. Those soils with high productivity include:
1. those associated with volcanic activities,
2. those clays formed from parent material derived partly or
wholly from basic amphibolite rocks,
3. those deep, red or brown, loam or clay loam, pediment soils,
4. deep sedimentary soils.
91
The above four groups are well represented all over the
country and have been cultivated extensively. There are also
those soils with low to nil productivity, which are shallow or
very sandy or sometimes with a lot of gravel. This group also
includes those montane soils on the slopes of mountains. These
are mostly in the north with localised patches and pockets in the
rest of the country.
Vegetation varies from rain forests to savanna and grassland
with a small patch of bushland and thicket in the extreme
northeastern part of the country in Karamoja.
Agriculture is the backbone of Uganda's economy, engaging
the majority of the population and currently accounting for about
95% of all foreign exchange earnings. Crop husbandry plays the
leading role in the overall farming system and is poorly
integrated with animal husbandry. Livestock are an important
factor in the ordinary life of the majority of Ugandans. Their
values to the owners range from wholly economical to
social/prestigious uses in marriages and other cultural
ceremonies and as security against shortcomings. The
productivity of these herds is, however, very low. The livestock
population density is greatest in the northern and northeastern
parts of the country. These areas also tend to have low rainfall
and can be classified as natural grasslands or rangelands.
According to the 1979 statistics of the Ministry of Animal
Industry and Fisheries, Uganda had 5,053,510 cattle, 2,453,528
goats and 962,315 sheep.
LOCATION OF PASTURE RESEARCH IN UGANDA
A number of institutions are involved in pasture research in
Uganda. These stem from two different parent ministries, namely
the Ministry of Agriculture and the Ministry of Animal Industry
and Fisheries (Veterinary Department) . Although pasture work can
be traced way back to the turn of this century, it was not until
1920 that a proper experimental station was established.
Currently three research institutions are engaged in pasture
research in the country in addition to the Faculty of Agriculture
of Makerere University. These are:
1. Serere Research Station - established in 1920 in the East to
carry out experimental work in the so-called "short-grass
areas" i.e. in the eastern and northern parts of the
country. This belongs to the Department of Agriculture and
carries out research on crops including pastures and some
animal work.
2. Kawanda Research Station - about 50 km north of Kampala was
established in 1937 in the "tall or elephant grass areas".
It belongs to the Department of Agriculture and it is the
seat of crop research in the country but with some animal
production work going on.
3. Animal Health Research Centre (AHRC) - situated south of
Kampala belongs to the Department of Veterinary Services and
92
Animal Industry. It carries out research and diagnostic
services in both animal health and production.
There are, however, some smaller agricultural research
stations and agricultural colleges which at some stage have been
involved in pasture work. These include Namulonge Research
Station, Bukalasa Agricultural College and Arapai Agricultural
College, all three belonging to the Ministry of Agriculture.
Because of the many institutions involved, which belong to
different parent ministries, there arises the problem of
coordination of research efforts to avoid duplication. Although
Uganda's third Five-Year Development Plan 1971/72-1975/76 stated
that "In order to avoid duplication and to effect a better
integration of efforts, in future all Government research on
livestock and pasture will be sponsored and administered by the
Ministry of Animal Industry," this has not fully been
implemented.
HISTORICAL HIGHLIGHTS OF
PASTURE RESEARCH PRIOR TO 19 60
A brief chronological listing of the highlights of pasture
research findings in Uganda up to 1960 has been given by
Henderlong (1973) as follows:
1906 Introduction of legume species primarily for vegetative
cover and soil conservation.
1925 A collection of pasture grasses started by Maitland near
Kampala.
1930 Selected species from the 1925 planting were established in
larger plots at Bukalasa for general observations on grazing
and feeding.
1931 Grass collections were started at Ngetta and later at
Serere.
1932 Continuous cultivation of arable crops even with green
manures was shown at Serere to result in decreased crop
yields when compared to crop yields under shifting
cultivation.
1933 All governmental farms in Uganda changed their crop
rotations to include a two- to four-year planted grass rest
(ley) phase. This policy gave the first real impetus to
research investigations with the local grasses.
1945 Stylosanthes gracilis was introduced in Uganda from
Australia by Thomas (1966) and planted at Kawanda and later
at Serere.
1948 Chloris gayana was recommended as the best overall grass
species for the grazed temporary ley throughout most of
Uganda. In some areas, elephant grass (Pennisetum
purpureum) was also commonly used for the temporary ley
phase, but it proved difficult to eradicate from cropland.
93
It required vegetative planting, and it was better adapted
as a fodder crop than for direct grazing.
1954 Detailed systematic work on grass and grass-legume mixtures
was initiated at Serere by Horrell (1958).
1956-1958 Introduction of legume species for the "forest area"
was initiated at Kawanda in addition to the earlier grass
museum and introductions at Kawanda by Thomas (1966) . These
initial studies included over 100 different grass species,
of both temperate and tropical origins, and approximately 60
different tropical and temperate legumes.
EVALUATION STUDIES OF PASTURE SPECIES
Much of the earlier detailed investigative work was done mainly
on grasses. Various texts were published on these grasses.
Initially they were investigated for the rest period after arable
cropping, and their importance as a source of grazing or fodder
was then subsidiary. The effects of the animals on soil
fertility were unclear until it was later shown that grazing did
not reduce yields of subsequent crops (Kerrham, 1947) .
The introduction of exotic animals after 1960 was an impetus
to extensive grassland improvement. Evaluation of both
indigenous and exotic pasture species was initiated as a result
of this development. These have, over the years been evaluated
in terms of management studies on establishment, persistence and
production fertilization and yields and defoliation. Evaluation
further concerned itself with chemical composition and nutritive
values, animal productivity and seed production.
Establishment, Persistence and Production
Horrell (1958, 1963) gives an account of a wide range of grasses
and legumes studied at Serere in nursery plots to determine their
suitability for leys in eastern and northern Uganda. He pays
special attention to persistence, production, habit, seeding
ability and to a lesser extent palatability. The origins of the
species in question are also given. From such studies, he was
able to recommend the following species for use in eastern and
northern Uganda.
Grasses:
Chloris gayana
Cenchrus ciliaris
Cynodon plectostachyus
Hyparrhenia rufa
Pennisetum purpureum
Setaria spp.
Paspalum notatum
Cynodon dactylon
for leys
for dry season
supplementary grazing
for lawns
Legumes: The most successfully adapted
these conditions included:
legumes under
94
Stylosanthes gracilis, Calopogonium orthocarpum,
Centrosema pubescens, Desmodium spp. , Pueraria
phaseoloides. Glycine javanica (now Neonotonia
wightii) .
All of the above except Neonotonia wightii were exotic to
Uganda. The study demonstrated the ability of a number of
tropical legumes to grow vigorously and reproduce in this
environment. A good growth of Stylosanthes gracilis with a crude
protein content of up to 24% has been achieved at Animal Health
Research Centre (AHRC Entebbe, unpublished data) . Chloris gayana
has also been successfully established under a silage crop cover
of sorghum at Entebbe while Melinis minutiflora was unable to
establish (Harker, 1954) . Pennisetum clandestinum was found to
grow well in pot experiments at AHRC, Entebbe (unpublished) .
Other pasture species have been tried in different environments
in Uganda, but the reports are difficult to come by.
Fertilization and Yields
The use of fertilizers was considered important in pasture work
in Uganda. Nitrogen fertilization was established to be
essential for good production of pure grass swards or leys
(Horrell and Bredon, 1963; Horrell and Tiley, 1970; Olsen, 1972).
It was found that production of grass species falls sharply in
the second and third years from planting but that yields could be
maintained at a steady level by application of N or by the
inclusion of a successful legume such as Desmodium intortum.
Phosphorus or superphosphate was proved valuable for the
establishment of legumes (Stobbs, 1969a; Olsen and Moe, 1971) .
They reported that liming was useful for the persistence of
lucerne and its nodulation but showed no general effects on the
establishment, persistence, production or nodulation of either
Desmodium or Stylosanthes.
Responses of Hyparrhenia rufa, Panicum maximum, Stylosanthes
guianensis, Centrosema pubescens and Macroptilium atropurpureum
to phosphorus, sulphur and potassium in eastern Uganda were
investigated by Wendt (1970) . All the species responded to P and
S to levels of application of about 70 kg/ha of P and 20-40 kg/ha
of S giving yield increases of 40-100%.
Much work has been done to evaluate the compatibility of
various grasses and legumes when planted in mixtures. Results of
Olsen and Tiharuhondi (1972) discouraged the sowing of Desmodium
intortum with Chloris gayana. Both D. int-rtum and Medicago
sativa made good association with Panicum maximum and Setaria
anceps (now sphacelata) .
Defoliation
Literature on this aspect of forage evaluation is scanty. It
appears that few studies on this aspect were undertaken. Horrell
and Bredon (1963) working on Panicum maximum at Serere reported
highest DM yields at cutting intervals of nine weeks while best
yields of CP were obtained at cutting intervals of between six
and nine weeks. Although there are no reports on legume
defoliation studies in Uganda, Henderlong (1973) observed that
95
the rate of regrowth of lucerne after defoliation was
considerably more rapid than for any of the adapted tropical
legumes.
Chemical Composition and Nutritive Values
A number of workers evaluated various species for their chemical
composition and nutritive values. Some of this work was carried
out in different ecological zones. Species were analysed for
crude protein, crude fat, crude fibre, ash and silica contents
and for the various mineral element contents. This was done on
both planted and rangeland pastures but mostly on grasses. The
various grass species analysed in greater details in this respect
were Pennisetum purpureum, Themeda triandra, Panicum maximum,
Chloris gayana, Cynodon dactylon, Digitaria spp., Hyparrhenia
spp., Melinis minutiflora, Setaria spp. and many other range
species.
Pennisetum purpureum, an indigenous grass of Uganda, has
been found to be of very high value as cattle feed with a CP%
content ranging between 9.2-20.5 and CF% of 29.3-37.7 (Marshall
and Bredon, 1963) . It is used both as a pasture plant as well as
fodder crop. Although pure pasture leys of individual herbage
species are unusual in Uganda, P. purpureum and Setaria
sphacelata have proved to be among the most productive pasture
grasses and have been proposed for commercial use (Morrison,
1971) .
Nutritive value assessments of pastures and forages have
been reported by Juko and Bredon (1961), Bredon et al , (1963),
Musangi (1965) and many others. The nutritional value of some
common cattle browse and fodder plants of Karamoja, northeastern
Uganda, has also been investigated by Wilson and Bredon (1963) .
Due to the role played by rangeland for the grazing animal,
several workers have attempted to evaluate the quality of range
pastures in different parts of the country. Reports of Reid et
al. (1973) indicate that natural grasslands in Uganda had lower
digestibilities than the planted species. But the higher
digestibilities in planted species declined more rapidly with
maturity than the natural grasslands. They also reported that
Brachiaria generally exhibited one of the highest levels of in
vitro digestibilities among the species evaluated which agrees
with the in vitro data of Soneji et al. (1971) which indicated
higher digestibilities for Brachiaria ruziziensis than either
Chloris gayana or Setaria sphacelata. Legume species on the
other hand generally had lower digestibilities than the grasses
in the early stages of growth, but the decline in legume
digestibility was considerably less marked than for the planted
grasses (Reid et al . , 1973).
Digestibility trials and chemical analyses of indigenous
grasses at AHRC indicated adequate starch equivalent values for
both maintenance and an element of production throughout the
year while protein was frequently limited (Marshall et al . , 1961;
Bredon and Torrell, 1962; Bredon and Marshall, 1962). Proximate
analyses on rangeland grasses of western Uganda indicated lower
CP values for Themeda triandra (4.14%) and Hyparrhenia
filipendula (4.41%) compared to Brachiaria spp. (6.49-7.64%),
96
Chloris gayana (5.96%), Panicum maximum (6.61%) and Cynodon
dactylon (8.27%).
Various workers have analysed various pasture plants for
their mineral contents in different localities. Minerals
analysed for include: N, Na, K, Mg, P, Mn, Cu, Zn, Co, Mb, S, Sn.
Investigations on the mineral status of the commonly used grasses
on farms in Buganda and Busoga and also in western Uganda in the
Queen Elizabeth National Park indicated adequate amounts of N, K,
Mg, Mn, Cu, Mb and S while levels of Na, Ca, P, Zn and Co were
marginal or deficient in many grasses thus requiring
supplementation (Long et al., 1969; Long et al., 1970).
Mean mineral composition (%) of some Uganda grasses as
reviewed by Harrington and Pratchett (1973) are given in Table 1.
Table 1. Mean mineral composition (%) of some Uganda grasses.
Sodium Potassium Magnesium Calcium Phosphorus
Themeda
triandra 0.034 0.50 0.09 0.31 0.11
Chloris
gayana 0.041 1.17 0.15 0.29 0.22
Setaria
sphacelata 0.046 1.82 0.13 0.41 0.26
Brachiaria
playtynota 0.034 1.19 0.19 0.40 0.17
B. brizantha 0.045 1.16 0.17 0.39 0.21
Ssekaalo (1972) working on eastern Uganda pastures,
concluded that grass species generally contained less cobalt than
legumes. Grasses had little difference in their cobalt content
and so did the legumes with the exception of Macroptilium
atropurpureum. The cobalt content was enough to meet animal
requirements except where Pennisetum purpureum was the main
fodder. Selenium status of pastures in Uganda has also been
analysed by Long and Marshall (1973) . The better quality
grasses, judged on the bases of CP content and DM digestibilities
such as Brachiaria spp. , Cynodon dactylon, Panicum maximum and
Setaria aequalis tended to have satisfactory mineral contents
while the more typical savanna type grasses, Hyparrhenia
tilipendula and Themeda triandra were generally low in essential
minerals (Long et al . , 1969).
Animal Productivity
Forage evaluation through livestock productivity has been studied
at Serere by Stobbs (1969c) and Otim (1973) . Higher values in
terms of animal production have been obtained when various
legumes have been included in the grazing systems. Stobbs
(1969b) investigated Stylosanthes gracilis and Centrosema
pubescens as mixtures, top dressed with single superphosphate
fertilizer. Stylosanthes gracilis because of its small hairy
leaves and woody stems generally had low intake while
Calopogonium mucunoides , one of the best legumes tested at
97
Serere, was completely unacceptable to the animal (Otim, 1973) .
Paspalum notatum was able to maintain excellent beef production
at Namulonge Station, indicating its relatively high
digestibility.
Seed Production
The availability of planting seed is a major hindrance to pasture
development efforts in Uganda. Preliminary evaluations of the
various species for seed production have indicated that most of
the various recommended species were able to produce seed.
Emphasis has been put on legume seed production. Stylosanthes,
Neonotonia, Centrosema and Macroptilium spp. have produced
reasonable seed yields at Serere while Desmodium and Medicago
spp. were not as successful (Wendt, 1970) .
RECOMMENDED PASTURE SPECIES
General planted pasture zones have been established
recommended species for each zone, as follows:
with
ZONE 1: Pennisetum purpureum
(Hoima, N. Toro, Mubende, Mpigi, S. Busoga, E. Masaka)
Grasses: Pennisetum purpureum
Panicum maximum
Chloris gayana
Brachiaria ruziziensis
Setaria sphacelata
Legumes: Desmodium intortum
Desmodium uncinatum
Medicago sativa
Neonotonia wightii
Stylosanthes gracilis (on poor or
soils)
Centrosema pubescens
Macroptilium atropurpureum
stony
zone 2 : Moist western areas
(S.E. Toro, N. and W. Ankole, N. Kigezi, S. West Nile)
Grasses: Chloris gayana
Setaria sphacelata
Brachiaria ruziziensis
Melinis minutiflora
Legumes: Desmodium intortum
Desmodium uncinatum
Macroptilium atropurpureum
Trifolium spp. (above 1700 metres) .
ZONE 3: Highland areas
(S. Kigezi, W. Toro, S. Sebei, E. Bugisu)
Grasses: Chloris gayana
Setaria sphacelata
Pennisetum clandestinum
Pennisetum purpureum (for fodder)
Melinis minutiflora
Legumes: Desmodium intortum
Desmodium uncinatum (lower altitudes)
Trifolium spp.
Medicago sativa
ZONE 4: Hyparrhenia spp.
(West Nile, Madi-Moyo, Acholi,
Teso, N. Sebei, W. Bugisu,
Luwero, N.E. Mubende, Masindi, N
W. Karamoja, Lango,
Tororo, N. Busoga,
Toro)
Grasses: Panicum maximum
Hyparrhenia rufa
Chloris gayana
Setaria sphacelata
Brachiaria spp.
Cynodon dactylon
ZONE 5: Themeda triandra
(Central Karamoja, S.W. Toro, E. and N.W. Ankole
N.W. Masaka, S.W. Mpigi, S. Mubende)
Grasses: Chloris gayana
Cenchrus ciliaris
ZONE 6:
Legumes : Stylosanthes gracilis
Macroptilium atropurpureum
Not suitable for leys
(E. Karamoja, Muntain summits)
DISCUSSION AND CONCLUSION
The climate of Uganda is generally considered as being good.
There is adequate well distributed rainfall in most parts of the
country. The soils are generally good. Hence many crops
including pasture are capable of growing in various parts of the
country.
Pasture work in Uganda can be traced as far back as the
beginning of this century. The initial aim of pasture
establishment was to assist in the regeneration of soil fertility
of continuously farmed land for the subsequent crops. The value
of grazing these pastures was by then unknown as it was believed
that this would interfere with the normal process of soil
regeneration and hence affect the yield of the following crops.
A journey across most parts of Uganda will reveal the
availability of ample amounts of unutilised herbage. These
natural pastures can easily be grazed by livestock. The
availability of such herbage has not been an inducement to
99
pasture improvement schemes involving farmers. Increasing
pressure on the land as already observed in parts like Kigezi in
the southwest, Mbale in the east and a few other places will
require efficient management of the resources so as to support
the increasing human and animal populationso. The provision of
better adapted, more nutritious forages for livestock will go a
long way in improving the productivity of national herds. Uganda
is rich in its plant genetic resources of forage potential which
are as yet to be exploited. Many indigenous grass speeies have
been evaluated for their forage value and many have been
recommended for various parts of the country. Indigenous legumes
on the other hand have not been exploited. Only Neonotonia
wightii has been studied in detail and recommended as a useful
pasture plant. The rest of the legumes evaluated and recommended
are exotic, most of them being of South American origin. They
have been evaluated in terms of their establishment and
persistence, chemical composition and nutritive values,
fertilization and yields, defoliation and seed production.
Livestock productivity from such pasture has also been evaluated.
Some of these trials have, however, been limited to the research
stations only, taking into account the two broad areas i.e. "long
grass areas" and "short grass areas" . Extensive multilocational
trials of various species have not been effected and this could
be the cause of poor performance in some areas. This should
therefore be carried out.
One of the current problems limiting forage evaluation
studies and requiring attention is the lack of planting seed.
Many of the species tried have been found to seed well. In the
past Kawanda and Serere have produced limited quantities of
pasture seed, but this has since stopped. Currently, pasture
seed must be imported from abroad if required. For researchers
needing small quantities of seed, the solution has been solved by
ILCA's FLAG programme which has gratefully supplied small
quantities of seed for local multiplication purposes.
Despite the important achievements in pasture research in
Uganda, one wonders to what extent this has contributed to the
overall livestock productivity in the country. The majority of
livestock owners are small peasant farmers who graze their herds
on communal land. These have obviously gained very little, and
the idea of pasture improvement could as well as be a myth to
them. There is lacking an integrated package which can be
presented to farmers in different parts of Uganda so as to
revolutionise their agriculture and enable them to increase their
productivity and incomes. Many factors obviously contribute to
the lack of an adequate and acceptable technology package of
improvements for farmers which include traditional beliefs of
large livestock numbers rather than productivity, land tenure
systems, tendency to divorce animal from crop husbandry, lack of
coordination between research and extension, and lack of
incentives and proper training for both research and extension
agents. There is still a need for integrated research into
systems of farming. Only a few progressive farmers and
government-owned livestock enterprises have gained from the
results obtained through research. With increasing population
pressure on the land and hence reduction in communal grazing
lands, it is likely that alternative methods of feeding and
improvements will be sought by the majority of livestock owners,
100
but this must be coupled with a good supply of planting seed to
farmers.
ACKNOWLEDGEMENTS
The author thanks the PANESA Coordinator for inviting him to
attend the Workshop and to present this paper. Thanks also go to
the Permanent Secretary, Ministry of Animal Industry, Uganda, for
permission to attend the Workshop.
REFERENCES
Bredon, R.M. and Marshall, B. 1962. Selective consumption of
stall fed cattle and its influence on the results of a
digestibility trial. E. Afric. Agric. For. J. 27(3) :168-
172.
Bredon, R.M. and Torrell, D.T. 1962. The chemical composition
and nutritive value of common grasses in Uganda. Part II.
The comparison of chemical composition and nutritive values
of grasses throughout the year, with special reference to
the later stages of growth. Trop. Agric. Trin. 39:13-17.
Bredon, R.M. , Harker, K.W. and Marshall, B. 1963. The nutritive
value of grasses grown in Uganda when fed to Zebu cattle,
II. The relation between crude fibre and nitrogen-free
extract and other nutrients. J. Agric. Sci. 61:105-108.
Harker, K.W. , 1954. The establishment of Chloris gayana under a
sorghum silage crop. E.A. Agric. J. 20:54-56.
Harrington, G.N. and Pratchett, D. 1973. Cattle diet on Ankole
rangeland at different seasons. Trop. Agric. Trin. 49:211-
219.
Henderlong, P.R. 1973. Pasture research in Uganda. Past,
present and future. In: Proceedings of the First Symposium
of the Uganda Society of Agronomy. Uganda Society of
Agronomy, pp. 30-47.
Horrell, C.R. 1958. Herbage plants at Serere, Uganda, 1957-
61. E. Afric. Agric. For. J. 24:41-46, 133-138.
Horrell, C.R. 1963. Herbage plants at Serere, Uganda, 1957-61.
E. Afric. Agric. For. J. 28:174-180.
Horrell, C.R. and Bredon, R.M. 1963. Management studies with
Panicum maximum in Uganda. 1. Effect of cutting interval
and nitrogen fertilizer on yield. Empire Journ. of Exper.
Agric. 31:335-342.
Horrell, C.R. and Tiley, G.E.D. 1970. Grassland. In: J.D.
Jameson (ed.), Agriculture in Uganda. Oxford University
Press, London.
Juko, C.D. and Bredon, R.M. 1961. The chemical composition of
leaves and whole plant as an indicator of the range of
101
available nutrients for selective grazing by cattle. Trop.
Agric. (Trin.) 38:179-187.
Kerrham, R.K. 1947. Elephant grass fallow in Uganda. 1.
Grazing during the fallow period. E.A. Agric. J. 13:3-7.
Long, M.I.E., Ndyanabo, W.K., Marshall, B. and Thornton, D.D.
1969. Nutritive value of grasses in Ankole and the Queen
Elizabeth National Park, Uganda. IV. Mineral content.
Trop. Agric. (Trin.) 46:201-209.
Long, M.I.E., Thornton, D.D., Ndyanabo, W.K., Marshall, B. and
Ssekaalo, H. 1970. The mineral status of dairy farms in the
parts of Buganda and Busoga bordering Lake Victoria, Uganda.
II - Nitrogen and mineral content of pastures. Trop. Agric.
(Trin.) 47:37-50.
Long, M.I.E. and Marshall, B. 1973. The selenium status of
pastures in Uganda. Trop. Agric. (Trin. J 50:121-128.
Marshall, B. , Bredon, R.M. and Juko, C.C. 1961. The nutrition of
Zebu cattle. Part IV. The intake of dry matter by cattle
with some notes on the digestibility of Rhodes grass and
star grass hays. J. Agric. Sci. 56:191-196.
Marshall, B. and Bredon, R.M. 1963. The chemical composition and
nutritive value of elephant grass (Pennisetum purpureum) .
Trop. Agric. (Trin.) 40:63-66.
Morrison, J. 1971. Pasture development in the high potential
areas of Kenya. Span 14:166-169.
Musangi, R.S. 1965. Feed intake studies in ruminants. 1. The
digestibility of tropical grass/legume herbage mixture
during the dry season. African Soils 10:313-317.
Olsen, F.J. and Moe, P.G. 1971. The effect of phosphate and lime
on the establishment, productivity, nodulation and
persistence of Desmodium intortum, Medicago sativa and
Stylosanthes gracilis. E. Afric. Agric. For. J. 36:29-37.
Olsen, F.J. and Tiharuhondi, E.R. 1972. The productivity and
botanical composition of some selected grass/legume mixtures
at different seeding rates. E. Afric. Agric. For. J. 38:16-
22.
Olsen, F.J. 1972. Effect of large applications of nitrogen
fertilizer on the productivity and protein content of four
tropical grasses in Uganda. Trop. Agric. (Trin.) 49:251-260.
Otim, J.J. 1973. M.Sc. thesis, Makerere University, Kampala,
Uganda.
Reid, R.L., Post, A.J., Olsen, F.J. and Mugerwa, J.S. 1973.
Studies on the nutritional quality of grasses and legumes in
Uganda. I - Application of in vitro digestibility techniques
to species and stage of growth effects. Trop. Agric.
(Trin.) 50:1-5.
102
Soneji, S.V., Musangi, R.S. and Olsen, F.J. 1971. Digestibility
and feed intake investigations at different stages of growth
of Brachiaria ruziziensis, Chloris gayana and Setaria
sphacelata using Corriedale wether sheep. E. Afric. Agric.
For. J. 37:125-128.
Ssekaalo, H. 1972. Cobalt and molybdenum contents of pasture
plants on dairy farms of the eastern region of Uganda.
Trop. Agric. (Trin.) 49:221-226.
Stobbs, T.H. 1969a. The effect of inorganic fertilizers upon the
adaptation, persistency and production of grass and
grass/legume swards in eastern Uganda. E. Afric. Agric.
For. J. 35:112-117.
Stobbs, T.H. 1969b. Animal Production from Hyparrhenia grassland
oversown with Stylosanthes gracilis. E. Afric. Agric. For.
J. 35:128-134.
Stobbs, T.H. 1969c. The value of Centrosema pubescens (Benth)
for increasing animal production and improving soil
fertility in northern Uganda. E. Afric. Agric. For. J.
35:197-202.
Thomas, A.S. 1966. Ecology and human influences. In: w. Davies
and C.L. Skidmore (eds) , Tropical pastures. Faber Ltd,
London. pp. 46-58.
Wendt, W.B. 1970. Pasture Agronomy Annual Report, Part II.
Serere Research Station, Uganda.
Wilson, J.G. and Bredon, R.M. 1963. Nutritional value of some
common cattle browse and fodder plants of Karamoja, N.
Province, Uganda. E. Afric. Agric. For. J. 28: 204-208.
103
FORAGE DRY-MATTER PRODUCTIVITY VARIATION OF
PANICUM AND CYNODON ECOTYPES IN MALAWI
H.D.C. Msiska
Lunyangwa Research Station
P.O. Box 59, Mzuzu, Malawi
B.H. Dzowela
ILCA, P.O. Box 46847, Nairobi, Kenya
and
J.T. Munthali
Chitedze Research Station
P.O. Box 150, Lilongwe, Malawi
Abstract
Both Panicum maximum and Cynodon nlemfuensis strains have great
variability in forage dry-matter productivity and responsiveness
to nitrogen fertilizer application. This variability has been
demonstrated and it warrants a wide selection potential for new
strains to replace old cultivars in these two species.
INTRODUCTION
In recent years many developing countries including Malawi have
realised the importance of local forage plant genetic resources.
However, the lack of funds to facilitate wide-scale collection
and detailed evaluation of the local germplasm has limited
exploitation of these local forage plant genetic resources.
Instead, imports of both legumes and grasses have been done from
Australia and other countries. Most of the imported materials
are, however, locally represented. To minimise expenses on
importations of seed, it was found necessary to critically look
at local ecotypes to come up with some cultivars for different
ecological niches. The evaluation work is limited by a lack of
experience in local forage exploration and by availability of
funds. Forage productivity evaluation is therefore limited to a
few most promising germplasms namely Neonotonia, Panicum and
Cynodon species. Only the work on Panicum maximum and Cynodon
nlemfuensis is reported here.
MATERIALS AND METHODS
During the 198 0/81 growing season, a forage plant collection trip
was made to some parts of the central and northern regions. The
target areas were chosen on the basis of climate and topography
hence, seasonally waterlogged valleys (dambos) , river courses and
the eastern escarpments of the highland zones. Dambo and river
courses were chosen to give an indication of forage species whose
seed originally came from the inaccessible hills and valleys
beyond the escarpments. Collections were made along roads
crossing dambos, rivers and in some other places of ecological
interest.
104
Vegetative materials were dug out and put in black polythene
bags; water was poured in and then the bags were tied at the
mouth. The most promising star and guinea grass materials based
on one growing season single row nursery observation were put in
two trials respectively as follows:
Trial 1: Star grass strains evaluation trial
Twenty-four entries from the southern African region and local
collections were put in a split-plot design trial with two
replicates. The plots which measured 9 square metres were split
two-ways into three nitrogen fertilizer application rates of 20,
40 and 60 kg N/ha applied in the form of calcium ammonium nitrate
(26% N) and three cutting regrowth intervals of 3, 6 and 9 weeks
respectively.
Trial 2: Guinea grass (Panicum maximum Jacq.) strains evaluation
trial
Fifteen collections, mostly indigenous to Malawi were planted in
the same design as described in Trial 1. The only difference was
that fertilizer rates of 0, 40 and 80 kg N/ha were applied. In
both trials, the grasses were cut, weighed fresh and subsampled
for the determination of dry matter after oven drying at 85°C for
twenty-four hours.
RESULTS AND DISCUSSION
1. Star grass
Yield data for the star grass strains for the 1984/85 and 1985/86
growing seasons are presented in Tables 1, 2, 3 and 4. Two
accessions, East Coast B and No. 343 Star gave the highest forage
dry-matter yields and these were in excess of 5000 kg/ha. These
are the more robust growing star grasses, with larger stolon and
leaf material than the small-slendered leaf and stolon materials.
Second ranking were those strains of 4000 kg/ha yield levels.
These were Chololo, Ichinga, Monzi, Misuku, Camper Down, Thunda
and Chinunkha. These were of intermediate stolon material. The
least amounts of forage were produced by the three Malawlan
collections, Nyungwe, Meru and S.E. Namwera all of which gave
yields of less than 2500 kg/ha. These were materials least
robust in growth and appearing less aggressive in that their
stolon growth was rather restricted; hence they formed open
turfs. The two commercially popular cultivars, Muguga and No. 2
Star, that have since been grown at Chitedze Research Station,
were intermediate, giving yields of about 3500 kg/ha.
The variability in forage productivity is explained by the
fact that there were two different species in this collection.
In the absence of critical taxonomic and botanical variety
description, it was obvious that the material demonstrated a lot
of physical differences. Plants that fitted the description with
respect to Cynodon dactylon (Harlan and de Wet, 1969) were less
robust in their growth habit. The more robust material probably
belonged to C. nlemfuensis (Clayton and Harlan, 1970; Harlan,
1970) .
105
Table 1. Forage DM yields (kg/ha) of Cynodon cultivars across
two replicates in response to cutting regrowth
intervals (1984/85 season) .
Cultivars 3 weeks 6 weeks 9 weeks Mean
1. Nyungwe 1840 2078 3303 2407
2. Chololo 2537 4088 5902 4142
3. East Court A+ 2790 3153 4640 3528
4. Durban+ 2053 2880 4828 3254
5. Thornville 2183 3070 4970 3408
6. Ichinga 2455 4450 5407 4104
7. Muguga 2393 3027 4690 3370
8. Monzi No. 2+ 2210 4860 6950 4680
9. Baka 2142 2618 4492 3083
10. Phoka Court 2150 3452 5375 3659
11. No. 171 Star+ 2810 3777 5538 4042
12. Coast Gross 11+ 2613 2658 4532 3268
13. No. 2 Star 1657 4043 4702 3467
14. Misuku 3597 5007 5758 4787
15. No. 161 Star+ 2625 3043 5565 3744
16. Chilambula 2335 2858 6027 3740
17. Camper Down+ 2832 5883 4893 4536
18. No. 343 Star+ 4570 4700 6107 5126
19. S . E . Namwera 3198 2889 2503 2863
20. Thunda 4175 3538 4677 4130
21. Monsi No.l+ 2748 3418 4472 3546
22. Chinunkha 3865 4203 4820 4296
23. Meru 2308 2213 2845 2456
24. East Court B+ 3105 4205 8105 5138
Means 2716 3589 5042
S.E. of cultivar means +640*
S.E. of cutting intervals means +121**
S.E. of interaction +209*
Overall C.V. 38%
+ introduced entries
Where * is significance at P=0.05
** is significance at P+0.01
Increasing nitrogen levels generally resulted in
significantly higher yields (P=0.05), but some strains were less
responsive to nitrogen depending on the cutting regrowth
intervals. Materials with a generally open turf responded poorly
and lacked a response to increasing nitrogen fertilization. This
was particularly true of the local Malawian ecotypes, Nyungwe,
Phoka Court, Chololo, Thunda, Meru and a few imported strains,
No. 161 Star, No. 2 and Thornville. There was also a significant
increase in forage DM yields in response to increasing cutting
regrowth intervals (P=0.01). This was no doubt a consequence of
increased forage maturity.
During the 1985/86 growing season, all strains responded to
increased cutting regrowth intervals from 3 to 9 weeks (P=0.001)
as shown in Table 3. There was also a significant response to
increasing levels of nitrogen fertilization, although the levels
of 20 and 40 kg/ha were not statistically different. The strains
106
Thornville, Baka, Misuku and No. 161 Star were not responsive to
increasing nitrogen fertilization (Table 4) . In the analysis,
the differences between cultivars did not turn out significant,
however, the largest amounts of forage, in excess of 8000 kg/ha,
were given by the strains East Coast A, Coast Cross II, No. 343
Star and Durban. The least amounts of forage, 6,500 kg/ha or
less, was given by the strains, Meru, East Coast B, Chilambula
and Phoka Court. The fact that in the second harvest year, No. 2
Star fell in the top ranking group while Muguga maintained its
position in the intermediate yielding category, shows that Star
No. 2, once it is fully established, has a high yielding
potential.
Table 2. Forage DM yields (kg/ha) of Cynodon cultivars across
two replicates in response to nitrogen fertilizer rates
(1984/85 season).
Cultivars Nl=20 N2=40 N3=60 Mean
1. Nyungwe 2360 2512 2349 2407
2. Chololo 4130 4208 4088 4142
3. East Court A+ 3075 3850 3658 3528
4. Durban+ 2835 3520 3407 3254
5. Thornville 3483 3473 3267 3408
6. Ichinga 3725 4358 4228 4104
7. Muguga 2972 3465 3673 3370
8. Monzi No. 2+ 4760 4605 4675 4680
9. Baka 2917 2712 3623 3083
10. Phoka Court 3114 4052 3812 3659
11. No. 171 Star+ 3355 4228 4542 4042
12. Coast Gross 11+ 2470 3692 3642 3268
13. No. 2 Star 3380 3625 3397 3467
14. Misuku 4087 5012 5263 4787
15. No. 161 Star+ 3978 3777 3478 3744
16. Chilambula 3198 3893 4128 3740
17. Camper Down+ 3787 4883 4938 4536
18. No. 343 Star+ 4813 5053 5510 5126
19. S . E . Namwera 2695 2675 3220 2863
20. Thunda 3802 4627 3962 4130
21. Monsi No.l+ 2822 4067 3750 3546
22. Chinunkha 4358 3978 4552 4296
23. Meru 2292 2483 2592 2456
24. East Court B+ 4902 4660 5853 5138
Means 3471 3892 3984
S.E. of cultivar means +640*
S.E. of Nitrogen means +137*
S.E. of Interaction +67 1NS
Overall C.v. 38%
+ Introduced entries
Where * is significance at P=0.05
NS is not significant.
107
2 . Guinea grass
Most of the locally collected strains have a higher yielding
potential than the local standard cultivars, Ntchisi Panic and
Hamil Panic. Notable in this case are selection H, selection O
and Chololo Panic (Tables 5 and 6) . These are generally leafier
material which are generally of a very good regrowth potential,
much better than either Ntchisi or Hamil Panic. The least
producing were selections X and Makueni Panic, both of which are
low growing and finer-leaved types. Both these strains were slow
to establish a full canopy; as a result, they suffered
considerably from weed competition during the establishment. The
fact that they were smaller plant types could be explained by
their belonging to a lower ploidy level than the majority of the
more robust types.
Table 3. Forage DM yields (kg/ha) of Cynodon cultivars across
two replicates in response to cutting regrowth
intervals (1985/86 season) .
Cultivars 3 weeks 6 weeks 9 weeks Mean
1. Nyungwe 4430 6022 11883 7445
2. Chololo 4498 6354 12934 7929
3 - East Court A+ 6081 9536 13545 9654
4. Durban+ 5022 7751 11335 8036
5. Thornville 4068 7865 9606 7180
6. Ichinga 4463 6614 11891 7656
7. Muguga 4357 6376 10390 7041
8. Monzi No. 2+ 3857 7644 8555 6685
9. Baka 3817 7320 10510 7215
10. Phoka Court 4202 6356 8962 6507
11. No. 171 Star+ 3408 7797 10349 7184
12. Coast Gross II+ 6658 8345 11196 8733
13. No. 2 Star 5951 6969 11851 8257
14. Misuku 5246 5538 9540 6774
15. No. 161 Star+ 3634 6903 11189 7242
16. Chilambula 5259 5756 8258 6424
17. Camper Down+ 5091 6775 12341 8369
18. No. 343 Star+ 4397 9006 1 918 8440
19. S.E. Namwera 4762 6826 9064 6864
20. Thunda 5157 5620 10762 7186
21. Monsi No.l+ 3985 8073 9536 7198
22. Chinunkha 5757 6393 11623 7924
23. Meru 3610 6182 9279 6357
24. East Court B+ 4942 6053 11556 6357
Means 4694 7003 10782
S.E. Cultivar +623NS
S.E. Cutting intervals
+ Introduced entries +3 31
Where *** is significance at P=0.001
NS is not significant.
108
All strains significantly increased in their forage
productivity in response to increasing cutting regrowth
intervals. While the response to increasing nitrogen
fertilization turned out significant, there was a tendency for
same strains to respond less to increasing nitrogen fertilization
beyond N2 which is 4 0 kg N/ha. Notable in this regard are
selections C, M, J, K, D, H, B, Ntchisi Panic and Hamil Panic.
Table A. Forage DM yields (kg/ha) across of Cynodon cultivars
two replicates in response to nitrogen fertilization
(1985/86 season).
Cultivars Nl=20 N2=40 N3=60 Mean
kg/ha kg/ha kg/ha
1. Nyungwe 7448 7256 7631 7445
2. Chololo 7083 7894 8809 7929
3. East Court A+ 8822 9975 10164 9654
4. Durban+ 7396 7346 9364 8036
5. Thornville 7283 7584 6673 7180
6. Ichinga 7997 7162 7809 7656
7 . Muguga 6616 7386 7123 7041
8. Monzi No. 2+ 5780 6865 7410 6685
9. Baka 6774 8003 6869 7215
10. Phoka Court 6230 6703 6588 6507
11. No. 171 Star+ 6540 6729 8284 7184
12. Coast Gross 11+ 8695 8413 9091 8733
13. No. 2 Star 6880 8656 9235 8257
14. Misuku 7472 6453 6399 6774
15. No. 161 Star+ 7436 7454 6836 7242
16. Chilambula 6415 6058 6799 6424
17. Camper Down+ 7785 8578 8744 8369
18. No. 343 Star+ 8961 8326 8033 8440
19. S . E . Namwera 6593 6198 7863 6884
20. Thunda 6569 7513 7476 7186
21. Monsi No.l+ 6414 6924 8256 7198
22. Chinunkha 7216 8716 7840 7924
23. Meru 6171 5840 7060 6357
24. East Court B+ 6698 7899 7954 7517
Means 7136 7497 7846
S.E. of Cultivars +623NS
S.E. of Nitrogen levels +141***
+ Introduced entries
Where *** is significance at P=0.001
NS is not significant.
109
•tUBOTJTUflTS)OUSISN
0100.P=)BsouBotftnBtsst**te;teu.M
*¥*582+S3{oeupaloot+
SN591+sntBris-aS.
79391239133929SUB8M
•1 Oniocteles 0691 988H 1Oln1 67112
•2 +OTUBdTSTUOtN 9A68 Z.9211 82111 86410
•£ MonT^oeles 69Z.8 41812 95021 n1511
•n TU0T5j313S 7119 55715 39715 LLtZl
o5 +OTUBdlneriXBM 9854 9655 3594 1536
•6 Xonlcteles 6863 58n5 3549 4056
*L Xonxctsles 2818 82310 9036 3097
"8 aU0T3O3l3S 6508 45161 89051 28111
•9 Honxctelss n845l 42326 42631 15221
*ol aonlctsels 92514 15315 86823 70961
•11 Ooniq.oeles 13710 16512 46517 VZLZl
•2T IonTq.oeles 0989 58013 00914 64711
•E1 »OTUBdXTmBH 01281 0Z.0Z.1 18313 0185T
*f1 OTUBdolo1ottD Z.8281 22361 66521 59214
•fit aonlcteles 55710 45261 6Z.121 95511
UB3M
BU./6i{
08=3N
BU./6i[
0»=2N0=1NSUTBJtS
•BX8A8X
noxtBZTXTtragneBoatjTU03esnodserntsatBoxxdar
OH^ssor0Ksutsi;soiucTnBdjo(bxi/£x)spxej^NQs5saoj'9eiqci
'010'0P=3oBeOUBOlJTUfeiSST***
X0'0P=taeonBDTfTnfiTSst**eertTM
s2(oeuolBooq+
»**582+slBAJteUT6nT))uoof'aS.
*»4951+sntBrtbofogS.
SUTBJtS S^een3 seken6 seken9 UB3M
•1 Sonftceles 3999 9999 1182X 61271
•z +OTUBdTSTupq.N 76X9 8488 4334X 86nO1
•£ MonTcteles 2209 LOZL 313i1 41511
j4 ronTcteles 1070 330XX 98322 LLVZl
•s +OTUBdTUerU[BM 5284 6526 5848 X536
•6 XonTcteles 0Z.5to 7066 8696 5406
•L XonTcteles 8947 98to9 16711 £097
•8 aonTcteles 096Z. 7289 37641 83X11
"9 HonTq.oeles £7215 8336X 3in23 X52X2
•10 Oonicteles 14110 51018 15022 Z.0961
*TT Oonicteles 6966 9540X X8902 \ZLZ\
*12 IonTcteles 885Z. 49210 82Z.61 V9LXl
"31 *OTUBdXTmBH 25910 260H 76272 10815
•n OTUBdololonC 9239 60n17 3908X 592n1
•51 aonicteles 1948 6Z.010 59615 955XX
SUB8M nX86 820XX 97515
•slvAaetnTfinxttncotesnodserntsetBOilder
on^ssocresntausmnoln»djo(sq/53()spxeiAMQbCsioj•SeX>TBI
o11
CONCLUSION
While the nitrogen fertilization regimes employed in this study
were not aimed at coming up with a complete nitrogen response
profile recommendation, it was important as a selection tool in
demonstrating the necessity of considering low fertilizer input
management. In the early evaluation stages it is important to
appreciate materials that may not positively respond to high
fertilization management as these have potential for eventual use
in smallholder farming systems where inorganic nitrogen
fertilizers may be lacking due to high cost.
The results show that the standard cultivars of star grass
namely No. 2 star and Muguga and the three standard guinea grass
cultivars, Ntchisi Panic, Hamil Panic and Makueni could be
replaced by some more superior locally selected strains.
Phoka Court and Misuku star grass strains which have not
demonstrated a response to high fertilization management were
collected from sites characterised by acid infertile soils. The
lack of positive response in these two and other similar strains
could be an adaptation to these poor growing conditions. This
potential ought to be exploited in greater detail.
This argument holds true also for the guinea grass strains
that showed this potential.
Panicum maximum strains have shown variability that warrants
selection to replace Ntchisi, Hamil and Makueni Panics by the
more robust, leafier ecotypes.
REFERENCES
Clayton, W.W. and Harlan, J.R. 1970. The genus Cynodon L.C. Rich
in tropical Africa. Kew Bull. 24:185-189.
Harlan, J. R. 1970. Cynodon species for grazing and hay. Herb.
Abstr. 40(3) :233-238.
Harlan, J .R. and de Wet, J.M.J. 1969. Sources of variation in
Cynodon dactylon (L.) pers. Crop Sci. 9:774-778.
1ll
EVALUATION OF DESMODIUM ADSCENDENS THROUGH VEGETATIVE
PROPAGATION IN NATURAL AND PLANTED PASTURES
E.N. Sabiiti
Crop Science Department, Faculty of Agriculture
and Forestry, Makerere University
P.O. Box 7062, Kampala, Uganda
P.R. Henderlong
Agronomy Department, Ohio State University
Columbus, Ohio 43210, U.S.A.
J.S. Mugerwa
Animal Science Department, Faculty of Agriculture
and Forestry, Makerere University
P.O. Box 7062, Kampala, Uganda
Abstract
The natural grasslands of Uganda comprise about 80% of the total
land area and are regarded as a national asset because they
support nearly 99% of Uganda's livestock and game animals. They
contain numerous indigenous forage species which have not been
developed to increase the productivity of these grasslands which
is below their potential.
The objective of the study was to evaluate a local forage
legume (Desmodium adscendens [Sw.] DC) when planted vegetatively
in ley and natural pastures.
The results indicated that D. adscendens established well
when vegetatively planted in ley and natural pastures. However,
it was more successful in ley than in natural pastures. The
legume has a high potential for dry-matter yield, % crude protein
and N2 fixation. D. adscendens exhibited remarkable persistence.
It is, thus, suggested that more detailed studies using an
interdisciplinary approach be conducted to determine the
contribution of the legume to pasture productivity, animal
production, soil fertility and subsequent crop yield.
INTRODUCTION
The natural grasslands of Uganda comprise about 80% of the total
land area (Horrell and Tiley, 1970) and are regarded as a
national asset because they support nearly 99% of Uganda's
livestock and game animals. Unfortunately, the productivity of
these grasslands in terms of herbage quality and quantity and
animal production is extremely below their potential. For
example, estimated milk production per cow ranges from 400-600
litres in a six- to seven-month lactation period, and this is
considerably below that of cows grazing improved grass/legume
pastures. Even the ley pastures which constitute a small
proportion of the natural pastures are below their potential.
Their potential could be increased by incorporating forage
legumes such as Stylosanthes gracilis, Desmodium intortum, D.
112
uncinatum, Centrosema pubescens, Macroptilium atropurpureum,
Medicago sativa and Indigofera hirsuta (Stobbs, 1969a, b; Otim,
1973; Horrell and Court, 1965; Mugerwa, unpublished; Sabiiti,
1979; UNDP/FAO, 1973).
However, one of the major constraints in the improvement of
pastures has long been the consistent lack of pasture seeds (both
legumes and grasses) . Nearly all the pasture seeds are imported
and are very expensive (about 5,000/= per kg of legume seed) and
are still scarce in the country. The situation was further
aggravated by political instability in the last 15 years or so
and nearly all the pasture germplasm is confined to research
stations or has disintegrated as at Kabanyolo Makerere University
Farm. Furthermore, there is evidence to show that most of the
recommended forage legumes fail to combine both productivity and
persistence for a long time under various pasture management
practices and ecological zones (Sabiiti, 1979) .
To overcome some of the above constraints, it is important
to evaluate local pasture species, especially legumes before they
are extinct under the current rapid environmental degradation in
the country. The need to develop local pasture species has long
been emphasised by Henderlong (197 3) and others.
One of the forage legumes which we have identified is
Desmodium adscendens (Sw.) DC. It has been observed to grow well
in the Lake Victoria crescent region of high rainfall (1000-1500
mm/annum) . The crescent region is important for dairy farming
but there is scarcity of legumes on these farms. The legume
appears to be grazed by livestock and game animals, especially
during the dry season when the grasses are very dry. It roots at
the nodes, and it could be established vegetatively to reduce the
dependence on importation of expensive seeds. This method has
been used successfully in Kenya to establish D. intortum and D.
uncinatum (exotic species) by Keya et al . (1971). However, there
is no data on this aspect. The legume makes excellent fodder and
could be used as a soil cover and green manure. In Indo-China,
Malaysia, East and South Africa, it is planted in tea and coffee
plantations as a cover and for green manure.
However, no work has been done on this legume here to
determine its contribution to pasture productivity. The
objectives of the study were to determine a method of
establishment, assess dry-matter yield, feeding value (CP) and
persistence of D. adscendens under various management practices.
MATERIALS AND METHODS
Study Area Description
The first study (under planted pastures) was conducted at
Kabanyolo Makerere University Farm. Kabanyolo is in the fertile
lake crescent area, 19 km north of Kampala (O°28'N, 32°37'E,
1,200 m.a.s.l.). The upland soils of Kabanyolo are classified as
Latosols, or ferrallitic soils. They are deep, highly drained
red soils. The pH varies from 5.0-5.8 (Mohiddin and Mukiibi,
unpublished; Reid et al . , 1973) with a very low level of
113
phosphorus (4 ppm P Truog) , medium potassium (18 mg/100 g) and
ample organic matter (3.1%) (Olsen and Moe, 1971).
The area has a moist tropical climate with mean maximum
temperatures varying from 28.5°C in January to 26.0°C in July and
minimum temperatures from 17.4°C in April to 15.9°C in July and
August. Mean annual rainfall is about 1,300 mm, with two peaks
in April and November and two periods of low rainfall in January
and July when the monthly mean drops to 60 mm.
The second study (under natural pasture conditions) was
conducted at Buwambo, 24 km north of Kampala (0°35'N, 32°34'E,
1300 m.a.s.l.). The soils and climate of the area are similar to
those of Kabanyolo. The vegetation is dominated by a natural low
medium tropical deciduous forest and abundant elephant grass (P.
purpureum) in areas that have been cleared. Other important
vegetation include perennial shrubs, native pasture legumes; D.
adscendens being the dominant one followed by Neonotonia wightii
and grass species Cymbopogon, Imperata, Digitaria, Panicum and
Brachiaria. The area is commonly grazed by cows, goats and
sheep.
METHODOLOGY
Study I: Evaluation of D. adscendens under planted pastures
Mature stem cuttings (15 cm long) were planted at a spacing of
0.5 x 0.5 m into an established pasture of Chloris gayana Kunth.,
Cynodon dactylon Linn., and Brachiaria brizantha at Kabanyolo
University Farm. The stem cuttings were collected from Buwambo,
24 km north of Kampala where the legume grows naturally in a
mixture of P. purpureum, Brachiaria spp. , Hyparrhenia spp.,
Jmperata cylindrica and Panicum spp. They were planted into
holes of about 7 cm deep and firmed with the soil. The pasture
paddock was divided into four plots (10 x 10 m) each and were
excluded from grazing for ten months. Planting was done during
the long rainy season in March in 1979 to ensure maximum
establishment. Before planting, single superphosphate at the
rate of 250 kg/ha was broadcast onto the plots to promote legume
root growth (Olsen and Moe, 1971) .
Establishment
Observations on the establishment of the legume were made after
three and six months from planting because after that the senior
author left the University. The total number of cuttings which
had sprouted were determined.
Dry-Matter Yield, Crude Protein and Persistence
These parameters were determined in 1986 following the senior
author's appointment at the University. However, the paddock had
been grazed several times in the past but no observations were
made. Conventional methods for herbage sampling of the old plots
and dry-matter estimation were used. Crude protein content was
determined by the Marcro-Kjeldhal method (AOAC, 1965) .
114
Study II: Evaluation of D. adscendens under natural pastures
Four plots (10 x 10 m) were randomly located at Buwambo, where
the legume grows naturally. Mature stem cuttings of 15 cm long
were planted onto a natural pasture dominated by B. brizantha, I.
cylindrica, H. rufa and P. maximum. Any existing legume was
removed before planting. The cuttings (one cutting/hole) at a
spacing of 0.5 x 0.5 m were planted in March 1979 during the
rainy season to ensure maximum growth. There was no application
of single superphosphate fertilizers.
Establishment
Observations on establishment of the legume were made after three
and six months from planting because of similar reasons as in the
above.
Dry-Matter Yield, % CP and Persistence
These were determined in 1986 from the old plots. Similar
sampling methods and dry-matter estimation and crude protein
content analysis were used as above. A ' t' test was used to
compare the treatment means of the two types of pastures as
outlined by Zar (1984) .
RESULTS
Establishment of D. adscendens in Ley and Natural Pastures
The results in Table 1 show that D. adscendens was successfully
established through vegetative propagation under both pastures
(ley and natural); however, a significantly (P<0.05) higher
percentage of establishment was obtained under ley pastures.
This is expected because of the excellent seedbed conditions
obtained in this method.
The cuttings started producing shoots after three weeks from
planting and by the fourth month most of the plants were fully
established with many leafy shoots (5-10 shoots/plant) . Growth
was vigorous but especially under ley pastures. Unfortunately,
there were no further observations on the performance of these
plants after 1979 because the senior author left Makerere
University after completing his M.Sc. research programme.
Persistence
The legume has persisted very well both at Kabanyolo and in
Buwambo under grazing, nearly eight years from planting. D.
adscendens is more abundant in nearly all the paddocks at
Kabanyolo than the introduced forage legumes mentioned in the
introduction. The same observation is true under natural
pastures. The legume has spread over large areas where it was
not abundant before. Also the legume nodulates profusely without
inoculation, especially under ley pastures, so it could be fixing
N, as most of the nodules exhibited red pigment when cut open.
Tne legume is prostrate and this makes it better adapted to
grazing conditions.
115
Table 1. Total number and percentage of vegetatively established
plants of D. adscendens in ley pastures at Kabanyolo
Makerere University Farm (A) and at Buwambo (B) in
1979.
Site Plot Number of %
established plants Establishment
A C 360 90.0
D 360 90.0
E 350 87.5
F 370 92.5
Mean 3 60+8 90.0+2
B G 265 66.3
H 258 64.5
I 300 75.0
J 290 72.5
Mean 278.3+20 69.6+5
<t' - test P<0.05, ±SD, n=4.
Dry-Matter Yield and % Crude Protein
It is clear from Table 2 that D. adscendens has a high potential
of productivity in terms of DM yield (2830 and 1620 kg/ha) and %
CP (22.5 and 17.8) in ley and natural pastures, respectively.
Table 2. Total DM yield (kg/m2) and % crude protein of D.
adscendens collected from ley pastures at Kabanyolo
Makerere University Farm (A) and from natural pastures
at Buwambo (B) in 1986.
DM yield
kg/m2Site Plot % CP
A l .218 21.5
2 .258 21.8
3 .357 24.3
4 .300 22.4
Mean .283+0. 06 22.5+1.3
B 1 .146 15.8
2 .178 18.5
3 .190 19.4
4 .135 17.6
Mean .162+0. 03 17.8+1.5
test P<0.05; +SD, n=4 .
116
The plants under ley pastures produced significantly (P<0.05)
higher DM yield and % CP. These are two of the major indices for
evaluating a new cultivar.
The plants under natural conditions mainly suffered from
severe competition from shrubs, trees and tall grasses of
Hyparrhenia spp. , P. purpureum and I. cylindrica. Management
under natural pastures would require removal of tall species and
topping of the existing vegetation to reduce shading and
competition.
DISCUSSION
The genus Desmodium has about 2 00 species (perennial or annual)
which occur in temperate and tropical regions where they are used
for grazing (Whyte et al . , 1953). Unfortunately, only two
species (D. intortum and D. uncinatum) have been studied in
detail in Uganda (Mugerwa, unpublished; Otim, 1973; Olsen, 1973;
Reid et al., 1973; Olsen and Moe, 1971; Ochodomuge, 1978). This
is also true for other countries (see Kategile, 1984; Anon,
1985) . Even then the success of these two legumes in pastures in
Uganda has been limited by lack of sufficient pasture seed and
other limitations which have been reviewed by Sabiiti (1979).
Therefore, the results we have reported on D. adscendens are
promising in that we can reduce dependence on imported seeds of
exotic species and rely on a local legume species that
establishes well by vegetative propagation and has a high
potential for DM yield and % CP. It has shown admirable
persistence in both ley and natural pastures under grazing
conditions. However, we do not know the stocking rate and
frequency of grazing that were used because no records were made.
The legume appears to be dispersed by animals through either the
seeds which stick on their skins and are dropped or by the broken
stems which get trapped between the hoof and are later dropped in
other paddocks. Also humans could have assisted in the spread of
the legume on the farm.
The profuse natural nodulation of the legume might be
important in increasing pasture productivity, soil fertility and
subsequent crop yields. Currently, we have no published data on
D. adscendens in East Africa to compare with although Keya et al.
(1971) have shown that other Desmodium species were easily
established from root splits (40-80%) . In South America, Lazier
unpublished did not clearly indicate how the legume was
established. In Taiwan, Byran (1969) successfully established D.
intortum and D. uncinatum by planting individual splits.
Since D. adscendens occurs abundantly in the regions that
are important for dairy industry in Uganda, farmers should be
encouraged to collect and plant the legume in their paddocks and
then apply single superphosphate to encourage root formation and
growth. The early establishment and higher survival rate in ley
pastures were partly attributed to single superphosphate. Once
the plants have established, they form several prostrate shoots
which grow rapidly, and these can again be used to plant in other
paddocks.
117
Future Research
Our present data are not yet conclusive and since there is little
information on this important legume, there is a need to
investigate further other indices of evaluation, namely, response
to defoliation (mechanical or biological), regrowth physiology to
understand mechanisms of persistence, nutrient analyses,
digestibility studies, grazing trials, (voluntary intake) and
animal production (live weight gain or milk production) , N2
fixation and soil fertility and crop production from pastures
previously planted to D. adscendens .
Our current evaluation programme of D. adscendens at the
University Farm is geared towards the above studies using an
interdisciplinary approach. We have got soil scientists,
microbiologists, animal nutritionists, agronomists and
physiologists because we believe we can achieve great success in
a short time.
ACKNOWLEDGEMENTS
We wish to acknowledge Dr. B. Dzowela (PANESA Coordinator) for
inviting the senior author to participate in the PANESA Workshop.
Professor J.L. Parsons, USAID Project Team Leader in the Faculty
of Agriculture and Forestry, is thanked for securing a USAID
grant to the senior author to travel to Kenya where he met Dr.
Dzowela. The Secretary, Sam Mubiru of Crop Science Department,
is thanked for typing the manuscript.
REFERENCES
Anon. 1985. Proceedings of the XV Intern. Grassland Cong.
August 24-31, 1985, Kyoto, Japan. The Science Council of
Japan, The Japanese Society of Grassland Science.
AOAC. 1965. Official methods of analysis (10th ed.). Association
of Official Analytical Chemists, Washington, D.C.
Byran, W.W. 1969. Desmodium intortum and Desmodium uncinatum.
Herbage Abstracts 39:183-191.
Henderlong, P.R. 1973. Pasture research in Uganda: Past,
present and future. In: P. Rubalhayo (ed.), Proc. 1st
Symposium of the Uganda Society of Agronomy. Makarere
University, Kampala. pp. 30-47.
Horrell, C.R. and Court, M.N. 1965. Effect of the legume
Stylosanthes gracilis on pasture yields at Serere, Uganda.
J. Brit. Grassld. Soc. 20:72.
Horrell, C.R. and Tiley, G.E.D. 1970. Grassland. In: J.D.
Jameson(ed. ) , Agriculture in Uganda. Oxford University
Press, London. pp. 318-332.
Kategile, J. A. 1984. Pasture improvement research in Eastern and
southern Africa. Proc. of a workshop held in Harare,
Zimbabwe, 17-21 September, 1984. Ottawa, Canada.
118
Keya, N.C.O., Olsen, F.J. and Holliday, R. 1971. The
establishment of desirable grass/legume mixtures in
unimproved grassland by planting root splits. E. Afr.
Agric. For. J. 38:220-223.
Ochodomuge, P.E. 1978. Effect of advancing maturity on the
productivity and feeding value of Desmodium intortum,
Macroptilium atropurpureum and Stylosanthes gracilis in
Uganda. M.Sc. thesis, Makerere University, Uganda.
Olsen, F.J. 1973. Effects of cutting management on a Desmodium
intortum (Mill.) Urb./Setaria sphacelata mixture. Agron. J.
65:714-716.
Olsen, F.J. and Moe, P.G. 1971. The effect of phosphate and lime
on the establishment, productivity, nodulation and
persistence of Desmodium intortum, Medicago sativa and
Stylosanthes gracilis. E. Afr. Agr. For. J. 37:29-37.
Otim, J.N. 1973. The potential of improved pastures for
fattening beef steers in Uganda. M.Sc. thesis, Makerere
University, Uganda.
Reid, R.J., Post, A.J., Olsen, F.J. and Mugerwa, J.S. 1973.
Studies on the nutritive quality of grasses and legumes in
Uganda. Trop. Agric. (Trin.) 50:1-15.
Sabiiti, E.N. 1979. The potential of Indigofera hirsuta L. as a
pasture legume in Uganda. M.Sc. thesis, Makerere
University, Uganda.
Stobbs, T.H. 1969a. The value of Centrosema pubescens (Benth.)
for increasing animal production and improving soil
fertility in northern Uganda. E. Afr. Agric. For. J.
35:197-202.
Stobbs, T.H. 1969b. Animal Production from Hyparrhenia grassland
oversown with Stylosanthes gracilis . E. Afr. Agric. For. J.
35:128-202.
UNDP/FAO. 1973. The value of oversown legumes in northern Uganda
pasture with special reference to Stylosanthes gracilis.
UNDP/FAO Working Paper No. 5 AGH.SF/UGA.8 .
Whyte, R.O. , Nilsson-Leissner , G. and Trumble, H.C. 1953. Legumes
in agriculture. FAO publication, Rome.
Zar, J.H. 1984. Biostatistical analysis. 2nd Edition.
Prentice-Hall Inc., Englewood Cliffs, N.J., U.S.A.
119
THE SCREENING OF PERENNIAL TRIFOLIUM SPECIES
MAINLY FROM THE ETHIOPIAN HIGHLANDS AND THEIR
POTENTIAL FOR USE IN PASTURE
J. Kahurananga
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
Abstract
Spaced plants of 98 accessions of perennial Trifolium spp. mainly
from the Ethiopian highlands were planted at Shola, Addis Ababa,
Ethiopia in June 1985 for initial screening. The objective was
to select those suitable for use on Vertisols which are subject
to waterlogging in highland pastures with an extended dry season
and periodic frosts. Total annual rainfall for 1985 and 1986 was
1050 mm and 1198 mm respectively. There were more frosts in
1985. T. cryptopodium followed by T. burchellianum had the
highest DM production and persistence during the dry season, but
the latter had poor flowering. T. semipilosum including cv
Safari had lower production but very good flowering. T. repens
included as a control had fair performance. None of the
accessions were affected by frost. Results indicate that these
clovers have potential for use in pastures particularly on fallow
areas and for soil conservation. Nine accessions were selected
for further evaluation in mixture with grass under cutting or
grazing.
INTRODUCTION
The highlands of Eastern Africa, in this context land above 1800
m, cover 70,000 km of which 60% is in Ethiopia. They contain
the highest human and livestock densities and are the most
intensively cultivated (Getahun, 1978; Morgan, 1982; Jahnke and
Asemenew, 1983) . The major function of cattle in the Ethiopian
highlands is traction. In the highlands of East Africa, cattle
are primarily kept for milk. In fact the largest proportion of
dairy production in Africa is in these highlands with Kenya
leading (Mukasa-Mugerwa, 1980). The highlands therefore
represent ideal smallholder mixed farming systems and have great
potential for livestock production.
Unfortunately, the highlands are plagued by the same
constraints as elsewhere in Africa of inadequate livestock feed
particularly during the dry season. Natural pastures provide
about 65% of the feed resources, and the rest are provided by
crop residues (Jahnke and Asemenew, 1983). So one way of
increasing livestock feed would be through pasture improvement .
However, periodic frosts, extended dry periods, high acidity,
extensive areas of Vertisols and Nitosols, each with its own
peculiarities present special problems to legume introduction.
On the other hand, highland pastures are rich in legume
composition particularly the Trifolium spp. (Gillett, 1952;
Gillett et al., 1971; Mengistu, 1975; Thulin, 1983). Evaluation
of annual Ethiopian clovers has shown that they have good
potential for hay production (Kahurananga and Tsehay, 1984). In
1985 work was started in screening perennial clovers as research
120
in Australia has shown that some of them have high potential for
improving pasture production (Mannetje, 1964; Jones and Cook,
1981) .
METHODS
The screening was conducted at ILCA Headguarters, Shola, Addis
Ababa, Ethiopia (9° 2'N, 38° 42 'E) at an altitude of 2380 m on a
gently sloping site with seasonally waterlogged black clay soil
typical of the bottomlands of the Ethiopian highlands
traditionally used for hay production and grazing. The soil is
slightly acidic (pH 5.8) and very deficient in P(8.68 ppm) .
Ten seedlings, each seven to nine weeks old of 98 accessions
of Trifolium africanum (2), T. burchellianum (23), T.
cryptopodium (13), T. semipilosum (57), including cv Safari for
comparison together with T. repens (1) (control) originally from
Ethiopia (83), Kenya (11), Tanzania (1) and southern Africa (3)
were transplanted into a well cultivated field at 1 m intervals
on 24 and 25 June, 1985. TSP was applied at a rate of 40 kg P
ha . Two handfuls each of sheep manure were put in the holes
and mixed with soil into which seedlings were transplanted.
Observations were taken once every three weeks and included
vigour, cover, stolon thickness, internode length, petiole
length, middle leaflet surface area, inflorescences per plant,
flowers per inflorescence and peduncle length. The plants were
harvested in May 1986 and again in September 1986 and the
material dried in an oven at 65 C for 24 hours for estimation of
dry matter.
Daily weather records were available at the ILCA Shola
weather station nearby.
RESULTS
The climatic data is presented in Table 1. The total annual
rainfall for 1985 and 1986 was 1058 mm and 1198 mm respectively
with a better spread in the later part of year (Showamare, 1986) .
Average maximum and minimum temperatures for both years were
similar but there were more frost nights in 1985 even during the
rainy season which stunted annual clovers in nearby plots.
The observations from the 980 plants are still being entered
into the computer for cluster analysis in order to group plants
according to their growth characteristics. Data so far recorded
for the most important observations for each species are
summarised in Table 2. The results show that T. cryptopodium had
the highest DM production followed by T. burchellianum var.
johnstonii . All varieties of T. semipilosum had lower DM with
var. semipilosum, glabrescens , including cv Safari, intermedium
and brunellii in that order. T. repens, T. africanum and T.
burchellianum var. blongum had the lowest DM production. Again
T. cryptopodium and T. burchellianum had the best vigour or green
leaf retention during the dry season. T. semipilosum varieties
shed most of their leaves during the dry season and those which
remained turned copper red.
1 2 1
(98-)eJiUiiftoqs:a--- -
---
■t:
98-
-T
-
S'L 6- Z- S-
- -fr -9
6-
9-T
i-
ti.i-aj-u--
98- L'6 O6 0"8
-fr -L -8
A-E
- 9O 6O
-O
O6
JO-
98- TZ O O O O O O O T -
-
01
-
ON
98-
9T
S S T Z Z O O O T
9T O -
--ST
9-T
O-- OT
-T
O9
-T
O- O9
-Z -2-
OZ-
--
O--
-e-fi-Ohav
D„3--q--J3duia^
98- g-g-
-Z-
O--
0"Z-
-'-
O'-
-O -O
OE
9O-
OE-
-E-
O-»PUi
98-
-9
--
-OT
-- --
8OT 6OT
fr'O
-6
T-
-L 9O
98-JO
98-
-8
EO OOT
O- O-
9-T
fr'O AOT
9OT *"6 -9
TO
Jiion-^i
XiW
98- O"EZ
-EZ -EZ
-O
A-EZ
-- -O
LOZ
OTZ
O"O
-O
6"ZZ
--Y
98- O"EZ
OEZ
-frZ
C-
0"EZ
-O
E-- i--
O- 9mTZ
T'ZZ
OZZ
-aq-odi-a (
98-
OT
-T
8-
OT
»9T
OT O
T-
-T
LLT 9-
L'T
-----a
mm)
98- 8- 9- L- O9T
OX
6--T
68 Z8 EET
9iT 9LT 9AT
Oq-ilUTTD 98- O frfr
O
LTZ
- OZ
89T
OO frZT LZ O O
8-T-iJUT-H (ana
)
98-
O
0 frZ O-
-T
-6 OZ OZ 8ET
LZ O 0
6-T
siq-O
--ir
-qaj
--eO
-adv AiW
--r
-i-r
---v
}d--
-^-o
--ON
-o-a
TE^OI
o]oaLj%
jT.
JSAOfJ
(jL-0o)j8
(oo-o)EZ
(j)j
(O-jO)6j
(oe)oe (jo)0o
(OT-O6)O
(jo)jo
(O:-j*)88
q:,30'*oAaq
6cTJSMOX3
(L0ZjLX)Z6j (ZO-Zj)oj
(6ET)6LT
(EEZ-Tj)Zj
(Ej)L8T
(jZ)lZZ
(Oj-OT)8O
(8ZZ-Z.j)OZ
(TEZ-8O)ojZ
^q*Te/iHa ^.j£TjJjj
(«6)
(OT,Z)j
(Z»E-TT)ZO
(XL-Xe)6O (jo-jr)frj
(jO-j)T8
(OT-W8s
(jZ-o)To
(jZT-oo)L8 (ZZZ-TZ)LH
OSajSAjq
jnoOTA
(j-T)
(fr-E)*
(6-fr)8 (o)o (8-Z)8 (o)o (T)T
U-T, (E-Z)E (L-T)t
30jeqvurij
SOTSS3DOa
E EZ X
O
X X j_ j e» O
AqbTJaApoaosToe>
ttoo^oouoCUaA
umoEf_Leuojnq us»6oo_qo-jaamnoaTLsuojeq TT.Tjjn.iqjaaA
asaosejrei'oj a
umoo_TjTuieo
omsoTTjTineo'J8A
umso_TjTniBo
Xa^OI
nmoaoTjja
unvfpoo^jAjo
ooejej
rnnooLfjTmeo umsoojfuieo
•I •L
"I
rJi •1 r1 •I, •I •I
•8O1Pja5 OjTaTOj.S}a
lunTTOjTJj,Xatoosaej,oooTq.EAsboqo^ e^ o u^q.30s6 eae EOE BAY*Z8T «I
The positions were reversed regarding flowering. T.
semipilosum had profuse flowering with varieties semipilosum and
glabrescens excelling. T. africanum had good flowering and T.
cryptopodium was average. The one accession of T. repens had
rather fair flowering. Both varieties of T. burchellianum had
the worst flowering with some accessions not flowering at all.
None of the accessions were affected by frost.
Besides the above characteristics, all the clovers showed
good soil binding characteristics, demonstrated by the fact that
most of the soil cracking during the dry season occurred mostly
on spots where there was no clover growth. T. cryptopodium with
its numerous matt-forming rhizomatous stems was most outstanding
in this respect.
DISCUSSION
This is the first time that such a large collection of perennial
Trifolium species from Ethiopia have been screened. It was
rather surprising that the smallish often overlooked T.
cryptopodium being evaluated for the first time surpassed the
others in DM. This is partly due to its dense matt of stems with
short internodes. It grows well in natural pastures with grasses
such as Andropogon abyssinica, A. amethystinus, A. distachyos,
Pennisetum thunbergii and P. clandestinum or Kikuyu grass. At
one time it was briefly referred to as forming a valuable close
sward under heavy grazing in association with Digitaria scalarum
on Mt. Elgon (Gillett, 1952) . The prostrate growth habit and its
small leaves and short internodes makes it ideal for sheep
grazing at high altitudes from about 2600 m - 3000 m. The matt-
forming growth habit which binds soil together also makes it
ideal for soil conservation. Its only problem is that its pods
shatter easily after ripening. There is therefore need to study
the seed ripening characteristics in order to devise efficient
means of harvesting.
The other species which showed good productivity and which
persisted well during the dry season was T. burchellianum var.
johnstonii. Its ability during the dry season was also noted in
early observations in Kenya and Australia (Bogdan, 1956; Britten,
1962; Mannetje, 1964). This characteristic is rather surprising
as this species is normally distributed in moist grasslands and
open glades in forests where rainfall is above 1000 mm (Gillett
et al. , 1971; Thulin, 1983). This species is found growing with
T. cryptopodium. Its major limitation is the poor flowering and
hence seed production. This was also observed in one of the
early glasshouse experiments where it failed to flower altogether
(Mannetje and Pritchard, 1968) . However, in this trial there
were two accessions, one from Ethiopia and the other from
Tanzania which had good flowering. There are therefore
possibilities for selection. This species has good potential for
use in pasture for both cattle and sheep. It is also suitable
for soil conservation.
Although the average productivity of T. semipilosum was
lower than the two preceeding species, there were some accessions
which did well. It is rather surprising that var. glabrescens
including cv Safari faired poorly. The latter is doing well in
124
subtropical areas of Australia (Jones and Cook, 1981) . Although
most of the accessions lost their leaves and those leaves which
remained turned copper red during the dry season, they responded
very fast to the onset of rain. This species had excellent
flowering, and the pods do not shatter so easily, making it more
suitable for seed production. It also grows well naturally with
grasses in pasture, and it is found in a very wide range of
altitudes, edaphic conditions for both cattle and sheep and for
soil conservation. It seems to withstand very heavy grazing in
communal natural pasture. The ability to persist under grazing
has also been noted under Australian conditions (Sproule et al.,
1983) .
This screening trial has shown that the perennial Trifolium
species have good potential for use in pasture. They could be
used in establishing new pastures or on fallow areas including
marginal areas. They could also be useful in reseeding degraded
areas. It is important to continue making selections as
evaluations continue. In this screening, four accessions of T.
cryptopodium (ILCA Nos.7621, 9391, 9398, 9461), three of T.
burchellianum var. johnstonii (ILCA Nos. 10179, 10207, 12481) and
of T. semipilosum var. semipilosum (ILCA No. 7642, 8025) were
selected for the next stage of evaluation in mixture with grasses
under cutting management and grazing.
ACKNOWLEDGEMENTS
I am grateful to ILCA field staff who have assisted me in the
screening and to Dr. J.C. Tothill for giving useful guidelines.
Also I thank ILCA's Director of Research, Dr. K. Peters and the
Director General, Dr. J. Walsh, for allowing me to attend the
workshop and present this paper.
REFERENCES
Bogdan, A.V. 1956. Indigenous clovers of Kenya. East African
Agricultural Journal 22:40-45.
Britten, E.J. 1962. Evaluation for pasture purposes of some
African clovers in a plant introduction program. Journal of
Range Management 15:329-333.
Getahun, A. 1978. Zonation of the highlands of tropical Africa.
ILCA, Addis Ababa, Ethiopia.
Gillett, J.B. 1952. The genus Trifolium in southern Arabia and
in Africa south of the Sahara. Kew Bulletin 7:367-404.
Gillett, J.B., Polhill, R.M. and Verdcourt, B. 1971. Leguminosae
(Part 4) Subfamily Papilionoidea (2) . In: E. Milne-Redhead
and R.M. Polhill (eds) , Flora of tropical East Africa. Crown
Agents, London. pp. 1016-103 6.
Jahnke, H.E. and Asemenew, G. 1983. Ethiopian highlands
reclamation study. An assessment of the recent past and
present livestock situation in the Ethiopian highlands.
125
Land Use Planning and Regulatory Department, Ministry of
Agriculture and Food and Agriculture Organization. Addis
Ababa, UTF/ETH/037/ETH Working Paper 8.
Jones, R.M. and Cook, B.G. 1981. Agronomy of Kenya white clover-
cultivar. Safari Information Services sheet No. 41-3, CSIRO
Division of Tropical Crops and Pastures.
Kahurananga, J. and Tsehay, A. 1984. Preliminary assessment of
some annual Ethiopian Trifolium spp. for hay production.
Tropical Grasslands 18:215-217.
Mannetje, L.T. 1964. The use of some African clovers as pasture
legumes in Queensland. Australian Journal of Experimental
Agriculture and Animal Husbandry 4:22-25.
Mannetje, L.T. and Pritchard. 1968. The effects of photoperiod
and night temperature on flowering and growth in some
African Trifolium species, flew Phytologist 67:257-263.
Mengistu, A. 1975. Grassland condition in the Chilalo Awraja,
Arussi Province, Ethiopia. M.Sc. thesis, Department of
Plant Husbandry, Agricultural College of Sweden, Uppsala.
Morgan, W.T. 1982. East Africa. Longmans, London. pp. 290-331.
Mukasa-Mugerwa, E. 1980. Small holder dairy development in the
highlands of tropical Africa. ILCA, Addis Ababa, Ethiopia.
Sproule, R.J., Shelton, H.M. and Jones, R.M. 1983. Effect of
summer grazing on growth habit of Kenya white clover
(Trifolium semipilosum cv Safari in a mixed sward) .
Tropical Grasslands 17:25-30.
Showamare, M. 1986. Climatic records for ILCA research sites
1985. ILCA, Addis Ababa, Ethiopia.
Thulin, M. 1983. Leguminosae in Ethiopia. Opera Botanica
68:92-223.
126
YIELD AND QUALITY OF LASLAB PURPUREUS ECOTYPES
IN HIGH ALTITUDE SEMI-ARID LANDS IN KENYA
m. Karachi
National Dryland Farming Research Station
P.O. Box 340, Machakos, Kenya
Abstract
Dry-matter production, forage nutritive value and seed yield of
Lablab purpureus ecotypes were studied over a two year period.
Total and green leaf DM production ranged from 2000 kg to 12000
kg ha~ and 400 kg to 3300 ha~ respectively. The leaf/stem
ratio declined by 3.7 to 4.5 times with maturity. In vitro dry-
matter digestibility and protein content of green leaf and stem
declined by 10% to 15% units and 3% to 5% units respectively.
Phosphorus content was notaffected by forage maturity. Seed
yields of 400 to 1800 Jcg ha with crude protein content of 24%
to 29% (DM basis) was recorded. It was concluded that the DM
yield, forage nutritive value and seed yield make Lablab an
attractive proposition in livestock/crop farming systems.
INTRODUCTION
Over 200 genotypes of Lablab purpureus are recognised
(Pulseglove, 1968) . This germplasm has been exploited as a
source of human food (Skerman, 1977) , forage for livestock
(Hamilton et al., 1970; Hendricksen and Myles, 1980) and as green
manure or cover crop (Pulseglove, 1968) .
In Kenya, Lablab finds its greatest use in the small-scale
sector mainly as a source of grain for human consumption.
However, the current trends in farming emphasise the use of
multipurpose crops that fit into integrated livestock/crop
farming patterns. The investigations reported examined the
potential contribution by Lablab in livestock/crop farming
systems.
MATERIALS AND METHODS
The experiments were conducted at the Lanet Beef Research
Station, Nakuru, altitude 18 60 m. Precipitation, mean maximum
and minimum temperatures during the trials periods were: 611 mm
and 623 mm; 24.5°C and 24.8°C; 9 . 5°C and 9.2°C in 1982 and 1983
respectively. The soil is a deep sandy loam with good water
holding capacity. The major nutrient status was: K(me) 1.0,
P(ppm) 1.9 and N(%) 0.10 with a pH(H20) of 5.5.
The cultivar Valo (black seeded) was introduced from India.
Nakuru (black and brown seeded) and Kakamega (black seeded) on
the other hand, are local ecotypes grown in the Rift Valley and
western Kenya respectively. Seeds were sown after 25 mm of
germination rain was received each year at a rate of 7 kg ha .
Plots (5 m ) were arranged in randomised complete blocks with
four replications. At sowing, fertilizers were applied at the
rates of 10 kg P ha" , 4 kg K ha and 0.40 kg Cu ha-1 as single
127
superphosphate, muriate of potash and copper sulphate
respectively .
Plots were harvested at monthly intervals after three-
months' growth from germination. Harvests were taken from 1 m2
quadrats excluding the guard row. Harvest four did not include
the seed yield components. Material was cut approximately 5 cm
above the ground and the bulk weight taken. Any shed leaves were
separately collected and weighed. A subsample from each
replicate was separated into leaf and stem fractions. The
separated fractions and a subsample from the abscissed leaves
were weighed and subsequently dried at 75°C for 24 hours for dry-
matter (DM) determination. Seed pods were included in the stem
DM yield in harvests one to three. The dried green leaf and stem
fractions were ground (2 mm) and used for determination of dry-
matter digestibility, crude protein (%N x 6.25) and phosphorus
contents. Nitrogen was estimated using the micro-Kjeldahl method
(Havilah et al., 1977) and phosphorus by X-ray fluorescence
spectroscopy. Dry-matter digestibility was estimated by in vitro
digestibility method calibrated with lucerne of known in vivo
digestibility. The data were subjected to analysis of variance.
RESULTS
All ecotypes nodulated effectively with cowpea rhizobia in the
soil; 23, 27, 22 and 19 effective nodules were recorded in Valo,
Kakamega, Nakuru (black seeded) and Nakuru (brown seeded)
respectively. Time to flowering differed by 20 to 30 days
(Figure 1) .
Total green DM yields ranged from 2000 to 12000 kg ha-1, and
most of the yield was stem (Figure 1) . Green leaf DM yields
ranged from 400 to 3300 kg ha . However. all ecotypes
maintained green leaf DM yields above 2000 kg ha"1 between three
and five months of growth. Leaf shedding increased with
maturity. This was associated with a progressive decline, 3.7 to
4.5 times with time, in leaf/stem ratio according to the
following models:
.80 (Valo)
.76 (Kakamega)
.74 (Nakuru, black seeded)
.90 (Nakuru, brown seeded)
Where Y is leaf/stem ratio and X is sampling time in months after
sowing .
Trends of dry-matter digestibility, protein and phosphorus
contents in green leaf and stem with time are shown in Table 1.
Digestibility was higher (P<0.01) in leaf than stem but declined
by 10% to 15% units over time in both fractions. The decline in
leaf digestibility was slightly higher than in the stem fraction:
0.17, 0.17, 0.10 vs 0.14, 0.09, 0.12 and 0.08% units day"1 in
Valo, Kakamega, Nakuru (black seeded) and Nakuru (brown seeded)
for leaf and stem components respectively between three and six
months of growth. Protein content was higher (P<0.01) in leaf
than stem, and declined by 3% to 5% units in both fraction over
1. Y = 1.51 - .33 X R, "
2. Y - 1.17 - .25 X *,
~
3. Y = 1.25 - .28 X R2 =
4. Y - 1.41 - .31 X R2 -
128
»I B HOXV CM rH CO ID cm o in ^ co i- cm in
g »I 0)
-*■*■** *r «* ■- T ** *» ■»- ■«- ■«- -* - -*
O 2 PCO o o o o o o o O o o o o o o o o
Xi O -
O. J3tW
to a,"
O to H
2 o 4-1 co vo in r~ CO CM 0\ VO rH CM CO \o in in co 01
o. £ 3
CO CO CO CO co co cm CO CO CO CO CO CO CO CO CO p
p•0 o o o o o o o O o o o o o o o o 0)
£ r-t
id
c
e O■H XI XI X! X! XI Xj XI 1a) c XI id id id Xl id id 10 xi id id id XI Xl id id
p -H
01
B
0)
P
inohH rH r> CM H <? \o *r 'o ^f in rH CM O
0 -O
h p ^ ■» CO C> CO CM rH CO CO CM rH r^ «no* p
a o
n —
CO i-l rH rH HHH H r-\ H rH rH rH id tA
0)
a O >1p
T5 Xl
2 01■0 01
0 2 id3
tfiiKe CO rH CO «r in »O t- rH O CO H CO •o o'
ai c
» id
u
rH ot o- r- o en r- r- OOM^ CM Ol c- r*
0 u-b CM i-l rH i-l CM rH rH rH CM r-t H <-t CM rH rH rH
>i r-i
p rH 0)
-H O r-t
i-l XI XI XI XI XlXl ih a
-•H
>1
p
B
p
Xi XI id id xi xi io CO xi id to io Xi id id id
p p
-H
-H -
-o *r io cm co o\ co H cm to a\ *r CO Tf ID CO
O rH
C 34-1 to -H io co en ^- cm o r- *T co in o c^ ID CO CM 01
s01 c rH to in in * «* in in -* -- in in in I1 in in in ■*
4) O -H — ai
tJl-H Xl*> &-M
-H P -H »-■a o P Xi XI Xi XI XI XI P c
CO Ul XI XI id id Xi Xi fl) 10 XI cd io cs XI 10 10 A O id
u u 0)
-H
it-i
id
ninn « c- co ^r lO CI CO ^ CO 01 CO CO P- 0 0
■ 0) c
0' 3
C C Q
91 tU
p 3 <f»HB mom o
CM O CO *■ - CM cn in
p oi Q t- vo vo in r» r- v> VO C- t~» VO ID t-, r~ U5 \D
io o' ■h id
E CO > O
1 >H O C P
r7iH
S CO -H
X! dai
•0 p * BP C
C en O •0 0X1 id 0X1X1 10 0 XI id id OXlXl id O -H-iH
- 01 N. -H U3 CM H CO CO r~ rH CO ID CO 01 rH r- rH O rH u z-a
O t»
-H U
p Q>
P CM f- - CM o ■* ■« CM H ^ CM CO HMOfl
o) o10 id
U3 r-l O o o H O O O H O O O H O O O U P oid X! 3 0
^ c P Xi id
-H C-H
B OUr.
w in ,9'. B P in
P op P H
c id -io c ^^
■H V\ ai <D
U PttM 4J B co ^ in **? co ^ in yo n ^r in io r) ^ in io
b m^ID C -H
S8 E-i id a
in oi
t co B 10 HH
H d) 0' 1 r-~ 1 ~ -rH 01-H
&
id
e
3 x-d 3 C 73
U 5 0)
rlST)
0) Kuai
H P O 10 3 id-O 3 O-0
XI o rH X X rH 0) iA U Q) ^^*
10 0 10 10 id XI 0) id Xl 0) H
H w > « S5 — CO S5--10 —
129
V « ° 3 o *
15;s*J TJ
o
u S o
Vi
o
3
QJ
1
u T3
O)a
<•
rt 3 O^ O
J H- ° ^
o 2 o0w
o
« l
c
—1 m
l» o
p -(
o «
e- >
C5
 
o
E
o>
E
130
time,
time.
Phosphorus content showed no significant changes with
Seed yield was a function of seed pods m . All ecotypes had
an average of 2 seeds pod . Nakuru (brown-seeded) out-yielded
(P<0.5) the other ecotypes (Table 2). Seed protein content was
similar (P>0.05) .
Table 2. Component of seed yield: Seed yield and
crude protein content in seed.
Ecotype Components of seed Crude
yield Seed yield protein
Pods m Seed m (kg/ha A) (%)
Valo 190a* 232a 733.7a 28.5
Kakamega 172a 292a 655.9a 27.3
Nakuru 166a 246a 409.4a 24.1
(black-
seeded)
Nakuru 325b 860b 1812.1b 28.2
(brown-
seeded)
OMeans within each column not followed by
common letter differ (P<0.05) according to
Duncan's multiple range test.
DISCUSSION
The variation in time to flowering confirms the existence of
physiological differences with Lablab purpureus ecotypes
(Pulseglove, 1968) . This variation could be exploited in
selecting germplasm suitable to a particular area.
Green total and leaf DM yields were higher than those
recorded for Lablab purpureus cv. Rongai (Murtagh and Dougherty,
1968; Hendricksen and Minson, 1986). The ceiling leaf DM yields
suggested by Murtagh and Dougherty (1968) are lower than that
contained in this study. Leaf retention characteristic exhibited
by the Nakuru (brown seeded) ecotype indicated a useful attribute
under grazing, and generally all ecotypes maintained more than
2000 kg ha green leaf content longer in the growing season than
cv. Rongai (Hendricksen and Minson, 1986) .
The decline in DM digestibility and crude protein content in
the foliage with maturity is consistent with reports on other
tropical pasture legumes (Jones, 1969; Norton, 1982). This
decline was also associated with increased rates of leaf fall.
However, the green leaf fraction maintained favourable levels of
protein and digestible fractions. This was probably due to
retranslocation of nutrients from the senescing leaves.
Phosphorus content varied least with maturity. Partly, this is
because phosphorus is not subjected to extensive retranslocation
between plant organs.
131
All ecotypes yielded more seed than was recorded by Davies
and Hutton (1970) but lower, except Nakuru (brown-seeded) , - than
The data on pods m-2 andwas reported by Pulseglove (1968).
seeds m and their effects on final seed yield appear to suggest
there were differences in seed size, hence a potential to select
for high seed yield among the existing materials. The protein
content in the seed further shows a legume with potential use as
human food.
A substantial amount of leaf abscission occurred late in the
season. This was the period (5 to 6 months of growth) when seeds
were approaching maturity. As the shed leaves would not be
available for grazing, studies are needed to investigate the time
to optimise both the DM and seed yields. Overall, the DM yields,
forage nutritive value, seed yield and its crude protein content
show that Lablab purpureus would contribute considerably in
livestock/crop farming systems.
ACKNOWLEDGEMENTS
The author is thankful to S. Gativiku for technical assistance.
Seeds of Valo ecotype were obtained from the Director, National
Seed Quality Control Station, Nakuru. This paper is published
with the permission of the Director of Research, Ministry of
Livestock Development, Kenya.
REFERENCES
Davies, J.G. and Hutton, E.M. 1970. Tropical and subtropical
pasture species. In:R.M. Moore (ed.), Australian grasslands.
ANU Press, Canberra.
Hamilton, R.I., Lambourne, L.J., Roe, R. and Minson, D.J. 1970.
Quality of tropical grasses for milk production. Proc. X1.
International Grasslands Congress Surfers Paradie. ,
Australia. pp. 860-864.
Havilah, E.J., Wallis, D.M. , Morris, R. and Woolnough, J. A. 1977.
A micro-colorimetric method for determination of ammonia in
Kjeldahl digests with a manual spectrophotometer. Lab.
Proc. 26:545-547.
Hendricksen, R.E. and Minson, D.J. 198 6. Growth, canopy
structure and chemical composition of Lablab purpureus cv
Rongai and Samford, S.E. Queensland. Trop. Grasslds. 19:81-
87.
Hendricksen, R.E. and Myles, D.J. 1980.
Lablab for beef cattle production.
Q.A.C. Lawes. p. 2 57.
Methods of using Rongai
Proc. Aust. Agron. Conf.
Jones, R.J. 1969. A note on the in vitro digestibility of two
tropical legumes. Phaseolus atropurpureus and Desmodium
intortum. J. Aust. Inst. Agric. Sci. 35:62-63.
Murtagh, G.J. and Dougherty, A.B. 1968. Relative
lablab and velvet bean. Trop. Grasslds. 2:57-63.
yields of
132
Norton, B.M. 1982. Differences between species in forage
quality. In: J.B. Hacker (ed.), Nutritional limits to
animal production from pastures. C.A.B. Farmhouse, London.
Pulseglove, J.W. 1968. Dicotyledons I. In: Tropical Crops. John
Wiley and Sons, New York, U.S.A. pp. 273-276.
Skerman, P.J. 1977. Lablab purpureus (L) sweet. In: Tropical
forage legumes. FAO plant production and protection series,
No. 2. pp. 314-322.
133
ADAPTATION OF FORAGE AND BROWSE LEGUMES TO THE
SEMI-ARID MID-ALTITUDE REGIONS OF KENYA -
AN INTERIM REPORT
L.K. Menin and D.M. Njarui
National Dryland Farming Research Centre, Katumani
P.O. Box 340, Machakos, Kenya
W.M. Beattie, b.a. Keating and R.K. Jones
ACIAR/CSIRO Dryland Project
P.O. Box 41567, Nairobi, Kenya
Abstract
The adaptation of 160 legumes of various growth forms is being
studied at four sites in the semi-arid, mid-altitude districts of
Machakos and Kitui in Kenya. The sites cover an altitude range
from 900 to 1600 m and mean annual rainfall range from 600 to
1000 mm. Establishment is by direct-seeding using a minimum
tillage approach. Phenological characters, susceptibility to
pests and diseases, and growth vigour are measured monthly, and
dry-matter production annually in row-plot experiments. Seasonal
yields of dry matter and nitrogen are also measured in companion
sward experiments using a "best bet" sub-set of the collection.
A number of legumes which are well-adapted to the
environments have been identified after three growing seasons and
appear to have potential for a variety of uses in the local
farming systems.
INTRODUCTION
Agriculture in the semi-arid midland regions of eastern Kenya is
characterised by very low capital inputs, low outputs (even in
good seasons) and a high frequency of seasons in which production
of staple foods is insufficient for local needs. Rapid increases
in rural population densities and in the intensity of
cultivation, fuelwood harvesting and grazing have resulted in
alarming rates of soil fertility depletion and soil erosion.
While rainfall is the principal limiting resource, it is usually
closely followed by soil fertility, hence the need to examine the
role of legumes in the cereal-based farming systems of the
regions as an alternative to the use of expensive nitrogen
fertilizers.
FAO/GK/NORAD projects since 1974 (Wilton, 1976; Van Soest,
1986; Ibrahim, 1981; Chabeda, 1976) have made extensive
collections of forage legumes within Kenya and evaluated them at
four sites - Kitale, Kiboko, Katumani and Mtwapa along with some
introduced material. However, much of the evaluation work is
unpublished or is of a semi-quantitative nature, and the full
potential of forage and browse legumes in Kenya's bimodal semi-
arid areas remains little understood. In addition, many new
species and accessions have become available from more recent
plant collecting and evaluation activities in other parts of the
world, and this material required testing in Kenya.
134
The objectives of the research programme reported here are:
1. to assemble an extensive seed collection of forage and
browse species with potential adaptation to semi-arid areas
of Kenya,
2. to assess the adaptation of this material in a series of
experiments to a range of climatic conditions in the
districts of Machakos and Kitui,
3. to select a nucleus of widely-adapted legumes that might fit
into the varied farming systems in these districts, and to
assist agronomists and animal production scientists in
further evaluating this material in "on-farm"
experimentation .
The work involves close collaboration between staff of the
Kenya Government's Ministry of Agriculture and staff of the
Australian Dryland Farming Project (ACIAR Project 8326) which is
being implemented by the CSIRO Division of Tropical Crops and
Pastures.
MATERIALS AND METHODS
Legumes
A total of 167 accessions from 23 genera are under evaluation.
Details of ecotypes/species, Commonwealth of Australia Plant
Introduction numbers, Kenya Introduction numbers, origin (where
known) and project sites planted are contained in Appendix 1.
Field Sites
The experiments were undertaken at four fenced sites as follows:
1. Katumani NDFRS , altitude 1600 m, mean annual rainfall 717
mm, mean annual temperature 19.6°C, soil type: chromic
luvisol, pH 6.5.
2. Kiboko RRS , 975 m, rf 595 mm, mean temperature 25.7°C,
rhodic Feralsol, pH 5.8.
3. Mua Hills, 1650 m, rf 1179 mm, estimated mean temperature
19.0°C, red sandy earth, pH 7.0.
4. Ithookwe (Kitui), 1160 m, rf 1080 mm, mean temperature
22.5°C, red sandy earth, pH 5.8.
Experimental Design and Layout
Row plot experiment. The experiments were laid in a randomised
complete block design with 104 accessions and 3 replicates.
Plots are 6.75 m and consist of two rows each 2.25 m long and
rows 1.5 m apart. Replicates run along the contour. One of each
pair of rows received a single dressing of a basal fertilizer.
The other row was unfertilized.
135
Sward experiments. For each season of planting, the experiments
were arranged as randomised blocks with 20 accessions and three
replicates. Individual plots were 16 m . Seed was not broadcast
over the entire plot but distributed along disturbed rows 0.5 m
apart. All plots received a basal fertilizer dressing at
establishment.
Seed Preparation
Row plot experiments. Seed viability was determined by
germinating scarified seed on moist filter pads in petri dishes.
Approximately 50% of all seeds/plot were scarified by sandpaper
or scalpel. Inoculation with the requisite rhizobia in a gum
arabic/distilled water solution was made a few days prior to
planting. Inoculated seed was kept under refrigeration.
Sward experiments. Seed preparation was similar to that for row
plots except that 100% seed scarification was attempted, and the
seeding rate was adjusted to give similar numbers of viable seeds
per plot (Table 2) .
Land Preparation, Sowing and Seedling Establishment
Land preparation was minimal except for Ithookwe (site 4) which
was ploughed just prior to planting. Vegetation was cut back and
burned off during the dry season and line rows pegged out. A
small chisel plough attached to a two-wheel Honda mini-tractor
marked a 4 cm deep furrow along each planting row. At the onset
of rains, when weed seedlings had emerged, the furrows were
dusted with Aldrin 2.5% @ 20 kg/ha. Legume seed was then hand-
sown, lightly covered and firmed with soil. Two days later or
just prior to legume germination the planted rows were sprayed
with Roundup herbicide 2 1/ha using a knapsack sprayer fitted
with a fan nozzle. Subsequent row weeding was done by hand or
with a rope wick containing Roundup.
Data Collection/Harvesting
Row plot experiments (Table 1) . Agronomic and phanological
characters recorded monthly from the second month after sowing
included plant population and development (vegetative, flowering
and seeding) , height, date of anesthesis, insect and disease
incidence and a visual bulk rating on a one-five scale. In early
October 1986, after 11 months of uninhibited growth, a measured
length of average plant material in each row plot was cut at 5 cm
above ground level (5 cm from the crown for trailing legumes),
oven dried at 60°C and weighed. Sample weight was converted to
yield (kg/ha) by the formula
{Sample wt)
10
{cut length x 1.5).
Sample experiments. Plant measurements similar to those given
above were made except that for visual bulk ratings. Harvesting
entailed cutting two randomised 1 m2 quadrats 5 cm above ground
level from each plot three times per year. The first harvest was
in late January, 1986 (end of short rains), the second in early
June, 1986 (end of long rains) and the third in late September,
136
1986 (end of dry season) . The latter harvest also included
measurements of regrowth from the first harvest. Cut material
was oven-dried at 60°C, weighed and subsampled for later N
analysis.
Fertilizer Applications
Two weeks after legume seedling emergence a basal fertilizer was
hand broadcast along one row of the row plots to a width of 1.5 m
and evenly broadcast over the swards. Rates in kg/ha were:
triple superphosphate 200, muriate of potash 41.7, dolomite 250,
manganese sulphate 15, borax, zinc sulphate and copper sulphate
10, respectively and sodium molybdate 0.36. The swards received
the same mixture broadcast over the entire plot.
Methodology for Experiments with Leucaena Accessions
At Katumani, a collection of Mexican high altitude Leucaena
accessions is being tested quantitatively against a suite of
commercial cultivars including cv Cunningham, cv Peru and K8.
Adaptation is progressively monitored throughout each year by
measuring shoot extension growth relative to soil moisture
content and air temperatures. Destructive harvests are made
twice and four times per year for wood, stem and leaf yield.
RESULTS
Row Plot Experiments
Data from 16 accessions which performed well at Site 1
(Katumani) , Site 2 (Kiboko) or both sites are presented in Table
1. Of the annuals, the short season Macrotyloma africanum with
densely twining fine stems, had high resistance to insect and
disease, good seed production and an excellent seedling
regenerative capacity. Another trailing legume Vigna unguiculata
CI60452 was more strongly persistent at both sites than other
entries from that genus. Alysicarpus rugosus established well
and had the highest edible dry matter (EDM) (1938 kg/ha) at the
end of the evaluation period at Site 1. The two Stylosanthes
hamata lines, cv Verano and K14363, are morphologically similar,
but the latter persisted later into the dry season. Both
accessions and the late flowering S. humilis are particularly
susceptible to termites. Lablab purpureus cv Rongai grew
vigorously at both sites and gave high visual bulk ratings. It
flowered profusely, but seed production was reduced by the
depradations of flower-eating beetles Mylabris and Coryna sp.
The selected perennial legumes (Table 1) have persisted
strongly at Site 1 but only Desmanthus virgatus, S. scabra cv
Fitzroy cv Seca and S. guianensis Alupe 'C survived the, harsher
environment at Site 2. These Stylosanthes accessions yielded >4
tonnes EDM/ha at Katumani and appear particularly well adapted at
that site. The semi-erect shrub legume Desmanthus virgatus
K14456 was slow to establish but showed promise later in the
experiment at both sites. Neonotonia wightii was also slow to
establish but grew vigorously after the second season. Cassia
rotundifolia cv Wynn flowered very early and continued to produce
seed throughout the year. It has persisted strongly at Site 1
137
■* A
5 a .
c^ ■n H
i
O m' ' 1 1 1 1 1 1 1o e n
00 CO o CM .-H O CO »H m r- m CM
i- r-i CM VO m CM C* m r-t o 1^ CM
i7- 1 ' -c 00 1o 1 CO r-
i-l
U") O
1M i-- --i « -T
o o o o o O o o o - o o o - O O
o o o o o cm H o o o o o o o o O
- - o o - o o --- o o - - o - o o
o CN - 1- 1- CM -
CM CM CM « i- 1o CM CM CM o - CM „ ^H 1- io
CM CM 1- io 1- i- tM io CM CM i- i- 1- « c- 1o
O O - O - O O o O o O O O o --» o CO i£i i0 r^ « 1- -» CJi -- CO o C7i
CM CM
o O m O - o - - - o O o o - - O
CD iH m iO iO m i- w1 CM rH - H o Ch o O(H rH H CM CM t-4 ^^ H H
O O n m m m in o m m o m o o m o
CM \o .H CM rH iH CM CM CM "^ CM " CM <■ i- m
O in - - m o m o in - - Tt - m m o
in in .H CM rA ft CM CM -» i- CM io -» "* -o
- - O O o o - o O - - O - - - -
CM o » CO r-i S CO CO ai-* rH CM CM H -"- CM CM CM CM CM CM
O o CM - 1o H O - i£> iD o iO \D m O -
i- m i- -s <T <r -* i- CM ^ o CM 1- -a- -J -»
o - o o O o o o o O o o o o o -
CO 0-i cji --- m o -» .H r-H ft -~t —4H CM CM CM •* CM io CM CM CM CM CM CM CM
i£ i- m - CO CO N O r-l m CO m 1- O - O
io -T i-
CO
in -» -T CI in 1- 1- -» in m
1 u
a
0
d rt
u
f,
9
4)
CM a
0
CO CM C -J 3 N -—l io ij m iO *i *i 4) -I > ^ urt -T « .0 i- ~2 m II u < 9
S > i- ■* - 3 m U »- * u. W
0
U- > -» « m > ■-t o « rt - i- > >
pi § o w h o U > 3 0
-4
C
CJ U
> g c| -
n
u fl) i-
0
u 4i 44 rt rt-* 0 u
c*
3 i S a u uCO u 4-1 u - -43 XI
iA
§
u - c» 4-1 -J 2 -4 tl rt-T 3 u
A £
a ^ 00 U J3 3 3 u 0
iH ■ »M x:
3
M 0 00 iw ClO 00 ifl
u; ti 1l T) > 4-1 53 rt w M u § c ifl w w 10i- s- 4i 4) » » 44 44 4J 4)3
|
A A A 3 --» V) rt J3 43 J3 A 43
3 ^ 4-1 ^ - 3 4-- 4J 4-1 4-1 4-1
a Crt s C I i-t j3 5 C■ 5 Crt s s g* m 4-1 ■du xi ^ M w it IT) C 4-1 w Ul n m M
-4 a 0 ■ 0 0 4 fl 0 0 0 0 O 0 0
Ui ►■ C w G E d h —i —1>. A CJ >i X M M w w 0 >s X >, ^rt rt 4-i 4-1 it 41 4) » 4-1 4-" hi w 4-t< J 3: w ») c/j > CJ Q Q Z lrt to to W W
« Jfl -• « -
:38
Table 2. List of accessions used in sward studies.
Present in planting
number
2 3 4 Mean
sowing
rate
kg/ha
Alysicarpus rugosus CPI 52351
Alysicarpus rugosus CPI 30034
Cassia rotundifolia Q10057
Cassia rotundifolia cv Wynn
Centrosema pascuorum cv Cavalade
Centrosema pascuorum CPI 65950
Centrosema schottii CPI 82271
Centrosema schottii CPI 76010
Centrosema virginianum CQ 2748
Clitoria ternatea CPI 48337
Clitoria ternatea CPI 47187
Clitoria ternatea CPI 49963
Desmanthus virgatus CPI 40071
Desmodium intortum cv Greenleaf
Dolichos Lablab K1002
Lablab purpureus cv Rongai
Lotononis bainesii cv Miles
Macrotyloma africanum CPI 24972
Macrotyloma africanum CPI 60207
Macrotyloma daltonii CPI 60303
Macroptilium atropurpureum
cv Siratro
Mucuna sp
Neonotonia wightii K75.2366
Rhyncosia minima CPI 52713
Rhyncosia malacophylla K18176
Stylosanthes capitata CPI 55843
Stylosanthes guianensis cv Graham
Stylosanthes guianensis cv Cook
Stylosanthes hamata cv Verano
Stylosanthes scabra cv Fitzroy
Stylosanthes scabra cv Seca
Stylosanthes fruticosa CPI 41219A
Vigna trilobata
CPI 13671
Vigna unguiculata cv M66
Vigna unguiculata cv Red Caloona
4.3
4.3
5.2
5.2
26.1
26.1
26.1
26.1
26.1
43.8
43.8
43.8
9.
7.
89.
9.
6.
6.
6.1
16.2
9.1
160.5
6.8
10.8
10.8
7.1
7.
7.
7.
7.
7.
7.
10.8
82.4
82.4
with useful seasonal seedling regeneration. Desmodium intortum
cv Greenleaf is susceptible to termites and while quite vigorous
during seasonal rains only just survived the long dry season at
Site 1.
Sward Experiments
The most complete data set currently available is for the first
planting at the Maruba site. The pattern of dry-matter
139
accumulation varied greatly between accessions as shown in Figure
la. Single-season grain legumes such as Vigna unguiculata cv Red
Caloona produced modest levels of dry matter after the first
rainy season (see Harvest 1) but none thereafter. The dual
purpose (forage/grain) legumes, such as Lablab purpureus (cv
Rongai) were the most productive after the first and second
rainy seasons (see Harvests 1 and 2) but declined thereafter (see
Harvest 3) . Maximum for the 18 other accessions tested was only
11% of the biomass produced by lablab purpureus . Legumes such as
Alysicarpus rugosus, Stylosanthes scabra and Neonotonia wightii
needed two rainy seasons to reach the production levels of an
accession like Lablab purpureus. Thereafter, strong perennial
species such as S. scabra and N. wightii had a clear advantage
whilst annuals such as A. rugosus declined over the long dry
season.
Figure la. Relative dry-matter yield (expressed as a percentage
of the maximum dry-matter yield) for five legume
swards grown at Maruba, Kenya. Swards were planted
on 30-10-85 and the third harvest was taken 327 days
later on 22-9-86. Absolute dry-matter yield in
tonnes ha-1 is shown in parentheses for each harvest.
%max. DM
yield
l 00o
90o
80 •
70-
60 •
50o
40
30
20
i0
0
Days from
plantiog
Seasooal
rainfall (mm)
(i.4)
86
_l_
(4.0)
Harvest oumber
 
u
296 3ll
Vigoa uoguiculata [ljljl ILablab purpurtut
(5.i) "
Neoootonia wightii
k/5.23662 cv Red Colooma cv Roogai
Alyticorpu* mgotuafTT.
CPl 5235i
Stylosaoto«s
cv Fitiroy
140
The first sward plantings at Katumani also provided an
opportunity to compare the productivity of a relatively large
number of Stylosanthes species under the cool dry conditions of
the midlands of eastern Kenya. In all, six accessions from five
species were included in this planting as shown in Figure lb.
None of the long season annuals, biennial or perennial species
produced any significant levels of biomass by the end of the
first season. After this time, however, the Stylos were all
relatively productive with the S. scabra cultivars, Fitzroy and
Seca the most outstanding (3.4 to 4.4 tonnes ha"?o) , followed by
S. guianensis cv Graham (2.5 to 3.5 tonnes ha"1). The least
adapted Stylo to this region was S. hamata cv Verano.
Figure lb. Relative dry-matter yield (expressed as a percentage
of the maximum for the particular harvest) for six
Stylosanthes accessions grown at Maresha, Kenya.
Other details as Figure la.
%max.DM
yield
l 00
90
80
70
60
50
40
30
20
l 0
(i.4) (4.0) t~,
E3-e TTTTiL
_i_
 (5.i)
 
Days from 86
plaotiog
Seasonal 296
rainfall
2i8 327
4i8
1
S. hamata
cv Veraoo
S. capltata
CPl 55843
S.scabra
cv Seca
S.f ruticosa
CPl 4l2l9A
S. guiaoeosis
cv Graham
S. scabra
cv Fitzroy
141
Leucaena Experiment
The first two destructive harvests from the Leucaena experiment
have now been completed. Dry matter data is shown in Figure 2.
From the preliminary data, the cultivar Peru appears to be the
most productive in terms of leaf production, whilst L.
diversifolia CPI 33820 was producing the largest amount of wood.
Figure 2. Dry weight data from first cut of Leucaena accessions.
Planted: 4/4/86; First cut: 29/12/86.
Dry Might
 
m
m
^1
Woody ttem
53•; Sreeo «t«m
Cat* from 9 reet of 4 meatvreaeiit
pi ooto (2 reot enly fer CoeeJefken
and IlJ1T]
Leaf on« petio4t
142
DISCUSSION
The potential role of forage and browse legumes in the semi-arid
midlands of Kenya is in smallholder farming systems. Legumes
could conceivably be used as inter-crops/alley-crops with
cereals, in erosion control, in terrace bank stabilisation, in
the improvement of degraded, non-productive grassland or as
simple cut-and-carry fodder banks. The range of growth habit and
persistence within the plant material being evaluated extends
from the short twining annual Macrotyloma africanum to the small
perennial tree Leucaena. Further experimentation (preferably on
farms) is required to sort out the respective roles of material
available, but the results from the experiments described here
indicate that a good range of well-adapted leguminous material is
now available.
Materials recommended for further testing in more favourable
environments in the region (as characterised by Site 1 at
Katumani NDFRS) are Cassia rotundifolia cv Wynn, Desmanthus
virgatus K14456, Lablab purpureus cv Rongai, Macrotyloma
africanum K14348, Neonotonia wightii K2366, Stylosanthes
guianensis 'Alupe C and Stylo scabra cv Fitzroy and 'Seca'. At
less favourable sites (similar to Kiboko, Site 2) this list is
reduced to Desmanthus virgatus and Stylosanthes scabra.
With the expansion of evaluation work to two other sites (3
and 4) and the inclusion of more plant material - from the Kenya
Plant Quarantine Station at Muguga, from a small, indigenous
plant collection and from further introduction from CSIRO,
Queensland - the range of accessions being tested has recently
increased. While evaluation of the following lines Aeschynomene
americana cv Glenn, Cassia pilosa CPI 57503, Lotononis angolensis
CPI 62202, Macrotyloma axillare cv Archer, Macrotyloma sp.
(indigenous), Mucuna pruriens aff . , Rhyncosia malacophylla
(Synon. sennarensis) , Rhyncosia totta and Stylo guianensis cv
Cook is incomplete, they all show considerable promise.
The critical role of legume seed nurseries cannot be over
emphasised. Once key lines have been identified, seed requests
from research stations and research or extension officers cannot
be met if supply is a constraint. The nurseries should be
established in areas where demands on irrigation, man-power and
supervision are easily met. Work is underway this season to
produce experimental quantities of seed of the promising
accessions noted above.
ACKNOWLEDGEMENTS
We gratefully acknowledge the support of Mr. S.N. Muturi,
Director of Agriculture, Kenya; Mr. P.K. Kusewa, Director of
Katumani, NDFRS; Dr. D. Okioga, Director of the Plant Quarantine
Station, KARI, Muguga and Dr. B. Woie, Director of Kiboko Range
Research Station. We also thank numerous officers and staff of
Katumani NDFRS. To ACIAR we are grateful for the funds provided
to carry out the project.
143
REFERENCES
Chabeda, A.E.o. 1976. Drought resistance in some Kenya grasses
and legumes. Annual Report. National Agricultural Research
Station, Kitale, Kenya.
Ibrahim, K.M. 1981. Forage plant introduction. Technical Report
No. 3 Project TF-KEN 29 (NOR). FAO, Rome.
Van Soest, L.J.M. 1986. Exploration and evaluation of grasses
for seed production Kenya. Eastern region of Kenya.
Working paper. Project TF-KEN 29 (NOR) . FAO, Rome.
Wilton, A.C. 1976. Exploration and evaluation of grasses for
seed production in Kenya. Working paper. Project TF-KEN 29
(NOR) . FAO, Rome.
144
Appendix 1. Legume-evaluation: Accession list.
SiteCPI Origin
Aeschynomene americana
Aeschynomene americana
Aeschynomene elegans
Aeschynomene falcata
Aeschynomene villosa
Aeschynomene villosa
Alysicarpus glumaceus
Alysicarpus hamosus
Alysicarpus longifolius
Alysicarpus monilifer
Alysicarpus rugosus
Alysicarpus rugosus
Alysicarpus rugosus
Alysicarpus rugosus
Alysicarpus rugosus
Alysicarpus rugosus
Alysicarpus vaginalis
Arachis monticola
Arachis pintoi
Cassia pilosa
Cassia rotundifolia
Cassia rotundifolia
Cassia rotundifolia
Cassia rotundifolia
Centrosema acutifolium
Centrosema acutifolium
Centrosema brasilianus
Centrosema macrocarpum aff
Centrosema pascuorum
Centrosema pascuorum
Centrosema plumieri
Centrosema plumieri
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema pubescens
Centrosema schottii
Centrosema schottii
Centrosema virginianum
Centrosema virginianum
Clitoria ternatea
Clitoria ternatea
Clitoria ternatea
Desmanthus bicocnutus
Desmanthus subulatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
91145 19712 Mexico 1
cv Glenn 19897 Mexico 1,4
37552 19714 Argentina 1
cv Bargoo 19715 Paraguay 1,4
37235 19716 Mexico 1
91113 19720 Mexico 1
52366 14449 Madagascar 1,2
94491 14374 Oman 1,2
94490 14376 Oman 1,2
52343 14377 Madagascar 1,2
30034 14382 India 1,2,• 3, 4
52348 14383 Zimbabwe 1
52351 14384 Malawi 1,2, 3, 4
76980 14447 Zambia 1
94489 14395 Ethiopia 1,2
- (50) Kenya 1,2
34149 14386 Nicaragua 1,2,.4
CQ990 14388 - 1
58113 14416 Brazil 1
57503 14451 Venezuela 1,2,,4
86172 19724 Mexico 1
86178 19725 Mexico 1
Q10057 14450 Brazil 1
cv . Wynn 18177 - 1,2, 3,,4
92874 14391 Brazil 1,2
94303 14390 Colombia 1
55698 14417 Brazil 1,2,,4
78358 20170 Brazil 4
65950 14392 Equador 1,2,*3,.4
cv Cavalcade 14418 - 1,2,* 3,o4
60477 14393 Brazil 1,2
82269 14394 Cuba 1,2
43197 20171 Colombia 4
46543 14395 Guatemala 1,2
58575 20172 Colombia 4
63895 14396 Brazil 1,2,,4
79630 20173 - 4
92721 14397 Colombia 1
cv Belalto 14332 - 1,2
cv Centro 14419 - 1,2
76010 14398 Mexico 1,2,,3
82271 14445 Cuba 1,2,,3
CQ2748 14399 - 1,2,.3,,4
91142 14400 Mexico 1,2,. 3,.4
47187 14402 Sudan 1,2,o3,,4
48337 14403 Tanzania 1,2,*3,.4
49963 14404 - 1,2,• 3,.4
91162 14438 Mexico 1
90857 14406 Mexico 1
40071 14456 Brazil 1,2,• 3, 4
55719 14407 Venezuela 1,2
65947 14405 Equador 1,2
78373 14408 Argentina 1,2
83570 14409 Brazil 1,2, 4
145
CPI Origin Site
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmanthus virgatus
Desmodium dichotosum
Desmodium intortum
Desmodium intortum
Desmodium oringlei
Desmodium prostratum aff
Desmodium setigeruam
Desmodium subsericeum
Desmodium wigginsii
Desmodium sp. D
Desmodium sp. D
Dolichos sp.
Dolichos sericeus aff
Lablab purpureus
Lablab purpureus
Lablab purpureus
Lablab purpureus
Lablab purpureus
Lespedeza striata
Lotononis angolensis
Lotononis bainesii
Macroptilium atropurpureum
Macroptilium atropurpureum
Macroptilium atropurpureum
Macroptilium atropurpureum
Macroptilium atropurpureum
Macroptilium atropurpureum
Macroptilium heterophyllum
Macroptilium heterophyllum
Macroptilium heterophyllum
Macroptilium lathyroides
Macroptilium longipendun-
culatum
Macroptilium martii
Macroptilium prostrata
Macrotyloma africanum
Macrotyloma africanum
Macrotyloma axillare
Macrotyloma daltonii
Macrotyloma daltonii
Macrotyloma uniflorum
Macrotyloma sp.
Medicago rugosa
Medicago rugosa
Medicago sativa
Medicago scutella
Medicago trunculata
Mucuna pruriens aff
Neonotonia wightii
Neonotonia wightii
84508 14410 Mexico 1,2
85178 14411 Mexico 1,2
90750 14412 Mexico 1,2
91146 14413 Mexico 1,2
91326 14414 Mexico 1,2, 4
92818 14415 Belize 1,2, 4
47186 - - 1,2
cv Greenleaf 14455 - 1,2, 3,4
91135 14338 Mexico 1,2
37232 14334 Mexico 1,2
91232 14336 Mexico 1,2
52431 14340 Malawi 1,2
78402 14335 Argentina 1,2
90418 14337 U.S.A. 1,2
91186 14453 Mexico 1
91212 14333 Mexico 1,2
24973 14452 Zimbabwe 1,2, 4
- (128) Kenya 1,4
30702 14466 Burma 1
41222 14341 Burma 1,2
cv Highworth 14463 - 1,2
cv Rongai 14420 Kenya 1,2, 3,4
- 1002 Kenya 1,2, 3,4
cv Kaloe 19899 U.S.A. 1,4
62202 14435 - 1,2, 4
cv Miles 14421 - 1,2, 4
84989 14343 Mexico 1,4
84999 14344 Mexico 1,2
90748 14468 Mexico 1,4
90776 14469 Mexico 1,4
90821 14465 Mexico 1
cv Siratro 14461 Mexico 1,2, 3,4
90448 14345 Mexico 1,4
91144 14467 Mexico 1
91222 14346 Mexico 1
cv Murray 14464 India 1,2
55751 14422 Brazil 1,2
49780 14347 Brazil 1,2, 4
78450 14342 Argentina 1
24972 14348 Zambia 1,2, 3,4
60207 14349 Zambia 1,2, 3,4
cv Archer 14462 - 1,4
60303 14350 Namibia 1,2,.3,4
94496 14351 - 1
cv Leichardt 14460 - 1,2
- (129) Kenya 1,4
cv Paraponto 19905 - 1,4
cv Sapo Gama 19906 - 1,4
cv H. River 19900 - 1,4
cv Sava 19907 - 1,4
cv Jemalong 19908 Australia 1,4
- (119) Kenya 1,4
- 2366(67) Kenya 1,2,.3,4
- (35) Kenya 1,2,.3,4
146
CPI Origin Site
Neonotonia wightii
Rhyncosia densiflora
Rhyncosia edulis
Rhyncosia malacophylla
Rhyncosia minima
Rhyncosia minima
Rhyncosia minima
Rhyncosia minima aff
Rhyncosia minima aff
Rhyncosia minima aff
Rhyncosia totta
Stylosanthes capitata
Stylosanthes fruticosa
Stylosanthes fruticosa
Stylosanthes guianensis
Stylosanthes guianensis
Stylosanthes guianensis
Stylosanthes guianensis
Stylosanthes guianensis
Stylosanthes guianensis
Stylosanthes hamata
Stylosanthes hamata
Stylosanthes hamata
Stylosanthes hamata
Stylosanthes humilis
Stylosanthes humilis
Stylosanthes scabra
Stylosanthes scabra
Stylosanthes scabra
Stylosanthes scabra
Stylosanthes scabra
Stylosanthes scabra
Stylosanthes subsericea
Stylosanthes sympodialis
Stylosanthes viscosa
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Stylosanthes sp.
Trifolium repens
Trifolium sp.
Vigna ambacensis
Vigna frutescens
Vigna luteola
Vigna oblongifolia
Vigna praecox
Vigna trilobata
Vigna unguiculata
vigna unguiculata
Vigna unguiculata
- (150) Kenya 1.4
52690 14353 Tanzania 1,2
52127 14352 Paraguay 1,2
- 18176 Tanzania 1,2, 3 4
52713 14355 Tanzania 1.2, 3
78473 14356 Argentina 1
84953 14357 Mexico 1,2
- (14) Kenya 1,2
- (41) Kenya 1,2
- (45) Kenya 1,2
52742 14358 Zambia 1,4
55843 14459 Brazil 1,2, 4
41219A 14426 Sudan 1,2, 3 4
- (33) Kenya 1.2
79637 14364 Brazil 1,2, 4
- Alupe c 1,2, 4
- Alupe I 1,2
cv Cook 18189 Colombia 1,2, 3 4
cv Graham 14427 Bolivia 1,2, 3 4
cv Oxley 19901 Argentina 1.4
49080 14365 Colombia 1,2
70522 14366 U.S.A. 1,2
73507 14367 Antigua 1,2
cv Verano 14428 Venezuela 1,2, 3 4
61674 14368 Venezuela 1,2
cv Greenvale 14429 - 1,2
55856 18190 Brazil 1,2
cv Fitzroy 14431 - 1,2, 3 4
cv Seca 14430 - 1,2, 3 4
- 8115 - 1,2
- 8111 - 1,2
- 82105 - 1,2
38605 14457 Guatemala 1,2
67704(B) 14369 Equador 1,2
34904 14371 Brazil 1,2
85895 14359 Mexico 1,2
85899 14360 Mexico 1,2
86137 14361 Mexico 1.2
87469 14458 Mexico 1.2
87479 14362 Mexico 1,2
87484 14423 Mexico 1,2
87485 14424 Mexico 1,2
87487 14425 Mexico 1,2
91138 14363 Mexico 1,2
cv Haifa 19902 Israel 1,4
- (26) Kenya 1,3
47188 14433 Sudan 1,2
- (126) Kenya 1
ILCA 113 18172 Belize 1,4
60430 14372 - 1,2
- (130) Kenya 4
13671 14434 - 1,2 3
60442 14313 - 1,2
60452 Kenya 1,2 4
cv Red Caloona 14436 - 1,2 3 4
147
Origin Site
Vigna unguiculata
Vigna vexillata
Voandzeia subterranea
Zornia sp.
M66 Kenya 1,2,3,4
(122) Kenya 1,4
18174 Mali 4
(123) Kenya 1
148
CHARACTERISATION AND PRELIMINARY EVALUATION OF
ACCESSIONS OF ZORNIA SPECIES FROM THE ILCA
COLLECTION
J.J. Hakiza, J.R. Lazier and A.R. Sayers
ILCA, P.O.Box 5689, Addis Ababa, Ethiopia
Abstract
The genus, Zornia has a tropical distribution mainly in America
and Africa. A collection of Zornia species is used to provide
natural grazing and hay during different seasons.
The genus has maintained its taxonomic notoriety. The
significance of classification of the species in this genus would
possibly be in the application of various groupings to agronomic
problems, particularly selection and/or breeding for specific
purposes.
As a guide to the understanding of the range of variation
available in the Zornia accessions at ILCA, a study was made on
the plants already growing at Soddo, Ethiopia, based on
morphological and agronomic characters. Analysis of the
collected data revealed that with the use of morphological
characters, it was possible to obtain groupings belonging to
particular species as exemplified by Zornia setosa, Z.
glochidiata, Z. glabra and Z. latifolia. Based on agronomic
characters a different pattern was obtained, but this narrowed
the collection into two major groups. One group was
characterised by prostrate to decumbent accessions while the
other group comprised accessions which ranged from decumbent to
erect in their growth forms. Other smaller groups with
particular agronomic attributes could also be identified.
It seemed imperative, however, that further studies
involving more attributes than those used here would provide more
systematic groupings of accessions with succinct characteristics.
INTRODUCTION
The genus Zornia is indigenous to savannas and similar areas in
tropical and subtropical regions of the world (Mohlenbrock, 1961;
Mclvor and Bray, 1983) . The number of species in this genus is
variable but commonly estimated to be around 80 and distributed
predominantly in America and Africa.
Some Zornia species (Bogdan, 1977) are of importance in
providing natural grazing and hay, especially in Brazil and the
neighbouring countries and also in the drier parts of West
Africa. Species like Z. glochidiata (synonym Z. diphylla)
provides some scant forage during the wet season and early in the
season but does not survive long into the dry season (Skerman,
1977) .
The genebank at the International Livestock Centre for
Africa (ILCA) currently holds at least 234 accessions of Zornia
species some of which were grown for seed production and
149
agronomic evaluation during mid-1985 at Soddo in Sidamo,
Ethiopia. In order to provide some quick information about these
collections, a study of these accessions was superimposed on the
existing evaluation programme. Using 22 attributes on 161
accessions of Zornia species data were collected on field-grown
plants that had been growing for just over one year. The 22
attributes comprised both morphological and agronomic characters
(Table 1) whose choice was partially based on taxonomic character
descriptions of Zornia species (Mohlenbrock, 1961) and some of
the forage evaluation characters used at ILCA. The purpose of
this work was to try and group these accessions on the basis of
their similarities and dissimilarities using both morphological
and agronomic characters. This was to assist in picking out
closely related accessions and also to obtain those accessions of
good agronomic characters that could be subjected to further
evaluation while awaiting the results from the ongoing evaluation
trial.
The collection data were classified using principal
component analysis (PCA) , cluster analysis (CA) and K-means
clustering.
Table 1. Codes for the characters (descriptors and descriptor
states) recorded on the Zornia sp. growing at Soddo,
Ethiopia.
1~. Number of leaflets per leaf: 1 = 2 leaflets
2=4 leaflets
2. Flower size: 1 = flowers shorter than
the bracts
2 = flowers longer than or
equal to the bracts
3. Shape of the leaflets of the lowest leaves:
1 = linear; 2 = filiform; 3 = lanceolate;
4 - ovate; 5 = elliptic; 6 = elliptic/
lanceolate; 7 = oblanceolate; 8 - obovate
4. Size of the lowest leaflets: 1 = leaflets
more than twice as long as broad.
2 = leaflets not more than twice as long as broad.
5. Shape of upper leaflets: 1 = linear;
2 = filiform; 3 = lanceolate;
4 = ovate; 5 - elliptic;
6 = elliptic/lanceolate; 7 = oblanceolate;
8 = obovate
6. Flower colour:
1 = corolla yellow with no red marks
2 = corolla yellow with few red marks
3 = corolla yellow with many red marks
4 = corolla red
7. Flower (corolla) size: 1 = very small;
2 = small; 3 = small-medium; 4 » medium;
5 = medium-large; 6 = large.
150
8. Plant growth form: 1 = creeping;
2 = prostrate; 3 = semi-prostrate;
4 = decumbent; 5 = semi-erect;
6 = semi-erect/erect; 7 = erect
9. Bract and leaf pubescence:
1 = glabrous; 2 = pilose
10. Colour of the bracts: 1 = green; 2 = green
with yellow marks; 3 = green with red marks;
4 = purple
11. Internode length: 1 = very short; 2 = short;
3 = medium; 4 = long
12. Internode thickness: 1 = very slender;
2 = slender; 3 = medium; 4 = thick
13. Inflorescence length of lowest inflorescence:
1 = very short; 2 = short; 3 = medium; 4 =• long
14. Spread of the bracts on the inflorescence:
1 - very sparse; 2 = sparse; 3 = medium;
4 t large
15. Bract size: 1 = very small; 2 = small;
3 = medium; 4 = large
16. Leafiness: 1 = negligible; 2 = very poor;
3 = poor; 4 = fair; 5 = good; 6 - very good
17. Plot cover: 1 = negligible; 2 = very poor;
3 = poor; 4 = fair; 5 - good; 6 = very good
18. Inflorescence density: 1 = negligible;
2 - very poor; 3 - poor; 4 - fair; 5 = high;
6 = very high
19. Branching intensity: 1 = negligible;
2 = very poor; 3 - poor;
4 = fair; 5 - high; 6 = very high
20. Vigour = 1 = negligible; 2 = very poor;
3 » poor; 4 - fair; 5 = good; 6 = very good
21. Productivity: 1 = negligible; 2 = very low;
3 = low; 4 = medium; 5 - high; 6 = very high
22. Adaptability: 1 = negligible;
2 = very poor; 3 = poor; 4 = fair; 5 = good;
6 = very good
METHODOLOGY
A total of 161 Zornia accessions which were growing at Soddo in
Sidamo Province of Ethiopia were observed for characterisation
and preliminary evaluation. Soddo is at an elevation of 1850 m
151
with acid Nitosols of pH 5.5 to 6.0. Rainfall is above 700 mm
per year while the minimum and maximum temperatures are 12.6°C
and 24.5°C respectively. The total length of the growing season
is 245 days.
Each accession has been planted on unreplicated raised plots
each containing six plants. The plot size was 2 m by 1.5 m with
7 5 cm pathways across the slope and 1 m pathways or drains down
the slope.
Twenty-two characters - morphological and agronomic - that
could be recorded at the time were selected based on their
taxonomic applicability. Each of the 22 characters was recorded
once for each accession.
An HP 3000 computer and BMDP software were used for analysis
of the collected data. For PCA the BMDP4M program was employed
while the CA used the BMDP2M program to produce the vertical and
horizontal dendrograms. The K-means clustering used BMDPKM
program (Engelman, 1980; Frane and Hill, 1974).
Only components with Eigenvalues greater than unity were
retained in the principal component analysis, and they were
rotated by the Varimax procedure to aid interpretation.
The BMDPKM program for K-means clustering used the
hierarchical agglomerative method (Engelman, 1980) . All data
were standardised before clustering. The BMDPKM program
iteratively reallocates cases into clusters until each case is in
the cluster whose centre is closest to the case. This procedure
is followed for each number of clusters specified. In this study
the specified clusters were 8, 10 and 12. The criterion of
clustering was to minimise the maximum distance of a case from
the centre of the cluster to which the case belonged. The word
"case" here is synonymous to "accession".
In all the above analyses the characters were considered
under morphological, agronomic and morphological-agronomic
categories.
RESULTS
Principal Component Analysis
Before carrying out principal component analysis it was necessary
to consider the correlation matrix for the characters under
study. There existed a number of high correlations among the
characters which confirmed the justification for carrying out
principal component analysis on the data (Tables 2, 3 and 4).
The level of significance at 5% and 1% were respectively >0.15
and >0.211 for morphological characters; >0.156 and >0.204 for
agronomic characters; and >0.214 for morphological-agronomic
(all) characters.
On application of the principal component analysis to the
data, four factors (components) were found to have eigenvalues
greater than unit for morphological characters, three for
agronomic characters and four for morphological-agronomic
152
fe 2
in 1o r^ iH *
 
en iu til w
W h h PQ t-t
153
X
Ei
CQ o
Eh <* o
CXi eg oa
Q rH
2
> *
Eh *
CJ o o
O co o **
Q CN o 01
O r
« rH
, 0, * *
a * *
in
5 n
o
o o
*rf
OJ
n) O cn o f» r~
•rH U r « •
e H ^oi
u >
o
N CO **
*
* «*z
c H O H o a
•H Q rH
SB CM
o
o
o en o
CO *£
in ID
h Cm rH
0) s*p
j 03 * * * «
(S a * * * «
H w o CO CO VD in<d Q o o 0i in r~ in
X! 1-5 M o r- ^r l£> KD
o
j
2 rH
•H
H
e « * * * * *
o > * oK oK * *
c o o en ID r~ in r- ,
j U <J> o in *r^ CO CO 10 * *
u EH rH o *o r* in A t^ ^*
CD
(S 3
a.
rH M3 j*
U
in o
j CO * * * * *
rH IN
>4H w * * * * * O O
X
w o n ID *cr T m r~ A A
2 CO o in CO vo in o en
-H fe rH o *<* O CM m ** r>
u <
•p 3 oH oV> <*>i0 in r-\
E
c
s * * * * * «
rt, * •« + * * # >•
0 o o cn o 01 in in 0~i oi 0)oH fct O o I- t rH CO o in o o*p « rH o IN ro\ m IN IN CN n c
(S
(-i a rH i (S0
0 & orH
u IH
u O CO 01 o rH (N m « •rH
j H rH M H ts CM (N (N c
o en*H
• £ CO Z co CO > X
B
m K CO 2 2 Eh Eh HH
O w O W w K O n 0
(1) fc, 2 u a Q D D i-i
rH a bo, En J ^ O Q cu rH
A 3 < O fc. is 0 O 2 0(B
Pi
W 3 2 fr{ H « Q >Eoi J a. H CQ > A < cu
154
K
s
s
o. a
a
id B
c
U
h B,
O
N
in
n
a)
4J
0
ID
U
10
£l
O
e
o
c
o
tn
10
i
i-(
10
-■H
o
&
O
s
M
O
iH
u
-p
10
B
c
o
-H■p
10
8 S
t- N
M CM
O C0 r-l
- i■ r-
- « « «* * * *
io ■i i■ -
-T /I iD iO
- *
§ « *
i- n r» -M o yi
O CM u")
-n©
- o
 
rH
11)
M
M
O
O
0)
fH
O- O t-l CM
3 S
O (J
U, H H
« O U
2 2
WW > Jw
i"- Z i ■ !■<
w u a: u to
a a 5 d h
u. 3 S o <
s= 2 m « g
h n > bi <
155
characters. The four factors explained 71.6% of the total
variance for morphological characters with the eigenvalues of
3.249, 2.421, 1.658 and 1.373 respectively. Three factors
explained 84.97% of the total variance for agronomic characters
with their respective eigenvalues of 2.897, 2.606 and 1.295. For
the combined morphological-agronomic characters the four factors
accounted for 72.09% of the total variance with the eigenvalues
of 5.045, 1.932, 1.739 and 1.376 respectively. The sorted
rotated factor loadings for morphological, agronomic and
morphological-agronomic characters are respectively presented in
Tables 5, 6 and 7.
In Tables 5, 6 and 7 the factor loading matrix has been
rearranged so that the columns appear in decreasing order or
variance explained by factors. The rows have been rearranged so
that for each successive factor, loadings greater than .5000
appear first. Loadings less than .2500 have been replaced by
zero.
The interpretations attached to these factors for the
morphological characters are that the positive end of factor one
is characterised by attributes responsible for the higher number
of leaflets per leaf, obovate in shape, yellow flowers with few
or no red marks and green bracts. Factor two represents
characters associated with large bracts and corollas, the
corollas being yellow in colour. The inflorescences are compact,
having very many flowers and bracts. The third factor expresses
thick and long internodes and long inflorescences. The fourth
factor is associated with characters responsible for the flowers
that are longer than the bracts.
For the agronomic characters the positive end of factor one
is associated with characters responsible for bushy plants which
are highly productive and adaptable while factor two indicates
plants which are vigorous and leafy with high productivity.
Factor three is associated with erect plants having very few
leaves.
Table 5. Sorted rotated factor loadings for morphological
characters in Zornia sp.
Factor Factor Factor Factor
1 2 3 4-
NLEAFPLF 3 0.878 0.000 -0.273 0.000
SHAPELL 5 0.864 0.000 0.000 0.000
SIZELL 7 0.858 0.000 0.000 0.000
BRACTISCO 12 -0.663 0.000 0.000 0.000
FLOWERCO 8 -0.643 -0.422 0.000 -0.325
BRACTS ZE 17 0.000 0.878 0.000 0.000
FLOWERSZ 9 0.000 0.789 0.000 0.298
SPBRINFL 16 -0.297 0.782 0.000 0.000
ININDEL 13 0.000 0.000 0.769 0.000
ININDETH 14 0.000 0.261 0.740 0.000
FLOWERSZ 4 0.000 0.000 0.000 0.813
INFLLENG 15 0.000 0.000 0.564 -0.624
VP 3.249 2.421 1.658 1.373
156
Table 6. Sorted rotated factor loadings for agronomic characters
in Zornia sp.
INFLDENS 20
BRANDENS 21
PLOTCOVR 19
VIGOUR 22
ADAPTBTY 24
PRODUCTV 23
PLGRFORM 10
LEAFNESS 18
VP
Factor Factor Factor
1 2 3
.905 .000 .000
.880 .295 .000
.721 .532 .000
.259 .837 .000
.570 .751 .000
.600 .721 .000
.000 .000 .887
.000 .607 -.655
2.897 2.606 1.295
Table 7. Sorted rotated factor loadings for morphological-
agronomi c characters in Zornia sp.
Factor Factor Factor Factor
1 2 3 4
PRODUCTV 23 .910 .000 .000 .000
ADAPTBTY 24 .900 .000 .000 .000
PLOTCOVR 19 .869 .000 .000 .000
VIGOUR 22 .797 .000 .000 .000
BRANDENS 21 .733 .000 -.364 .000
INFLDENS 20 .632 .295 -.308 -.295
LEAFNESS 18 .613 -.567 .000 .000
FLOWERSZ 9 .538 .428 .000 .000
PLGRFORM 10 .000 .782 .000 .000
BRACTSCO 12 .000 .748 .000 .000
SIZEUL 7 .000 .000 .843 .000
SHAPELL 5 .000 .000 .817 .000
FLOWERSZ 4 .000 .000 .000 .892
BRACTSZE 17 .577 .000 .000 -.618
VP 5.045 1.932 1.739 1.376
Considering the morphological-agronomic characters combined,
the positive end of factor one has some resemblance to factor one
for agronomic characters. It is characterised by high
productivity, adaptability, plot cover, vigour, branching
intensity, inflorescence density, leafiness and large flowers
(corollas) and bracts. Factor two, on the other hand, is mainly
characterised by plants having reddish or purple bracts, semi-
erect to erect but scanty foliage and flowers of medium size.
Factor three is associated with plants having obovate leaflets
with low branching intensity and inflorescence density. Factor
four is mainly characterised by flowers that are larger than the
bracts.
157
The factor scores of individual accessions on the first four
factors were plotted against one another under morphological ,
agronomic and all characters combined (Figures 1, 2 and 3).
Clustering could be identified especially at the ends of the axes
with most of the accessions scattered randomly at the point of
origin.
Based on the morphological characters four clusters, (A, B,
C and D in Figure 1) on the plot of factor 1 versus factor 2 were
formed. Cluster A comprised Zornia setosa accessions, cluster B
was composed of Z. brasiliensis while cluster C comprised
accessions 11411 and 11424. These two accessions have not been
identified, but they could possibly belong to Z. lanata. Cluster
p was composed mainly of Z. glabra. Accessions belonging to Z.
diphylla and Z. pratensis tended to form a continuum, which is
not surprising since Z. diphylla is normally applied to a complex
of species, both perennial and annual (Bogdan, 1977). This
complex of species usually embraces such species as Z. latifolia,
Z. gracilis, Z. perforata, Z. pratensis and several others.
Groupings of accessions or clusters based on agronomic
characters were almost similar to those based on the combined
morphological-agronomic characters (Figures 2 and 3) . Other
species formed their own groups based on their stature and plant
growth forms. Based on the clustering pattern of these
accessions and the factor scores responsible for some characters,
particular groups with particular characteristics can be picked
out for further evaluation.
Cluster Analysis
As in the principal component analysis, the data were categorised
into morphological, agronomic and morphological-agronomic
characters. Using morphological characters, a minimum of ten
clusters was formed with some non-conformist groups which later
joined the rest of the clusters at higher amalgamation distances.
The amalgamation distances ranged from 0.000 to 10.330 thus
producing simple to complex fusions of accessions that produced a
very large vertical tree dendrogram. Table 8 shows the
composition of various clusters by accessions.
Some characteristic features of some clusters had a
definitive role for particular species. Cluster 1, for example,
was characterised by having four leaflets per leaf, obovate upper
and lower leaflets, small yellow flowers with few or no red
marks, very short and slender internodes and very short
inflorescences and bracts. This cluster comprised all accessions
of Z. setosa and one accession, ILCA 11476, of Z. brasiliensis.
Cluster 2 had accessions whose flowers were shorter than the
bracts. They had ovate to elliptic leaflets with a few
accessions having lanceolate leaflets. They also tended to have
medium to long inflorescences, medium to large spread of bracts
or inflorescences and medium sized bracts. Accessions in this
cluster belonged to Z. glochidiata and Z. orbiculata. Cluster
10, which was the largest, comprising 28 accessions, was
characterised by medium to large flowers with inflorescences of
medium length, leaflets that were more than twice as long as they
were broad and medium-to-large spread of bracts on inflorescences
158
Figure».PCAmorph---gica-charac er-.
Factor- ore-
x-Axi-i-factor1,yAxi-sfactor2
O21--- i2 1l i i i
-°—<©2~P\~*
 
DOOOO O
jrox■e
-1
■1-
 
■Axi-i-fXTtor1,y-Axisfa■orn 210-.^-_._2.
D
 
 
-1
-1-
Sca-eTfr-m-3-o+3.Fa - rore-g e er-han3rep o- d.Fac-o e--eh n3
arep-o--ed-3.
Fioure2.PCAaorjomiccharacOe s.
Factoro oreo
xoA ooofoct rl,y-Axosfacto2
jg.1qo\4
x-Axooofact r1,y-Axosj
 
3,z!o12jf>
 
Scaleisfrom-3Oo+3.FacO rscor soreaOerOh nepl O ds3Fac orsco slesa
ploOOedas-3.
Figure3.PCAa--charac-ers.
Factor- ores
x-A i-ifactor1,y-2
x-A i-isfactor1,yO
 
-O i0
O2-|
cO
Sca-eifrom-3-+3.Fac- r-core-grea-erh n3arp o-- doe
p-o--eda-3.
tOET'68EO
OO-OT-T6EXX'88EOEEOEOT'8ZZO90»E8EETT"0896'»SEOOOE6'8-OIS8EOT«TT
89E-' 9ETT-OOT
OOSO'8EEOT6ETT0ZS
80»O'9£EO-S■
OtO6EO'06ESETT
6VEO'ZOSO6T»OE*6EE-a O' 9EO8*
E6EO'ZO6OI8ETTESEO
ZTUT'08»O80SZSE-E-'S8ETTZZVOOUO
Z9EO'8 »O
80SO'06ETT»6EO
ZE»OSSETT
6Z»ZTOSEO' O6OI0Z88E6'09Z6-ZOTT- O
O>rOT'EZ»O-»6E>O
80»O' 060T89EO6
0E»O'STSO60»-E«ZO'8Z»O9t-6
llii-'8TOT>VEOZO»SO■T'£ZOZZS96EOEZO'OSOSO»9T*
8SEO'9tO8OOi
Z6EO'88EO» »ZOO
6T»O'9T-O
EZ»O' OS-
'8SEO9Z»
ZSEO'VOSO
S8»O'E9EO6E»9O
08EO' OOOO»696ETT'ZOO-E ■T-O»O
EZTTT' ZSOOS1-*
99E-'0-»OZOS O
»S«'0896
--O'68E6O SOZI»SO6
6E8E-'6I OIO8Z»O6 E6
6IEOI'6 8ET
'0896ZESOO»
'ZIEO,EO8Z»
9LOT
8ST6'09Z6ZEO8 ZOTO-OXT'LO
Z6EO'0iEO88£6-E0E I'8ZZ TOOOZEO6SES IO O
6E8ET'- OOO8 0T'UZO6SZOT'S8Z O-Z8 6T
EJiq3E-Ei[;iO1[u■-a8vE-^os puO-a-iquj■o)1jyc^bn0e-«t^--[e--|3--[-qd:i■wit- -■titXjOU*J3 --|3rOtia tuj:)--f BA3uT3-3O3t-oaioadtOOOK -z■-it ov'8a leT
»8»TT'ZE» T
Z6ETT'U TT8 »0ZS
6SO»OOTT
OO1;u'6SETT»S-OE»»
E»9O'ZS-TT8*»O9STTT' OTTO»-E
SOSTT'09» TS-»ESSETT' 6E TZ I-E
Z»ETT'0-E T
SOHT-T»-'6E TiO-
O*TT'S8» T0 ETE II
S■T»'SE»TT0060»ETSL
»ZUT«T'-»TT
SO»TT'8SE TV E9ZUT
-STT'Z9E T0OS
»ZSTT'6TS TUIT98E
TOSTTO»» T'66ET9 EE-TT'T06OIOOS T99»
OmiO
»TSTT'O» T.I6OV
OTVTT' SE TZ^6
6EETT'0 » T-O8VE
»8»TT'STS TT6E- T
»T»TT'ZT» T-09'ZS
9£ETT'TT»ZTOOS68E89ETT'ESE T» »u-E
SSSTT6EET »
6SZOT' iE TZOSES
8E»TT'O» T6iE<!-U80STTOOO' »E T8
SE»TT'U» T08»T
T9»TT8E'ZeTT
L»»O'0TSTTZ »T Z
Z»ETTOT-T'O■9»T
EZ»TT'»OET6EES
OOT'6Z»ZT
8-TT'96E T»SZZS
6Z»ZT
6S-T'ETS T0060»»E
SE»TT'SiEOTt Tm
08ETT'-*O-S8» TOSE
ZETTT'6-TSE T0 V
ia■e-■po-jui---Sv
--tj-w-eOpO-a iim--)i y
E-i-eJtiiD-[O ^a-^-^d-w}---[O
Op,3»»0O)-g*V\*l
imi■Zm'66ETT
8EETTO»ZO'T9EET0 T
LO-O'S»ETT6iZ-'OZZO»T-O--TTS'»9O»OEZ0TU-OO'1»ZZ»
L<i<rO ZS»U
99»TT -»TT
6TSTT
E-TT TO-T
9»»TT
OSO
'T»SOT 'TfTTT 'SOSTT
'OETT
'E9EO
'««TT
'-»TT '-m
-»E»TT
'6SSTT
'9O-
;es»Ti
'90STT
'00»TT
'0OSTT O»TT
'LS»-
'O»TT
ZZ»-
'aiXX O9>O- '9S»TT '6TEOI --s-i 'T0-T
'66ETT
'ES»TT
-»»TT
'9OTT
-»UP"»<OZ O'L»Z-
6«»U'*-II6SETT9OO
ZSOT'»S-T0S»8 1ESm'6**««»-S fTT
9S-TT'8»»TTET6E-
09-T'S*» TZL»»9-
TOTT'Z»» TOSOSZ9UT'-E T0T-ZLZO
98ETTOTSiE O
TOSH'68E6ZIE0T'L»Z-
»OET
98ETT'E-9OIV9E-99» TOSO'88ETTZ6 'LSZ-
-rOOOP"«9Z»TT>EZ0TET' OS06- S6
TZ»O' LT<r9Z-
S6ETT'Z»E TTT»ZE---
---'0OS-O
SWTOZS-T-6O-' O
-»OPOTZZUTE6ET'S8E T0iZT>l-8
c-i-e■^odOuj---Sv
--3333C2*ip( a-iiro--)-jy- i v-vip-[eo p8- [-^d--nJan n -
O-,3-O)gaiisi
which were either purple in colour or had red marks. Cluster 10
was composed of Z. glabra and Z. latifolia. Other clusters
tended to be an assortment of various species, namely Z.
diphylla, Z. latifolia, Z. pratensis and other unidentified
Zornia spp.
For clustering based on agronomic and morphological-
agronomic characters, ten clusters could be identified in each
case. However, based on agronomic characters the whole
population could be divided into two major units. One group
comprised accessions that ranged from creeping to decumbent and
were of small stature (clusters 1 to 7) . The second group was
characterised by being decumbent to erect, of fair to very good
leafiness, good to very good branching intensity, vigour,
productivity and plot cover. The first group comprised such
species as Z. setosa, Z. glochidiata, Z. diphylla and some
unidentified species while the second group comprised mainly Z.
glabra and Z. latifolia.
K-means Clustering
In the final K-means clustering analysis, divisions into 8, 10
and 12 clusters were specified under morphological, agronomic and
morphological-agronomic characters (Appendix 1) . This number was
arrived at on the basis of the previous observation which
indicated that during cluster analysis at least 10 clusters could
be identified while in the principal component analysis a minimum
of six clusters including the continuum of accessions scattered
at the point of origin could be identified.
The chaTacters or variables which were important in
determining clusters were those with relatively high "F-ratios"
of the between to the within cluster mean squares. The
characters which had high "F-ratios" (>30) were adaptability,
productivity, inflorescence density, branching intensity and plot
cover. These characters determined clusters under agronomic
characterisation while those under morphological characters were
flower size, size of the lowest leaflets and shape of the
leaflets of the lowest leaves. For morphological-agronomic
characters were flower size, size of the lowest leaflets and
shape of the leaflets of the lowest leaves. For morphological-
agronomic characters the important variables responsible for
clustering included the number of leaflets per leaf, flower
(corolla) size, size of the lowest leaflets, productivity and
adaptability.
Some clusters were either wholly or partially composed of
accessions belonging to particular or closely related species.
This was exemplified by the clusters under morphological
characters. Cluster 1 is, for instance, composed of accessions
belonging to Z. brasiliensis and Z. setosa only. Z. brasiliensis
is represented in the ILCA collection by accessions 11347 and
11476 only. Cluster 2 is dominated by Z. glabra and Z. latifolia
while cluster 3 is composed of the accessions 11411 and 11424
(possibly belonging to Z. lanata) . Cluster 6 comprises
accessions belonging to Z. glochidiata while cluster 7 has an
assortment of species including Z. diphylla, Z. pratensis and Z.
glochidiata.
165
DISCUSSION
Zornia is a genus of known notorious taxonomic difficulty. The
significance of classifications of the species in this genus
would possibly be in the application of various groupings to
agronomic problems, particularly selection and/or breeding for
specific purposes. The results of classification do not
establish but merely suggest the potential of a given accession.
Plant evaluation is expensive in terms of time and effort,
and it is known that certain species or genera are more likely to
be of value than others. Selection of the materials to be used
in the improvement of pasture germplasm depends on the available
information about the germplasm. Coupled with the information on
preliminary evaluation is the knowledge on characterisation of
these materials both of which help in grouping the germplasm on
the basis of morphological and agronomic characters.
Agronomic information is, of course, essential for an
understanding of the range of variation available in the
accessions as a guide to existing desirable combination of
characters and to the beneficial recombinations that may be
possible with genetic manipulation. A pure morphological or
morphological-agronomic classification is liable to be
insufficiently sensitive to agronomic heterogeneity. A pure
agronomic classification is too crude to permit identification,
owing to the limited numbers of attributes that can be devised
for collection. There is ample evidence that classification
provides the single most profitable technique for obtaining a
preliminary synoptic view of a complex system (Williams et al.,
1973) . In a study of this nature the groups of accessions may be
regarded as centres of variation and allow the information to be
presented in a more simple form than enumerating the properties
of all individuals. So if a collection like this Zornia one is
too large to permit succinct description, a classification in
groups of known mean characteristics will still be necessary if
the information is to be summarised and communicated to others.
Under morphological ordination and classification,
accessions belonging to particular species or closely related
species tended to form groups. Zornia setosa and Z. brasiliensis
formed their own group; so did Z. glabra and Z. latifolia (with a
few exceptions) ; and Z. glochidiata. A continuum of accessions
belonging to Z. diphylla, Z. pratensis, some Z. latifolia and
other unidentified Zornia spp. was formed. This was consistently
so under the three types of analyses (PCA, CA and K-means
clustering) . Since morphological classification is basically
very valuable for identification and description, such
classification has provided a useful tool for that purpose.
Whereas a mixed morphological-agronomic classification is useful
for description, it also provides some guide to the agronomic
potentiality of some of the groups. The groupings in this study
tended to conform to this observation. Agronomic classification
tended to differ from the other two. Here groupings were purely
based on agronomic similarities. However, the clusters
comprising some accessions of Z. glabra and Z. latifolia were
similar in composition for the three categories of
classification. Accessions of z. diphylla, Z. pratensis, Z.
setosa and Z. glochidiata tended to exhibit characteristics of
 
 
166
poor agronomic potential while Z. glabra and Z. latifolia
generally showed better agronomic potentiality. For example, in
principal component analysis and cluster analysis accessions of
Z. glabra, Z. latifolia, some Z. glochidiata and some
unidentified Zornia spp. showed a high expressivity of branching
intensity, inflorescence density, plot cover, productivity and
adaptability. Good performance in a spaced-plant trial, however,
does not guarantee good performance in a grazed pasture. Grazing
of Z. brasilensis accessions, for example, had to be terminated
at any point in time (CIAT, 1983-1985) , at Guilichao, CIAT,
although the preliminary evaluation of these materials had been
promising.
However, because it is often uneconomic or even
impracticable to carry out field trials of all available
material, the function of methods such as those outlined in this
paper is to select those accessions most likely to justify more
intensive trials. Groupings of accessions and/or species have
been obtained in this study and some of their characteristics
spelt out. Choice of particular groupings will depend upon the
use to which these materials are to be put. Groups comprising Z.
setosa, Z. glabra, Z. latifolia and Z. glochidiata including
their characteristics could give a guideline as to what use these
selections could be put should they be advanced to further
evaluation trials. The numerical methods of groupings are of
particular value in the tropical genera which may require
taxonomic revision like Zornia. Groupings defined on the basis
of overall similarities of agronomic importance and morphology
could well be of more value for practical purposes than
stringently defined species and could be used equally well for
communication purposes.
ACKNOWLEDGEMENTS
We wish to acknowledge the technical assistance of Mr. James
Ochang in growing and looking after the plants at Soddo, Ethiopia
and members of the Forage Agronomy Group (FLAG) who provided
transport and other necessary assistance.
One of the authors (J.J. Hakiza) wishes to acknowledge the
financial support and authorisation to carry out the study by the
International Board for Plant Genetic Resources (IBPGR) and the
International Livestock Centre for Africa (ILCA) . He is equally
grateful to IBPGR for allowing him to attend the PANESA workshop
at Arusha, Tanzania.
REFERENCES
Bogdan, A.V. 1977. Tropical Pasture and Fodder Plants. Longman
Group Limited, London.
CIAT, Tropical Pastures Programme, 1983-1985. Annual Reports,
Cali, Colombia.
Engelman, Laszlo. 1980. Annotated computer output for BMDPKM/K-
means clustering. BMDP Technical Report No. 71. University
of California, Los Angeles, U.S.A.
167
Frane, J.M. and Hill, M.A. 1974. Annotated computer output for
factor analysis: A supplement to the write up for computer
program BMDP4M. BMDP Technical Report No. 8. University of
California, Los Angeles, U.S.A.
Mclvor, J.G. and Bray, R.A. (eds) . 1983. Genetic resources of
forage plants. CSIRO, Melbourne, Australia.
Mohlenbrock, R.J. 1961. A monograph of the leguminous genus
Zornia. Webbia 16:1-141.
Skerman, P.J. 1977. Tropical forage legumes. FAO Plant
Production and Protection series No. 2. FAO, Rome.
Williams, W.T., Edye, L.A. , Burt, R.L. , and Grof, B. 1973. The
use of ordination techniques in the preliminary evaluation
of Stylosanthes accessions. Aust. J. Agric. Res. 24:715-
731.
168
tu tu
O O
01 tfl
c ai
O-H
■M u
m o
Bl CJ<
0) 0)
0 -p
0 10
id 0
>iT1
XI C
10
c
O 10
-H M
-P 0)
3-P
-P w
■H 3
-PrH
0) 0
c
O tH
U O
10- -
- 3
10 O
H
« o<
-H
S3
O O
N-H
u
U 10
O >iu
u
cr' o
C t-l■H
u to
0) k
-P a)
01 4J
3 10
■H 3
O rH
0 -
tM n
O -P ^
B 01
01 01 -P
c u 0
10 0) 10
01 iU n
S"n 10
i -h x:
U5-0 0
c
01
<
en u-i fi
H H "-1 H H ,H
^ «-l
* ~ - * -
^ ^ n
ra O - o -H rt -- -- H -
o-i
to
c- ^ - c- 0" -
o■ n■
^
£ to £ -
13 CJ - -■"■ H H H H rt
--> ■- o -O
I O N iO ■»
■ - in » Ifl a ■i i
■■»n4*4'44*0-«-H4Ifl
i - o ■i -
i in fi e■ 4 m Is- Oi
i ■» - h n * m ri
r-i ■* Mt tO iH
) CM in r- CO ■
O O O 9i O O
n n « 9i n - n
- u"i r- in ■> «- in
tH O- r-i - c> 0- -
i O O O 0"i - -
i - o Oi -
1 tC ■i BO O 00
i r- i- ■n « m
14 MS S N
-i -i m 0--
O r-l r-l r-l r-l
IO--O-liMi0!yiH
) i- r"- CO ■» ci p- c^
\C CO A 4 --- ~
i0it0oN0--CO,01. J~ -I ■Jf ■-" X i/l -J u", l
■ ■» in i■ i- -» in in i
■ 4 CI a ib - CI 4 i
169
-O-
0060»LZT-
0EETT9E»TTOETTIE»TT
ETSTT
6EETTZO60T6Z>;ZT
'SEUTZ---
'8T»TT
E0UT
96ETT
'6E»TT
EEETTS8»TT
O»TT
08ETT
S8ETT
Oe-O»EO
mTT
0»ETT
Ot-
006OI
EZ»TT
0EETT9E-T
SUTT
IE-T
ETSTT
6EETTZO60T6Z»ZT
SE»TT'Z--
'8T»TT
'E0»TT
'96ETT'6tO
'SZETT'S8»TT'Xe»Xx
'08EO
'S8ETTOSETT'TKTT
ETSTT-SE» T
0»EO'ZZSTT
OO»
006OI
TZi-
0EETT9EYO
'8T»TT
E0»TT
'96EO
'6E»TT
SEETT
SUTT'S8» TTtOO8»TTETSTT'08EO
6OETT'S8E TZO60T'»SE T6Z»ZT'T E T
SOSTTO S T'ZOSZES0T'*ej-0T68£6
ZOSTT68/6
SOSTT
'£ES0»
'ETUT
ET-T'0TS TTOSTT'9 -TSOTT'98E T
6TEOI'09»TT
66ETT'ZIE0T
ET»TTTOSTTSOTT
6IEOI66ETT
0TSTT
'--T
98ETT
09»TT
'ZIE0T
0TSTT' OS T
-»TT'SSOT
OEET'98ETT6TEOT'09» T
66ETT'ZIE0T
ZTSTTZTUT
6E»TT
'SO»TT
0060T
'ETSTT
ZTSTTSO»ZT»TT'0060T
08ETT'ZVE T
6E»TT
08ETT
ETSTT'ZUTT
UETT
50»TT
SE»TTZZSTT
8T»TT
E6ETT
»ZSTT
ZZUT
96ETT
6E1TT
SZETTSUTTTE»TTETSTT
6ZETTZO60T
ZT»TT
-ZTZO
ZTSTT
'8ZUT'6TSTT
'0UTT
-ZZ»TT
'0060T
-ST»TT'ZO»TT
'-STT
'0ZETT^ETTT'0OSTT
'S8»TT'TSO
'08ETT
'S8ETT'ZVETT
'»SETT
'T»ETT
ZT
-»TT
6EETT
Z6ETT
9T»TT
ZE-T
8SETT
t-iaiae-eqo-[-[y
d--O
'9ZO
'E9ETT
'6TOT
'-O
'EZ»TT
'SO-T
0T
-»TT
6EETT
Z6ETT
9T»TT
ZE»TT
8SETT
OE»TT
6»ETT
d--Q
'9Z»TT
'E9ETT
'6OTT
'TZ»TT 'EZ»TT 'SO»TT 'ZE»TT 'EZ»TT
-»TT
6EETT
ZSETT TSETT
»TSTT 6T»TT
8UTT
9T»TT
Z6ETT
8
'9Z»TT
'E9ETT 'ZOSTT
'0T»TT
'80STT '60TTT
'»6ETT
'TZ»TT
'8E»TT
d--^
6SETT
ZZETT
6EETT
SSETT Z9»TT TT»TT
ZT
taa^oBJei--il d--Q
'991TT
'6S1TT
'T«0T
'9ZTTT
'EZ»TT
'90STT
0T
6SETT
ZZETT
6EETT
SSETT Z9»TT TT»TT
jj----v
d--
'99»TT
'6OTT
'T»50T
'O»TT
'EZ»TT
'90STT
6SETT
ZiETT
6EETT
SSETT Z9»TT TT-T
8
'99»TT
'6SUT
'T»50T
'9E»TT
'EZ»TT
'905TT
---B-^n.
d-J3
»8»TT
E9ETT
6T»TT
TE»TT Z9»TT
ZT
d-ag
'9Z»TT
'-»TT
'ZO5TT
'E6ETT
'EZ»TT
■[1I3T8---qd-w
0T
'JS»TT E9ETT
6EETT
ZSETT TSETT
»TSTT
»6ETT
TE»TT
0TSTT6T»TT
TTSTT
d--o
'9Z9TT
'-»TT
'ZE»TT 'ZOSTT
'0T»TT
'80STT
'0E'ITT
'Z6ETT
'8E»TT
'60»TT
'EZ»TT
9T»TT
»8»TT
E9ETT
6EETT
ZSETT TSETT
»TSTT 0TSTT
6OTT »6ETT
TTSTT
8
'9Z-T
'-»TT
'ZE»TT 'ZOSTT
'0T»TT
'80STT
'0E-T
'Z6ETT
-8E»TT
'60»TT
'EZO
9T»TT
d-JO
Oa---^-
-qiu-u -
ZT
OSE6
ZTE0T
0896 EESOT
6E8ET
-aa-e-eqoTT»
d-jzg '8z- 'T»SOT 'OiETT
'68Z6 '6IEOI
ZZ»TT
0»
TSE6
ZTE0T
0896 i-OT
6E8ET
d-ag
'8Z»TT
'T»SOT
'OTT
'68^6
'6TEOI
ZZ»TT
»SE6 ZTE0T
0896
iESOT 6E8ET
99OT
/Oi,TT
-ETT
E»-»
»E»TT
T9EET
66E-
TOSTT
Z»ETT
90STT
T06OI
0TSTT
8
'8Z»TT
'T«0T OiETT '68£6
'6TEOT 'O»TT 'TT»ZT 'ET-TT
'98ETT
'ZT»TT
■-TT 'E-TT
'6TSTT '0OSTT
'Z9ETT
'UITT
'ET»TTd--
T8»TT S8»TT OTVTT
TTSTT
99ETT
SO»TT EZ»TT
9Z»TT
8Z»TT
SUTT
6SZOT
STSTT
»EZOT
0»ETT
iOSTT
-ETT
SE»TT ESETT Oll-
ZZO
ZT
-J9»D8JF1J33-tUIOUO-3v
d--o 'ESETT 'ZEO
'VOSTT
'ZO»TT
'»T»TT
'Z-TT 'EO»Ti
'{8ETT
'6«TT
'OETT
'ZSETT
'80STT
'6-O
'6EOET
'8«TT
'TE»TT
'08-O '.TT
'6ZETT 8E»TT
0T
9Z»TT
ZE»TT
»OSTT
ZO»TT
TTSTT
99ETT
Z6ETT SO»TT E0»TT E8ETT
»8»TT
T6ETT
8Z»TT
SEETT
6SZOT
STSTT
»EZOT
0VETT
nsTT 0EETT SE»TT ESETT UETT
d--o
'KETT 'O»TT
'S8»TT ei-TT e-TT '»iO 'ZZSST 'UOT
'-O
'UETT
'eejiO
'8EETT
'-ETT
'OETT
'ZSETT
'80STT
'60»TT
'6E8ET
'8»ETT
'TE»TT
'08-O
'U»TT
'6£ETT 8E»TT
9Z»TT
ZE»TT
»OSTT
ZO»TT
TTSTT
99ETT
Z6ETT SO»TT EO»TT E8ETT
»8»TT
T6ETT
8Z»TT
SiETT
6SZO»
STSTT OEZOT
0»ETT
EISTT
-ETT
SE»TT ESETT iZ»TT
8E»TT
8
'OSETT 'T8»TT 'S8»TT
'0T»TT '0ZSTT '»TUT
'ZZSST 'LlO-
'8UIT
'TiETT 'EZ»TT
'8EETT
'6»ETT
'S»ETT 'ZSETT
'80STT
'-»TT
'6E8ET
'8»ETT
'TE»TT
'08»TT
'Hi- '6EETT
'^ETT
--ioB-e^-
d--
»S£6 ZTE0T
0896
£ES0T 6E8ET
ZT
d--
'8Z»TT
'T»SOT
'OTT '68E6
'6TEOT
ZZ»TT
■jeoTa-i
0»
»SZ6 ZTE0T
0896
£ESOI 6C8ET
■qd-H
d--ig
'8Z»TT
'T»SOT
'WETT '6OE6 '6TEOI
ZZ»TT
OSE6
ZTE0T
0896 iESOT
6E8ET
LSETT
99TTT
OEZOT
88ETT 80TTT
T060T Z6ETT EO»TT OOET
ZO»TT
ET90T Z9ETT OEZOT
METT
8
'OZ»TT9 'T»S0T 'OiETT '6-6
'6TEOT 'O»TTL '8TSTT
O9ETT OOSTT
'66ETT
'T9EET
'-T»TT
'ETVTT 'ZZ»TT V8ETT
'98ETT
'TOSTT
'08»TT
'8SETT
333511»0
d--Q-qum-
'
LT»O ZOOZO
-«-
'TT»ZT 'ET-T OE»TT ESETT
TTUT iT»O EO0T
»E»TT OEET
66ETT
TOSTT
Z»ETT
-STT
T060T
0TSTT
'99»TT
'/OSTT
'98ETT
'ZT»TT
'«STT '/9ETT
'6TSTT '0OSTT 'OETT
's-»- 'EWTT
O»TT
9E»TT'8S 6
//--
OE»TT
ZO60T
ETSTT OZE0T
»EZ0T
9E»TT 9iETT
-/6
89ETT 88ETT
0896
-»TT
//Z0T
EOE0T
'»S/6 '8ST6
'S8ETT
'68ETT
'O»ETT
'06ETT '»6ETT
'8ZZOT
'TSETT
'»TSTT '8H/6
'09/6
'80OT
'8SZOT /Z»TT
8TSTT
/9ETT
/OSTT
66ETT
T9EET
/T»TT
ETUT
»OET
ZO»TT EO0T Z9ETT
»EZ0T
»»ETT
'/SETT
'99»TT
'»EZOT
'88ETT
'80»TT
'T060T 'Z6ETT
'EO»TT
'/8ETT '98ETT
'TOSTT
'08»TT
'8SETT OUT
8TSTT
/9ETT
/OSTT
66ETT
T9EET LT»- STOT
»OET
ZO»TT EO0T Z9ETT
»EZ0T
»»ETT
'/SETT
'99»TT
OEZ0T
'88ETT
'80»TT
'T060T 'Z6ETT 'E0»TT
'/8ETT '98ETT
'TOSTT
'08»TT
'8OTT
O»TT
ZT
T6ETT OZSTT
»8»TT
9iETT 99ETT
fT»TT
»EZ0T
89ETT
80UT
/SETT
EOUT
8TSTT
d-- O»ETT
'8EETT
'Z6ETT 'TiETT
'90»TT
'ZO>O-
'8/»TT
OOSTT
'88ETT
OOET 'E6ETT
'/SETT
0»
T6ETT
OSTT
»8»TT
9iETT 99ETT
»T»TT »EZ0T
89ETT
80»TT
/8ETT
EO»TT
-STT
d-- O»ETT
'8EETT
'Z6ETT 'T/STT
'-»TT
'ZO»TT
'8/»TT
OOSTT
'88ETT
OOET 'E6E-T
'/SETT
--e- 8SETT
OZSTT -s--
9iETT 99EO
»T»TT
SO»TT
»OSTT
88ETT »OET
E6ETT
/SETT
8
'68ETT
'T6ETT
'8EETT
'Z6ETT 'TiETT '9- 'ZO<lO
'8/»TT
OEZ0T
'89ETT
'80»TT
'/se- 'EO»TT 8TSTTd-^o
8ES0T'SOETO
8ES0TSOE
T»ETTOOEll*O'0060-E6ETT'EOE0TOSZ0T'//Z0TSO»TT'- TT09/6'08/6
88/6' 8ETT
»TSTT'89E T
TSETT'O/68E
8ZZOT'9iE TO ST
»6ETT'9 » T6»06ETTOEZ0T-»TT
OETT'0ZE0T»T»T
68ETT'ETS T99
S8ETT'ZO60TT6
8ST6'0E»TT0 »T»S/6'6Z»ZTE»TT
9iETT89ETT
TTTZTESETT
»OSTT
8»ETT
STSTT
06ETT68ETT
8EETT
OZSTTe'lETT
90»TT
O»O
99ETT
T6ETT
00»TT
»E»TT
'9iETT'89ETT
'TT'lZT'ESETT
'»OSTT
'8»ETT
'STSTT
'06ETT'68ETT
OEETT'9iE TOZSTT'89E T
6»ETT
-»TT
<1T<OO
99ETT
T6ETT
TT»ZTESETT
»OSTT
8»ETT
STSTT
OO-T'06E T
»E»TT'68E T
ZTd--
0Td--
8d-ij)
ZTd-ig
0Td--
gd---qiun
-at--[O
--333«eqo[jy
-ia■E..Ooj■---v
--]3wci-[w-iS-j-nd-w
Z-TT Z-TT
-ni
6SETT
-VTT
E»*TT
TOTT
»OO
»-O}
8OO
OETT
8OU
0-TT
SO-TT
6OTT 98ETT
T9EET
66ETT
TOSTT Z9ETT S- EI»TT I9OI
'-SETT
'--ETT
'ZZETT
'E»»TT 'O»TT
eO»n
'6¥»TT
'ZOTT
'-»TT
'E»9TT
'SSt)E OOTT
'TTT
'9On
'»*»TT
OOSTT
'»ZSTT
O9ETT
'6TSTT
'OO-TT
'90-TT
'T06OI
'0TSTT
ZO6OI
EI-TT OZEOO
»£ZOT »6ETT
8ZZOT
TSETT
»TSTT
'0E»TT
'S0ETT
'68ETT 'miO '06ETT 'OETT
'O6E6
'89ETT
-E»T
UITT'»ZET
E0»TTOZ»
90-TT' iEO
9E»TT'SiET
00SI'OCT
T8»TT'6EEIS8ETT'HI
»8-TT
8EETT'.,T
8E»TT'TZ»O
Z6EIO'0ZSTTZ6ETT'0ZS T
»-ETT
S8»TT ii»TT
0»ETT mix 6EETT 6EETT OI»TT
OOS»
60»TT
OI-TT
»6ET» 06ETT
'6Z»ZT
'TE»TT
'96ETT
'ZZSTT
'8-*TT
'ZE»TT 'ZE»TT 'ZSETT 'ISETT 'OTSTT
'8OTT
'0E»TT
'iisn
»SETT ZO6OI
6£ETT
IE-T -tETT -iETT
-ETT 90STT
iO»TT ST»TT iE»TT
0»ETT
6I-TT UETT
'6Z»ZT
'ZOTT 'S8ETT 'O8»U 'S8»TT
'0OSTT
'9E»TT '96ETT
OZ»TT
'»Z-TT
'E6ETT 'ZZSTT 'SE»TT
'-»TT
UETT
ZT
ZSETT TSETT
»TSTT 8E»U
ZE»TT
-ETT
TT-TT
08»TT
8»ETT
--g
'6EETT
'0HTT '80STT
'60»TT
OETT '0EUI '»EZOT
'06ETT
'-ISTT
0»
ZSETT T-ETT
-n
8E»TT
ZEYTT
-ETT
TTSTT
OOTT
8»ETT 68ETT
O-g
'6EETT
eO-
'80-TT '60»TT
'»6ETT '0E»TT
'»EZOT
'06ETT
'STSU
'68ETT
ZT
OO-TT
9-O 6»»TT
ZS»TT
-»TT
I06OI OS»TT
-ETT
»-»TT
-I»TT
T9EEI
rf-j;) O9£TT
'0OTT
'00-O
'TOTT
'»ZSTT
'8»*TT
'{OTT
'6T-TT
'ZZ»TT
't9TT 'E»»TT
0»
OOSTT
-»TT -»TT
ZOTT EOTT I06OI
8OTT -ETT »OTT
-T»TT
I9EET
d--g 'Z9ETT
'0-»TT '00»U
'IS»TT
'U-TT
'8»»TT
'Z-»TT
'6TSTT
'ZZ»TT 'E-TT 'E»»TT
ed-ag
gdm9I-
Z\d--g
0»d---
--a-Jaqtu--
tjjijcjtqaH¥
tj»n■.qo■iox>u■j8v
tjjpFjeqj^v3-|-■10i]dj■H
CHARACTERISATION AND EVALUATION OF FORAGE LEGUMES IN ETHIOPIA:
PRELIMINARY EXAMINATION OF VARIATION BETWEEN
ACCESSIONS OF STYLOSANTHES FRUTICOSA (RETZ.) ALSTON
J.J. Hakiza, J.R. Lazier and A.R. Sayers
ILCA, P.O. Box 5689, Addis Ababa, Ethiopia
Abstract
A study was carried out on accessions undergoing agronomic
evaluation and seed multiplication. A total of 93 accessions
grown as spaced plants were involved and 18 simple morphological
and agronomic data recorded. Principal component analysis,
cluster analysis and K-means clustering were used on the data.
It was shown that although there existed a high level of
agronomic heterogeneity and genetic variability among the
accessions, more collections were inevitable to widen the genetic
base of S. fruticosa and cover areas particularly in Africa where
this species is prevalent. It was also realised that a more
systematic study of these accessions for fairly long periods
might give a better understanding of these materials.
Characteristics of various groupings are discussed.
INTRODUCTION
The importance of a germplasm collection to those utilising it is
very dependent on the availability of accurate descriptions of
the accessions and on the taxonomic identification of the
germplasm (Engels, 1986) . Where there is a substantial number of
accessions of one species or genus within the collection, efforts
should be made to classify these into like groups. The objective
of this classification is to allow the reduction of the number of
accessions that have to be grown in the next phase.
The genus Stylosanthes comprises 30 to 40 species (Burt et
al., 1983; Stace and Edye, 1984), many of which are either
agronomically unattractive or are of rare occurrence. By the
beginning of 1984 the world collection of Stylosanthes spp. and
S. fruticosa were as shown in Table 1. Stylosanthes species are
adapted to quite different ecological situations and range from
annuals to perennials. They also have different agronomic forms.
This exemplifies the high range of variability within the genus.
S. fruticosa is regarded as an excellent forage legume in
some parts of Africa. Its deep and strong tap root system makes
it resistant to grazing. The prostrate forms have some value in
protecting the soil against erosion. S. fruticosa is a species
which has received less attention than its South American allies
S. guianensis and S. humilis, particularly in Australia. It must
be presumed to have considerable potential as a forage legume in
regions with a single dry season. It is distributed in many
countries in Africa as seen in Figure 1.
The latitude and longitude ranges within the current S.
fruticosa collections were respectively between 12°00'N and
13°24'N and 02°19'E and 09°03'E for Niger, and for Ethiopia
174
Table 1. World collection of Stylosanthes fruticosa and
Stylosanthes spp.
Institution1' S. fruticosa All Stylosanthes spp.
CIAT 10 2241
CSIRO 88 1451
EMBRAPA 0 1296
EPAMIG 1 62 6
University of Florida 23 838
ILCA* 164 858
17 CIAT - Centro International de Agricultura Tropical,
Colombia.
CSIRO = Commonwealth Scientific and Industrial Research
Organization, Australia.
EMBRAPA - Empresa Brasileira de Pesquisa Agropecuaria,
Brazil.
EPAMIG - Empresa de Pesquisa Agropecuaria de Minas Gerais,
Brazil.
ILCA - International Livestock Centre for Africa,
Ethiopia.
OThe figures for ILCA are as of 31/12/86, and these were not
included in the original table.
Source: Stace and Edye (1984) .
between 04°54'N and 07°ll'N and 37°22'E and 39°30'E. Williams et
al (1984), Lazier (1984), Skerman (1977), Thulin (1983) and IBPGR
(1984) show that S. fruticosa is distributed in at least 23
African countries.
Despite this wide distribution of S. fruticosa, especially
in Africa, the current collection at the International Livestock
Centre for Africa (ILCA) is mainly from Ethiopia and Niger. Two
accessions were acquired from the Commonwealth Scientific and
Industrial Research Organization (CSIRO) , Australia. In order to
search for materials of agronomic potential, it is essential to
collect sufficiently variable germplasm from many diverse
environments and evaluate their forage potential .
Some morphological and agronomic characters were studied for
both identification purposes and assessment of agronomic value.
METHODOLOGY
Of the 164 accessions of S. fruticosa in the genebank at ILCA,
102 accessions were planted for agronomic evaluation and seed
multiplication at Soddo in Sidamo, Ethiopia during April, 1985.
The geographical and climatic description of Soddo and the
planting procedure were as given in the Zornia paper in these
proceedings .
These S. fruticosa accessions had been collected mainly from
Ethiopia and Niger (Appendix 1) . Those from Ethiopia were
175
Figure 1. Distribution of Stylosanthes fruticosa in Africa.
 
_20f l-
S.fruticoso («»o lodio, Sri Lonko)|
\ 1 S.ertcto \
I I | S.suborbiculoto
Source: Williams et al . (1984).
predominantly from Wolayta in Sidamo. In order to assess the
existence of variations between the accessions that were
undergoing agronomic evaluation, a study was superimposed on this
trial. Observations were made when the plants were about 18
months old at the end of the rainy season. Eighteen characters
(morphological and agronomic) on 93 accessions were selected
based on the practical and field considerations at that
particular time. These are given in Table 2 with their character
state codes.
Principal component analysis and cluster analyses were done
on the HP 3000 computer as described in the Zornia paper in these
proceedings.
176
Table 2. Morphological and agronomic characters taken at Soddo,
Ethiopia, on Stylosanthes fruticosa and their character
states.
Tl Plant growth form:
1 = erect; 2 = semi-erect; 3 - ascendant;
4 = ascendant/decumbent; 5 = decumbent,
6 = decumbent/prostrate; 7 » prostrate
2 . Plant height at centre of the plant excluding the central
shoot (cm)
3. Plant height to highest point of central shoot (cm)
4. Canopy spread at the widest point (cm)
5. Canopy spread at the narrowest point (cm)
6. Mean number of branches originating from the main stem of
all plants at or below 5 cm
7. Length of the longest primary branch (cm)
8 . Average length of the spike (mm)
9. Adaptability:
1 — negligible; 2 - very poor; 3 — poor;
4 = fair; 5 - good; 6 - very good
10. Length of central leaflet of the leaf nearest to the spike
(mm)
11. Width of the central leaflet of the leaf nearest to the
spike (mm)
12. Central leaflet length/width ratio
13. Distance between base of centre leaflet to junction point of
the two outer leaflets:
1 = <1 mm; 2 = <2 mm; 3 = >2-<mm; 4 - >3-<4 mm;
5 = >4-<5 mm
14. Stipule sheath length (mm)
15. Plot cover:
1 = negligible; 2 = very poor; 3 = poor; 4 = fair;
5 — good; 6 «» very good
16. Height or distance of shoot tips from the ground level (mm)
17. Expressivity of the central (main) shoot:
1 = negligible; 2 = very poor; 3 = poor;
4 = fair; 5 = good; 6 = very good
18. Petiole length of the leaf nearest to the spike (mm)
177
RESULTS
Principal Component Analysis
The correlation matrix for the characters under study (Table 3)
shows a number of high correlations. Plant growth form was most
highly correlated with the height or distance of shoot tips from
the ground but had otherwise low correlations with other
characters. Other high correlations included canopy spread at
the widest point with the length of the longest primary branch;
canopy spread at the narrowest point and adaptability; and
adaptability with plot cover. These correlations indicate that
principal component analysis may provide a useful summary of the
variability in the data.
When principal component analysis was applied to the data,
four components or factors were found to have eigenvalues greater
than unity. These four factors explained 71.6% of the total
variance with 4.176% eigenvalues of 4.176, 2.606, 1.920 and 1.318
respectively. Based on the sorted rotated factor loading (Table
4) the following interpretations were attached to the factors.
The positive end of factor one is associated with the
attributes responsible for characters of agronomic interest such
as plants which are highly branched and quite bushy. Factor two
represents characters associated with large leaflets while factor
three expresses the high plant growth structure. The fourth
factor is concerned with high leaflet length to width ratio.
The factor scores of the individual accessions on the first
three factors were plotted against each other. Some clustering
could be identified, although most of the accessions appeared to
be scattered randomly about the origin. On the plot of factor 1
versus factor 2, for example (Figures 2 and 3), about six
clusters could be identified.
Cluster 1 was composed of six accessions. These were
characterised by long and wide central leaflet of the leaf
nearest to the spike, petioles of the leaflet nearest to the
spike being long and with distance between base of centre leaflet
to junction point of the outer leaflets being between 1 and 2 mm.
Cluster 2 comprises ten accessions. These accessions have
very good agronomic characters tending to those of S. scabra in
growth form. They are quite bushy with a high branching pattern
and very high adaptability and plot cover.
Cluster 3 is a uni-membered cluster composed of the
accession 10539 whose central leaflet is very narrow and rather
short. The accession, however, has similar agronomic features to
those of the accessions in cluster 2, especially its wide canopy
spread. This accession happens to be the one collected from the
furthest east in the collection. It was collected from near
Negele Borana in Sidamo, Ethiopia along the longitude of 39°30'E.
Cluster 4 is composed of ten accessions. Their main
characteristic features are poor to fair adaptability and
extremely poor to very poor plot cover. Exceptional cases are
the accessions 13932, 13929 and 13973. However, 13973 has
178
O**)XO"*a,03oo:'bd boOJToSisj_aAai
0O'X3HI3 l
03=HI3*H
o8LOTTT'_6'*L
00OT680o* TO
OOjTT'
T»T'-OjV¥*T0VEO->lLl'jO
/LOZj'j1
00jT¥»'*3'-OjL**'*8
O0oT*¥805o-*SO
OOT»»8LO
OOX
00jT
OX'
TEj- Lilo
0j'
**8f>f 8LX'
ESj
00jT
¥¥OL OT' j0"
SL0o
8O'
*»*Z?'
Tj' <lL0o
OOjT
»*j*"
OT3T
0*O-
bT'-
T»T'
6SO
OOo
ZIj- 830-
OOjT
**OL **jj
OZ8O
ETTo Oeo' L5O SjT' *»SSSo ono
+OZo
OOjT ¥¥S6j ¥¥38*'
OZZj OOO
33O
jO- OO
0O'
O6EL
TTT' Lj'
OO
EO'
¥¥»To
Zjj
0O'
O»8o
80T'
8LO
Z.0O
6*T'
O<lTj O6Lj
TO'
T¥L33'
OO
EO'
58O »O'
¥O3'
XSj
6O'
TO' -Zj' ¥ZL3'
80T'
¥¥0ZO
3O'
X - - - - - -
Z S 8 L 8 6 TT O
ZT
*T O
LX
j
03
IHIHldWJHOlO MOSHVD NOSNVD
H8jj
H8HOl LOvav
1JT
MIT
IVHLl
LoLa
AOOlO IHSXH HLaai
LroLr
j
AjOI
e
LoLa
O
IVH3T
»T
MdT ZT
131
O
Ldvav
TT
HMdT
6
HOON 8
L8SE
NOSNVDMIHIHlOKIHS
•esooijtujseT4iuesoo[X5oxof:iie iooi^vjau g•j» 1"X
extremely poor plot cover. The rest of the accessions generally
have few branches originating from the main stems at or below
5 cm.
Although this cluster exhibits a negative association of
factors 1 and 2 , there is no consistent pattern of expression of
these characters for all the accessions in the cluster. These
accessions, however, have a less than 1 mm distance between base
of central leaflet to function point of the two outer leaflets.
The central leaflets of the leaves nearest to the spikes are
generally narrow and short, with a few exceptions being of medium
length and fairly wide. These accessions other than those
originating from Ethiopia have the longest spikes in the S.
fruticosa collection at ILCA. The above mentioned features tend
to separate accessions in cluster 4 from those of cluster 5.
Cluster 5 comprises nine accessions. These accessions have
extremely poor to very poor plot cover and adaptability. They
are of small stature and tend to form small bushes. Their
leaflet and petiole characters are, however, almost similar to
those of some accessions in clusters 1 and 2 . It is interesting
to note that although most of these accessions show poor
adaptability, they were collected from Soddo, Sidamo in the
neighbourhood of where they are currently being evaluated.
Cluster 6 is another uni-membered cluster comprising the
accession 13924 from Niger. Like the other accessions appearing
on the negative side of factor 1, it has very poor agronomic
characters. It is also the only accession in the collection
which was collected as far south in Niger as latitude 12°00'N.
Cluster Analysis
The 93 accessions of S. fruticosa in the final dendrogram were
divided into ten clusters which distinctly are grouped into two
main blocks comprising 15 and 77 accessions which will be
referred to as groups A and B respectively. Group B comprising
77 accessions is solely composed of accessions collected from
Ethiopia while group A comprises accessions from Ethiopia, Niger
and CSIRO, Australia.
In group A the accessions which form the main cluster are
13942, 13850, 13860, 13893, 13890, 10829, 10523, 13919, 10431 and
10453. These are some of the accessions found with negative
scores on factor 1 in the principal component analysis. These
accessions are part of clusters 4 and 5 in the plot of PCA factor
1 versus factor 2 and are characterised by long spikes and short
and narrow central leaflets of the leaf nearest to the spike.
These accessions also have small bushes, very poor plot cover and
adaptability.
Accessions in group B are characterised by large canopy
spread, a fairly high number of branches originating from the
main stem at or below 5 cm and good to very good adaptability.
180
Table 4. Sorted rotated factor loadings for Stylosanthes
fruticosaLa
FACTOR FACTOR FACTOR FACTOR
1 2 3 4
CANSPW 6 .907 .000 .000 .000
LPRBR 9 .890 .000 .000 .000
CANSPN 7 .882 .000 .000 .000
ADAPT 11 .813 .000 .000 .000
PLCOV 17 .740 .000 .000 .000
NOBR 8 .619 .000 .000 .000
LFW 13 .000 .890 .000 .000
LFL 12 .000 .767 .000 .565
DIST 15 .000 .739 .000 .000
LPETR 20 .000 .718 .000 .000
HTSHT 18 .000 .000 .858 .000
PLGRFM 3 .000 .000 -.790 .000
PLHTHT 5 .296 .000 .640 .000
LFRAT 14 .000 -.266 .000 .906
VP 4.176 2.606 1.920 1.318
The above factor-loading matrix has been rearranged so that the
columns appear in decreasing order or variance explained by
factors. The rows have been rearranged so that for each
successive factor, loadings greater than 0.5000 appear first.
Loadings less than 0.2500 have been replaced by zero.
K-means Clustering
In the final K-means clustering analysis, divisions into 10 and
12 clusters were specified (Table 5) . The relative importance of
characters or variables in determining clusters can be judged by
the "F-ratio" of the between to the within cluster mean squares.
The most important variables were the average length of the
spike, canopy spread at the narrowest point, the distance between
base of centre leaflet to junction point of the two outer
leaflets, canopy spread at the widest point, length of the
longest primary branch, petiole length of the leaf nearest to the
spike, plot cover and adaptability. The F-ratios for these
variables were all above 10.0 with the highest being 28.0 for the
average length of the spike. This can be illustrated by
comparing clusters 2, 4 and 8. Cluster 2 is composed solely of
accession 6745 which has the longest spike (of about 50 mm) among
the collection; cluster 4 is also solely composed of 13342 which
has the shortest spike (about 5 mm) and cluster 8 comprises
accessions 13942, 13929, 13893, 13890, 13850 and 13860 whose
spike lengths vary from 25 to 35 mm. The rest of the accessions
had spike lengths ranging from 10 to 20 mm.
181
* —
w
X
<
r\i
X i
l
rt i
n i
o
U rO
•A
U
a rO
>H i
n i
o)
X
C\J
l1o1
1drt
0 OJ
o o l
i
o
w w i
i-i
oJ X
o
<
rt
irt ,_" o
CD Z i
U to T"
o
— l
< </> i
u X i
< i
I
PL,
 
c
i3
c
oI
si
182
IP-
O
ro
o
ID
1 *
(M
O
c
01
ro
CD
<t IDro- <*
C0 o
CO
= rO
O
o
CD
01 ,
rO
ro
CD
8^
— ro
*s
CM*—
CDo
rO
• O
CD
CM
CO
CMo ■
i
on
CD
i
ro
CDO
i*- 101 ■ 'i
i* i
(M C0
1^r ro
ro
o -
oO i
rO
CD
O
CD°
i
lro
ro
ro CD
in co i
o° 95 i— CO o
<jo oo
f* I
o i
s
u
<
u
a,
3
U.
C\J
o
rO-
I
l
i
1
CM
V
i
1
1
1
o
1
1
l
rO-
CM ro
cfoO
CD -
rO°
O.
*
Q
o
co
= <t
m „cd
„ro rO
CM
ro
ro-
C£>
8.
CO"
ro
cm ,
„o
CM" •
in • „
- s°2>QoJcD
to
CO
vl
in
o m
— „tD
°rO
CD
CD
=-'•$5
CD
ro
O
CD
00
CD
rO
<9
o o
ro
00
in
o
o =
° ro°
o o in o I
O
o CD °ro ■ nin
CM 2
i»» oj —
ro 9Xe ro f
oo— r^-
0> to,
9,°* 1}
* o CD)
CD rO uro O —
m
m j*
—o 0m *
S - ro
CO
r0°
• r:
C£> i
i00 CD
§° $ im CD CO
i* e '
I
o oO
Table 5. Accessions constituting various clusters in the 10 and
12 group of K-means clustering.
Clusters Group of 10 clusters Group of 12 clusters
Cluster 1 23 accessions 10 accessions
Cluster 2
Cluster 3
Cluster 4
Cluster 5
Cluster 6
Cluster 7
Cluster 8
Cluster 9
10388, 10417, 10424, 10437 10428, 10493, 13947, 13970,
10433, 10432, 10493, 10511 10396, 10482, 10391, 13959,
10513, 13947, 13970, 10311 10304, 10470.
10322, 10396, 10420, 10482
10304, 10470, 10451, 13958
10391, 13971.
1 accession
6745
1 accession
6745
5 accessions 4 accessions
10445, 10440, 10514, 13932 10445, 10440, 13932, 10464
1 accession
13342
18 accessions
10501, 10506, 10515, 13949
13951, 13953, 13978, 13961
13954, 9761, 9796, 10364
10393, 10539, 9799, 10381
13973, 9678
19 accessions
10423, 10507, 13950, 13976
13975, 13972, 13968, 13966
13965, 13962, 13959, 9752
10280, 10288, 10415, 10452
10517, 10471, 13964
8 accessions
10449, 10448, 10431, 10523
13945, 13960, 10404, 10829
6 accessions
13942, 13929, 13893, 13890
13850, 13860
5 accessions
10468, 10453, 10516, 13924
13919
1 accession
13342
4 accessions
9761, 10364, 10539, 9799
17 accessions
10507, 13950, 13976, 13975
13972, 13968, 13966, 13965
13962, 9752, 10280, 10288
10415, 10452, 10517, 10471
13964
10449,
13945,
13942,
13850,
10468,
13919
8 accessions
10448, 10431, 10523
13960, 10829
6 accessions
13929, 13893, 13890
13860
5 accessions
10453, 10516, 13924
Cluster 10 7 accessions
10450, 10828, 10425, 10488 10450,
10474, 13946, 9753 10474,
7 accessions
10828, 10425, 10488
13946, 9753
184
Table 5 (cont'd) .
Clusters Group of 10 clusters Group of 12 clusters
Cluster 11
Cluster 12
12 accessions
10506, 10515, 13949, 13951
13953, 13961, 13954, 9796
10393, 10318, 13973, 9678
10501, 10514, 13978
15 accessions
10417, 10424, 10437, 10433
10432, 10511, 10513, 10311
10322, 10420, 10451, 13958
13971, 10388, 10423
DISCUSSION
Cluster analysis, principal component analysis and K-means
clustering were used to group similar accessions. The 93
accessions available for study were agronomically and
morphologically diverse. With cluster analysis, some strong
clusters were formed. However, about one-third of the accessions
were non-conformist. Accessions collected from Niger tended to
form a group of their own while the Ethiopian materials also
tended to form theirs. There were exceptions to this, however.
Group A in the cluster analysis which had all the Niger
accessions, except 13929, had materials of Ethiopian origin too.
However, a sub-cluster existed that comprised wholly of materials
from Niger. This cluster was also consistent in both principal
component analysis and K-means clustering. It was not possible
to record some of the most important characters in Stylosanthes
species such as flowering, seed and fruit characters in this
study .
For proper characterisation and evaluation of plant genetic
resources it is necessary to start from seeding to harvesting in
order to obtain more meaningful inferences. Where time is
limiting, however, a study such as is described here can give
guidelines on what characters to use during evaluation and
possible variations existing among the materials. All the
analyses gave similar grouping of accessions. These groups were
related to the geographical location or types of environments
from which these plants were collected rather than variations
between accessions from similar geographical location. For
instance, the Niger material tended to be grouped together, while
the Ethiopian material tended to form "agrotypes" .
Since ILCA's collections of S. fruticosa is from such a
limited part of its range, further collections from other African
countries are highly recommended on the basis of the following
observations :
185
1. There is need to collect and preserve genetic materials that
are in danger of disappearing as a consequence of the
destruction of natural vegetation.
2. There is need to increase the variability of collection from
particular areas in relation to geographic, climatic and/or
edaphic relevance.
3. There is need to collect genetic material in particular
areas which in the past have been neglected or overlooked
with regard to collecting activities.
Whereas it has been possible to obtain groups of accessions
with particular characteristics based on the descriptions used in
this study, the practical application of the findings has some
limitations because of the omission of some important attributes.
The variability existing within the species could therefore have
been better described if more descriptors were used over a longer
period of time. Despite the fact that little study has been done
on S. fruticosa compared to the other Stylosanthes species, this
characterisation and preliminary evaluation gives an indication
of the variability or heterogeneity existing in S. fruticosa
accessions currently available at ILCA.
ACKNOWLEDGEMENTS
We wish to acknowledge the technical assistance of Mr. James
Ochang in growing and looking after the plants at Soddo, Ethiopia
and other members of the Forage Legume Agronomy Group (FLAG) who
provided transport and other necessary assistance.
One of the authors (J.J. Hakiza) wishes to acknowledge the
financial support and authorisation to carry out the study by the
International Board for Plant Genetic Resources (IBPGR) and the
International Livestock Centre for Africa (ILCA) .
REFERENCES
Burt, R.L. , Rotar, P.P., Walker, J.L. and Silvey, M.W. (eds.)
1983. The role of Centrosema, Desmodium and Stylosanthes in
improving tropical pastures. Westview Tropical Agriculture,
Series No. 6.
Engels, J.M.M. 1986. The systematic description of cacao clones
and its significance for taxonomy and plant breeding.
TBPrR, Royal Botanic Gardens, Kew. 1984. Forage and browse
plants for arid and semi-arid Africa. IBPGR Secretariat,
Rome.
Lazier, J.R. 1984. Global ventures in Stylosanthes . IV. West
Africa. In: H.M Stace and L.A. Edye (eds) , The biology and
agronomy of Stylosanthes. Academic Press, Melbourne,
Australia.
186
Skennan, P.J. 1977. Tropical forage legumes. FAO Plant
Production and Protection Series No. 2. FAO, Rome.
Stace, H.M. and Edye, L.A. (eds) . 1984. The biology and agronomy
of Stylosanthes . Academic Press, Melbourne, Australia.
Thulin, M. 1983. Leguminosae of Ethiopia. Opera Botanica 68.
Williams, R.J., Reid, R. , Schultze-Kraft, N.M. , Coasta, S. and
Thomas, B.D. 1984. Natural distribution of Stylosanthes.
In: H. M. Stace and L. A. Edye (eds), The biology and
agronomy of Stylosanthes. Academic Press, Melbourne,
Australia, p. 94.
187
ujwujujujujujuj W W UJ W UJ
in c»1 ci en vO tN Io ci ci ci
ciMciocicicioociociciofoocirtciotn
ifl m ifl
C7lCJii7*C7lO*C7*<7lOt iT) CJI Q"> CP> CT*
v£> VO v£ VO ID I£> i£) \£)
Z Z Z Z z z z z z z z zzzzzzzz
uuujutJuuuuuuujjujjjujujju
J _] -J _I -J _] ..J _] J -J J _J _] _] J _1 _J _J _] _1 _J -J _J .J „j
ucjjujjujjjjOjtJdj0tJtJujujuu
J -1 -J J J J *J J -J J _1_1_1_i_1_1_],-i_]
UJ UJ UJ U] UJ W UJ UJ UJ UJ W Ul w
o o o o 0 o
c c d n C a
4-1
w UJ UJ UJ UJ UJ
a a a a
uj w w w uj UJ
ri r* <fl ofl <n 4 (Tt 4 rt ffl <* m rt ofl rt (11 ofl o1 ffl 4
w VI V) W n « w w n to V) W w o o w o V) W
o o o o n Q o o o o o o o o o 0 o o o
o o o CJ o U o u u y o o CJ o o CJ u o o o
4o> 4ot *J
2 2 2 2 2 2 2 2 2 2 2 2 2 5 2 2 8 2 2 2 2 2 2 2 2
4h m lt **4 lw m 4-H 4uoH 4o4 4uo4 4o<
xi jZ Ji Ji J3 J^j3j3^j3^j3j3j3j3j3j3xj3xj3j3j3x:
ofl ofl ofl ofl
CAC/lMCACAC/JCAC/JCAHCAC/JCAC/jC/JCAC/ICACACACACACA(7jC/J
0 CO oT m oo tN
CM CN tN tN tN Co m 10 m ln
10 >£> r- co a> rH tN PI
188
U U W W W W W UUUUUUUUUUUUU
- r-. iO m co to fM fM »- ^ tf1 if) co CO - S iTl in in
r- r^
-£
r- r-
- co co CO co co co co CO CO co co co CO co co m co co co c: co CO co
= Z Z 2 Z Z Z z Z Z Z Z Z Z ■ z Z Z Z z Z Z z z
miA
u
LA
CM CN) CM N CM
tr\ in 1/1 in in 1/1 in
z z z z z z Z SB Z Z Z Z
< < < <u u u -
■< < <
(J - o
J J J -J ^J -J -] -J _J .4 J -J ^]
3 .
UUUUUUUCJt3ucJUUUOUUOOUUi-tOUU
a-.aaaaaaaa.aa.aaa.
4J4J4J4J4J4J4-14J-J4J-JjJ-J4J4J4JtJ4J4J4J4J
WUUWUUUWUUWUUUUUUUUUU
-o .g j:
W M W
nl 11 a rt * m i1 m m i« i1 ti ii ii X « rt fl 1 fl - n rt i* rt
tfi w V) w Wl to ■n w i- w
f) 0 o n - 0 - - - - - o o - ■ - o - 0
U o u u o u o 0 u o o o u o o u o o o u - CJ u u u
2 2 2 2 2 2 2 2 2 2 2 2 5
M-l i4H iH i-4 -H i--- "4-1
2 2 2 2 2 2 2 2 2 2
1M -w m
.c j3 js j3 j: s:
S ut S
,c.c.c.c.c.e.c.c.c.c.c.e.c.c.c.c
Sd c d d d d a c a
XXX XX XX XXXXXXXXXXX X X X X X X
WWUlt/)UlWWC/lW1/)WWWUll/lWWWWlflWl/lC/lW
in in in in in in
-*■*-*■* m m m
CJi c^ c^ c^ c^ c^ c^
CO CO CO CO CI CI co
co in ■■
CM CM CM OJ CO CI co co co d co
189
uuuuuuuuuu uuuuuuuuuuuuu
yi m cn
CI CO CI o eg ci eg o
CI CO CI CI CI CI CI CI CI CI CI CI
Z BE Z Z
CI CI CM CM rH
u-i O-i u-, o") inmu"iu-iiomioin «- ci cm
M> « « «
z z z z z z z z z z z z z z
-uu-uuuuuuuu-33cju-u-uuuu-o
J l-J J J J ,J,J,J,J,J,Jt-ll-l.J -1,J,J-1.J,J-|J
SS55S5SS5S5SSSS5S5SSS5SSS
J,J,JIJ1J-),J1J■J,J JJJJJJJUJJJJ
0.0.0-0.0-0.0.0.0,0.0.0.0.0.0. a a a a a a
4J -J
J3J3J3J3J3J3J3J3J3J3J3J3 JS J3 J3 J3 J3 J3 J3
UUUUUUUUUUUUUUUUUUUU1
o) 11 4 1 4 4 4 rt rt 4 - 4 4 11 4 4 n 4 4 4 4 4 4 4 4
i/i ifl ei w «] i- w VI « W in w V) w w w w l/I w W ») w m
0 0 O 0 0 - n - - n - O O ■ - - n 0 O O - - n O 0
0 u u - U u u 0 0 0 u O 0 0 lJ - u y lJ 0 u o u - -
2 2 2 2 2 2 2 2 2
4J 4J u 4J 4J -J -J
2 2 2 2 2 2 2
M-l IW -M iw iw iw m
2 2 2 2 2 2
m m i+( y-i i-4 m
2 2
j3j3j:J3J3j:j3j3jSjSj3J3 J3js.ej3j3-i5.ej3.i3.i3
s s s s s
js js♦J 4J
s s
4J4J-J-J4J*J-J-J-J-J4-I-J4J4J-J4J-J-J
WWWWU)WWt/)WWWU/)Wt/)U)Wt/)C/lto to w t/i to w
0"i Oi
0■ a> in
\£i ^ \0 U~l U"l
Oi C^ CTi Qi O^
in in Ifl Ifl
a- o CO CO
CI
CO
iN
O
00
CM
O
C)
oo o -
- - CM CI -3- u-'
190
38ZZo'S1<lZO
3£.1OM,j
jooozoi*T
jo?i}8inuori3ojj
jEZZ08LZOSVSZZ0j66ZO
38*ZoiO
nOE,8'M3VOoj860O
3Ooj00
3Z06'N<)ZET
3Z"ZZo,ZjZO
3j8EVO
j8EZ•;S0jO6LSO
38ZLjZ»36ZLEj»*
jTEL10
3S*LEj8_
VOTT»XOOI
aooO aoon
8UO|]
aooO
»oojf
aoof]
8nOf| 8nO{]
aooN aoou aoojj
auoh
jn0H
»nO{] »oon 8ooN
aooN
jXX
VOX VOX VOX VOX VOX VOX VDXX VOX
jlX
V311 VOX VOX
aoOOvjn aoOsLvjn
VOX
aoOO/vjn
aoaO/vjn aoOO/vjn HOOO/jlX
VOX VOX VOX VOX VOX VOX VOX VOX VOX VOX VOX
O11
VOX
oaioo
VOX
llSfJOJ0
«OOoio^ia *lOoinia *fOoi^jj
eiOoiioa eiOoima *iOomoa «TO<>TMi3 «T<oomi3 *TOoima
*^Ooiioa
*fOoimo
eiOoi^a ETOoTHjo
oa8o[|j
j8T(j X
*o[OolHi3
Je8l|J
saTeOssj ^
«sool^ejj
esooi^jjj
esoo^ieJJ esoo"jieoj
FsOOi^ruj esooT^»»aj
*soo"jirLlj «sooi^»uj
esooT^joj
esooofir»jj
esooT^njj FOooT^joj esooiijoj esoojruj
esooi^joj «Oj"JiejJ
esooT^»uj esooT^njj
sai^oesofjo samoesoo[jo sa\]aouso"[j5 san^o*sooX^o saipoesofjo sai^oesoijo saioiouso^Xio sa^ooFsoo[Xio samuesoojo sa^^oesooX^o sanioesofXso saoj^oesofX^o sai^oesoiX^o sai:joeso"[XjS saiioEsojXu samoesofX^S sanioeso"[Xio siipoHio"[A o
Oj1X
ZZE0T
»j0X
E6E0X
86E0T
6ES0T
8O0T 6O0T
Z*6ET ZE6ET
-^O
O6ET
E68ET
068ET 088ET
0OEX
ZjEEX ZEV0T
6Z6ET
8L
LL 81 6/ 08 18
O
E8g V8 58 88 L8 88 68 06 T6 66 Z6
aDjOsjo
JjOmoaq.e:nisXriooooe
oipOSofoej o^iojnos"[FolSoj.Ojjo e
(e(iooo)*TXTeoaOOy
FODDER PRODUCTION POTENTIAL FOR ZERO-GRAZING
SYSTEMS IN THE BIMODAL WEST USAHBARA MOUNTAINS OF TANZANIA
Reinhard Pfeiffer
TIRDEP-SECAP-LUSHOTO, P.O. Box 72, Tanzania
Abstract
Overpopulation and increasing land pressure are resulting more
and more in the deterioration of the agricultural potential in
the hitherto highly productive agricultural areas of the West
Usambara Mountains. To counteract this sad development, multiple
measures have been applied. This paper presents the production
potential data of fodders planted along contourlines and gives an
outlook on environmental effects of such fodder lines in zero-
grazed systems as part of an integrated approach to resolving the
land pressure problems.
INTRODUCTION
The Western Usambara highlands are among the highly populated
areas in Tanzania. Lundgren (1978) estimated the human
population density at 180 people/km . Backeus (1982) predicted
that by the year 1990 the population will have risen to 300
people/km2 . The high population has considerably increased
pressure on land for agriculture. In certain areas people keep
large herds of cattle which graze freely on the slopes as the
communal grazing areas at the valley bottoms are used extensively
for horticultural crops.
Free grazing has caused considerable surface runoff and soil
erosion resulting in site deterioration with a consequent
reduction in crop production. Lushoto district has tried to
tackle the problem by implementing a major project on
agroforestry throughout the Western Usambara highlands since
1980. The Soil Erosion Control Agroforestry Project (SECAP)
applies an integrated extension approach which takes into account
forestry, plant production and livestock measures using the
macro-contourline approach. This approach integrates livestock
production through "cutting and carrying" of fodder, zero-grazing
and stall-feeding of livestock. It also integrates plant
production through farmyard manure application due to stall-
feeding operations, and it has a forestry component for fuel,
construction and soil erosion control measures.
MATERIALS AND METHODS
In order to simulate the many site conditions in the West
Usambara Mountains, four on-station trial fields covering various
ecoclimatic conditions have been established since November 1985.
Screening, elimination and management trials have been carried
out using the randomised block designs. The macro-contourline is
essentially an erosion control barrier consisting of four
components, namely: a fodder bushline (preferably a legume) ,
creeping legume line, fodder grass line and an agroforestry-tree
line. This has been prepared and planted accordingly. The
192
fodders are subjected to intensive cutting regimes. All the
fodder components are cut half a year later after planting as
this is what arr average farmer would normally do. Thereafter the
fodders are cut intermittently throughout the year once there is
sufficient amount of regrowth on them.
Detailed ecological and meteorological measurements have
been carried out in between the macro-contourlines. These have
included:
Z : A zero treatment in which no soil
control measure has been effected.
MI: Sunflower planted along the contourline.
T: Grevillea robusta line planted at 2m
spacing along the countour.
GT: Guatemala grass with Grevillea robusta
planted.
I: Maize planted as a sole crop.
II: Groundnuts planted as a sole crop.
Ill: Maize/groundnut mix stand planted.
IV: Maize/oil radish mix stand planted.
Erosion (-) and accumulation (+) of soil was measured in mm
at soil erosion measuring stakes in between two contourlines.
RESULTS AND DISCUSSION
The data in the accompanying pages suggest that the macro-
contourlines produced considerable amounts of biomass of both
grass and legume species. In Table 1 for instance, the high
productivity potential of the macro-contourlines has been
demonstrated over sole stands. The average dry-matter (DM)
yields are nearly three times higher on macro-contourlines than
on sole grasslands in Kenya. The macro-contourlines produced
considerably more fodder in subsequent years than the first
(Table 2). The grass yields were 22% whereas the yields of
fodder bushes were 86% more in the second year. This was most
probably due to better established root systems and more
intensive cutting regimes after the first year.
Because of the diversity in topography and ecoclimatic
conditions the most feasible fodder combinations for zero grazing
systems in the West Usambara Mountains are presented in Table 3.
The macro-contourlines appear to serve as basic fodder sources
for zero grazing systems in the West Usambara Mountains. Due to
stall-feeding of dairy cows with macro-contourline fodders, the
milk yields in the West Usambara Mountains have increased already
from 3 liters/day for local farmers to 7 liters/day for SECAP
farmers.
193
Table 1. Fodder DM yields (t/ha) of various grasses.
Sole grasses
in Kenya
Macro-contour1 ines
in Tanzania
Species/
location Kisii Kakamega Lushoto
Banagrass 18.2 20.0
Guatemala
grass 3.1 3.1
51.1
25.5
Table 2. Fodder DM yields (t/ha) of different fodder types.
Fodder type/
year 1st year
Mean of 2nd
& 3rd year
Fodder grasses
Fodder bushes
2.10
1.75
3.30
3.25
Table 3. Feasible fodder combinations adapted to ecoclimatic
zones/sites.
Site/
ecoclimatic
Mbwei in
warm/dry
Ubiri and
Silvici in
humid/warm
Nywelo in
dry/cold
Fodder combinations
Bana grass
Desmodium intortum, Calliandra
cal lothyrsus ,
Leucaena diversifolia, Neonotonia
wightii,
Stylosanthes guianensis.
Napier grass
Neonotonia wightii, Desmodium
intortum,
Desmodium uncinatum, Cajanus cajan,
Leucaena leucocephala.
Tripsacum laxum, Desmodium intortum,
Sesbania aculeata.
The macro-contourlines also appear to have good
environmental attributes based on detailed ecological and
meteorological measurements carried out in cropped fields between
macro-contourlines. As shown in Table 4, most of the soil was
transported from treatments MI, T and GT; these were treatments
194
that offered small erosion control because of poor ground cover.
The zero treatment which was densely covered by natural fallow
weeds had little soil movement.
Table 4. Soil loss as a result of macro-contourline plantings on
a slope inclination of 21° between contourlines.
Decreasing soil movements >
MI G III II IV I Z
2.86(+) 2.44(+) 1.60(+) 1.42(+) 1.41(+) 1.17(+) 0.62(+) 0.26(-)
CONCLUSION
The contourline fodder production studies in the West Usambara
Mountains have indicated potential for integrated approach to
livestock production, crop production, soil erosion control and
agroforestry . It has provided considerable amount of forage for
zero-grazing systems by more than doubling the milk production
capacity of the small-scale dairy farmers in the area. The
contourline fodder and tree plantation also has helped to
consolidate the erosion control role of these structures.
REFERENCES
Backeus, I. 1982. Report on study tour to the indigenous
forests of the West Usambara Mountains. Vaxtbiologiska
Institutionen, Uppsala.
Lundgren, B. 1978. Soil condition and nutrient cycling under
natural plantation forests in Tanzania Highlands. Reports
in Forest Ecology and Forest Soils No. 31, Uppsala.
195
THE CURRENT STATUS OF KNOWLEDGE ON THE FEED
VALUE OF CROTALARIA SPECIES
S.V. Sarwatt and F.E.J. Mkiwa
Department of Animal Science and Production
Faculty of Agriculture
Sokoine University of Agriculture
P.O. Box 3004, Morogoro, Tanzania
Abstract
The potential of Crotalaria ochroleuca in farming systems in
Tanzania is reviewed. Possibility of its use as a livestock feed
is discussed in terms of its nutritive value in comparison to
that of other legumes. However, there is no documented
scientific information on its utilisation as a livestock feed in
Tanzania. Factors that limit its wide utilisation as a feed are
given. The paper suggests areas of research emphasising a
multidisciplinary approach in attempting to appreciate the role
of this wild legume in our agricultural systems.
INTRODUCTION
The potential and use of forage legumes in farming systems and
particularly in livestock nutrition has been well summarised by
Tothill (1986) . However research on indigenous wild legumes has
been very limited compared to improved legumes. Tropical legumes
which are used for forage have been cultivated during the last 50
years (Williams, 1983) yet the volume of unexplored genetic
resource remains vast.
Crotalaria ochroleuca locally known as "marejea" in Tanzania
is a well-known leguminous plant in the southern part of the
country. As early as 1939, the legume was cultivated on small
scale by Benedictine Fathers at Peramiho (Rupper, in press) . It
was mainly cultivated for the purpose of restoring fertility to
the soil and to combat weeds. During shortages of fodder the
legume was fed to dairy cows and calves. Although no quantified
data is available, the performances of the animals were observed
to be good.
The aim of this paper is therefore to highlight research
work related with Crotalaria spp. in the field of animal
production. Areas that need immediate investigations are
emphasised.
BOTANICAL DESCRIPTION
Crotalaria , the genus to which Crotalaria ochroleuca belongs is
the largest in tropical Africa and is commonly encountered in
open places from mountains to semi-deserts, with over 500 species
(Polhill, 1982). Most of the species are either annuals, semi-
perennial, perennials, herbs or shrubs (Martin and Leonard,
1970) . The leaves are either simple or compound with 3,5 or 7
leaflets (Polhill, 1982).
196
Most of the species have numerous yellow flowers which later
on bear tough-skinned seed pods that are inflated (Polhill,
1982) . The number of seeds contained in a single pod depends on
species but ranges from 5 to 50 seeds. The seeds are kidney-
shaped, and their colour varies from olive-green to either
yellow-red or brown.
CHEMICAL COMPOSITION
Dry-Matter Yield
Total dry matter is one of the factors which to a large extent
determines the carrying capacity of a forage. In many selection
programmes dry-matter yield is a key criteria for selection
(Whiteman, 1980) . Since Crotalaria spp. are among the high
nitrogen fixers, this partly explains their yield potential
because total dry matter and nitrogen fixation in legumes are
known to be positively correlated.
In Upper Volta, it has been reported that C. juncea
produced 10 tonnes/ha of dry matter without application of
fertilizer (Anon, 1965) . In Surinam when C. quinquefolia was
grown for fodder, Ubels (1960) recorded an average of 17
tonnes/ha of dry matter. In the southern part of Tanzania,
Mukurasi (1986) harvested 12 tonnes/ha of dry matter from C.
zanzibarica. Comparatively, lucerne (Medicago sativa) , which is
one of the most important perennial fodders used in many regions
of the world produces yields of 8-9 tonnes/ha (Whiteman, 1980) .
Crude Protein, Minerals and Other Nutrients
Table 1 shows that an average crude protein content of Crotalaria
spp. ranges from 10-35% with a mean of 24.1%. Crude protein
content of the whole plant of C. ochroleuca is 28.1% and that of
leaves is 34.5%. Although it is not possible to rank tropical
legumes in terms of their protein content, according to Minson
(1977) the mean crude protein of tropical legumes is 17.2%. The
range may lie between 5.6% for Stylosanthes humilis to 35.8% for
Leucaena leucocephala. It is therefore obvious that Crotalaria
ochroleuca ranks high amongst other legumes in terms of crude
protein content (Table 2).
The crude fibre content of a forage is another important
measure of dry-matter intake and in vivo dry-matter
digestibility. The efficiency of ruminant microbes during
digestion is influenced to a large extent by lignin available in
the fibre component. Van Soest (1965) showed that the proportion
of potentially digestible components decline as the fibrous
content increases. Thus the relatively fair crude fibre content
in Crotalaria (Table 1) suggests a possible high intake and dry-
matter digestibility.
Phosphorus and calcium content in Crotalaria spp. has been
shown in Table 1. According to NRC (1978) the minimum level of
phosphorus requirements for most animals is about 0.15%, which is
below the amount available in Crotalaria. The calcium content is
observed to be fairly high compared to the other legumes.
197
0EO8oTZ0jZj«VZZVLoOZO
OE6Xj OEOj OEOj OEOj OE6Xj OE6Xj OE6Xj OE6Xj OEOj OEOj OE6Xj :»LOjo»misTJ5lJ»3*3(m!A jS6Xj jE6Xj jEOj jE6Xj
oop,
oejj
jjj ee3j ooj
o^jj oejog
ojj uo8j
iro8j
ajj ojj
eotoejie:n?o[T?g
O8Oj oOio
JB^eH
o^eo
OOOj*i»in
2e})OOo
O86X: OE6X!
5 3 *s O V 3 3 3 5 *s *S l• eiX*T*eH Pj1O88(1 eoeo[*an eo*[«»« 1: 1:
fESOa Oe8OQ leSOQ ie8Oa
1«8Oa
*[eSOQ O«SOQ O*8OQ T«8joQ T^SOQ ie8OQ
sejmfoH OKM>S
nt/uos
spotjla}l
OeA*31
o»i*lOaT°Mn
& 8Z0 8o0
82o
O0
LO0 So0
j0
ZEo0
8j0 8Z0
8Zo
EZ0
- - -
SE0
-
*D
LS0
!?6o
180 Zo.0
«o
SL0
Oj O0 jT
«o
L<fX
- - - 08o0 -
*jS
O0
60o
Tj0
«o
Lj0
j0
ESO
HO 6E0 8Z0 - - - - - -
ijs» ZOZ 8ZL O6
066 168
»Z8 6S0T O'L 8O6 O'Z -
00'ET
- 6o8 6oZ »OS -
ieS ZS8
0V8
j3j
6S6E
E8ZZ
O8Z
88SZ
E0o
58SS
ZZT»
8X'LZ
<rC81 X6LO
oll OLZ -
E00V
06e 80"OV
-
ZTSE EWE
-
O
LXZ EoZ Z87 ToE E6T
68Z
jZ jZ
LOZ
j7 6"»
T8Z
- ao OZ oT -
LOX E6oZ
-
L0
Z08Z Z0TZ
j*O
80j
ZLSE LXXZ X0'O SLZE
0V0Z 6Z9Z
01'8
Oj
00»T
OX<l VX6E
001*
- 0fO 8Z"»T 0XjE
OD
00EZ
TSOE
6Z.EZ
588Z
X<fXZ
58XE
a'8T
ZE0Z
0STZ
£6SZ
LZO S6»Z
00SE
00o0T
00'^T
O'O - 0VVX 0SO
01jZ
wa 88 Z6 00 00 ZE'
joe8t^s
iVi0j8
eA"X3C4e8i/l
jJAOo'3
eAT3E3e8aA
M"O,
:JXX, „
eAO3*1e8eA
''*OO_ '*OX"3
*Tji™ - *TjXX<n&
oZ SZ Sj oj
<!Z
*'*OX"3 -
OeTo3
saefeX8Fj
F^^oo[FJ
*n*inT
^oreijioinT *o»eijioaaui
»^«oojjoi
e»rjijeO
«oef^i^»O
esoojaj
WfOO"J"J"[*A
wueo
waojf
COTjeiojsos sl1Ti»ieOe
»5«TJit
EorpajuaioT
vj^ojajbofb
«2«ooj
«jale»qoo *jifo »po
OS *3 '3 '3 '3
•3 •3
'3 *3
•3 •3 •3 •3
3 *3
•3
3
•3 •3 •3 •j
*saofeOsb6"jeo[L0*^oiojjojjoT^isoOmooo[^olOOOeO^3eTQ X
(OO)Toaje^nw
(8O1)H5/UaS
(jj)Appea
(LL6T)jaULXj5OS
(jO)Appea
(88O)jamrtSj
U86X)pjojTTM
(8jO)ucpEogXEfceog
__-jjT
oo-6OTj'*Oj
OOfrjLT__X EO
Z_T8O6frO8M
_O8o_jM-VOj5Z
T'6_jfrjOZ Z
-----»T
--8oEM
--8jZ
aO
aje_ojiooo•oaoneLojqoo*
aeoonCaTjaLE^ jo
EXaqjeoooneLEoea eeqEXEijjeoooee_aseEone^
aATHESofcaOfpSH
STLTmrmseu^oaoo_A^o
E^acXOuemnTpouio aa^EcToonumTpou sea
y>
33j
33
O
OOHQ
jeepoj
(oTSaiHQ%)aTJEXE^ojo<nemnOe_japp j^oejej Tp3n_BAaA ^Tj^noooT T ojmooic ,
UTILISATION AS FEED FOR LIVESTOCK
Research on the use of Crotalaria as a feed for livestock is very
limited. The limited data that is available does not give
quantitative investigations to establish the potential of
Crotalaria spp. as a fodder plant. The only work that gives some
detail is that of Balaraman and Vankaterkrishman (1974) who
determined the nutritive value of C. juncea by conducting a
metabolism trial using rams. The chemical composition and
digestibility coefficients of sunnhemp hay used is shown in Table
3. Intake on average was 2.57 kg of dry matter per 100 kg body
weight. The digestibility coefficients on consumption basis
showed that the portion consumed was highly digestible. However,
on feed basis the digestibility coefficients were very poor due
to the fact that the rams chose the leafier parts of the plant
and rejected the stems.
Table 3. Chemical composition and digestibility coefficients of
C. juncea fed to sheep (% of DM) .
Nutrients Composition Consumption Feed
basis basis
Dry matter - 61.90 31.49
Organic matter - 65.07 32.50
Crude protein 24.95 75.88 19.95
Crude fibre 31.65 59.53 53.68
Other extracts 2.81 35.62 24.39
NFE 27.62 57.30 27.51
Ash 13.00 - -
Calcium 1.47 - -
Phosphorus 0.36 — —
Source : Balaraman and Vankaterkrishman (1974)
TOXICITY OF CROTALARIA SPP.
About 20 species of Crotalaria in tropical Africa alone are known
to cause poisoning to cattle (Polhill, 1982). The symptoms seem
to vary but common syndromes include inflammation of the horn,
lameness and occasional starvation due to refusal by the animals
to eat. The more lethal forms of poisoning affects the nervous
systems, lungs and liver (Polhill, 1982). Poisoning seems to be
serious at flowering and seed stages. Probably this is due to
accumulation of the toxic alkaloid substance called Crotaline in
these parts of the plant at maturity. Steyn and van de Walt
(1945) reported loss of wool when sheep were fed C. juncea hay
that was cut at flowering stage. They suggested that
supplementation level in livestock feeds should not exceed 10% of
the daily ration. Rupper (1984) observed that when C. ochroleuca
seeds are sprinkled between the bags containing cereal grains,
storage pests were killed instantly. They reappear six to nine
months later. The fact that Crotalaria spp. have been observed
to be toxic in one location but non-toxic in another, may explain
the contradicting reports. The reasons why one species can be
toxic in one location and non-toxic in another is still not yet
understood. Crotalaria ochroleuca may be poisonous at flowering
200
stage and the seeds may contain some poison, however, the legume
seems to be non-toxic before flowering (Rupper, 1984) . Further
research on this area is needed to establish the effect of
feeding the legume on livestock performance and health.
CONCLUSION
Crotalaria ochroleuca is regarded as a multi-purpose crop in
agricultural production in Tanzania. Its importance as a
livestock feed has been underscored because the plant is
cultivated mainly in the area where a small number of livestock
are kept and the land is not a limiting factor (Sarwatt, in
press). In areas where cultivable land is a problem (e.g. in
Arusha and Kilimanjaro regions) the legumes can be intercropped
with regular food or cash crops.
The legume will provide nitrogen to the crops intercropped
with and assist in the control of weeds and nematodes. The green
part can be cut two to three times and fed to livestock. Every
time the legume is cut, the nodules senescent thus releasing
more nitrogen and other nutrients which can be utilised by the
accompanying crops.
Research on the multi-purpose use of "marejea" is very
limited in Tanzania. Most of the information available is mainly
from the farmers' experience, which fails to answer questions
like "how much N is fixed by marejea under the present farming
system," or "how much marejea should be fed to a dairy cow for
efficient milk production". For these questions to be answered,
a multi-disciplinary approach is needed. Lack of scientific
information on its feed value as regards feed intake,
digestibility coefficients and safe levels of supplementation
calls for an urgent need to assess the animal production
potential of this plant as a cheap fodder legume.
REFERENCES
Anon. 1965. La crotalaire, legumineuse fourragere annuelle
possible en region Mossi. (Crotalaria, a possible annual
fodder crop for the Mossi area) . Cah. Agr. Pratique Pays
Chands 2. Tropical Abstracts 21:93-95.
Balaraman, N. and Vankaterkrishman, R. 1974. Nutritive value of
sunnhemp (Crotalaria juncea Linn) hay for sheep. Indian
Veterinary Journal 51 (5) : 337-341.
Dougal, H.W. and Bogdan, A.V. 1936. The chemical composition of
some leguminous plants grown in the herbage nursery at
Kitale, Kenya. East African Journal of Agriculture and
Forestry 32(1): 45-49.
Martin, J.H. and Leonard, W.H. 1970. Principles of field crop
production. Macmillan Co., New York.
201
Milford, R. 1967. Nutritive value and chemical composition of
seven tropical legumes and lucerne grown in sub-tropical
south eastern Queensland. Australian Journal of
Experimental Agriculture and Animal Husbandry 7:540-545.
Minson, D.J. 1977. Chemical composition and nutritive value of
tropical legumes. In: D.J. Skerman (ed.), Tropical forage
1egumes . FAO , Rome .
Mukurasi, N.J. 1986. Agricultural attributes of Crotalaria
zanzibaries. Uyole Agric. Centre, Mbeya, Tanzania.
NAS. 1979. Tropical forage legumes. Natural Academy of Sciences,
Washington, D.C.
Neal, W.M. and Becker, R.B. 1933. A type of laboratory silo and
its use with Crotalaria. Journal of Agricultural Research
47:617-625.
Newman, D.M.R. 1968. A comparison of the nutritional value of
Stylosanthes humilis, S. guyanensis and introduced grass in
Northern Rhodesia. Australian Journal of Agricultural
Research 28:921-935.
nrc (National Requirements Council). 1978. Nutrient
requirements of domestic animals. Nutrient requirements of
dairy cattle. 5th edition. National Academy of Sciences,
Washington, D.C.
Polhill, R.M. 1982. Crotalaria in Africa and Madagascar. A. A.
Balkema Co., Rotterdam.
Reddy, M.R. 1969. Sunnhemp hay can cut down concentrate needs of
cattle. Indian Farming 18:45-46.
Rupper, G.P. 1984. Marejea Rafiki ya Wakulima. Benedictine Pub.
Ndanda - Peramiho.
Rupper, G.P. (in press) . Cultivation of Crotalaria ochroleuca
around Peramiho. In: Proc. of Marejea cultivation in
Tanzania held at Peramiho, Songea, 15th-21st October, 1986.
Sarwatt, S.V. 1986. Chemical composition and in vitro dry matter
digestibility of Crotalaria ochroleuca. Dept. of Animal
Science, Sokoine Univ. of Agriculture, Tanzania.
Sarwatt, S.V. (in press) . The value of Crotalaria ochroleuca as
a livestock feed. Proc. of Marejea cultivation in Tanzania
held at Peramiho, Songea, 15th-21st October, 1986.
Skerman, P.J. 1977. Lablab purpureas (L) Sweet. In: Tropical
forage legumes. FAO Plant Production and Protection Series,
No. 2. FAO, Rome.
Steyn, D.G. and van der Walt, S.J. 1945. Farming in South
Africa 20:445-447.
202
Tothill, J.C. 1986. The role of legumes in farming systems of
sub-Saharan Africa. In: I. Haque, S. Jutzi and P.J.H. Neate
(eds) , Potentials of forage legumes in farming systems of
sub-Saharan Africa. Proceedings of a workshop held at ILCA,
Addis Ababa, Ethiopia, 16-19 September 1985. ILCA, Addis
Ababa, Ethiopia.
Ubels, E.R. 1960. Green manuring. Herb. Abst. 31(1) :144.
Van Soest, P.J. 1965. Symposium on factors influencing voluntary
intake of herbage by ruminants. Voluntary intake in
relation to chemical composition and digestibility. Journal
of Animal Science 24:834-843.
Whiteman, P.C. 1980. Tropical pasture science. Oxford
University Press, London.
Williams, R.J. 1983. Tropical legumes. In: J.G. Mclvor and R.A.
Bray (eds) , Genetic resources of forage plants. CSIRO,
Melbourne, Victoria, Australia.
203
A REVIEW OF PASTURE AND FODDER CROPS GERMPLASH EVALUATION
AT UYOLE, MBEYA IN TANZANIA
T.J. Myoya and N.J. Mukurasi
Uyole Agricultural Centre, P.O. Box 400, Mbeya, Tanzania
Abstract
Since 1970 different pasture grasses, perennial legumes and
annual fodders have teen introduced and evaluated at Uyole.
Initial introductions were temperate species among which only
Trifolium repens proved adapted to the Kitulo plateau where they
are presently grown on a large scale. Further evaluaticns
involved both tropical and subtropical species with Rhodes grass
("Chloris gayana Kunth) as a locally adapted reference species.
At Uyole several introduced species have outyielded Rhodes grass
and future work will look at the chemical composition of these
species at different growth stages. Among the legumes, Desmodium
spp. cv Greenleaf and Silverleaf have shown good adaptation at
Uyole and a few of the substations. These species have mixed
well with most grasses. Neonotonia wightii cv Tinaroo and
Macroptilium atropurpureum have also shown good promise at Uyole.
Amongst the perennial fodders, Napier grass (Pennisetum
purpureum Schumach) has shown a wide adaptation to different
localities and is currently used by most small-scale dairy
farmers for fodder.
Lupinus spp. among the annual fodders has grown and stood
well into the dry season at Uyole when planted late in the
growing season. Oats cultivars have lot teen successful due to
poor crown rust resistance. Currer-L evaluations have included
Triticale cultivars which have outyielded oats. Fodder beets
have been evaluated but have failed to bolt.
Animal production evaluations at Uyole have just been
initiated and preliminary results have indicated that a grazing
pressure of as high as 32 cows/ha can be achieved on some of the
forages and fodders without affecting milk production
drastically.
INTRODUCTION
The Southern Highlands of Tanzania cover vast areas of natural
grasslands with varying land topography where the overstorey
comprises Miombo woodland. On the hillsides and on undisturbed
flatlands Hyparrhenia and Themeda spp. form a dense cover. These
grasslands often lack local legumes, and therefore increased
animal production (beef or milk) cannot be expected. The
increasing population pressure especially in Mbozi, Rungwe and
Mbeya districts have forced farmers to develop a forage-feeding
system where very small areas could be grown with improved
pastures.
Due to the low yield and quality of natural pastures a
search into better quality grasses was initiated in 1970 at Uyole
to cater for the four regions of Iringa, Mbeya, Rukwa and Ruvuma.
204
The work reported is mostly from Uyole, and it is only recently
that germplasm evaluation has been started in Iringa and Rukwa
Regions.
This paper broadly covers germplasm evaluations carried out
at Uyole (Long. 33° 32 'E; Lat. 8° 55'W) at an altitude of 1800 m
with an average rainfall of 870 mm. Mean maximum temperatures
range from 21-27°C and mean minimum temperatures of 8-14°C
(Heady, 1978) . The soils developed from volcanic parent material
and contain pumice. They have recently been classified as
Eutrandepts (Soil survey USDA classification) (Kamasho and Singh,
1982). They have a medium to slightly acid reaction (pH 6.5) and
have a light texture (clay-loam) . Phosphorus levels hardly reach
5 ppm. Copper is deficient in such soils.
PASTURE PLANT EVALUATION
Grasses
Initial grass introduction started in 1970 at Uyole from Sweden.
They were mainly temperate grasses. A list of first
introductions is shown in Table 1. In two seasons most temperate
grasses proved a failure at Uyole despite a high nitrogen
application rate of 120 kg N/ha. Only perennial and annual
ryegrasses were worth harvesting. However these did not survive
the dry season and due to poor seed set they had to be abandoned.
However, Lolium spp. have proved a success on Kitulo plateau
(2900 m.a.s.l.), and the species is now grown on a large scale
for dairy production.
During the same season Rhodes grass and Nandi setaria were
introduced from Kitale and compared at Uyole under different
nitrogen levels. Further comparisons with Congo signal, Italian
ryegrass and perennial ryegrass showed the latter to be low-
yielding in terms of dry matter as indicated by Table 2. Among
all grasses Nandi setaria showed a much higher nitrogen
efficiency followed by Rhodes grass. The two temperate species
showed both low yields and low nitrogen efficiency. After the
second season the temperate species were attacked by cutworms and
had to be dropped out. Their seeding gualities were also poor.
No chemical analysis was made on the different grass species
due to lack of analytical facilities at the centre. Further
evaluations had then to be based on Rhodes grass and Nandi
setaria for their proved high yield with a good nitrogen
response. Further comparisons were made involving molasses grass
and coloured guinea grass (Anon, 1976) . Both coloured guinea and
molasses grass gave 5-6 t DM/ha in three cuts at 4 0 kg N/ha
fertilizer rate compared with 6 t/ha for Rhodes grass.
During 1974 further introductions were made from Kitale,
Kenya where Pokot Rhodes, Mbarara Rhodes, Masaba Rhodes and
Molasses grass were compared with the locally growing Rhodes
(Mbeya ecotype) . All Rhodes cultivars did not yield
significantly higher than Mbeya ecotype at all nitrogen levels
(0-350 kg N/ha) while the rest of the grasses actually gave lower
205
Table 1. A list of grass species introduction at Uyole in 1970.
Species Cultivars DM yield kg/ha
1970/71
1971/72
Phleum pratense
(Timothy) Vallo - -
Kaumpe II - -
Omria - -
Engmo - -
Bottnia II - -
Festuca pratensis
Vanadis - -
(Meadow fescue) Mimer - -
Sena - -
Bottnia II - -
Dactylis glomerata
Loeken - -
(Cocksfoot) Frode - -
Poa pratensis
Tardus II - -
(Bluegrass) Flyking - -
Lolium perenne
(Perennial ryegrass) Delta 3490 -
Viva 2720 -
Lolium multiflorum
Viris 3240 -
(Annual ryegrass) Imperial 6580 1770
Novita 5000 1400
Barwoltra 5700 -
Barmultra 5270 -
Tawera 5540 1500
Tetrone 5410 -
S.V. 02050 - 1810
H.G. 7222 — 1470
yields. During 1975/76 season a chemical analysis based on %
crude protein and in vitro dry-matter digestibility were carried
out on Pokot Rhodes, Nandi setaria and Mbeya ecotype Rhodes as
shown in Table 3 for three boot-stage cuts.
On the whole, Nandi setaria contained more crude protein at
most nitrogen levels. Variation among cuts was not expected
since the cutting interval and growth stage was the same for all
cuts. INVODMD seemed not to be affected by the nitrogen level.
Quality changes were not pronounced up to 150 kg N/ha implying
higher N dressing requirements on the low N status soils of
Uyole. Pokot Rhodes gave lower protein content than Mbeya Rhodes
which also displayed lower INVODMD. To date, seed increase and
distribution is based on Mbeya Rhodes because of its superiority
over the introduced cultivars from Kenya.
206
$3
S
o
u
a
I
5
207
Table 3. Crude protein (% of DM) and dry-matter digestibility of
three grasses under nitrogen application (Trial 116/74) .
1st cut
(Jan. 1976)
CP INVODMD
(%) (%)
2nd cut
fMarch 1976)
3rd cut
Nitrogen
KgN/ha/
cut
fMav 1976)
Grass species CP INVODMD
(%) (%)
CP INVODMD
(*) (%)
0 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
9.2
9.6
9.3
57.5
57.7
66.3
63.3
75.0
70.4
10.6 52.9
58.0
64.6
9.2
- 10.6
50 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
8.1
7.2
8.6
59.4
60.1
64.8
8.0
7.8
8.1
54.6
59.1
65.6
7.9
5.8
8.3
63.5
59.1
66.2
100 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
9.4
8.7
63.3
65.4
60.4
14.2
8.2
9.6
59.3
59.5
55.8
8.0
6.5
9.0
55.8
66.3
64.410.3
150 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
11.0
11.9
14.2
63.8
67.2
61.4
13.0
10.9
12.6
53.5
66.1
57.0
15.5
9.1
53.2
60.0
67.311.4
200 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
15.7 60.0
63.9
61.4
10.5
14.6
17.3
63.2
68.4
71.8
12.1 58.9
66.1
65.1
15.5
16.6
9.6
11.9
250 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
17.5 64.7
62.2
60.2
18.3
16.2
16.8
69.4
61.5
60.5
13.1
12.2
15.2
57.8
68.8
67.6
15.4
19.2
300 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
17.0
16.1
18.7
62.0
59.7
64.1
16.9
15.1
16.5
60.4
64.2
11.3
15.8
55.8
64.2
71.1
350 Mbeya
Pokot
Nandi
Rhodes
Rhodes
Setaria
17.6
16.7
19.9
61.1
63.5
61.3
17.7
17.4
18.3
59.2
60.5
63.2
12.1
12.9
15.5
53.2
68.9
67.9
Legumes
Due to the increasing number of small-scale dairy farmers in the
Southern Highlands, the reliance on fertilizer nitrogen was
considered not possible as in most cases the capital available is
not enough even for the food crops. The highlands' natural
pastures lack local legumes and most of them had to be
introduced. Introduced legumes include Greenleaf Desmodium,
Silverleaf Desmodium, Siratro, Centro, Puero, Calopo, Clitoria,
Neonotonia, Kenya white clover, Subterranean clover, Cook stylo,
Hamata stylo, Seca stylo, Red clover, Lablab and Vicia or common
vetch. Initial legume evaluations started in 1972. Most legumes
did not germinate (Trial 105/72) except Greenleaf and Silverleaf
Desmodium. Most of these were again introduced from Australia in
1981 as will be discussed under mixtures. Both cultivars of
Desmodium have proved well suited to Uyole (1800 m.a.s.l.) Mbimba
(1525 m.a.s.l.), Rungwe district (Mbeya) and Njombe district
(Iringa) . Further evaluations are now being carried out at
203
Mitalula (1055 m.a.s.l.) Nkundi (1975 m.a.s.l.) and Ismani (1370
m.a.s.l.). Seed increase Is now being carried out for the two
cultivars for further evaluation in different localities and
distribution to areas where the legume has already been tested.
Mixtures
During 1971/72 season all the above legumes were grown in mixture
with either Rhodes grass or Nandi Setaria. Most legumes did not
germinate and those that germinated did not form an adequate
ground cover, and some did not tolerate the dry season. Further
mixture evaluations were made in 1979-1986 season (Trial 505-79)
and results are shown in Table 4 .
Table 4. Dry-matter yields (kg/ha) of mixtures and their
components in pure stands (yield aggregates for three
cuts/annum) .
Year Mean pure Rhodes/ Rhodes Silver- Green- Silver-
grass Greenleaf leaf Desmodium leaf leaf
yield Desmodium
mixture
mixture
1980 11670 15580 14230 13810 11810
1981 3095 5980 5290 6760 5150
1982 4935 5330 4190 5870 4290
1983 8700 7250 6580 7340 7650
1984 4210 4130 4740 3920 5280
1985 3665 3650 3870 4100 4040
1986 6115 5420 5370 5510 5300
Means 6055 6760 6320 6760 6220
Standard
deviation +3100 ±4063 +3597 +3355 +2728
The overall mixture effects were masked especially after the
second year for the two legume cultivars, and this was worsened
after the fifth year due to ingress Hyparrhenia sp. During the
first year a chemical analysis was made on all the three cuts as
shown in Table 5. Only crude protein was determined due to lack
of facilities for other analyses.
Irrespective of cut number, there was an increase in the
crude protein content when Desmodium was included. Pure legumes,
however gave higher values than the mixture.
209
Table 5. Crude protein content (% dry-matter bases) of pure
Rhodes, Rhodes/Desmodium and pure Desmodium at Dyole.
Cut No. Pure Rhodes/ Greenleaf Pure Rhodes/ Silver-
Thodes Greenleaf Desmodium Rhodes Silverleaf leaf
Desmodium Desmodium
1st cut
9/1/80
2nd cut
17/3/80
3rd cut
18/6/80
11.3
8.8
7.0
13.7
13.3
10.2
19.2 10.8
13.9
13.7
8.5
5.9
13.2
13.5
10.2
18.0
19.2
9.0
During the 1981/82 growing season different grass and
legumes species and cultivars were introduced from Australia.
Among the grasses we had:
Cenchrus ciliaris cv Biloela
Panicum maximum var. typica (common Guinea)
cv Riversdale.
Panicum maximum cv Gatton
Panicum maximum cv Hasil
Setaria anceps cv Narok
Setaria anceps cv Nandi
Setaria anceps cv Kazungula
Brachiaria decumbens (Signal grass) cv Basilisk.
Paspalum plicatulum (Plicatulum) cv Rodd's Bay
Paspalum plicatulum (Plicatulum) cv Bryan
Urochloa mosambicensis (Sabi grass) cv Nixon.
The legumes included:
Centrosema pubescens cv Common Centro
Neonotonia wightii cv Tinarro
Macroptilium atropurpureum cv Siratro
Medicago sativa (Lucerne) cv Hunter River
Stylosanthes guianensis cv Cook
Stylosanthes hamata (Caribbean stylo) cv Verano
Stylosanthes scabra (Shrubby stylo) cv Seca
Trifolium semipilosum (Kenya white clover) cv Safari
Trifolium repens (white clover) cv Haifa.
The grasses and legumes were planted in the field at Uyole,
the grasses being either in pure stands or in mixtures with the
different legumes. Due to seed shortage plot sizes had to be
reduced to 9 m unreplicated and arranged in a split-block
design. Weed control was done by the hand hoe during the first
year of establishment and depending on the growth habit of the
species, no weeding was carried out during the second year.
210
Centro was very slow to establish compared to most other
legumes. Stylos showed good performance during the first year
only while lucerne was completely eliminated by most grasses.
Though suited to the drier inland regions Buffel grass cv Biloela
performed adequately at an altitude as high as 1800 m.a.s.l.
All grasses proved to be drought resistant and although they
were susceptible to night frosts in July/August they recovered
rapidly. An exception was Urochloa mosambiscensis which did not
recover after the fourth year. In addition Urochloa was
completely shaded and killed by both Desmodium and Neonotonia
spp. Most grasses were able to control weeds effectively except
Urochloa, Paspalum plicatulum cv Bryan and Setaria sphacelata cv
Narok. Paspalum was also completely shaded by both Desmodium
spp. or Silverleaf. The only grasses that tended to show a good
balance with the desmodiums were Brachlaria decumbens and Setaria
sphacelata cv Kazungula.
Among the legumes only Siratro, the Desmodium and Neonotonia
spp. showed promise. Legumes like Centro, clovers and the stylos
were all suppressed by the grasses. Despite the promise shown by
Neonotonia, the legume failed to flower at Uyole and at Mitahila
(1055 m.a.s.l.). One Centro has proved to flower in three months
from planting at that altitude.
Further grass evaluation in pure swards are currently being
done on Rompha grass, canary grass,' Entolasia imbricata, Cynodon
nlemfuensis , Cocksfoot and Panicum coloratum. They are compared
with Rhodes grass and Themeda triandra as standard checks. With
the legumes, further evaluations include Crotalaria intermedia,
Archer dolichos, Berseem, Rhynchosia and Neonotonia wightii
(Mbeya type) .
Evaluation in terms of animal production has just been
started at Uyole and there are indications that on a fertilized
Rhodes/Desmodium mixture a grazing pressure of 32 animals/ha can
be used without effecting milk production.
Perennial Fodder Crops
The major problem the small- and large-scale farmers face in the
highlands is shortage of green fodder during the six-month dry
season. To solve this problem fodder crops have been introduced
and evaluated. These are discussed below.
Napier grass
This crop was introduced at Uyole in 1971 from the surrounding
districts in Mbeya Region. Evaluations at the Centre showed it
to be high yielding. Due to the increase in fertilizer prices,
the evaluation included comparisons with Silverleaf Desmodium.
Three-years' average data are shown in Table 6.
This kind of evaluation indicated benefits for the mixture
in dry-matter yield which could help farmers release inorganic
fertilizers for other crops. For two consecutive seasons a
chemical analysis was done on the first cut of the mixture, and
it was clear that protein content increased with the inclusion of
the legume as indicated in Table 7 .
211
Table 6. Dry-matter yield (kg/ha) of Napier grass
mixture at two nitrogen levels.
Desmodium
Mixtures
Nitrogen application-
kg N/ha/season
0 8 ON Mean
Napier grass alone
Napier grass + Desmodium
within rows
Napier grass + Desmodium
between Napier rows
5050 10140 7590
12580 15730 14150
11880 15440 13650
Source: Myoya (1980)
Table 7. Crude protein content of fodder Napier when fertilized
or grown in association with Silverleaf Desmodium.
Fodder type
Crude protein
fon % DM basis)
1973/74 1974/75 Means
Without NJ Napier alone
4 0 kg/ha N, Napier alone
Without N, Napier + Desmodium
4 0 N/ha N, Napier + Desmodium
5.4 4.3 4.9
5.9 4.6 5.3
12.9 11.4 5.3
10.4 9.8 10.1
Applications of 40 kg N resulted in a small crude protein
increase as compared to the legume mixture effect. Though the
crop was hit by night frosts in July, it recovered at the end of
August. Growth commenced immediately, and it became available for
feeding in the October-November period.
Due to much interest in Napier fodder, four new cultivars
were introduced from Kitale, Kenya in 1975 and dry-matter yield
assessed for five consecutive years. Only average data is shown
in Table 8 (1977-1981). Despite the high yield obtained from
Gold Coast, which almost approached that of Mbeya ecotype, its
hairiness limits its use as they produce skin irritations
especially when used under cut-and-carry . French Camerouns
produced relatively high yields and large amounts have already
been planted by the surrounding small-scale dairy farmers. The
crop has successfully been introduced in other localities like
Mbozi (1525 m.a.s.l.), Nkundi in Sumbawanga (1750 m.a.s.l.) and
Mitalula (1055 m.a.s. 1. ) .
212
Table 8. Dry-matter yield of different Napier grass cultivars at
varying nitrogen application rates (kg/ha) .
Nitrogen
Cultivars
level
(kg N/ha) Gold
Coast
French Mlingano Ethiopia Mbeya
Camerouns
0 3720 3860 1910
2430
4580
6380
1400
2290
3800
7970
3850
60 4950 5890 5960
120 7780 8260 6910
240 14290 13260 15800
Russian Comfrey (Symphytum peregrinum)
This is reported as a suitable fodder crop for cattle, horses,
sheep, pigs and poultry. Observations at Uyole confirm Strange' s
(1959) findings that it is never touched by cattle when freshly
cut. The crop was introduced from Tengeru in Arusha in 1977.
Dry-matter yield averaged 6 t/ha under nitrogen applications of
160 kg N/ha. A chemical analysis of the crop indicated 17.9%
crude protein, 10.9% crude fibre, 3.2% crude fat and 24.4% ash
all on dry-matter basis.
Lucerne: Among the perennial fodder legumes that have received
much attention at Uyole is lucerne. Initial work in Tanzania
started in Northern Tanzania where the cultivar Hairy Peruvian
was compared with the cultivars Saladina and Hunter River. Work
at Uyole started in 1971 with introductions of cultivars Tuna and
Du Puitis (Anon, 1972) from Sweden. Due to poor germination and
heavy weed competition these cultivars were lost.
During the 1974/75 season 16 lucerne cultivars were re
introduced from Sweden. Evaluations based on dry-matter
productivity showed that at the onset of the dry season good
quality hay can be produced from lucerne.
However, during the 1973/74 season the cultivar Hairy
Peruvian was planted in replacement series with Rhodes grass cv
Pokot and nitrogen application included in order to measure the
possible yield contribution of the mixture. During the first
season a seed rate mixture of 3.5 kg Rhodes grass and 4 kg
lucerne produced the highest yield. The two-year average data is
shown in Table 9.
The results not only demonstrated the advantages of the
mixture but also indicated that lucerne benefited from small
quantities of nitrogen application. Without nitrogen the mixture
had a 12-14% CP content while with nitrogen application this was
13.9-15.8% CP, showing a small increase. Further evaluations
stopped due to seed unavailability. At this altitude (1800
m.a.s.l.) there were not enough pollinators for the crop and
hence no seed yield.
213
Table 9. Dry-matter yield of Rhodes-lucerne mixture, kg/ha.
Nitrogen application, kg N/ha
60 Means
Pure Rhodes
Rhodes lucerne
Pure lucerne
3860
6890
6070
8210
9940
8260
6035
8415
7165
Source: Anon (1977, 1980).
Annual Fodder Crops
These are short-term crops that cannot only be used to overcome
seasonal feed shortages but also as main producers of roughage.
They can be used as pioneer crops in new lands and they can
easily be integrated into mixed crop-livestock enterprises. They
can be used for direct grazing, made into silage or fed as green
chop.
Oats (Avena sativa)
At higher altitudes in the highlands cool-season crops like oats
can be grown. During 1970/71, ten oats cultivars were introduced
from Sweden, Finland and Holland. These were planted in single
rows and their seed yield potential assessed. Most cultivars
showed a vigorous vegetative growth, but due to disease attack
very few panicles were formed and hence low seed yields were
recorded (Table 10) . Comparisons were made with the locally
grown cultivar "Suregrain" from Canada which was introduced at
Uyole in 1970 for use as a companion crop in the establishment of
grass legume pastures, silage and green fodder production.
Table 10. Seed yield of different oats cultivars at
matter yield and rust score trial.
Dyole, dry-
197 0/71 1971/72
Dry-matter Rust score"
Cultivar Days to Seed yield yield
heading (kg/ha) (kg/ha) (1-70)
SOIL II - - 5,400 5
Pendek 104 250 5,720 -
Hannes 109 500 - -
Same 104 130 - -
Nil 83 420 5,380 3
Titus 104 310 4,570 6
Stormogul - 170 - -
Sorbo - 130 4,330 10
Linda 104 780 5,920 6
Condor 111 740 5,650 5
Selma 103 760 6,620 5
Suregrain 93 480 4,050 0
Nina " " 5,380 5
a Crown rust assessment: 0 = No attack; 10 = Very severe attack.
214
During the following season two new cultivars Nina and Sol
II were added to the evaluation and the dry-matter yield measured
at milk ripe stage. Rust assessment was made on all cultivars
and there was an indication that most introduced cultivars had
severe to very severe attack. The locally grown oats had no
attack at all. These assessments tended to show that higher seed
and dry matter yielding oats can be obtained but their problem
was attack by crown and stem rust, resulting in a complete crop
loss in cooler years.
During the 1975/76 season 363 spring oats cultivars were
introduced from the Royal Veterinary and Agricultural University,
Sweden. These were planted in single rows whose purpose was to
use intra-varietal diversification in the form of multiline
cultivars to supplement the conventional resistant cultivars as
part of the programme of crown rust control. The cultivars that
did not set seed due to crown rust attack were abandoned in the
first season. Only 25 cultivars were then selected based on
resistance to crown rust. These were grown for four consecutive
seasons and their mean dry matter, seed yield and other
characters are as shown in Table 11. The cultivars Palestine and
Orient were outstanding for earliness of maturity which was under
100 days. Based on crown rust assessment, Suregrain was the most
resistant cultivar followed by C. Iowa, Clinton, Gidgee, C.I. 3821
and Ballidu.
During the 1982/83 season, the locally adapted Suregrain
cultivar was tested for its nutritive quality in comparison with
Napier grass (Table 12) . Two trials were conducted at different
growth stages of oats.
The results showed oats to be highly digestible, but
digestibility falls with age. The normal harvest stage for
Suregrain is about 135 days where it could be at milk-ripe stage.
Due to crown rust attack on the oats, different Triticale
cultivars have been introduced, and assessments are in progress.
So far, it has been shown that most Triticale cultivars are
resistant to rusts and they give much higher dry-matter yield.
Triticale cultivars have also shown a much higher seed-yield
potential (Table 13) . Most oats and Triticale cultivars
outyielded the standard cultivar Suregrain.
215
^-na
9'O. (O-O)
9-
fE
£--
z-g
- -
OO 9-
-
T'O
£-- -
c-g L- E- Z-
-9
O6
-
L-
- -- -9 --
O- g-o
Ti■H
q■-q■-
(%)
--
-TE -Ofr
Z-fr
OEE
T'frE
8-E 8-E
-OE
O'-
OEE
--ev -i.Z
e-se
--
>-og
L--
--
-9»
--Z
-Ll-
0"-
--e
frOE
--
frOE
OO'T
inX»UJ95(
(6)
-frZ
OTZ O- OO
fr'O SOE O-E
-O -O -O
O-
-O
OO
frOE EOE LOZ EOE
6"O -O
xm-z EOE
-TZ
--E -- --
OOE
-q^uoH
S&BJOJO3
3S9A2-L(
- -
T-g T-g
- --
c-g
9'-
g-
-E 6O -
9-
--
O»
- -
ft vg
T-g
-E
9'-
-- -» -»
9-
PTS-X---
(iq/fi*)
AXZ ZAT 8- OO
8O
O- OE OT O-
EO
-Z
LZ- S9C EiE 9- EO
-8T --
OT
OfrE LLE
8O
9-
OZ
EOfr OO
O
(eii/63()
OOfr OfrEE
OfrE
OTOfr OiiE
O8E
OOE
O>E O-
OOE
OO
OOE
O-
O-Z
OO
O-frE OgO OOZ OiO OT- OT- OOE
Ofr£
O9-
-iE
O-
-ba-^xho
xbCv
i/HOI-0 UO^UTTO
iAOI'H-W
i«OI-iJJ'I'H
--fi--o
i-x-H
iT9SJiM
3Tia
^u3i-o
"PT-^a
uirnifix-j
q---g-x iea
O--I'OBUT^ -x-d
O--I'O
TZ8E"I-O
-OUTS--qiv -OUTSS-qi'»
e-uiq-
-su-q-ca
3UT^-S-Tid
frZfrZ-O-o
UTijfil-j (.19:5(00)uiijfix-J
8-L-I'O
UTc-fie-n-
-(T8--8A6T)i^ p-fii^e-iji-- noj)-Ji T x--
--sq-03T3Sjo^u ^■oTT™1P<«iID--/ix u ^tOO'T' x f^- se. ia
-TTSTqcj,
fM
Table 12. Dry-matter digestibility and intaXe of oats and Napier
grass at different growth stages.
Trial I Trial II
Parameter
Oats Oats
82 days 156 days
Napier
grass
Oats
97 days
Oats Napier
171 days grass
Digesti
bility, %
intake
(g/kg W0-75
64.5 55.2 32.9 54.7 49.8 47.1
77.0 41.0 34.9 47.2 50.0 45.8
Table 13. Comparison of selected forage oats and Triticale
cultivars for dry-matter and seed yield (kg/ha) .
Species Cultivars
Dry-matter yield Seed
(kg/ha I yield
kg/ha
1983 1984 1985 Mean 1983
2930 5900 7650 5490 300
4350 6910 8530 6600 730
4470 5340 8240 6020 1030
4850 4680 8770 6100 900
4670 6020 7180 5950 800
5520 4690 7800 6000 1630
5640 4920 7370 5980 2370
5780 5040 8980 6600 1540
5800 6760 7850 6800 1450
5740 5620 9230 6860 1170
Oats Suregrain
C.I. 3920
C.I. 7488
Avena abyssinica
Palestine
Triticale Til - 152
Til - 234
Til - 158
T10 - 9
T10 - 211 - 8
Sorghum
No sorghum germplasm introductions were made at the Centre until
1982 when seed of Sorghum almum cv Sugardrip was obtained from
Australia. Since it is a crop that can be rationed, it was grown
either in pure stand or in mixture with Greenleaf and Silverleaf
Desmodium, Seca Stylo, Cook Stylo, Siratro and Tinaroo
Neonotonia. Average data for two seasons indicated sorghum to
give 8 tonnes DM per hectare in pure stand when cut three times
per season while the yield was drastically reduced in mixtures
with legumes as indicated by the data in Table 14. Such mixtures
proved a failure during the third season as most sorghum plants
were completely smothered by the legumes.
217
8090 - 8090
1330 1700 3030
390 1270 1660
710 1110 1820
410 12780 13190
170 16670 16840
1030 7870 8900
Table 14 . Dry-matter yield of Sorghum almum cv Sugardrip when
grown in pure stand or in mixture with different
legumes.
Mixtures: Sorghum yield Legume yield Total
kg DM/ha kg DM/ha
Pure sorghum
Sorghum + Seca stylo
Sorghum + Cook stylo
Sorghum + Siratro
Sorghum + Silverleaf Desmodium
Sorghum + Greenleaf Desmodium
Sorghum + Neonotonia
Lupines
The introduction of lupines at Uyole was aimed at improving soil
fertility through crop rotations, increasing quality of the
available fodders and supplying green feed during the early to
mid-part of the dry season. During the 1971/72 season two
species of lupines (Lupinus albus and Lupinus angustifolius) were
introduced from Sweden and subjected to a planting dates trial
the results of which are presented in Table 15.
In this particular evaluation it was shown that lupines
could be planted as late as mid-March without affecting either
seed or forage yield. Lupines planted mid-March would be ready
for feeding in mid-July under Uyole conditions. For grain
production, late planted lupines grew longer into the dry season
resulting in much shattering of the seed.
Having established that lupines could help solve feed
quality and supply problems, more cultivars were introduced in
1974 for further evaluation. Results showed the L. albus
cultivar Kulina to be the most outstanding. The most promising
ten cultivars are currently being further tested for dry-matter
yield and alkaloid content.
218
Table IS. Dry matter, seed yield (kg/ha) and 1000 kernel weight
of two lupine species under different planting dates.
Planting DM yield
kg/ha
Seed yield
kg/ha
1000 kernel
weight (g)
1. Blue sweet lupines L. albas
2.12.71 9500 1565 118.6
22.12.71 7330 1077 130.6
17.1.72 7930 67 -
1.2.72 8360 358 148.8
19.2.72 9860 - -
10.3.72 6890 450 -
2. White Sweet Lupine (L. angustlfolius cv uniwhite)
2.12.71 9180 1047 116.3
22.12.71 6070 962 138.5
17.1.72 7780 617 144.5
1.2.72 8320 973 157.1
19.2.72 9160 865 157.1
10.3.72 9270 1300 —
Source: Anon (1973) .
Fodder-Sugar Beet*
Germplasm evaluation started in 1971 at Uyole with the
introduction of two beet cultivars (Aring Barres fodder-beet and
Triumf sugarbeet) from Sweden. The sugar beet Triumf outyielded
the fodder beet Aring Barres in a progressive harvesting date
trial (Table 16) . The dry-matter content of leaves varied from
9% for the first harvest data to 60% during the last for both
cultivars; while that of roots varied from 9* to 19%. This trial
demonstrated the advantages of growing beet for fodder. when
planted in January the crop can be left in the field up to
November without very much affecting the yield.
Thirty-three cultivars of fodder beets and 42 cultivars of
sugar beets were further introduced. The sugar beets gave much
higher yields than fodder beets, and in most cases sugar beets
had a dry-matter content of above 25% while fodder beets had
below 20%. There was a tendency for declining yield with the
progressing years which could probably be due to inadequate
phosphorus fertilization which was at the rate of 44 kg P/ha.
Beets have been reported from California to require rates of
between 49.3 kg to 148 kg P/ha (de Geus, 1973). Sugar beets gave
yields ranging from 3-7 t DM ha, whereas the fodder beets gave a
yield range in excess of 6 t DM/ha. The major problem
encountered with this crop is seed production as the plants
cannot bolt in Tanzania.
219
Table 16. Dry-matter yield of fodder-sugar beets under delayed
harvesting at Uyole (kg/ha) roots and leaves (planted
in January 1971) .
Harvesting
date Aring Bares Triumf
5/4 4360 5630
3/5 6580 9330
1/6 9340 12390
1/7 8100 11890
2/8 7520 10890
2/9 8480 10370
1/10 8160 11880
8/11 7670 10250
7526 SD + 1506 10329 SD ± 2154
Fodder Shrubs
Leucaena
This is one of the first fodder shrubs to be introduced at Uyole.
Introductions started in 1981 from coffee plantations in Mbozi
district where it was planted as a shade tree for coffee.
Further introductions were made in 1982 from Australia. These
were var. Peruvian ex. Aust. Yates; K28; K8 ; Cunningham; var.
Peruvian Ex. Tengeru, Kongwa and Morogoro. All the introduced
cultivars were shrubs except that collected from Mbozi which
grows to 7m high when undisturbed. So far no data have been
collected on dry-matter yield, but from observations var.
Peruvian Ex. Aust. Yates has a more prolific branching habit and
is leafier. Seed production of all cultivars is quite good.
Galega orientalis (Goats rue)
This particular shrub legume was introduced from Helsinki
University, Finland in 1986. Growth during the first year was
very slow. Information from Helsinki (Varis, 1986) indicates
that its quality can exceed that of lucerne. Observations on it
are taking place.
Desmodium ransonii
This legume was introduced from West Kilimanjaro in 1984. It has
proved to be a non-branching legume that grows to a height of
about 8 m. It has a very weak stem and even at that height it is
less leafy.
CONCLUSION
The suite of germplasm introduced and evaluated is wide. It
ranges from the annual perennial fodders of temperate origin to
the tropical/subtropical grasses and legumes including
shrubs/trees. The promise of some of the species has to be
matched with feed requirements of the existing livestock
220
production systems in the Southern Highlands region of Tanzania.
Research efforts have logically to be directed towards studying
the animal production potentials of some of these forage and
fodder germplasm.
REFERENCES
Anon. 1972. Annual Research Report (Arti-Mbeya) 1970/71.
Anon. 1973 . Annual Research Report (Arti-Mbeya) 1971/72.
Anon. 1976. Annual Research Report (Arti-Mbeya) 1972/73; 1973/74.
Anon. 1977. Annual Research Report (Arti-Mbeya) 1974/75.
Anon. 1980. Annual Research Report (Arti-Mbeya) 1975/76.
Anon. 1986. Annual Research Report (Arti-Mbeya) 1982/83.
de Geus Jan. G. 1973. Fertilizer guide for the tropics and
subtropics. Centre d'Etude de L'Azote. Zurich.
Gohl. 1975. Tropical feeds: Feeds information summaries and
nutritive values. FAO Feeds Information Centre, Animal.
Prod. and Health Div., Rome.
Heady, H.F. 1978. Range and wildlife management in the tropics.
Longman , London .
Kamasho, J. A. and Singh, B.R. 1982. A case study of copper and
zinc in Mbeya district. J. of Pedology, Ghent.
Mukurasi, N.J. 1984. Alfalfa and grass production in a high
altitude monsoonal climate; Tanzania. Swedish Univ. of
Agric. Sci. Publication Report No. 131, Uppsala.
Myoya, T.J. 1980. Pasture research 1970/78. Research Report No.
28, UAC.
Steward D. 1961. New ways with seeds of sugarbeet In: Seeds:
The Yearbook of Agriculture, US Dept. of Agric, Washington,
D.C. pp. 199-205.
Varis, E. 1986. Galega orientalis (Goats rue) , a potential
pasture legume for temperate conditions. Journal of Agric.
sci. in Finland 58:83-101.
221
REVIEW ON EVALUATION AND AGRONOMY OF PROMISING
LUPINE VARIETIES IN KENYA
F.N. Muyekho
Forage Agronomist
National Agricultural Research Station
P.O. Box 450, Kitale, Kenya
Abstract
This review deals with a series of agronomic and evaluation work
on promising lupine varieties in major agroecological zones of
Kenya. The trials examined regional performance of different
varieties of lupine and the effects of inoculation and phosphatic
fertilization. Results from early observation trials indicated
that lupines performed better in the medium and high altitude,
high potential areas. Other areas appeared unsuitable for this
crop partly due to inadequate moisture levels.
Reported observations showed that cultivars Unicrop and
Ultra were early maturing and were relatively shorter in height
compared to Uniwhite and Uniharvest both of which were taller and
later maturing. Moreover, further work revealed that cultivars
Uniharvest, Ultra and Uniwhite had higher grain yield and higher
protein (42.8%, 40.7% and 35.8% respectively) compared to Unicrop
(31.0%). Cultivar Ultra was the highest seed yielder. Only
small increases in seed yield and quality were reported when
inoculants and phosphatic fertilizers were used suggesting that
these treatments may not be beneficial under certain conditions.
INTRODUCTION
Lupines are annual or perennial legumes belonging to the genus
Lupinus which contains 250 species (Highes et al . , 1952). Most
of these are native to America. The large-seeded species of
commercial importance, however, are all native to the
Mediterranean basin in Europe. These include blue lupine (L.
angustifolius and L. hirsutus) , white lupine (L. albus) and the
yellow lupine (L. luteus) . Others are the Spanish lupine (L.
hispanicus) and Texas lupine (L. mutabilis) which have been used
for more than 3000 years as a subsistence crop in the Andean
Highlands (Lees, 1979) . Some of the cultivated lupines are
bitter and unpalatable but excellent producers of nitrogen.
In Kenya, there are two species of sweet lupines whose
agronomy and nutritive evaluation has received considerable
attention. These are the blue lupine cultivars Uniwhite,
Uniharvest and Unicrop and the white lupines represented by the
cultivar Ultra (Wanjala, 1979) . All of them are annual in
nature. These varieties with high foliage, grain yields and high
crude protein content were shown to take 4-7 months to mature
depending on variety (Gladstones, 1970; Wanjala, 1979). The main
objective of conducted work on lupine in Kenya was to establish a
cheap source of protein as a concentrate for livestock
supplementation. Trials undertaken at the National Agricultural
Research Station, Kitale, included testing the adaptability of
lupine varieties in different agroecological zones (Bauer and
222
Oketch, 1976) , the observation of physiological characters of
lupine varieties grown under uniform conditions (Kusewa et al . ,
1977), the comparison of herbage dry-matter yield (Kusewa et al.,
1980, 1981) and the investigation on the effect of phosphatic
fertilizers and inoculation on seed production of the lupine
variety Ultra (Kusewa et al . , 1980, 1981).
RESULTS
Lupine Regional Adaptability
Based on rainfall and altitude Bauer and Oketch (1976) have
reported on the adaptability of lupines in nine sites in Kenya
(Table 1) . The lupines were planted at a recommended spacing of
50 x 30 cm, Bauer and Oketch (1975) in unreplicated plots.
Table 1. Performance of lupine varieties in different agro-
ecological zones of Kenya.
Altitude
(m)
Rainfall2
(mm)
Height
(cm)
Variety yield
Seed
t/ha Remarks
Molo 2740 1015-1270 110 Ultra Slow to
swell
Eldoret 2130 1015-1270 119 Ultra 1.8 Lodging
Nyandarua 2331 760-1015 100 Ultra - -
Kitale 1980 1015-1270 60-140 Ultra 1.7
2.2
1.8
1.2
Uniharvest
Uniwhite
Unicrop
Lodged
Shortest
Kakamega 1580 1780-2030 87 Ultra 1.4 Seed shat
tering
Embu 1460 760-1015 79 Ultra 0.81
Katumani1 1680 510-760 41 Ultra
Uniwhite -
All poor
and
diseased
Uniharvest
Unicrop
Kiboko1 1000 300-650 39 Ultra,
Unicrop -
Pods were
weak
Mtwapa 12 1015-1270 25 Ultra,
Unicrop
Uniwhite,
Uniharvest -
Poor and
diseased
Source: Bauer and Oketch (1976)
1 Seed yield not available.
2 Source of mean annual rainfall: National Atlas of Kenya,
third edition, 1970.
223
Medium and high altitude, high potential regions such as Kitale,
Molo, Eldoret, Kakamega and Embu proved to be the most promising
sites for growing sweet lupines. Hot and dry areas such as
Katumani and Kiboko as well as the hot and wet areas such as
Mtwapa (Mombasa) proved unsuitable. There are indications that
for better yields, lupines need 5 months free of serious moisture
stress and mean monthly maximum temperatures ranging from 15 to
25°C (Wanjala, 1979) . These conditions are typical of the areas
where lupines produced favourable results.
Physiological Development of Lupine Varieties
Physiological development of four lupine varieties grown under
uniform conditions was observed at Kitale (Kusewa et al . , 1977).
The four lupine varieties Uniwhite, Unicrop, Ultra and Uniharvest
were inoculated with group "G" inoculant before planting. Single
superphosphate fertilizer at the rate of 40 kg/ha P2°5 was
applied at planting. A spacing of 50 x 30 cm was used with two
seeds per hill at planting.
It was shown (Table 2) that Unicrop and Ultra were early
maturing and were short and medium in height, respectively. On
the contrary, Uniwhite and Uniharvest were tall and late
maturing. Uniwhite was more susceptible to lodging than any of
the other varieties studied. Field observations also indicated
that Unicrop was the most susceptible to powdery mildew.
Considering seed weight. Ultra had a higher weight per 100 seeds
than all the other varieties. It also had more pods/plant than
the others.
Table 2. Physiological development of lupine varieties.
Variety
Observation Uniwhite Unicrop Uniharvest Unicrop
Days to germinate 5 5 5 5
Days to 80% flowering 125 49 91 49
Lodging (%) 30 0 10 15
Days to maturity 202 120 202 150
Shattering of seed (%) 30 5 5 5
Plant height (cm) 140 60 115 110
Pods per plant 28 19 27 45
100 seed weight (g) 17. 3 19.3 17.5 43.5
Source: Kusewa et al. (1977)
Productivity and Evaluation of Lupine Varieties
Kusewa et al . (1980, 1981) conducted a trial at Kitale to
determine herbage dry matter, seed production and nutritive value
of four lupine varieties; Unicrop, Uniwhite, Uniharvest and
Ultra. They were planted under recommended establishment
practices similar to those of the physiological development
study. The lupines were harvested at three stages of growth i.e.
2, 3 and 4 months after establishment. A single plot was
included per variety for the determination of seed at
physiological maturity.
224
Significant differences (P<0.01) were observed in both
herbage dry-matter and seed yield among varieties and harvesting
stages (Table 3). Variety Ultra was outstanding in both herbage
and seed yield. During 1980, maximum herbage yield for Ultra,
Uniharvest and Uniwhite were obtained when they were cut at four
months but highest yield for Unicrop was attained at three
months. In 1981, however, all the varieties had the highest
herbage yield when harvested at four months. Crude protein (CP)
values and in vitro digestibility values (Table 4 and 5) were
higher in the grains than in the herbage in all the four lupine
varieties investigated.
Table 3. Performance of lupine varieties harvested at different
stages.
Herbage DM production kg/ha
2 months 3 months
Harvesting 1980 1981 1980 1981
stage
4 months Seed yield
(Kg/ha)
1980 1981 1980 1981
Unicrop
Uniwhite
Uniharvest
Ultra
760 140 1612 610 1260 990 350 260
510 120 2764 410 3074 1620 947 890
616 180 2227 820 4426 2590 750 1190
692 166 3400 2180 4300 6180 3344 4000
Source: Kusewa et al . (1980, 1981).
Table 4. Effect of stage harvesting on the crude protein content
of lupine varieties.
Harvesting stages (CP %) Seed at physio-
cal maturity
2 months
1980 1981
3 months
1980 1981
4 months
1980 1981
logi
(CP %1
Variety 1981
Unicrop
Uniwhite
Uniharvest
Ultra
21.9
19.0
21.0
20.2
25.4
25.0
20.4
16.6
16.4
17.8
18.9
15.9
26.8
28.6
10.7
23.5
14.3
14.3
17.3
14.3
23.9 31.0
8.2
19.6
13.6
35.8
42.8
40.7
Source: Kusewa et al . (1980, 1981)
225
Table 5. Effect of s
values of lup
tage of harvesting
ine varieties.
on the digestibility
Variety D-value D-value D-value D-value
Unicrop 62.2
Uniwhite 64.3
Uniharvest 62.7
Ultra 66.2
65.2
65.0
64.1
67.5
57.6
56.0
54.7
74.2
85.5
88.8
88.7
84.0
Source: Kusewa et al . (1980 )•
Phosphate fertilizer and inoculation requirements of lupines
A trial conducted by Kusewa et al . (1980, 1981) examined the
seed production of lupine cv Ultra when inoculated and fertilized
with phosphates at the rates of, 100, 200, 300 kg/ha P2°5
Inoculation resulted in more nodule formation per plant whereas
phosphate fertilizer application did not have any effect on
nodulation between the inoculated and non-inoculated lupine
plants (Table 6) . Differences in seed yield and percent protein
content were not significant suggesting that inoculation and
phosphatic fertilization were not beneficial under Kitale
conditions.
DISCUSSION
The four sweet lupine varieties tested were shown to do well in
medium and high altitude, high potential areas of Kenya. The
lupines did not do well in hot and dry areas because of the
moisture stress. These results are in agreement with those of
Heath et al . (1975) who had reported that lupines required cool
weather for best development. Savile and Wright (1958) reported
that sweet lupines grow at altitudes between 2225 and 2743 m and
bitter lupines from 1981 to 2743m. It would, however, appear
that sweet lupines can still do well at lower altitudes than
those recommended by Savile and Wright as evidenced by studies
conducted by Heilbutch (1952) ; Poultney (1963) ; Bauer and Oketch
(1976) and Wanjala (1979) . They reported promising results in
Kitale at a lower altitude of 1980 m.
The cultivar Ultra outyielded other varieties (Table 3) due
to development of numerous pods and heavier seeds. This result
would appear to contradict earlier observation (Table 1) where
Ultra was outyielded by Uniharvest and Uniwhite. However,
experiments conducted by Bauer and Oketch (1975, 1976) ; Kusewa et
al. (1977) and Wanjala (1979) confirm that Ultra is superior to
the others in seed yield. Two farmers in Kitale reported a seed
yield of 780-1100 kg/ha in 1951 and 780-1900 kg/ha in 1952 for
sweet blue lupines (Heilbutch, 1952) . This is within the range
reported in Table 1.
226
*(jO'0O1j•"[«ie*Jiasjj:aeeoo
EjE17Z6XVoO80ZTS8VSOX3
Z0XO0oZEOEO"ZOE6T8»''"M/iPT»T^»jXLTZX1jOZE1iOTTO/s»TjP<>N
80SOEEj8ZEo6Z*X6S"TZZOEX
j0S0o0oEO'OV6LoEZ/6"ZZOLE»i/lP1»T*»jLjoTjOZZEin»o[O/s»_je°N
OO"JooooLeaw00EO«»»W1A
OSj0(ZjOeil/8^<S)0OZjO"H/**
sobbidjo3"Sooi^eo[joo f-uoojo Ti»Te o X
•»Ji1OoTOjijooouwsij^oaja^e Oso OVwooon jjoo ojs^aa^x'TTX
Lack of response of sweet lupine varieties to inoculation
and phosphate fertilization indicates that lupines can still be
grown in Kitale without such treatments. The lack of response
could be attributed to the fact that soils in Kitale have levels
of phosphate adequate to support lupine growth. The results
agree with those of Henderson and Preston (19 59) who recommended
that application of phosphate should only be done on soils of
very low fertility. Work by Bauer (1976) on lupine cv Ultra gave
similar results.
Inoculation should have resulted in greater nodulation hence
higher seed yields. However, this was not so and could be
attributed to ineffective nodulation. This argument is in line
with that of Henderson and Preston (1959) who reported good
nodulation of uninoculated crops in most areas of Kenya, although
they noted that most of the nodules were ineffective. Work
carried out by Wanjala et al . (1983) in one farmer's field showed
a significant seed-yield response of sweet lupine cv Ultra to
inoculation but not to phosphate fertilizer. A more
comprehensive study by de Souza (1969) when working on blue
lupines found effective natural nodules at twelve sites and no
natural nodulation at nine sites.
Experiences gained from this study show that there is need
for lupine variety testing trials to be conducted in different
ecological zones for herbage and seed yield. There is also need
for phosphate and inoculation studies to be conducted in a wide
range of sites. Future agronomic interventions of lupine studies
should involve a study on the relationship of critical elements
like sulphur and molybdenum with seed production and
investigations on the effect of planting dates on seed yield for
both early and late maturing varieties to avoid seed losses
during wet weather. Other aspects to be considered should be
intercropping lupine with food crop or other forages with the aim
of maximising the utilisation of the ever-diminishing land
holdings in the high potential areas.
CONCLUSION
The high dry-matter yield, percent crude protein and high D-
values especially in seed for the various lupine cultivars
indicate that there is great potential for the utilisation of
lupines as a livestock feed supplement. Further studies aimed at
the utilisation of sweet lupines as a feeding material will
elucidate the importance of these lupines in the livestock
industry.
ACKNOWLEDGEMENTS
The author wishes to thank Mr. C.W. Mwendia for his technical
advice in the writing up of this paper and Mr. R.E. Kamudi for
the statistical analyses. Appreciation is extended to Mr. P.K.
Kusewa for his work which formed the backbone of the agronomic
review of lupine. Appreciation also is extended to the Deputy
Director of Research of the Kenya Agricultural Research Institute
for permission to publish these results and to Mr. D.K. Muthoka,
Director of National Agricultural Research Station, Kitale for
his assistance logistically .
228
REFERENCES
Bauer, P.J. 1976. Lupines in East Africa. A review. National
Agricultural Research Station, Kitale. Technical Report No.
12, 1976.
Bauer, P.J. and Oketch, G.O. 1974, 1975, 1976. Annual Reports.
National Agricultural Research Station, Kitale.
de Souza, D.I. A. 1969. Legume nodulation and fixation studies in
Kenya. East African Agricultural and Forestry Journal
34:299-305.
Gladstones, J.S. 1970. Lupines are crop plants. Field crop
abstracts 23 (2) : 123-148 .
Heath, M.E., Metcalfe, D.S. and Barnes, R.F. 1975.
Forages. Iowa State University Press, Iowa.
Heilbutch, W. 1952. Notes on sweet lupines as crop plants.
National Agricultural Research Station, Kitale. Lupine
correspondence file.
Henderson, G.R. and Preston, P.T. 19 59. Fodder farming in Kenya.
East African Literature Bureau, pp. 105-106.
Highes, H.D., Heath, M.E. and Metcalfe, D.S. 1952. Forages.
Iowa State College Press. Ames, Iowa.
Kusewa, P.K., Wanjala, B.W. and Kamau, R.I. 1977. Annual Report.
National Agricultural Research Station, Kitale.
Kusewa, P.K., Wanjala, B.W. , Kamau, R.I. and Kamau, J.N. 1980,
1981. Annual Reports. National Agricultural Research
Station, Kitale.
Lees, P. 1979. Many uses highlight the agricultural lupine.
World Farming 21:16-19.
National Atlas of Kenya 1970. Third edition. Government of
Kenya, Nairobi.
Poultney, R.G. 1963. A comparison of direct seeding and
undersowing on establishment of grass, and the effect on the
cover crop. East African Agriculture and Forestry Journal
29:26-30.
Savile, A.H. and Wright, W.A. 1958. Notes on Kenya Agriculture
III, Oil seeds, pulses, legumes and root crops. -East
African Agriculture and Forestry Journal 24:1-9.
Wanjala, B.W. 1979. An introduction to lupin cultivation. Kenya
Farmer. January. English Press, Nairobi, p. 17.
Wanjala, B.W. , Wandera, J.L., Muyekho, F.N. and Kamau, J.N. 1983.
Annual Report. National Agricultural Research Station,
Kitale.
229
THE EFFECT OF LUPINE/MAIZE MEAL SUPPLEMENTATION
ON THE PERFORMANCE OF LACTATINO FRIESIAN CONS
e.a. Mukisira
National Agricultural Research Station
P.O. Box 450, Kitale, Kenya
Abstract
Two groups each of five lactating Friesian cows were supplemented
with varying proportions of lupine and maize meal. The five
rations contained lupine and maize meal percent proportions,
respectively as follows: A(100,0), B(75,25), C(50,50), D(25,75)
and E (0,100) . The milk yield and its quality for each cow were
determined along with supplemental intake and percent crude
protein (CP) .
The rations had no effect on the milk butterfat, solids-not-
fat, calcium, phosphorus and total solids. Animals on ration
D(17.4% CP) had the highest milk yield of 15 kg/day and an intake
of 3.9 kg/day. Ration E with 10.0% CP gave the highest intake at
4.0 kg/day with a yield of 13.4 kg Fat Corrected Milk (FCM) per
day. This production was superior to that recorded for rations
A, B and C. Invariably, ration A with the highest percent CP of
32.7 gave the least intake of 2.0 kg/day and the least milk yield
of 10.3 kg FCM/day. This unexpected observation was attributed
to some inherent factor in the lupine meal that negates intake
with the resultant effect of lowering milk yield.
INTRODUCTION
In most smallholder dairy systems in the high potential areas of
Kenya, provision of enough feed to lactating animals is a major
constraint to production. Fresh herbage and crop residues fed
during the dry season are generally of low quality and hence
contain substandard energy and protein levels that can hardly
meet maintenance requirements of lactating cows. In these
circumstances, lupine grain could prove suitable as a feed
supplement (Kenney, 1980) . As well as having a comparatively
high protein content (over 30% in dry matter) , lupine is easy to
feed and convenient to store for a considerable period of time.
Commercial feeds manufactured and compounded by some feed
industries form the bulk of the supplements in dairy enterprises.
These are expensive for the average Kenyan farmer, hence the need
to have supplement substitutes that can provide adequate energy
and protein levels that do meet production requirements of
lactating animals. The cost of lupine and its availability and
value as a feed is such that it could be mixed with other cereal
grains particularly maize for supplementary feeding. A grain
with such a high protein content could be suitable as a
supplement for lactating cows for which a diet with a protein
content of about 16% is recommended for maximum production (NRC,
1978) .
Protein ingredients in most supplements are the most
expensive and hence pose major problems in most on-farm rations;
230
this is because humans compete for the same for food. This,
however, is not the case with lupine which is currently being
used solely as feed. This study which involved the feeding of
lupine (Lupinus albus cv Ultra) and maize meal was designed to
solve the supplementation problem of nearly 20,000 farmers
(Stotz, 1983) located mostly in the Kenyan highlands.
Lupine, however, has shortcomings which include lupinosis
known to cause liver disfunction through the damaging of liver
tissues. The dangers of lupinosis have not been reported when
seed is fed as a concentrate to ruminants, but was reported on
the non-ruminants by Godfrey et al . (1985) and Wanjala (1979).
Little work has been reported on the performance of dairy
cows when supplemented with lupine alone or when incorporated in
high energy diets. The primary objective of this trial was to
study the effect of supplementing varying proportions of maize
and lupine meal on the milk yield of lactating Friesian cows.
METHODOLOGY
The lactating Friesian cows, divided into two groups of five,
were selected based on stage of lactation (animals in second or
third lactation being preferred) , average yield of Fat Corrected
Milk (FCM) , days postpartum and liveweight. The cows were then
randomly assigned to five treatments containing varying
proportions of maize and lupine meals. The percentage
proportions of the two ration ingredients were A(100,0),
B(75,25), C(50,50), D(25,75) and E(0,100) of lupine and maize
meal, respectively.
A change-over design in two 5x5 latin squares was used in
the study. For each group of cows there was an adjustment period
and a collection period of 14 days and 7 days, respectively. The
cows were grazed on good quality Chloris gayana (Rhodes grass)
pasture with supplementation being done daily during milkings at
08:00 and 16:00 hours. About 4.0 kg of each ration was
supplemented daily to each animal.
For each cow, the quantity of FCM produced daily was
recorded. Samples of milk were taken twice a week during the
collection period. The Gerber method was used in the
determination of percent butterfat, on a portion of the sample,
immediately after milking. The remaining sample was bulked for
each cow within each period and later analysed for milk protein
(CP) , solids-not-fat (SNF) , total solids (TS) , calcium and
phosphorus .
RESULTS
Milk Yield and Quality Assessments
A summary of the average milk yield and its quality values are
shown in Table 1. Milk yield expressed on a fat corrected basis
was highest (15.0 kg/day) for cows supplemented with treatment D,
followed with treatment E which gave 13.4 kg FCM/day. Non
significant differences (P>0.05) were observed among treatments
231
Table 1. Summary of average milk production and milk quality
values.
Experi
mental
Milk Milk quality (%)d
yield
rations (kg)
CP1 BF2 SNF3 Ca P TS4
A 10.3° 27. 9a 4.0 7.2 .88 .95 11.2
B 11.5° 26. 7b 3.9 7.4 .91 .89 11.3
C 10. 7C 25. 8b 4.0 7.4 .86 .88 11.4
D 15. 0a 24. 5C 4.1 7.3 .87 .86 11.4
E 13. 4b 24.1° 3.7 7.3 .91 .87 10.9
SE
LSD
CV %
0.5
1.4
8.4
0.3
0.9
2.5
0.1 0.1 .02
NS
5.4 6
.03
NS
.5
0.1
NS NS NS
6.8 1.8 6.8
abc Means in the same columns with different superscripts are
significantly different.
d Averaged for each ration between experimental periods.
1 Crude protein
2 Butterfat
3 Solids-Not-Fat
4 Total solids
A, B and C (10.3, 11.5, 10.5 kg FCM/day, respectively). The milk
protein from cows fed on treatment ration B and C were higher
(P<0.05) than that of milk from cows fed on treatments D and E.
Other milk quality factors (SNF, BF, TS, Ca, P) did not show any
differences (P>0.05) among treatments.
Supplement Quality Analyses and Intake
The per cent CP content and intake values of the various ration
formulations are presented in Table 2 . Ration A had the highest
CP (P<0.05) value of 32.7 but the least intake (P<0.05) of 2.0
kg/day. The results showed a declining trend in intake with
increasing ration CP content.
232
Table 2. Crude protein and Intake values of the concentrate
rations.
% CP
Experimentsll Periods Intake
rations I II III IV V Mean (kg/day)
A 33,.5 32.,9 34.1 31.9 31,,1 32. 7a 2.0C
3.2bB 27,.9 26..9 25.9 28.1 27,,2 27. 2b
C 24..6 23,.9 21.9 23.2 21..1 22.9°
17. 4d
3.5a
D 18,.6 16.,8 16.9 17.2 17,,6 3.9a
E 9.,7 9.,6 10.1 10.5 9.,9 10. 0e 4.0a
S.E. 0.4 0.2
LSD ,05 1.2 0.6
C.V. % 4.2 13.0
abcde Means in the same columns with different superscripts
are significantly different (P<0.05).
1 Values have been averaged for each ration between
periods.
DISCUSSION
The results indicate that ration D with a percent CP of 17.4 and
an average daily intake of 3.9 kg/day produced favourable milk
yields. It would thus appear that this ration had optimal
nutritive status to increase and sustain high milk-yield levels.
Cows fed on pure lupine showed some reluctance in consuming it;
this being reflected in occasional observation of nil daily
intakes particularly with decreasing levels of crude protein in
the rations. This observation appears to conflict with the
generally accepted concept that crude protein is the one
nutritive factor that does consistently increase intake as
evidenced by studies conducted by Ammerman et al . (1972),
Mukisira and Hembry (1985).
Inference from the observations made supposes that a factor
inherent in the lupine meal could be responsible for the
unexpected low levels of supplemental intake with increasing
ration percent CP values. There is some possibility that
alkaloids result in an undesirable taste to the cows. Inclusion
of maize meal in the rations tends to decrease the bitter taste
with the resultant effect of increasing supplemental intake and
to some extent animal performance. Considering this argument,
cows supplemented with pure maize meal had the highest intake of
4.0 kg/day albeit their average milk yield was inferior to that
from cows fed treatment D. This could be attributed to the lower
percent CP value in the pure maize meal ration.
CONCLUSION
Data drawn from this trial show that ration D containing 75%
maize and 25% lupine meals is desirable since animals on this
233
treatment produced the highest milk yields. It also manifested
fairly high intakes. Treatment rations with high lupine
inclusions tended to have low intakes. This was attributed to
the presence of alkaloids which impart an undesirable taste to
lupine meals. It is hoped that the next stage of the study will
investigate the actual alkaloid levels in the sweet lupine and to
some extent establish its lupinosis effect, if any, on the liver
cells of yearling steers.
ACKNOWLEDGEMENTS
The author would like to thank Mr. C.W. Mwendia for his technical
advice when writing up the paper and R.E. Kamidi for the
statistical analyses. Appreciation is also extended to the
Deputy Director of Research of the Kenya Agricultural Research
Institute for permission to publish these results. Last but not
least, to Mr. D.K. Muthoka, Director, National Agricultural
Research Station, Kitale for his assistance in the entire study.
REFERENCES
Ammerman, C.B., Verde, G.J., Moore, J.E., Burns, W.C. and Chicco,
C.F. 1972. Biuret, urea and natural proteins as nitrogen
supplement for low quality roughages for sheep. Journal of
Animal Science 35:121-124.
Godfrey, N.W. , Mercy, A.R. , Emms, Y. and Payne, H.G. 1985.
Tolerance of growing pigs to lupin alkaloids. Australian
Journal of Experimental Agriculture 25:791-795.
Kenney, P. A. 1980. Intake and production of lambs fed rations of
oats with varying amounts of lupins. Proceedings of the
Australian Society of Animal Production 13:253-256.
Mukisira, E.A. and Hembry, F.G. 1985. A nutritional assessment
of warm season forages by in vivo and in vitro techniques.
Proceedings of the fourth small ruminant CRSP. Kenya
Workshop, pp. 226-230.
Nutritional Requirements Council. 1978. The nutritional
requirements of domestic animals, pp. 36-37.
Stotz, D. 1983. Production techniques and economics of
smallholder livestock production systems in Kenya Ministry
of Livestock Development. Animal Production Division,
Nairobi. In: Farm Management Handbook of Kenya. Volume IV,
p. 48.
Wanjala, B.W. 1979. An introduction to lupin cultivation. Kenya
Farmer. January. English Press, Nairobi, p. 17.
234
SESSION III: EXTENSIVE LIVESTOCK PRODUCTION SYSTEMS
235

LIVESTOCK PRODUCTION SYSTEMS IN CHOKWE ,
SOUTHERN MOZAMBIQUE
Jonathan Timberlake
UNDP/FAO Project MOZ/81/015
Land and Water Department
National Agronomic Research Institute (INIA)
C.P. 3658, Maputo, Mozambique
Abstract
The livestock components of the mixed crop-livestock system of
Chokwe District are described. This system has been undergoing
changes in recent years due to an influx of people and cattle and
redistribution of land in the irrigation scheme, leading to
severe localised overgrazing, cattle deaths and an increased
demand for animal draft power.
A pasture/soil/dry season water points survey of the area
is described and the livestock carrying capacity of the various
pasture types was calculated using a livestock carrying capacity
model. Total livestock capacity of rangeland areas was
calculated as 50340 animal units (AU) and the feed resources of
the irrigation area (fallow fields and maize stover) sufficient
for a further 15700 AU. Actual livestock population is 37400 AU.
The problem of the area is shown to be distribution of cattle,
not absolute numbers, with excessive concentrations of animals
around the villages on the edge of the irrigation scheme.
The utilisation of rice straw and the introduction of
appropriate forage legumes into the existing farming systems are
given as ways of intensifying the production system, and these
require some applied research and testing.
INTRODUCTION
Out of a total cattle population in Mozambique of 1.4 million in
1981 some 72% were found in the southern three provinces of Gaza,
Inhambane and Maputo, which account for about 20% of the total
area of the country. Of these cattle 86% (881,000) were in the
traditional family (or communal) sector, raised on extensive
grazing with some use of crop residues. The major reason for the
concentration of cattle in the south is the presence of tsetse
north of 22°S, a line running through the north of Gaza and
Inhambane provinces.
The southern three provinces can be roughly divided into
three livestock zones based on rainfall, which shows a steep
gradient from the coast to the interior. The area is also
bisected by three major rivers: the Limpopo (including the
Elefantes and Changane rivers) , the Incomati and the Maputo.
Most of the family sector cattle are found along these rivers
especially in:
(a) the drier interior parts (under 600 mm rainfall/year)
where crop cultivation is very risky and cattle raising
237
becomes the more important economic activity, in areas
where human population densities are quite low;
(b) the area between 600 and 800 mm rainfall/year which is
principally an area of mixed farming, especially on
slightly heavier soils. The major activity of the
family sector is crop production (principally maize and
cowpeas) but cattle are kept to provide draft power,
milk for family consumption, meat, economic security in
bad harvest years, and as a means of investment in
years of surplus; and
(c) the wetter coastal areas with their higher and more
reliable rainfall (800 to 1100 mm/year) , sandy soils
and dense human populations. Crop production (cashew,
maize, cassava and groundnuts) is very important and
livestock much less so although cattle are frequently
used for ploughing.
The mixed crop-livestock farming system is a very important
one, involving not only a sizable portion of the national cattle
herd but also a large portion of the rural population of the
southern provinces.
The district of Chokwe lies in the mixed farming zone on the
southern side of the Limpopo river. There has been recent
immigration of cattle and people into this relatively safe
district from surrounding areas affected by the war, and
concentration of a previously more scattered population into
villages on the edge of a large irrigation scheme. This has
caused many cattle to die of hunger in the dry season. The
pasture and animal feed potentials of the district are large, and
there has been a big increase in demand for draft oxen for use in
the irrigation scheme. The need for intensification of livestock
production is clear. Because of this situation a study was
carried out (Timberlake et al , 1986) into the carrying capacity
of the pastures and the distribution of dry season watering
points in the area. This would assist in a more rational use of
existing feed resources and provide a basis for land use planning
for livestock production.
This paper is in two parts. The first part describes the
recent survey mentioned above, while the second part discusses
these results, and results from other surveys in the area, from a
farming systems perspective and gives some suggestions as to
future lines of investigation.
STUDY AREA
Chgkwe District in Gaza Province, in which the study area of 1830
km* is located, is primarily an agricultural district situated on
the south side of the Limpopo river some 100 km from its mouth.
It is characterised by a low and variable rainfall (averaging 623
mm/year, mostly falling from November to March) and high
evapotranspiration (1413 mm/year) leading to high agricultural
risk in dryland farming areas. A large irrigation scheme of
approximately 37,000 ha, mostly for rice, maize and vegetables,
is the major feature of the district. Since 1983 some 13,000
238
families have been granted plots of 0.5 to 1 ha in 9,000 ha of
the scheme. Surrounding this are dryland grazing areas, some of
which are cultivated in the form of small family plots. Maize
and cowpeas are the major crops, but cassava, groundnut and sweet
potato are also grown. Cattle are an important component of the
fanning systems of the region, principally providing draft power.
Goats are common but not numerous.
The area is flat to gently undulating and consists basically
of marine deposits overlain in places by more recent colluvial
and alluvial deposits. Soils close to the river are sandy but
fertile, while the rest are sandy loam in texture. These latter
areas are dissected by large shallow depressions of clay soils
which form water-courses for a few weeks of the year. In the
southeast of the study area, rising above the rest, are interior
dunes of reddish sands.
METHODOLOGY
The study was in three parts: a soil survey, a pastures/
vegetation survey and a dry season water points survey.
Soil Survey
The soil survey was at a reconnaissance level. A photo-
interpretation map was prepared based on aerial photographs at a
scale of 1:40,000 flown in 1973. On this map the main
physiographic/soil units were distinguished. Fieldwork consisted
of 89 soil auger observations to a depth of 1.2 m in locations
thought to be representative of the various physiographic units.
The following characteristics were recorded: physiography,
slope, drainage class, actual land use, soil texture, soil colour
and mottling, soil consistency, presence of calcium carbonate,
salinity and depth of groundwater table. Some 76 soil samples
were also taken for laboratory analysis including pH and mineral
content. On the basis of the fieldwork the preliminary photo-
interpretation map was adjusted and a final soil map prepared at
a scale of 1:75,000.
Pasture/Vegetation Survey
The pasture/vegetation survey was done at the same reconnaissance
level. Using the same aerial photographs, 100 observation points
were chosen, representative of the varying physiographic and
vegetation units distinguished. At each observation point the
tree, shrub and grass species present were noted with an
indication of their cover/abundance. The observation points
covered an average of 0.5 ha each and were not sited in heavily
disturbed areas or inside the irrigation scheme.
A matrix was later made of observation points along one axis
and species along the other. This matrix was rearranged
repeatedly on a micro-computer so that species which tend to be
associated, and observations of similar species composition, were
placed together. After various rearrangements clear groupings of
observation points emerge and these form the basis of the
vegetation units. The pasture/vegetation map was made using a
modified version of the physiographic/soil map.
239
From the map the areas of each pasture/vegetation type were
determined. Potential livestock carrying capacities of each type
were determined using the potential pasture productivity and
livestock carrying capacity model of Timberlake and Reddy (1986).
The assumptions used are given in Tables 1 and 2.
Table 1. Livestock carrying capacity calculations for vegetation
units in Chokwe ' .
Vegetation Estimated tree and ha/head3/
unit potential shrub
primary cover
productivity
(t DM/ha/year) 2>
(%)
Broad-leaved woodland 6.0 10 1.0
Broad-leaved woodland
- light bush 2.2 20 3.6
- heavy bush 2.2 40 7.1
Acacia savanna
- light bush 2.5 20 3.2
- heavy bush 2.5 40 6.3
- very heavy bush 2.5 50 8.8
Open grassland
- Themeda 3.0 0 1.8
- Dactyloctenium 2.0 0 2.7
- wet area 6.0 0 0.9
1/ Using the model of Timberlake and Reddy (1986) and assuming
50% use factor and an animal feed intake of 2740 kg
DM/head/year for a 350-kg animal with 100 gm/day LWG.
2/ 75% probability value.
3/ 1 head - 1 adult 350-kg animal.
Source: Timberlake et al. (1986).
Water Point Survey
In determining dry season water points all points where water
appeared to accumulate were noted on aerial photographs. Local
information was used in locating those points (less than half)
which still had water during September-October (dry season) , and
this proved quite reliable. A questionnaire was filled in at
each point concerning the presence of water, type, present
condition, age, size and depth, where the cattle came from which
use it, overgrazing/erosion, and whether it sometimes dries up.
The system of canals inside and on the edge of the irrigation
scheme were not included as access by cattle is possible along
most of their length.
240
Table 2. Areas of vegetation types and livestock carrying
capacities in Chokwe.
Vegetation/land type Area Carrying Total
(km2) capacity Livestock
(ha/head) capacity
('000 head)
Riverine woodland 61.0 3.0a 2.03
Broad-leaved woodland 226.9
849. 9b
6.0 3.78
Acacia savanna 5.0 15.10
Open grassland 294.3 1.0 29.43
SUB TOTAL 1423.1 50.34
Irrigation scheme 370.7 - 15.70°
Plantation forest 12.4 - -
Open water 6.8 - -
Urban areas 8.3 - -
TOTAL 1830.3 66.04
Assuming that only half of the area can be grazed; rest is
cultivated.
b. Assuming 0.5 ha cultivated in this unit per family and
19,000 families, leaving 754.9 km available for livestock.
c. Uncultivated and saline areas, and maize stover from Table 3.
Source: Timberlake et al. (1986).
RESULTS
Soil Survey
The description of the soil units is based partly on data
collected during the study, but also on the descriptions of
Touber and Noort (1985) who carried out a detailed survey of the
irrigation scheme. The 14 soil units determined can be divided
into four major groups:
soils of the interior dunes
soils of the marine Pleistocene sediments on elevated
areas
- soils of the marine Pleistocene sediments in the broad
depressions
soils of the recent fluvial sediments of the Limpopo
river.
In general, the soils are quite fertile (except those of the
interior dunes) , but the soils from marine sediments in the broad
depressions often show rooting limitations due to the heavy clay
and to periodic inundation. A good correlation was seen in the
field between soil type and pasture type, although the length of
inundation was also an important factor.
241
Pastures/Vegetation
Four major pasture/vegetation units were determined and
described:
1. Open Riverine Woodland. Characterised by large spreading
trees of Ficus sycamorus , Trichilia emetica, Acacia albida
and Sclerocarya caffra. Grass cover is mostly composed of
Urochloa mosambicensis , Panicum maximum and Cynodon
dactylon, with Setaria incrassata and Ishaemum afrum on the
black clay soils. Most of the area is of light alluvial,
well-drained soils, nearly all used for traditional
cultivation.
2. Broad-leaved Woodland. A woodland or bushland type with
Terminalia sericea, Sclerocarya caffra, Albizzia versicolor,
Strychnos spinosa, S. madagascariensis, Tabernaemontana
elegans and occasionally Hyphaene natalensis. Grass cover
is mostly of Panicum maximum, Cynodon dactylon and Digitaria
eriantha. Soils are sandy, and much is used for traditional
cultivation. Burning is frequent.
3. Acacia Savanna or Woodland. A woodland to tree/bush savanna
with Acacia welwitschii, Euclea undulata, A. senegal, A.
nilotica, Omocarpum trichocarpum and Albizzia petersiana.
A. xanthophloea is found in wetter areas. In the far
northwest of the study area mopane (Colophosperum mopane)
woodland occurs. Grass cover varies from Dactyloctenium
geminatum and Sporobolus nitens to Themeda triandra
depending on soil type. Terraitaria sp. are often present.
Soils are mostly sandy loams and cultivation is limited to
the edges of the broad depressions.
4. Open Grassland. An open grassland with very few trees or
shrubs of Acacia xanthophloea and A. nilotica. Grass cover
is of Echinochloa colona, Eriochloa stapfiana, Panicum
coloratum, Eragrostis cf. inamoena and Cyperus spp.
Termitaria sp. are generally absent. These areas are
flooded for varying lengths of time; soils are heavy black
clays. There is no cultivation and the grasslands are only
used for grazing.
Results from the livestock carrying capacity calculations
are shown in Table 1, and total livestock carrying capacities of
the different pasture/vegetation types are shown in Table 2.
Gross estimates were also made of feed resources inside the
irrigation scheme and are shown in Table 3.
Water Points
Excluding irrigation canals and the Limpopo river, 70 water
points were found with water in the dry season. An additional 90
points which appear to contain water in the early dry season were
noted from aerial photographs. The 70 points with water were
divided into nine types: lakes, rivers, irrigation canals, small
dams, small reservoirs, road excavations, boreholes, wells and
natural depressions. Twenty major water points were noted and it
was seen that the irrigation system, directly or indirectly
(through raised groundwater levels filling excavations) , is by
242
far the most important source of water. There are 4 0 small
reservoirs specifically made for use by cattle, but only 14 now
contain water through the dry season due to silting up and lack
of maintenance. The Mazimuchopes river dries up to a few pools
in the dry season.
Table 3. Estimated feed resources inside the Chokwe irrigation
area.
Feed Cattle*
DM available supporting
Area production/ ( 000 tons Use capacity
Uncultivated
(ha) ha/year DM/yr) factor ('000 head)
areas 5000 25000 12.5 0.7 3.2
Saline areas 2000 1000 2 0.3 0.2
Maize residue 7000 6000° 42 0.8 12.3°
Rice straw n/a 3700d 43e 0.8 11. 5f
Total - - 99.5 - 27.2
b.
Assuming 2740 kg DM/year animal intake for a 350 kg animal.
ILACO (1981, p. 476). Actual recorded grain yields are 1.8
t/ha with 13% moisture in the cool season, and 0.6 t/ha in
the hot season (Woodhouse et al., 1986).
Assuming 6% protein.
Assuming 1:1 ratio of grain: rice. State farm production
data averages 4 t/ha grain if a 50:50 ratio of the two
varieties grown is used, however var. C4-63 has less
vegetative growth (Woodhouse et al., 1986).
Assuming 1:1 grain: rice
production data.
ratio and commercialised rice
f. Assuming full urea treatment and 3 t DM/year animal intake.
In practice urea-treated rice straw would be used as a
supplement.
Source: Timberlake et al. (1986).
A notable feature of the distribution of water points is
that most of them are situated on or near the edge of the
irrigation scheme. Between the irrigation scheme and the
Mazimuchopes rivers some 13 to 30 km distant, there is virtually
no source of water, so cattle are overgrazing some areas and
underutilising other good pasture areas 8 to 15 km away.
243
DISCUSSION
Importance of Draft Power and other Livestock Fractions
The number and class of cattle in the area is reliably recorded
every year on the basis of the dip tanks. It is probably an
underestimate due to immigration of unregistered cattle and the
possibility that owners with one or two head do not bring their
animals to be dipped. Table 4 gives the numbers and classes of
animals and shows the high percentage of oxen in the family
sector herds compared to commercial herds. These cattle,
however, are not evenly distributed, and great concentrations are
found associated with villages bordering the northwest edge of
the irrigation scheme. Average herd sizes have been calculated
from livestock census data for different localities. They are
very variable, ranging from 4.4 to 38.7 head/family, with a mean
of 9 head (Vet. Fac. , 1986).
Table 4. Cattle population, Chokwe district, 1986.
Oxen Cows Calves <1 yr Total
% of % of
sector
% of % of
No. totalSector No. sector No. No. sector
Family 9179
156
18.9
1.8
16316 33.5
35.7
49.7
7362 15.1
32.9
15.7
48650 84.5
Private 3081 2842 8636 15.0
State 158 50 318 0.5
TOTAL 9335 19555 10254 57604
A recent study by Rocha (report in preparation) on
traditional livestock practices carried out in three villages in
the southwest of the study area and away from any major influence
of the irrigation scheme, shows that only 2.1% of livestock
owners interviewed did not have oxen, although 100% had a plough.
The strong positive relationship between number of draught
animals and area cultivated noted in this study is shown in Table
5. A Veterinary Faculty study (1986) into livestock practices in
Chokwe District, also carried out by questionnaire, found that
27% of family sector animals were used for ploughing and/or
transport, of which only 76% were oxen. The other 24% were
principally cows but also included bulls and steers. Within the
irrigation scheme a study into agronomic practices of families
with small plots (Woodhouse et al., 1986) showed the great
importance of animal draught power there, which varied with the
cropping system and averaged 83% (Table 6) .
Subsistence production of milk is widespread, although
previously not fully recognised. Rocha's study found that 10.6%
of livestock owners in Mazimuchopes do not have cows, but of
those that do 93% milk in the wet season. Data on yields are
still not available, but it can be assumed that yields are around
1 to 1.5 litres/day/producing cow. Milking is much less
practiced in the dry season; questioned in July (dry season) ,
244
only 2 6.5% of cattle owners said they milked their animals (Vet.
Fac, 1986), but it is not clear if they were speaking generally
or referring to that month in particular.
Table 5. Relationship between draught animal ownership and area
cultivated around Mazimuchopes.
Number of draft
animals per family
Mean area
cultivated (ha)
1-2
3-5
6-9
9 +
1.1
1.9
3.6
2.6
Source: Rocha (unpublished data) ,
Table 6. Ownership and importance of oxen in two major cropping
systems in the Chokwe irrigation scheme.
Maize Rice Mean
Plots ploughed with oxen (%) 92 65 83
Ownership of:
oxen rainfed plot (%)
+ + 34 25 31
+ 24 9 19
+ 16 10 14
— — 25 55 35
Source: Woodhouse et al. (1986)
Supplementary feeding appears to be an important practice of
traditional livestock owners. In the Mazimuchopes study Rocha
found that 32% of livestock owners gave maize stover and/or
failed maize crops, 9% gave cut grass in addition to maize
stover, and 2% make hay for dry season use. In the Chokwe
irrigation scheme in 1987 farmers were selling maize stover at
1000 meticais (around 25% of the minimum monthly agricultural
wage) per cart load (approximately 200-300 kg) at the farm, i.e.
transport is not included. This however was at a time when
grazing was poor because of a mid-season dry spell.
A measure of the sophistication of a farming system is its
use of salaried labour. It is therefore interesting to note that
32% of livestock owners employ salaried labour for herding
(Rocha, unpublished data) , although the influence of receipts
from migrant labour in South Africa probably plays an important
role.
Carrying Capacity of Natural Pastures and Grazing Limitations
The calculated carrying capacities given in Table 1 show the good
grazing potentials of the area. The importance of the open
grassy depressions is very clear, although for some months of the
245
year they are inundated and not utilised. This however is in the
wet season when sufficient grazing is available in the Acacia
savanna pasture type. Actual primary production data from these
areas are not available and it is important to collect some.
If grazing were evenly distributed the livestock carrying
capacity of the natural pastures in the study area is 50,340
adult head, perhaps 64,000 head with the existing herd structure.
The number of animals in the study area (which does not cover the
whole of Chokwe District) is estimated at around 47,500 head,
giving a theoretical possibility of increase of 16,500 head. The
problem of course is that grazing is not consistent with the
distribution of grazing resources for three major reasons:
1. owners are afraid to graze the relatively unpopulated areas
away from the irrigation scheme because of cattle thieving
and the civil war situation;
2. most of the cattle are associated with the villages along
the edge of the irrigation scheme and kraaled there at
night ; and
3 . the great lack of dry season water points away from the
irrigation scheme and the resulting long distances from
grazing to water.
Overgrazing around the major villages is severe and some
hundreds of animals die from hunger near water points in the late
dry season. The two most practical ways of overcoming the
problem are (a) to intensify livestock production with the use of
forages or treatment of crop residues, or (b) provide water in
the dry season in presently underutilized areas where this,
rather than security or distance to the night kraal, is the
limiting factor. The pasture study has made recommendations on
some suitable locations for water points, which make use of
runoff water that can be captured and stored in a small
(approximately 100 x 50 x 3 m) reservoir. These reservoirs, many
of which were constructed in the past but have not been
maintained, only require re-excavation every two or three years
and otherwise have no running costs.
Bush encroachment is found in the Acacia savanna and broad-
leaved woodland vegetation types, particularly the former. Here
it seems to be more acute in areas that are or were commercial
ranches. However, some of the species causing bush encroachment
in the Acacia savanna type are suitable for browsing.
Feed Resources in the Irrigation Scheme
The irrigation scheme is the centre of nearly all agricultural
activity in the district, and thus produces large quantities of
crop residues, principally rice straw and maize stover. There
are, in addition, large areas of fallowland; estimates are as
high as 13,000 ha, but 5,000 ha would seem a safer figure, and
perhaps a further 2,000 ha of abandoned saline areas. In theory
cattle are not allowed to enter the irrigation scheme except
working animals because of the damage they cause to the canal
banks and crops they might eat. But this is not enforced and
large numbers of cattle can be seen inside feeding on maize
246
stover at the end of the season, in fallow fields, and grazing on
weeds and regrowth in the rice fields after harvest. It is
rather difficult to quantify these feed resources, but estimates
were made (Table 3). The theoretical carrying capacity was
calculated assuming that the animals would be living entirely off
these feed resources, which is far from the real situation as
they are not available year round. However, their importance can
be seen, not only at present when fallow areas and much of the
maize stover is utilised, but also the potentials of the large
quantities of rice straw which are now normally burnt. This
requires an applied research programme into the best and most
cost-effective methods of treatment to increase its acceptability
and feed value at the level of small-scale livestock producers.
Cereals are not the only crops grown in the irrigation
scheme. Cowpea (Vigna unguiculata) , common bean (Phaseolus
vulgaris) , Lablab purpureus and pigeon pea (Cajanus cajan) are
frequently found grown with maize. These increase the protein
value of crop residues when grazed together, but foliage
productivity data are not available. Lablab in particular shows
promise as a dual-purpose crop-grain for human consumption and
foliage for cattle. It is normally planted with maize under
dryland conditions at the end of the main rains (March/April) ,
but the plants are small. If planted one month earlier foliage
production is much higher (Heemskerk, pers. com.). Some trials
were carried out on Melilotus alba under irrigation as a cool-
season crop after rice (Woodhouse et al., 1986). Forage dry-
matter yields averaged 3.6 t/ha. Lucerne (Medicago sativa) has
been grown commercially inside the scheme.
Outside of the irrigation scheme Cajanus, Lablab and
Leucaena spp. would appear, from personal observation, to offer
the best prospects as forages or dual-purpose crops that could be
utilised as animal fodder by traditional farmers.
CONCLUSIONS AND FUTURE LINES FOR INVESTIGATION
Chokwe District is undergoing a challenge as regards livestock
production. It is an area with a strong livestock tradition in
which the provision of draft power is of the greatest importance,
and an area with a large agricultural potential due to fertile
soils and a large irrigation scheme. The challenge is due to an
unstable and rapidly changing situation caused by various
factors:
(a) the civil war situation, resulting in a large influx of
cattle and people from less-secure surrounding areas
and a fear of grazing animals far from the villages;
(b) the recent concentration of villages, and therefore
cattle, along the periphery of the irrigation scheme
leading to localised but severe overgrazing;
(c) the impending return of much of the migrant labour
force from South Africa which provides much of the cash
and goods to the family economies of the area;
247
(d) land use conflicts between livestock and dryland arable
fanning concentrated around the villages;
(e) the conflict between livestock and the management of
the irrigation scheme which severely limits grazing of
cattle within its limits (before a major dry-season
grazing area) due to damage caused to canals and crops;
(f) the insufficient quantity and quality of feed during
the dry season for the draught animals which carry out
an important part of the agricultural work inside and
outside the irrigation scheme; and
(g) the poor distribution of dry-season water points away
from the irrigation scheme which limits accessibility
to pastures because of excessive distances from grazing
to water points and finally to the night kraal.
With increasing land use pressure there is now a need for a
more rational use of available resources (grazing, water, crop
residues, etc.)/ and the need for intensification of traditional
livestock production is clear. In the foreseeable future most
agriculture will continue to be carried out by draft animals and
will be in or close to the irrigation scheme. These animals need
good feeding to maintain their strength, and the scheme will have
a big role in providing their feed. Forages grown within the
existing smallholder farming systems and better utilisation of
crop residues are the main ways to achieve this.
Recent studies discussed in this paper give a general idea
of traditional livestock practices in the area, the principal
problems and the feed resources available. Some future lines for
investigation regarding animal feeding can be drawn from this:
(a) Two or three detailed village-level studies need to be
carried out in environmentally different zones in the
region to determine and quantify livestock practices
and constraints to production. These should include
size and composition of individual herds, grazing
practices, use of crop residues and quantification of
weight losses (especially cows and oxen) . This will
ensure that future interventions by the extension
service are appropriate and have a good impact.
(b) The relative importance and problems of goats which
should be better adapted to shortage of grass and
water.
(c) Determination of the relationship between cultivation
in the irrigated areas and cultivation under dryland
conditions, as most smallholders in the irrigation
scheme also have a plot outside and animal draught
power is used in both.
(d) Determination of pasture and animal productivity from
both natural pastures and crop residues and the
determination of carrying capacities on a more detailed
level.
248
(e) Determination and testing of appropriate methods of
improving acceptability of rice straw, i.e. urea
treatment or treatment with molasses. This testing
should preferably be done through farmer-managed
trials.
(f) Suitable methods of storing maize stover and rice straw
for use later on in the season without losses in
nutritive value.
(g) Intercropping of maize with Lablab, cowpeas and other
suitable dual-purpose legumes with a view to improving
the protein levels of crop residues. Some of the
leafier legume varieties available from ICRISAT, IITA
and other institutions presently available in the
country can be used.
(h) The use of Cajanus as forage including selection of
leafier varieties, its place in the cropping pattern
and management techniques.
(i) The use of Leucaena as a forage reserve in dryland
areas and around the village. Any limitations to its
use (e.g. productive losses due to mimosine) should be
investigated also.
(j) Economics and production effects of supplementation
with urea/molasses and phosphate blocks.
The area of Chokwe is somewhat unique due to the influence
of the irrigation scheme, but many of the traditional
agricultural practices there are found over a large and important
area of southern Mozambique. The impact of livestock or forage-
related research, and good extension packages, on this farming
system would be large. It would also greatly improve crop
production in the area through a more efficient use of animal
draught power, thus the rural family economies would become more
stable.
The problems are now reasonably clear but still require some
further quantification. Solutions have been suggested, but work
has not really commenced on their viability. Given the limited
resources in research a cooperative approach is required, with
soil surveyors, land use planners, pasture agronomists, crop
agronomists and animal production specialists working in
coordination.
ACKNOWLEDGEMENTS
I am very grateful to Dr. Antonio Rocha for permission to use his
unpublished data. Thanks are due to Drs. Celia Jordao and wilhem
Heemskerk for comments on the draft.
REFERENCES
ILACO. 1981. Agricultural compendium for rural development in
the tropics and sub-tropics. Elsevier, Amsterdam.
249
Timberlake, J., Jordao, C. and Serno, G. 1986. Pasture and soil
survey of chokwe. Comunicacao 50 (Serle Terra e Auga) ,
INIA, Maputo.
Timberlake, J. and Reddy, S.J. 1986. Potential pasture
productivity and livestock carrying capacity over
Mozambique. Comunicacao 49 (Serie Terra e Agua) . INIA,
Maputo .
Touber, L. and Noort, L. 1985. Avaliacao de terra para
agriculture regada na area do "SIREMO", Vale do Limpopo.
Vols I, II. Comunicacao 26 (Serie Terra e Agua) . INIA,
Maputo .
Veterinary Faculty. 1986. Relatorio final das actividades de
Julho de Brigada 7 - Chokwe. Veterinary Faculty, Eduardo
Mondlane University, Maputo.
Woodhouse, P., Jimenez, H. , Heemskerk, H. , Spittel, M. and
Slobbe, W. 1986. Smallholder farming systems research in
the Chokwe irrigation area. Field Document 4, UNDP/FAO
project MOZ/81/014. INIA, Maputo.
250
PRODUCTION OF THE MALAGASY EXTENSIVE LIVESTOCK
J.H. Rasambainarivo and R. Razaf indratsita
Departement de Recherches Zootechniques et Veterinaires
Division Alimentation Animale
B.P. 4 - Antananarivo, Malagasy Republic
INTRODUCTION
Ruminant production is far more important than that of non-
ruminant in the Malagasy state. The livestock population is
distributed as follows (MPAEF, 1986) :
Cattle 10,485,000
Sheep and goats 1,829,000
Pigs 1,350,000
Of the 12,341,000 ruminants, more than 90% are raised in an
extensive system. This is predominant in the western and
southern parts of the island, where land is still available for
grazing, as the human population density is rather low between
six to eight persons/km .
The Malagasy cattle owner is also a rice producer. He lives
in a sedentary system and his priorities for keeping animals are
obviously different from those in developed countries. So
productivity may be understood as the high number of zebu cattle
which reflect his wealth and the ability to sell and/or slaughter
at the occasion of traditional festivals. Productivity also
means the working capacities of the zebu herds for draught, their
manure and milk yield. For the national economy on the other
hand, livestock productivity means the meat for domestic
consumption and export.
This paper analyses the different patterns of zebu cattle
productivity in order to make some suggestions for research
priorities and productivity improvements.
METHODOLOGY
Data presented here were collected from two sources: the first
was from Ministry surveys (MDRRA, 1979; MPAEF, 1986) and the
second from the Miadana Research Station.
Ministry Surveys
Two surveys were carried out during the last ten years. The
first was from November 1978 to February 1979; the second from
June to December 1984. The purpose of these studies was to
acquire some data about the main features of the livestock
population. The surveys used the stratified sampling procedure.
It is clear that the collected information was not complete and
accurate. But by such periodic appraisal it is possible to get
the major trends of the national herd.
251
Experimental Station
Miadana Station (15° 36'S; 46° 38 'E) is the only research station
covering the western and southern zones. It is located 60 km
from Mahajanga, and the mean annual rainfall is 1,450 mm. Some
data on the zebu weight gains and reproduction performances of
improved management systems were collected from this station.
RESULTS
The extensive livestock system is well spread all over
Madagascar, but the western and southern provinces are the most
representative of the system. These two provinces cover 318,000
tan , nearly 55% of the island's total area, and 61% of the
national zebu cattle herd is raised in these two provinces. The
climate varies from subhumid in the northern part to sub-arid in
the southern region. The rains fall between November and April,
and the dry season may last six to nine months of the year.
Feed Resources
The most important feed resources are natural pastures, crop
residues and to a limited extent, improved pastures. The grazing
lands are dominated by four grass species: Heteropogon contortus,
Hyparrhenia rufa, Chrysopogon serulatus and Aristida rufescens
(Figure 1) (Cabanys and Razafindrastsita, 1971; Granier and
Bigot, 1972; Morat, 1972; Suttie, 1974a). In the south the
grasses are replaced by shrubs.
Generally the edible dry matter available during the rainy
season is sufficient as the stock is not able to eat all the
grass produced. After the rainy season there is a rapid change
of nutritive value in the grassland. Crude protein content
particularly may be as low as 2% in September (Figure 2) . The
management of this natural grassland is not controlled, except in
the Androy region (south) , where there is a traditional deferred
grazing area ranging from 2 to 200 ha, which is only grazed in
the dry season. This deferred grazing land is called Adily
( Rats imba-RaJohn and Razafindratsita, 1977) . It belongs to a
family or a local community, and their livestock are the only
ones authorised to graze on it.
Several pasture improvement tests were performed (Borget,
1971; Suttie and Hablutzel, 1975; Rasambainarivo et al . , 1980),
particularly with leguminous species. The best adapted are
Stylosanthes guianensis, Stylosanthes humilis, Stylosanthes
hamata and Macroptilium atropurpureum (siratro) , but they failed
in practical extension due to socio-economic reasons.
In the south, thorny cactus is a traditional multipurpose
plant. It is firstly used as fence around villages and fields.
It is also cut, burnt and distributed to livestock in the dry
season. The mature fruit is appreciated by humans. Some
thornless cacti were tested (Suttie and Matzernmiller, personal
communication) ; they give higher yield in terms of fodder
although their management is more difficult, and they cannot be
used as fences, so their extension is very limited.
252
Figure 1. Grazing lands in western and southern Madagascar.
Legend
E
Hyparrhenia
Heteropogon
Cenchrus
Aristido
Shrub
ITULEAR
 
Since rice is planted all over the island, rice straw is
therefore the most common crop residue. It is usually left in
the field after threshing and consumed by zebu cattle without any
treatment. Weeds growing in the rice fields are also grazed
253
after harvest. Other crop residues are maize and sorghum
stovers, cassava, sweet potatoes, vines and groundnut haulms. A
survey carried out in 1977 indicated that 70% of the cattle
owners utilise crop residues for animal feeding.
Agro-industrial by-products are not available for extensive
livestock due to their high cost and transport problems.
Herd Structure
The Malagasy zebu is the major breed constituting the extensive
cattle population. Herd composition (Table 1) indicates that
females are more numerous than males and the bull: cow ratio is
about 1:18. No culling of infertile females is practiced because
legally they are not to be slaughtered. More than half of the
stock are four years old and over, and about 11% are more than
ten years old. In this old group the majority are castrated
males.
Figure 2. Crude protein contents of natural pasture in
and southern Madagascar provinces.
western
Crude protein
%DM
60
50-
40
30-
20-
10-
 
Heteropogon contortus
Heteropogon contort us
Chrysopogon sarrulotus
[Heteropogon cantor tut")
<Hyparrhenio dtssoluto k
[Chrysopogon serrulatusl
A A
 
—A
—r-
M J J
Months
-r-
0
254
Table 1. Composition of household herd in Faritany Mohajanga in
1984.
Age (years) Males Females Total
Less than 1 8.4
1-3 12.0
4-9 15.3*
10 and over 3.6*
9.2
14.0
30.0
7.4
17.6
26.0
45.3
11.0
Total 39.4 60.6 100.0
* 11.8% are castrated males.
Source: Sarniguet and Hartier (1986) -
Reproduction Performances
Reproduction performance data are presented in Table 2 . The data
were collected all over Mahananga Province. The data for Miadana
Station are averages for a ten-year period. The calving rate is
quite variable in the traditional management system, ranging from
44.7% to 68.6%. In the "Miadana" conditions, the mean rate is
67.3% (Rakotobe-Ralako, 1987).
Table 2. Extensive reproduction performance.
Regional
Mahajanga
survey
Miadana
station
Calving rates (%)
Age at first calving
(months)
Calving interval (days)
58
60
420
67 +8
54
377
In the extensive situation the age at first calving is
longer. The calving interval is also longer. These data are in
agreement with the report by Rakotobe-Ralako, (personal
communication) who reported that in well-managed herds of zebu
cattle age at first calving may be as short as 28 months compared
to a range of 48 to 68 months under extensive management. The
calving interval was extended in extensive management systems.
Mortality varies widely depending on the age. The most
important mortality is that of calves less than one year old; 21%
of these calves die in the extensive system conditions. The
mortality rate decreases as the zebu get older, and the overall
mortality is estimated to stabilise at 7.5%. All these
255
reproduction data will vary depending on seasonal availability of
nutrition.
Liveweight Changes
The general pattern of Malagasy zebu growth in extensive system
is presented in Figure 3 which compares smallscale owner and
research station cattle. As expected, the rate of growth of the
better managed station cattle is higher than that of the
extensively managed. Even at the end of 48 months after birth
the station-managed cattle are heavier by about 50-75 kg
(Ratovonirina, 1983) .
Considering the growth rates within the first 36 months, it
is evident that the worst period lies between the 6th and 18th
month of age. This period coincides with weaning, a period when
daily weight gain decreases from 458 g to only 90 g/day.
Figure 3. Weights of Malagasy Zebu in extensive and station
conditions.
4O0
300-
200-
100
kg
x
 
o o Males poysons ouest
x x Males temoint M DA
0--i—i—i 1—i 1 1 1 1 1—i—i—i—i 1—i 1 1—i 1
0 3 6 9 12 6 18 21 24 27 30 33 36 39 42 45 48 51 54 57
ft t fAgeenmois
2 1 3 I 4 I 5
Source: Ratovonirina (1983) .
256
Comparative livestock productivity studies have shown that
stocking natural grasslands at one animal/ha induces a weight
loss whereas on legume reinforced planted pastures stocked at the
rate of three animals/ha the animals were able to gain at 0.3
kg/day. These data show the potential for improvement of the
extensive grazing land resources in Madagascar (Rasambainarivo et
al., 1985).
Slaughter Performances
Nearly all the cattle slaughtered are zebu coming from extensive
systems. These slaughters represent a national carcass
consumption of 13.1 kg/ha/annum. This figure is doubtful as
there are a lot of cattle slaughtered but not recorded. The mean
carcass weight of Malagasy zebu cattle is 172 kg. The majority
of slaughtered zebu are more than eight years old. Nevertheless,
in a better environment the zebu carcass may weigh 200 kg
(Gilibert, 1971) .
Milk Production
Milk production is low in extensive systems. Data were collected
in the suburbs of Mahajanga. The mean individual cow production
is estimated at 350 litres for a 200-day lactation (Planchenault,
1986) . In the deep south of the Androy region, milk is also
produced and traditionally processed, but milk production records
are not available.
conclusion
From the data reviewed, it is evident that the major factor
limiting production of the zebu cattle in Madagascar under
extensive production system is inadequate nutrition which leads
to poor reproduction and weight gain performance. The zebu
cattle may not be expressing their full potential as a result of
the nutrition problem under these conditions. Notwithstanding
the socio-economic constraints to improving livestock production,
an improvement of the feed resource would invariably increase the
reproduction performance of cows and reduce calf mortality.
More accurate information about the different management
systems is required, and more comprehensive surveys must be
carried out. Additional research information should be gathered
about the status of bovines on-farm. And all of these data have
to be used with a computer model for a production system
simulation. As beef production is a rather complex enterprise,
improvement can only be successful if it is carried out by inter
disciplinary teams on national and regional level.
REFERENCES
Borget, M. 1971. Recherche et production fourragere en
Republique Malgache. I.R.A.T., Paris.
Cabanys, Y. and Razafindratsita, R. 1971. Reconnaissance
agrostologique du Sud de Madagascar. I.E.M.V.T. Region de
Recherches de Madagascar.
257
FO.FI.FA. 1971. Pre-enquete sur 1 'utilisation des residua des
recoltes et de la potentialite de fustier de pore dans
l'Androy. FO.FI.FA/Division Economic Rurale.
Gilibert, J. 1971. Valuers boucheres des zebus a Madagascar.
Rev. Elev. Med. Vet. Pays Trop. 24 (3) :445-465.
Granier, P. 1971. Problemes poses par le developpement de
l'elevage dans le Sud de Madagascar. Etudes preliminaires.
Rapport destine a la Sogreah. I.E.M.V.T. - Region de
Recherches de Madagascar.
Granier, P. and Bigot, A. 1972. Donnees preliminaires pour une
etude de l'Elevage bovin dans la province de Majunga.
I.E.M.V.T. Region de Recherches de Madagascar.
M.D.R.R.A. 1979. Ministere du Developpement Rural et de la
Re forme Agraric.
Morat, P.H. 1972. Contribution a l'etude des savanes du Sud-
Ouest de Madagascar. These de Docteur-es-Sciences
Nature lies, Paris-Sud, Contre d' Orsay.
M.P.A.E.F. 1986. Rapport d'activites de la Direction de
1 ' Elevage .
Planchenault, D. 1986. Etude de la speculation. Production
laitiere autour de la ville de Mahajanga. Inst. d'Elev. et
de Med. Vet. des Pays Trop. - Maisons-Alfort.
Quesnel, M. and Sarniguet, J. 1979. Etude de la
commercialisation due betail et des prix de la viande a
Madagascar. Tome I : Estimation et caracteristiques due
cheptel. M.D.R.R.A. Sedes.
Rasambainarivo, J.H., Rakotozandirndrainy, R. and Razafindrata,
R. 1985. Les sous-produits de coton. Utilization en
alimentation animal (croissance et embouche bovines) .
D.R.Z.V. Division Alimentation Animale.
Rasambainarivo, J.H. , Rakotozandrindrainy, R. Razafindratsita, R.
and Rabehanitriniony, M. 1986. Productions des paturages
ameliores dans deux regions due Faritany de Mahajanga.
Academie Malgache et Colloque International sur
1' Amelioration de l'Elevage en zone tropical: cas de
Madagascar.
Ratovonirina, E. 1983. Amelioration genetique du zebu malgache.
Memoire de fin d'etudes, E.E.S.S.A. Universite de
Madagascar.
Ratsimba-Rajohn, R. and Razafindratsita, R. 1977. Rapport de
tournee dans l'Androy. D.R. Z.V. /Division Alimentation
Animale.
Sarniguet, J. and Hartier, P. 1986. Recensement du cheptel des
Faitany d' Antananarivo et Mahajanga. Enquete 1984.
M.P.A.E.F./ Direction de l'Elevage et Sedes.
258
Sarniguet, J. and Vindrinet, R. 1979. Etude de la
commercialisation du betail et des prix de la viande a
Madagascar. Tome 2 : Analyse economique des filieres et
propositions. M.D.R.R.A. et Sedes.
Suttie, J.M. 1974. La situation agrostologique et les problemes
du developpement de l'elevage dans la zone d' intervention de
la SODEMO. Compte-rendu de mission. Project FAO-CMCF
MAD. 16.
Suttie, J.M. 1974a. Les paturages naturels et les plantes
fourrageres de l'Androy Sedimentaires. Leurs roles dans la
production animale. FAO-PNUD.
Suttie, J.M. and Hablutzel, H. 1975. La promotion de la culture
de la Raketa (Opuntia ficus indlca) dans l'Androy en vue de
1 'amelioration de 1 'alimentation du troupeau des ruminants.
FAO-PNUD.
259
THE PREDICTION OF RANGELAND PRODUCTION FROM
RAINFALL DATA IN ARID AND SEMI-ARID EASTERN AFRICA
P.N. de Leeuw and R. Nyambaka
International Livestock Centre for Africa
P.O. Box 46847, Nairobi, Kenya
Abstract
Relationships between rainfall and peak biomass were analyzed for
eight sites in arid and semi-arid areas in Kenya and northern
Tanzania. These sites received average annual rainfall ranging
from 200-600 mm falling mostly in a bimodal pattern. Regression
equations of seasonal biomass were significant for all sites
explaining 56 to 99% of the variation. Predicted seasonal
biomass was 1200-2000 kg DM ha 2 at 300 mm and 2100-3900 kg DM at
500 mm of rainfall. Differences between sites were due to
species composition, soil type and herbaceous cover density.
These data sets were compared to similar regressions for west and
southern Africa. It was concluded that productivity per unit of
rainfall in East Africa was higher than these other regions
because of bimodal rainfall and higher soil fertility.
INTRODUCTION
In recent years relationships between seasonal rainfall and end
of season herbaceous biomass have been published for several
regions in Africa. For the Sahelo-Sudanian region data were
compiled by Le Houerou and Hoste (1977) followed by studies in
Mali by Penning de Vries and Djiteye (1982). For Eastern and
southern Africa regressions between annual rainfall and
herbaceous biomass were developed by Deshmuck (1984) which showed
that per unit of rainfall biomass production in this region was
twice as high as in West Africa. However, similar regressions
for southern Africa given by Rutherford (1978) indicate that
productivity of rangelands in Zimbabwe and Botswana is much lower
than in East Africa.
The purpose of this paper is to focus on the productivity of
rangelands in arid and semi-arid Eastern Africa and indicate
which factors are responsible for the substantial higher levels
of productivity in this region as compared to southern and
western Africa.
METHODS AND DATA SOURCES
Relationships between rainfall and peak biomass were analysed for
sites in Kenya and Tanzania. All sites were located in areas with
arid or semi-arid climates having seasonal rainfall ranging from
100 to 600 mm falling mostly in a bimodal pattern due to their
location around the equator (Tables 1 and 2) .
Raw data were obtained for three sites in semi-arid Kenya:
(1) Kiboko Research Station (Too, 1985) ; (2) Athi plains
(Potter, 1985); and (3) Tsavo National Park (van Wijngaarden,
1985) . These three sites are all located within the pastoral
260
land use zone (cf . Jaetzold and Schmidt, 1983) . Additional data
sets were used as supplementary sources: Amboseli (Western and
Grlmsdell, 1979), Serengeti (Sinclair, 1979; Braun, 1973).
Table 1. Relationships between rainfall and herbaceous standing
above-ground biomass in bimodal rainfall areas in Kenya
and Tantania.
Average
Area annual Principal 100 300 500
rainfallb/ genera *g DM ha"
Amboseli (IV)b/ 300- 550 Sporobolus-
Pennisetum 10 760 1510
Kiboko (V) 450- 900 Digitaria-
Chloris 670 1490 2300
Serengeti (IV) 600-1100 Sporobolus -a/ 1190 2150
Tsavo (VI) 300- 550 Chloris-
Chysopogon 420 2020 3620
Serengeti (IV) 600-1100 Themeda -a/ 1570 3700
Serengeti (IV) 600-1100 Themeda in
woodlands 470 1790 3100
Athi (V) 450- 900 Themeda 10 1370 3860
Serengeti (IV) 600-1100 Andropogon -a/ 1520 3230
a/ The data did not cover rainfall below 200-250 mm.
b/ IV-VI Moisture availability zones and the annual
according to Braun (in Sombroek et al, 1982).
rainfall
Table 2. Regression equations for seasonal rainfall on standing
biomass in East Africa.
Area Equation Reference
Amboseli Y- -367+3. 8X (N» 6; R2 - 0.99) Western and
Grimsdell (1979)
Kiboko Y- +262+4.41 (N- 38) R2 - 0.78) Too (1985)
Serengeti Y- 262+4. 8X (N- 7; R2 - 0.93) Braun (1973)
Tsavo Y- 380+8. OX (N- 89; R2 - 0.65) van Wijngaarden
(1985)
Serengeti Yi= -1644+10. 7X (N- 12; R2 - 0.62)
R2 - 0.90)
Braun (1973)
Serengeti Y- -185+6. 6X (N- 24; Sinclair (1979)
Athi Yt -251+1. 2X+0. 01X2 (N- 24; R2 - 0.95) Potter (1985)
Serengeti Y- -1052+8. 6X (N= 10; R2 - 0.56) Braun (1973)
Biomass in kg DM ha-1. rainfall in mm.
261
In Kiboko a clipping trial was carried out from February
1982 to December 1984 in four neighbouring vegetation types. The
plant cover consisted mainly of perennial grasses with a medium
to light cover (10-30%) of Acacia and Commiphora shrubs and low
trees. Seven different cutting regimes were applied together
with two defoliation intensities (cutting at 5 and 12.5 cm
height) . Within each vegetation type the 14 treatment
combinations were repeated on land that was burned prior to the
start of the trial and also on unburned land. Only data from the
high intensity defoliation treatments were used assuming that
harvesting at 5 cm height best represented available biomass. As
the effect on herbaceous biomass growth of burning and vegetation
type or their interactions were not significant (Too, 1985) ,
these were combined in the re-analysis, giving 38 pairs of data
of cumulative seasonal rainfall and resulting biomass.
The study in Tsavo National Park was designed to investigate
and describe the ecosystem and to "quantify the climate-soil-
vegetation-large- herbivore interactions and to develop a model
describing these relationships quantitatively" (van Wijngaarden,
1985, p. 149) . To this end, herbage productivity was measured in
representative vegetation types over five rainy seasons during
1976-78. Data on seasonal rainfall, clipped herbaceous biomass,
cover density of the perennial and annual grass as well as the
woody component were made available for 88 sample plots, which
were used for regression analysis.
The site in the Athi plains was located in the treeless
Themeda grasslands. The trial design was a factorial with five
defoliation frequencies (3-, 6-, 9-, 12- and 15-week cutting
intervals) each at three heights of cutting (5, 10 and 15 cm) .
Treatments were applied from April 1974 to December 1978 covering
ten growing seasons. Data on rainfall, evapotranspiration and
biomass growth were given for three-weekly periods for the entire
duration of the trial. As for Kiboko, only data for the high
defoliation intensity (i.e. 5 cm) were used for the analysis. As
the medium defoliation frequencies produced the highest biomass,
only the cutting intervals of six and nine weeks were included in
the analysis. This provided 24 pairs of cumulative seasonal
rainfall and its resultant grass biomass production.
For the other locations, relationships between standing
biomass and annual instead of seasonal rainfall were available.
For the Amboseli plains (Figure 1) the data covered the period
1973-78 (Western and Grimsdell, 1979). Similarly for the
Serengeti plains, data were extracted and re-analyzed from
Sinclair (1979) and from Braun (1973) .
All sites were either protected or were located where
grazing pressure was low. Standing biomass was measured through
clipping either at ground level or at 5 cm above ground. In most
areas, end of season biomass was recorded except in Kiboko (Too,
1985) and in the Athi plains where several defoliation regimes
were combined.
262
Figure 1. The relationship between rainfall and biomass yield at
two sites in semi-arid rangeland in Kenya.
Blomciit kg/ho
6j00 i-
sooo
4000
sooo -
2000 -
i000
 
- sooo
- 4000
- sooo
- 2000
- i000
0 50 i00 lSO 200 230 300 3S0 400 4S0 300 SSO «j0 «S0
Staional rainfall (mail
RESULTS AND DISCUSSION
Effect of Rainfall
The results of regression analyses of seasonal rainfall on
biomass are shown in Tables 1 and 2 . The proportion of the
variation explained by rainfall ranged from 56 to 99% and was
significant in all cases. Linear regressions were as good as
quadratic ones except for the data from the Athi plains (Table
2) . The response to rainfall increases positively with the total
amount per season. In the arid zone (Amboseli) biomass yield is
only 3.8 kg DM ha-1 per mm of rainfall increasing to 10.7 kg in
the tall Themeda grasslands in the Serengeti.
However, responses to rainfall in Tsavo were twice as high
as in Kiboko (8.0 and 4 . 1 kg DM/mm of rain respectively) although
the latter site receives more rainfall. This is most likely due
to differences in grass cover density. The average density in
Tsavo was 60%, whereas Too (1985) reported that herbaceous cover
in his trial area varied from 15-40%.
Biomass yield at the lower range of rainfall (100-200 mm)
was much higher in Tsavo and Kiboko than in the Athi plains
(Figure 1) . In Kiboko harvesting was confined mainly to rainy
seasons; hence for some of the cutting regimes, further growth
263
occurred after the last cut which was carried over into the next
season inflating the response to the next rains. However, in
Tsavo only new season growth was recorded.
In the Athi plains, a quadratic function gave the best fit,
because the data set represented a series of very dry and very
wet growing seasons. The trial started towards the end of a long
dry period (1972-76) , and during the five seasons from December
1974 to January 1977, seasonal rainfall averaged 200 mm producing
only 200 kg DM/ha or only 1 kg DM ha-1/1™ of rain. Rain fell in
small, isolated showers; hence the moisture index (P/Etj (cf.
Sombroek et al., 1982) for any three-week period never exceeded
0.5. This long dry period was followed by 20 months of good
rains. From March 1977 up to December 1978 seasonal rains
averaged 365 mm and in each of the four seasons there was a
-noisture surplus (P/Et>l) for at least six weeks. Early in 1978
over 500 mm of rain fell within 3.5 months producing close to 5 t
DM biomass (Figure 1) .
Soil and Herbaceous Cover
The effect of rainfall on productivity is influenced by soil type
and the density of the herbaceous cover. The high biomass of the
Themeda grasslands in the Serengeti and Athi plains is attributed
to the relatively high fertility of the deep Vertisols over
basalt (Sombroek et al , 1982). Standing biomass of 3-4 t DM/ha
were recorded on similar soils in Kajiado (Page et al, 1975;
Karue, 1975) and Masai Mara (Boutton and Tieszen n.d.). The
effect of soil type on primary production was studied in Tsavo.
For the same rainfall and plant density standing biomass on deep
well drained sandy clays was 30-55% percent higher than on
shallow gravelly soils (Table 3).
Table 3. The effect of seasonal rainfall, plant cover and
herbaceous soil type on end of season standing biomass
in Tsavo National Park, Xenya.
100
Rainfall (mm;)
300
Percent cover
Perennial grass
Annual grass
10 20 40 10
30 40 10
20 40
10 30 40
Standing biomass, kg DM ha"
■1
Deep soil*/
Shallow soil"/
200
150
450
290
760 600
590 450
1350
860
2270
1760
Derived from van Wijngaarden (1985) ,
a/ Ferral- and Luvisols.
b/ Cambisols .
264
The density of herbaceous cover is another important factor
influencing primary productivity. In Tsavo, primary productivity
increased threefold when grass cover increased from 20 to 80%
(Table 3). Low plant cover partially explains the poor response
to rainfall in the Kiboko and Amboseli areas. In the latter area
herbaceous cover is usually below 10% (Lamprey, personal
communication) .
Comparison with West and Southern Africa
Re-analysis of the data from West Africa for the rainfall range
of 100-600 mm produced a regression equation Y - -152 + 3.2 x (N
- 26; R - 0.73) giving a standing biomass of 800 and 1450 kg
DM/ha-1 for rainfalls 300 mm and 500 mm respectively. Similar
relationships were given by Penning de Vries and Djiteye (1982)
for the Sahel in Mali; or very close to those for Amboseli (Table
1) . The low Sahel primary productivity was attributed to (1)
poor and often sandy soils deficient in nitrogen and phosphorus;
(2) high rates of evaporation during the growing season; (3)
soil capping increasing runoff and impeding infiltration. As a
result, the vegetation consists mainly of short-lived annuals
with shallow root systems (Le Houerou and Hosts, 1977) , while
plant density is usually below 35% and end-of-season standing
biomass of annual grasses maximally was 1.0-1.5 t/ha when density
was 20-35% (Wagenaar and de Ridder, 1985) .
Rutherford (1978) showed that for the Colophospermum mopane
savanna in Zimbabwe, 300 mm and 500 mm rainfall produced
approximately 1000 and 1500 kg DM ha or close to the West
African data. Much lower productivity was recorded in Botswana
(22-23°S) during 1983-84 by Prince and Astle (1986). Although
total seasonal rainfall varied from 360 to 530 mm, standing green
biomass ranged only from 100-600 kg DM/ha (in 3 sites) with very
low cover value (6-20%) . High woody cover (mean of 30%) and poor
distribution of the rainfall were held responsible for the low
herbage growth. Dancy et al. (1986) for the same period and at
the same latitude in Botswana reported peak standing biomass
yields (in May 1984) from 200-1700 kg/ha . Low yields were
related to low plant density, which was in turn caused by high
grazing pressure. Ungrazed areas had dense cover (65%) and
yields of 1700 kg DM ha , which at an annual rainfall of 550 mm
conforms to the equation of Rutherford (1978) . This level of
productivity is similar to the average peak biomass of 1900 kg DM
ha-1 reported by APRU (1977) for seven moderately grazed ranches
receiving an average rainfall of 450 mm. Earlier work in
Botswana confirmed the effect of stocking rate on peak biomass.
McKay (1968) found that rangelands with an average rainfall of
300 mm grazed at high and moderate stocking rates for an
eight-year period produced 400 and 1360 kg DM ha-1 respectively.
The relationships established in Table 1 indicate that
productivity is generally lower than given by the equation of
Deshmukh( 1984) (300 mm:2350 kg DM ha-1, 500 mm:4050 kg DM ha-1),
but also that productivity for unit of seasonal rainfall in East
Africa is higher than that in either West or southern Africa.
The equation of Deshmukh (1984) is distorted by the inclusion of
annual and perennial grasslands together from very different
environments with both monomodal and bimodal rainfall regimes
extending from latitudes 23°S to 4°N. Furthermore, the
265
selective use of data sets of questionable value reduces its
predictive value. '
CONCLUSIONS
In conclusion, it appears that East African rangelands are more
productive per unit of rainfall than those in West and Southern
Africa. A combination of factors contribute to this: bimodal
rainfall usually falls in well determined seasons of two-three
months in length broken by a short dry season. This pattern
allows perennial grasses to survive across seasons but encourages
short-lived annuals to fill soace adding to the overall biomass
yield.
Soils are generally more fertile as they are derived from
basaltic parent material or when underlain by basement complex
surface horizons and are often enriched by volcanic ash deposits.
Finally, extreme caution is required when rainfall-biomass
equations that are derived from protected or ungrazed sites are
used for prediction of rangeland resources under heavy grazing
pressure. Such equations should be adjusted for actual plant
cover which is a prime determinant in the rainfall response
levels of rangeland vegetations
REFERENCES
APRU. 1977. An integrated programme of beef cattle and range
research in Botswana, 1970-77. Gaborone: Ministry of
Agriculture, Botswana.
Boutton, T.w. and Tieszen, L.L. n.d. Estimation of vegetative
biomass by spectral reflectance in an East Africa grassland.
J. Range Management (in press?) .
1/ A large part of the equation is determined by nine data in
Namibia and Marsabit in northern Kenya (cf Herlocher and
Dolan, 1980). According to Rutherford (1978, p. 629) "...the
data from Namibia are maximal values for localized areas.
It appears that these data be recognized as accumulated
plant mass present at the time of sampling and not as annual
production which might be in the order of half or less..."
The ten data sets from Marsabit include sparse annual
grasslands with 200 mm of rain together with perennial
grasslands producing up to 8 t DM ha-1 with 700 mm of rain.
These perennial grasslands are often found near riverbeds
(with run-on moisture )or on isolated mountain sites at 2000
m elevations. These two data sets represent 58% of the
total data used to construct the equation. Another six
points have been derived from Braun (1973, Figure 5); this
figure showed 63 data points covering four different
vegetation types two of which were used in Tables 1 and 2
(see equations 5 and 8) .
266
Braun, H.M.H. 1973. Primary production in the Serengeti:
Purpose, methods and some results of research. Ann. Univ.
d'Abidjan (E) 6:171-188.
Dancy, K.J., Webster, R. and Abel, N.O.J. 1986. Estimating and
mapping grass cover and biomass from low-level photographic
sampling. Int. J. Remote Sensing 7:1679-1704.
Deshmukh, I.K. 1984. A common relationship between precipitation
and Grassland peak biomass for East and Southern Africa.
Afr. J. Ecol. 22:181-186.
Herlocher, D.J. and Dolan, R.A. 1980. Primary productivity of
the herb layer and its relation to rainfall. IPAL Tech:
Rep. A-3, MAB Nairobi, pp. 22-29.
Jaetzold, R. and Schmidt, H. 1983. Farm Management in Kenya,
Vol. II C East Kenya. Ministry of Agriculture, Nairobi,
Kenya .
Karue, C.N. 1975. The nutritive value of herbage in semi-arid
lands in East Africa. II. Seasonal influence on the
nutritive value of Themeda trlandra. E. Afr. Agric. For. J.
40:372-387.
Le Houerou, H.N. and Hoste, C.H. 1977. Rangeland production and
annual rainfall relations in the Mediterranean basin and in
the African Sahelo-Sudanian zone. J. Range Management
30:163-178.
McKay, A.D. 1968. Rangeland productivity in Botswana. E. Afr.
Agric. For. J. 34:178-193.
Page, C.P., Lusigi, W.J. and Mumukha, P.W. 1975. Vegetation
studies in selected areas in Kajiado District. Mimeo,
UNDP/FAO Wildlife Management Project, Nairobi.
Penning de Vries, F.W.T. and Djiteye, M.A. (eds) . 1982. La
productivite des paturages sahe liens. Agric. Res. Reports
918. PUDOC, Wageningen, The Netherlands.
Potter, H.L. 1985. Aspects of climate, herbage growth and animal
production in a semi-arid area of Kenya. Ph.D. thesis,
University of Nairobi, Nairobi, Kenya.
Prince, S.D. and Astle, W.L. 1986. Satellite remote sensing of
rangelands in Botswana. Int. J. Remote Sensing 7:1533-1553.
Rutherford, M.C. 1978. Primary production ecology in Southern
Africa. In: M.J. A. Werger (ed.), Bibliography and ecology
of southern Africa. Dr. W. Junk Publishers, The Hague,
pp. 621-659.
Sinclair, A.R.E. 1979. The eruption of the ruminants. In:
A.R.E. Sinclair and M. Norton-Griffiths (eds), Serengeti,
dynamics of an ecosystem. University of Chicago Press,
Chicago. pp. 82-103.
267
Sombroek, W.G., Braun, H.M.H. and van der Fouw, B.J. A. 1982.
Exploratory Soil Map and Agro-climatic Zone Map of Kenya
1980. Kenya Soil Survey, Rep. EI, Nairobi.,
Too, D.K. 1985. Effects of defoliation frequency and intensity
on production of four burned and unburned bushed grassland
communities in South-Central Kenya. M.Sc. thesis, Texas A &
M University, Texas, USA.
Wagenaar, K.T. and de Ridder, N. 1985. Estimates of biomass
production and distribution in the Integrated Livestock
Project (ILP) in the Republic of Niger. Internal Report,
ILCA, Addis Ababa.
Western, D. and Grimsdell, J.J.R. 1979. Measuring the
distribution of animals in relation to the environment.
Handbook No. 2, African Wildlife Leadership Foundation,
Nairobi.
Wijngaarden, W. van. 1985. Elephants-Trees-Grass-Grazers:
relationships between climate, soil, vegetation and large
herbivores in a semi-arid savanna ecosystem. ITC Publ.
No . 4 , Enschede , Netherlands .
268
PRODUCTIVITY OF BEE? STEERS UNDER RANGELAND CONDITIONS
IN A SEMI-ARID AREA OF KENYA1'
H.L. Potter
Kenya Agricultural Research Institute
P.O. Box 21, Kikuyu, Kenya
Abstract
Results from 13 years of continuous grazing of a semi-arid
grassland by beef cattle are presented. Liveweight gain per head
was found to be independent of stocking rate in the range 2-5 ha
per animal . This is because grazing selection to improve dietary
intake quality above the gross level in the sward was possible
even at the highest stocking rate. Analysis of the rainfall and
growth data for the 37 seasons of the study to date showed that
78% of the variation in liveweight gain could be accounted for by
the relationship:
Liveweight gain per day (g/hd) = 247.5 x rainfall (mm/day) - 44.5
where the liveweight gain and rainfall figures are derived from
the average for each season. The relevance of such a predictive
relationship to requirements of settled pastoralists for methods
of objective estimation of the long term productive potential of
their holdings is discussed.
INTRODUCTION
Approximately 85% of Kenya can be classified as arid to semi-
arid, receiving less than 750 mm rainfall in four out of every
five years (Griffiths, 1962). The traditional land use in these
areas has been nomadic pastoralism, as with the unacceptably high
risk of crop failure, settled crop production was not practicable
under the low and very variable rainfall regime. However,
nomadic pastoralism can only provide for security of food supply
in such a very inhospitable environment if the necessary free
movement of people and their stock is possible through low
population density and where the land is not registered for
ownership (Henning, 1960; Allan, 1965). Efforts by both
Government and private development agencies towards improvement
of human health and veterinary services, particularly over the
last 30 years have considerably increased the pressure on the
land in the arid and semi-arid regions of Kenya. This has been
combined with the efforts by the Government to register land
titles, either on an individual or group basis, in order to
provide collateral for development loans and to make the
provision of government services easier in a settled situation.
A major result of these actions has been to limit the
possibilities for movement as a strategy to cope with the effects
of drought on feed and water supply to stock. Problems of
periodic shortages of grazing; and of water under the settlement
^ Views expressed in this paper are not to be considered as
representing the official policy of Kenya Agricultural
Research Institute or the Government of Kenya.
269
programme have to be dealt with on the individual holding in
situ.
Although the subsistence requirements for the settled
pastoralists in the semi-arid areas are likely to remain centred
on milk, there is evidence that provision of good marketing
facilities can result in stock sales and considerable changes in
herd management and in food purchases and consumption patterns
(White and Meadows, 1981) . Past strategies of maintaining
apparently high stock numbers to reduce the risks of complete
stock losses in drought, together with a high proportion of
mature females in the herd to provide for milk production on a
year-round basis are less justifiable or practical in a settled
situation. Where stock numbers are high in a settled situation,
drought effects may be expected to be severe, so it is desirable
that the long-term animal production potential of particular land
holdings are estimated using objective methods. This may provide
a basis for recommendations as to the stock numbers and herd
structure appropriate to the level of risk acceptable to those
depending on the holding for their livelihood.
The present paper presents data from a long-term study in a
semi-arid area of Kenya of the growth of beef steers in relation
to the pattern of rainfall distribution and amounts. It will be
suggested that the results of the study indicate a possible
methodology for the objective assessment of the long-term
production potential of such rangeland sites, where the climatic
database is adequate.
METHODS
Field data were collected at the Rohet Ranch sub-station of the
Kenya Agricultural Research Institute, situated about 15 km from
the Athi River Township (Lat. 1°20'S Long. 37°05'E). The site is
about 1500 m altitude. Rainfall is bimodal, falling mainly in
November/December and March/April/May with an average of about
580 mm being recorded over a period of fifteen years. The
rainfall is typically highly variable both annually and
seasonally, and vegetation growth closely follows its
distribution. Evaporation potential is generally high, with less
than 5% of the months having a Rainfall/Eo ratio greater than
unity. Under the existing standard classification of East
African rangelands the site lies in Ecozone IV (Semi-arid) (Pratt
et al, 1966). In the more recent agro-ecological zone
classification adopted by the Kenya Soil Survey the site lies on
the boundary between the Livestock-Sorghum Zone (UMS) and the
Upper Midland Ranching zone (UM6) (Jaetzold and Schmidt, 1983).
The soils at the site are generally shallow sandy-clay loams
derived from transported materials (Scott, 1962; Sombroek et al.,
1982). Vegetation is an open grassland dominated by rhemeda
triandra Forsk. The conditions at the site are representative of
extensive areas of Kajiado, Narok and parts of Machakos
districts, traditionally used by the Maasai tribe.
The work was designed to measure the growth performance of
beef steers under a range of set stock rates. For the first
eleven years of the trial rates were 2, 3, 4 and 5 hectares per
270
steer. For reasons which will be discussed further below, these
stocking rates were increased to 1.33, 2.0, 2.67 and 3.33
hectares per head thereafter. The initial rates were chosen
following discussions with local ranchers and extension personnel
and agree with the estimates made for ecozone IV (Pratt et al.,
1966; Jaetzold and Schmidt, 1983) and with estimates based on
rainfall distribution (Jahnke, 1982) .
The animals used were F1 cross steers from Bos indicus
(local Boran type) dams and Bos taurus (Hereford type) bulls,
such crosses being typical of the type of animals to be found in
the commercial ranch in this part of Kenya. The animals entered
the trial at about 15 months of age and were removed at about 33-
36 months or at a weight of 450-500 kg. Liveweight gain of the
animals was considered a good indicator of pasture condition.
During the 13 years of the trial, eight series of animals were
used. Set stocking was used throughout the trial as the limited
data available from tropical areas did not show any evidence in
favour of rotational grazing (jt Mannetje et al., 1976).
Also, set stocking relates closely to current land usage and
will provide baseline data, with the more resource-demanding
rotational systems being tested in later studies. Because of
predator risks the experimental animals were removed to a secure
night paddock during darkness, so they grazed from 0630 to 1830
hours each day.
For reasons of cost and availability of land, only a single
paddock for each stocking rate was used, with four animals per
stocking rate initially, raised to six in the twelfth year. All
animals were sprayed against ticks twice weekly and were
routinely vaccinated against prevalent diseases. Weights for
each animal were recorded at the same time each week following
overnight fasting and withdrawal of water.
Samples for estimation of dietary composition were obtained
from oesophageal fistulated animals similar to those used in the
grazing study. Sampling was carried out for about 30 minutes at
each sampling, with either three or four animals being used twice
a day for three days.
Analysis of the collected material was confined to
laboratory chemical analysis using standard methods (Van Soest,
1963; and AOAC, 1970) with the data being calculated on an ash-
free basis in an attempt to reduce the effects of salivary
contamination (Marshall et al., 1967). Twenty 2 m x 1 m sample
guadrats were harvested at random per paddock by cutting at about
5 cm height to give an indication of the composition of the
overall sward at the time of the oesophageal sampling.
RESULTS
Figures 1 to 8 indicate the trend of liveweight gain for each of
the eight series of animals for the four stocking rates used,
with the cumulative rainfall being also shown on each of the
figures for the relevant period. The weights are indicated on a
three-week basis for clarity of presentation. The generally
close similarity in performance between steers on the different
271
stocking rates is notable. The pattern of growth was closely
associated with the pattern of rainfall.
Figure 1. Liveweight gain/rainfall,
(3.6.74-20.1.75) .
animal series 1
Cumulative liveweight
goin (kg/hd)
2 ha/hd
Cumulative rainfall
(mm) 
4 5 6 7 8
Three-week periods
272
Figure 2. Liveveight gain/rainfall, animal aeries 2
117. 2. 75-30. 8. 76) .
Cumulative liveweight Cumulative rainfall
gainlkg/hd) (mm)
230 2 ha/hd 920
210 - ho/hd 840
A. kn /kH A
190 *>no/no 760
170
5 ho/hd . 660
150
n Season
600
130 " 9 520
1 10 ;
Season
(wet)
440
90 8
(dry)
360
70 T/^Seoson7 7
280
50 (wet) 200
30
l'r*j$ft (dry)
5 yrT/ C
1 20
10
■ rEVi i i—i_i i—l_ i i i i i i j i i-j i— ■ ■ ■ ■ -
40
2 4 10 12 14 16 18 20 22 24 26 28
Three- week periods
273
Figure 3. Liveweight gain/rainfall, animal series 3
(20.7.77-11.9.77).
Cumulative liveweight
gain(kg/hd)
150
140
130
120
no
100
90
80
70
60
50
40
30
20
10
0
-10
-20
Cumulative rainfall
(mm)
2 ha/hd
3ha/hd
4 ha/hd
-o 5 haAd -< h
-
: A A ifSeason //
'3 Vif--
("">//» /
Season
////J
-
12
(dry)
Season Jt-o/// yY/A r-
ii wy\J/ \yx /
Season
(wet)£/ O If 1
10
(dry) ~^<r Ai l-
- y
-
i i i V i i i i i i i i i i
700
650
600
550
500
450
400
350
300
250
200
150
100
50
2 3 4 5 6 7 8 9 10 II 12 13 14
Three -week periods
274
Figure 4. Liveweight gain/rainfall, animal series 4
(1.8.77-8.1.79) .
Cumutotive liv* weight
goin(kg/hd)
Cumulative rainfall
(mm)
 
8 10 12 14 t6 18 20 22 24 26
Three -week periods
275
Figure 5. Livawaight gain/rainfall, animal series
(11.1.79-22.3.81).
Cumulative Nveweight
goin(Kg/hd)
340
Cumulat've rainfall
(mm)
 
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38
Three- week period*
276
Figure 6. Livoweight gain/rainfall, animal series
(16.3.81-21.6.82).
Cumulative liveweight
gain( kg/hd )
Cumulative rainfall
(mm)
 
-2 0' ' ' '
6 8 10 12 14 16 18 20 22
Three-week periods
277
Figure 7. Livoweight gain/rainfall, animal series
112.7.82-23.7.85) .
Cumulative livawalght
gainOig/hd)
Cumiiafive rainfall
(mm) 
6 10 14 18 22 26 30 34 38 42 46 30 34
Three-week period*
278
Figure 8. Livewoight gain/rainfall, animal series
(9.7.85-30.3.87) .
Cumdative liveweight Cumulal ive rainfall
gain [kg/nd) ( mm)
200
133 ho/hd --*"" i
1050
200 ho/hd
267 ho/hd
- 950
850
180 A A
-
ifin 333 ho/hd , -+ 750
140 Rainfall
650
120
550
inn
Season Season 450
80 Season
(dry)
36
(wet)
37 ■
350
L 34bU O Season [intermediate)
4n
33 //Season
250
. Wry) if 35 150
20 ( (wet)
50
0
1 1 1 1 1 1 1 111 1 1 1 1 1 1 L ± .J- A- 1 1 1 1 1 1 ■ ■ 1
2 4 6 8 C 12 14 16 18 20 22 24 26 28 30
Three -week periods
Tables 1 to 6 indicate the results of the dietary quality
examination, using crude protein as the quality parameter,
derived from the fistula samples, together with an indication of
dietary selection practised by the animals as represented by the
differences between sward and fistula samples.
Table 1. Protein content of sward and oesophageal fistula
samples, June 1974.
Component Stocking rate ha/head
(Ash free basis) Mean S.E.
Crude protein % fistula 8.0a 8.2a 8.5b 8.1a 8.2 0.1
% sward 5.8a 6.0a 6.0a 5.9a 5.9 o.l
% increase 37.9 36.7 41.7 37.3 37.0
279
Table 2. Protein content of sward and oesophageal fistula
samples, October 1974.
Component Stocking rate ha/head
(Ash free basis) 2 3 4 5 Mean S.E.
Crude protein % fistula 4.5a 5.3b 6.2c 4.7a 5.2 0.1
% sward 3.5a 3.5a 3.8a 3.4a 3.6 0.2-
% increase 28.6 54.4 63.2 38.2 44.4
Table 3. Protein content of sward and oesophageal fistula
samples. May 1975.
Component Stocking rate ha/head
(Ash free basis) 2 3 4 5 Mean S.E.
Crude protein t fistula 12.1a 12.4a 11.1b 12.6a 12.1 0.3
% sward 9.4a 9.5a 9.5a 9.6a 9.5 0.3
% increase 28.7 30.5 16.8 31.3 27.4
Means with the same subscript letter do not differ significantly
at P-0.05 (Duncan, 1955).
Table 4. Protein content of sward and oesophageal fistula
samples, August 1975.
Component Stocking rate ha/head
(Ash free basis) 2 3 4 5 Mean S.E.
Crude protein % fistula 6.2ab 6.5b 6.6b 5.5a 6.2 0.3
% sward 4.1a 4.2a 4.4a 4.1a 4.2 0.3
% increase 51.2 54.8 50.0 34.1 47.6
280
Table 5. Protein content of sward and oesophageal fistula
samples, September 1974.
Component Stocking rate ha/head
(Ash free basis) 2 3 4 5 Mean S.E.
Crude protein % fistula
% sward
% increase
7.3a
5.1a
43.1
8.0b
6.2b
29.0
8.1b
5.4a
50.0
6.8a
5.1a
33.3
7.6 0.3
0.35.5
38.2
Table 6. Protein content of sward and oesophageal fistula
samples, June 1977.
Component Stocking rate ha/head
(Ash free basis) Mean S.E.
Crude protein % fistula 10.9a 10.8a 10.5b 10.5a 10.7 0.3
% sward 8.4a 8.5a 8.6a 6.5a 8.5 0.4
% increase 29.8 27.1 22.1 23.5 25.9
Means with the same letter do not differ significantly at P=0.05
(Duncan, 1955) .
DISCUSSION
The results for the dietary composition and selection presented
in Tables 1 to 6 indicate that there was apparently very little
difference in the quality of diet eaten by the animals on the
different stocking rates at different seasons of the year
covering dry, wet and intermediate season conditions. Throughout
the whole of the initial eleven years of the trial there was no
indication of any statistically significant differences in per
animal performance across the stocking rates so that performance
at 2 hectares per head was as good as at 5 hectares. The
inference to be drawn from this is that quality of the diet had
the greatest effect on performance, with quantity of herbage
available being of lesser importance, provided it was sufficient
so that dietary selection could be practised effectively. It has
been suggested that the relationship between animal performance
and stocking rate would be highly site-specific and not conform
to the simpler linear (Jones and Sandland, 1974) or culvilinear
(Mott, 1960) relationships proposed earlier. To test what would
happen to the sward and animal performance an increase in
stocking rate in the present trial was made at the end of the
281
eleventh year in order to determine whether selection would still
be effective at higher stocking rates. Since long-term
degradation of the sward had not been observed up to that point
it was postulated that it might occur as at higher stocking rates
together with sward damage by trampling and soiling by dung and
urine.
The results of the eighth series of animals to date (Figure
8) do not show any clear indication of a reduction of performance
in the 1.3 3 ha/head stocked paddock so that the effects just
mentioned may not yet become evident over the time span of 18
months .
The general association between the performance of the
animals and the pattern of rainfall distribution suggests that a
predictive relationship may be derived from the results so far
obtained. The growth curves were broken down into a series of
wet seasons associated with the presence of rainfall,
interspersed with periods of little or no rainfall in the dry
seasons, the seasons being indicated in Figures 1 to 8. An
earlier' analysis of the growth and rainfall data (Potter, 1985)
indicated a time lag effect in the relationship between rainfall
and liveweight gain. Due to the time taken for the vegetation to
respond to the onset of the rains, any rain falling in the three-
week period immediately before the weight measurements appeared
to be less influential than rain that had fallen three weeks
earlier, that is from three to six weeks before weighing.
Indeed, the analysis indicated that rainfall in the period
immediately before the weighing was negatively correlated with
growth. This may be explained by a loss in weight associated
with the onset of rains resulting from a combination of reduced
temperatures and a great change in the solid/water ratio and
nutritive content of the diet (French, 1956; Denis et al., 1976).
Using a three-week lag as discussed above, the data
presented in Figures 1 to 8 were analysed further by examination
of the growth rates in particular seasons in relation to the
quantity of rainfall during the season. The seasons used are
indicated on the respective growth curves as indicated on the
figures. Since the results for all four stocking rates were
combined, analysis of variance for the performance of the
separate rates for each series of animals indicated no
statistically significant differences (P=0.05). Eq. 1 indicates
the relationship derived between daily rainfall during the
seasons and liveweight gain per day. The data for the eighth
series of animals were included in the derivation of the
relationship even though the mean stocking rate was 2.3 3 ha per
head as opposed to 3.66 ha per head for the earlier series as
analysis indicated that all eight series could be considered part
of one homogeneous data set.
Y = 247.5 (S.E. 22.4) X - 44.5 (S.E. 49.9)
(-Equ. 1) R2 = 0.78
where Y represents daily liveweight gain in grammes per head and
X represents daily rainfall in mm. A linear relationship, which
accounts for 80% of the variation in liveweight gain suggests
that rainfall data may be useful as a basis for predicting animal
performance. Kenya is fortunate among developing countries in
282
having a relatively large database of long-term rainfall
(up to 80 years) even in some of the semi-arid areas.
records
Even though there may be an element of bias in the siting of
the rainfall stations in positions which may be somewhat wetter
than the surrounding area (Potter, 1987) , the data were used to
calculate the probability of seasonal rainfall for the Rohet
Ranch site combined with the Athi River Railway Station site some
15 km distant over a period of 65 years. Using a relationship
derived as above the probability of annual livestock production
levels were calculated on a long-term basis. Figure 9 indicates
the probability of achieving a particular level of annual
productivity per animal using the long-term rainfall data.
Figure 9. Long-term probability estimate for annual liveweight
gain per head for beef steers at Rohet Ranch.
Annual liveweight
gain (kg/hd)
160
130
140
130
120
110
100
90
80
70
60
50
0 10 20 30 40 50 60 70 80 90 100
Probability of achieving a given liveweight gain (% of years )
The above analysis assumes that the quantity of herbage
available to the animal is not limiting selective grazing during
any one year. Up to the highest level of stocking used to derive
the relationship (Eq. 1) no restriction in selective ability was
evident, at least for the first eleven years of the study. The
term changes in the sward vigour which may result in a decline in
quantity of production or alteration in sward composition and
therefore selection possibilities, but to date these have not
become apparent. Direct measurements of sward productivity may
not be adequate to indicate any such decline as the method of
harvest may have a considerable effect on the yield figure
measured. This point has been discussed in a related paper
(Potter and Said, in press).
283
It is intended that this study will be followed by others
which will provide the necessary data to test the methodology at
other sites, as no similar data are at present available from any
similar long-term study in the region. Should the methodology
prove to be generally applicable, even outside East Africa, the
potential productivity for sites will be predictable to a level
of accuracy useful for management planning recommendations.
REFERENCES
Allan, W. 1965. The African husbandman. Oliver and Boyd,
Edinburgh and London.
AOAC. 1970. Official methods of analysis of the Association of
Official Analytical Chemists, llth Edition. Edited by H.
Horowitz. AOAC, Washington.
Denis, J. P., Blancou, J. and Thiongane, A.I. 1976. Note
preliminaire sur la "crise de juillet". Resultats de
l'experience 1976 au C.R.Z de Dahra. Research Note C.R.Z.
Dahra, Senegal. (mimeo)
Duncan, D.B. 1955. Multiple range and multiple F tests.
Biometrics 11:1-42.
French, M.H. 1956. The importance of water in the management of
cattle. E. Afr. Agric. F. J. 21:171-181.
Griffiths, J.F. 1962. The climate of East Africa. In: W.E.
Russell (ed.), The natural resources of East Africa. D.A.
Hawkins Ltd. in association with East African Literature
Bureau, Nairobi.
Henning, R.O. 1960. Range management in the pastoral areas of
Kenya. J. African Admin. 13.
Jaetzold, R. and Schmidt, H. 1983. Farm management handbook of
Kenya: Natural conditions and farm management information.
Ministry of Agriculture, Nairobi.
Jahnke, H.E. 1982. Livestock production systems and livestock
development in tropical Africa. Kieler Wissenschaftsverlag,
Kiel.
Jones, R.J. and Sandland, R.L. 1974. The relation between animal
gain and stocking rate. Derivation of the relation from the
results of grazing trials. J. Agric. Sci. (Camb.) 83:335-
342.
't Mannetje, L. , Jones, R.J. and Stobbs, T.H. 1976. Pasture
evaluation by grazing experiments. In: H.H. Shaw and W.W.
Bryan (ed.), Tropical pasture research. W.W. Bull. 51.
Commonwealth Bureau Pastures and Field Crops, Hurley, U.K.
Marshall, B. , Torell, D.T. and Bredon, R.M. 1967. Comparison of
tropical forages of known composition with samples of these
forages collected by oesophageal fistulated animals. J.
Range Mgmt. 20:317-320.
284
Mott, G.O. 1960. Grazing pressure and the measurement of pasture
production. Proc. 8th Int. Grassld. Cong., Reading,
pp. 606-11.
Potter, H.L. 1985. Aspects of climate, herbage growth and animal
production in a semi-arid area of Kenya. Ph.D. thesis,
University of Nairobi, Kenya.
Potter, H.L. 1987. The effects of climatic variability on
livestock production. In: M.M. Parry, T.R. Carter and N.T.
Konijn (eds) , The impact of climatic variations on
agriculture. Volume 2. Assessments in semi-arid regions.
Reidel, Dordrecht, The Netherlands.
Potter, H.L. and Said, A.N. (in press). Effects of defoliation on
vegetation growth in a semi-arid area of Kenya. 1. Annual
dry matter production. E. Afr. Agric. For. J.
Pratt, D.J., Greenway, P.J. and Gwynne, M.D. 1966. A
classification of East African rangeland - with an appendix
on terminology. J. Appl. Ecol. 3:369-382.
Scott, R.M. 1962. The soils of East Africa. In: E.W. Russell
(ed.), The natural resources of East Africa. D.A.
Hawkins in association with East African Literature Bureau,
Nairobi.
Sombroek, W.G., Braun, H.M.H. and van der Pouw, B.J. A. 1982.
Exploratory soil map and agro-climatic zone map of Kenya
1980, Kenya Soil Survey, Rep. EI, Nairobi.
Van Soest, P.J. 1963. The use of detergents in the analysis of
fibrous feeds. II. A rapid method for the determination of
fibre and lignin. J. Ass. Off. Agric. Chem. 46:829-835.
White, J.M. and Meadows, S.J. 1981. Evaluation of the
contribution of group and individual ranches in Kajiado
District, Kenya, to economic development and pastoral
production strategies. Min. of Livestock Development,
Nairobi.
285
CLASSIFICATION OF KENYA RANGELAND
Moses Ooro Olang
Tana and Athi Rivers Development Authority
P.O. Box 47309, Nairobi, Kenya
Abstract
There is need for another committee to be set up to review the
boundaries of the existing ecological zones. The East African
Range Classification based their classification on ecological
potential of the land and the description of the existing
vegetation with emphasis on nature and relative contribution of
woody plants and grass.
Detailed work must be carried out on broad eco-climatic
zones so that they are subdivided on physiographic and edaphic
criteria to yield precise ecological land units. The author has
tried to subdivide the existing ecological zones into 29
ecological units based on total and monthly rainfall
distribution. Also used is the livestock utilisation of these
areas. The latter criterion is useful when it comes to planning.
Rainfall controls soil formation and its potential, and that is
why it became a useful tool in the subdivision also.
INTRODUCTION
The East African Range Classification Committee which was set up
to classify rangeland ran into a problem because those who had
tried to classify rangeland before them used grassland
composition (Pratt and Gwynne, 1977) . This approach was not
satisfactory because the variation in grass composition in East
Africa is so great that to classify rangeland solely by reference
to one or two of its constituent grass species would give an
incomplete idea of its character.
The committee then used three characteristics to classify
rangeland; soil moisture availability, climax vegetation and land
use. They came up with six ecological zones. Rangeland was
divided into three zones i.e. IV, V and VI. There has always
been a need to review these zones because variation in
productivity is too great. For example, two areas may receive a
total of 500 mm of rainfall annually.
One area may receive all that rain in one month while
another area may receive it in two months. These two areas will
differ in their land potential and grass composition. There has
therefore been a need to review these ecological zones for proper
planning and effective management.
Kenya Rangeland Ecological Monitoring Unit (KREMU) divided
rangelands of Kenya into 44 eco-units based on area occupied by
animals belonging to a group of people all year round. This
criterion of land use was not enough to delineate rangeland into
ecological units which could later be used for planning purposes.
Other factors like soil types, rainfall, soil moisture, and
vegetation types had to be considered.
286
METHODOLOGY
The methodology involved looking at those factors already
mentioned to see how they control or affect land potential.
Soil Types
Soil undergoes a series of development stages as climate and
organisms act upon the original rock or parent material (Stoddard
and Smith, 1955) . Thus soil alone cannot influence the
productivity of an area because its development is controlled by
other factors. Soil which has developed through interaction of
high temperature, low rainfall and little organism will be poor
in plant nutrient.
Vegetation Types
Vegetation is the product of its environment. The environment
includes land form, soil, climate, animal and man. Often, of
course, the present vegetation represents a stage of regression
from a more highly developed or rigorous community which has been
brought under stress, perhaps through use by man.
Rainfall
Rainfall readings were divided into seven segments of 100 mm
interval from 200 mm to 800 mm. Isohyets were drawn to join all
areas with the same readings. Another rainfall parameter which
was used was the monthly distribution.
Soil Moisture Pattern
Soil moisture is closely related to grass production. Cassady
(1973) reported that storms which produced 175 mm to 280 mm of
rain within 30-60 days soaked the soil profile to depths of 120-
180 cm. Thus soil which is capable of retaining most of the rain
water and make it available to plants has higher plant production
than either sandy soil or soil with hard layer which does not
allow much percolation. Soil moisture was not used because not
much research has been done on the major soils.
Land Use
Generally land use in most areas of rangeland is pastoralism.
The area was used by either one ethnic group or a section of that
group without coming out of it unless it was very necessary.
RESULTS
Isohyets of 500 mm, 400 mm, 300 mm and 200 mm were used as
boundary lines. Another boundary line was added showing
livestock movement by one ethnic group or a section of a group.
Thus a few modifications were made to the isohyets to accommodate
these movement areas. For instance, Ewaso Ngiro was curved out
of Mandera and Khorof Harar because of increased soil moisture
from the underground river.
287
The whole rangeland was divided into 29 ecological units, as
shown in Figures 1, 2 and 3. The present 29 ecological units are
manageable units as they are being used at the moment. There is
some relationship between the boundaries of land potential and
rainfall monthly distribution.
Figure 1. Rainfall map of Kenya.
 
Mton annual rainfall
Ecological Units
1. Lotikipi Plain
This eco-unit covers the upper half of Turkana District. The
rainfall ranges from 300-500 mm annually, and it decreases from
the Sudan border to Lake Turkana. The driest months are January
288
through March while the wettes
other months receive a little ra
with rainfall while the Lotikipi
The western bushland has grasses
and Eragrostis racemosa. The
palatable grasses like Echinochl
but when misused, the area
aegyptium, Aristida mutabilis
accanthoides and Cleome hirta (0
t are April through June. The
in. Bush density also decreases
Plain itself is open grassland.
dominated by Sporobolus confinis
central plain is dominated by
oa haploclada and Chloris virgata
is invaded by Dactyloctenium
Pupalia lappacea, Barleria
lang, 1984b) .
Figure 2. Ecological zones of Kenya.
 
289
Livestock concentrate in the highland around Lake Turkana
during the dry season of January through March but move down in
the plain during the rainy season. From there they move towards
the Sudan border from July onwards, but cattle rustling
interferes with the grazing in that area.
2 . Lokichar Eco-unit
This is the lower part of Turkana District. Rainfall ranges from
500 mm on the western border to below 100 mm around the town.
The dry central scrubland is divided almost in half by the
Turkwel river. The area receives rainfall in April and May and
very little along the Uganda border in June. The short rains of
November and December are hardly experienced because of the
prevailing wind which blows from Lake Turkana to the western
border from August through December.
Figure 3. Ecological units in Kenya.
 
290
Riverine vegetation are dominated by Acacia tortilis,
Hyphaene coriacea and Salvadora persica. Areas away from the
river are dominated by A. reficiens , whose density and tree
height increases with increasing rainfall.
The area is dominated by annual plants like Digitaria
velutina, Aristida mutabilis, Dicoma tomentosa, Pupalia lappacea,
Zaleya pentandra and Eragrostis cilianensis.
D. velutina dominates the western border, while A. mutabilis
dominates rocky areas and E. cilianensis is the common grass
found in the remaining areas. Further to the west and the
southern borders perennial grasses like Panicum maximum,
Digitaria milanjiana and Echinochloa haploclada are found. When
the area is misused, it is invaded by A. mutabilis and Cleome
monophylla.
3 . Pokot Eco-unit
This eco-unit forms the northern end of the Charangani hills.
Rainfall ranges between 500 mm and 700 mm annually. The area has
thick bush dominated by Acacia tortilis, A. reficiens, A.
brevispica and A. mellifera. Grasses are mainly desirable
species like P. maximum, Chloris gayana and Cenchrus ciliaris.
4. Maralal Eco-unit
This area runs from Sukuta Marmar through Maralal town up the
northern end of the district. There is a ragged valley bordering
Turkana District where vegetation is still in its potential.
Rainfall ranges from 300 mm to 600 mm. The area east of Maralal
hill receives the lowest amount of rainfall. Rainfall falls
during the months of March to April and October to November.
Desirable grasses like Themeda triandra, C. dactylon and D.
milanjiana are systematically being replaced by Aristida adoensis
and Harpachne schimperi which are increasers and Microchloa
kunthii which is an invader.
5. Cherangani Eco-unit
This eco-unit mainly comprises the eastern footslopes of the
Cherangani hills and the Kerio Valley. The western side of the
river is dominated by Acacia tortilis while the eastern dry side
is dominated by A. mellifera, and the grasses are mainly C.
dactylon, A. mutabilis and Urochloa brachyphylla. There is high
concentration of livestock in the valley because of water
availability.
6. Baringo Eco-unit
This eco-unit lies between the Tugen hills and Laikipia
escarpment with Lake Baringo centrally located. The vegetation
cover is good to the south around Chemogoch where grasses like T.
triandra, E. superba and C. dactylon are still available. The
vegetation cover and palatability improves again north of Lake
Baringo. The area around the lake has been overgrazed and is
dominated by Acacia adoensis, Sporobolus fimbriatus, Microchloa
kunthii. Tragus terrestris and Dactylocteneum aegyptium.
291
Rainfall ranges between 300 nun and 500 mm, decreasing from south
to north.
7. Laikipia Eco-unit
This eco-unit is bordered by the Ewaso Nyiro river on the north,
Aberdares mountains to the south and Mt. Kenya to the east. Its
rainfall ranges from 400 mm to 600 mm annually, decreasing
northward. The grass cover is dominated by Themeda triandra and
C. dactylon. Cynodon increases as Themeda is overgrazed, but the
area is generally still in good condition as most of it is
divided into ranches which are well managed. Acacia
drepanolobium is the dominant woody species. Areas which remain
dry for longer periods support short flat-topped A. drepanolobium
while the more fertile areas that remain moist for longer periods
support very tall A. drepanolobium. Areas which are overgrazed,
seen mostly towards Rumuruti, are dominated by unpalatable
species such as H. schimperi and A. adoensis.
8 . Mukogodo Eco-unit
This is the eastern end of Laikipia district where the Ndorobo
ethnic group lives. In this hilly area human population has
greatly increased.
Although the area gets good rainfall, overgrazing and the
hilly nature of the area has greatly accelerated soil erosion.
Woody vegetation is being depleted through charcoal burning.
Harpachne schimperi and Microchloa kunthii are replacing
palatable species such as T . triandra.
9. Isiolo Eco-unit
This eco-unit runs from the Somali strip on the western end
through Shaba and Buffalo Game Reserve to Modogashe town.
Rainfall and land potentiality decreases from west to east. The
long rains fall between April and May while the short rains fall
in November. Herbaceous plants are dominated by T. triandra, P.
maximum, Cenchrus ciliaris, Sporobolus marginatus, Panicum
coloratum and Chloris roxburghiana. C. roxburghiana and C.
ciliaris dominate around Garba Tula while C. dactylon dominate in
the northeast of Merti town.
Areas which have been overgrazed around Garba Tula are
dominated by Aristida papposa and Digitaria velutina. The woody
plants are dominated by Commiphora sp. while areas around Kula
Mawe through to the Shaba Game Reserve is dominated by Acacia sp.
10. Chalbi Desert Eco-unit
This eco-unit can easily be divided into three sub-units. The
Kaisut area has quaternary sediment and lies south of Marsabit
hill. The second is the old lake bed of Lake Chalbi which is
mostly covered by saline alkaline alluvial materials. The last is
the recent alluvial and aeolian material overlying the eastern
and north-western edges of the Chalbi desert (Herlocker, 1979) .
The desert forms a swamp of an interior drainage system with
perennial springs at the edge of the desert.
292
Long rains fall in April and May while short rains fall in
November. Evapotranspiration in this area is so high that the
rains support mostly annual species. The desert supports salt-
loving species like Sueda monica while other areas support annual
grasses such as Aristida papposa. While one view was that the
area was dominated by Chrysopogon/Aristida , Heady (1960) thought
Aristida sp. was the most dominant grass. Pratt and Gwynne
(1977) called the area barren land. Composition and production
of grasses vary greatly from place to place and from time to time
reflecting high variability in rainfall (Herlocker, 1979) .
11. Marsabit Eco-unit
This eco-unit is found in two areas: around Marsabit hill and Mt.
Kulal. It is restricted to the upper elevation of these
mountains.
(a) Marsabit: The southern and eastern slopes of the
mountain are wetter than the others (Herlocker, 1979) .
The deciduous trees consist of species like Commiphora
spp. and Croton dichogamus. Harrisonia abyssinica is a
common shrub. Fire seems to be an important factor in
degrading the vegetation. Grasses are mostly
Dichanthium/Themeda , but overgrazed areas have
unpalatable herbs or grasses such as Eragrostis
tenuifolia and Chenopodium spp.
(b) Mt. Kulal: Like Marsabit rainfall is higher and better
distributed than other parts of the district.
Herlocker (1979) noticed that perennial grasses grow
where openings occur in the bush canopy. Grasses are
especially like those found in Marsabit. Most of the
pastoralists who bring their livestock up-hill in the
morning during the dry season descend in the evening
because of cold nights.
12 . samburu Eco-unit
This eco-unit runs from the Maralal (Karisia) hills to the Wamba
and Oldoinyo hills to the east of the Elbarta plains. Animals
graze in the Suare and Ilponyeki plains during the wet season but
come to water at Seyia river. During dry seasons most of the
animals are taken into the hills while others go to Kom area.
Range conditions in this area seem to be decreasing as readings
on herbaceous cover showed a change from 42% in 1961 to 31% in
1970, then 22% in 1980. Plant composition of desirable species
also changed from 68% in 1961 to 49% in 1970, then to 33% in 1980
(Skovlin, 1980).
Ilponyeki plain is dominated by Cynodon dactylon and
Aristida papposa.
13. Ewaso Ngiro Eco-unit
This eco-unit receives the same amount of rainfall as Mandera and
Khorof eco-units except it gets more water through the drainage
system. Lagh Bogal which starts from the border of Marsabit and
Wajir around Tamsa swamp runs through the Boji plain and ends at
the Kenya/Somalia border. The other is Lagh Awaro which forms
293
the northern end of the unit. Lagh Dera is actually the lower
reaches of Ewaso Ngiro where water from the river runs
underground. It has the highest number of perennial grasses. P.
coloratum, Sporobolus spp., A. adoensis, Rhynchetrum spp.,
Enteropogon macrostachys and Eragrostis caespitosa have been
recorded (Olang and Karime, 1981) .
14 . Mandera and Khorof Eco-units
From the Ethiopian border the area is mostly rugged with a lot of
small water streams from west to east. There are a number of
watering points along the international borders with Ethiopia and
Somalia. Herbaceous cover in the south is dominated by Justicia
flava, Aristida kenyensis and Aristida adscensionis . Rainfall
ranges between 300 mm to 400 mm annually although the central
part falls below 300 mm.
15. Tupo Eco-unit
This eco-unit lies below Marsabit and Mandera/Khorof eco-units.
It remains wet longer than the areas around. The good grass
cover found in the Lagh bogal drainage system starts from Tamsa
swamp which is in the southern end of the unit. C. dactylon is
one of the most important perennial species.
16. Huri and Sololo Eco-unit
The hills are the southern end of the Ethiopian hills with the
rainfall near the border being more than 800 mm. The hills do
not have extensive areas of closed forest, but many valleys,
gullies and steep hillsides have dense cover. Among the common
trees are Croton macrostachyus , Euphorobia candelabrum, Olea
africana and Bauhinia tomentosa.
Open hills and gentle slopes in the wetter western slopes
carry plants like Combretum molle, Lannea spp. and Ozoroa spp.
Much of the Hurri hills is grass covered which is edaphically
controlled. Soils are heavy clay, with impeded drainage and
frequent flooding (Synott, 1979) . Themeda/Chrysopogon grassland
occurs at the eastern base and to the northeast of Hurri the
hills (Herlocker, 1979) .
17. Ileret Eco-unit
This eco-unit includes Sibiloi National Park where the herbaceous
cover is dominated by Dactylocteneum aegyptium and Aristida
mutabilis which are both annual with one perennial Sporobolus
ioclauds (Odhiambo, 1981) . Other grass species include
Enneapogon cenchroides, Aristida adoensis, Tetrapogon tenellus,
Tragus berteroniaus and Stipagrostis uniplumis. Further north
towards Sabarai the number of species and percent cover
decreases. A. mutabilis becomes dominant. Other species like
Tetrapogon tenellus and Tragus berteronianus are found in low
depression areas where soil moisture is higher.
18. Kitui Eco-unit
This eco-unit covers the whole of Kitui District, the northern
part of Machakos and parts of Embu and Meru districts. Livestock
294
movement is not so conspicuous here because of cultivation and
some improved ranches .
Kitui District used to support perennial grasses like P.
maximum, Sehima nervosum, H. contortus, E. superba, E.
macrostachyus, D. milanjiana, C. roxburghiana and T. triandra
(Skovlin, 1980) . Overgrazing and shifting cultivation has
resulted in increased bare ground and decreased cover by
perennial grasses. Annual grasses and herbs like Brachiaria
leersioides, Justicia exigua, E. cilianensis, Tetrapogon tenellus
and A. adscensionis dominate most of the overgrazed areas
(Muchoki, 1982).
19 . Nakuru Eco-unit
This eco-unit covers the Nakuru, Naivasha and Elementaita area
where most of the savanna has been turned into wheat farms.
Dominant grass in its potential is T. triandra, but with
increased grazing, C, dactylon and Sida cuneifolia become
dominant. With overgrazing, H. schimperi and Micqsrochloa
kunthii increase or invade. Vegetation cover is changing from
grassland to bush through the encroachment of Tarchonanthus
camphoratus in the Gilgil area.
20. Narok Eco-unit
The Loit hills form the southeastern border while Mau escarpment
forms the northern border. These two hills attract rain from
south and west making the unit wetter than normal . The
northeastern part of the plain receives fairly little rain. The
Acacia/Pennisetum shrubland in the black cotton soil has a
mixture of Pennisetum mezianum, C. dactylon, Achyropsis
greenwayii, Justicia elliotii and Dychoristes radicans.
Themeqsda triandra dominates in red or sandy loam soil (Olang,
1984b) . Overgrazing in black cotton soil leads to increase in
cover by J. elliotii and D. radicans . Overgrazing in Themeda-
dominated areas leads to increase in woody plants such as Lippia.
spp. , Indigofera spp., Justicia exigua and grasses such as M.
kunthii and H. schimperi.
21. Magadi Eco-unit
From Suswa dam down to the beginning of Lake Magadi is the Kedong
valley which has good stands of Cynodon plectostachyus and T.
triandra. The area west of Lake Magadi is dominated by
Sporobolus helvolus which covers the whole area during the rainy
season, but almost all of it is grazed during the dry season.
The area east of Lake Magadi is covered with rock pebbles
with scattered Aristida spp. during the rainy season, but all of
it disappears as the dry season comes. Areas further east and
southwards are dominated by T. triandra, C. dactylon, P. maximum
and P. mezianum.
22. Tsavo Eco-unit
The soils are weathered and strongly leached. The area receives
rainfall between 400 mm and 600 mm annually. Areas receiving the
295
highest amount of rain are Voi hills and Mtito Andei.
is a gradual decrease eastwards.
Then there
Vegetation cover and composition also respond to this
rainfall pattern. The tree cover is Acacia/Commiphora bushland,
and grasses around Mtito Andei are dominated by C. roxburghiana
and C. ciliaris . Around Voi are C. ciliaris and Eragrostis
superba while further south around Mackinnon Road are E.
macrostachyus, P. maximum, E. superba.
Moving eastwards towards the Galana river composition
changes and Chrysopogon plumulosus and a few Tetrapogon
bidentatus are present. Further still, A. mutabilis becomes
dominant during the dry season.
REFERENCES
Cassady, J.T. 1973. The effect of rainfall, soil moisture and
harvesting intensity on grass production on two rangeland
sites in Kenya. E. African Agricultural and Forestry
Journal 39(1).
East African Meteorological Department. 1974. Summary of
rainfall in Kenya for the year 1972. Nairobi, Kenya.
Heady, H.F.
Kenya .
1960. Range management in East Africa. Nairobi,
Herlocker, D. 1979. Vegetation of southern Marsabit District,
Kenya. IPAL Technical Report No.D-1. UNEP-MAB Integrated
Project in Arid Lands. UNEP-MAB, Nairobi.
Muchoki, C.H.K. 1982. Baseline data for vegetation stations in
Taita-Taveta and Kitui Districts. KREMU Technical Report
No. 53, Nairobi.
National Atlas of Kenya 1970. Third edition.
Kenya , Na i rob i .
Government of
Olang, M.O. 1984a. Vegetation cover assessment in Turkana
District, Kenya. Proceedings of workshop on Land Evaluation
for Extensive Grazing (LEGG) . Nairobi, Kenya. pp. 183-193.
Olang, M.O. 1984b. Endangered grassland of the Maasai Mara and
Loita plains. Endangered Resources for Development.
Proceedings of a workshop on the Status and Options for
Management of Plant Communities in Kenya. pp. 227-232.
Pratt, D.J. and Gwynne, M.D. 1977. Rangeland management
ecology in East Africa. Hodder & Stoughton, London.
and
Skovlin, J.M. 1980. An evaluation of 10-year rangeland changes
at selected sites throughout Kenya. A report submitted to
the Head of Range Management, Ministry of Agriculture and
Livestock Development, Kenya.
Stoddard, L.A. , Smith, A.D. and Box, T.W. 1955. Range management.
Third edition. McGraw-Hill Book Company. New York, USA.
296
Synott, T.J. 1979. A report on the status, importance and
protection of the montane forest. IPAL Technical report No.
D-2a. UNEP-MAB Integrated Project in Arid Lands. UNEP-MAB,
Nairobi.
297
IDENTIFICATION AND RANKING OF PRODUCTION CONSTRAINTS IN
AGROPASTORAL SYSTEMS OF MACHAKOS DISTRICT IN KENYA
Adrian Wekulo Mukhebi
Winrock International
Small Ruminant Collaborative Research Support (SR-CRSP) Kenya
P.O. Box 252, Maseno, Kenya
Antony Mwai Gitunu
Kiboko National Range Research Station
P.O. Box 12, Makindu, Kenya
Abstract
The agropastoral systems of southern Machakos District of Kenya
are in a semi-arid area with an annual rainfall of 500 mm to 700
mm falling in two seasons. A mean household consists of 15
persons, and owns 20 ha of land, 36 head of cattle, 13 sheep and
45 goats. A variety of crops are grown during the long and short
rainy seasons, mainly in mixed stands of maize + pigeon peas,
maize + beans, maize + cow peas, and maize + green grams.
Farmers perceive the following as the constraints which
limit, in order of their severity, increased production of crops
and livestock in their systems: drought, capital scarcity,
livestock diseases, crop diseases and pests, poor infrastructure,
inadequate extension service, lack of improved inputs, wildlife,
land scarcity, labour scarcity, marketing, land tenure, bush
encroachment, soil erosion, theft, large family and witchcraft.
The ranking of these constraints varies slightly by farm size and
survey site. Most farmers view the government as the source of
solutions to most constraints. Admittedly many of the
constraints would require government action such as improvements
in infrastructure and marketing, land tenure, research and
extension as pre-requisites for further development.
Researchable constraints of immediate interest to the Kiboko
National Range Research Station in order of their importance to
farmers are: drought, livestock diseases, crop diseases and
pests, and improved crop varieties and livestock breeds.
INTRODUCTION
This paper reports some of the results of a survey conducted by
the Socioeconomics Division (SEA) of the Kiboko National Range
Research Station (NRRS) to identify and rank production
constraints as perceived by the agropastoral ists of southern
Machakos District of Kenya (Mukhebi et al . , 1985).
The Kiboko NRRS has a mandate for conducting research
directed at developing technologies for increasing rangeland
productivity in Kenya. To make such technologies relevant,
production constraints that reflect the felt needs and
aspirations of rangeland inhabitants are a pre-requisite.
One of the reasons for the dismal performance of many
development projects and programmes is that they are often
298
designed to address problems of target populations as perceived
by government officials and other outsiders rather than as
perceived by the people themselves. The danger with this
approach is that in many cases, considerable amount of resources
are expended on trivial problems while priority problems of
people are left unattended. The result is often that as soon as
external funding and assistance are withdrawn, the initiated
development activities which are supposed to be continued by the
people are not carried beyond the life of the target group. This
happens because sometimes a technical problem is implemented to
provide a technical solution, when, in fact, that technical
problem ranks low in the minds of the target population. The
people may be having a different priority problem on their minds
that requires a technical or non-technical solution as a
necessary and sufficient pre-condition for any other development
activity or project.
METHODOLOGY
The study was conducted in the southern part of Machakos District
stretching from Salama through Makindu to Mtito Andei townships
(Figure 1) . The area is semi-arid with an annual rainfall of
about 500 to 700 mm, falling in two seasons: long rains (March to
May) and short rains (September to October) . It is inhabited by
the Akamba agropastoralists. On average, they own about 20 (+41)
ha of land, keep 36 (±56) head of cattle, 13 (+16) sheep and 45
(+48) goats per household of 15 (+13) persons. They grow a
variety of food crops during the long and short rainy seasons,
mainly in mixed stands of maize + pigeon peas, maize + beans,
maize + cowpeas, and maize + green grams.
Three sites were selected for sampling: Kasikeu location
near Salama, Makindu location in the middle of the target area
and Mtito Andei location at the southern end of the target area.
These areas are representative of the mixed farming systems of
southern Machakos: relatively intensive systems of Kasikeu, less
intensive systems of Makindu and rather extensive systems of
Mtito Andei.
On the basis of the above stratification by location, lists
of farmers were obtained from area assistant chiefs and
agricultural extension officers. (There was no official Land
Registry list of farmers as land registration has not been
undertaken in the area) . A random sample of a pre-determined
number of farms was selected from each site. In total, 106 farms
were selected: 32 in Kasikeu, 43 in Makindu and 31 in Mtito
Andei.
A survey team from SED with the assistance of the extension
staff interviewed sampled farmers through single-day visits. The
interviews were conducted during May 1985 using a pre-tested
questionnaire. Among other variables, respondents were asked an
open-ended question: What problems do you consider to be limiting
increased production of crops and livestock on your farm? After
enumerating all the problems, the respondent was asked how he/she
felt they could be solved, one by one.
299
Figure 1. Surveyed areas in Kajiado and Southern
Districts in Kenya.
Machakos
 
Indian
Ocean
RESULTS
Responses regarding felt problems are summarised in Table 1 by
farm size and Table 2 by survey site. The problems or
constraints to increased crop and livestock production are ranked
on the basis of their response frequency.
300
For all the farms, 17 constraints are identified and ranked
in the following order of their severity starting with the most
severe: Drought (expressed severally as lack of rainfall,
insufficient rainfall, lack of water, crop failure, lack of
forage during dry season, etc.); capital scarcity (lack/ shortage
of cash, loans); livestock diseases; crop diseases and pests;
poor infrastructure (lack of or poor roads, hospitals,
transport) ; inadequate extension service (lack of or insufficient
skills) ; lack of improved inputs (lack or unavailability of grade
livestock, hybrid seeds, fertilizers, machinery, low yields) ;
Table 1. Listing and ranking of production constraints by
farm size. South Machakos District, Kenya, 1985.
Farm size All farm
:es
Small (a) Medium (b) Large (c)
si:
(n •= 106)
(n - 44) (n - 28) (n = 34)
Constraint % Rank % Rank % Rank % Rank
Land 23 5 14 9 15 11 18 9
Labour 23 5 7 12 12 12 15 10
Capital 84 2 86 2 74 3 81 2
Extension services 16 8 25 7 32 6 24 6
Improved inputs 18 7 29 6 18 8 21 7
Drought 87 1 93 1 88 2 90 1
Infrastructure 14 9 39 5 32 6 26 5
Soil erosion 5 13 7 12 3 14 5 14
Crop diseases/pests 50 4 46 4 62 4 53 4
Livestock diseases 64 3 57 3 100 1 74 3
Land tenure 9 10 11 10 9 13 9 12
Marketing 9 10 11 10 18 8 10 11
Bush 5 13 4 14 18 8 8 13
Wildlife 9 10 21 8 35 5 21 7
Theft 2 15 4 14 0 - 2 15
Large family 0 - 0 - 3 14 1 16
Witchcraft 0 " 4 14 0 " 1 16
a Small farm size
b Medium farm size
c Large farm size
0 - 9.9 ha.
10 - 19.9 ha.
20 + ha.
wildlife (crop damage by wild animals, live
wild animals) ; land scarcity (land shortage,
area; overstocking); labour scarcity (labour
problem, lack of enough time to work on th
(lack of or poor markets, low output prices,
delayed payment for marketed crops) ; land
land, lack of title deeds, trespassing) ; bush
erosion; theft; large family; and witchcraft.
stock predation by
shortage of grazing
shortage, weeding
e farm) ; marketing
price fluctuations,
tenure (unsurveyed
encroachment; soil
Farmers' suggested solutions to the above constraints are
summarised in Appendices 1-15. The percentage of respondents (n -
106) suggesting solutions to a given constraint is greater than
100 when some respondents suggested more than one solution, and
less than 100 when less than 106 respondents suggested any
solution. The majority of the respondents view government as the
major source of solutions to most constraints.
301
Table 2. Listing and ranking of production constraints by
surveyed area. South Machakos District, Kenya, 1985.
Surveyed area All
si;
(n
irveyed
Kasikeu Makindu Mtito Andei = 106)
(n = 32) (n = 43) (n = 31)
Constraint % Rank % Rank % Rank % Rank
Land 22 6 21 7 10 12 18 9
Labour 28 5 7 12 13 11 15 10
Capital 91 1 84 3 68 3 81 2
Extension services 9 10 30 5 29 7 24 6
Improved inputs 19 7 14 9 32 6 21 7
Drought 88 2 93 1 87 1 90 1
Infrastructure 19 7 30 5 29 7 26 5
Soil erosion 6 12 5 13 3 14 5 14
Crop diseases/pests 56 3 51 4 52 4 53 4
Livestock diseases 53 4 88 2 74 2 74 3
Land tenure 9 10 5 13 16 9 9 12
Marketing 16 9 2 15 16 9 10 11
Bush 3 14 12 10 10 12 8 13
Wildlife 6 12 21 7 35 5 21 7
Theft 3 15 2 15 0 - 2 15
Large family 3 14 0 - 0 - 1 16
Witchcraft 0 - 0 - 3 14 1 16
DISCUSSION
Among the five severest constraints for all farms, drought is
ranked first as expected in this semi-arid area. Capital
scarcity is rated second. Many respondents expressed lack of
cash for purchasing or hiring farm inputs such as seed or oxen
for ploughing. This in turn is reportedly due to low volumes and
prices of marketable product surpluses and inaccessibility to
commercial credit. Livestock diseases is ranked as the third
severest constraint. Respondents attributed this problem largely
to inadequate veterinary services, lack of drugs and insufficient
livestock dips. Tick-borne diseases such as East Coast Fever and
heartwater are reportedly the most prevalent. Crop diseases and
pests are the fourth most important constraint in these
agropastoral systems. Frequent pests include birds and army
worms, the latter coming especially at the on-set of rains
following prolonged dry spells. The fifth constraint is poor
infrastructure, i.e. lack of roads in many areas, poor roads that
are impassable during wet seasons, lack of public transport
vehicles, lack of sufficient hospitals and health clinics.
Constraints are identified and ranked by farm size and
survey site to determine any differences between farm size and
spatial distribution. There is no difference in the list of
identified constraints by farm size and site, but there is a
slight variation in ranking. While drought is ranked number one
constraint in small (0-9.9 ha) and medium (10-19.9 ha) size
farms, it is ranked second in the large (20+ ha) size farms, in
which livestock diseases are rated as the first constraint.
302
Smaller farms rely more on crops than livestock for livelihood
and vice versa for larger farms. Furthermore, while the farmers
of Kasikeu perceive capital as their number one problem (due to
more intensive systems) , it is ranked number three in Makindu and
Mtito Andei locations, where drought is ranked first. Spatial
variation in type and ranking of constraints could be expected to
become more pronounced with distance. This would call for more
caution generalising constraints over large geographical areas.
One striking observation about the suggested solutions is
that most respondents view government as the source of solutions
to most of their problems. This view is not conducive to
development. It encourages people to sit and wait for the
government to provide solutions to problems that they themselves
can solve or do something about. People ought to be educated "by
the government" about the need to find or initiate solutions to
constraints facing them by themselves rather than wait for the
government to provide the answers. This could be effected
through encouragement of community self-help (harambee) efforts
and through cooperative societies as a few of the respondents
suggest for solutions. Admittedly, many of the constraints
identified would require government action such as improvements
in infrastructure and marketing, land tenure, research and
extension as pre-conditions or pre-requisites for any significant
development initiated by the people themselves.
Many of the constraints identified are interrelated.
Solutions to some may alleviate others. For instance capital is
ranked the second constraint considering all farms together. But
the capital constraint is related to the land tenure, large
family and marketing constraints. For instance, a freehold land
tenure system with title deeds would enable farmers to have
access to commercial credit by offering the title deeds as
security for loans; smaller family size would lessen household
consumption expenditure and save cash for purchase of farm
inputs; and improved marketing would improve producer prices or
increase volume of marketed farm produce, resulting in higher
cash in-flow. Each one of these actions would have the tendency
of relaxing the household capital constraint.
From the NRRS point of interest, researchable constraints,
in order of their priority to the farmers are: drought, livestock
diseases, crop diseases and pests, and improved crop varieties
and livestock breeds that would be more productive than the
current ones. However, the solutions suggested by farmers are
important and need to be explored thoroughly before research is
designed and undertaken to develop or test technologies that may
not contribute significantly to the solution of the constraint in
the eyes of the farmers. In fact, researchers would need to
gather more detailed field information about these constraints
before launching any research, for instance, the magnitude and
nature of each constraint. Such information would help in
focusing research and developing technologies on more specific
aspects of the constraints.
It should be borne in mind that ultimately, whatever
solutions (technologies) are devised to address the four and
other "researchable" constraints above, farmers will need capital
to adopt them. Since capital is ranked the second most severe
303
constraint, it would need to be addressed before farmers would be
expected to adopt the new technologies. Furthermore, since the
listed constraints are interrelated as already mentioned and
their solutions are geared towards the same goal of increasing
agricultural production, an interdisciplinary approach in the
context of rangeland systems research (RSR) would be the
recommended strategy to adopt in addressing the constraints.
CONCLUSIONS
The agropastoralists of southern Machakos District perceive the
following as the constraints limiting increased crop and
livestock production in their systems, listed from the most to
least severe: drought, capital, livestock diseases, crop diseases
and pests, infrastructure, extension services, improved inputs,
wildlife, land scarcity, labour scarcity, marketing, land tenure,
bush encroachment, soil erosion, theft, large family and
witchcraft.
The ranking of these constraints varies slightly among
different farm size groups and survey sites. The majority of
farmers view the government as the major source of solutions to
most of the constraints. Admittedly in infrastructure and
marketing, land tenure, research and extension services as pre
requisites for any significant development initiated
endogenously.
Researchable constraints of immediate interest to the Kiboko
National Range Research Station, in order of their importance to
farmers are drought, livestock diseases, crop diseases and pests,
and improved crop varieties and livestock breeds. More detailed
data about these constraints would have to be gathered by
researchers in order to focus research on, and develop
technologies for more specific aspects of the constraints. As
most of the constraints are interrelated and affecting the same
goal of increasing agricultural production, an interdisciplinary
approach in the context of rangeland systems research would be an
appropriate strategy to follow in addressing the constraints.
REFERENCES
Mukhebi, A.W. , Gitunu, M. , Kavoi, J. and Iroha, J. 1985.
Agropastoral systems of Southern Machakos District of Kenya.
Technical Paper No. 7, Socioeconomics Division, Kiboko
National Range Research Station, Ministry of Agriculture and
Livestock Development, Kenya.
304
APPENDIX
Appendix 1. Farmer-suggested solutions to drought constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Government to construct new dams and
repair old ones 47
Government to construct boreholes 32
Government to assist reforestation and
soil conservation 38
Government to implement irrigation project 14
Farmers to feed livestock with hay and
supplements 2
Farmers to construct water tanks to store
rain water 5
People to organise harambee water project 4
Farmers to practice proper stocking rates 3
Farmers to store food 7
Government to provide seed for drought
resistant crops 2
People to practice family planning 1
People to pray to God 8
Appendix 2. Farmer-suggested solutions to capital constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n - 106)
Government to provide loans 3 3
Government to increase producer crop prices 6
Government to increase employment
opportunities in rural areas 8
Government to issue land title deeds for
loan security 7
Government to develop and improve markets
for farm produce 2
Government to increase salaries 2
People to form cooperative societies that
can provide loans 9
305
Appendix 3. Farmer-suggested solutions to livestock disease
constraint, southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n » 106)
Farmers to maintain proper dip strength 4
Government to make drugs more available 14
Government to control stock movement 2
Government to provide more veterinary
officers 29
Farmers to dip and spray livestock more
frequently 24
Government to research on livestock
diseases and drugs 9
Government to construct more dips 32
Government to lower drug prices 2
Government to control tsetse flies 3
Extension officers to visit and advise
farmers more frequently 12
Farmers to improve livestock management 2
Appendix 4. Farmer-suggested solutions to crop diseases and pests
constraints, southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Government to make pesticides more available 19
Government to lower prices of pesticides 14
Farmers to increase use of pesticides 37
Government to research on crop diseases and
pests 18
Government to provide more extension officers 18
Extension officers to visit and advise farmers
more frequently 19
Appendix 5. Farmer-suggested solutions to infrastructure
constraint, southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n - 106)
Government to construct more health centres,
schools, markets, roads and bridges 25
Government to provide loans for purchasing
transport vehicles 15
People to organize harambees for
constructing health centres and schools 16
People to buy more public transport vehicles 8
Government to supply more drugs in hospitals 17
Government to educate people more on
development 4
Government to provide more doctors 10
306
Appendix 6. Farmer-suggested solutions to extension services
constraint, Southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Government to provide more extension
officers 23
Government to provide more transport to
extension officers 6
Government to build more farmers'
training centres 5
Extension officers to visit and advise
farmers more frequently 19
Government to station extension officers
near to farmers 10
Government to organise seminars for
farmers 5
Appendix 7. Farmer-suggested solutions to improved inputs
constraint, southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Government to provide more grade
livestock and hybrid crop seeds 20
Government to test soils for fertility 3
Government to provide and distribute
fertilizers more timely 18
Government to provide loans for purchase
of grade animals 9
Government to make agricultural chemicals
more available 7
Government to provide tractors and loans
for their purchase 14
Appendix 8. Farmer-suggested solutions to wildlife constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n - 106)
Government to chase away wild animals 18
Government to kill wild animals 12
Government to compensate farmers for
wild animal damage 19
Government to confine all wild animals
in game parks and reserves 16
Government to provide loans to farmers for
fencing their farms against wild animals 11
Game personnel to patrol more frequently 13
307
Appendix 9.
Solution
Farmer-suggested solutions to land constraint,
southern Machakos District, Kenya, 1985.
% of farmers
suggesting (n - 106)
Farmers to purchase more land 14
Farmers to preserve livestock forage 8
Government to provide land for settlement 17
Government to provide loans for land purchase 2
People to practice family planning 3
Government to survey and provide individual
land title deeds 7
Farmers to intensify production 4
Government to control ants that destroy
livestock forage 2
Appendix 10. Farmer-suggested solutions to labour constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n - 106)
Government to provide loans for hiring labour
Government to provide herbicides to minimise
hand weeding
Government to provide loans for purchasing
farm machinery
Government to provide loans for hiring oxen
13
4
Appendix 11. Farmer-suggested solutions to marketing constraints,
southern Machakos District, 1985.
Solution % of farmers
suggesting (n = 106)
Government to improve marketing systems
for farm produce
Government to pay cash for delivered farm
produce
Government to increase prices of farm
produce
Government to reduce prices of farm inputs
Government to construct auction rings for
livestock
Farmers to form marketing cooperatives
Government to supply weighing scales for
cattle
308
Appendix 12. Farmer-suggested solutions to land tenure constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Government to survey land and issue title
deeds 9
Farmers to fence their farms 1
Government to provide loans for fencing 2
Appendix 13. Farmer-suggested solutions to bush constraint,
southern Machakos District, Kenya, 1985.
Solution % of farmers
suggesting (n = 106)
Farmers to clear bush by hand 4
Government to provide chemicals for
clearing bush 1
Government to provide loans for
clearing bush 3
Appendix 14. Farmer-suggested solutions to Soil erosion
constraint in southern Machakos District, Kenya
1985.
Solution % of farmers suggesting
(n = 106)
Government to instruct and assist farmers to
construct terraces and benches 5
Government to carry out reafforestation
and reseeding 3
Appendix 15. Farmer-suggested solutions to soil erosion
constraint, southern Machakos District, Kenya, 1985.
Constraint Solution % of farmers
suggesting (n = 106)
Theft Government to increase
security 2
Government to increase
employment opportunities 1
Large family People to practice family
planning 1
Witchcraft Churches to teach religion 1
Government to arrest witchdoctors 1
309
RELATIONSHIP BETWEEN PHYTOMASS PRODUCTIVITY, ANIMAL FORAGE
PREFERENCE AND ENVIRONMENTAL FACTORS ON SEMI-ARID GRAZED AND
UNGRAZED KEY RANGE PLANTS IN TANZANIA
M.M.S. Lugenja
Tanzania Livestock Research Organization
Livestock Research Centre, P.O. Box 5016, Tanga, Tanzania
R.N. Mero
Tanzania Livestock Research Organization
Livestock Research Centre, P.O. Box 5, Kongwa, Tanzania
A.R. Kajuni
P.O. Box 446, Tukuyu, Tanzania
Abstract
The relationship between phytomass productivity, animal forage
preference and environmental factors on semi-arid grazed and
ungrazed range plants in Tanzania was investigated from 1979 to
1982, on three range sites at: Pasture Research Centre, Kongwa;
Livestock Multiplication Unit, Mabuki, and Livestock Research
Centre, West Kilimanjaro. Phytomass production was obtained from
nine random quadrats harvested at middle and end of rains.
Height-weight relationship from grazed and ungrazed plants
coupled with timed observation on grazing animals, and CP content
was used to determine animal preference. Phytomass productivity
from grazed paddocks ranged from 2.20 to 4.74 tonnes DM ha as
compared to 2.04 to 3.95 tonnes on ungrazed ones. The combined
yield of grasses and browse plants increased phytomass
productivity to about 15 tonnes per hectare. There was
significant difference (P>0.05) in the utilisation of the key
range plants at Kongwa, West Kilimanjaro and Mabuki. Cattle
preferred grazing on perennial grasses while goats preferred
trees and shrubs. Regression of animal forage preference on
phytomass productivity, crude protein and environmental factors,
and phytomass productivity on environmental factors tor Kongwa
and West Kilimanjaro were nonsignificant (P>0.05) . To maximise
range resources for sustained animal productivity mixed stocking
is suggested.
INTRODUCTION
Tanzania has an area under rangeland estimated to be about 60 m
ha which supports about 20 million stock units (Anon, 1983) .
This level of animal production has, however, been below the
national nutritional demand (Akilimali, 1983; Chagula, 1983).
Several factors are responsible for the low animal
production levels, but the most important of all is shortage of
adequate quantity and quality feed in the dry season. This is
manifested in overgrazing which is a common feature in all
livestock farming systems.
High levels of animal production can, however, be attained
from these rangelands provided good range management principles
are applied. Practices such as grazing at optimum stocking rate,
310
use of proper grazing management systems and the use of the right
kind of animals restore balance between forage supply and animal
nutritional requirements.
Grazing animals are known to prefer certain grasses and
legumes at some stages of growth, and such behaviour was observed
in palatability studies (Johnstone-Wallace and Kennedy, 1944) .
As a result of selective grazing chemical and botanical
composition, herbage consumed by grazing animals differs from the
average of the composition of the sward in the field (Johnstone-
Wallace and Kennedy, 1944) . The amount of green herbage on offer
influenced the grazing animal's intake (Johnstone-Wallace and
Kennedy, 1944) . Similar findings were reported by Chacon and
Stobbs (197 6) , and that intake and total yield of herbage on
offer was high and positively correlated.
Cattle also prefer certain plant species at certain stages
of growth and various parts of an individual plant (Hafez and
Bouissou, 1975) . It was further reported that the taste of
ingested materials probably provided ultimate basis upon which
the animals formed a decision as to whether the material was
highly preferred, merely tolerated or completely rejected. On
the other hand, the grazing habits of goats was described by
Knight (1964) , who showed that grass formed the major food
preference for East African goats. However, these findings were
not in agreement with those obtained by Staples et al., (1942),
Jordan (1957), Wilson (1957), Edward (1948), and Lugenja and
Kajuni (1979) , who concluded that goat diet comprised mainly
trees and shrubs.
Although Hancock (1954) found no relationship between
temperature and grazing behaviours, Stricklin et al. (1976)
reported that occurrence of thunderstorms forced animals to stop
grazing. Similar observations were reported by Arave and
Albright (1981). However, Smith (1959) did not observe any
increase in the proportion of the night grazing of individual
cattle under high daylight temperature.
In this study attempts have been made to establish the
relationship between phytomass productivity (PP) , animal forage
preference (AFP) and environmental factors (EF) on semi-arid
grazed and ungrazed plants in Tanzania.
MATERIALS AND METHODS
A total of 15.4 ha in 14 paddocks, each 105 m x 105 m at Kongwa
Pasture Research Centre (PRC) , the West Kilimanjaro Livestock
Research Centre (LRC) and the Mabuki Livestock Multiplication
Unit (LMU) were used for the study. The mean annual rainfall and
temperatures were: 250 to 500 mm and 27° to 33°C (PRC, Kongwa),
300 to 400 mm and 20° to 30°C (LRC, West Kilimanjaro) and 700 to
800 mm and 29° to 32°C (LMU, Mabuki) . The soils are derived from
granite and gneisses at PRC, Kongwa (Owen, 1964) ; neugene
alkaline volcanic materials at LRC, West Kilimanjaro (Anderson
and Naveh, 1965) and alluvial lacustrine origin at LMU, Mabuki
(Hathout, 1983). In general, all locations are dominated by
Commiphora and Acacia trees, associated with Aristida, Pennisetum
and Cynodon grass species.
311
Primary productivity (PP) of grasses in tonnes dry matter
(DM) was obtained at the middle and end of rains from 1 m2
quadrats, randomly located in the paddocks. However, PP of trees
and shrubs was estimated from twigs of current year's growth of
key browse plant within 1.67 metres height. All twigs in
predetermined direction were harvested. The bulked-up samples
were analysed for crude protein, (CP) calcium (Ca) and phosphorus
(P) . The grass data were pooled for one season and analysed as
split plot by way of ANOVA (Steel and Torrie, 1980) . Duncan's
multiple range test (DMRT) was used to separate means, and chi-
square of independence test was used for chemical analyses data
(Sokal and Rohlf , 1969) . Tree and shrub yield data were
inadequate for statistical analyses.
AFP was determined by recording the time spent by animals on
each plant species. Five to ten cows (zebu cattle) and 30-40
goats (blended castrates and females) were used at Kongwa and
West Kilimanjaro locations. The species were observed on
different dates between 0800-1200 hours at each location and site
during the middle and end of rains. Cattle alone were used at
LHU, Mabuki. The animals were observed for an hour at all
locations, and different species of animals were observed on
different days. The percent time spent by cattle and goats
grazing on different plant species were summarised under four
major plant classifications, namely: annual grasses (Urochloa
trichopus, Aristida adscensionis, Chloris pycnothrix, Chloris
virgata, Setarla verticillata, Dactyloctenium aegyptium,
Digitaria velutina, Brachiaria deflexa) ; perennial grasses
(Bothriochloa insculpta, Chloris gayana, Cenchrus ciliaris,
Cynodon dactylon, Pennisetum spp. , Sporobolus spp. , Hyparrhenia
spp. , Themeda triandra) ; herbaceous plants (Astriporaoea
hyosciamoides , Monechma blepharis spp., Dalbegia spp., Rynchosia
sennaarensis, Sida spp., Tephrosia spp.); and trees and shrubs
(Acacia mizera, A. tortilis, Blepharispermum zangubaricum,
Bauhinia spp., Cadaba spp., Combretum spp., Euphorbia cuneata.
Indigofera spp.).
Forage utilisation of key range plants (grasses) at Kongwa
and West Kilimanjaro was determined from locally compiled height-
weight tables as described by Lammasson and Jamson (1942).
Phosphorus, calcium and crude protein were determined using
the Molybdenum blue Fiske and Subbarow method, the Oxalate method
and the Kjeldahl method respectively. This was in turn used to
assess AFP.
Environmental factors (percent relative humidity - % RH,
maximum temperatures - Tmax. , evaporation - £ and rainfall - R)
were collected twice daily at 09.00 and 15.00. Percent relative
humidity was calculated from standard meteorological tables
(Goodchild, 1983) .
Regression of AFP (pooled for cattle and goats, grazing
animal diet) on PP, AFP on CP, EF on AFP and PP on EF, for Kongwa
and West Kilimanjaro locations, were used to establish the
relationship between different factors (Sokal and Rohlf, 1969) .
Data from LMU, Mabuki were too scanty to be included in the
regression analyses.
312
RESULTS
Phytomass Productivity (PP)
Ungrazed paddocks at Kongwa produced 14% more phytomass in tonnes
DM ha (P<0.05) than that which was produced on grazed ones
(Table la) . On the other hand, grazed paddocks at West
Kilimanjaro and Mabuki produced 12% more phytomass in tonnes DM
ha (P>0.05) than that which was produced on ungrazed paddocks,
respectively (Tables lb and lc) . Sites 2 and 3 at Kongwa
produced significantly more phytomass in tonnes DM ha"1 (P<0.05)
than that produced on site 1. There was no significant
difference (P>0.05) in the phytomass in tonnes DM ha produced
on the three sites at West Kilimanjaro. However, site 3 produced
60% and 99% more phytomass in tonnes DM ha than that which was
produced on sites 1 and 2, respectively (Table lb).
Table la. Mean phytomass productivity in tonnes DM ha-1
at Kongwa.
Paddocks
Sites
1 2 3 Mean
Grazed 2.61 3.57 2.97 3.05a*
Ungrazed 2.74 3.95w3.76b
3.75
3.48b
+ 0.10
Mean 2.68a 3.36b
SE + 0.57
SE - Standard error.
* Means with different superscripts differ significantly
at 5% by DMRT.
Table lb. Mean phytomass productivity in tones DM ha'1 at West
Kilimanjaro.
Sites
Paddocks 12 3 Mean
Grazed 2.75 0.89 4.74 2.79a
Ungrazed 1.91 2.89 2.72 2.49a + 1.20
Mean SE + 1.01
SE - Standard error.
* Means with different superscripts differ significantly
at 5% by DMRT.
Tree and shrub data from Kongwa location (Table 1d) ,
indicated that high (15.04 tonnes) DM ha-1 was produced from site
1. Site 3 was the least productive, with only 0.7 6 tonnes.
Acacia mizera produced the highest mean DM (8.52 tonnes) at site
2. The least productive shrub was Cadaba kirkii , with mean DM
yield of 0.24 tonnes.
313
Table lc. Mean phytomass productivity in tonnes DM ha"1 at
Mabuki .
Sites
Paddocks
Grazed 2.59
Ungrazed 2.04
Mean 2.32 NS
SE ± 0.30
SD - Standard deviation.
Means with different superscripts differ
significantly at 5% by DMRT.
Table Id. Mean phytomass productivity of trees and shrubs in
tonnes DM ha-1 at Kongwa.
Sites
Species
Euphorbia cuneata 1.58 1 70
Vittex spp. 1.63 -
Grewia spp. 0.60 -
Dirichletia pubescens 2.51 -
Blepharispermum zangubaricum 8.40 -
Acacia tortilis - 2 56
Acacia mizera - 8 52
Cadaba kirkii 0.32 0 24 0.76
Animal Forage Preference (AFB)
Cattle at Kongwa spent 26% more time grazing on perennial grass
species than on annual grasses during mid-grains (Table 2a) .
Least time (14%) was spent on herbaceous plants, and trees and
shrubs. A similar trend was observed at the end of the rains
(Table 2b) . Cattle at West Kilimanjaro spent 66% more time
grazing on perennial grasses than they did on annual grasses and
herbaceous species (Tables 2c and 2d) . However, observations
made at Mabuki (Table 2e) indicated that cattle spent 16% more
time grazing on perennial grasses than they did on herbaceous
plants and annual grasses.
314
Table 2a. Animal preference (as percent time spent on annual and
perennial grasses, and herbaceous plants and trees and
shrubs) at Kongwa in January to March 1980, 1981 and 1982.
Sites
Plant 1 2 3
classif icaition
Cattle Goats Cattle Goats Cattle Goats
Annual gra sses 49 18 21 2 46 2
Perennial grasses 23 21 75 18 46 6
Herbaceousl plants 4 21 0 28 4 18
Trees and shrubs 10 30 0 40 0 70
Table 2b. Animal preference (as percent time spent on annual and
perennial grasses and herbaceous plants and trees and
shrubs) at Kongwa in June to July 1980, 1981 and 1982.
Sites
Plant 1 2 3
classif iceition
Cattle Goats Cattle Goats Cattle Goats
Annual grasses 31 8 33 0 29 9
Perennial grasses 36 15 55 29 46 30
Herbaceous plants 17 34 6 23 7 8
Trees and shrubs 5 15 5 48 14 57
Although cattle spent 17% more time grazing on perennial
grasses in 1981 than they did in 1982 (Table 2e) , there was no
significant difference (P>0.05) in the way cattle grazed in 1981
and 1982. On the other hand, goats spent 50% of the time
browsing on trees and shrubs at Kongwa during mid-rains, and 50%
of the time at West Kilimanjaro at the end of the dry season.
Least time was spent on grasses as shown in Tables 2a and 2b.
Similar trends were observed at the end of the rains (Table 2b) .
315
Table 2c. Animal preference (as percent time spent on annual and
perennial grasses, and herbaceous plants and trees and
shrubs) at West Kilimanjaro in January to March 1980
and 1981.
Sites
Plant
classification
1 2 3
Cattle Goats Cattle Goats Cattle Goats
Annual grasses
Perennial grasses
Herbaceous plants
Trees and shrubs
1 1 35 4
26
8 0
99 88 60
3
0
86
0
42
0 1 0
40
2
0 3 1 39
Table 2d. Animal preferences (as percent time spent on annual and
perennial grasses, herbaceous plants and trees and
shrubs) at West Kilimanjaro in October 1982.
Sites
Plant
classification
1 2 3
Cattle Goats Cattle Goats Cattle Goats
Annual grasses
Perennial grasses
Herbaceous plants
Trees and shrubs
21 4
18
40 4
18
17 8
71
0
55 70 18
0
53
0 0
51
0 0
1 5 12 44
Table 2e. Cattle preference (as percent time spent on annual
and perennial grasses, herbaceous plants, and trees
and shrubs) at Mabuki in July to September.
Plant
classification
Year
Mean
1981 1982
Annual grasses
Perennial grasses
Herbaceous plants
4 14 9
57
35
40
11
49
23
Species utilisation: At the West Kilimanjaro location, plant
species were 2.27 and 5.67 more utilised than Kongwa and Mabuki
locations, respectively (Table 3) . However, there was no
significant difference (P>0.05) in the way species in the three
locations were utilised. Urochloa trichopus had the highest
(38.3%) utilisation of all the key species at the three
locations. Pennisetum spp. and Chloris virgata followed closely.
Bothriochloa insculpta was the least (28.05%) utilised species.
316
Table 3. Mean percent utilisation of key range plants (grasses)
for the years 1980-1982.
Locations
Species Kongwa West Kilimanjaro Mabuki
Aristida adscensions 28.55 31.40 -
Bothriochloa insculpta 28.05 35.50 -
Cenchrus ciliaris 32.91 30.20 -
Cynodon dactylon 30.63 28.20 -
Chloris virgata 36.25 - -
Urochloa trichopus - 38.40 25.75
Pennisetum spp. — 37.60 30.00
Chemical composition: Analysis of herbage materials, (Table 4a)
shows that site 3 at Kongwa had the highest (14.72%) mean percent
CP compared to that which was obtained at site 1 (10.75%) and
site 2 (9.2%) .
Table 4a. Chemical composition (percent, on DM basis) of forage
plants in grazed paddocks at Kongwa.
Sites
1981 1982 Mean
CP Ca P CP Ca P CP Ca P
1
2
3
9.28
13.40
16.65
0.49
1.42
1.27
0.-10
-l3
-20
12.22 0.78
0.34
0.55
0.15
0.08
0.12
10.75 0.64
0.88
0.91
0.13
0.11
0.16
0.
0.
5.13 9.27
12.78 14.72
The highest (11.65%) mean percent CP, at West Kilimanjaro
(Table 4b) was obtained from site 1. This was closely followed
by site 2 (9.97%) and site 3 (9.38%). There was no pattern in
the way Ca and P occurred in the three sites at the three
locations. Chi-square of independence indicated that year of
grazing had no influence (P>0.05) on the chemical composition.
However, sites influenced chemical composition at the two
locations (P<0.05).
Table 4b. Chemical composition percent of DM of pasture plants
on grazed paddock at West Kilimanjaro.
Sites
1981 1982 Mean
CP Ca P CP Ca P CP Ca P
1
2
3
17.10 0.62
0.88
1.25
0.23
0.17
0.26
6.70
6.35
0.54
0.30
0.07
0.06
11.56
9.97
9.38
0.56
0.59
1.25
0.13
0.12
0.26
13.39
9.38
317
Environmental Factors
Kongwa location experienced high (30.87°C) but less variable (CV
- 3.98%) mean temperatures when compared to that at West
Kilimanjaro (26.21°C) and Mabuki (30.15°C). However, it was
23.23% and 27.09% more humid at West Kilimanjaro than it was at
Kongwa and Mabuki locations respectively. The amount of moisture
lost through evaporation ranged from 3.0 to 25.0 mls. West
Kilimanjaro received 67% and 16% more precipitation, than that
received at Kongwa and Mabuki respectively (Table 5) .
Table 5. A summary of environmental factors.
Relative
humidity
Maximum
tempera
ture
(°C)
Evapo
ration
Rainfall
(mm)
Wind
velocity
(%) (mls)
Kongwa : Range 36.00 28.50 3.00 8.60
to
138.80
-
to
66.00
to
32.60
to
6.00
Mean 32.40 30.90 5.00 67.70 -
CV 6.03 3.98 23.07 11.01 -
W/Kilimanjaro : Range 68 . 00
to
88.00
22.30 4.00 4.70
to
239.00
5.70
to
36.10
to to
24.70 44.40
Mean
CV
75.70
3.38
26.20
7.97
9.90
25.81
113.20
7.82
98.40
11.42
Mabuki : Range 28.40 29.30 - 7.50
to
196.20
-
to
76.40
to
31.90
Mean 48.50 30.15 - 78.80 -
CV 7.87 4.11 - 9.63 -
Regression Analyses
Relationship between AFP and PP, CP, EF and PP and EF are
summarised in Table 6.
The results show that T max, R and E were major factors in
determining AFP at Kongwa (P>0.05). Phytomass productivity (PP) ,
CP and percent RH were not the major factors determining AFP at
that location. On the other hand, PP, CP, T max and R were major
determinant factors of AFP (P>0.05), while percent RH and E were
less important factors at West Kilimanjaro location.
318
Table 6. Relationship between AFP and PP, AFP and CP, AFP and EF
and PP and EF at Kongwa and west Kilimanjaro locations.
Depen- Indepen
dent dent
Regression
equation
Coefficient Level
of determi- of
nation (r2) significant
AFP
AFP
AFP
AFP
AFP
AFP
PP
PP
PP
PP
PP AFP
CP AFP
%RH AFP
Tmax. AFP
R AFP
E AFP
RH PP
Tmax. PP
R PP
E PP
21.97+0.55 PP 0.07
21.32+0.20 CP 0.24
25.18-0.02 %RH 0.03
87.20+3.56 Tmax. 0.79
44.55-0.30 R 0.94
9.90+3.80 E 0.56
7.66-0.07 %RH 0.98
33.09+1.35 Tmax. 0.48
8. 31-0. 08 R 0.26
6.65-1.00 E 0.18
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
West Kilimanjaro
AFP PP AFP - 21.70+0.90 PP 0.77
AFP CP AFP - 15.90+0.58 CP 0.77
AFP %RH AFP - 15.91+0.08 %RH 0.14
AFP Tmax. AFP - 30.72-0.31 Tmax. 0.52
AFP R AFP - 20.93+0.02 R 0.98
AFP E AFP - 25.47-0.40 E 0.16
PP RH PP - 35.26-0.44 %RH 0.67
PP Tmax. PP - 12.44+1.35 Tmax. 0.26
PP R PP ■= 1.94+0.01 R 0.08
PP E PP - 14.68+1.99 E 0.66
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS
NS = non significant at P<0.05.
DISCUSSION
Phytomass Productivity (PP)
At Kongwa ungrazed paddocks produced 14% more mean PP in tonnes
DM ha than grazed paddocks, in contrast to Mabuki and West
Kilimanjaro locations (Tables la, b, and c) . The relatively low
mean PP produced on grazed paddocks at the two locations were
probably due to less grazing effect. The high phytomass produced
on ungrazed paddocks at Kongwa is comparable to that obtained by
Pearson (19651 who reported that grazed areas produced about 1.78
tonnes DM ha annual top growth when compared to protected areas
(1.99 tonnes DM ha ) . The low mean PP obtained on ungrazed
paddocks at Mabuki and West Kilimanjaro locations might be due to
low moisture infiltration rates resulting from soil surface
crusting. The high mean PP produced by A. mizera (8.52 tonnes)
and B. zangubaricum (8.40 tonnes) shown in Table 1d is probably
due to the growth habit of the shrubs. A. mizera tends to branch
about a metre from the surface of the soil and these branches
usually terminate into productive branchlets within browsing
height. B. zangubaricum tends to sprout profusely at the base,
forming large clumps, with abundant forage within browsing
height. On the other hand E. cuneata, Vittex spp. , Grewia spp.,
D. pubescens , and C. kirkii are either single-stemmed, or branch
high up, thereby contributing less to grazing.
319
Animal Forage Preference (AFP)
The high preference for annual and perennial grasses by cattle,
and herbaceous plants and trees and shrubs by goats, observed at
Kongwa, West Kilimanjaro and Mabuki (Table 2) confirm those of
Staples et al . (1942), Edward (1948), Jordan (1957), Wilson
(1957) , and Lugenja and Kajuni (1979) . In New South Wales, goats
were observed to eat Acacia aneura and Cabtis spp. , while cattle
had a high preference for Calotis spp. and Chenopodium
anidiphylum (Squires, 1982) . The highest preference for
perennial grasses exhibited by cattle (99%) on site 1 at West
Kilimanjaro (Table 2c) was probably due to high percent CP
content in the grazing (Table 4), which was mainly made up of
Cynodon and Bothriochloa spp. This trend was, however, not
observed on other sites and locations. In a study of cattle
browsing behaviour in semi-arid area of Tanganyika, Payne (1963)
observed that nutritive value and succulence determined the
species that were browsed. The highest preference for perennial
grasses exhibited by goats (88%) (Table 2c) was possibly due to
the absence of trees and shrubs with forage within browsing
height.
Utilisation
The highest mean percent utilisation (33.55%) observed at West
Kilimanjaro is probably due to common use in practice at the
location. The non-significant differences in the utilisation of
key range plants at the three locations suggest that the
experimental animals had probably low grazing action.
Environmental Factors (EF)
The high mean maximum temperature (30.9°C) observed at Kongwa
location, relative to those at West Kilimanjaro and Mabuki (Table
5) , suggest that the location experienced more adverse conditions
for plant growth than either West Kilimanjaro or Mabuki. The
high coefficient of variation (25.81%) observed at West
Kilimanjaro suggest that the location was probably influenced by
the mountain's effect. On the other hand, the high percent
relative humidity observed at West Kilimanjaro was probably due
to high precipitation coupled by low temperatures (Table 5) . The
high wind velocities (98.4 km/hour) were probably responsible for
the high moisture (99.0 mls) lost through evaporation whereas,
the high amount of precipitation received probably suggests that
the experimental period fell in more than average rainfall years.
Regression Analyses
The variation in AFP associated with differences in PP and CP at
West Kilimanjaro (r - 0.77) is probably due to decrease in leaf
availability as grazing progressed in the season. Between
January and March pastures at West Kilimanjaro and Kongwa are
usually mature and dry and although the animals graze
selectively, the choice at West Kilimanjaro is limited by the
abundancy of less palatable species of Pennisetum. These
observations are similar to those by Chacon and Stobbs (1976)
working with cattle in Australia. The authors showed that the
number of bites and intake declined when leaf decreased. They
attributed these observations to lack of desire by ruminants to
320
harvest feed, to nitrogen or mineral deficiencies, or to bulk in
the rumen thus preventing the development of an eating drive.
The variation in AFP associated with differences in T max
(r - 0.79) and rainfall R(r - 0.94) for Kongwa and T max (r2
0.52) and rainfall (r2 = 0.98) for West Kilimanjaro, is probably
due to favourable influence of high ambient temperatures and
rainfall they have on PP (Table 6) . However, these factors tend
to decrease CP and subsequently digestibility (Deinum and van
Soest, 1968) . The low AFP might also be due to rumen pH
depression associated with high ambient temperature and humidity
(Mishra et al . , 1970) .
The high, but non-significant effect of T max on AFP (Tables
5 and 6) is probably due to inherent differences in the herds
used in the study at Kongwa and West Kilimanjaro locations. The
herd at Kongwa seems to have been more adapted to temperature
conditions than that at West Kilimanjaro. There was no pattern
in the way R and E affected AFP and PP at the two locations.
However, percent relative humidity seem to have been negatively
correlated with PP at these locations. The high but non
significant negative relationship between AFP and R at Kongwa
probably suggests the detrimental effect of non-effective
precipitation on grazing animals under semi-arid conditions.
Non-effective precipitation encourages termite activity,
resulting in low standing matter (Lugenja and Kajuni, 1979).
The high but non-significant positive relationship between
AFP and E at Kongwa and West Kilimanjaro is probably due to
conducive environmental conditions experienced as a result of
evaporation. The two locations are characterised by high wind
velocities which probably bring about cooling effect to grazing
animals. On the other hand, the non-significant positive
relationship between PP and E at the two locations, is in
agreement with the established fact that DM production depends on
the amount of moisture lost through evaporation and transpiration
processes (Lane, 1976) .
CONCLUSION
The results show that the mean PP in tonnes DM ha"1 of the range
increased from 4 to 15 tonnes DM ha , when grasses and browse
plant productivity is taken into consideration. The results also
show that cattle grazed mainly on grasses while goats grazed
mainly on herbaceous plants, trees and shrubs in this trial at
Kongwa and West Kilimanjaro. The results further show that
cattle and goats grazed grasses and browse plants to some degree,
suggesting that there is a dietary overlap between the two animal
species. These observations imply that for better and uniform
range utilisation, cattle and goats should be mixed during
grazing periods. In future intensive studies on stocking rate
will be carried out for the different ecological zones in order
to achieve optimum multiple range use, and identification of the
major browse plants and establishing their roles in ruminant
nutrition will be done.
321
ACKNOWLEDGEMENTS
The authors wish to thank Messrs Raphael Kamate and Mwinyishe
Abdallah for collecting the data and Mr. S. Kaganda and Miss Rose
Kutona for compiling the data. The International Foundation for
Science (IFS) is acknowledged for the provision of funds to carry
out this experiment. Finally our thanks are due to Dr. A.M.
Macha, Director General, TALIRO, for permission to attend the
PANESA Workshop and present the paper.
REFERENCES
Akilimali, B.M. 1983. The present and potential role of beef
production in meeting nutritional requirements of the
nation. Proc. 10th Scientific Conference of Tanzania Society
of Animal Production, Morogoro, Tanzania. pp. 44-65.
Anderson, G.D. and Naveh, Z. 1965. The influence of climate and
soil on the productivity of certain promising grasses and
legumes in Northern Tanzania. Afr. Soils 10:241-251.
Anon. 1983. Livestock Policy. Ministry of Livestock
Development, Dar es Salaam. Government Printers, Tanzania.
Arave, C.W. and Albright, J.L. 1981. Cattle behaviour. J. Dairy
Sci. 64:1318-1327.
Chacon, E. and Stobbs, T.H. 1976. Influence of progressive
defoliation of a grass sward on the eating behaviour of
cattle. Aust. J. Agric. Res. 27:709-727.
Chagula, A.M. 1983. The present and potential role of beef
production in the traditional sector in meeting nutritional
requirements of the nation. Proc. 10th Scientific
Conference of Tanzania Society of Animal Production,
Morogoro, Tanzania. pp. 72-74.
Deinum, B. , van Es, A.J.H. and van Soest, P.J. 1968. Climate,
nitrogen and grass. II. The influence of light intensity,
temperature and nitrogen on in vitro digestibility of grass
and the prediction of these effects from some chemical
procedures. Neth. J. Agric. Sci. 16:217-223.
Edward, L.C. 1948. Some notes on the food of goats in a semi-
arid area. E. Afric. Agric. J. 13: 221-223.
Goodchild, A.V. 1983. A new approximation for saturation vapour
pressure in humidity calculation. Proceedings of the Sixth
meeting of the National Agrometeorological Committee, Dar es
Salaam, 23-24 November, 1983. Tanzania.
Hafez, E.S.E. and Bouissou, M.F. 1975. Behaviour of cattle. In:
E.S.E. Hafez. (ed.), The behaviour of domestic animals.
The Williams and wilkins Co., Baltimore, pp. 203-211.
Hancock, J. 1954. Studies of grazing behaviour in relation to
grassland management. 1. Variation in grazing habits of
dairy cattle. J. Agric. Sci. 44:240-432.
322
Hathout, S.A. 1983. Soil Atlas of Tanzania. Dar es Salaam.
Tanzania Publishing House, Map 7.
Johnstone-Wallace, D.B. and Kennedy, K. 1944. Grazing management
practices and their relationship to the behaviour and
grazing habits of cattle. J. Agric. Sci. 34:190-196.
Jordan, S.M. 1957. Reclamation and pasture management in the
semi-arid areas of Kitui District, Kenya. E. Afr. Agr. For.
J. 23:84-88.
Knight, J. 1964. Some observations on the feeding habits of
goats in the Baringo District of Kenya. E. Afr. Agric. For.
J. 30:182-188.
Lammasson, T. and Jamson, C. 1942. Determining utilization of
range grasses from height-weight tables. Mimeo, Department
of Agriculture, U.S.A.
Lane, I.R. 1976. A consideration of factors determining
potential stocking rate in Tanzania. Proceedings of the 3rd
Annual Scientific Conference of the Tanzania Society of
Animal Production 17-19 May, 1976, Morogoro, Tanzania.
Lugenja, M.M.S. and Kajuni, A.R. 1979. Range management in
central Tanzania: Results of grazing management systems at
Pasture Research Station, Kongwa. Proceedings of the
Tanzania Society of Animal Production, Morogoro 6, 24-51.
Mishra, M. , Martz, F.A. , Stanley, R.w. , H.D. Campbell, J.R. and
Hilderbrand, E. 1970. Effect of diet and ambient
temperature-humidity on ruminal pH, oxidation reduction
potential, ammonia and lactic acid in lactating cows. J.
Anim. Sci. 31 (2) : 1023-1028 .
Owen, M.A. 1964. A study of grassland problems on the upland
soils of the Kongwa Plain, central Tanzania, with
investigations into some pasture establishment techniques.
M.Sc. thesis, University of Wales, U.K.
Payne, W.J. A. 1963. A brief study of cattle browsing behaviour
in semi-arid area of Tanganyika. E. Afr. Agric. For. J.
29:131-133.
Pearson, L.C. 1965. Primary production in grazed and ungrazed
desert communities of eastern Idaho. Ecol. 46:278-285.
Smith, C.A. 1959. Studies on the Northern Rhodesia Hyparrhenia
Veld. I. The grazing behaviour of indigenous cattle grazed
at light and heavy stocking rates. J. Agr. Sci. 52: 369-
375.
Sokal, R. and Rohlf, F. 1969. Introduction to biostatistics.
W.H. Freeman and Company, San Francisco. Chap. 12 and 13.
Squires, V.R. 1982. Dietary overlap between sheep, cattle and
goats when grazing in common. J. Range Manage. 35:116-119.
323
Staples, R.R. , Hornby, H.E. and Hornby, Robina, M. 1942. A study
of the comparative effects of goats and cattle on a mixed
grass-bush pasture. E. Afr. Agric. J. 8:62-70.
Steel, r.g.d. and Torrle, J.H. 1980. Principles and procedures
of statistics - A blometrical approach. 2nd edition.
McGraw-Hill Book Company, New York, USA.
Stricklin, W.R. , Wilson, L.I. and Graves, H.B. 1976. Feeding
behaviour of Angus and Charolais. Angus cows during summer
and winter. J. Anim. Sci. 43:721-731.
Wilson, P.N. 1957. Studies of the browsing and reproductive
behaviour of the East African dwarf goat. E. Afr. Agric. J.
23:138-147.
324
GRAZING BEHAVIOUR OF MAASAI CATTLE
P.P. Semenye
P.O. Box 252, Maseno, Kenya
Abstract
The grazing behaviour of Maasai cattle was influenced by season,
area and grazing orbit. In a grazing day the cattle spent 1, 8,
14, 15 and 62% of their time drinking, ruminating, resting,
walking and grazing respectively. The mean length of the grazing
day of Maasai cattle is defined, as from the time they left their
bomas in the morning until they returned for the night; this
duration was 10.8 ±0.6 and 10.4 ±0.6 hours for cattle and
calves respectively. The mean grazing orbit was 15.5 and 10.8 km
for adult cattle and calves respectively. A vibracorder proved
quite accurate in timing cattle activities.
INTRODUCTION
Daily intake of grazing cattle is determined by the product of
time spent grazing, the rate of biting and forage intake per bite
(Hodgson, 1982) . It therefore follows that grazing time
influences intake directly, while other grazing habits such as
drinking, ruminating, playing and resting may influence intake
either positively or negatively. In Africa pastoral herdsmen
exercise a major influence on the grazing behaviour of their
cattle through herding. By herding, man decides when and where
cattle are to go for grazing. So the skill of a herdsman coupled
with environmental factors are the determinants of a successful
grazing day in terms of forage intake.
The endeavour to understand the grazing behaviour of the
Maasai cattle was part of a study of the Maasai pastoral
production system. This study was initiated by the International
Livestock Centre for Africa (ILCA) . The study was carried out in
three group ranches which were selected on the bases of their
ecology and development gradients in Kajiado District, Kenya.
MATERIALS AND METHODS
The five-minute method was used to study grazing behaviour, as
suggested by Hancock (1953) and used by Lampkin et al . (1958) at
Muguga, Kenya. At the end of each five-minute period, an
observation was made and the behaviour of each selected animal
recorded. Behaviour activities recorded were: species of grass
or browse grazed, resting while either standing or lying,
ruminating while standing or lying, walking and drinking. With
the exception of the species grazed the other behaviour patterns
were confirmed if they continued for ten or more seconds.
In theory, a total of 144 observations could be made from
7.05 to 19.05 hours which is the normal daylight period in Kenya
at this latitude. However, the grazing period does not always
extend over the entire daylight period, but rather is modified
according to the schedule and needs of the individual livestock
325
owner. In each herd, three animals were selected according to
the role each cow appeared to play in the herd: the herd leader,
an ordinary cow and a laggard. Each animal selected was followed
by one observer during the entire grazing day.
The observers used pedometers to monitor the distance
covered during each grazing day. At the end of each day the
distance walked was taken as the average of the three pedometers
used. Pedometers were calibrated regularly against a known
distance. A vibracorder was used as described by Stobbs (1970)
to monitor total time spent actively, in grazing, walking,
drinking and ruminating. For statistical analysis, SPSS and
least squares procedures of Harvey (1977) were used.
RESULTS
Grazing Day Length
The mean length of the grazing day of Maasai cattle, defined as
from the time they left their bomas in the morning until they
returned for the night, was 10.8 + 0.6 and 10.4 + 0.6 hours for
adult cattle and calves respectively. The difference between the
adults and the calves of 0.4 hour was not significant (P>0.05).
Although the 28 grazing days recorded by the five-minute
method and the vibracorder did not overlap entirely, the latter
also recorded an average of 10.6 + 0.8 hours. During green
seasons grazing days were longer than during dry ones; the
averages for the green and the dry were 11.1 + 0.4 and 10.3 + 0.8
hours respectively but the difference was not significant (P >
0.05). The measurements of the grazing day with an ordinary
watch and the vibracorder were very similar and the differences
between them were within the experimental error. A vibracorder
had the advantage over human observers that, once the instrument
was set with a chart covering seven days' records, it needed no
further handling until the expiry of the seven days. The
vibracorder also recorded night activities between 18.30 to 07.30
hours. The average duration of night activity for the green and
dry seasons was 2.0 + 1.3 hours and the seasonal difference was
not significant (P>0.05). The night activity was assumed to be
mainly rumination. Total time spent ruminating on the average
was 2 . 8 hours divided into 2 hours at night and 0 . 8 hours during
the day.
Distance Walked per Grazing Day
The mean distance walked per grazing orbit was 15.5 + 5.0 and
10.8 + 4.8 km for adult cattle and calves respectively. Hence
adult cattle walked 4.7 km more than calves. The difference was
36% of the overall mean and significant (P<0.05). This
difference in walking distance between the calves and the adult
cattle was due to grazing management. Calves grazed in reserved
areas near the homestead (Olopololi) , while the adults utilised
the communal grazing areas. Calves had more opportunity for high
intake since in the homestead herbage was relatively more
abundant than in the communal areas, necessitating less walking
in search of forage. For the adult cattle the distance walked
326
was dependent on the herder's skills, on the grazing orbit
followed, and whether it was a watering day or not.
specific Grazing Habits
The overall percentage time spent in a grazing day in specific
activities were 1, 8, 14, 15 and 62% in drinking, ruminating,
resting, walking and grazing respectively. The percentage time
spent was different between seasons, areas and grazing orbits and
there were significant interactions of season x area and season x
grazing orbit.
Seasonal effects were highly significant (P<0.01) while
animal type was not (P>0.05) for any of the activities. The
coefficient of variations of the activities were large, and they
ranged from 23 to 98.
Table 1 presents various activities as affected by seasonal
changes. More time was spent drinking in the green than in the
dry and the dry/green seasons. Cattle ruminated less in the
green than in the other two seasons, while rumination time during
the day was 8% longer in the dry than in the green season.
Cattle rested 13% of the day longer in the dry than in the green
or dry/green seasons, but they walked for a longer period in the
green and dry/green than dry season. Proportion of time spent on
grazing was longest in the green season (66%) , followed by
dry/green (62%) and 57% for the dry season.
Ratios between rumination and grazing time were 1:8, 1:9,
1:17 and 1:5 for the overall mean, dry/green, green and dry
respectively. Although this calculation of the rumination to
grazing time ratio was only for the day-light period when viewed
together with Table 1, a trend emerged linked to forage quantity
and intake. Grazing and ruminating times appeared to be
Table 1. Percentage of grazing day spent on different activities
by season.
Season N Drinking Ruminating Resting Walking Grazing
Overall mean 61 1 8 14 15 62
(6) (48) (84) (90) (372)
(0.1) (0.8) (1.4) (1.5) (6.1)
Dry/green lb 7a 13a 17a 62ab
Feb/Mar 1983 21 (6) (42) (78) (102) (372)
(0.1) (0.7) (1.3) (1.7) (6.2)
Green 3a 4° 8b 19a 66a
May/June 1983 28 (18) (24) (48) (114) (396)
(0,3)
lb
(0.4)
12b
(0.8) (1.9) (6.6)
Dry 21c 9 57
Sept/Oct 1983 12 (6) (72) (126) (54) (342)
(0.1) (1.2) (2.1) (0.9) (5.7)
abc Except for the overall means other means having no
superscript letter in common differ at the P<0.05.
(In brackets - time in hours) .
327
interrelated; low quality forage in the dry season resulted in
longer ruminating and shorter grazing (intake) time, while the
reverse was found in the better conditions of the green season.
It is accepted that intake is limited by forage quality. But in
this production system it appeared that rumination time was
negatively related to intake.
DISCUSSION
That the vibracorder can be used successfully to record grazing
time (Stobbs, 1970) was confirmed in this present study. It has
many obvious logistics advantages over observers with watches.
It is a tool that could be used to compare herdsmen and their
herds. The vibracorder was also effective in recording time
spent on rumination during the night. Therefore the total
ruminating time was calculated by adding the night time recorded
by the vibracorder to that of the day time recorded by observers.
The total ruminating time in 24 hours of 2.8 hours fell at the
lower end of the range of 2.5 to 10.4 hours, reported by Harker
et al (1953) and van Soest (1982).
Maasai cattle walked shorter distances than pastoral Fulani
cattle in northern Nigeria studied by van Raay and de Leeuw
(1974) who recorded daily averages ranging from 16 to 30 km.
Maasai cattle walked less because of better grazing resources due
to the bimodal rainfall and a greater degree of sedentarisation
and a closer spacing of water sources, all leading to a lesser
need for walking.
The percentage time spent grazing of 62% or 6.7 hours was
intermediate between times reported for herded cattle (65 to 89%,
Semenye 1981; Otchere, 1986) and non-herded cattle (55%, Harker
et al., 1953; Lampkin et al., 1958; Lampkin and Quarterman,
1962) .
Lack of significant difference between animal types within a
single herd means that the behaviours for the three different
categories were similar. In a situation where cattle are not
herded, similarity between the categories is less likely. Where
cattle are herded the influence of the herder dominated the
individual expression of cattle. However, when it came to
behaviour on which the herder had less influence such as
rumination, significant differences between grazing areas were
observed due to differences in forage quality and water
availability. Rumination time is known to vary with the forage
quality (Hancock, 1953; van Soest, 1982), which varied across
areas and seasons.
More time was spent drinking in the green season, when there
was plenty of surface water. Reasons have been advanced above
for expecting longer rumination periods in the dry season when
forage was at its lowest quality. Since they had access to
forage of high quality and plenty of water in the green season,
cattle should enjoy more resting time instead of less. However,
it is possible that more resting occurred in the dry/green and
dry seasons than in the green because of higher prevailing
temperatures and the desire of either the cattle or the herder to
avoid unnecessary effort. With increasing temperature, cattle
328
rest for longer periods by seeking shade. The trials of Lampkin
and Quarterman (1962) with steers at Muguga and Mariakani with
temperatures below and above 3 0°C respectively, showed that time
spent under the shade was much greater in the hotter environment.
A similar seasonal trend in walking time was observed, but it
should be noted that in the dry season cattle devoted their
entire walking time simply to walking whereas in other seasons
grazing and walking occurred simultaneously in some instances.
Cattle grazed for the shortest time in the dry season, when
quantity and quality of forage were lowest but spent a relatively
longer period in rumination.
CONCLUSIONS
The Maasai herdsmen herded their cattle with great skill despite
severe limitations of livestock facilities such as adequate water
sources, troughs and dips; and lack of paddocks and perimeter
fences. The need for increasing the length of the grazing day to
enhance intake was uppermost in their minds. However, due to
predators this was not possible, hence the total grazing time
coincided with sunrise and sunset.
Rumination time was found to vary with the forage quality.
Therefore, in dry seasons longer rumination time necessitated by
poor forage quality could limit intake.
A vibracorder timing of grazing days was found to be as
accurate as watch timing. Due to the convenience of a
vibracorder over observers, it is an ideal tool for grazing
behaviour studies.
ACKNOWLEDGEMENTS
The author wishes to thank ILCA for funds supporting this
research, the pastoralists for allowing me to study their cattle,
Drs. A. Carles, A. Said, J. Lambourne and Mr. P. de Leeuw for
advice and guidance, Messrs. P. Chara, E. Rugema, T. Tukai and E.
Wacira for their good work as field assistants and Ms. P. Pacha
for typing the manuscript.
REFERENCES
Hancock, J. 1953. Grazing behaviour of cattle. Animal Breed
Abstr. 21:1.
Harker, K.W. , Taylor, J.I. and Rollinson, D.H.L. 1953. Studies
on habits of zebu cattle. I. Preliminary observations on
habits on grazing habits. Journal Agric. Sci. 44:13.
Harvey, W.R. 1977. User's guide for least-squares and maximum
likelihood computer program. Ohio State University,
Columbus, Ohio.
Hodgson, J. 1982. Ingestive behaviour in herbage intake
handbook 6, 113.
329
Lampkin, G.H., Quarterman, J. and Kidner, M. 1958. Observations
on the grazing habits of grade and zebu steers in a high
altitude temperate climate. J. Agric. Sci. 50:211.
Lampkin, G.H. and Quarterman, J. 19 62. Observation on the
grazing habits of grade and zebu cattle. II. Their
behaviours under favourable conditions in the tropics. J.
Agric. Sci. 57:119.
Otchere, E.O. 1986. Traditional cattle production in the
subhumid zone of Nigeria. In: R. von Kaufmann, S. Chater and
R. Blench (eds) , Livestock systems research in Nigeria's
subhumid zone. Proceedings of the second ILCA/NAPRI
Symposium, Kaduna, Nigeria, 29 October - 2 November 1984.
ILCA, Addis Ababa. pp. 110-140.
Semenye, P. P. 1981. Practical application of range areas
monitoring. Proceedings of the Animal Production Society of
Kenya 13, 28.
Stobbs, T.H. 1970. Components of grazing bahaviour of dairy cows
on some tropical and temperate pastures. Proc. Aust. Soc.
Anim. Prod. 10:299.
van Raay, H.G.T. and de Leeuw, P.N. 1974. Fodder resources and
grazing management in a savanna environment: an ecosystem
approach. Samuru Research Bulletin No. 224, Zaria, Nigeria.
van Soest, P.J. 1982. Nutritional ecology of the ruminant. O &
B Books, Inc., Corvallis, Oregon.
330
THE ROLE OF BODY WEIGHT CHANGES AND OTHER FACTORS IN THE CONTROL
OF FERTILITY OF BEEF CATTLE AT ATHI RIVER RANCH, KENYA
E.K. Odhuba
Animal Production Research Department
Kenya Agricultural Research Institute (KARI) , Muguga
P.O. Box 30148, Nairobi, Kenya
Abstract
A study of breed and environmental factors affecting calving
interval of range-fed beef cattle was conducted at Athi River
Ranch in the Kenyan rangelands. In experiment 1, conception
period, rainfall and body weight changes prior to conception were
evaluated from a total of 308 calving intervals. In experiment
2, 1985 in-calf females were supplemented during the dry season
in 1978, 1979 and 1980. In experiment 3, 122 cows were allowed
to wean their calves at various ages at the end of September,
October and November/December in 1977, 1978 and 1979.
Short mean calving intervals were associated with conception
between November and January. Increased rainfall during the
month prior to conception was correlated with a linear and highly
significant (P<0.01) reduction in calving interval depending on
conception period. Year effects were highly significant
(P<0.005) . The target joining weight was around 318 kg. Above
the target weight, fertility became a function of the absolute
body weight such that heavier cows at joining were more fertile
(P<0.01) than lighter ones. Below the target weight, fertility
was associated with females that gained weight or suffered less
body weight loss during the month prior to conception.
Previously, dry cows tended to gain weight prior to conception to
the detriment of fertility (P<0.005) . Effect of calving month
was highly significant (P<0 . 005) depending on age of breeding
females. Influence of dry season supplementation was significant
but was modified by a number of factors such as age, year and
previous parity. There was no significant difference in
fertility of cows that were supplemented with energy or energy
plus urea before calving. Generally, cows that were supplemented
after calving had a shorter mean calving interval. Effects of
weaning period, month and year were marginally significant
(P<0 . 05) . Fertility improved as the dam's age increased from
three years onwards but declined after about nine to ten years.
Effects of previous parity and breed were marginal (P<0.05)
depending on conception period. Nith supplementation, Boran
crosses had a significantly (P<0.01) shorter mean calving
interval than the pure Boran females. In general, effects of
breed were minor compared to environmental factors especially
those associated with conception period.
INTRODUCTION
More than 50% of the 8.7 million beef cattle in Kenya are reared
in semi-arid to arid areas or rangelands (Livestock Development
Division, 1983) . The overall offtake of beef is not adequate to
meet domestic demand and at the same time sustain an export
market. There is need, therefore, to increase not only
331
efficiency of production but also to multiply the beef population
quickly. The latter objective calls for a higher level of
fertility. Not only does low fertility reduce productivity but
it also limits improvement of the herd genetically. For
instance, when calving rate is low, nearly all replacement
heifers must be used to maintain the herd numbers, and this
curtails selection intensity thereby lowering the rate of genetic
improvement. Besides, a decline in fertility would result in a
compensatory increase in the number of unproductive and aged cows
with adverse consequences of overstocking leading to
environmental degradation which is often irreversible.
Fertility can be evaluated in terms of calving rate, i.e.
the number of calves born in relation to the number of breeding
females exposed to the bulls per annum. This measurement is
ideal for survey studies. For selection purposes, individual cow
data are required for which the parameter evaluated is the
calving interval, i.e. the period between two successive
parturitions. Although several factors may influence fertility,
there is overwhelming evidence that inadequate nutrition is the
most important single factor mitigating reproductive efficiency
in beef cattle (Topps, 1977) . This observation is particularly
relevant to tropical rangelands where pronounced seasonal changes
in rainfall (French, 1957) contribute to a very precarious feed
supply. Topps (1977) has associated seasonal changes in feed
supply with corresponding changes in body weight of grazing beef
cattle with adverse consequences on reproductive performance.
The objective of this study was to characterise the influence of
body weight changes and related factors that are associated
directly or indirectly with the nutritional status of beef cows
grazing natural pasture on calving interval.
MATERIALS AND METHODS
Site
The study was conducted at KARI ' s field station near Athi River.
The ranch lies in a semi-arid zone with a bimodally distributed
rainfall averaging 565 mm per year (Figure 1) . The main grass
species is Themeda triandra Forsk on gently undulating mosaic of
ridges and shallow valleys of black clay loams (Gethin-Jones and
Scott, 1955; Ledger et al., 1969).
Breeding Females
The genotypes were the Boran, small East African Shorthorn Zebu
(EASZ) and their F± progeny from Hereford bulls. All animals
were run as one herd and were grazed on natural pasture. Routine
measures to control ticks and vaccinations against foot-and-
mouth, contagious abortion, rinderpest, blackquarter and anthrax
were undertaken. Random mating was followed throughout 1973.
From 1974 until 1976, mating was confined between June and
August, but this was subsequently brought forward between May and
July. Heifers were first mated after attaining a liveweight of
250 kg.
332
Figure 1. Rainfall and evapotranspiration (mm) at Athi River
Ranch.
 
.BaaL
 
Cvopotrootpirotion
"J (ETo)
/ Rointoll
A
i978
mm
 
/
/,
/y
/y/
//
JOYFLrT,m
 
J FMAMJJYASONOJ FM 1 M J JTt S ON D J F a 1UJ jr« S 0 N D
i972 i973 i974
* ETo doto not ovilobU
333
Experiment 1: Evaluation of Breed and Environmental Factors
Affecting Calving Interval
Conception Period
Each calendar year was divided into six two-month periods to
coincide as much as possible with the seasonal pattern of
rainfall and temperature. Date of "second" conception was
estimated from date of calving and assuming a mean gestation
period of 283 days (Hutchison and Macfarlane, 1958) . An
appropriate conception period was then assigned to each calving
interval.
soil Moisture Index
Evaporative loss (E-) was estimated using McCulloch's (1965)
Tables from data collected at the meteorological site at the
ranch. E- values were multiplied by a crop coefficient of 0.86
derived by Pereira and McCulloch (1962) to predict
evapotranspiration (ETQ) from a vegetative cover. A soil
moisture index (SMI) was calculated from the equation:
SMI - Total monthly rainfall fmm)
Estimated ETQ (mm/month)
which, for the purposes of this study, gave an indication of soil
water adequacy for pasture growth after taking into account the
biggest water loss (ETQ) . Indices were calculated for each
calving interval for the period of one month (SMI1) and one to
two months (SMI2) prior to "second" conception.
Orouping of Females
Arising from variation in the breeding season, females were
divided into five groups depending on the period taken after
"first" calving before joining with the bulls. Group 1, regarded
as the normal class, had bulls within 60 days post-calving. In
groups 2, 3 and 4, joining took place between 61-150, 151-230 and
231 or more days after calving respectively. In group 5, females
did not conceive during the "first" breeding season but did so
during the subsequent mating period. Preliminary analysis
indicated that there was no additional information lost by
excluding groups 3 and 5. Consequently, only groups 1, 2 and 4
were retained for final evaluation.
Experiment 2 : Effect of
Interval
Strategic Supplementation on Calving
Three experimental rations were prepared for supplementing in-
calf cows during the dry period between January and March in
1978, 1979 and 1980 as follows:
Ration A: 2 kg chopped fresh grass from the paddocks.
Ration B: 2 kg ground sorghum grain (Sorghum vulgare)
mixed with 500 g of sugarcane molasses (Energy) .
Ration C: 2 kg ground sorghum grain mixed with 500 g
molasses and 75 g of urea (Energy + Nitrogen) .
334
The supplemented regimes were:
(i) chopped grass (control) - Ration A
(ii) pre-parturient supplementation
(a) Energy alone - Ration B
(b) Energy + Nitrogen - Ration C
(iii) Post-parturient supplementation - Ration 3.
Cows on feeding regime (i) and (ii) were supplemented during
the last two months of pregnancy for varying periods until they
calved down while those on regime (iii) were supplemented up to
60 days.
Experiment 3: Effect of Weaning on Calving Interval
Breeding females were suckling their calves each year. The
calves were weaned in three batches at the end of September,
October and November/December during 1977, 1978 and 1979 when
they were between 5 and 9 1/2 months old.
Data
Each female was regarded as an entity during the breeding season
and data were compiled to include breed, birth and calving dates,
previous parous state i.e. whether the animal was a heifer, dry
cow or suckling a calf at conception prior to the calving
interval in question and conception period. The covariables
included body weight at weaning (WWN) , "earlier" calving (WCV) ,
one month after calving (W1ACV) , one and two months prior to
"second" conception (W1BC and W2BC respectively) and at "second"
conception (WCP) , SMI1 and SMI2. From this information, age at
"earlier" calving, calving interval and weight changes WWN-WCV,
W1ACV-WCP, W1BC-W2BC and W1BC-WCP were computed. Factorial
effects in experiments 1, 2 and 3 are summarised in Tables 1, 2
and 3 respectively.
335
Table 1. Data structure and number of females in experiment 1.
Groups
Class Levels 1 2 4 Total
1. Breed type Boran 44 29 58 131
EASZ 42 47 33 122
Crosses* 28 20 7 55
2 . Dam ' s age at 3 or below 18 16 22 56
earlier calving 4-8 65 38 71 174
(years) 9 and above 31 42 5 78
3 . "Second" conception Mid Jan-mid March 1 7 33 41
period Mid March-mid May 2 - 8 10
Mid May-mid July 52 59 19 130
Mid July-mid Sept 37 26 1 64
Mid Sept-mid Nov 6 - - 6
Mid Nov-mid Jan 16 4 37 57
4. Year of "second" 1973 (early) 2 11 70 83
conception 1973 (late) 26 - 3 29
1974 2 3 - 5
1975 28 59 19 106
1976 23 23 - 46
1977 33 - 6 39
5. Dam's previous Heifer 17 17 31 65
parity Dry cow 39 14 49 102
Suckled 58 65 18 141
Total 114 96 98 308
OCrosses were F± from Boran and EASZ cows x Hereford bulls.
Table 2. Data structure and number of females in experiment 2.
Class Levels No. of observations
1. Breed type
Dam's age at
earlier
calving (yrs)
Dam's previous
parous state
Feeding regime
Month of
earlier
calving
Year of
earlier
calving
Boran 34
Boran x Hereford (F^ 52
EASZ 48
EASZ x Hereford (F,) 51
4 22
5-9 100
10 and above 63
Dry cow 72
Suckled 113
Chopped fresh grass 40
Pre-parturient supplementation 57
(a) Energy 43
(b) Energy + Nitrogen 45
Post-parturient supplementation 57
February 83
March 58
April/May 44
1978 58
1979 53
1980 74
Total 185
336
Table 3. Data structure and number of females in experiment 3.
Class
1. Breed type
2 . Dam ' s age at
weaning (yrs)
3. Weaning month
4. Weaning year
5. Calving month*
Levels No. of
observations
Boran 21
Boran x Hereford (F,) 42
EASZ 29
EASZ x Hereford (F, ) 30
4 14
5-9 70
10 and above 38
End of September 35
End of October 38
End of Nov/December 49
1977 28
1978 43
1979 51
Jan/February 46
March 36
April/May 40
Total 122
OMonth of calving following weaning in question.
Statistical Analysis
The sources of variance and covariance for calving interval were
analysed using the generalised least-squares regression methods
for multiple classifications and non-orthogonal data as described
by Harvey (1960) and adapted by Siebeck (1976) in his computer
programme, SYSNOVA. All effects were considered fixed. The
estimate of the mean squares attributable to any effect was
computed after all other effects in the evaluation model had been
fitted and was tested against that of the residual.
The generalised model used was:
Yijkl " + ai + bj + cjk + <ab>ij + <ad>i + d<Dijkl-D) + eijkl
where,
Y^oi{^ - calving interval of an individual cow
- effect common to all cows
a^ - effect of the ith A class after removal of
bj - effect of the jth B class after removal of
coijj = effect of the kth C class within the jth B class after
removal of the jth B class. This was hierarchical
effect of nutrients within the pre-parturient feeding
regime.
(abJioi - effect of the ijth AB sub-class after the average
effects of A and B have been removed.
337
(ad) ± - effect of an interaction between A and the continuous
variate Di-jkl ■ Tili-a interaction between a treatment
and a covarlate enables particular slopes to be fitted.
- partial regression coefficient of calving interval
Yijkl on Dijkl-
eijkl random errors.
RESULTS
Weight Change During the Month Prior to "Second" Conception
The effect of weight change on calving interval was influenced
significantly (P<0.01) by the period during which females
reconceived and is depicted in Figure 2. During the May-July
period, animals reconceived at an average weight of 331 kg, and
weight changes did not affect calving interval to any appreciable
degree. Weight gains up to 10 kg prior to conception between
July and September were associated with short calving intervals
(Figure 2) . Females reconceived between September and January at
an average weight of 308 kg, and long intervals were associated
with weight loss in excess of 10-15 kg. Influence of weight
change also varied significantly (P<0.005) and curvilinearly
depending on year of "second" conception as shown in Figure 3.
While weight gain during 1976 resulted in systematic and sharp
reduction in calving intervals, this effect was marginal during
1975. Weight gain during 1973/74 caused considerable variation
in calving interval.
Weight Change Between One to Two Months Prior to "Second"
Conception
The effect of this parameter was linear and highly seasonal
(P<0.005) and is depicted in Figure 4. Females that gained
weight prior to conception between July and September had calving
intervals which were reduced by 0.7 days for each kg increase in
body weight. This trend was reversed in the case of animals that
conceived during periods other than July-September such that
females that gained weight also had longer calving intervals.
However, effect of weight gain prior to conception between May
and July was marginal.
338
Figure 2. Relationship between calving interval and weight (W1BC-
WCP) within conception period of group 1 cows.
Y= 356.9+0.79X+0.I4X
 
339
Figure 3. Relationship between calving interval and weight (WlBC-
WCP) within conception year of group 2 cows.
1973/74
1973
1976
Y= 369 + 5. 70X-0.33X'
Y= 369»0.I2X-0.0IX'
1
 
Y=369- 1. 77*0.04 X*
Waight (kg)
340
Figure 4. Relationship between calving interval and weight (WICP-
W2CF) within conception period of group 2 females.
4i0 Y = 369+ 4.66X
£ Jan-Mor
 
-l0 0 10 20
Wtight (kg)
30 40
341
Weight Chang* Between On* Month After Calving and Reoonception
Influence of this factor varied significantly (P<0.05) depending
on the year when reconception occurred as depicted in Figure 5.
During 1978, weight gains up to 25 kg were associated with short
calving intervals. During 1979, effect of weight change was
marginal. Weight gains during 1980 resulted in long intervals.
Mean calving interval was reduced by 0 . 7 days for each kg
increase in body weight following calving during March; otherwise
the overall effect of weight change did not vary across calving
month .
Figure 5. Relationship between calving interval and weight
(WIACV-WCP) within year of earlier calving of
supplemental cattle.
440
Y= 362 + 0.74 X + 0.03X* 
10 20
Weight (kg)
342
Body Weight at "Second" Conception
In the normal class of females, the optimal weight at "second"
conception was 318 kg with a range between 310 and 330 kg (Figure
2) . However, joining weight did not affect calving interval
significantly. Group 4 females reconceived at an average weight
of 311 kg, and heavier animals had a significantly (P<0.05)
shorter mean calving interval. In the weaning trial, animals
reconceived at an average weight of 351 kg, and for each kg
increase in liveweight above the overall mean, there was a
corresponding reduction (P<0.01) in calving interval of 1.1 days
(Figure 6) .
Figure 6. Relationship between calving interval and weight at
"second" conception.
\ 440
\ 400 -
*- \o \
\ Y = 360.7-I.07X
>
£350c
\
c
>
oo
\
1
u
300
i i - 1 i i i
-75 -50 -25 0 25
Weight (kg) above /below mean weight
50 75
343
Soil Moisture Index
The effect of this factor on calving interval was linear and
highly significant (P<0.005) but was modified by conception
period as shown in Figure 7. In general, increase in SMI during
the month prior to "second" conception was associated with a
shorter mean calving interval . The response was greatest during
the September-January conception period but was relatively
marginal when conception occurred between May and September.
Figure 7. Relationship between calving interval and soil moisture
index (SMI I) within conception period of group l
females.
350
340
330
o 320
(I
:
a
e
" 310
300
 
\ Y = 356.9-46.2SMTT
\ May -Jul
\ Jul - Sept
N S«pt - Jon
"^-Y^e. 9-248. 2SMII \
-1- -l- -,_ -i
0.2 0.4 0.6 0.6 IO 1.2 1.4 1.6
Soil moisture index
344
Aga of Breading Pemelas
When evaluated within conception period, age of animals had a
highly significant (P<0.005) influence on calving interval
(Figure 8) . The three-year-old females that conceived between
January and July had a longer mean calving interval compared to
the 4 to 8-year-old cowa. The shortest intervals were achieved
by the 4 to 8-year-old cows which conceived between November and
January. Regardless of age, all animals that reconceived between
July and September had long intervals.
Figure 8. Effect of age at calving (years) within conception
period on calving interval of group 2 females.
60 -
40 -
20
m~°
?
°B 0
o >
Si
-20
-40 -
-60
3 years
 
Jan/May May/jul Jui/S«pt
Conception period
Nov/Jon
345
Dam's Previous Parity
The effect of this parameter became marginally significant
(P<0.05) when evaluated in the context of conception period
(Figure 9) . Heifers that conceived between January and July had
longer calving intervals compared to previously dry cows.
Reconception between November and January was associated with
shorter calving intervals regardless of previous parity.
Year of "Second" Conception
Year effects on calving interval were highly significant
(P<0.005). Females that conceived during 1973 (late), 1975 and
1977 had shorter intervals compared to those that conceived
during 1973 (early) and 1976.
Role of Supplementation
Dry season supplementation had a significant effect on calving
interval, but this was modified by a number of factors:
Age: Without supplementation, the 5 to 9-year-old cows had
calving intervals which averaged 38 days shorter than the mean
for the 4-year-old females (Figure 10) . Supplementation, whether
before or after calving, only maintained calving intervals of the
mature cows at average levels. However, there was a dramatic
reduction in the interval when the 4-year-old females were
supplemented after calving.
Previous parity: Females that suckled calves during the previous
year responded positively to supplementation especially when
given after calving (Figure 11) .
year of supplementation: While supplementation reduced calving
intervals during 1978 and 1979 (post-parturient) , it was
virtually useless during 1980 (Figure 12) .
Month of "earlier" calving: Females that calved between March
and early May had shorter calving intervals compared to those
that calved during January and February (Figure 13) . However,
with no supplementation, the 4-year-old females that calved
during April and May had long intervals (Figure 14) .
Energy versus energy + urea: Differences in the effects of these
nutrients when given pre-partum were not significant.
346
Figure 9. Effect of previous parity within conception period on
calving interval of group 4 females.
200
150
s
o■o
— 100
50
o
c
o
: -so
£ -too
8
s
o
I -150
-200
Heifer
Dry cow
Suckled —o—
 
-i-
Jon-Mor Mor-Jul
Conception period
Nov -Jon
347
Figure 10. Effect of aga calving within faading regime on oalving
interval of supplemental oattla.
20
I 0
-5
■10
-20
Z -30
2
-40
4 years
5-9 ytors
lOyears —o-
 
Ch-G = Chopped grass
Pre = Pre-porturient
Post = Post-porturient
Ch.G Pre-
Feeding regime
Post-
348
Figure 11. Effect of previous parity within feeding regime on
calving interval.
20 r Suckled
Ory cow —o—
 
Ch.G Pre-
Fetding regime
Pott-
349
Figure 12. Effect of feeding regime within calving year on
calving interval.
Chopped gross —
Pre -parturient —
Post-parturient — x —
 
1978 1979
Calving year
1980
350
Figure 13. Effect of age (years) within month of earlier calving
on calving interval of supplemented cattle.
 
Fab May Apr/May
Month of earlier calving
351
Figure 14. Effect of dam's age at weaning within subsequent
calving month on calving interval.
 
Jan/Feb Mar
Colving month
Apr/May
352
Effects of Weaning
Calving intervals increased marginally (P<0.05) by 0.2 days for
each day increase in the period before weaning was effected.
There was some indication that the 4-year-old females that weaned
calves at the end of September had a shorter mean calving
interval than that of females that weaned calves at the end of
October, otherwise the effect of weaning month per se was
marginal (P<0.05). As weaning progressed from year to year, the
10-year-old-and-above cows had calving intervals which averaged
32 and 21 days longer than for the 4- and 5 to 9-year-old females
respectively. In general, effects of weaning on calving interval
were marginal.
Dam's Genotype
The influence of this parameter was significantly (P<0.05)
modified by conception period (Figure 15) . Both the EASZ and the
Boran females had similar calving intervals except during the
September-January conception period when the latter breed had
shorter intervals. The crosses that conceived between May and
July had calving intervals which averaged 3 0 days below the
adjusted overall mean of 357 days. Animals of all genotypes
which conceived between July and September had long intervals.
With supplementation, the Boran crosses had significantly
(P<0.01) shorter calving intervals than pure Borans.
353
Figure 15. Effect of breed type within conception period on
calving interval of group 1 females.
o
*a
20
i0-
I °
: -io
m
3
•o
o
E
o
_ -20
•30
Boron
EASZ
Crosses — oo
 
/
May- Jul Jul- Sept
Conception period
Sept-Jan
354
DISCUSSION
The highly significant main effects and interactions with body
weight changes, soil moisture index, age, previous parity and
dam's genotype made conception period one of the most important
factors affecting fertility of grazing beef cattle at Athi River
Ranch. High fertility was achieved during the November-January
conception period which normally covers the "short rains" with
high temperatures. The high fertility observed between January
and March (Figure 8) could be due to increased sexual activity
mainly as a result of high temperatures (Anderson, 1944) .
However, in rather striking contrast, group 4 females that
conceived during the same period had a long fertility (Figure 9) .
Considering that the number of females that conceived during this
period in group 4 was much greater than that in group 2 (Table
1) , it would appear that inadequate grazing had a more marked
adverse effect on fertility than the influence of increased
temperature and sexual activity. Low fertility observed during
the cold and dry period between July and September could be due
to a level of intake sufficient to bring about ovulation but
grossly inadequate for full manifestation of oestrus.
The highly significant year effects reflected important
variation in fertility from year to year. Admittedly, a large
proportion of year effects would be confounded with variation
associated with conception period. But since it is possible that
factors other than conception period could contribute
significantly to the observed variation, inclusion of year
effects in the evaluation model would still be justified. Age of
bulls and management practices relating to stocking rate, mineral
supplementation and culling programmes are some of the factors
which could vary from year to year.
Bishop (1978) reported high conception rates following
rainfall during the previous year. However, in Kenyan rangelands
with such tremendous variability in rainfall both annually and
seasonally (French, 1957) , it would be desirable to correlate
fertility with rainfall within the immediate past so that
corrective measures could be instituted in good time. In
Australia, Andrews (1976) reported a high correlation between
conception rate and rainfall with a one-month lag. Working at
Athi River Ranch, Potter (1985) observed that liveweight gain was
correlated with rainfall occurring between three to six weeks
previously. The period of SMI evaluation in this study coincided
with the one used by Andrews (1976) but did not quite correspond
with the one evaluated by Potter (1985). Further work is needed
to pin-point this period more precisely. Potter (1985) has
suggested that use of rainfall data was just as good as the more
elaborate and often cumbersome water balance information
involving evapotranspiration data. Although rainfall may be an
important factor, Andrews (1976) contended that a period of one
month was not sufficient time for pastures to grow and improve
body condition and hence conception rate. He postulated that
increased fertility was likely to be due to either a change in
the nutritional composition of the pastures or an increase in the
rate of intake or both through the action of the hypothalmus
and/or the pituitary. However, it is important to note that
rangeland vegetation at Athi River Ranch grows very fast and
355
matures very quickly (Karue, 1972; 1974), and so it may have a
short-term direct effect on body condition and hence fertility.
The debate continues as to whether the animal's ability to
conceive was a function of its absolute body weight per se
(Lammond, 1970; Grosskopf, 1980) or weight change prior to
conception (Elliot, 1964, cited by Topps, 1977; Capper et aj . ,
1977; Thorpe et al., 1981). While suggesting the concept of a
target joining weight, Lammond (1970) had intimated that for each
cow, depending on breed, age and year, there was a certain range
in body weight and body condition required for conception. By
implication, therefore, it did not matter whether animals lost or
gained weight during the breeding season provided their
liveweight was maintained above a certain minimum. The ability
to maintain weight would depend, among other things, on the
current lactational stress on the animal, hence the reason why
researchers evaluate body weight in heifers, dry or lactating
cows (Kidner, 1966; Sacker et al . , 1971; Morley et al , 1976).
Such evaluation would be possible if fertility was considered in
terms of calving date rather than calving interval. Another
aspect is whether the various workers were evaluating absolute or
proportionate weight changes. A proportionate weight change such
as that used in this study with weight at "second" conception as
a reference point would render changes in small animals (though
of the same breed) comparable to similar changes in bigger ones.
It seems as if both body weight and weight change operate singly
or simultaneously to influence fertility depending on the
lactational and/or nutritional status of the animal. For
instance, although animals in the normal class reconceived at an
average weight of 320 kg well within the optimum range of 310 to
330 kg for this group of animals, influence of body weight was
not significant. This could have been due to the adverse effect
of suckling reported by Wiltbank and Cook (1958) and Hutchison
(1983) .
Group 4 females had dried off at the time they reconceived,
and the effect of joining weight was able to be manifested,
though marginally due to the fact that animals conceived at an
average weight of 311 kg which was at the bottom of the required
range in body weight. Under these circumstances, both body
weight at joining and weight change affected fertility
simultaneously. Animals reconceived between September and
January and during 1976 at average weights of 308 and 298 kg
respectively. Both weights were below the minimum weight
required for conception, and animals that had a higher fertility
were those that gained or lost less body weight during the month
prior to conception. With favourable nutritional conditions
between May and July (Figure 2) , during 1975 (Figure 3) and 1979
(Figure 5) , animals reconceived at average weights of 330, 311
(half the number of females were the small EASZ) and 352 kg
respectively. Under these conditions, effect of weight change
was marginal and animals could gain or lose weight without
jeopardising their fertility. The picture during 1973/74 (Figure
3) was somewhat confusing probably due to the fact that fertility
was evaluated over a two-year period. In any case, the sample
size during 1973 and 1974 was far too small to warrant any
serious consideration of the effects during these two years.
Curvilinearity in response of weight change could also be
ignored.
356
Dry cows (group 4) and females that had settled down after
calving before they were re-introduced to the bulls (group 2)
tended to gain weight prior to conception to the detriment of
fertility. However, in normal ranching situations, animals would
be lactating at the time they reconceive with minimum chances of
overweight. Failure to register significant effects of weight
change between one month after calving and conception could have
been due to the rather limited variation in climatic conditions
during the pre-selected breeding season between May and July.
However, it was evident that weight changes occurring near
conception exerted a much greater effect on fertility than
changes far removed from conception. What is, perhaps,
perplexing is the revelation by Thorpe et al . (1981) that
liveweight was merely an approximate indicator of the nutritional
status of a beef animal. These workers postulated the existence
of an intricate relationship between nutrition, hormonal control,
lactation and fertility in which the post-partum period with its
associated lactational stress was considered more important than
variation due to liveweight. Since this relationship is still
vaguely understood, monitoring of body condition as suggested by
Andrews (1976) together with proportionate body weight changes
during the month prior to breeding would give a reasonable
indication of the fertility status of a grazing beef animal.
Supplementation of beef cattle with energy (Wiltbank et al
1962 and 1964; Dunn et al . , 1969) or nitrogen (Elliot, cited by
Christie, 1962; Siebert et al., 1976; Holroyd et al . , 1977) has
been a normal practice to improve fertility but results have not
been consistent. For instance, in Botswana, Capper et al. (1977)
observed no response to stimulatory licks of molasses and urea by
dry cows or cows with calves older than five months. More
recently, Holroyd et al . (1983) supplemented beef cows with urea
and molasses but observed no effect on fertility. In this study,
improved fertility was observed in the four-year-old and
previously suckled females. It seems imperative, therefore, that
assessment of the effect of supplementation should be done in the
context of age and previous parity. Failure to detect
differences in the effect of grain versus grain plus urea feeding
could be attributed to the favourable climatic conditions
especially during 1979 and 1980 which made it possible for
animals to derive most of their nutrient requirements from
pasture alone. It is worth noting that the last trimester of
pregnancies at the ranch coincided with the period covering the
"short rains" between November and December. Under such
conditions, Orskov (1982) suggested that nitrogen requirements
for pregnancy could be met by microbial protein synthesis alone.
However, there is a tremendous increase in the nutritional
requirements for lactation and reconception (Crampton and Lloyd,
1959) , hence the significance of post-calving supplementation.
It would, therefore, be of interest to investigate the effect of
nitrogen and energy sources when given post-partum. Since all
indications pointed at conception as the most critical period, it
would be necessary to pinpoint more precisely how soon after
calving should supplementation be given. Because of the
intricate relationship between non-protein nitrogen (NPN) and
carbohydrate metabolism and utilisation (Orskov, 1982) , further
investigations should be focused not on the relative importance
of NPN and energy but rather on the proportion of rumen
357
degradable and undegradable nitrogen relative to metabolisable
energy intake.
The effect of age on fertility of beef cattle is an indirect
one through its influence on body condition to which young cattle
are more sensitive (Andrews, 1976) . As body condition was
related to liveweight, Wiltbank and Spitzer (1978) have stressed
the significance of allowing young cattle to attain a target
weight and age before first mating - the weight and age varying
according to breed. Under East African conditions, Macfarlane
and Worrall (1970) had indicated that puberty in Boran heifers
occurred at approximately 60% of the mature body weight.
However, mere attainment of puberty is not enough as young cattle
are expected to conceive, carry the foetus to full term, rear the
calf successfully and reconceive at the earliest opportunity. It
is possible that the 250 kg liveweight stipulated for first
breeding of heifers at the ranch was lower than the threshold
joining weight especially in crosses. This would not only delay
reconception but could curtail their productive life. If
anything, it would be better to breed heifers too late than too
soon. Young beef cattle were more vulnerable to adverse
nutritional conditions at post-partum oestrus. However, their
fertility improved as the post-calving period before joining with
the bulls increased from about five months onwards. There was
evidence from the weaning trial that highest fertility was
achieved by the four-year-old females. The implication of these
observations is that young cattle should be allowed about five
months after calving before joining with bulls for their second
mating season. This would mean that first breeding in heifers
and that of the main herd would be out-of-phase, although all
indications were in favour of seasonal mating. However, earlier
breeding in heifers would necessitate their separation from the
main herd which would facilitate closer attention and feeding to
ensure high fertility. A suggested package for young beef cattle
would be (a) mating between November and January or, in general,
immediately after a rainy period (b) separating and mating
approximately three months before the main herd to synchronize
second mating with the rest of the herd and (c) providing a
supplementary diet after first calving.
The timing of the breeding programme at the ranch did not
permit evaluation of the effects of calving months other than
those between late January and early May. However, it was
evident that animals that calved down during the dry season in
January and February had a low fertility. This could be due to
failure to regain lost body weight (Kidner, 1966) but as Christie
(19 62) had indicated, dry season loss in body weight per se did
not, necessarily, reduce fertility as long as the animal was able
to recoup during the subsequent wet season. Animals that calved
down during March had a high fertility which confirmed Daly's
(1971) suggestion that calves should be dropped at the end of a
dry season. The older cattle that calved down during the rainy
season in April-May seemed to get away with less favourable
nutritional conditions at post-partum oestrus probably because of
their ability to draw on body reserves accumulated during the wet
period (Orskov, 1982) .
There were indications to suggest that previous parity had
an influence, albeit small, on fertility. Results from various
358
workers reporting on this parameter are rather inconsistent
(Christie, 1962; Buck et al., 1976; Thorpe et al . , 1981). It
would appear that in order to make meaningful comparison in
fertility, one would have to consider previous parity in the
context of conception period. When nutrition was adequate, for
instance during November-January conception period, or when
supplementation was given after calving, influence of previous
parity became insignificant. Generally, carry-over effects
arising from previous parity were of less importance in the
fertility complex. This should not be confused with the over
whelming influence from current lactational stress on the animal
(Crampton and Lloyd, 1959; Trail, 1968), and the whole idea
behind weaning is to remove this stress thereby ensuring that
animals attain the body weight and body condition required for
reconception. Lammond (1970) had suggested increasing body
weight and condition before calving and attempting to hold this
condition up to the breeding season. This could be achieved by
early weaning but because of seasonality in pasture productivity
(Karue, 1972, 1974), it would be difficult to maintain body
weight. Besides, results of this study indicated that fertility
was not always correlated with body weight at calving. The other
alternative also suggested by Lammond (1970) was to allow cows to
calve down in reasonable body condition and feed heavily up to
conception. This would be a much more practical method which was
supported by results of this study relating to post-partum
supplementation.
The question of weaning month necessitated some
consideration because of the observed sensitity to pasture by the
early weaned calves (Schotter and Williams, 1975) on one hand and
the need to prepare the breeding female for subsequent calving
and, more importantly, reconception on the other. However,
effects of weaning month were, at best, only marginal probably
because of a closed breeding period and hence weaning season.
Thus, in considering the effects of lactation, it would be more
important to pay particular attention to when animals are due to
calve down so that they can be supplemented for early
reconception rather than attempting to increase body weight and
body condition at calving by early weaning.
The EASZ females exhibited fairly consistent fertility
levels throughout the year, an observation that was consistent
with the findings of Galukande et al. (1962). Zebu cattle are
indigenous in East Africa (Mason and Maule, 1960) and would be
expected to have adapted themselves to fluctuations in feed
supply and related factors from out-of-season rains and prolonged
droughts. Adaptability has been attributed to their ability to
lay down fat on a lower plane of nutrition (Mason and
Buvanendran, 1982) . Fertility of the Borans was, by and large,
similar to that of the EASZ which supported the observations
reported by Trail (1968) at Ruhengere Ranch in Uganda. However,
there was evidence to indicate that the Borans had a higher
fertility during the last quarter of the year when pasture
productivity was adequate.
Results of cross-breeding to improve fertility have been
inconsistent. Some reports have indicated a reduction in age at
first calving (Mahadevan and Hutchison, 1964) and increased
manifestation of oestrus in cross-bred zebu cattle (Swensson et
359
al., 1981). Observations from this study tended to tip the
balance in favour of Boran crosses under conditions of adequate
nutrition. High fertility could be due to heterosis. However,
this could not be confirmed due to lack of reciprocal crosses.
Although it is desirable to maintain the indigenous zebu
germplasm, the prospects of increased beef productivity from
rangelands based on the Boran and its crosses look very
promising.
CONCLUSIONS
Fertility in range-fed beef cattle is a very complex
physiological process involving the interplay of a number of
factors. These factors are overwhelmingly climatic in origin
which means that environmental conditions especially nutrition
and management have to be tailored for the benefit of the animal.
Nutritional and management factors immediately before and during
the mating season had the most significant and far-reaching
effects on fertility. Even during the conception period,
influence of nutrition and related factors such as rainfall, body
weight and dam's age varied from season to season and was
modified by other factors which, all working in concert,
complicated the situation even further.
Fertility was seasonal. Consequently, seasonal mating would
be advocated. High fertility was observed between November and
January followed by May-July. The significant year effects
reflected mainly seasonal effects although variation across years
could be attributed to changes in feeding, mating practices and
age of bulls. Increased rainfall during the month before
conception was associated with high fertility. The effect was
seasonal which underscored the significance of conception period
and seasonal mating.
The target joining weight was around 318 kg with a range
between 310 and 330 kg. Heavier cows at conception were more
fertile than lighter ones, but the depressing effect of suckling
could counteract that of body weight unless joining weight was
well above 318 kg. Animals that gained body weight or suffered
less weight loss during the month prior to conception had a
higher fertility. Previously dry cows tended to gain weight
prior to conception and gave a better indication of fertility
status than weight changes far-removed from conception.
Dry season supplementation only helped to maintain fertility
of older cows at average levels. However, it improved fertility
of young and previously suckled females especially when given
post-partum. There was no significant difference in the
fertility of cows that were supplemented before calving with
energy or energy plus urea.
Fertility improved marginally following early weaning. Some
benefit was achieved when young females weaned their calves in
September following a dry and cold season. Generally, effects of
weaning were of less significance in the fertility complex.
Fertility also improved as the dam's age increased from
three years onwards but declined after about nine to ten years.
360
Young cattle were more vulnerable to adverse post-partum
nutritional conditions. However, their fertility improved when
they were allowed about five months after first calving before
mating for the second parity. This would mean separation and
mating of heifers some three months or so before breeding in the
main herd to be able to synchronise their second mating with the
rest of the herd.
Effect of previous parity on fertility was marginal
depending on the nutritional level during the breeding season.
Under grazing conditions without supplementation, dry cows were
more fertile than previously suckled animals. This trend was
reversed with supplementation. Females that conceived between
November and January had a high fertility regardless of previous
parity. Fertility levels of the East African shorthorn Zebu and
Boran cows were similar except during the last quarter of the
year when the latter breed had a better performance. Generally,
crosses had a high fertility following a rainy season. However,
with dry season supplementation, performance of the Boran crosses
excelled the rest of the genotypes.
ACKNOWLEDGEMENT
This paper was prepared for presentation at the PANESA 1987
workshop with the permission of the Director, K.A.R.I.
REFERENCES
Anderson, J. 1944. The periodicity and duration of oestrus inb
Zebu and grade cattle. Journal of Agricultural Science
(Cambridge) 34:57-68.
Andrews, L.G. 1976. Reproductive performance of beef cattle in
the Northern Territory. M.Sc. thesis, James Cook
University, Australia.
Bishop, E.J.B. 1978. The performance of a beef breeding herd
subjected to continuous mating in the Valley Bushveld of the
Eastern Cape. South African Journal of Animal Science 8:15-
25.
Buck, N.G., Light, D. , Rutherford, A., Miller, M. , Rennie, T.W.,
Pratchett, D. , Capper, B.S. and Trail, J. CM. 1976.
Environmental factors affecting beef cow reproductive
performance in Botswana. Animal Production 23:357-363.
Capper, B.S., Pratchett, D. , Rennie, T.W., Light, D. , Rutherford,
A., Miller, M. , Buck, N.G. and Trail, J. CM. 1977. Effect
of rumen stimulating licks on reproductive performance and
liveweight gain of beef cattle in Botswana. Animal
Production 24:49-55.
Christie, G.J. 1962. Sterility or reduced fertility due to
nutritional causes. Bulletin of Epizootic Diseases of
Africa 10:161-167.
361
Crampton, E.W. and Lloyd, L.E. 1959. General nutritional aspects
of reproduction and lactation. In: Fundamentals of
nutrition. W.H. Freeman and Co. , San Francisco and London,
pp. 386-393.
Daly, J.J. 1971. A breeding policy for beef herds. Queensland
Agricultural Journal 97: 171-175; 237-242.
Dunn, T.G., Ingalls, J.E., Zimmerman, D.R. and Wiltbank, J.N.
1969. Reproductive performance of two-year-old Hereford and
Angus heifers as influenced by pre- and post-calving energy
intake. Journal of Animal Science 29:719-726.
French, M.H. 1957. Nutritional value of tropical grasses and
fodders. Herbage Abstracts 27:1-9.
Galukande, E.B., Mahadevan, P. and Black, J.G. 1962. Milk
production in East African Zebu cattle. Animal Production
4:329-336.
Gethin-Jones, G.H. and Scott, R.M. 1955. A second provisional
soil map of Kenya and Tanganyika. East African Agriculture
and Forestry Research Organization, Muguga, Kenya.
Grosskopf, J.F. 1980. Non-pathogenic factors associated with the
reconception of beef cows under extensive conditions. In:
IV World Conference on Animal Production. The Argentina
Association of Animal Production, Buenos Aires, pp. 511-525.
Harvey, W.R. 1960. Least squares analysis of data with unequal
subclass numbers. Agricultural Research Service, U.S.D.A.,
Wash. , D.C.
Holroyd, R.G., Allan, P.J. and O'Rourke, P.K. 1977. Effect of
pasture type and supplementary feeding on the reproductive
performance of cattle in the dry tropics of North
Queensland. Australian Journal of Experimental Agriculture
and Animal Husbandry 17:197-206.
Holroyd, R.G., O'Rourke, P.K., Clarke, M.R. and Loxton, I.D.
1983. Influence of pasture type and supplement on fertility
and liveweight of cows and progeny growth rate in the dry
tropics of Northern Queensland. Australian Journal of
Experimental Agriculture and Animal Husbandry 23:4-13.
Hutchison, H.G. 1983. Physiological limitations on reproduction
performance: In: World Conference on Animal Production.
European Association for Animal Production, Rome. pp. 147-148
Hutchison, H.G. and Macfarlane, J.S. 1958. Variation in
gestation periods of Zebu cattle under ranch conditions.
East African Agricultural and Forestry Journal 24:148-152.
Karue, C.N. 1972. Supplementary feeding of beef cattle. Paper
presented to the Animal Production Society of Kenya,
October.
362
Karue, C.N. 1974. The nutritive value of herbage in semi-arid
lands of East Africa. 1. Chemical composition. East African
Agricultural and Forestry Journal 40:89-95.
Kidner, E.M. 1966. Seasonal body weight changes in suckled Boran
cows at pasture and their probable association with interval
between calvings. East African Agricultural and Forestry
Journal 31:399-404.
Lammond, D.R. 1970. The influence of undernutrition on
production in the cow. Animal Breeding Abstracts 38:359-372.
Ledger, H.P., Odero, J.C. and Ndugire, N. 1969. Athi River Field
Station. Record of Research Annual Report. East African
Agriculture and Forestry Research Organization, Muguga,
Kenya. pp. 118-119.
Livestock Development Division. 1983. Annual Report. Ministry
of Agriculture and Livestock Development, Kenya. pp. 25-27.
Macfarlane, J.S. and Worrall, K. 1970. Observations on the
occurrence of puberty in Bos indicus heifers. East African
Agricultural and Forestry Journal 15:409-410.
Mahadevan, P. and Hutchison, H.G. 1964. The performance of
crosses of Bos taurus cattle for milk production in the
coastal region of Tanganyika. Animal Production 6:331-336.
Mason, l.L. and Maule, J. P. 1960. The indigenous livestock of
Eastern and Southern Africa. Commonwealth Agricultural
Bureaux, Farnham Royal, Buckinghamshire.
Mason, l.L. and Buvanendran, V. 1982. Breeding plans for
ruminant livestock in the tropics. FAO, Rome.
McCulloch, J.S.G. 1965. Tables for the rapid computation of the
Penman estimate of evaporation. .East African Agricultural
and Forestry Journal 30:286-295.
Morley, F.H.W., Axelsen, A. and Cunningham, R.B. 1976. Livestock
at joining and fertility in beef cattle. Proceedings of the
Australian Society of Animal Production 11:201-204.
Orskov, E.R. 1982. Host animal protein requirements and protein
utilization. In: Protein nutrition in ruminants.
Academic Press Inc. (London) Ltd., London. pp. 85-136 151.
Pereira, H.C. and McCulloch J.S.G. 1962. Water requirements of
East African crops. In: E.W. Russell (ed.), The national
resources of East Africa. D.A. Hawkins Ltd., Nairobi,
pp. 88-91.
Potter, H.L. 1985. Aspects of climate, herbage growth and animal
production in a semi-arid area of Kenya. Ph.D. thesis,
University of Nairobi.
Sacker, G.D., Trail, J. C.M. and Fisher, l.L. 1971. Crossbreeding
beef cattle in Western Uganda. 4. Calving percentage and
mothering ability. Animal Production 13:165-170.
363
Schotter, J.H. and Williams, W.J. 1975. The effects of early
weaning of Brahman-cross calves on calf growth and
reproductive performance of the dam. Australian Journal of
Experimental Agriculture and Animal Husbandry 15:456-459.
Siebeck, R.M. 1976. SYSNOVA version: a reference manual.
Tropical Research Centre, Division of Animal Production,
C.S.I.R.O. Rockhampton, Queensland.
Siebert, B.D., Playne, M.J. and Edye, L.A. 1976. The effects of
climate and nutrient supplementation on the fertility of
heifers in North Queensland. Proceedings of the Australian
Society of Animal Production 11:249-253.
Swensson, C. , Schaar, J., Brannang, E. and Meskell, L.B. 1981.
Breeding activities of the Ethio-Swedish integrated rural
development project. III. Reproduction performance of Zebu
and crossbred cattle. World Animal Review (F.A.O.) 38:31-36.
Thorpe, W. , Cruickshank, D.K.R. and Thompson, R. 1981. Genetic
and environmental influences on beef cattle production in
Zambia. Animal Production 33:165-177.
Topps, J.H. 1977. The relationship between reproduction and
undernutrition in beef cattle. World Review of Animal
Production 13:43-49.
Trail, J. C.M. 1968. Beef breeding at the Ruhengere Field
Station. Reproductive performance. In: G.D. Sacker and
J. C.M. Trail (eds) . Proceedings of the beef cattle breeding
and ranching development conference, Mbarara. Ministry of
Animal Industry, Game and Fisheries, Uganda. pp. 44-49.
Wiltbank, J.N. and Cook, A.C. 1958. The comparative reproductive
performance of nursed cows and milked cows. Journal of
Animal Science 17:640-648.
Wiltbank, J.N., Rowden, W.W. , Ingalls, J.E., Gregory, K.E. and
Koch, R.M. 1962. Effect of energy level on reproductive
phenomena of mature Hereford cows. Journal of Animal
Science 21:219-225.
Wiltbank, J.N. , Rowden, W.W. , Ingalls, J.E. and Zimmerman, D.R.
1964. Influence of post-partum energy level on reproductive
performance of Hereford cows restricted in energy intake
prior to calving. Journal of Animal Science 23:1049-1053.
Wiltbank, J.N. and Spitzer, J.C 1978. Recent research on
controlled reproduction in beef cattle. Practical
application. World Animal Review (F.A.O.) 27:30-35.
364
THE ROLE OF HATER PULSES ON THE PRODUCTIVITY OF SAVANNA
ECOSYSTEMS IN KENYA
E.K. Maranga
National Range Research Station, Kiboko
P.O. Box 12, Makindu, Kenya
Abstract
Seasonal patterns of transpiration, leaf conductance and leaf
water potential of herbaceous and shrub plant species in semi-
arid savanna ecosystems were determined over a growing season.
Transpirational use of water for the shrub species ranged between
20 micrograms/cm~ S and 30 micrograms/cm S~ . The deep-
rooted shrub species maintained favourable leaf water balance and
remained green over the entire growing season. The water
response curves for the shallow-rooted plants indicated that they
had little control over their water economy. Environmental
processes leading to bush encroachment seem to be closely related
to water response patterns of semi-arid savanna ecosystems in
Kenya's pastoral grazing lands.
INTRODUCTION
Savanna grasslands in Kenya occupy areas of marginal rainfall
that constitute over one half of the total area of the country
(Pratt et al . , 1966). Water availability is the major driving
variable controlling biological activity and regulating
productivity of savanna ecosystems (Walker et al . , 1981; Maranga
et al., 1983; Maranga, 1986). Rainfall events are sporadic and
often occur in irregular pulsations of short-lived storms
(Griffiths, 1972). Effective rainfall recharges soil moisture
for a specific period of time. The time elapsing between a
change in soil-water status and an observed change in a plant
process constitutes the response period (Sala et al. , 1982). The
response period of water-controlled processes is not fixed but
is, at least partially, a function of the degree of water stress
to which plants were previously exposed. The efficiency of
utilisation of water input will decrease as the response period
increases and therefore will depend on the water status (leaf
water potential) of the plants prior to a rainfall event (Sala et
al., 1982).
This paper attempts to evaluate water responses of native
shrub and grass species to fluctuating soil water in the semi-
arid lands of Kenya. Plant variables used in this study to gauge
plant water responses include leaf water potential and
transpiration both of which are important determinants of plant
productivity in semi-arid areas. The adaptive mechanisms that
explain the observed responses will be deciphered.
MATERIALS AND METHODS
Data and conclusions presented here come from field experiments
conducted at the National Range Research Station, Kiboko. Mean
annual rainfall is 600 mm based on over 70 years of data from
365
Makindu Meteorological Station located near the research station.
Rainfall is bimodally distributed. The "long" rains last from
March to May; the "short" rains begin in or around October and
end in December.
Detailed descriptions of the vegetation physiognomy of the
study area are documented by Michieka and van der Pouw (1977) .
The vegetation of the experimental area can be described
generally as open-wooded grasslands. There are scattered trees
of Acacia seyal. Acacia tortilis and Commiphora riparia. There
is a patchy understorey of shrubs of Grewia sp. , Duosperma sp.
and other genera scattered in a matrix of grassland characterised
by Themeda triandra, Panicum maximum, Digitaria macroblephara and
Chloris roxburghiana .
Field studies of soil water potential distribution with
depth on sites that were dominated by Digitaria macroblephara,
Panicum maximum. Acacia tortilis. Commiphora riparia, Grewia
villosa and Hermania alhensis were carried out by means of a
modified version of Spanner's (Spanner, 1951) Peltier
thermocouple psychrometer installed at 10, 30, 60, 90 and 120 cm
depths. The Peltier psychrometer system consisted of a
microvoltmeter system, a control unit to supply current for
Peltier cooling and a thermo couple psychrometer. The Hewlett
Packard Model 419A (Hewlett Packard Englewood, Colorado) dew
point microvoltmeter was used in conjunction with the
thermocouple psychrometers calibrated for the five depths.
Transpiration and diffusive resistance were measured by means of
a steady state porometer (Li-Cor LI-1600 diffusive resistance
porometer) between 10.00 a.m. and 2.00 p.m.
RESULTS
Effective rainfall contribution recharged the dry soils
increasing soil water potentials from -10 MPa in early December
to field capacity conditions by mid-December in the surface
depths of 10 cm and 30 cm. Soil water potential in the deeper
depths of 60-120 cm oscillated between approximately -7 MPa and -
3 MPa over the same period (Figure 1) . The wetting phase in the
short rainfall season maintained field capacity conditions in the
surface depths for a period of approximately three weeks. As
soil water input declined, a sharp drop in soil moisture
potentials occurred for the shallower soil profiles with only a
gradual decline for the deeper soil horizons.
Leaf water potentials for the shrub and herbaceous plant
species (see Figure 2) in mid-December were characteristic of
non-water deficit conditions. In the course of the wetting
phase, leaf water potentials barely dropped below -4 MPa.
Thereafter, all other plant species except Commiphora riparia
exhibited fluctuating trends of leaf water potentials indicating
a cessation of soil water input coupled by soil water depletion
to meet transpiration demand. The lowest leaf water potential of
-7 MPa was recorded for Digitaria macroblephara and Grewia
villosa between January and February. Commiphora riparia
maintained leaf water potentials at levels greater than -2 MPa
throughout the growth cycle. Leaf water potentials for Acacia
tortilis fluctuated between -4 MPa and -1 MPa over the same
366
oo
a.
< ,
fO'
.e cvj
u
o
2 _.
&2H
J3
c
tl
o
o.
u
SO
in ji
E E E E E
o u o o o
o o o o o
— rO 10 <xi CJ
 
T
I 2" £- fr- S- 9-
(°dW ) loiiuatod J3JDM |ios
z- 8" oi-
367
a.
*
o
U
V
H
j3
o1
-I
1
p
DO
 
t—r
fi V £
(08S/W3) SDUDpnpuOO
1 1 1 1 1 1 1 l l l
1-2- E- fr- S- 9- I- 8" 6-0i-
(OdW) lDljU«40d ja{OM jd«-l
368
period. Stomatal conductance exhibited a pattern that was
approximately inversely related to leaf water potential at least
for the herbaceous plant species studied. Stomatal conductance
values peaked in December coinciding with the phase of active
soil water replenishment.
Transpiration rates of the shrub species in the course of
the wetting phase ranged between 20 micrograms/cm S-1 and 30
micrograms/cm S whereas that of Digitaria macroblephara rarely
exceeded 20 micrograms/cm S_1 (see Figure 3) .
369
u
0
p.
o
11
a
»)
>-.
d
o
--i
-s
IIIIIIl
o E
o
u
o E
in
in
o O
CVJQ. o
.2 c a
o a> -Q i i
Q. O o o
a.in o
in
O
<
u
u
o
a.
O E k.
O o '>
o
c
*-
■&
c
o
o o
o E 'i £ o
* *
'o
o E 4) k. '5>o o w 0)
5
o o
< a e> I 'A V)
 
-i 1 1 1—i—i—i—r
OS 0t> 0£ 02 01 0 I- 2- C-fr- S- 9- Z- 8- 6-01-
(oesujo/firt) uo^ojidsuDJi (OdW) IDijuojod jsjdm nos
370
DISCUSSION
In the course of the wetting phase, water was available within
the rooting zone and all species maintained a favourable water
balance. However, as soil water depletion continued to meet the
rising evapotranspiration demand as the dry season advanced,
moisture supplies were exhausted from the top soil profiles.
Shallow-rooted Digitaria macroblephara and Grewia villosa
registered unfavourable water balance. The deep-rooted Acacia
tortilis appeared green for the entire growth cycle indicating
that water was available within its rooting zone. Commiphora
riparia drops most of its leaves and maintains a succulent stem
in the course of the dry season. In this study, Commiphora
riparia maintained a favourable stable leaf water potential
suggesting an efficient regulation of its water economy
facilitated by water conservation mechanisms. Feinner (1981) in
a study of two Commiphora sp., Adonsonia digitata and Sterculia
rhynchocarpa reported that these species are relatively efficient
in conserving their water content.
The specificity in water use patterns of semi-arid savanna
ecosystems places important restrictions in their responses to
management interventions. Excessive removal of herbaceous
biomass by grazing animals would influence the total water input
reaching the various soil profiles. Walker et al. (1981)
reported that on medium to heavy textured savanna soils,
decreases in infiltration rates resulting from capping of the
soil surface pores were associated with a decline in grass and
litter cover.
In most of the Kenyan pastoral grazing lands, excessive
depletion of grass cover through overgrazing has resulted in a
serious problem of bush encroachment. The environmental
processes leading to bush encroachment seem to be closely tied to
the differential water use patterns associated with morphological
characteristics facilitating exclusive use of water available at
different soil profiles.
CONCLUSIONS
Soil water potential distribution in the soil profiles of semi-
arid ecosystems is closely related to the nature of precipitation
events and soil water depletion characteristics. The soil
moisture response curve is characterised by rapid oscillations in
the levels of soil water potential in the top soil profiles.
Gradual declines in soil water potential occur with increase in
depth and progressive advancement of the dry season.
Shallow-rooted plants like Digitaria macroblephara and
Grewia villosa tend to take exclusive advantage of water
availability in the top soil profiles when precipitation events
occur. As the dry season progresses, these plants quickly show
unfavourable water balance suggesting little control of their
water economy.
In general the seasonal patterns of transpiration, leaf
conductance and leaf water potential suggested that the shrub
371
species utilised
plant species.
more water per unit area than the herbaceous
ACKNOWLEDGEMENTS
I wish to express my appreciation to the technical support staff
of the National Range Research Station, Kiboko for assistance in
field work. Professional advice rendered by Drs. Dennis Child
and M.J. Trlica of Winrock International is sincerely
appreciated. This work was supported by funding from USAID and
the Government of Kenya.
REFERENCES
Feinner, M. 1981. Water loss from cut shoots of arid bushland
species in Kenya. Kenya Journal of Science and Technology
(B) 2:9-13.
Griffiths, J.F. 1972. Climate.
Its peoples and resources.
In: W.T.W. Morgani East Africa.
Oxford University Press, London
Maranga, E.K., Trlica, M.J. and Smeins, F. 1983. Water relations
of Panicum maximum and Digitaria macroblephara on a semi-
arid rangeland in Kenya. E. Afr. Agric. For. J. 48:74-80.
Maranga, E.K. 1986. An ecological perspective of the
phytosociology of woody - Acacia herbaceous - understorey
complex. Proceedings of a Conference on Range Development
and Research in Kenya, Egerton College, Njoro, Kenya.
Winrock International Institute for Agricultural
Development.
Michieka, D.O. and van der Pouw, B.J. A. 1977. Soil and
vegetation of the Kiboko Range Research Station, semi-
detailed soil report, No. 53, Draft Edition. Kenya Soil
Survey, NAL, Ministry of Agriculture, Nairobi, Kenya.
Pratt, D.J., Greenway, P.J. and Gwynne, M.D. 1966. A
classification of East African rangelands. J. Appl. Ecol.
3:369-382.
Sala, O.E., Lauenroth, W.K. and Parton, W.J. 1982. Plant
recovery following prolonged drought in a short-grass
steppe. Agric. Meteorol. 27:49-58.
Spanner, D.C. 1951. The Peltier effect and
measurement ofsuction pressure. J. Exper.
its use
Botany 2:
in the
145-168.
Walker, B.H. , Ludwig, D. , Holling, C.S. and Peterman, R.M. 1981.
Stability of semi-arid savanna grazing systems. Journal of
Ecology 69:473-498.
372
NUTRITIONAL VALUES OF THE DIETS OF MAASAI CATTLE
P.P. Semenye
P.O. Box 252, Maseno, Kenya
Abstract
The most grazed grass species were Chloris roxburghiana,
Digitaria macroblephara and Pennisetum mezianum. Season, area
and the interaction between area and season were significant
(P<0.05) sources of variation for the diet nutritive values. The
least squares means of the diet nutritional values were: 8.1, 69,
49, 0.26, 0.5 and 0.22% for CP, NDF, IVDMD, P, Ca, Kg, and 1703,
6.2, and 67 ppm for Fe, Cu and Mn respectively.
INTRODUCTION
The performance of pastoral cattle is usually described as poor.
For instance, weaning weights of calves are less than 65 leg
(Nicholson, 1983; Diallo et al . , 1981), calving intervals are
longer than eighteen months (Wagenaar et al., 1986; Otchere,
1986) and mortality rates of calves in some years are as high as
50% (Williamson and Payne, 1978; Mahadevan, 1966). However, it
is known that the performance of cattle is a function of the
breed, forage intake and the environment, including management
and climatic factors. In turn forage intake by cattle grazing
extensive rangelands is influenced by the age and physiological
status of the forage on offer. In extensive production systems,
of the three factors determining intake, forage quality is of
greatest biological importance as compared to the others, because
man has the least influence. This is so because forage quality is
determined by climatic and soil conditions. To measure the
forage quality and its implications on forage intake in a
pastoral system a research was initiated by the International
Livestock Centre for Africa (ILCA) as part of a three-year study
of the Maasai livestock production. It comprised three group
ranches with five grazing areas, located in the eastern area of
Kajiado District in Kenya.
MATERIALS AND METHODS
Oesophagally fistulated (OF) cows were used for forage sampling.
Sampling was done twice in a day to coincide with AM and PM
grazing peaks. Sampling during grazing peaks were preferred in
order to minimise contamination of the extrusa with regurgitated
material. During each sampling session three OF cows were used
and extrusa of approximately one kg was collected at each
sampling from each OF cow. The collection period averaged 25
minutes. Extrusa samples collected in wet seasons were
transferred into plastic bags and placed in a cool-box containing
dry ice before being shipped to Nairobi for oven drying. In dry
seasons the extrusa samples were first placed on wire mesh trays
for sun drying. All extrusa samples on reaching Nairobi were
dried in a Memmert oven at 60° C. Before milling by a Christy
mill through a one-mm screen, the samples were fractionated into
whole sample, leaves, and sheaths together with stems.
373
Laboratory analysis consisted of Kjeldahl, fibre analysis
(Goering and van Soest, 1970) , dry-matter solubility using
double enzyme method of pepsin and cellulase, and atomic
absorption spectrophotometry and flame photometry for minerals.
The five-minute method was used to mimic grazing of the
cattle. At the end of each five-minute period the species of
grass or browse grazed was identified and recorded. Six areas and
four seasons were studied.
For statistical analysis SPSS and least squares procedure of
Harvey (1977) were used.
Botanical Composition of the Diet
Over the 84 cow days, 74 different species of herbs grasses and
shrubs were foraged by cattle. The observation period covered
four seasons and six areas. Since one of the seasons represented
the peak of the drought of 1984, the Kiboko area was included in
this season because the majority of the cattle were moved there.
It was claimed the area was endowed with more herbage than the
home grazing orbits. Tables 1 and 2 show botanical compositions
by season and area respectively.
The dominant species in the diet were Chloris roxburghiana,
Pennisetum mezianum and Digitaria macroblephara compromising at
least 50% of the diet during the four seasons of 1983.
Prominence of the first two species persisted even in the drought
of 1984.
In the dry season cattle did not browse but grazed on other
species more and for the first time on the litter. Compared with
the dry season of 1983, in the drought season Chloris
roxburghiana, Pennisetum mezianum, Digitaria macroblephara,
Aristida keniensis, Bothriochloa insculpta and Cenchrus ciliaris
were grazed less frequently and Cynodon plectostachyus,
Sporobolus fimbriatus, Chrysopogon aucheri, Setaria sphacelata,
browse, litter and others more frequently.
As for the seasons of 1983, the three main species in the
diet of cattle in all five areas of sampling were Chloris
roxburghiana, Pennisetum mezianum and Digitaria
macroblephara. The rest of the diet varied as certain species
were either dropped or included from respective areas. The
average diet composition of the five areas was very different
from Kiboko ' s where the cattle were sent during the drought of
1984. For Kiboko Pennisetum mezianum and Digitaria macroblephara
ceased to be one of the three major species. Instead there came
browse species and the litter with the former contributing 34% of
the diet. For this area the diet consisted of less than 65%
grass.
374
Table 1. Composition of diet selected by Maasai cattle by season
(%).
Season Dry/green Green Dry Mean Drought
Feb/Mar May/June Sept/Oct for July/Aug
Year 1983 1983 1983 1983 1984
Number of
cow days 21 28 12 20 23
Chloris
roxburghiana 20 24 19 21 13
Pennisetum
mezianum 18 18 20 19 18
Digitaria
macroblephara 20 18 11 16 7
Aristida
kenyensis 8 7 4 6 NG
Cynodon
plectostachyus 7 8 2 6 4
Sporobolus
fimbriatus 10 6 NG 5 5
Chrysopogon
aucheri 3 4 6 4 7
Bothriochloa
insculpta 3 NG 8 4 3
Cenchrus
ciliaris NG 4 2 2 NG
Setaria
sphacelata NG NG NG NG 3
Browse 3 5 NG 3 4
Litter NG NG 15 5 17
Others 8 6 13 9 19
NG not grazed.
The diet preference index (Taylor, 1973) for the three
major species ranged from 1.4 to 3.6, and their ground cover
within the grazing orbits was more than the other species.
Availability of these species was important in the determination
of their selection. It should, however, be noted that these
species are of high palatability compared with the others. Their
selection, particularly for Digitaria macroblephara and Chloris
roxburghiana, was due to availability and palatability while that
of the less palatable Pennisetum mezianum was due only to
availability.
375
Table 2. Composition of diet selected by Maasai cattle by area
Kavati 01- Ilkelu- Oiti Isupa- Mean Kiboko
Area karkar nyeti nani drought
Number of
cow days 16 8 12 12 13 12 9
Chloris
roxburghiana 13 21 21 30 17 20 17
Pennisetum
mezianum 18 11 25 18 25 19 NG
Digitaria
macrobl ephara 20 18 8 23 10 16 2
Cynodon
plectostachyus 9 12 5 3 4 7 NG
Sporobolus
fimbriatus 13 10 3 NG 4 6 NG
Aristida
kenyensis 2 NG 8 11 9 6 NG
Chrysopogon
aucheri NG NG 3 1 10 3 15
Bothriochloa
insculpta NG 3 4 4 NG 2 10
Cenchrus
ciliaris 2 6 NG NG NG 2 NG
Ischaemum spp 2 2 NG NG NG 1 NG
Setaria
sphacelata 2 NG NG NG NG 0. 4 NG
Browse NG 3 2 7 2 3 34
Litter 6 4 9 NG 6 5 17
Others 13 10 12 3 13 10 5
Total 100 100 100 100 100 100 100
NG - not grazed.
Diet Quality
Oesophagally fistulated cows were used to determine forage
quality by season and the study area. Analysis of 126 samples
for crude protein (CP) , neutral-detergent fibre (NDF) , in-vivo
dry-matter digestibility (IVDMD) , phosphorus (P) , calcium (Ca) ,
magnesium (Mg) , iron (Fe) , copper (Cu) and manganese (Mn) are
presented in Table 4. Season, area and the interaction of season
and area (except for two cases of P and Mg) were significant
sources of variation (P<0.05). Differences between times of
sampling were consistently non-significant for any of the
variables. Differences among grazing orbits and cows showed no
consistent trend. The interaction of time by season was
significant (P<0.05) for CP only.
376
Table 3 . Least squares means of nutritional values of Maasai
cattle diet.
N CP NDF IVDMD P* Ca Mg
%%%%%%
Fe
ppm
Cu Mn
ppm ppm
Overall
mean 126 8.1 69 49 0.26 0.5 0.22 1703 6.2 67
Season
Feb/Mar'83
Green/
dry =5 33
May/June' 83
Green =6 45 10.7
Sept/Oct'83
Dry =9 48 5.5
8.1a 68a 50a 0.26 0.4a 0.23a 1637a 6.5a 116*
68c
711-
531- 6.4C
45c 0.27 0.5C 0.17c 2351° 5 . 6C
38*
48^
Area
Kavati 21
Olkarkar 26
Ilkelunyeti 21
Oiti 31
Isupanani 27
10.1" 70
1.7*- 70
7.4f
a 51a 0.31a 0.5a 0.23a 2214a
51a 0.26b 0.4b 0.21ab 2002a
a 47b 0.24° 0.5a 0.24a 1536b
51a 0.28b 0.5a 0.20b 1242b - . -
ac 48b 0.21d 0.5a 0.24a 1523b 5.8°
89f
6.3" 73"
5.4.c 58c
c
c58"-
abed Means having no letter in common differ at P<0.05.
1 Corrected for saliva.
Table 4. Least squares means of nutritional values of Maasai
cattle diet: whole sample, leaves, and stems and
sheaths.
Vari-
able N CP NDF IVDMD Ca Mg Fe Cu Mn
wS 126
L 105
SS 104
8.1=
7.8£
5.7C
69 =
70s
77c
49°
42fc
0.26a 0.5a 0.22a 1651a 6.2!
a 0.20b 1742a 5.9'0.241-
0.18c
0.5
0.31 0.151- 955b 4.5--
67 =
63 =
53s
abc means having no letter in common differ at P<0.05.
% - N, CP, NDF, IVDMD, Ca, Mg; ppm - Fe, Cu, Mn.
WS - whole sample; L - leaves; SS - stems and sheaths.
Nutritive Values of Diet Fractions
Least squares means of nutritional values of Maasai cattle forage
fractions are presented in Table 5. In all cases the analyses of
the whole sample (WS) and leaves (L) were significantly higher
than those of stems and sheaths (SS) .
The WS was higher than L by 0.3, 1.0, 0.02 and 0.02% in CP,
NDF, P and Mg content respectively (P<0.05). Other differences
between WS and L in IVDMD, Ca, Fe, Cu and Mn were not significant
(P>0.05) .
377
Table 5. Least squares means of nutritional values of Maasai
cattle diet: fraction x season.
Season N CP NDF IVDMD Ca Mg Fe Cu Mn
Green/drv
WS 33
Leaf 33
SS 28
Green
WS
Leaf
SS
Dry
WS
Leaf
SS
8.2°
7. 5*
5.9C
45 10.6°
40 10.0?
37 6.8*
48
36
39
5.6°
4.4*
68°
7 6r
68 =
71E
78c
71°
71a
78c
5°t49£
42c
54°
54"
46
44"
46C
38c
0.26°
0.23*
0.19c
0.26°
0.231
b 0.14c
0.27s
0.25?
0.20*
0
4h
0 23a 1434a 6 4a 114
0 5b 0 2lf 1767a 6 of 98
0 3C 0 18b 681b 4 8b 71
0 6a 0
27b3
1097 6 4a 38
0
6b3
0 1089 6
3b2
42
0 0 16° 808 5 39
0 5a 0 17a 2421a 5 7a 47
0
5b3
0 17S 2370a 5 2f 49
0 0 llb 1375b 3 4b 48
WS - whole sample.
SS - stems and sheaths.
abc = means within the same season having no letter
in common differ at P<0.05.
% - N, CP, NDF, IVDMD, P, Ca, Mg, PPM - Fe, Cu, Mn.
1 = corrected.
The L fraction exceeded SS nutritively by 2.1, 7, 8, 0.06,
0.2 and 0.05%, 787, 1.4 and 10 ppm or by 26, 10, 16, 23, 40, 23,
48, 23 and 15% for WS in CP, NDF, IVDMD, P, Ca Mg, Fe, Cu and Mn
respectively.
For the seasons represented, the Fe concentration followed a
different trend from the other elements for the green season, as
differences between the fractions were not significant
(P>0.05). The peculiar trend for the green season was probably
because the Fe analysed in plant fractions was partly endogenous
and partly exogenous arising from soil contamination and came
from dust and soil ingestion during grazing; the green season
dust was minimal and cattle did graze close enough to the ground
to ingest soil or dust. In all seasons the SS fraction had the
lowest concentration of Fe due to its smaller surface area on
which dust could land.
DISCUSSION
In the normal seasons of 1983 three grass species dominated. The
three species, Digitaria macroblephara, Chloris roxburghiana and
Pe/inisetum mezianum were preferred probably by virtue of being
perennial, and having greater ground cover than other grass
species. For the first two species Kibet (1984) has advanced
similar reasons for their selection by cattle. Another
stabilizing factor between the seasons' diets was the bimodal
rainfall received. In the extreme conditions that prevailed in
the drought of 1984 changes were forced in the diet selection.
378
Cattle selected more of the less palatable and dead grass as
litter. For the seasons by area, species composition varied
little, but in contrast the quality of the diet was influenced
much, due to environmental differences of rainfall, soils and
ambient temperatures. The decision of the pastoralists to move
their cattle to Kiboko area was wise nutritionally. The cattle
browsed more and, presumably, got sufficient N to enhance their
utilisation of the litter and the less palatable species that
were available.
The mean and the seasonal variation of CP was consistent
with what has been reported of clipped and OF samples from Kenya
(Karue, 1974; Potter, 1981; Reed, 1983; Tessema and Emojong,
1984) . This means the problem of rapid decline in CP is not an
isolated case for the Maasai rangelands alone. This is probably
explained by the high temperatures in the growing seasons that is
characteristic of Kenyan rangelands, which accelerates plant
maturation leading to lignification. If protein supplementation
is contemplated in these areas, for maximum return, it should be
restricted to September/October when the mean value of CP was
lowest (5.5%) .
The highest value of HDF (73%) obtained in this study is
equal to the mean value of Tessema and Emojong (1984) and lower
than 79% of Reed (1983) from comparable rangelands. These
differences may have occurred due to sampling technique of OF
versus clipping.
Digestibility is known to limit voluntary forage intake
(Campling et al . , 1962; Blaxter and Wilson, 1962; Karue et al . ,
1973) , and within certain limits there is a positive correlation
between digestibility and intake. This was shown by Elliot and
Fokkema (1961) over a range of digestibilities of forage from 42
to 64%. The digestibility of Maasai cattle forage fell within
this range (42-59%) , and it may be assumed that their voluntary
intake was limited by low digestibility.
Except for Ca and Fe all other elements investigated were
only marginally above the critical level. For increased
productivity of cattle in this production system serious
considerations must be given towards supplementation with mineral
salts. For P and Cu, apart from being marginally above the
critical value, their availability is hampered by other factors
prevalent in the study area. With regards to P, the Ca:P ratio
does influence availability and utilization. Ratios of Ca:P of
above 2:1 could limit availability and utilization of P (Conrad
et al . , 1982) and such ratios were observed in certain seasons
and areas of the present study. On the other hand Cu
requirements for livestock increases where Mo concentrations in
the diets is high. In the study area Mo is high as reported by
Whittington and Ward (1983) . This implies there is need for Cu
supplementation in order to cover the shortfall from marginal
concentration in the forage and extra requirements triggered by
excessive concentration of Mo.
Differences in the nutritive values of the L and SS
fractions were large, and the diets were made-up of 80 to 89% L.
This means cattle selected the most nutritive parts of the forage
available. Thus when OF samples and clipped samples are taken at
379
the same, time the former commonly exceeds the latter in
nutritive value (Pratchett et al . , 1977; Holechek et al . , 1982).
However, the nutritive edge of WS over L was unexpected, because
the former contained SS fraction which had the lowest values.
These results must have been influenced by the high content of L
in WS. However, the edge in the nutritive value of WS over L was
probably due to the fact that the most nutritive part of a plant
at any time is the growing point be it stem or leaf. As the
growing points of the stems and sheaths were part of the WS, they
contributed positively to its nutritive value. The same happened
to the SS fraction, but due to its greater proportion of
senescent stems and sheaths, the growing points contribution was
insignificant. Lambourne and Reardon (1962) reported the
digestibility of stem was initially higher than that of leaf but
falling more rapidly after ear emergence. So the longer any part
of the plant, in this case grass, remains in the growing stage,
the better. Wilson (1983) cited 24 authors whose work showed
that when herbage was stressed by lack of water, digestibility
either increased or showed no change. Moisture stress slows
growth and hence delays senescence.
CONCLUSIONS
Forage intake by Maasai cattle was limited in the dry seasons by
low values of crude protein and dry matter digestibility and high
values of neutral detergent fibre. In addition to the low
nutritional value of the diet being a limiting factor to intake,
the dry seasons were characterized by scarcity of forage on
offer.
Supplementation with phosphorus, magnesium and copper is
recommended during the dry seasons because of low forage
intake. In green seasons supplementation is not necessary at the
current production level of the cattle. On the other hand iron
level in the diet was extremely high. As this high level (of
over 300% above the recommended level) could be toxic, there is
need for investigations on its effect and its interaction with
other elements.
Digitaria macroblephara and Chloris roxburghiana contributed
greatly to the diet of Maasai cattle. Due to the fact that they
are perennials of high palatability, drought resistance and
nutritive value, they are ideal species for either introduction
or re-seeding in arid and semi-arid areas.
ACKNOWLDGEMENTS
The author wishes to thank ILCA for funds supporting this
research; the pastoral ists for allowing me to study their cattle;
Drs. A. Carles, A. Said, J. Lambourne and Mr. de Leeuw for
advice and guidance; Messrs. P. Chara, E. Rugema, T. Tukai and E.
Wacira for their good work as field assistants and Ms. P. Pacha
for typing the manuscript.
380
REFERENCES
Blaxter, K.L. and Wilson, R.S. 1962. The voluntary intake of
roughages by steers. Anim. Prod. 4: 351.
Campling, R.C., Freer, M. and Balch, C.C. 1962. Factors
affecting the voluntary intake of food by cows. 3. The
effect of area on the voluntary intake of oat straw. Brit.
J. Nutr. 16:115.
Conrad, J.H., McDowell, L.R.M. and Loosli, J.K. 1982. In:
Mohamed K. Yousef (ed.). Animal production in the tropics.
Praeger.
Diallo, A., Wagenaar, K. and de Leeuw, P.N. 1981. Livestock
productivity and nutrition in systems research in the arid
zones of Mali. ILCA Research Report No. 5. ILCA, Addis
Ababa.
Elliot, R.C. and Fokkema, K. 1961. Herbage consumption studies
on beef cattle. Rhod. Agr. J. 58:49.
Goering, H.K. and van Soest, P.J. 1970. Forage fiber analyses
(apparatus, reagents, procedures and some applications) .
Agriculture Handbook No. 379. United States Department of
Agriculture. Washington, D.C.
Harvey, W.R. 1977. User's guide for least-squares and maximum
likelihood computer program. Ohio State University,
Columbus, Ohio.
Holechek, J.L. , Vavra, M. and Pieper, R.D. 1982. Methods for
determining the nutritive quality of range ruminant diets.
A review. Journal of Ani. Sci. 54:363.
Karue, C.N. 1974. The nutritive value of herbage in semi-arid
lands of East Africa. 1. Chemical composition. E. Afr.
Agric. J. 40:89-95.
Karue, G.M. , Evans, J.L. and Tillman, A.D. 1973. Voluntary
intake of dry matter by African zebu cattle: Factors
influencing dry matter intake. E. Afr. Agr. For. J. 38:352.
Kibet, P.K. 1984. Influence of available browse on cattle diets
in an Acacia savannah of East Africa. M.Sc. thesis, Texas
A and M University, Texas.
Lambourne, L.J. and Reardon, T.F. 1962. Use of 'seasonal'
regression in measuring feed intake of grazing animals.
Nature 196:961.
Mahadevan, P. 1966. Breeding for milk production in tropical
cattle. Technical Communication No. 17, Commonwealth Bureau
of Animal Breeding and Genetics, Edinburgh.
Nicholson, M.J.L. 1983. Calf growth, milk offtake and estimated
lactation yields of Borana cattle in the southern rangelands
of Ethiopia. Joint ILCA/RDP Ethiopian Pastoral Systems
Study Programme. JEPSS Research Report No. 6.
381
Otchere, E.O. 1986. Traditional cattle production in the
subhumid zone of Nigeria. In: R. von Kaufmann, S.Chater and
R. Blench (eds) , Livestock systems research in Nigeria's
subhumid zone. Proceedings of the second ILCA/NAPRI
symposium held in Kaduna, Nigeria, 29 October-2 November
1984, ILCA, Addis Ababa. pp. 110-140.
Potter, H.L. 1981. Strategies for the settled pastoralists.
Proceedings of the Animal Production Society of Kenya 13:18.
Pratchett, D. , Capper, B.G., Light, D.E., Milles, M.D.,
Rutherford, A.S., Rebbue, T.W., Buck, N.G. and Trail J.C.
1977. Factors limiting liveweight gain of beef cattle on
rangeland in Botswana. Journal for Range Management 30:442.
Reed, J.D. 1983. The nutritional ecology of game and cattle on
a Kenyan ranch. Ph. D. thesis, Cornell University, Ithaca,
New York.
Taylor, B.R. 1973. The botanical composition of cattle diets on
a 7-pasture, high-intensity, low-frequency grazing system.
M.Sc. thesis, Texas A&M University, Texas.
Tessema, S. and Emojong, E.E. 1984. The influence of stocking
rates and grazing management on liveweight change in cattle,
sheep and goats grazing natural pastures. East African
Agricultural and Forestry Journal, special issue 44:383.
Wagenaar, K.T., Diallo, A. and Sayers, A.R. 1986. Productivity
of transhumant Fulani cattle in the inner Niger delta of
Mali. ILCA Research Report No. 13. ILCA, Addis Ababa
Whittington, D.L. and Ward, R.G. 1983. Mineral availability of
eleven grasses and two shrubs collected during the dry and
wet season at N.R.R.S. Kiboko. Proceedings of the second
Small Ruminant CRSP, Kenya Workshop.
Williamson, G.^and Payne, W.J. A. 1978. Animal husbandry in the
tropics . Longmans , London .
Wilson, J.R. 1983. Effect of water stress on in vitro dry
matter digestibility and chemical composition of herbage of
tropical pasture species. Aust. J. Agric. Res. 34:377.
382
LIVESTOCK FEEDING POTENTIALS, CONSTRAINTS
AND POSSIBLE IMPROVEMENTS IN THE SEMI-ARID AREAS
OF CENTRAL TANZANIA
A.J. Kitalyi and M.A. Kabatange
Department of Animal Nutrition, Tanzania
Livestock Research Organization P/Bag, Mpwapwa
Abstract
Based on a diagnostic survey carried out in Mpwapwa District, the
main features of the farming system are presented. This is
followed by a description of the grazing lands which supply the
major feed component of the ruminant population. The existing
grazing pattern and farmer feeding strategies over the seasons
are discussed.
The farmers are agropastoralists producing mainly for
subsistence. Livestock sales is the main source of cash income
which in turn is spent on purchase of grain. There are very low
investments in agricultural production resulting in low
production coefficients.
Despite the uneven distribution and poor quality of feed
resource, farmers do manage to take most of their stock through
the dry season by employing various strategic measures. Loose
grazing of animals in the crop fields after harvest is the main
form of supplementation (60% of the respondent n = 152) .
However, it is evident that poor animal nutrition in the dry
season is a serious problem in the livestock production
subsystem. The causes of the problem are discussed and research
opportunities are highlighted.
INTRODUCTION
Central Tanzania covers an area of 140,000 km in the central
semi-arid plateau, between latitudes 4° and 7°S and longitudes
33° and 37°E (Figure 1) . The plateau lies at an elevation of
750-1500 m, the eastern boundaries of which are steeper
escarpments rising to 1750 m above sea level. The area has a dry
savannah type of climate with a long dry season between April and
November, and a short wet season from November to April (Figure
2) . The total annual rainfall ranges between 250 mm to 750 mm
with an average of 600 mm. The distribution is very erratic and
reliability is low. Average temperatures are 20-25°C with a
diurnal range of 12°C. Apart from the extensive areas of
alluvium the two main soil types are a greyish sandy and red
sandy loam textured. Severe leaching and laterisation are a
common feature of the area. Natural vegetation consists of a
dense deciduous thicket with Commiphora and Acacia as the
dominant genera.
Typical of semi-arid areas most of the land is of marginal
agricultural productivity with the exception of small pockets of
high potential along the eastern boundaries. The dwellers are
agro-pastoral ists integrating crop production and livestock
383
Figure 1. Central zone of Tanzania (Bingida and Dodoma regions),
 
384
Figure 2. Average rainfall distribution (1974-1984),
region.
Dodoma
--■• Mpwapwa
Rainfall (mm) —■•—— Dodoma
160- • Kongwo
150 -
A
* Hombolo
140 - r
/ 1
1
\
lI \
130 - i
i
I*
i 1 ■,\\
120 -
'if \S
\ i
110- §\ \ i\ i
'/ '- \'\ ..-■ «./ \ t\ i
i i
100^ I
\ \
'"• \ i
\ i
\ i
9C -
'/;■' \ \
■ \ i
\ \ t
'- \ '
80 - 1 \
70- 1I
\ 4
60-
$
t4 i
w t
50- I/h
li40-
li30- / E
r fir
if?
20 - i{£
10 -
/Jf
o- 1 1— i i i i 1 1 T T — T -
Oct Nov Dec Jan Feb Mar Apr May June July Aug Sept
Month
385
keeping in different degrees depending on socio-economic and
cultural situations. According to the 1984 human population
census, there are over two million people of various ethnic,
groups namely: Gogo, Nguu, Hehe, Nyaturu, Nyiramba, Kaguru and
Rangi. All these groups are of Bantu origin with an exception of
a few pure Hamites, Maasai and Barbaig who have migrated from the
neighbouring regions. The population is sparsely distributed in
the area except for the few areas of higher soil fertility.
Adoption of improved technology in agricultural production
is minimal, and very low investments are put into the sector.
Consequently, production coefficients both in cropping and
livestock farming are low. Land productivity potential, however,
is higher for extensive livestock production than for cropping.
The area has over 15% of the national ruminant population. Moore
(1971) reported that whereas 88% of the land is used for grazing,
only 10% is used for cropping. Consequently, livestock is an
important resource which if well tapped can improve the standard
of living of the people.
The production coefficients of cattle in the area have been
reported to be generally low; calving rate of females over three
years old of 45%, calf mortality to one year of age of 12%,
mortality of other cattle of 10%, selling age of males of seven
years and a market offtake of 13% (Stockes, 1976) . Major
problems of this extensive livestock production system are
seasonal shortages of good quality feed and lack of acceptable
range management practices. However, given the nature of the
livestock production systems which produce several products and
use many resources, a thorough description of the system is a
pre-requisite to testing of any improvement strategies.
A team of scientists from the Tanzania Livestock Research
Organization (TALIRO) formulated a Farming Systems Research (FSR)
project entitled "Improvement of Animal Production Systems in
Central Tanzania". The main objective of the project was to gain
an understanding of the existing livestock farming system and
develop improved pasture management, livestock feeding, herd
composition and management practices that are acceptable to
farmers in the traditional sector of Central Tanzania (Anon,
1986) . This paper attempts to describe and discuss the livestock
feeding aspects of the study area as drawn from a diagnostic
survey of Mpwapwa District.
386
METHODOLOGY
The diagnostic survey work was carried out in the 1985/86 growing
season following the FsR methodology. Collation of background
information was followed by an informal survey which resulted in
stratification of the study area into three zones from which
villages and households for formal survey were selected. There
were three villages from each zone and 5% of the village
households were interviewed using a structured questionnaire.
The questionnaire was designed to cover all aspects of the
farming system e.g. socio-economic background, cropping pattern,
livestock species kept and livestock management practices (Anon,
1986) . More emphasis was placed on livestock enterprise.
Trained enumerators with certificate level of education in
agriculture were used. A total of 152 questionnaires were filled
in from the nine villages.
The data were coded, tabulated and analysed using hand
calculators.
RESULTS AND DISCUSSION
Main Features of the Farming System
From survey data, the main ethnic groups in the area were Gogo
44%, Kaguru 25% and Hehe 21%. Forty-one percent of the
respondents had been living in the area for over 2 5 years whereas
47% had moved in during the national villagisation "Ujamaa"
programme (197 0-1974) . It was revealed that the farmers in the
area were agro-pastoral ists with over 99% of the respondents
keeping livestock as well as engaging in crop farming.
The farm plots were fragmented, the highest number of farm
plots per household in the last growing season being six, and on
average the furthest farm plots were 4.63 km from the homesteads.
The average total hectarage was 5 ha, which may be considered to
be rather large given the low level of mechanisation involved.
Survey reports in other smallholder systems have reported farm
sizes of 3.23 ha and 2.4 ha for draft cattle owners and non-
owners respectively (Anon, 1982). A survey carried out in
Northern Tanzania indicated the majority of farmers to own 1.2 ha
of land (Urio and Mlay, 1984) . The tendency for farmers in
Central Tanzania to own large farm plots may be explained by the
relatively low labour requirement in the preparation of seed beds
in this area.
Ninety-two per cent of the respondents acquired their
cropping land by clearing bush, thus individual land ownership is
temporal . This reflects the existing land tenure system which is
further indicated by the farmers' responses on the adequacy of
grazing land. The majority of the respondents indicated that
grazing land was abundant. However, the responses may be biased
in view of the existing free movement of livestock from one area
to another without limitation.
As indicated by the secondary information, adoption or use
of improved technologies in agricultural production was found to
be negligible. Only 1% and 6% of the respondents had used
387
tractors and draft power respectively in the previous cropping
season. Agricultural production was mainly for subsistence.
Seven percent of the respondents said they would sell their
surplus yields, and the main source of cash income for the
majority (79%) was livestock sales. Hence, food ranked first
(35%) in the expenditure list, followed by clothing (32%) and
crop farming (18%) .
The livestock species kept were cattle, goats, sheep,
donkeys, pigs and poultry. On average one farmer kept 38 head of
cattle, 18 goats, 7 sheep, 4 donkeys and 17 chickens. The cattle
kept had been acquired through inheritance, exchange of
smallstock with grain, dowry or direct purchasing. There were
high variations in herd sizes across the area, the majority (76%)
having between 10 and 60 head of cattle. The herd sizes declared
by the farmers were lower than those reported in the literature
and from general observation. The discrepancies may have been
due to the well known reluctance of pastoralists to reveal such
information to strangers.
Grazing Lands and Feeding Strategy
Herding is the main method of supplying feed for the ruminant
population in the study area. The operation is carried out by
adult male members of the household. It was evident that herding
is a well-planned activity among the livestock keepers. Some
farmers indicated herding their animals collectively while others
did it individually. In most cases there is strategic herding to
combat cattle rustling which was indicated to be a common problem
among the livestock keepers (72% n - 152) .
Describing the grazing lands of the study area can be quite
difficult for researchers. For instance, there were no clear cut
boundaries between the cropping land and the grazing lands.
Moreover, after crop harvest the crop fields turn into grazing
lands. The herdsmen, however, could identify their specific
grazing areas, and grazing in crop fields was considered to be
feed supplementation. Follow-up observation on the grazing
pattern revealed that the individual or group herds were grazed
in specific areas. Grazing times are determined by the herdsmen,
basing it on the animal behaviour. Discussion with the herdsmen
revealed that they had a good knowledge of the rangeland, could
tell the most palatable species and that the choice of grazing
areas was based on their knowledge and experience.
The main grazing lands are in the secondary thicket which is
an open community that has invaded abandoned crop land as a
consequence of the shifting cultivation common in the area. It
is composed of a mixed association of grass species and herbs.
The common grass species were: Vrochloa trichopus, Dactyloctenium
sp. , Aristida sp., Eragrostis sp. and Chloris virgata. Perennial
grass species found in isolated areas were: Cynodon sp. , Cenchrus
sp. , Hyparrhenia sp. and Panicum sp. Identified annual herbs
included Astripomoea sp. , Crotalaria sp., Commelina sp., Cleome
sp. and Ipomea sp. The common perennial weeds were: Solanum
pandoroeforme, Sida grewiodes and Tephrosia incana while common
thicket species were Cassia sp. , Croton sp. and Acacia sp. Most
of the vegetation including weeds and shrubs were readily eaten
388
by all classes of stock, a feature common in most semi-arid areas
(Dyne et al . , 1984) .
The common grass species are generally known for their low
nutritive value. Work done by Karue (1974) indicated that most
of the grass species cannot meet the nutrient requirement of the
ruminant animals. He deduced that even the best of grasses
notably Cynodon dactylon and Digitaria setivala could only supply
about 30% of the protein required by the beef animal. This can
lead one to conclude that the shrubs, weeds and other forage play
a big role in the nutrition of the animals. In recognition of
this, herdsmen cut down branches of browse for easy access to the
animals.
Availability and quality of the feed resource fluctuates
during the year between the wet and dry seasons respectively.
While in the wet season forage is abundant, in the dry season it
is both scanty and poor in quality. Various strategies are
adopted by the herdsmen in overcoming feed shortage problems in
the dry season. Of main importance is the grazing in the crop
fields after the harvest to utilise crop residues. About 60% of
the respondents indicated doing so. A smaller proportion (40%)
indicated burning; leaving the crop residue in the field and
burning it when preparing land for the next growing season. In
practice most of the residue is utilised by the animals, for no
fencing was observed in the crop fields. Some farmers were
reluctant to allow grazing in crop fields allegedly because it
would result in reduced crop yield. Collection of crop residues
for feeding was not a common practice probably because of
limitations of transport.
In summary, crop residues play an invaluable role in dry
season for it is only when it is exhausted that herdsmen opt for
other strategies. These include movement of animals to distant
grazing lands for varying periods depending on the severity of
dry periods; this involves a daily trekking of animals to
distances as long as 6 km or more in search of pastures and
water.
RESEARCH OPPORTUNITIES
Poor animal nutrition particularly in the dry season came out as
one of the major constraints to livestock productivity in the
study area. This view is supported by the low milk production in
the dry season. Whereas, 1.4 litre per cow per day could be
produced in the wet season, milk production in the dry season was
negligible. The slow growth rates leading to attainment of
mature weights in seven years could partly be due to poor
nutrition also. Furthermore, movement of animals to distant
grazing areas in the dry season results from the feed shortages.
This problem can be described with the causes shown in ovals
diagrammatically (Figure 3) . There can be many causes to this
problem and some of the causes are interrelated. Three major
causes to the problem have been identified in the current study:
inadequate feed supply, inadequate grazing time and lack of
adequate supplementation.
389
6a
a
u
o
g.
3
o)
8o
CO
o«
CD
N
c CO
*o
o> CD
k. c
> o>
O m
 
390
The main causes could be further analysed to the root causes
which can be used to identify research opportunities. Some of
these could be difficult for researchers, for instance the big
herd sizes, the land tenure system, cattle rustling and water
shortage. These would require a closer relationship between
researchers, policy makers and other national development
agencies. On water shortage, improvement of the environment
through afforestation programmes and use of water harvesting
technologies could alleviate the problem. Water conserving
fruits e.g. watermelons are being used by some farmers for
feeding sensitive classes e.g. calves and sicklings. Wider use
of these could also reduce the problem but this has to be
directed to specific classes depending on production targets.
Various techniques for improving rangelands have been
suggested and some tested by range scientists. However, as
Nestel et al . , (1973) correctly puts it, "the major constraint
lies in introducing change into the existing socio-economic
systems, exacerbated by inexperience in adopting technology to
suit local conditions" . Exploitation of fodder trees and shrubs
might involve introduction of more drought resistant and high
quality species. The FSR team working on the research area is
trying to introduce browse spp. such as Desmanthus virgatus,
Leucaena leucocephala, Sesbania sesban, Sesbania spinosa and
Rhynchosia senaarensis. However, a parallel comparative study of
the nutritive value of the indigenous and introduced browse
species is worthwhile for the former is already adapted.
Information on the potential of the browse spp. as a forage
resource in the area is scanty and most of the research work on
this has not been conclusive (Kabatange et al., 1984). Its
establishment might raise some problems because of the existing
land tenure system. Furthermore, to introduce these as fodder
banks might involve transportation which will again require wider
use of draught power. As noted earlier, the use of draught power
was not common particularly for the Gogo ethnic group who own the
largest proportion of the cattle. Compassion is the main latent
reason for not using animals for draught power.
Improvement of the natural rangelands by introduction of
drought resistant better quality grass genera such as Cenchrus
and Cynodon would be a good alternative, but the problem of land
tenure would limit its success. However, such introductions
could be effected in the terminal year of cropping before the
farm is left fallow. Another alternative is to improve the
nutritive value of crop residues with minimal investment. This
has been a subject of discussion (with ILCA scientists probably
leading the way) , and a number of field experiments have given
positive results (Butterworth and Mosi, 1986; Saleem, 1985).
These experiments have been based on inclusion of annual legumes
in the crop fields. Of importance to consider here is the time
to undersow the legume in relation to grain yield and farmer
labour requirements.
The main grain legume in the study area was groundnuts.
This legume is harvested earlier than the cereal, and the method
of harvesting i.e. uprooting the whole plant, does not make it a
good legume for consideration. Furthermore, since the crop is
harvested when the shoot is green and it is piled up, in most
cases it gets mouldy and might not be a good supplement. Other
391
common legumes grown in the area are cowpea and Lablab spp.
Since the leaves of cowpeas are used as a vegetable, Lablab spp.
remain the best alternative. Thus, inclusion of this legume in
the cropping systems and its effect on crop yield, quality of
stover and livestock productivity could be investigated.
ACKNOWLEDGEMENTS
The authors wish to thank the PANESA 1987 workshop organising
committee for the invitation and permission to present the paper
at the workshop. We are thankful to the Director General of
tarilo for permission to attend the workshop. We are indebted to
IDRC and World Food Programme (Tanzania) for funding the FSR
Project. Finally we acknowledge the Livestock Production
Research Institute FSR project executing team for their
contribution to the work reported.
REFERENCES
Anon. 1982. Demonstrations of an interdisciplinary approach to
planning adaptive research programmes. Report No. 5. RSS,
Ministry of Agriculture, Zimbabwe.
Anon. 1986. Annual Report 1985/86. Livestock Production Systems
(Tanzania) Project. Tanzania Livestock Research
Organisation/IDRC.
Butterworth, M.H. and Mosi, A.K. 1986. The voluntary intake and
digestibility of combinations of cereal crop residues and
legume hay for sheep. ILCA Bulletin 24:14-17.
Dyne, G.M. van, Burch, W. , Fairfax, S.K. and Huey, W. 1984.
Forage allocation on arid and semi-arid public grazing
lands: Summary and Recommendations. In: Development
strategies for rangeland management. National Research
Council, Westview Press, Boulder, Co., USA.
Kabatange, M.A., Rukantabula, A. A. and Kitalyi, A.J. 1984.
Research experience on feeding of livestock in the dry
season at Mpwapwa. Proceedings of a seminar held at
Livestock Production Research Institute (LPRI) , Mpwapwa,
Tanzania, September 20-22, 1984.
Karue, C.N. 1974. The nutritive value of herbage in semi-arid
lands of East Africa. 1. Chemical composition. E. Afr.
Agric. J. 40:89-95.
Moore, J.E. 1971. Rural population carrying capacity. BRAIUP
Research Paper No. 18, University of Dar-es-Salaam, Dar-es-
Salaam, Tanzania.
Nestel, B., Pratt, D. J. , Thome, M. and Tribe, D.E. 1973. Animal
production and research in tropical Africa. Report of the
task force commissioned by the African Livestock Sub
committee of CGIAR. Washington D.C.
392
Saleem, M.A.M. 1985. Effect of sowing time on the grain yield
and fodder potential of sorghum undersown with stylo in the
subhumid zone of Nigeria. Trop. Agric. (Trinidad) 62(2).
Stockes, K.W. 1976. Livestock marketing existing situation and
recommendations for change. Proceedings of the Third
Scientific Conference of Tanzania Society of Animal
Production Mwanza, 17-19 May, 1976. Dar-es-Salaam,
Tanzania.
Urio, N.A. and Mlay, G.I. 1984. Dairy feeding systems (Tanzania)
progress report on diagnostic survey among small holder
farmers in Hai, Morogoro, Tanzania.
393
THE POTENTIAL OF PASTURE LEGUMES AND THEIR ROLE IN
IMPROVING FARMING PRACTICES UNDER
EXTENSIVE LIVESTOCK PRODUCTION SYSTEMS IN ZAMBIA
J . Kulich
Nanga National Irrigation Research Station
P/Bag S 3, Mazabuka, Zambia
Abstract
The development of improved pastures in Zambia is reviewed on the
basis of research and farmer experiences. Pasture research
programmes have provided much basic information about development
of pasture systems for both dairy and beef cattle. There is need
to know much more about how to maintain stable pastures, the
persistence of legumes, establishment of improved species,
maintenance fertilizer requirements and pasture management; more
information is needed on the economic aspects of these systems.
To increase production with minimal increase in costs seems
possible only with an integrated usage of improved and native
pastures.
INTRODUCTION
The most important constraints to improving livestock production
in Zambia are related to animal nutrition. The feed resources
mainly based on the natural pastures, are poorly managed
throughout the country. Due to the long dry season, there are
feed shortages so that animals lose weight, suffer high death
losses and exhibit low fertility. When animals are under so much
stress, they are also more susceptible to diseases. Overstocking
and overgrazing, which are very common in many grazing areas in
Zambia, have resulted in serious land degradation, the
disappearance of valuable pasture species and the spread of
unpalatable species. When natural veld pastures are developed
with a legume-base and fertilized mainly with superphosphate,
most of them can yield large quantities of low-cost animal
products. Legumes supply nitrogen cheaply for the associate
grasses and grazing ruminants, and their large scale use in the
future will ensure that adequate quantities of beef and milk will
be within the financial reach of the rapidly expanding population
of Zambia.
In many areas of Central, Southern and Lusaka provinces in
Zambia, inorganic nitrogen-grass systems are used for dairying
and intensive beef production, but with increasing costs and the
removal of subsidy and possible shortages of energy, there has
been created a potential instability in price/supply of
fertilizer N, thus making this practice uneconomical. There is,
therefore, a swing in these areas to legume-grass systems.
One of the most important facets of tropical pasture
research is to find legumes suitable to the Zambian environment.
Pastures based on these legumes have the potential for both
higher stocking rates and faster cattle growth rates which can
result in a significant increase in animal production.
394
PAST AND PRESENT SITUATION
The volume of pasture research, and hence of literature, on the
improvement of pasture production in Zambia, is very great, and
no attempt is made here to review it fully. There have been
other reviews by Prins (1970) , Craufurd (1979) , Kulich and Kaluba
(1985), Kulich (1985), Kulich and Nambayo (1985), and Kulich
(1986) . During the past 30 years there have been significant
contributions on pasture legume research notably by Smith (1962,
1963), van Rensburg (1967, 1969a, 1969b), Potter (1972), Verdoom
(1965), Prins (1972, 1975), Peterson (1975), Shalwindi (1978),
Craufurd (1978), J. and S. Kulich (1973, 1976) and by Kulich
(1977, 1981). Despite the efforts of these people to stimulate
commercial interest, farmers have not responded in any
significant way.
During the '50s and early '60s beef production was not
particularly profitable; the beef enterprise played second fiddle
to tobacco and maize. Although farmers were planting pasture
grasses, they were not prepared to spend money on the more
expensive legumes. In the '70s, the situation started to change
with an increase in consumer demand coupled with the banning of
beef imports. When beef became more profitable there was
increasing interest in pasture development. This interest,
however, was not matched with a corresponding interest by most
farmers with pasture legumes. The farmers relied on inorganic
nitrogen which was heavily subsidised. As costs of N fertilizer
increased with removal of subsidy in the 80s interest in pasture
legumes was renewed and a high research priority was given aimed
at finding stable legume-grass systems. There was an increased
emphasis on pasture seed production as a result.
Locations
The investigations were carried out at two locations:
1. Mount Makulu Research Station, situated 16 km south of
Lusaka at an altitude of 1,206 m. The average annual
rainfall at this site is about 803 mm, falling during the
rainy season that spans the November-March period. The mean
annual temperature is 20.7°C with the coldest months, June
and July, experiencing occasional ground frosts.
2. Chimyamauni Farm, Mkushi, situated 300 km north of Lusaka at
an altitude of 1,300 m. This site has an average annual
rainfall of 900 mm and a mean annual temperature of 19.8°C.
RESULTS AND DISCUSSION
Excellent results have been obtained from three pasture legumes:
Siratro, Stylo and Glycine at Mount Makulu. They have been grown
in pure stands and in pasture mixtures for a number of years, and
their suitability has been well established. For instance, two-
year-old Chloris gayana (Giant Rhodes grass) was oversown with
Macroptllium atropurpureum (cv Siratro) , Stylosanthes guianensis
(Stylo) and Neonotonia wightii (Glycine) . The pasture received a
fertilizer application of 400 kg/ha of single superphosphate and
200 kg/ha of muriate of potash. All three legumes developed
395
satisfactorily and formed a considerable proportion of the
herbage as demonstrated in Table 1.
total
Table l. Herbage yield from Rhodes grass/legume pastures.
July
Pastures mixtures kg DM/ha kg DM/ha kg DM/ha
grass legume
Legume
January as
percent
kg DM/ha of
total total
Rhodes grass/Glycine 1,307 771 144 915 15.7
Rhodes grass/Siratro 1,840 1,442 260 1,702 15.3
Rhodes grass/Stylo 1,743 1,504 96 1,600 6.0
Glycine and Siratro recovered rapidly on a section that was
burnt accidentally in May. Stylo suffered most from the burning
and the plants were subsequently damaged by termites. Siratro
spread evenly in a very striking manner throughout the plots
producing the most satisfactory and vigorous growth.
All three species were compatible in Rhodes grass/legume
pastures and were observed even to grow well with Star grass and
in natural pasture swards. In the latter case, establishment and
development were relatively small. The legumes were capable of
producing well over 6,000 kg/DM/ha and over 1,000 kg CP/ha (16.6%
CP) . They produced high-quality herbage throughout the season
with crude protein values ranging from approximately 10 to 20%
generally two to three times higher than the value for grasses.
Glycine and Siratro also produced a considerable amount of dry-
season growth and yielded 500 to over 1,000 kg DM/ha between June
and October when good-quality forage is generally very scarce.
These legumes stood up very well to heavy intensive stocking when
they were allowed adequate rest between grazing cycles.
Rhodes grass and native grass oversown to legume and stocked
at 0.50 to 0.75 ha/steer supported steers for about seven to nine
months. The steers gained 90 to 100 kg during the dry period.
Glycine and Siratro are fire-tolerant and produce new growth soon
after burning. Where they were completely burnt in June, the two
species produced vigorous dry-season growth, yielding about 1,160
kg DM/ha. Stylo which is more drought-resistant is not fire-
tolerant. The roots are readily killed by burning. It will,
however, grow and re-establish itself during the rainy season
following the fire from fallen seed.
The above-ground herbage of all three species is killed by
frost, but their root-stock is frost-tolerant. Both Glycine and
Siratro produced vigorous new growth soon after severe frost
which occurred in July, and they continued to produce excellent
growth throughout the rest of the dry season. Regrowth of Stylo
on the other hand seemed slower.
On loamy soils of medium and heavy texture the most suitable
species are Glycine and Siratro, while on the light-textured
sandy soils Siratro and Stylo seem to be more suitable. Other
396
cultivated and indigenous species such as Silverleaf desmodium,
Macrotyloma and Rhynchosia, can also play an important role in
pasture and forage improvement.
Prior to 1974, large areas in the Mkushi District had been
sown to both Rhodes grass and Star grass and some farmers had
experimented with Torpedo grass (Panicum repens) in their
'dambos'. As a result of the increased profitability of beef,
one farmer between 1979 and 1982 was able to clear 600 hectares
of land along Mkushi River. The whole area was cleanly wind
rowed and ploughed. Grass seed (Rhodes grass, Signal grass and
Green panic) was sown with the legumes (Silverleaf desmodium,
Greenleaf desmodium, Stylo cvs. Cook, Endeavour, Graham, and
Seca, Lotononis and Siratro) into a rough seedbed and lightly
harrowed.
In the 'dambos' the area was disced during the dry season.
Torpedo grass was planted vegetatively on a metre square spacing
and Lotononis oversown into the rough seedbed. Since
establishment, paddocks have been grazed at an overall average
stocking rate of about 1.6 LU/ha. The pastures were heavily
grazed during the December to February period and then either
rested or lightly grazed for the next three to four months. In
the latter part of the dry season they were again subjected to
heavy grazing. Results of this exercise are presented in
Table 2.
In all cases the legumes were slow to establish although the
companion grasses did well. Greenleaf desmodium started well but
was quickly grazed out. The Silverleaf desmodium was the most
encouraging, quickly establishing itself after the first year and
maintaining or even increasing its share of the sward over the
following four to five year period.
Of the Stylosanthes spp. both Cook and Endeavour did well,
but they suffered from termite attack during the last two years.
In the rainy season of 1984/85 for the first time in Zambia the
incidence of Anthracnose was confirmed. Worst affected were the
cvs Endeavour and Cook in Rhodes grass. The one stand of cv
Graham looked most promising having increased steadily over four
years. It seemed less affected by termites and devoid of
Anthracnose.
The one plot of cv Seca (Stylosanthes scabra) started very
slowly but had increased steadily despite being heavily frosted
in 1981 and 1982. Seca plants spread into adjacent pastures,
presumably through dung of grazing animals. The Lotononis had
done exceptionally well in conjunction with Torpedo grass in the
'dambo'. Because of its palatability, it tended to be very
heavily grazed in the dry months, but despite this it still
managed to maintain itself.
CONCLUSIONS
It has been proved during the last decade that many of the common
pasture legumes will grow in Lusaka, Central and Southern
provinces in Zambia. A good balanced pasture should contain both
twining and bush-type legumes. The former group is represented
397
o
ST j Sfr j oO 8 j
j jO oo o j o O oj
OO
it 8*
o j O O o
OO
TT o j O O j O
S
j O
8 _ L _
- _ _ _ _ _ _
ssaj6sapoirassejOsaeoiassaj6sapoifa ssejL>sapoira
osej6oeejaoL
ssajBO^uBto
ssaj6oeej oj,
ssej6sepoin ssej6Oao6to
OOjOOjOOj OOj OOj OOj jOj jOj OOj
OOO O
»
O
8 08
O
meaouoeejajOjsaOto
O^^S5OjD
OJ^aJTS
oOZ}Saoso
oTsoso^cyL
raeieoraoeeia _oiaATTS
STOO^CTX
UmipomOBeJajTcjjJS
oLa^SuiaXoEao
OO Z
O O
L s
j8O
O
O j O
»
oO S frT
O O
E
OO 8
O
SZ
j
Z
jO
O
OT ST ET T
OO L Z E E -
ssajdsoeoiassaj6sepoirassaj6saeoifassej6sepoirassajBsepontf
6LOj 6LOj 6Z.Oj 6LOj 6LOj
O O O O O
meTeorasaeiaa_oaa o
jse_aeATTS oLa^Sxooo
oTAq.oJeoAaaesa
sjooo^,
ejEwsmeSaT%,e^EmT^s
OOEjO
OTcajlUj
^seissoSTLiE^sah
jojajaa av
eumbeq
"OO-6LOjEiiuiEZ'TUoO^W_TseEuiEAfflTc.3_^se sTiiE^sae ieO Ta e^sa*j_ Ej,
by Silverleaf desmodium and Siratro while Seca and Graham Stylos
are the most promising of the latter group. Lotononis continues
to be the main hope for ' dambo' improvement. There is no point
in growing grass and legume pastures unless they can be properly
utilised. It is likely that they would grow and thrive well if
they are adequately managed. If pastures are allowed to grow
unchecked during the entire growing season, the herbage becomes
mature and coarse, and it rapidly deteriorates in feeding value.
Pastures which are subjected to intensive rotational grazing
remain relatively green and leafy and improve in basal cover.
The major constraint to the expansion of legume/grass pastures is
availability of seed. The seed production programme is being
carried out under the auspices of the Government of Zambia and
the Swedish International Development Authority, Agriculture
Sector Support Programme. This programme has demonstrated the
potential that there is for local seed production of different
kinds of pasture species if a good management and good permanent
maintenance breeding programme is introduced.
REFERENCES
Craufurd, R.Q. 1978. Plant breeding pasture grasses and legumes,
1967-1978. Research Memorandum No. 26. Ministry of Rural
Development. Research Branch, Mount Makulu, Zambia.
Craufurd, R.Q.
1933-1979.
Zambia.
1979. Ecology and pasture research in Zambia
Mount Makulu Research Station, April, 1979,
Kulich, J. 1977. Effect of fertilization and irrigation at
different frequencies on the growth, yield, and quality of
the natural flood plain sward of the Zambezi area, Zambia.
Proc. XIII Int. Grassld. Congress, Leipzig, German
Democratic Republic, 1186-1189.
Kulich, J. 1981. Selecting species and cultivars for pastures in
Zambia. Zambia Seed Company, Lusaka, Zambia.
Kulich, J. 1985. The potential for seed production of tropical
pasture species in Zambia. Proc. XV Int. Grassld. Congress,
Kyoto, Japan. pp. 262-263.
Kulich, J. 198 6. Brief review of past and actual pasture
research work in Zambia. Proc. of a Workshop CDA Mwase,
Lundazi, Zambia.
Kulich, J. and Kaluba, E.M. 1985. Pasture research and
development in Zambia. In: J. A. Kategile (ed.), Proc. of
workshop held in Harare, Zimbabwe, 17-21 September, 1984.
IDRC, Ottawa 1985 pp.508.
Kulich, J. and Kulich, S. 1973. Production and fertilization of
the natural flood plain under irrigation in the Western
Province. Research Branch, Mongu, Zambia 633, 2.
Kulich, J. and Kulich S. 1976. A review of pasture research work
in the Western Province during the period 1971-1976.
399
Research Memorandum No. 17, Ministry of Rural Development.
Research Branch, Mount Makulu, Zambia.
Kulich, J. and Nambayo, G. S. 1986. Advances in pasture research
and development in Zambia. In: I.Haque, S.Jutzi and
P.J.H.Neate (eds) , Potentials of forage legumes in farming
systems of sub-Saharan Africa. Proceedings of a workshop
held at ILCA, Addis Ababa, Ethiopia, 16-19 September 1985.
ILCA, Addis Ababa, pp.306.
Peterson, R.T. 1975. End of contract report. Ministry of Rural
Development, Research Branch, Mount Makulu, Zambia.
Potter, H.L. 1972. Pasture research for beef cattle in the
Southern Province sandveldt. Farming in Zambia 7(2).
Prins, W.H. 1970. A review of pasture research work in the Mount
Makulu and Mazabuka areas, 1950-1969. Research Memorandum
No. 12, Ministry of Rural Development Research Branch, Mount
Makulu, Zambia.
Prins, W.H. 1972. Legumes in pastures: some recommendations.
Farming in Zambia 7(3).
Prins, W.H. 1975. Plant production and protection with special
reference to pasture research and extension in Zambia.
Terminal Report AG:DP/ZAM/68/008. FAO, Zambia.
Rensburg, H.J. van. 1967. Pasture legumes and grasses in Zambia.
Government Printer, Lusaka, Zambia.
Rensburg, H.J. van. 1969a. Legume/grass pastures in Zambia.
Farming in Zambia 4(2).
Rensburg, H.J. van. 1969b. Introduced legumes in natural
grassland and in established pasture in Zambia. Government
Printer, Lusaka, Zambia.
Shalwindi, f.k.m. 1978. Planted dryland pastures. Farming in
Zambia 12 (1) .
Smith, C.A. 1962. Tropical grass/legume pstures in Northern
Rhodesia. Journal of Agricultural Science 59:111-118.
Smith, C.A. 1963. Oversowing pasture legumes in the Hyparrhenia
grassland of Northern Rhodesia. Nature 200:811-812.
Verdoom, W.C. 1965. Legumes in Zambia. Trop. Agric. (Trin.)
42(3).
400
LAND-USE CHANGES IN RANCHES WHICH WERE SET
UP IN MARGINAL AREAS OF KENYA
Moses O. Olang
Tana and Athi River Development Authority
P.O. Box 47309, Nairobi, Kenya
Thouri Njoka
Lecturer, University of Nairobi
Range Management Division
P.O.Box 30197, Nairobi, Kenya
Abstract
A number of ranches were set up in marginal areas during colonial
days, and they were the main source of steers for the beef
market. Most of these ranches were bought by local people after
forming cooperative societies. These ranches are being
subdivided by the society members for dry land farming.
This may lead to a drop in beef production and without
proper land-use planning, may lead to soil erosion. Already an
area east of Lake Nalvasha is experiencing wind erosion during
low rainfall seasons. Once the land-use has changed from
ranching to crop production, the people in range management
should move out of the area and concentrate in semi-arid areas
where livestock production should be intensified.
INTRODUCTION
The British colonial settlers established excellent commercial
ranches, mainly in Machakos, Nakuru and Laikipia districts.
After independence most of the white settlers offered their
ranches for sale to Africans. The sale prices for those ranches
were so high that Africans could not raise the required money at
that time. Loans were made available at 80% of the sale price.
The local people decided to form cooperative societies
whereby each member was to raise Shs. 2,000/- to buy one share.
Some members still could not raise the Shs. 2,000/- and had to
incorporate other members. Thus one share was bought by a number
of people but only one of them had his name registered. The
others were called shadow members. For some time the high
quality standard of livestock was maintained and a number of the
ranches supplied the local farmers with improved bulls. The
Kenya Meat Commission also continued to get high grade steers for
slaughter.
The current study was carried out to determine the effect of
subdivision to livestock development and ecology of the range.
COOPERATIVE RANCHING SOCIETIES
A cooperative ranching society is where members own a ranch
through the purchase of shares. A general meeting is held once
to elect a committee of ten who run the ranch on behalf of the
401
members. A general manager, who is supposed to
husbandry, is recruited by the committee.
know livestock
When shareholders bought the ranches most of them thought
that the benefits coming from the ranch would be immediate and
substantial. Others became members just to see the colonialists
go, while still others thought they would graze their own
livestock side by side with the society ones.
Most of the societies started experiencing problems brought
by members. Committee members were the worst. Some of them took
money from non-members then allowed them to graze their cattle on
the society's land, while other members brought their relatives
to live in the ranch as squatters. Some of the white settlers
promised their workers part of the ranch when it was sold. This
made employees become squatters in some ranches. A case in point
is Kiu Co-operative Ranch where workers occupied 4000 acres since
1971 when the ranch was sold. The biggest problem to the
societies, especially those in Machakos District, came from
shadow members. These members were not allowed to participate in
annual general meetings. Whenever a dividend was declared, it
was paid only to registered members. The shadow members became a
force which was soon recognised by the Government. They were
later allowed to become members by increasing their share
contribution .
Cooperative Ranches in Machakos District
By 1964 local people in Machakos had formed 55 cooperative
societies. Some of the ranch societies in the district are
listed in Table 1. The problem with shadow members was
experienced in Machakos only. Some committee members were
illiterate, so instead of guiding the general member, they
retarded his development plans. Some of the managers were not
fit for the posts as their employment was influenced by some
committee members. Because of all these problems it became
necessary to subdivide the ranches.
Table 1. Cooperative ranching sooieties in Machakos District.
Name Size Date Original Land area
(acres) began member per member
in acres
Katelembo 11,954 1972 3446 3.5
Kiu 11,330 1971 994 11.4
Nguu 34,000 1967 1600 21.3
Konza 35,658 1964 1604 22.2
Kima 9,108 1973 1604 5.7
Lukenya 60,000 1963 743 80.8
Drumvalle 11,330 1964 1462 7.7
Mountain View 1 1967 1462 1
Koma-Rock 36,190 1963 279 129.7
Katheka-Kai 16,945 1971 188 90.1
Kalembwani 5,085 854
Aimi Ma Kilungu 35,425 1422 24.9
402
Table 2 shows the members' plot sizes after subdivision.
Officers in range management have always tried to stop the
subdivision of the ranches because they anticipate land
degradation and reduced livestock numbers in the whole country.
Table 2. Effects of subdivision on livestock production under
cooperative management.
Ranch Acreage Members 1986
35,658 3,546 841
11,954 932 467
11,330 1,407 955
16,945 1,124
246
1,061
34,000 3,309 2,096
60,000 4,042 3,460
36,190 2,595 197
Konza
Katelembo
Kiu
Katheka-Kai "B"
"A"
Nguu
Lukenya
Koma Rock
IMPLICATION OF SUBDIVISION
The effect of subdivision is already being felt in the country.
Kenya Meat Commission
During colonial days KMC made arrangements with the ranchers on
the sale of finished steers for slaughter. Ranchers finished
their steers with a market already assured while KMC officials
also made all their purchases by telephone. KMC was thus able to
operate at its full capacity.
As these ranches changed hands and subdivision started, sale
by telephone came to an end and KMC is almost closing its two
factories in Mombasa and Athi River near Nairobi because of lack
of animals to slaughter.
Land Use Changes
Ranches which are already subdivided are being turned to crop
production on subsistence level. Wildlife from Nairobi National
Park and Hopecraft Game Ranch graze in these ranches during dry
season. There will be continued crop damage by wildlife.
Land Degradation
An area east of Lake Naivasha called Mai Mahiu which had been
subdivided much earlier and whose members are growing maize and
beans is being degraded through erosion. Members harvest crops
successfully once in five years. Wind erosion becomes serious
when rains fail to come after land has been prepared.
403
Shortage of steers for Slaughter
Most of the trained range management officers look upon pastoral
areas (especially those in ecological zones V and VI) as being
difficult areas. No proper cattle sales are organised and
instead there is permanent quarantine over cattle movement
because of contagious Bovine Pleuropneumonia. Thus there are no
plans for improvement regarding such problems.
POSSIBLE SOLUTION
Improved Livestock Feed
Most of the vast beef production areas in Kenya fall under
ecological zones V and VI, where rainfall is erratic and the
larger part of the year is dry. Livestock spend most of their
time feeding on dry and protein-deficient forage. Some research
should be carried out to find either deep-rooted legumes or
annuals which will grow within a short time and dry up with the
protein intact. Land preparation should be limited to scratching
the top soil just to stop water from carrying the seeds away.
404
A STRATEGY OF LIVESTOCK RESEARCH ADAPTED TO
SEMI-ARID SMALL-SCALE MIXED FARMING SYSTEM -
THE KATUMANI EXPERIENCE
S. Tessema, E.E. Emojong and M. Maluti
National Dryland Farming Research Station
P.O. Box 34 0, Machakos, Kenya
P.N. de Leeuw
International Livestock Centre for Africa
P.O. Box 46847, Nairobi, Kenya
Abstract
The traditional methods used in livestock and pasture research
usually involve component approaches such as grazing management
and stocking rate studies; dry-matter yields and nutritional
value, growth rates and milk production measurements. However,
because of the complementary nature of the relationship between
these components, an integrated research effort is a mandatory
requisite if relevant and applicable innovations are to be
generated. This is particularly true for small-scale farms under
mixed farming situations.
The livestock and pasture research strategy at the National
Dryland Farming Research Station, Katumani adopted the farming
systems approach with four hierarchical levels: base-line study
of the traditional system; experimentation; subsystem simulation
study and on-farm research. This paper discusses the details of
the latter two levels including methodologies used and
preliminary observations made. The technical innovations being
tested in the sub-system simulation study and the on-farm
research are designed to encompass a range of alternative
technologies that could increase output and improve livestock
productivity in small farm systems of semi-arid areas.
INTRODUCTION
Research in livestock and pasture production at the National
Dryland Farming Research Station (NDFRS) Katumani, is principally
aimed at increasing feed resources of smallholder agricultural
systems of semi-arid areas and developing technologies and
management systems that would raise livestock production from the
subsistence level to a more commercial or a semi-specialised
enterprise system. The research programme is carried out at four
hierarchical levels: 1) base-line studies of the traditional
livestock production system, 2) experimentation, 3) subsystem
simulation study and 4) on-farm research.
The base-line study of the traditional farming system
included comprehensive farm surveys in different parts of the
semi-arid zone in Eastern Kenya in order to describe the existing
farming systems and identify the critical constraints faced by
farmers of the region (Rukandema et al . , 1984; Tessema et al. ,
1985) . These studies showed that traditional farmers of the
dryland areas in Eastern Province avert risks by mixed cropping
and livestock raising with the aim of achieving self sufficiency
405
in food production. The land tenure in these semi-arid areas is
a freehold system and farm sizes range from 1 to 20 hectares with
a mean of 7.5 ha, out of which 2 to 3 hectares are cropped and 4-
6 ha are left as natural rangelands. In the cropping subsystem,
the principal crop grown is maize, usually intercropped with
beans, pigeon peas and cowpeas. In the livestock subsystem,
cattle, goats and sheep are kept by almost all farms, usually in
numbers that exceed the carrying capacities of the farms. This
has resulted in poor livestock productivity (e.g. in cattle, 50%
calving rate; 18 months calving interval, 65 kg weaning weight
and 450 kg milked yield per year per lactation) . The large
number of livestock kept is related to the threat of recurrent
food shortages due to crop failures on account of erratic and
unreliable rainfall patterns. Thus, livestock are the only means
of hedging against such crop failures. They serve as savings
and to maintain cash liquidity. Their functions as sources of
draught power, manure, meat and milk are secondary. Given these
circumstances, however, these smallscale farmers attempt to
operate rationally and optimally. Given appropriate technologies
and the means to implement them, these farmers are quick to
change and do raise their productivity considerably.
RESEARCH BACKGROUND
Experimental research has been pursued on a number of areas.
studies of the natural pasture through simple techniques such as
selective bush clearing, burning and reseeding have shown that
with a bimodal pattern of rainfall, reasonably high dry-matter
accumulation of herbage can be achieved allowing higher stocking
rates (2-3 ha/LU) than were normally accepted (6 ha/LU) for the
dryland areas (Tessema and Emojong, 1984a) . Improvement of crop
residues utilisation has shown good possibilities to solve dry
season feeding problems. The quality of crop residues was
improved through physical treatment and supplementation with high
protein Leucaena and pigeon pea leaves (Tessema and Emojong,
1984b) . The economics of milk production was found to be
profitable through feeding cultivated fodders such as Napier
grass (Pennisetum purpureum) and Bana, (Pennisetum purpureum x
Pennisetum typhoides) to crossbred dairy cows (Thairu and
Tessema, 1987). Draught Zebu oxen, when properly managed and
adequately supplemented, provide sufficient draught power to
break the soil and provide good tilth for planting crops before
the rains thus making it possible to prepare more land for food
and feed production (Tessema and Emojong, 1984c) . Such findings
were used for developing a set of improved practices that can be
brought together and tested as a package.
In the past, research efforts to improve the productivity of
small-scale farms have had little success because of the failure
to consider an integrated approach as a research strategy. An
integrated research approach is even more important for the
livestock system than it is for crops. Improved technological
alternatives in crop production such as new crop varieties, new
rates of seeding and different planting dates may each be
effectively adopted individually without substantially requiring
LU = 250 kg weight
406
accompanying technologies or affecting other activities on the
farm. Innovations in livestock production, however, may not be
considered individually or in isolation without due consideration
of their effects on other activities of the farm. For example,
the adoption of cultivated pasture and fodder crops depends upon
the extent to which the activity is conflicting or competing with
food crop farming. Cultivated fodder crops could only be
economically utilised by livestock that are highly productive and
with good genetic potential for such economic traits as milk
production. As the genetic potential of the indigenous zebu
cattle for milk production is rather low, the use of cultivated
fodder crops for increased milk production would at least require
the introduction of crossbred dairy cows in the system.
Furthermore, for increased milk production from intensively
managed pastures to be fully attained, an adequate water supply,
proper animal health care and disease control, appropriate
housing and shelter are required. Thus, a single innovation,
such as the introduction of cultivated fodder crops in the
farming system entails the adoption of a multitude of other
activities. Therefore, while a step-wise improvement strategy is
possible in crop improvement programmes, the livestock production
subsystem must have the 'critical mix' of technological
alternatives that need to be implemented and studied
simultaneously in order to determine relationships, interactions,
competitiveness and/or complementarity of innovations.
At NDFRS in Katumani this is being accomplished at two
hierarchical levels in a farming systems framework: subsystem
simulation study and on-farm research.
SUBSYSTEM SIMULATION STUDY
In this study the most promising technological innovations
emanating from component research in livestock and pasture
production are brought together and tested under simulated farm
conditions at the experiment station, using the average
resources available to a typical farmer in terms of land, labour
and capital. The objectives of the study are the following:
1. to identify real constraints of the improved livestock
production system when compared with the traditional system;
2. to evaluate inputs, outputs and economics of livestock
production under specific farming conditions;
3. to formulate new and economically sound technological
packages ;
4 . to determine the competition and complementarity in resource
use between the livestock and cropping enterprises.
Design of the Study
An area of land comparable to the average-sized farm (7.5
hectares) was demarcated at the station. Leaving 2.5 hectares
for food crop production, the remaining 5 hectares were put under
the following improved practices of livestock production:
407
a) improvement of the natural pasture through selective bush
clearing and burning and reseeding with Stylosanthes
scabra. The purpose is to improve the quality and quantity
of dry-matter production and increase carrying capacity;
b) improved utilisation of crop residues for dry season feeding
through better collection and storage, treatment and
supplementation with Leucaena leucocephala and pigeon peas;
c) planting of cultivated leys (Rhodes grass and Stylosanthes
mixture) and fodder crops (Bana/BajraVNapier) ;
d) appropriate feeding of draught oxen during the dry season;
e) the conservation of excess forage as hay and silage;
f) provision of an appropriate stock shed and other facilities
for better feeding, water provision and manure collection;
g) maintenance of a systematic disease prevention and control
schedule;
h) adjustment of livestock numbers to the available feed
resource ;
i) introduction of crossbred (Zebu x Friesian) dairy cows with
high milk production potential.
The central theme of the package is to increase feed
resources and improve the livestock feeding system by
intensifying land use and by integrating feed resources so that
continuity of feed supply is assured throughout the year.
Livestock production is intensified through the introduction of
dairying for commercial milk production based on crossbred cows
maintained on improved natural pasture, cultivated fodder crops
and supplemented with home grown and conserved feedstuffs.
From this subsystem study, data have been collected over a
three-year period on pasture and animal production, including
productivity of natural and improved pasture, milk production,
calf, lamb and kid growth, weight gains and losses in mature
stock, disease occurrence and mortality and livestock offtake.
Along with livestock production operation all input data are also
recorded: i.e. labour needed for various activities and cost
prices of all items used.
Inputs and Investments
The initial and annual costs are given in Table 1 and included
the following activities:
Bush clearing
The five hectares of bushy grazing land was selectively cleared.
Reseeding with Stylo and Rhodes grass using minimum cultivation
methods is still under study.
Pennisetum typhoides
408
Table 1. Estimated cost of inputs.
Activities Labour min
man days
Initial
cost
Annual
5.0 ha of bush clearing 60 1500 7
1.0 ha of fodder crops
planting
- weeding (2X)
25
50
625
1250 625
0.5 ha of fodder legumes
planting 10
10
300
- weeding 300 300
0.5 ha of grass/
legume ley
seed
- planting
weeding
10
10
100
300
300 300
2.0 ha crop residues
collection,
transport and storage 20 600 600
Ensilage of 5 tonnes
- pit digging 30
140
900
4200- filling 4200
Hay making
baling per tonne 20 600
Establishment of grass/legume leys
The grass/legume ley is composed of Rhodes grass and stylo and
requires one thorough weeding each season. This forage is grazed
or cut and conserved as hay for dry season feeding.
Establishment of fodder grasses (Bana Napier)
The established plot is expected to have a life span of four to
five years and would require one weeding each season (short and
long rains) . The fodder is zero-grazed during the growing season
and any excess is ensiled.
Establishment of fodder legume (Leucaena)
The established plot is expected to have a long life 6pan
requiring one thorough weeding each season. The fodder is cut
and fed to zero-grazed animals both during the growing season and
the dry season.
Utilisation of crop residues
Crop residues from maize, pigeon pea and sorghum are collected
properly each season and used during the dry season. Various
techniques are used to improve the efficiency of their
409
utilisation, i.e. chopping; treating with urea and
supplementation with legumes such as Leucaena or pigeon pea
leaves and stems.
Forage conservation
Silage-making
The forage material used to make silage in this study was
obtained from a one-hectare Bana Napier field which is not part
of the five hectares assigned to the subsystem unit. To adopt
this silage-making technology, it would be necessary to increase
the fodder area of the unit (Table 1) from one to two hectares in
order to get the bulk required.
Circular pit silos, 3 deep and 2 in diameter containing 5
tonnes of chopped forage material packed properly by continuous
manual filling and trampling have been made. The inside of the
silo is lined with a polythene sheeting to prevent contamination
with soil and help in excluding air.
Although this method of silage-making was part of the
package tested in the subsystem study, silage-making at small-
scale farms is recommended only as a communal effort as it may
not be possible for a farm to have the necessary bulk of forage
crops or the necessary family labour force to do the work, nor
would it be economical to do it by hired labour. If made on a
communal basis, it is assumed that it will also be shared equally
among those who contributed in the effort.
Hay-making
This was done using a simple wooden box hand-baler developed at
the station and costing KShs. 260/- . Three men can cut and make
10 bales of hay per day each weighing 15 kg on the average.
The following are the purchased investments included in the
package :
Ox-cart
A locally made wooden body ox-cart made from an old car axle and
wheels and tyres filled with saw dust cost KShs. 2000/-. These
carts are made by a small firm called Ndume in Gilgil near Nakuru
town.
Hand chopper
A hand-operated chopper for cutting fodder crops and crop
residues cost Kshs. 4 000/-. This machine is available in Nairobi.
To keep animals in reasonable comfort, to eliminate unhygienic
situations and to feed and manage them properly and separately, a
shed with partitions for dairy cows, oxen, calves and small stock
^■One US dollar - 16 KShs.
410
Livestock shed
was constructed. Locally available materials were used in the
construction of the shed. If corrugated iron sheet roofing is
used, it can also serve to collect rain water for use by animals.
The bulk of water requirement is, however, hauled from a river
source using ox-cart and drums. The livestock shed is estimated
to cost KShs. 4000/-.
Number and types of livestock
Based on the feed resources developed on the five hectares of
land as given in Table 2, the following livestock were kept in
productive state throughout the year:
- three crossbred dairy cows (Zebu x Friesian) ,
- one replacement heifer, and two oxen (indigenous) ,
- seven sheep and their followers,
- seven goats and their followers.
Table 2. Estimated total fodder resources.
Type of resource Yield t DM ha1 Total t DM
5 ha improved rangelands
0.5 ha grass/legume ley
1.0 ha planted fodder
0.5 ha Leucaena
2.0 ha crop residues
Total resources 26.9
Total yield per annum from two growing seasons.
An in-calf crossbred heifer is generally estimated at
approximately KShs. 4500/- when directly purchased. But the
crossbred cows can be obtained through artificial insemination of
indigenous cows, thus eliminating this high investment cost.
A stocking rate of approximately one livestock unit (250 kg)
per hectare of farm holding (7.5 hectares) is achieved while the
mixture of livestock kept will meet the farmer's needs for
draught power, milk production (for sale and home consumption)
and small stock for meat and periodic sale to meet large payments
such as school fees, etc.
Labour required
The labour required to carry out the activities of this livestock
subsystem is shown in Table 3. The sum total of man-hours
required per day amounts to 11.8 hours. Out of these, seven
hours are needed for herding, which under real farm situations
can be undertaken by 7 to 10-year-old children who go to school
only part of the day. Therefore, these hours need not be
considered in terms of adult labour force. In the subsystem
2.3 11.5
6.0 3.0
7.0 7.0
4.0 2.0
3.4 3.4
411
unit, grazing areas were fenced, thus the labour requirement for
grazing was only imputed. The real requirements of labour hours
for the livestock activities in terms of adult equivalents were
therefore 4.8 hours.
Table 3. Labour requirement for livestock work.
Activities hrs/day
Milking: 3 cows at 0.3 hrs/cow 1.0
Feeding: cutting, transporting,
chopping, feeding 2.7
Cleaning barn and transporting manure 0.6
Watering, hauling water (1 km) in
drums by oxcart 0.5
Herding of cattle and smallstock 7.0
Total 11.8
Other expenditures
These include:
- drugs for all animals(- KShs. 980.00),
- mineral and protein supplements (= KShs. 544.00),
- dipping costs(- KShs. 760. 00) .
Outputs
Fodder resources
The total amount of feedstuffs produced is shown in Table 2 and
indicated that 27 tonnes of dry matter were potentially available
in an average year.
Milk production
The total average lactation yield from three cows over two
lactations was 1870 kg per cow without the calf at foot. This
was achieved under semi-zero grazing where the cows were grazed
on natural pasture during the day and, in the evenings, were
offered chopped Napier grass (ad libitum) during the growing
season and treated maize stover mixed with green Leucaena (at 3 0%
of DM intake) during the dry season.
When 500 liters of the milk produced is fed to the calf over
a six-months period, 1370 kg of milk per cow is left for sale or
home consumption. This is three times the amount produced by
the average cow under traditional management. At a current value
of milk in the local market of KShs. 3/- per liter, the value of
milk offtake from three cows amounted to KShs. 12, 330 over a 14-
month period.
412
Livestock growth and offtake
Annual average weight changes of all livestock over a three-year
period is shown in Table 4 below.
Table 4.
Class of animals
Annual liveweight gains of different classes of stock.
Class of animalADG
(g)
ADG
(g)
Cows <3 years old
Heifers <2 years old
Calves 0-12 months
264
493
319
Oxen 5-6 years old 0
Shoats, young 0-5
months 80
Shoats, adults
> 5 months 30
The annual offtakes consisted of three lambs weighing an
average of 28 kg, with a value of KShs. 672/- (at 8 KShs/kg) and
four weaner goats weighing 30 kg with a value of KShs. 1080 (at 9
KShs. liveweight).
Draught power output
The two oxen provided adequate traction power to plough two to
three hectares of land each season, before rains started. They
also pulled an ox-cart whenever required to draw water or to
transport fodder. The opportunity cost of renting a team of oxen
to plough one hectare of land in the area is KShs. 550/-.
Manure
A total of 30 tonnes of manure (a compost made of 80% dung and
20% soil and other plant material) was collected yearly from the
herd. The monetary value of this manure, according to current
sale price is KShs. 2400 (at KShs. 80/- per tonne) .
The technological improvements tested in this study
certainly required high inputs. But the outputs are also
correspondingly high. It is, however, felt that more data would
be required to make an appropriate assessment of the
profitability of the enterprise. Calving, kidding and lambing
intervals; death rates; the rate of decline in yield of
established fodder fields; value of replacement heifers sold or
kept etc. are some of the vital data that are still required to
make a comprehensive assessment. It is estimated that a period
of five years would be adequate to make such an assessment.
The existence of strong interactions between crops and
livestock in these small-farm systems is well recognised. While
the crop subsystem is being studied in a separate simulation
programme, the on- farm research programme discussed below is,
however, carried out with an integrated multi-disciplinary
approach.
413
ON-FARM RESEARCH
The most promising results from the multi-disciplinary on-station
research were used to formulate technological improvements that
would appear to increase the productivity of livestock on small-
scale farms in semi-arid areas. Preliminary results of the
profitability of these improvements when tested as a package
under simulated farming systems also appeared to be very
promising. The on-farm research programme was therefore designed
to test the validity of these technological innovations under
actual farm conditions.
There are many factors which determine how rapidly a
practice will be adopted by small-scale farmers. These include
the costs and risks involved, the ignoring of potential returns,
the complexity of the practice, the time it takes for better
results, the compatability of the practice relative to cultural
customs and established pattern of agriculture. Farmers are
expected to respond differently to a given economic opportunity
depending upon their subjective evaluation of the innovation. The
smaller the potential increase in income from a particular
recommendation, the more varied the response of different farmers
is likely to be. For example, an increase of 10% to 20% in milk
yield or in growth rate above the common experience can hardly be
distinguished from changes in yields resulting from the usual
seasonal or annual fluctuations. Incentives which in the initial
stages bring about increases of 100% or more have a much better
chance of adoption. Thus, the concept of the on-farm research
programme has necessitated a somewhat radical approach, namely
the introduction of a commercialised livestock enterprise in
small-farm systems.
Smallholder dairying based on crossbred cows maintained
almost entirely on well managed forages and pasture grasses and
supplemented with home grown and conserved feedstuffs would
indeed seem to be a feasible innovation worth extensive on-farm
testing. Except for the problems of feed resources and diseases
that may limit level of production, the economic aspects of
dairying in low potential areas would not be too different from
that of the high potential regions when expressed in terms of
returns to labour and capital investment. However, a successful
dairying enterprise needs to adapt to the cyclical nature of the
environment. Therefore a year- round feed supply of sufficient
quality needs to be guaranteed.
The feeding of dairy cattle for improved productivity is
best achieved by the integration of herbage from natural
rangeland, sown pasture, planted fodders and crop residues.
Without the combined use of these different feed resources,
increased and sustained milk productivity cannot be achieved. In
order to attain this objective a change must come about from the
present system of keeping livestock as a separate enterprise to
that of fully integrating them in the farming system. As soon as
farmers accept that grasses and forages need the same care and
management of food crops, a transition from livestock keeping for
subsistence to that for productive purposes will be effected.
414
Design of On-Farm Research
The technological innovations are introduced step-by-step into a
number of farms representing the main ecological areas, farming
systems and size classes. An optimum plan is made to each
individual farm, incorporating as many of the recommended
technologies as possible. The recommended improvements in
pasture and livestock production in particular are those that
require the minimum capital input and the maximum use of family
labour force. Since the farms selected are being considered as
experimental units, a protective mechanism has been built into
the on-farm research programme whereby the inputs are provided
to the farmer on the basis of loans. The participating farmers
are required to repay these loans only after it has been proven
that the innovations have contributed substantially to the net
income of each farm. Thus, participating farmers are not
subjected to unnecessary risks and losses due to the research
exercise being carried out on their farms.
On the basis of rainfall and existing farming systems, the
study area is divided into three domains encompassing parts of
Machakos and Kitui districts. Domains have been identified:
1. Most favourable, 700-800 mm rainfall;
2. Medium favourable, 600-700 mm rainfall;
3. Least favourable, 500-600 mm rainfall.
Farms are grouped according to farm size in small (5 ha) ,
medium (5 to 10 ha) and large (>10 ha) .
Two farms from each domain and each class are selected,
making a total of 18 farms. Another group of 18 farms within the
same domains are selected to serve as controls where only
comparative data are collected.
Each farm is surveyed and appropriate technologies are
developed for each domain's classes. Ex ante analyses are made
by computer modelling and simulation to assess potential
applicability, input requirements and constraints and outputs.
The technical coefficients used for these analyses were generated
from the subsystem simulation study discussed above.
The technological packages being tested on the farms are the
same four enumerated under the subsystem simulation study.
Improvement packages vary according to each domain e.g. dairy
cattle for most favourable rainfall zones and dairy goats for
drier zones. Farm plans are discussed with farmers and modified
to reflect their preferences and willingness to test innovations.
Budgeting and sensitivity analysis techniques are used to
determine optimal farm plans.
To date, the on-farm research programme has been implemented
only on six participating farms. For the six farms, data have
been collected on pasture and animal production including
production of natural and improved pasture, milk production;
calf, lamb and kid growth; weight gains and losses in mature
stock; disease occurrences and mortality. Along with livestock
production operations, selected input/output data on all farm
enterprises are monitored continually to quantify labour,
415
traction and material inputs as well as the returns for various
enterprises, and to assess the pressure on farm resources and the
levels of risk of different enterprise combinations faced by
farmers. The data will be analysed by linear programming,
regression equations and variance analysis to evaluate results
and identify variables. Cost-benefit analyses will be made to
test economic feasibility.
Achievements of applicable results from livestock and
pasture research are lengthy and difficult processes. While
substantial data are available from the subsystem simulation
study, collected over the past three years, data from the six
farms on the on-farm research programme are scanty. Preliminary
observations of the research approach used, however, show that
the researcher is more knowledgeable of the real constraints of
the production system and therefore can design more relevant and
effective programmes. It seems also that the desired end results
of the research efforts can be arrived at in a reasonably shorter
period of time.
REFERENCES
Rukandema, M. 1984. Farming systems of semi-arid Kenya, a
comparison. East African Agric. and For. Journal 44:422-35.
Tessema, S. and Emojong, E.E. 1984a. The influence of stocking
rates and grazing management on liveweight changes in
cattle, sheep and goats grazing natural pastures. East
African Agric. and For. Journal. Special Issue. Vol. 44.
Tessema, S. and Emojong, E.E. 1984b. Utilization of maize and
sorghum stores by sheep and goats after feed quality
improvement by various treatments and supplements. East
African Agric. and For. Journal. Special Issue. Vol. 44.
Tessema, S. and Emojong, E.E. 1984c. Feeding of draft oxen for
improved and more efficient power. East African Agric. and
For. Journal. Special Issue. Vol. 44.
Tessema, S., Emojong, E.E., Wander a, F.P. and Nderito, M. 1985.
Features of traditional farming systems as they affect
livestock production. A case study of 18 small-scale farms
in the dryland areas of the Eastern Province of Kenya
(Machakos, Kitui and lower parts of Embu districts) .
Dryland Farming Research and Development Project, KEN/8/011,
Machakos . Document No . 6 .
Thairu, D.M. and Tessema, S. 1987. Research on animal feed
resources: Medium potential areas. In: J. A. Kategile, A.N.
Said and B.H.Dzowela (eds) , Animal feed resources for small-
scale livestock producers. Proceedings of the Second PANESA
Workshop on Animal Feed Resources for Smallscale Livestock
Producers held in Nairobi, Kenya, 11 - 15 November 1985.
IDRC-MR 165e. International Development Research Centre,
Ottawa, Canada. pp. 125-148.
416
LEGUME-BASED CROPPING:
A POSSIBLE REMEDY TO LAND TENURE CONSTRAINT
TO RUMINANT PRODUCTION IN THE SUBHUMID ZONE
OF CENTRAL NIGERIA
G. Tarawali, M.A. Mohamed-Saleem and R. von Kaufmann
International Livestock Centre for Africa
Subhumid Zone Programme, Kaduna, Nigeria
Abstract
There is a critical shortage of crude protein in fallow and range
grazing during the dry season in the subhumid zone of Nigeria.
Supply shortages and high cost preclude the use of purchased feed
and research was directed towards the production of forage
legumes.
A package for creating fenced legume pastures ('fodder
banks') has been developed, but since most land is owned by
cultivators who do not keep cattle, pastoralists have found it
difficult to find the necessary land.
Productivity of the pasture is reduced over time by the
build-up of soil nitrogen that encourages invasion by
nitrophilous grasses, thereby excluding the desired legume
species. This build-up can be reduced by periodic cropping with
nitrogen-demanding cereals in order to favour legume dominance.
The cereals used for this purpose yield better than natural
fallow because of the nitrogen and other improvements in the
characteristics of the soil.
This situation can be exploited to increase crop/livestock
integration even where they are generally practised by different
groups; in this case, farmers and pastoralists . The present
paper describes a study carried out to test the hypothesis that
legume-based cropping is beneficial.
INTRODUCTION
The ILCA Subhumid Programme, adopting a farming systems research
approach, identified poor nutrition in the dry season as the main
constraint to cattle production in the subhumid zone of Nigeria.
During this period, the natural pasture is not only scarce but
also poor in quality, with a crude protein (CP) content of about
3%. This is below the critical 7% CP level required in the diet
of ruminants. Cattle may lose up to 15-20% of their body weight
during this season (Otchere, 1986) . Also, milk yields are low,
calf mortality is high and many cows are unable to conceive
because of nutritional anoestrus.
A contributory factor to the low quality and productivity of
the herbage in the subhumid zone is the poor nature of savanna
soils. Intensive methods of agricultural production such as
continuous cropping, land clearing and mechanised farming lead to
nutrient losses through soil erosion, runoff and leaching. The
organic matter of these soils is very low because the poor
savanna vegetation contributes very little to the soil to
417
replenish the fast decomposing organic matter. Burning and the
removal of crop residues from the site for livestock feeding
aggravate the situation.
Any attempt to promote livestock production in the subhumid
zone should therefore consider a programme of soil fertility
maintenance in addition to improving the nutritional value of the
pasture. For instance, the poor carbon and nitrogen content of
these soils could be improved by the incorporation of forage
legumes into the cropping and fallow systems. The use of agro-
industrial by-products such as cottonseed cake, groundnut meal,
urea and molasses can improve the productivity of lactating and
pregnant cows. However, supplies of these feeds are not readily
available and prices are escalating. In view of these ecological
and financial constraints, ILCA considered a sustainable
enterprise such as planted forage legumes (fodder banks) to be a
more appropriate long-term option.
Research is being conducted in three main case study areas.
Kurmin Biri is a grazing reserve established by the government to
assist pastoral ist settlement. Abet is an intensively cultivated
area where pastoral ists settle amongst the arable farmers who are
the landowners. Ganawuri is also intensively cultivated but by
mixed farmers, who own both land and cattle.
THE FODDER BANK CONCEPT
A fodder bank is a concentrated unit of forage legumes
established and managed by pastoralists near their homesteads for
the dry season supplementation of selected animals. These
legumes can fix soil nitrogen, and the protein content can stay
above 8% for a greater part of the dry season. At present,
Stylosanthes guianensis cv Cook and S. hamata cv Verano are the
two main species recommended.
The guidelines for the establishment, management and
utilisation of fodder banks are as follows:
(a) Depending on availability of land and number of animals,
select an area (normally about 4 ha) close to the homestead.
(b) Prepare the seedbed by confining a herd overnight in the
area for several weeks.
(c) Broadcast scarified seeds mixed with 150 kg/ha of single
super phosphate fertilizer.
(d) Control fast-growing grasses by early season grazing.
(e) Allow the forage to bulk up by deferring grazing until the
dry season.
(f) Ration the fodder bank by selecting the appropriate type and
number of animals and limiting grazing to 2.5 hours per day.
(g) Ensure sufficient seed drop and adequate stubble for stylo
regeneration in the following season.
418
Experiments have shown that supplementary feeding with 1 kg
of cotton-seed cake (crude protein - 30%) daily, maintained the
bodyweight of Bunaji cows throughout the dry season (Otchere
1986). It was predicted that cattle fed with 2.5 kg of dry stylo
(crude protein - 12%) would derive similar benefits. With a
potential stylo dry-matter yield of 4 to 5 tonnes per hectare, it
was estimated that a 4-ha fodder bank should adequately meet the
supplementary requirements of 15-20 pregnant and lactating cows
for the six-month dry season if grazing is limited to 2.5 hours a
day.
Benefits of the Fodder Bank
Animals with access to fodder banks in the dry season of 1983
produced more milk, lost less weight and their calves survived
better (Bayer, 1986: Figures 1 and 2). This is supported by
evidence from the comparison of herds with and without fodder
banks (Table 1) . The advantages of owning fodder banks is
acknowledged by cattle owners as evidenced by an increase in the
number of fodder banks in the subhumid zone from two in 1980 to
77 in 1986 (Figure 3) , and financial support has been provided by
the World Bank for the establishment of an additional 2000 fodder
banks over the next five years.
Table 1. Effect of dam supplement.
No
Trait supplement Supplement Improvement
Cow survival (%) 92.2 96.0 4.1%
Calving percentage 53.8 58.1 8.0%
Calf survival (%) 71.8 86.3 20.2%
Calf weight at
1 year (kg) 98.1 103.4 5.4%
Lactation
yield (kg) 300.2 312.5 4.1%
Source: Mani (1986) .
Major Constraints to Fodder Bank Expansion
Social problems (land tenure system)
Among the three basic resources (land, labour and capital)
required for the development of fodder banks, the difficulty of
acquiring land by pastoralists has been identified as the most
important limiting factor. The FulBe agropastoral cattle owners
who were the original target group have no legal land rights in
this part of the zone. Most indigenous farming communities
generally have no direct interest in cattle production.
419
•H
oa
*5
B
S
I
*5
o
-I
w
o
IT
c
O
*P
o 1
o
o
 
<
—I—
O
o
5
—i 1 r—
O O
CJ IO
1 1
i < r
O O
5Jo IO
1 1
O a
1
(1ouj1uo/6>|) seBuoga tMDl>H
420
cc
o
B
c
10
u
3
-p
■O
c
HI
u
3
■P
U]
10
a
■P
Ul 10
01 -H
> ^
.-I -P
«
- 0'
c
O N
a
w u
c tP
-H
10 W
tJicO
■P ■*
J3 CO
m -h
> m
■P C
38
6 3
3 «
1
3201
O0O
o o >->-
pp R "'/ A
,'/ /\
'/ / \/ A \
? // f"" /V i // /3
' I I / 1soO
W fQ-O /
"
>- i \ \ /tn 1 /
\ Ds \
N^x \r
O X
\N
*
D P \ ^0 ' ~\N /
1 S /
V1 yix>
?A
a-
c
o
' / / /
p
o
3
' / '/o
2
—1 1 1 1 1 1 1 1 1i 1
o
5
ai
-H
ooooooooo o
(iHuoui/6>|) suioB m6|»M OAUDinixino
I
O
I
i
10
CO
421
FLgure3.Numbeoffodderbanksundd ffer ntlevelsma ag ment.
8o
jo*ue
6oo
£5o+ o4o o>3o
E
12o
1oo
Reoearcher-managedReoeaFarm r-managed Reoearcher-executedF m r-xecutedrm r-xecute
2ooo
EZZ3
ProjecOed
I7\
noun
 
idb.
1jo1
1j2
1jj4
Year
1j56
1jj
Source:Mohamed-Saleem(1986)
FulBe who are willing to settle on government grazing
reserves find it easy to obtain land for fodder banks but there
is only a limited number of grazing reserves. Most
agropastoralists actually prefer to settle near crop farmers to
have access to crop residues, better marketing facilities and
social services. With an expanding population and consequent
increasing demand for land for cropping, the pastoralists will
face greater difficulties because the farmers retain the best
crop land for themselves.
Moreover, they are typically reluctant to lease their unused
land to cattle owners for pasture development because of their
tedious procedures and safeguards that have to be negotiated
under customary law. In certain areas, pastoralists are worried
that farmers will go to the extent of hindering cattle owners
from acquiring individual pasture plots even in grazing reserves.
This clearly indicates that fodder bank technology will not be as
widely adopted as desired in the subhumid zone unless it is of
significant benefit to the farmers, as well as to the livestock.
Ecological problems (nitrophilous grasses)
Invasion of leguminous pastures by nitrophilous grasses (Gardner,
1984) reduces fodder bank productivity over time. This is
because such pastures accumulate soil nitrogen through fixation
of atmospheric N (Vallis and Gardner, 1984) which makes
nitrophilous grasses very competitive. Since grasses grow faster
than legumes, the latter are likely to be smothered, thereby
lowering the nutritional value of the pasture especially if the
grasses are not controlled. This was observed in a four-year-old
S. hamata cv Verano fodder bank in the Kachia grazing reserve,
which showed a rapid build-up of fast-growing grasses (Table 2) .
In a one-year-old fodder bank, stylo productivity, soil N and
crude protein declined with an increase in grass density (Figure
4) . In Thailand, mixed legume/Setaria sphacelata swards changed
from legume dominance to grass dominance within two years
(Andrews and Gibson, 1985) . This implies that a system must be
devised whereby the excess N can be periodically flushed out to
check the invasion of nitrophilous grasses.
Cropping as a Solution
Cropping uses the accrued N, thereby restricting the growth of
nitrophilous plants and favouring the legume component of the
pasture. Nitrogen is the most limiting element for cereal crops
and one of the most expensive fertilizers (FAO, 1984) .
Exploiting legume-fixed nitrogen through cropping will,
therefore, interest the farmers and will encourage them to lend
their land more readily to the FulBe for pasture development.
Previous work on legume-based cropping (Mohamed-Saleem,
1986) showed that maize-growing on fodder banks of various ages
produced higher grain yields than on previously cropped and
uncropped soils. Improvement in fertility and physical
properties of the soil contributed to the increase in yield
within the stylo area. Measurements on soils where the above
experiments were carried out showed differences in bulk density,
gravel, organic matter and nitrogen contents. ILCA's approach is
to exploit this feature in order to encourage farmers to
423
Table 2. Frequency distribution (%) of major grasses in a four-
year S. hamata cv Verano fodder bank at Kachia grazing
reserve.
Species Natural Fodder
vegetation bank
Andropogon spp . 6.2 13.3
Brachiaria ssp. 8.3
Digitaria ssp. 0.8
Hyparrhenia spp. 11.4 16.1
Loudetia spp. 40.7 4.0
Panicum spp. 0.8
Paspalum spp. 1.4 0.4
Pennisetum spp. 0.5
Setaria spp. 0.6 0.8
Others 24.4 5.6
Legume
S. hamata cv Verano 59.3
Source : Mohamed-Saleem et al (1986) .
cooperate more readily with pastoral ists seeking land for stylo
establishment. Moreover, once the arable community recognises
Stylosanthes as a soil conditioner, it will command the same care
and respect as that given to crops. This will also indirectly
alleviate problems such as the deliberate burning and
indiscriminate grazing of fodder banks.
METHODOLOGY
s conducted duri:
30'N, 8° 25'E, rainfall 1300 mm between April-October), one of
ILCA's case study areas. Two fodder banks of Stylosanthes hamata
cv Verano established three and four years previously and grazed
during all consecutive dry seasons, were identified. Four plots
each of 15 m x 10 m were laid out inside each fodder bank. The
plots were set up in sites in which the legume component was
representative of that found in fodder banks of the subhumid zone
(50-60% stylo) . Each plot formed a replication. Comparable
plots which had remained as natural grass fallow for at least
four years were also demarcated immediately outside the fodder
banks .
The plots were ridged to a height of 30 cm in May at one-
metre intervals and planted with three maize seeds (Variety TZPB)
per hole at a spacing of 0.25 and thinned to one seedling after
emergence. A uniform dose of N, P and K at the rates of 60, 60
and 60 kg/ha, respectively was applied as compound fertilizer (N
p15 K15^ prior to planting and another 60 kg N/ha as urea was
applied six weeks later. Plots were weeded five and eight weeks
after sowing. At harvest, the cobs were first plucked then
dehusked, dried and shelled to determine the grain yield. The
stalks were cut and dried at 80°C to determine the dry weight of
the crop residues. All data collected were analysed according to
variance technique.
424
(%) N l!OS 1D(01
0)*P
o
u
a
o1
*o
*a
c
i0
*P
o
•p
*p
•p
u
p
*p
c
0 •
>i'e
*j o
•H o)
o1 1o
C <D
a a
iO
*p
oi a;
o1 3
a
u in
trcD
o
10
*poh
o a
a o
<m io
 
O»H
P
lT
o O O o O o o O o
IT) lD in in m in in in inCT) CD N C0 m * ro C\J
(dl1/6>() inatojd apnjo 1Otoi
425
RESULTS
The grain yield and crop residues of the maize planted Inside the
fodder banks almost doubled that of maize sown outside the fodder
banks (Tables 3 and 4) .
Table 3. Grain yield (kg/ha) of maize growing inside and outside
two fodder banks in Abet.
Location Site 1 Site 2 Mean
Inside
Outside
LSD (0.01)
4785 4533 4659
25452411 2679
452 235
Table 4. Crop residue yield (kg/l
outside two fodder banks
ia)
in
of maize
Abet.
growing inside and
Location Site 1 Site 2 Mean
Inside
Outside
LSD (0.01)
7212
4018
2331
7586 7399
42714523
592
DISCUSSION
Even at the vegetative growth stages it could be seen that the
maize planted in the legume-based soils was performing better
than that grown on natural fallow. These visible differences led
to the organisation of a field day for over 100 farmers and
pastoralists in Abet on 12th August, 1985. They were quick to
note the superior growth of maize inside the stylo area. The
incremental income to farmers growing crops in fodder banks in
Abet would be 124% of normal crop revenue resulting from a
natural fallow (Kaufmann and Blench, 1986) .
Despite years of contact with fodder banks, the
demonstration trials enabled farmers for the first time to see in
a direct, visual manner that legume-based pastures can be
exploited for crops as well as livestock. As a result, farmers
asked for Stylosanthes pastures to improve soil fertility in
their fallows. Five of these have been established and more
farmers are becoming interested in the package (Table 5) .
Initially, the small hectarage will be used for improving feed
for small ruminants, but later it is hoped that the farmers will
see the advantage of letting pastoralists seed and manage larger
areas for them.
426
Table 5. Fodder banks for small ruminants in Abet.
Farmer No. of Size of Produc %
agro- ruminants fodder tivity Stylo
pastoralist bank
(ha)
(kg/ha)
Magaji Bala 8 0.1 3,604 58
Bayei Kaburuk 7 0.1 3,589 60
Tanko Agau 4 0.05 2,614 42
Achie Gariya 6 0.05 2,832 52
Bammi Buldi 30 1.0 4,836 62
There already exist farmer-pastoralist exchange
relationships whereby pastoralists charge lower rates for
manuring fields to farmers who ensure them first access to crop
residues. At other times farmers arrange for herds of
pastoralists to fertilize crop fields, so as to enjoy the benefit
of cropping the area during the first year. In return, the
pastoralists are allowed to cultivate the areas in subsequent
seasons. This symbiosis could be developed further by
incorporating forage legumes. The farmers could lend their
fallow land to pastoralists for the establishment of fodder banks
and thus improve their crop yields at the same time.
Future work will include the determination of the minimum
area required to produce fodder that will support the average
number of small ruminants kept by farmers. Management guidelines
for the utilisation of the fenced pastures for both wet and dry
season supplementation will be worked out. Also, the
contribution of stylo to soil fertility and the performance of
animals grazing on the fodder banks will be monitored.
Studies of the response of cereals to fixed nitrogen in
fodder banks will be continued for newly opened and continuously
cropped stylo soils in order to determine the most responsive
crop and the fertilizer/labour requirements for cropping such
soils. A crop-legume rotational system will be introduced in the
larger (4 ha) fodder banks.
CONCLUSION
Stylosanthes-based cropping serves the dual purpose of
maintaining forage quality and improving crop yields. Thus, it
can benefit both farmers and agropastoralists. Cropping stylo
areas, hitherto considered as optional in fodder bank management,
should now be regarded as an essential practice.
Farmers and agropastoralists in Kurmin Biri, Abet and
Ganawuri (ILCA case study areas) have seen this potential for
cropping stylo areas. Since tilling these soils does not require
more labour than a natural fallow (Tarawali et al . , 1986) they
have started to crop part of their fodder banks and legume
fallows. While Stylosanthes pastures were originally meant to
promote livestock production, their continued spread across the
subhumid zone will be a boost to crop production as well.
427
REFERENCES
Andrews, A.C. and Gibson, T.A. 1985. Adaptation of improved
pasture species in the Highlands of Northern Thailand. 2.
The 800 m Zone. Tropical Grasslands 19:17-24.
Bayer, W. 1986. Utilization of fodder banks. In: R.von
Kaufmann, S.Chater and R. Blench (eds) , Livestock systems
research in Nigeria's subhumid zone. Proceedings of the
Second ILCA/NAPRI Symposium held in Kaduna, Nigeria, 29
October - 2 November 1984, ILCA, Addis Ababa. pp. 351-364.
Food and Agricultural Organization of the United Nations. 1984.
Legume inoculants and their use. A pocket manual jointly
prepared by Nitrogen Fixation for Tropical Agricultural
Legumes (NifTal) Project, USA and FAO Fertilizer and Plant
Nutrition Services, Land and Water Development Division in
association with FAO Crop and Grassland Production Service,
Rome.
Gardner, C.J. 1984. The dynamics of Stylosanthes pastures. In:
H.M. Stace and L.A. Edye (eds) , The biology and agronomy of
Stylosanthes. Academic Press, Australia, pp. 334-356.
Kaufmann, von R. and Blench, R. 1986. The subhumid zone of
Nigeria: A systems study. The Subhumid Zone Programme,
ILCA, P.M.B. 2248, Kaduna, Nigeria.
Mani, R.I. 1986. Cattle productivity. Paper presented to ILCA
External Programme Review (8-10 April) . ILCA Subhumid Zone
Programme, P.M.B. 2248, Kaduna, Nigeria.
Mohamed-Saleem, M.A. 1986. Integration of forage legumes into
the cropping systems of Nigeria's subhumid zone. In: R.von
Kaufmann, S.Chater and R. Blench (eds) . Livestock systems
research in Nigeria' s subhumid zone. Proceedings of the
Second ILCA/NAPRI Symposium held in Kaduna, Nigeria, 3 0
October - 2 November, 1984. ILCA, Addis Ababa. pp. 304-325.
Mohamed-Saleem, M.A. , Suleiman, H. and Otsyina, R.M. 1986.
Fodder banks: For pastoralists or farmers. In: I. Hague,
S. Jutzi and P.J.H. Neate (eds), Potentials of forage
legumes in farming systems of sub-Saharan Africa.
Proceedings of a Workshop held at ILCA, Addis Ababa,
Ethiopia, 16-19 September 1985. ILCA, Addis Ababa.
pp. 420-437.
Otchere, E.O. 1986. The effects of supplementary feeding of
Bunaji cattle in the subhumid zone of Nigeria. In: R.von
Kaufmann, S.Chater and R. Blench (eds), Livestock systems
research in Nigeria' s subhumid zone. Proceedings of the
Second ILCA/NAPRI Symposium held in Kaduna, Nigeria, 30
October - 2nd November, 1984. ILCA, Addis Ababa. pp. 204-212.
Tarawali, G., Mohamed-Saleem, M.A. and von Kaufmann, R. 1987.
Comparison of labour requirements for cropping maize on land
previously used for Stylosanthes fodder banks and on
adjacent fallows in the subhumid zone of Nigeria. ILCA
Bulletin 26:36-39.
428
Vallis, I. and Gardner, C.J. 1984. Nitrogen inputs into
agricultural systems by Stylosanthes. In: H.M. Stace and
L.A. Edye (eds) , The biology and agronomy of Stylosanthes .
Academic Press, Australia. pp. 359-379.
429
ROLE OF CROP RESIDUES AS LIVESTOCK FEED
IN ETHIOPIAN HIGHLANDS
Daniel Keftaso
Pastures and Forage Crops
Ministry of Agriculture South-Eastern Zone, Ethiopia
Abstract
Ethiopia is known for large livestock populations of which 80%
are raised in the highlands where intensive crop farming is also
undertaken using ox traction. Crop residues particularly cereal
straws are the major livestock feed particularly in the dry
seasons, providing 40-50% of the total annual livestock feed.
Evaluation of nutritional value of crop residues shows that
they are generally low in digestibility and protein content.
Improving intake and digestibility of crop residues by
physical and chemical treatments (NaOH NH3) are technically
possible but they do not have practical value in Ethiopia at
present. Supplementation of crop residues with molasses, oilseed
meals and leguminous fodder crops have been observed to increase
both intake and digestibility of crop residues, and the system is
being developed.
Intercropping cereals with leguminous fodder crops and
supplementation of crop residues with foliages of Leucaena,
Sesbania, etc. and hays of Trifolium, alfalfa and vetch appear to
have much practical value in mixed farming systems of the
Ethiopian highlands.
INTRODUCTION
Ethiopia has the largest livestock population in Africa with
approximately 26 million head of cattle including about six
million draught oxen. About 80% of these animals are raised in
the intensively cultivated highland part of the country. The
Ethiopian highlands cover 490,000 km or about 40% of the country
and almost half of the total African highland area (Gryseels and
Anderson, 1983) .
Ethiopia has unique agricultural features such as the use of
teff (Eragrostis tef) as a cereal and enset (Musa ensete) as a
root crop, an indigenous system of ox traction and the extensive
use of equines (seven million) for transport (Jahnke, 1982) .
Land use is dominated by mixed smallholder rainfed
agriculture producing cereals, pulses and livestock. Crop
production and livestock husbandry are commonly integrated in the
mixed farming system of the medium-highland zones of Ethiopia.
Oxen are important suppliers of draught power for land
development, tillage, threshing and transport which means that
the more crop production is increased, more and more oxen are
required. This fact is even more realistic in most parts of high
potential agricultural zones of Ethiopian highlands due to the
uneven nature of the landscape. In Ethiopia, about six million
430
ha of land are put under annual grain crops; mainly wheat,
barley, teff, maize, sorghum and pulses.
As more and more land is put under crop production,
livestock feed becomes scarce and crop residues particularly
cereal straws remain the major feed source for the animals
particularly during the dry period of the year (which spans the
November to May period) . Some estimates indicate that crop
residues provide 40-50% of the annual livestock feed requirement.
This paper contains current practices of straw utilisation
and reviews efforts made to improve nutritional quality of crop
residues in Ethiopia.
CURRENT PRACTICES
Straw Utilisation
From about 6 million ha of farmland about 12 million tonnes of
crop residues (grain yield 10 qt/ha and 2:1 straw to grain ratio)
are annually produced in Ethiopia of which about one-third is
left in the field for aftermath grazing (stubbles) .
Farmers use crop residues for different purposes, e.g. for
fuel as firewood and minor constructions, especially maize and
sorghum stovers; for roofing local houses, in the case of wheat,
oat and barley straws; as binding material for walls of local
houses, especially teff straw. But the major use is for
livestock feed particularly for draught oxen during dry seasons.
Animals feed on crop residues mainly in two ways. The
residues are piled in stacks near homesteads and animals are let
to eat from the stacks or given small quantities in the morning
and evening, or for working oxen, before and after work.
Alternatively, the residues are left in the threshing ground and
consumed by animals together with the standing straws which are
left for aftermath grazing.
In some parts of the Ethiopian central highlands farms and
government-owned fattening feed lots use straws with molasses and
urea. There is a strong tendency towards improving the
utilisation of crop residues by supplementing with molasses
and/or urea at beef farms. And at some farmer cooperatives pen
fattening of cull cows and old oxen is practiced on straw-based
diets.
Production of Crop Residues From Different Crops
Grain and straw yields of different crops are shown on Table 1.
Yields of grain and crop residues are highly correlated. High
grain yields are the result of high vegetative growth which is
associated with high production of crop residues. The straw to
grain ratio for wheat varieties ranges from 1.8 to 2.9 and of
barley from 1.9 to 2.6. It appears that varieties of low yield
potential produce the highest straw to grain ratio i.e. have the
lowest harvest index.
431
Table 1. Mean yields of grain, residues, harvest indices, CP and
CF (% of DM) of different crops in Arsi Region,
Ethiopia.
Crop varieties Grain Residue Total Residue Harvest % of DM
yield kg/ha kg/ha to grain index CP CF
kg/ha ratio %c
Wheat Enkoy 4585 9808 14393 2.1 32 4.99 38.0
Romany 4000 10872 14872 2.7 27 5.04 40.0
K6295-4A 2890 7969 10859 2.7 27 4.85 35.5
Bulk 3822 9400 13222 2.5 29 5.28 36.5
Dashen 5156 9138 14294 1.8 36 4.36 37.5
Gara 4448 9813 14261 2.2 31 5.28 35.0
Batu 3953 8590 12543 2.2 32 5.72 31.3
Local 2392 6862 9254 2.9 26 4.70 40.0
Barley Improved
(food)a
Improved (malt)
2737 7004 9741 2.6 28 4.4 36.6
2918 5972 8890 2.0 33 5.7 36.2
Arusso 3180 5900 9080 1.9 35 4.30 39.0
Local 1982 4665 6647 2.4 30 5.76 35.0
Tef DZ-354 2100 7100 9200 3.4 23 3.83 29.0
Horsebeans 2200 3900 6100 1.8 36 7.67 48.0
Fieldpeas 2340 12000 14340 5.1 16 13.8 41.0
a IAR 485, A-HOR
b HB37, HB42, Holker.
c Harvest Index i- Graini vield (xq)
Total above-ground weight (kg)
Samples were collected from trial plots and the Seed
Multiplication Farm, so that the yield figures for different
crops might deviate from what is normally expected. However, the
figures shown for residue to grain ratio, harvest index and
contents of crude protein and crude fibre can safely be used for
rough estimation.
Reed et al. (1986) have also demonstrated that high grain
yielding varieties of sorghum (in Ethiopia) can also yield
reasonably high amounts of crop residue with high nutritive
value.
Nutritional Quality
As shown in Table 1 there are considerable variations in the
contents of crude protein and crude fibre among different crop
residues and varieties of the same crop.
Nutritional values of different crop residues and some
forages are shown in Table 2. Chemical composition of the cereal
crop residues contain a large proportion of lignocellulosic cell
wall constituents and have low crude protein contents. Thus one
could expect both voluntary intake and digestibility of the crop
residues to be low. Generally, cereal crop residues have an
energy content of less than 7.0 MJ/kg dm. As shown in Table 2
the digestible protein content of cereals is below 20 mg/kg which
is only about 25% of that from grazing grasses. Legume straws
432
contained less crude fibre, higher digestible protein and calcium
than cereal straws.
Quite a remarkable difference between the crude protein
contents of cereal and legume straws was observed. Comparing
Tables 2 and 3 it can be estimated that the digestible protein
contents of wheat and barley straws is about 40 and that of
pulses is about 45% of the total crude protein contents which
implies that figures of crude protein contents are not the best
measurement of quality but that digestibility is more important.
Table 2. Nutritional values of some straws and fodder crops.
Fibre Digestible Ca P Energy
% of protein g/kg g/kg MJ/kg
DM g/kg DM DM DM
Barley straw 39.5 18 3.0 0.8 6.8
Wheat straw 41.0 14 2.4 0.7 6.2
Tef straw 35.9 19 4.1 0.5 6.8
Pea straw 26.0 40 13.2 0.6 7.1
Bean straw 40.1 20 5.2 0.5 6.8
Native grasses 28.9 87 3.9 1.6 8.9
Improved grasses
(tropical) **
(temperate)
28.2 79 4.3 1.4 8.7
25.4 119 4.8 1.9 9.1
Vicia1 33.2 146 12.0 1.4 9.1
Alfalfa1 22.2 186 28.0 2.4 10.4
Included for comparison.
Source: Evaldson (1970) .
Table 3. Chemical analyses of different straws at Chilalo,
Ethiopia.
Crops/varieties
%
Dry
matter
%
Crude
protein
%
Crude
fibre
%
Ash
Barley improved
(varieties)
Barley local (varieties)
Wheat improved (varieties)
Wheat local (varieties)
Triticale
Tef
Maize (cobs, stalks)
Sorghum
Horsebeans
Fieldpea
Lentiles
Flax
Chickpea
Vicia dasycarpa
92.7 4.6 41.4 10.6
93.5 5.4 43.5 11.8
92.5 3.7 43.0 9.9
93.0 4 .8 45.4 10.8
91.0 2.4 41.5 10.4
91.3 4.2 38.8 6.7
92 3.5 33.6 4.6
91.5 1.7 40.0 4.9
92.2 4.0 43.4 10.3
93.5 9.7 47.3 9.6
91.5 5.1 52.5 4.4
90.5 4.9 49.5 4.2
93.0 9.2 49.8 4.3
93.5 9.8 41.2 11.2
433
Improvement in Nutritional Quality of Crop Residues
Straws are poor quality feed because they have a low ME content
(often less than 7.5 MJ/kg/ DM). Variations are caused by
varietal differences, season of planting, harvesting conditions
and height of cut.
The feeding value of straw depends on intake and
digestibility. To achieve maximum intake of straw a crude
protein content of 66-85 g/kg DM is necessary. Maximum intake of
DM has been observed when crop residues form 16-35% of the diet.
From studies at ILCA in Ethiopia (Mosi and Butterworth,
1985) showed that cereal crop residues are often less than 50%
digestible and intake is usually 50 g/kg LW 0.75 or less.
Preston and Leng (1987) reported that straws from various
species of grain crops appear to be highly variable in in vitro
digestibility and the crude protein content is always low
(Table 4) .
Table 4. In vitro OM digestibility and N content of straws from
different crops.
In vitro OM dig.
%
N content
% DM
Wheat straw
Oat straw
ice straw
Barley straw
Maize
Sorghum leaves
28-58
34-68
40-52
34-61
31-50
65-78
0.4-1.0
0.4-1.0
0.5-1.0
0.4-1.0
0.5-1.2
0.5-0.8
Generally, crop residues are poor quality feed so that their
digestibility is low, and also they contain low N which limits
voluntary intake and nutritional value.
Straw Treatments
Physical treatments
The most commonly used physical treatment has been grinding. It
reduces particle size and increases the surface area of the
straws exposed to rumen microbial action. Straws have the
disadvantage of being extremely bulky, making costs of
transportation high per unit of nutritive value. One advantage
of grinding or fine chopping is to reduce the bulk of the
materials.
Soaking straw in water just before feeding has also been
practised to increase dry-matter intake and hence digestible
organic-matter consumption (Chaturvedi et al . , 1973; Holzer et
al., 1975).
434
Sundstol (1981) and Nicholson (1981) have reported that high
pressure steam, high temperature treatment and irradiation of
straws - are effective in increasing the digestibility under
laboratory conditions. This practice appears to be costly and
might not be used in the foreseeable future at small-scale
livestock husbandry systems such as those in Ethiopia.
Chemical treatments
The most widely used chemicals to treat straws have been sodium
hydroxide (NaOH) and ammonia (NH3) . It has long been known that
soaking in lime (alkaline solution) improves intake and
digestibility of straw. Butterworth and Mosi (1986) (quoting
Jaasuriya) reported that treating roughages with sodium hydroxide
can increase their digestibility by 10-20 percentage points and
increase intake by 30-50%. Studies conducted in U.K. (FAO, 1977)
show that inclusion of up to 30% treated straw (with 5-6% NaOH)
in complete diets has not depressed production of milk or beef.
Ammoniation of crop residues using gaseous ammonia or
through wet ensiling (for 10-30 days) with urea (4-5%) has been
found applicable for practical use at present.
Preston and Leng (1987) reported that this practice
increases digestibility by 5-10% units, nitrogen content of the
straw, and acceptability and voluntary intake of the treated
straw by 25-50% as compared to untreated straw.
In Ethiopia, urea (46% N) is dissolved in water and
sprinkled over the residues and fed at the rate of 75-100
gm/day/animal in some beef production feedlots.
Supplementation
With molasses
In Ethiopia, over 50,000 tonnes of molasses is produced annually
of which about 55% is exported (Alemayehu Mengistu, 1985) . The
practice of using molasses as a supplementary feed for livestock
has not been well understood at present. Due to the drought
problem in 1984/85, the Ministry of Agriculture and ILCA jointly
conducted some studies and as a result of this work molasses/urea
mixtures are being used as a strategic feed reserve in the
drought stricken areas and as a routine feed in cattle fattening
farms. Molasses is used as a palatable carrier for urea and
minerals for improving the utilisation of crop residues.
Molasses/urea mixtures are used also as the basis of a supplement
to crop residues for routine feeding during the dry season. In
cattle fattening enterprises, Preston (1985) recommended that
molasses (at free choice) should be supplemented with 2.5% urea,
cereal straw (0.8% liveweight on dry-matter basis) and protein
meal about 0.25% of liveweight basis.
With oil cakes
Most of the oil seed plants such as noug (Guizotia abyssinica) ,
linseed, groundnuts, rapeseed, sesame and sunflower are widely
grown in Ethiopia. The cakes of these crops are widely used as a
protein supplement to low quality hays and crop residues.
435
With leguminous fodders
Leguminous forage crops which are rich in protein and usually
high in digestibility have been found beneficial as a supplement
to improve intake and digestibility of crop residues. From
studies conducted in Ethiopia, Mosi and Butterworth (1985) found
that addition of 20-25% Trifolium tembense hay to teff straw
increased feed intake of sheep by 20-30%. In another study
Butterworth and Mosi (1986) reported that where hay from
Trifolium is supplemented to residues of teff, oat, wheat and
maize at the rate of 30% by weight, dry-matter digestibility was
increased by about 10 percent units as compared to straws alone
(Figure 1) .
Olayiwole et al. (1986) have also obtained satisfactory
growth rates of crossbred dairy heifers by feeding cereal crop
residues supplemented with small amounts of maize and urea and
additional protein from Trifolium hay and noug meal (Table 5) .
Improving cereal crop residues by supplementing with green
forage legumes such as foliage from Leucaena, Sesbania , pigeon
pea or hays from Trifolium, alfalfa and vetch, is the most
practical method which can be adopted locally by small-scale
farmers. Growing forage legumes on land that could otherwise be
fallowed has the additional benefit of increasing the yield of
subsequent crops through the nitrogen fixed by the legume.
Table 5. Pooled mean values of total dry-matter intake (g/kg
LW0.75j and weight gain (g/day) by animals fed
different combinations of cereal crop residues (teff
and wheat), Trifolium hay, urea and noug meal.
-Trif +Trif No urea Urea Noug
or no
DM intake
noug
(gm/kg LW75)
Wt gain g/day
76.5
133
80.7
227
72.2
72
71.7
179
86.8
289
Source: Olayiwole et al. (1986).
Intercropping cereals with legumes such as Dolichos with
maize or sorghum (Haque, ILCA, Addis Ababa, Ethiopia, unpublished
data) can improve the nutritional value of the residues. This
practice appears more appropriate and a cost-effective way as
compared to other methods of improving the nutritional value of
crop residues.
436
Figure 1. Effect of addition of Trifolium hay on dry-matter
digestibility of cereal straws.
digestibility %
 
10 20 30 4O 5O
Trifolium hay content of feed (%)
Source: Butterworth and Mosi (1986)
437
CONCLUSION
In high potential agricultural zones of the Ethiopian highlands
more and more land is being cultivated for crop production every
year to satisfy the increasing human demand for food.
Consequently, the grazing lands are becoming scarce and crop
residues remain the main feed source for livestock. The limiting
factors with the utilisation of crop residues as livestock feed
are low voluntary intake and poor digestibility. Emphasis is
being given to assess implications of a more widespread use of
crop residues as livestock feed and possibilities of improving
nutritional value.
Various methods of physical and chemical treatments of
straws have been developed. Both treatment methods have been
found to require heavy investment on machinery followed by
regular availability or supply of replacement parts and chemicals
which would not be feasible for small-farmer level at present in
Ethiopia. Growing fodder legumes and use of these fodders as
supplement to crop residue is the most practical and cost-
effective method of improving nutritional value of crop residues.
This creates an opportunity for crop residues to play a useful
role in reducing weight losses of animals particularly during dry
periods.
Better use of crop residues ought to be stressed, and
research efforts must be directed to develop methodologies by
which the utilisation of this resource could be increased. Plant
breeders seldom consider the fact that crop residues are the
most, often the only, available livestock feed resource when
selecting crop varieties for high grain yields. Straw yield and
nutritional quality might have to be considered when new
varieties are developed.
Possibilities of sowing legumes under cereals and
improvement of the nutritional qualities of crop residues by
supplementations with leguminous fodders should be emphasised in
the future .
REFERENCES
Alemayehu Mengistu. 1985. Feed resources in Ethiopia. A paper
presented at the Workshop on Animal Feed Resources for Small
Scale livestock Producers, 11-15 November 1985, Nairobi,
Kenya. 12 pp.
Butterworth, M.H. and Mosi, A.K. 1986. The voluntary intake and
digestibility of combinations of cereal crop residues and
legume hay for sheep. ILCA Bulletin 24:14-17.
Chaturvedi, M.L., Singh, U.B. and Ranjhan, S.K. 1973. Effect of
feeding water soaked and dry wheat straw on feed intake
digestibility of nutrients and VFA production on growing
zebu and buffalo calves. J. Agric. Sci. (Cambridge) 80:393-
397.
Evaldson, O. 1970. An inventory of feeding system and feed
stuffs, Chilalo Awraja, Ethiopia. Minor Research Task at
438
CADU No. 5. CADU (Chilalo Agricultural Development Unit),
Asella, Ethiopia.
FAO. 1977. New feed resources. Proceedings of a technical
consultation held in Rome, 22-24 November 1976. FAO Animal
Production and Health Paper No. 4. FAO, Rome. 300 pp.
Gryseels, G. and Anderson, F.M. 1983. Research on farm and
livestock productivity in the central Ethiopian highlands,
1977-1980. Research Report No. 4. ILCA, Addis Ababa.
Holzer, Z., Levy, D. , Tagari, H. and Volcani, R. 1975. Soaking
of complete fattening rations high in poor roughage. I.
The effect of moisture content and spontaneous fermentation
on nutritional value. Anim. Prod. 21:323-335.
Jahnke, H.E. 1982. Livestock production systems and livestock
development in tropical Africa. Kieler Wissenschaftsverlag
Vauk, Kiel, Germany. 253 pp.
Mosi, A.K. and Butterworth, M.H. The voluntary intake and
digestibility of diets containing different proportions of
Teff (Eragrostis tef) , straw and Trifolium tembense hay when
fed to sheep. Tropical Animal Production. 10(l):19-22.
Nicholson, J.W.G. 1981. Nutritional and feeding aspects of the
utilization of processed lignocellulosic waste materials by
animals. Agric. Environ. 6:205-228.
Nuwanyakpa, M. , Preston, T.R. and Sayers, R. 1986. Effects of
watering frequency on the response of sheep to
supplementation with molasses/urea and noug cake (Guizotia
abyssinica) . In: T.R. Preston and M.Y. Nuwanyakpa (eds) ,
Towards optimal feeding of agricultural by products to
livestock in Africa. ILCA, Addis Ababa. pp. 93-101.
Olayiwole, M. , Butterworth, M.H., Sayers, A.R. and Olorunju,
S.A.S. 1986. The effect of supplementing cereal straws with
urea, Trifolium hay and noug meal on feed intake and
liveweight gain of growing crossbred heifers. ILCA Bulletin
24:18-19.
Preston, T.R. 1985. Molasses as animal feed supplement with
particular references to Ethiopia. Forage Network
Newsletter No. 7, ILCA, Addis Ababa.
Preston, T.R. and Leng, R.A. 1987. Matching ruminant production
systems with available resources in the tropics and
subtropics. Penambul Books, Armidale, Australia. 245 pp.
Reed, J.D., Tedla, A. and Jutzi, S. 1986. Large differences in
digestibility of crop residues from sorghum varieties.
ILCA, Addis Ababa.
Sundstol, F. 1981. Methods for treatment of low quality
roughages. In: J. A. Kategile, A.N. Said and F. Sundstol
(ed) , Utilization of low quality roughages in Africa. AUN-
Agri. Dev. Report 1. Aas, Norway, pp. 61-80.
439
NUTRIENT QUALITY OF FORAGES IN ETHIOPIA
WITH PARTICULAR REFERENCE TO MINERAL ELEMENTS
E. Kabaija and D.A. Little
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
Abstract
Hays, range grasses, crop residues and browse trees from
different parts of Ethiopia were analysed for nitrogen, fibre
components and mineral concentrations. Pasture grasses and crop
residues studied had very low crude protein contents, while the
browse on the average had more than 15% crude protein. Almost
all the forages had sufficient levels of K, Ca, Mg, Kn and Zn to
meet requirements of ruminant animals, but the occurrence of
marginal to deficient supplies of Na, Cu and P appear very
likely. There is an urgent requirement for experimentation on
supplementation regimes involving these minerals, so that
appropriate recommendations can be formulated.
INTRODUCTION
In Ethiopia, as elsewhere in Africa, malnutrition impairs
livestock production. Ethiopia's overall livestock productivity
is below average. Although 12.7% of Africa's 524.61 million
cattle, sheep and goats are found in Ethiopia, the country
produces only 7.3% and 5.1% of Africa's total meat and milk
production respectively (FAO, 1985) . Grazing animals in Ethiopia
subsist mainly on poor quality feedstuffs in the form of poor
quality pastures in arid and semi-arid areas and hays and/or crop
residues in the arable areas. In a few of the animal production
centres where improved management is undertaken, the main
supplements are energy and protein in the form of agro-industrial
by-products such as cereal brans, molasses and oilseed cakes.
Scant attention is given to the mineral content and nutritional
balance of such diets. It has been widely established that
available energy and protein of a feed are of primary importance
to any animal but optimal performance is only possible if there
is an adequate supply of minerals and vitamins (McDowell, 1985) .
In Ethiopia, cattle rarely receive mineral supplements
except occasionally common salt. Pastures are thus the main
source of minerals, and only rarely can forages completely
satisfy all mineral requirements of livestock (Miles and
McDowell, 1983) . Mineral status of grazing animals in Ethiopia
and other African countries has received very little attention.
This may be due partly to the fact that the methodology of
mineral nutrition studies, especially of trace elements, is
rather complicated and signs of marginal mineral deficiencies
are not easily detected but frequently it is mistakenly assumed
that the grazing animal will obtain its mineral needs from the
pasture. Mineral deficiencies, even if marginal, can result in
depression of animal performance. Subclinical mineral
deficiencies are often widespread and are responsible for as yet
unestimated, but probably great, economic losses in livestock
production.
440
An earlier mineral study in Ethiopia (Faye et al., 1983)
indicated probable widespread copper and zinc deficiencies, and
although much more information on the supplies of essential
minerals for livestock is required, it is very likely that
mineral deficiencies contribute to the poor performance of
livestock in Ethiopia. Before measures can be undertaken to
correct these deficiencies, it is necessary to assess the
mineral status of forages and of the grazing animals as well as
their production responses to mineral supplementation. This
study was undertaken to investigate the nutrient status, with
special regard to minerals, of the common crop residues, pasture
grasses, grass hays and browse trees used as livestock feeds in
Ethiopia.
METHODOLOGY
Samples
Samples of meadow hay were collected from an ILCA site in Addis
Ababa and from two other sites at Debre Berhan (an ILCAstation
120 km east of Addis Ababa and a Government sheep breeding
station 132 km from Addis Ababa) . Samples were also taken from a
commercial source in Addis Ababa, whose supplies came from the
Addis Ababa region. All the samples were from the highlands
(2500-3000 m) and were obtained by pooling core samples from ten
positions within the stack. Samples of oats hay (whole plant
with grain) , ryegrass hay and improved pasture hay (mostly
ryegrass and native Trifolium species) were taken as described
above from stacks of hay at ILCA sites in Debre Berhan harvested
between September and November, 1986.
Sesbania sesban grown at ILCA, Addis Ababa and in Yabello in
the southern Ethiopian rangelands of Sidamo were harvested in
February, 1987. Samples were taken by pooling leaves from mature
trees (50 in Addis Ababa and 15 in Sidamo) . Leaves from the
trees in Addis Ababa were further separated into young (first ten
leaves on twig) and old. Pods with seeds on trees from Addis
Ababa were first air-dried and then separated into pods and
seeds. Samples of Leucaena leucocephala cv Peru were taken in
February, 1987 from a lowland area at a Government research
station (Institute of Agricultural Research, Melka Worer) in
eastern Ethiopia using the same procedures as for Sesbania.
Grass and browse species were collected from the Sidamo
southern rangelands in February, 1987 by hand plucking during
grazing observations over three days. The species collected
included: Cenchrus ciliaris, Themeda triandra, Chrysopogon
aucheri, Pennisetum mezianum. Acacia brevispica and Acacia
nilotica. The Acacia samples were of the species mainly browsed
by camels and goats.
Samples of crop residues were taken using the same
procedures as for hays. The crop residues were of highland
origin, and most of them were being used in feeding experiments
at ILCA, Debre Zeit and ILCA, Addis Ababa.
441
chemical Analysis
Dry matter was determined after drying in a microwave oven at
60 c for 24 hours. The dried samples were ground using a one-mm
sieve and stored in plastic bottles ready for analysis. Neutral-
detergent fibre (NDF) , acid-detergent fibre (ADF) , lignin and ADF
ash were determined with a microfibre apparatus using the
procedures of Goering and van Soest (1970) .
Samples were prepared for mineral analysis by the wet
digestion method using concentrated sulphuric acid in presence of
hydrogen peroxide. Nitrogen (N) and phosphorus (P)
concentrations in the extracts were determined with the
autoanalyser which uses ascorbic acid as a reducing agent for P
determination. The concentrations of potassium (K) , sodium (Na) ,
calcium (Ca) , magnesium (Mg) , iron (Fe) , manganese (Mn) , zinc
(Zn) and copper (Cu) were determined by atomic absorption
spectroscopy. Sulphur (S) in the samples was quantified using
the turbidimetric method of Cottenie et al. (1982).
RESULTS AND DISCUSSION
The analytical results for hay and range grasses are shown in
Tables 1 and 2 , for crop residues in Tables 3 and 4 and for
leguminous trees in Tables 5 and 6.
Table 1. Dry matter, crude protein, NDF, lignin, ash and ADF ash
of some hays from Ethiopian highlands and range grasses
from southern Ethiopian rangelands.
Sample description DM CP NDF ADF Lignin Ash ADF
ash
Highland hays
Meadow hay, DBSBS 93.4 5.3 69.7 37.9 4.0 7.5 2.4
Meadow hay, ILCA-DB 93.1 4.4 69.2 37.9 4.9 7.9 2.8
Oats hay, ILCA-DB 91.2 5.1 71.0 38.1 4.5 4.6 1.5
Rye grass hay, ILCA-DB 92.5 6.2 57.5 35.7 5.0 7.4 2.6
Improved pasture hay, 92.0 5.3 58.7 32.7 4.0 6.5 3.1
ILCA-DB
Meadow hay, ILCA-Addis 94.1 4.6 72.2 43.7 5.7 8.7 3.4
Meadow hay from 91.9 3.7 77.0 41.1 5.1 8.4 3.4
commercial centre
Range grasses
Cenchrus ciliaris 58.8 7.5 66.2 37.2 5.1 11.3 4.06
Themeda triandra 84.7 5.0 77.2 50.2 8.0 10.9 7.61
Chrysopogon aucheri 59.1 6.0 71.2 36.0 5.2 11.5 5.80
Pennisetum mezianum 78.0 6.3 74.1 46.0 8.5 10.7 4.60
442
Table 2. Mineral content of some hays from Ethiopian highlands
and range grasses from Ethiopian Sidamo southern
rangelands.
Sample description K Na Ca Mg P S Fe Mn Zn cu
Highland hays
g/kg
Meadow hay, DBSBS 17 6 0.4 3.4 1.6 1 6 3 0 695 207 45 5.9
Meadow hay, ILCA-DB 18 4 0.3 4.4 1.6 1 4 2 6 974 274 39 6.1
Oats hay, ILCA-DB 12 6 0.9 1.6 1.0 1 8 1 8 192 96 30 7.0
Ryegrass hay, 16 3 1.1 3.8 1.7 2 2 2 1 397 214 32 7.9
ILCA-DB
Improved pasture 20 8 1.2 3.7 1.7 1 6 2 6 358 206 28 7.1
hay, ILCA-DB
Meadow hay, 17 5 0.6 4.9 1.7 1 4 2 4 191 246 29 6.2
ILCA-Addis
Meadow hay, from 13 1 0.5 4.6 2.1 1 6 3 0 589 322 37 7.9
commercial centre
Range grasses
Cenchrus ciliaris 23 5 1.0 2.0 1.6 1 5 2 1 830 45 24 3.6
Themeda triandra 12 0 0.1 4.1 1.5 1 2 1 0 625 72 31 3.5
Chrysopogon aucheri 1 0 0.08 2.2 1.6 1 0 1 1 882 63 32 5.6
Pennisetum mezianum 12 0 0.1 3.2 2.0 1 4 1 1 452 72 40 4.7
Table 3. Dry matter, crude protein, ndf, adf, lignin, ash and
ADF ash of some crop residues from Ethiopian highlands.
DM CP NDF ADF
(%)
Lignin Ash ADF ash
Wheat straw 90.5 1.6 70.7 50.6 11.1 10.8 5.5
Barley straw 88.2 1.8 65.4 56.2 7.0 7.7 1.7
Teff straw 89.7 1.7 71.3 49.1 7.9 7.8 3.7
Oats straw 88.9 1.5 57.2 41.9 6.0 8.3 2.1
Linseed straw 93.0 3.0 68.7 - - 8.4 3.1
Maize stover 87.2 1.5 68.1 47.1 7.9 9.2 3.1
Table 4. Mineral concentration of some crop residues from the
Ethiopian highlands.
Wheat straw
Barley straw
Teff straw
Oats straw
Linseed straw
Maize stover
K Na Ca Mg P S Fe Mn Zn Cu
g ——— —— — ———— -•-.. mg/kg
14.8 0.3 4.1 1.5 1.3 1.4 325 78 11 3.0
10.7 0.5 4.6 1.4 1.9 1.2 1175 90 12 5.0
11.7 0.3 4.3 1.9 1.6 1.6 170 59 26 6.5
17.7 0.4 3.9 1.8 1.7 1.7 196 191 17 14.0
10.7 0.6 7.7 2.2 1.3 1.6 103 71 24 14.0
17.8 0.5 3.3 2.5 1.7 0.9 408 61 24 5.9
443
Table 5. Dry matter, crude protein, NDF, AOF, lignin, ash and
ADF ash of Leucaena leucocephala, Sesbania sesban and
Acacia species.
Sample DM CP NDF ADF Lignin Ash ADF
description ash
Leucaena:
Young leaves 20.7 31.2 25.1 10.6 0.3 6.5 -
Old leaves 34.1 28.6 23.4 13.4 2.4 8.9 -
Pods 79.1 7.9 67.8 61.1 19.6 5.7 -
Seeds 97.3 34.8 33.9 21.2 2.1 4.2
Sesbania:
Small twigs with
leaves 29.0 27.5 14.0 12.4 2.8 12.4 0.3
First ten leaves 31.3 28.5 13.3 9.2 2.4 6.3 0.2
Old leaves 30.4 28.5 12.7 9.8 2.6 7.8 0.4
Pods with seeds 27.5 21.0 50.4 40.4 6.6 2.7 0.5
Pods alone 90.8* 10.6 60.9 52.6 13.9 6.4 0.3
Seeds alone 89.8** 34.6 33.0 22.4 2.6 3.7 0.3
Old and young
leaves* 33.6 25.9 13.0 9.9 1.7 13.6 0.8
Acacia:
Acacia brevispica 38.9 26.3 47.3 29.0 13.0 8.4 0.60
Acacia nilotica 40.1 15.0 30.7 28.1 11.0 5.9 0.60
Taken from the low-mid altitude southern Ethiopia rangelands.
First air-dried.
Table 6. Mineral content of Leucaena leucocephala,
sesban, and Acacia species.
Sesbania
Sample K Na Ca Mg P S Fe Mn Zn CU
description -g/kg
Leucaena :
Young leaves 20..4 0.3 3.7 1.9 3.1 3.4 181 31 39 10.0
Old leaves 16,,7 0.3 18.4 3.9 2.1 2.5 214 53 19 11.8
Pods 13,,6 0.4 15.4 0.6 0.4 - 198 43 11 3.4
Seeds 15,,6 0.3 3.4 1.9 3.8 4.9 131 29 40 11.0
Sesbania:
Small twigs
with leaves 13,,0 1.2 21.3 2.8 2.1 2.3 274 165 41 9.5
First ten
leaves 10. 9 1.2 23.4 2.7 2.1 2.3 285 200 48 8.1
Pods with
seeds 19.,9 1.0 6.3 1.2 2.6 1.7 109 45 37 5.2
Pods alone 12,.8 1.1 12.0 0.9 1.1 1.2 581 63 31 5.3
Seeds alone 9.,3 0.8 3.7 1.4 4.6 1.9 181 52 52 7.6
Old and young
leaves 13.,2 1.0 28.3 4.9 2.4 3.2 280 410 62 6.4
Acacia
brevispica 16.,9 1.04 7.7 2.9 2.0 3.4 151 137 46 6.9
Acacia nilotica 8..1 0.5 18.0 1.9 2.2 2.3 664 66 31 9.0
* Taken from the low-mid altitude southern Ethiopian rangelands.
444
All the pasture grasses and crop residues studied had very
low crude protein contents which is characteristic of mature
tropical forages whose protein content declines rapidly following
the rapid growth of the rainy season. Those figures were all
below the level of 7.5% considered to be required for optimum
rumen function (van Soest, 1982) .
For reasonable levels of production, animals subsisting on
these forages will therefore require supplementary protein which
may be in the form of oilseed cakes and/or NPN sources such as
urea which could be easily obtained locally. While use of such
sources is practicable on smallholder units, it is not easy to do
so for cattle on range in pastoral areas. In such situations it
is possible to achieve some success through the use of high
protein browse and tree crops such as Sesbania, Leucaena,
Gliricidia and Acacia species (Table 5) . On smallholdings in the
highlands, where feeding hay is practised, effort could be made
to improve the quality of the hay by practising early cutting,
although the problem here is to compromise quantity with quality.
Since hay is now being marketed in Ethiopia, a degree of quality
control could be achieved if analyses were carried out and a
system of price premium for good quality was introduced. The
high levels of ADF ash in most of the grasses indicates the
presence of large amounts of silica which may seriously reduce
digestibility (van Soest, 1982) .
In relation to the essential mineral elements, the level of
K in all the forages was above the level of 8 g/kg recommended
for grazing animals (Underwood, 1981) . It has, however, been
suggested (McDowell, 1985) that weaned calves and high producing
dairy cows under stress, such as heat stress, may require K level
above 10 g/kg, but since only the straws of barley, teff and
linseed approached this figure, it seems most unlikely that
problems of K deficiency are likely to arise. The same
conclusion appears valid for S, Fe and Mn, the contents of which
in almost all the forages studied were above the levels of 1
g/kg, 50 mg/kg and 40 mg/kg respectively proposed as adequate for
grazing animals (McDowell, 1985). The requirement for S is
basically related to that for N and S-containing amino acids,
such that the ARC (1980) suggested that an N:S ratio of 14:1 is
indicative of adequacy of S; in this regard, all of the materials
examined are satisfactory.
With reference to Na, there is some debate in the literature
concerning the dietary concentration required. While Underwood
(1981) recommended 1 g/kg for most grazing animals, the findings
of Morris (1980) and Little (1987) indicate that 0.7 g/kg is
adequate for non-lactating cattle, and the ARC (1980) suggested
that a level of 1.4 g/kg is required by sheep. The present data
show that with the exception of Sesbania, most of the materials
examined were very poor sources of Na, such that routine
supplementation is likely to be necessary. This characteristic
of Sesbania has been recorded elsewhere in the tropics (Little,
1986) .
while dietary Ca concentrations of 2-6 g/kg, with higher
requirements for lactation have been variously recommended for
cattle and sheep (NRC, 1978, 1984, 1985; ARC, 1980), the findings
of Sykes and Field (1972) suggest that levels of 2.5-3.0 g/kg are
445
adequate in most circumstances. The results indicate that
problems of Ca deficiency would not be expected. It is still
widely believed that a large excess of Ca over P, resulting in
Ca:P ratios approximating 10 or more, is deleterious to
ruminants, although much conflicting evidence occurs in the
literature (reviewed by Little, 1970) . In this context it is
noteworthy that the ARC (1980) concluded that "...it la not
possible to state the optimal ratio of calcium to phosphorus for
animal performance or whether such a ratio actually exists."
Where wide ratios occur, the dietary concentration of P per se is
almost certain to be inadequate.
Underwood (1981) considered a dietary P level of 1.7 g/kg to
be marginal for grazing animals, in essential agreement with work
of Little (1980, 1985) which indicated that a figure of 1.4 g/kg
should be regarded as minimal for growing cattle. The data in
Tables 2 and 4 show that most grasses and crop residues examined
were marginal to deficient in P, and supplementation with P is
likely to be beneficial. For animals grazing on range, this
would best be supplied through incorporation into complete
mineral licks or in mixtures with common salt and bone meal.
Animals on smallholder units, which may receive some concentrate
feeds, may receive adequate P if offered high protein oilseed
cakes such as cotton seed cake and noug cake.
McDowell et al (1978) considered 30 mg/kg to be a critical
level of dietary Zn, although the ARC (198 0) has suggested that
concentrations of 12-20 mg/kg are adequate for growing cattle.
The crop residues may thus constitute a marginal supply of Zn
(Table 4) ; the necessity for supplementary Zn needs to be kept
under review particularly for sheep, which require some 35 mg
Zn/kg diet (ARC, 1980) .
It is commonly accepted that the dietary requirement of
cattle for Cu lies in the range 8-14 mg/kg (ARC, 1980; NRC, 1978,
1984) . Clearly most of the materials examined will provide a
marginal to deficient supply of this mineral. This situation may
be even further complicated by high levels of dietary Fe which
can be elevated by soil ingestion during grazing. Humpries et
al. (1981) showed that dietary concentrations exceeding 1 g Fe/kg
can profoundly reduce the availability of ingested Cu; relatively
slight and doubtless common levels of dietary soil contamination
can produce Fe concentrations of this order (Healy, 1973) ,
necessitating the serious consideration of supplying
supplementary Cu, either by injectable preparations, oral dosing
with copper oxide needles or providing mineral licks containing
Cu. In this context it should be remembered that the Cu
requirement of sheep is only about 5 mg/kg however, so that the
provision of supplementary Cu to this species is much less likely
to be required.
It should be stressed that data of the type presented here
can provide only an indication of the existence of potential
mineral deficiency problems, since animal selectivity usually
results in the consumption of material of somewhat higher quality
than that of the total available, and conclusive diagnosis must
be based on the occurrence of a positive response to
supplementary supply of the mineral in question. However, such
data are vital in the formulation of critical supplementation
446
experiments, upon which recommendations for practical
supplementation regimes should be based.
CONCLUSION
The pasture and crop residues examined had very low crude protein
contents, and for increased levels of animal production the
primary need is for supplementary protein. The most appropriate
means of providing this will vary with local circumstances;
protein-rich oilseed cakes or urea-containing blocks are valuable
sources where they are available. But there is great potential
for encouraging the use of leguminous trees, as well as the
establishment of forage legumes as crops or oversown in the
pasture.
The minerals most widely present in inadequate amounts are
Na, Cu and P, and supplementation regimes involving these
elements are very likely to produce beneficial results. There is
an urgent need for appropriate experimentation so that
soundly-based supplementation packages can be devised.
REFERENCES
ARC. 1980. The nutrient requirements of ruminant livestock.
Commonwealth Agricultural Bureaux, Farnham Royal, U.K.
Cottenie, A., Veroo, M. , Jiekens, L. , Velgha, G. and Camerlynch,
R. 1982. Chemical analysis of plants and soils. I.W.O.N.,
Ghent-Belgium.
FAO. 1985. Production yearbook 39. FAO, Rome.
Faye, B. , Grillet, C. and Tessema, A. 1983. Report on survey of
trace element status of forages and blood of domestic
ruminants in Ethiopia. National Veterinary Institute, Debre
Zeit, Ethiopia.
Goering, H.K. and van Soest, P.J. 1970. Forage fibre analysis,
Agric. Handbook 379. Agricultural Research Service, USDA 0
ARS, Beltsville, MD.
Healy, W.B. 1973. Nutritional aspects of soil ingestion by
grazing animals. In: G.W. Butler and R.W. Bailey (eds) ,
Chemistry and biochemistry of herbage. Vol. 1. Academic
Press, London, pp. 567-588.
Humpries, W.R., Young, B.W., Phillippo, M. and Bremner, I. 1981.
The effects of iron and molybdenum on copper metabolism in
cattle. Proceedings of the Nutrition Society 40:68A.
Little, D.A. 1970. Factors of importance in the phosphorus
nutrition of beef cattle in northern Australia. Australian
Veterinary Journal 46:241-248.
Little, D.A. 1980. Observations on the phosphorus requirement of
cattle for growth. Research in Veterinary Science 28:258-
260.
447
Little, D.A. 1985. The dietary mineral requirements of
ruminants; implications for the utilization of tropical
fibrous agricultural residues. In: P.T. Doyle (ed.), The
utilization of fibrous agricultural residues as animal
feeds. IDP, Canberra, Australia. pp. 34-43.
Little, D.A. 1986. The mineral content of mineral feeds and the
potential for mineral supplementation in South-East Asia,
with particular reference to Indonesia. In: Ruminant
feeding systems utilizing fibrous agricultural residues,
1985. IDP, Canberra, Australia. pp. 77-86.
McDowell, L.R. 1985. Nutrition of grazing ruminants in warm
climates. Academic Press, New York.
McDowell, L.R., Houser, R.H. and Tick, K.R. 1978. Iron, zinc and
manganese in ruminant nutrition. In: J.H. Conrad and L.R.
McDowell (eds) , Latin American symposium on mineral
nutrition research with grazing ruminants. University of
Florida, Gainesville, Florida. pp. 108-116.
Miles, E.H. and McDowell, L.R. 1983. Mineral deficiencies in the
Llanos rangelands of Colombia. World Animal Review 46:2-10.
Morris, J.G. 1980. Assessment of sodium requirements of grazing
beef cattle: a review. Journal of Animal Science 50:145-152.
NRC. 1978. Nutrient requirements of domestic animals. Nutrient
requirements of dairy cattle. 5th edition. National
Academy of Sciences, Washington, D.C.
NRC. 1984. Nutrient requirements of domestic animals. Nutrient
requirements of sheep. 6th edition. National Academy of
Sciences, Washington, D.C.
NRC. 1985. Nutrient requirements of sheep. 6th edition. National
Research Council, National Academy Press, Washington, D.C.
Sykes, A.R. and Field, A.C. 1972. Effects of dietary
deficiencies of energy, protein and calcium on the pregnant
ewe. I. Body composition and mineral content of the ewes.
Journal of Agricultural Science (Cambridge) 78:109-117.
Underwood, E.J. 1981. The mineral nutrition of livestock. 2nd
edition. Commonwealth Agricultural Bureau, London.
Van Soest, P.J. 1982. Nutritional ecology of the ruminant. 0
and B Books, Oregon.
448
EFFECT OF PASTURE MINERAL LEVELS ON EXTENSIVE
CATTLE PRODUCTION IN KENYA
Aggrey Abate
Department of Animal Production
University of Nairobi
P.O. Box 29053, Nairobi, Kenya
Abstract
Mineral levels of Kenyan pastures were examined with the
objective of determining whether they were sufficient to meet the
requirements of grazing cattle. Potassium (K) was found to be
adequate in the herbage while grazing animals are likely to meet
their requirements for iron (Fe) through incidental ingestion of
soil. The status of magnesium (Mg) and calcium (Ca) needs
resolving because adequacy and deficiency have both been
reported. Other minerals were, however, found to limit extensive
cattle production because they were deficient particularly in the
dry season. These included phosphorus (P) , sodium (Na) , copper
(Cu) and cobalt (Co) . There was little information on minerals
in the rangelands and further research in this area was
suggested. Of the preventive measures recommended, provision of
mineral supplements was found most appropriate. It was, however,
emphasised that for effectiveness of any mineral nutrition
programmes, adequate energy and protein nutrition has to be
maintained particularly in the dry season.
INTRODUCTION
In Kenya, as in many developing countries natural pastures are
the main source of nutrients for cattle. As a consequence,
seasonality in quantity and quality of available herbage has a
marked influence on animal productivity. Poor animal performance
has usually been interpreted in terms of low dry-matter (DM)
intake and inadequacy of energy and protein in the DM (Glover and
Dougall, 1961; Abate et al . , 1981; Abate, 1985; Abate and Abate,
in press) . Little mention is made of the effect essential
elements have on production mainly because under controlled
experimental conditions mineral supplements are normally fed.
Yet surveys carried out in the country have revealed mineral
deficiencies in Kenyan pastures (Burdin and Howard, 1963; Howard,
1969; Mwakatundu, 1977) with obvious consequences on animal
development and reproduction (Todd, 1954; Howard et al., 1962).
This paper aims to examine mineral levels in Kenyan pastures
with emphasis on the effect of deficiencies on cattle production.
EARTH LICKS AS SOURCES OF MINERALS
Early indications of mineral deficiency in Kenyan pastures
followed observations on earth eating and oesteofagia by cattle
and wild ruminants (Todd, 1954; French, 1955; Howard, 1963).
Table 1 shows the composition of two such natural licks. Hudson
(1944), French (1955) and Howard (1963) concluded from other
similar analyses that natural licks contained insufficient sodium
449
(Na) while the levels of phosphorus (P) and calcium (Ca) were too
low to alleviate a deficiency. Concentrations of as high as 15.1
and 78.1% of sodium chloride (NaCl) have, however, been reported
in the DM of natural salt licks in parts of North Eastern
Province (Hudson, 1944) . The availability of mineral elements
from earth licks is also probably low since only up to 6% of the
organic matter is soluble in dilute acid (Table 1) . In spite of
their unlikelihood as sources of minerals, cattle owners still
regarded such salt licks as essential to the well being of their
animals (Hudson, 1944) .
Table 1. Mineral composition (% DM) of salt
different areas in Machakoa District.
licks from two
Email Koru
Total organic matter, %
Inorganic matter soluble
in 0.125N HC1, %
Percent DM
Calcium
Magnesium
Sodium
Potassium
Phosphorus
Iron
Copper
98.6
1.8
0.03
0.40
0.02
0
Trace
0.04
0.0008
96.8
6.5
0.80
0.10
0.10
Trace
0.01
0.18
0.001
Source: Hudson (1944) -
BORDERLINE TO SUFFICIENT MINERALS
Potassium (K) concentration in Kenyan forages are quite high
(Abate, 1984; unpublished data); levels compare well to those in
productive grasses elsewhere in the world (Howard, 1963) and are
adequate in meeting the requirements of grazing livestock (Howard
et al., 1962). Because of large quantities in the soil, animals
are also likely to augment their iron (Fe) supplies through
direct ingestion of soil or from soil-contaminated herbage
(French, 1955; Hodgson et al., 1962) and would, therefore, suffer
no deficiency. For some minerals, the risk is either little or
there is no evidence of deficiency in grazing animals (French,
1955; Mwakatundu, 1977). Depending on the area, other minerals
may be borderline to sufficient. Thus, Howard (1963) has
indicated that there may be a need to topdress Kenyan pastures
with magnesium (Mg) fertilizers because levels of the element in
grazings were similar to those that predispose animals to grass
tetany. But Mwakatundu (1977) maintained that Mg was not
limiting in Kenya because the element was sufficient in soils,
pastures and bovine plasma. Also a suspected case of
hypomagnesemia in the Naivasha area was not confirmed by serum
analysis (Howard, 1963) . The position with regard to Ca equally
requires defining. Howard (1963) found Ca levels in Kenyan
pastures satisfactory but Mwakatundu (1977) reported that areas
in Central and Rift Valley provinces were deficient in the
mineral. A national average value of 0.53% has been quoted
450
(Howard, 1963) although a lower concentration of 0.39% has also
been reported for ley and natural pastures at Muguga, Central
Kenya (Howard et al., 1962). On account of a comparatively lower
productivity, a Ca deficiency is unlikely to occur in beef herds
because of the ability to mobilise body depots. A proper Ca:P
ratio is, however, physiologically desirable for high fertility
and adequate rate of growth irrespective of level of production.
Results of studies carried out in Baraton, Egerton, Kabete and
Molo suggested that the Ca:P ratios of the grazings were
satisfactory for ruminants (Mwakatundu, 1977) .
DEFICIENT MINERALS
Based on the available literature, the minerals most likely to
limit cattle production in Kenya are P, Na, copper (Cu) and
cobalt (Co) . Certain trace minerals like selenium (Se) and zinc
(Zn) could, however, also be limiting.
Phosphorus
In a mineral survey Howard (1963) reported that 36% of natural
pastures analysed contained less than 0.3% P while the average
value for the country was 0.41%. Levels were lower in the dry
than in the wet season. Todd (1954) analysed samples of Chloris
gayana, Bothriochloa insculpta and Brachiaria dictyoneura and
found them to contain as low as 0.13% P in the DM in the dry
season. These results were similar to those of Howard et al.,
(1962) who reported a range in P content of 0.072-0.252% in
grazed pastures in the high potential areas of Kenya; at the
lower level P was deficient and this occurred during dry spells.
Mwakatundu (1977) also confirmed that in Kenya P deficiency was
widespread as shown by low levels in soil, pasture and bovine
plasma. The concentration of P depended largely upon the extent
of recent precipitation and was highly correlated (r = 0.72**) to
CP levels in the pasture (Howard et al . , 1962). Mwakatundu
(1977) has reported similar results.
The requirements for P by grazing Kenyan cattle are not
known. It is probable, however, that pasture levels of P are
sufficient to maintain slow maturing and low producing indigenous
animals but is inadequate for high producing improved breeds
(French, 1955) . Suboptimal P levels in grazings have resulted in
the following in Kenyan cattle: loss of appetite hence reduced
feed intake, retarded growth, interference with regularity of
heat hence reproductive rates, lowered milk production and, in
extreme cases, depraved appetite or pica (Howard, 1963) .
Sodium
The concentrations of Na in pastures throughout Kenya are very
low (Howard, 1963) , particularly in the dry season (French,
1955) . Wet season grass may, however, also contain Na levels
that are insufficient in meeting animal requirements (French,
1955) . This is because rapid gain in weight by animals during
the wet season induces high mineral requirements (McDowell et
al., 1984). Of all pastures analysed in Kenya, 77% contained
0.01% or less Na; levels which are inadequate to cover the
demands of dairy and improved beef cattle. A range of 0.007
451
0.022% Na has been reported for samples obtained from ley and
permanent pastures of Cynodon dactylon and Chloris gayana in
Muguga, Central Kenya (Howard et al . , 1962). Severe symptoms of
Na deficiency have not been reported in Kenya mainly because most
farmers feed salt to their animals.
Copper
Disorders such as stilted gait, rusty coat colour and, in severe
cases, fragility of bones in calves have been identified in
Kenya as due to Cu deficiency (French, 1955; Howard, 1963).
Copper concentrations in Kenyan pastures seem to lie between 4.0
and 12.2 ppm (Howard et al., 1962). Deficiency normally shows
itself at levels lower than 5.0 ppm, and this has been shown in
about 35% of pastures analysed in Kenya (Howard, 1963) . Shortage
of Cu has been recorded in grazings along the Rift Valley and in
parts of Central and Rift Valley provinces (Howard, 1963) because
of a Cu deficiency in the soil (Pinkerton et al., 1965). Soils
in which the underlying rock is ash and pumice may be expected to
be deficient in Cu (Nyandat and Ochieng, 1976) . Mwakatundu
(1977) also found subclinical Cu deficiency widespread in Kenya
as a result of Cu deficiency either in the soil or pasture or
bovine plasma and is more pronounced in dry periods (Nyandat and
Ochieng, 1976) . The functioning of dietary Cu can be inhibited
by excess molybdenum (Mo) resulting in a conditioned Cu
deficiency. Animals consuming forages with Mo concentrations
above 15 to 2 0 ppm showed Cu deficiency symptoms even though the
Cu levels in the forage were higher than 5 ppm (Hodgson et al.,
1962). Conditioned Cu deficiency does not seem a problem in
Kenya because Mo was found deficient in several areas surveyed
(Mwakatundu, 1977).
Cobalt
Cobalt is a serious mineral limitation to livestock because even
when grazing is abundant deficiency will lead to chronic
starvation or wasting which is often indistinguishable from
energy and protein malnutrition (French, 1952; Howard, 1963;
McDowell et al . , 1984). It is rare for grasses to contain Co in
concentrations that meet the demands of grazing animals (Hodgson
et al . , 1962). When the content in the pastures herbage is 0.10
ppm or less (Hodgson et al., 1962) grazing animals are likely to
suffer from Co deficiency. In Kenya, the condition is known as
"Nakuruitis" (French, 1952; Howard, 1963) or as "Narurasha" among
Maasai herdsmen (Hudson, 1944; French, 1952). Cobalt deficiency
is common along the Rift Valley and is seasonal in character with
symptoms usually appearing after the rains when grazing is
plentiful and green (Hudson, 1944) . Mwakatundu (1977) also
confirmed Co deficiency in the Egerton, Kabete and Molo areas of
Kenya. Animals suffering from Co deficiency lose appetite and
condition, may abort if in calf or may have difficulty to
conceive again; the condition seems to affect lactating cows more
than any other type of stock (Hudson, 1944; French, 1952).
MINERAL LEVELS IN THE RAKGELAND8
The literature cited above refers mainly to mineral deficiencies
in the high potential areas of Kenya which are characterised by
452
about 1200 mm of rainfall per annum. Reports on the effects of
minerals on animal production in the rangelands of Kenya are
lacking.
About 85% of the total land area of Kenya is rangeland
unsuitable for arable cropping because it receives only 500-750
mm of rainfall per year. The rangelands are, however, important
because they contain about 50% of the total livestock population
in the country. The cattle population here consists mainly of
low production indigenous stock. In these areas grazing forms
the only source of feed and the natural vegetation is dominated
by Commiphora, Acacia and Themeda associations with a carrying
capacity of 4 ha to a stock unit.
In Table 2 are given lists of preferred grass species in the
rangelands of Kiboko (Hatch et al . , 1984) and around Isiolo
(Schwartz, H.J. pers. comm.). For most of the year the quality
of the herbage is low. Karue (1974) stated that the level of
digestible energy for most of the grasses in the dry zones of
Kenya is inadequate to meet the demands for maintenance of
grazing stock in the wet and dry seasons. Tessema (1986)
reported that the grasses of the semi-arid areas of Eastern
Province had high proportions of structural carbohydrates which
were deposited in the plant tissues at an early vegetational
stage; the protein content on the other hand dropped from 11 to
4% at six months of age.
Table 2. Composition of the most important grass species in
rangelands around Kiboko and Isiolo.
the
Kiboko Isiolo
Digitaria macroblephara
Cenchrus ciliaris
Themeda triandra
Chrysopogon plectostachyus
Chloris roxburghiana
Eragrostis caespitosa
Panicum maximum
Enteropogon macrostachys
Sporobolus fimbriatus
Setaria verticillata
Sporobolus nervosa
Aristida adscensiomis
Eragrostis ciliaris
Cynodon dactylon
Tetrapogon cenchriformis
Pennisetum mezianum
Chrysopogon plumulosus
Sources: Hatch et al (1984) ; Shwartz (unpublished data) .
The mineral composition of wet season grasses around Isiolo
is given in Table 3 . This is compared to the data from high
potential areas of Mwakatundu (1977) given in Table 4. The mean
values for the range area are comparable to those reported for
the high potential areas. The data suggest that wet season
grasses in the rangelands of Isiolo contain adequate levels of
minerals to meet the demands of grazing animals. Consumption of
sufficient quantities may, however, be limited by herbage intake
since in the rangelands, CP levels fall to about 7% in only 8-10
weeks (Tessema, 1986) .
453
Table 3. Mineral composition (% DM) of some grass species in
the isiolo rangelands.
Mg
-% M-
Na
Pennisetum mezianum 0.17 0.19 0.10 0.62
Setaria vertilillata 0.27 0.41 0.10 2.62
Aristida adscensionis 0.14 0.15 0.11 0.51
Tetrapogon cenchriformis 0.24 0.22 0.14 1.22
Cynodon dactylon 0.44 1.10 0.11 1.86
Eragrostis ciliaris 0.39 0.28 0.10 1.44
Chrysopogon plumulosus 0.16 0.21 0.06 0.69
Sporobolus nervosas 0.40 0.28 0.08 1.21
Mean + S.D. 0.28+0.12 0.36+0.31 0.1+0.02 1.27+0.27
Table 4 . Average mineral composition in the DM of pastures
frcm selected areas of Kenya.
Season StationCa P Mg
0.32 0.24 0.17
0.59 0.28 0.24
0.53 0.35 1.18
0.55 0.27 0.17
0.37 0.45 0.26
0.39 0.26 0.10
0.38 0.16 0.18
0.53 0.20 0.18
Cu Co
ppm
7.0 0.142
11.0 0.210
14.0 0.061
9.0 0.036
7.0 0.142
6.0 0.098
8.0 0.086
11.0 0.027
Wet Baraton
Dry
Wet
Dry
Wet
Egerton
Kabete
Dry
Wet Molo
Dry
Source: Mwakatundu (1977)
MEETINO MINERAL REQUIREMENTS 07 GRAZING ANIMALS
An adequate supply of energy and protein are no doubt important
in the maintenance of high productivity in any animal production
system. In the rangelands of Kenya benefits would be even
greater if in addition, appropriate strategies were adopted to
meet the mineral requirements of the grazing animal. McDowell et
al. (1984) have observed that cattle grazing forages in severe P-
Co- or Cu-deficient areas are even more limited by lack of these
elements than by lack of either energy or protein.
In the past the practice has been to move cattle to natural
earth licks at certain seasons of the year (Hudson, 1944; Howard,
1963) , but as has been indicated earlier, the value of such licks
as sources of minerals is low since they are devoid of reasonable
quantities of either salt, Ca or P. Movement of animals to
unaffected areas has been shown to cure Co deficiency (Hudson,
1944) . This is unlikely to remain a solution in the future in
view of the present increase in population which will take more
land for food crop production. Use of mineral supplements on the
454
other hand increased milk yields, produced heavier calves and
maintained better animal condition than unsupplemented controls,
particularly in the dry season (Howard, 1963) .
It is evident from the above observations that there is need
for increased attention to be paid to mineral nutrition. Farmers
need to be informed that energy, protein and minerals are
interrelated in body maintenance, growth and health. They should
be made aware of the possible incidence of mineral deficiencies
because parts of Kenya are lacking in a number of the mineral
elements essential in animal nutrition. Periodic mineral surveys
which should include analyses of soil, water, plant and animal
tissue are necessary in order to detect inadequacies.
Inclusion of legumes in the pasture provides a mixture that
ensures adequate intake of Co by grazing animals because legumes
are efficient in accumulating Co in amounts above animal
requirements (Hodgson et al., 1962).
A mineral deficiency in pasture herbage is often a result of
a deficiency in soil. Thus, the potential role of fertilizers in
meeting animal requirements for minerals needs to be studied so
as to develop a fertilizer package that is location specific.
For example, Cu deficiency in animals may be treated by top
dressing pastures with copper sulphate (Hodgson et al., 1962).
It should be noted, however, that inadequate or erratic rainfall
may be dominant over soil effects in determining the mineral
concentrations of the herbage. Individual minerals may be
administered as a drench, injection or slow release pellets. The
use of "copper bullets" has been recommended for range livestock
(Hodgson et al., 1962), but this would require professional
supervision.
By far the most important method of meeting requirements
lies in mineral supplementation. Economic and physiological
responses to mineral supplements have been recorded (French,
1955; Howard, 1963) , and this should be encouraged through the
widespread use of free choice compound mineral licks. Several
ranches in Kenya are known to offer mineral licks to their
animals although there is no guarantee that all animals will
consume enough of the lick at all times. What is desirable,
however, is that such mineral mixtures should ensure that levels
of P, Na, Cu and Co are particularly adequate. Finally, it is
important to note that the measures recommended above are only
effective if the general level of nutrition with regard to energy
and protein is maintained particularly in the dry season.
REFERENCES
Abate, A. 1985. In vitro estimation of the energy content of
some feeds commonly fed to livestock in Kenya. East African
Agricultural and Forestry Journal 45:255-260.
Abate, A.N. and Abate, A. (in press). Cassava as a source of
energy in supplementary rations for weaner beef calves.
East Afric. Agric. For. J.
455
Abate, A., Kayongo-Male, H. and Karue, C.N. 1981. Dry matter,
protein, energy and fibre intake by dairy heifers grazing a
Rhodes (Chloris gayana) pasture. Animal Feed Science and
Technology 6:15-26.
Burdin, M.L. and Howard, D.A. 1963. Mineral survey of Kenya.
Kenya Veterinary Research Laboratories, Kabete, Kenya.
French, M.H. 1952. Mineral deficiencies in farm livestock. East
African Agricultural and Forestry Journal 10:4-8.
French, M.H. 1955. Earth eating and the mineral needs of
livestock. East African Agricultural and Forestry Journal
20:168-175.
Glover, J. and Dougall, H.W. 1961. Milk production from
pastures. Journal of Agricultural Science (Cambridge)
56:201-204.
Hatch, S.L., Morden, C.W. and Woie, B.M. 1984. The grasses of
the National Range Research Station, Kiboko (Kenya) . S.M.
Tracey Herbarium, Department of Range Science, Texas A & M
University, Texas.
Hodgson, J.F., Leach, R.M. and Allway, W.H. 1962. Micronutrients
in soils and plants in relation to animal nutrition.
Journal of Agriculture Food and Chemistry 10:171-174.
Howard, D.A. 1963. Notes on animal diseases. XXIII. Mineral
deficiency diseases. East African Agricultural and
Forestry Journal 28:191-194.
Howard, D.A. 1969. Notes on animal diseases. Mimeographed
report. Veterinary Research Laboratories, Kabete, Kenya.
Howard, D.A., Burdin, M.L. and Lampkin, G.H. 1962. Variation in
mineral and crude protein content of pastures at Muguga in
the Kenya Highlands. Journal of Agricultural Science
(Cambridge) 59:251-256.
Hudson, J.R. 1944. Notes on animal diseases. XXIII. Deficiency
diseases. East African Agricultural and Forestry Journal
10:51-55.
Karue, C.N. 1974. The nutritive value of herbage in semi-arid
lands of East Africa. 1. Chemical composition. East
African Agricultural and Forestry Journal 40:89-95.
McDowell, L.R. , Conrad, J.H. and Ellis, G.L. 1984. Mineral
deficiencies and imbalances and their diagnosis. In: F.C.M.
Gilchrist and R.I. Mackie (eds) , Herbi»ore nutrition in the
subtropics and tropics. The Science Press.
Mwakatundu, A.G.K. 1977. A study of the nutritive value of East
African pastures for ruminant livestock with special
reference to mineral nutrition in grazing dairy cattle.
Ph.D. thesis, University of Nairobi, Kenya.
456
Nyandat, N.N. and Ochieng, P.N. 1976. Copper content and
availability in soils: a survey of arable and range areas of
Kenya. East African Agricultural and Forestry Journal 42:
1-7.
Pinkerton, A., Barret, M.W. and Guthrie, E.J. 1965. A note on
copper deficiency in the Njoro area, Kenya. East African
Agricultural and Forestry Journal 30:257-258.
Tessema, S. 1986. Chemical and in vitro dry-matter digestibility
of herbage. In: R.M. Hansen, B.M. Woie and R.D. Child
(eds) , Proceedings of a conference on Range Development and
Research in Kenya held at Egerton College, Njoro, Kenya, 1-5
April 1988. Winrock International Institute, Morrilton,
Arkansas, USA. pp. 85-102.
Todd, J.R. 1954. The phosphorus status of cattle in parts of
Central Province, Kenya. East African Agricultural and
Forestry Journal 20:66-68.
457
EFFECT OF SEASON, FERTILIZER APPLICATION AND AGE OF REGROWTH
ON MINERAL CONTENT OF GUINEA GRASS (PANICUM MAXIMUM, 8CHUM)
AND GIANT STAR GRASS (CTOODON NLEMFUENSIS , CHEDDA)
E. Kabaija
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia
O.B. Smith
Department of Animal Science, University of Ife
Ile-Ife, Nigeria
Abstract
A study was conducted to evaluate the effect of season (wet/dry) ,
age of regrowth (2, 4, 6, 8, 10, 12 weeks) and fertilizer
application (with/without SPK) on the mineral content of Guinea
grass and giant star grass. Mineral concentrations in both
grasses were significantly affected by the main factors and the
age of regrowth effects were often modified by both season and
fertilizer effects. The concentrations of potassium and
phosphorus and that of copper (only during the wet season)
declined with age of regrowth. Fertilizer application resulted
in increase of potassium and copper concentrations. The levels
of potassium and copper in both grasses were lower during the dry
season while those of magnesium, zinc, iron and manganese were
higher. Despite these fluctuations, the forages contained
adequate amounts of all minerals for livestock requirements
except for phosphorus, sodium and copper which were deficient.
In production systems where the two grasses constitute the main
diet of grazing animals it would be necessary to offer
supplementary sodium, phosphorus and copper.
INTRODUCTION
All plants depend upon the soil for their supply of mineral
nutrients, and grazing ruminant animals obtain the majority of
their mineral nutrients from plants grown on these soils.
Concentrations of mineral elements in forage are dependent upon
the interaction of a number of factors, including soil, plant
species, stage of maturity, yield, pasture management and climate
(McDowell et al., 1983). Most naturally occurring mineral
deficiencies in herbivores are associated with specific regions
and are directly related to soil characteristics (Underwood,
1981; McDowell et al., 1983). When mineral nutrients in herbage
are marginal in respect of animal requirements, changes in
concentrations brought about by climatic, managerial or, seasonal
influences as well as plant maturity can be significant factors
in incidence or severity of deficiency states by livestock wholly
or largely dependent on these plants (Underwood, 1981) .
In West Africa, as in the rest of tropical Africa, forages
still serve largely as a source of essential elements for grazing
animals. Little coordinated research has been done in these
areas to identify places of endemic mineral deficiencies and
toxicities. Little information is available on the mineral
status of grazing animals and of the forages upon which they
458
subsist. Since the mineral status of these forages is a function
of multiple factors which interact with one another to produce
varied effects, it is vital to investigate how such factors
influence the mineral status of different forages in a tropical
environment. This study was thus planned to investigate the
effect of age of regrowth, season and fertilizer application on
the mineral profile of two grasses predominant in the humid zone
of West Africa, namely Guinea grass and giant star grass.
MATERIALS AMD METHODS
The experiment was conducted at the University of Ife teaching
and research farm situated in southwestern Nigeria at coordinates
7°28'N and 4°33'E at an altitude of 240 m above sea level. The
average annual rainfall is about 1290 mm and the climate is
subhumid. The rainy (wet) season stretches from about April to
October while the dry season is from November to March.
Four plots each 10 m x 10 m, two for Guinea grass (Panicum
maximum, var. S112, Nchisi, Schum.) and two for giant star grass
(Cynodon nlemfuensis var. nlemfuensis, IB8, Chedda) were cut back
at heights of 15 and 5 cm respectively during the early wet
season and early dry season. On the day of cutting back, one
plot for each grass was fertilized with NPK (15-15-15) at the
rate of 100 kg/ha. Samples of regrowth cut at a height of 15 and
5 cm for Guinea grass and giant star grass respectively were
taken at 2 , 4, 6, 8, 10 and 12 weeks after initial cutting back.
The samples were harvested from five randomly selected areas
(each 1 m in size) . The samples were taken to the laboratory,
chopped into pieces, mixed and dried at 85°C over night before
being ground in a laboratory stainless steel mill using a one-mm
sieve. The dried samples were ashed at 500°C for 8 hours and
extracted with 6 M HC1 as described by Buchanan-Smith et al .
(1964) . Potassium and sodium concentrations in the extracts were
analysed by flame photometer while calcium, magnesium, manganese,
iron, zinc and copper were determined by atomic absorption
spectroscopy. Phosphorus concentration was determined by
calorimetry (Fiske and Subbarow, 1925) .
Soil samples were randomly taken from the unfertilized plots
during both seasons and prepared for mineral analysis as
described by Black (19 65) . The mineral concentrations in the
extracts were determined using the same procedures described
above.
Data on forage mineral content was analysed, as a 2 x 2 x 6
factorial experiment with two fertilizer application levels, two
seasons and six ages of regrowth. Soil data were subjected to
one-way analysis of variance. Significant differences between
treatment means were tested with Tukey's procedure (Steel and
Torrie, 1960) .
RESULTS
The results on soil chemical composition indicated the soil pH
was acidic (Table 1) . Soil total nitrogen and exchangeable
potassium and magnesium concentrations were higher (P<0.01)
459
during the dry season than during the wet season. Available P,
Ca, Na and Zn were lower (P<0.01) during the dry season than
during the wet season.
Table 1. Chemical composition of soils from experimental site
at different periods.
Period of sampling
Early wet Late wet Late dry
season season season Mean SE
Soil pH 5o5 h
0.28°
5o4 v,
0.31°
5.9 5.6
0.3
0.18
0.04Soil nitrogen (%)
Available P (g/g)
Calcium ( g/g)
Manganese ( g/g)
55. 5a 15.3° 10.4° 27.1 17.68
87.08
17.4
1611.7* 1452° u 1372.3° 1478.7
241.3° 251.3° 287. 7a 260.1
Potassium ( g/g) 105. 4C 171.9° 212. 2a
8.5°
163.2 38.5
Sodium ( g/g) 11.4* 10. 6a 10.2 1.07
Manganese ( g/g) 575 550 525.
225°
550 14.7
Iron ( g/g) 275a 300 .
32.5°
266.7 27.28
Zinc ( g/g) 70a 12c 38.2 17.30
Copper ( g/g) 7.1 6.4 6.1 6.5 0.37
a,b,c in the same row, means bearing different superscripts
are different (P<0.01).
Mineral concentrations in both grasses were significantly
affected (P<0.0.1) by the main factors. In most cases, the
effects of the main factors were dependent on one another. Thus
two-way interaction effects were observed for nearly all the
minerals. The age of regrowth effects were often modified by
both season and fertilizer effects.
The K concentration in both grasses was higher (P<0.01)
during the wet season particularly during the early period of
regrowth. With both grasses there was a general decline in K
concentration after six weeks of regrowth and the concentration
of this element was lowest during the dry season though the
differences were narrow towards the end of the regrowth period
(Figure 1) . The concentration of Na in both grasses increased up
to four weeks and thereafter the changes were inconsistent. With
both Guinea grass and giant star grass, the concentration of Ha
was higher during the dry season. There was an increase in the
concentration of Ca during both seasons in Guinea grass while in
giant star grass the changes were irregular and no overall
increase was observed over the whole period. Except for K where
lower concentrations of the element were observed in unfertilized
plots, fertilizer effect on Na and Ca concentrations were
inconsistent.
The concentration of P in Guinea grass during both seasons
and in giant star grass during the wet season declined with age
of regrowth. No remarkable differences were observed due to
either season or fertilizer application (Figure 2) . Overall,
460
Figure 1. Potassium, calcium and sodium concentrations of P.
maximum and C. nlemfuensis over 12 weeks of regrowth as
influenced by season and fertilizer application.
Wet itotoo
fertilized
Wit teutoo
not fertili zed
Dry aeoaon
fertilized
Dry uoioo
oot fertilized
C . otemfueoeit
 
»gt of regrooto (aeeke)
there was an increase with age of regrowth in the concentration
of Mg in Guinea grass during both seasons and in giant star grass
during the dry season. The concentration of Mg in both grasses
was higher during the dry season. Fertilizer application was
associated with higher Mg concentration during the dry seasons in
giant star grass while the reverse was observed in the case of
Guinea grass.
The concentration of Fe in Guinea grass was fairly constant
during the wet season but rose sharply with age of regrowth
461
during the dry season. The changes in iron concentration with
age of regrowth in giant star grass were inconsistent. Fe
concentration was greatly increased by fertilizer application
during the dry season. Except for giant star grass during the
dry season when there was a decline in concentration with age of
regrowth, the Mn concentration in both grasses exhibited slight
Figure 2 . Magnesium, phosphorus and iron concentrations of P.
maximum and C. nlemfuensis over 12 weeks of regrowth as
influenced by season and fertilizer application.
Wtt IIQIoo
ftrtlliztd
Wtt liiiiao
not ftrtiliztd
Dry itaton
oot ftrtiliztd
MO
450-
350
250
lSO
| 50
J. 50-i
"
P. mailmum
Mn . sv:
rrrrr^
\
i
 
 
2 4 6 8 C
Agt of rtgrowth (wttkt)
4 • • i0 i2
Age of rtgrowth (wttkl)
462
variations over the 12-week period. In both grasses the
concentration of Mn was higher during the dry season. No
remarkable differences in concentration of Mn were associated
with fertilizer effect (Figure 3) . The changes in Zn
concentration in Guinea grass during the wet season over the 12-
week period were slight while there was a significant increase
during the dry season. The concentration of Zn was high in
Guinea grass during the dry season and only in unfertilized giant
star grass during the same period. There was decline in Cu
concentration during the wet season in both grasses with age of
regrowth while little variation occurred with age of regrowth
during the dry season. Fertilizer application on both grasses
during the wet season resulted in higher Cu content.
DISCUSSION
Potassium, phosphorus and copper concentrations declined with age
of regrowth. With temperate grasses, Whitehead (1966) observed
that P content and that of most trace elements in herbage
declined with age. Potassium and P declining with age has been
reported in tropical forages (Perdomo et al . , 1977; Gomide,
1978) . This decline may be due to the effect of dilution of
these elements in a great quantity of dry matter that is produced
and accumulated with advancing age. Gomide (1978) associated the
low concentration of P and K in young tissues to their being
mobile and thus easily translocated from the oldest tissues to
the young ones .
In this study Na and Ca concentrations in both grasses did
not change appreciably with increase in age of regrowth at most
times. Perdomo et al. (1977) observed that Ca concentration in
Guinea grass and Bermuda grass did not change with increasing
maturity. They however observed that Na concentration in Guinea
grass increased with age. Fleming (1973) observed that Ca levels
in forages remained relatively constant with advancing maturity.
The concentration of Mg exhibited slight variations with age of
regrowth during the wet season but increased during the dry
season. Fleming and Murphy (1968) reported that Mg content in
temperate grasses exhibited relatively little variation with
advance in maturity. On the contrary, Jones (1963) in Zimbabwe
noted that Mg levels in Guinea grass decreased with advancing
maturity.
Except for Cu, whose concentration declined with age of
regrowth during the wet season in both grasses, the changes
exhibited by the other trace elements were inconsistent. The
decline in Cu concentration with age of regrowth has been earlier
reported in temperate grasses (Thomas et al., 1952) and tropical
grasses (Gomide et al., 1969). The inconsistencies in changes in
trace element concentration observed in this study agreed with
the statement by Conrad (1978) that trace minerals in plants may
increase, decrease or show no consistent change with stage of
growth, plant species, soil or seasonal conditions.
In this study only K and Cu were remarkably affected by NPK
fertilizer application. When NPK fertilizer was applied on
Guinea grass and giant star grass in central Brazil (Gomide et
al., 1969) no effect was observed on any of the minerals studied
463
Figure 3. Manganese, line and copper concentrations of P. maximum
and C. nlemfuensis over 12 weeks of regrowth as
influenced by season and fertilizer application.
wtr seoe0o Wit ttaton
fertilized oot fertilized
0.2 5o
0.20o P. maximum
Dry leaioo
not fertilized
 
2 4 6 6 i0 i2
Age of regrowth (weeke)
-i i00
2 4 6 8 i0
Age of regrowth ( weeks)
464
except Mn, whose concentration increased. Hemingway (1962) found
that application of nitrogen fertilizer to grasses increased
their Cu content while neither superphosphate nor muriate of
potash influenced the Cu content of the herbage. Fleming and
Murphy (1968) reported that application of K fertilizers resulted
in increased Mg concentration in the herbage while the level of K
subsequently increased. In this study, the failure of NPK
fertilizer to influence Mg concentration considerably could be
due to the combined effect of K and P, as Andrew and Robins
(1969) reported that P fertilizer application resulted in
increased Mg concentration while K fertilizer caused a reversed
effect.
According to NRC (1978, 1984, 1985) mineral requirement
recommendations for cattle and sheep, the two forages were
adequate in K, Ca, Mg, Mn, Fe and Zn. The latter element was
marginal in Guinea grass during the wet season. The grasses were
deficient in P, Na and Cu during both seasons. The Cu deficiency
was more severe during the dry season. Low levels of Na, P and
Cu in tropical forages have earlier been reported (Oyenuga and
Hill, 1966; Ademosun and Baumgardt, 1967) . Concentrations of Mn
as high as 430 mg/kg that were observed in Guinea grass could
probably be detrimental to animal health through interference
with metabolism of other minerals (Thomas, 1970) . Whether such
high levels are of any economic significance in Africa is an
aspect yet to be investigated.
CONCLUSION
A study of the effect of season, age of regrowth and NPK
fertilizer application on the mineral profile of Guinea grass and
giant star grass indicated significant effects of the main
factors, particularly of season and age of regrowth on almost all
the elements. The two grasses were adequate in all minerals
measured for livestock feed, except for Na, P and Cu. In case
the two grasses are exclusively fed to livestock, it is
recommendable to supplement with these deficient minerals. Other
measures that could be taken to improve the mineral status of
these forages include fertilizer application in case of P, Zn and
Cu as well as foliar dusting in case of Zn and Cu as long as
precautions are taken not to deposit excess amounts of these
minerals on the vegetation. Effectiveness of such corrective
measures under African field conditions need to be investigated.
In both grasses, the Mn and Fe concentrations occurred in
quantities far in excess of livestock requirements. It is
recommended that further research be done to elucidate any
deleterious effects to livestock that might be associated with
such high mineral concentrations.
REFERENCES
Ademosun, A. A. and Baumgardt, B.R. 1967. Studies on the
assessment of the nutritive value of some Nigerian forages
by analytical methods. Nigerian Agricultural Journal 4:1-7.
465
Andrew, C.S. and Robins, M.E. 1969. The effect of phosphorus on
the growth and chemical composition of some tropical pasture
legumes. II. Nitrogen, calcium, magnesium, potassium and
sodium contents. Australian Journal of Agricultural
Research 20:675-685.
Black, C.A. 1965. Methods of soil analysis. Agronomy No. 9(2).
American Society of Agronomy, Madison, Wisconsin.
Buchanan-Smith, J.G., Evans, E. and Poluch, S.O. 1964. Mineral
analyses of corn silage in Ontario. Canadian Journal of
Animal Science 54:253-256.
Conrad, J.H. 1978. Soil, plant and animal tissues as predictors
of the mineral status of ruminants. In: J.H. Conrad and
L.R. McDowell (eds) , Latin American symposium on Mineral
Nutrition Research with Grazing Ruminants. University of
Florida, Gainesville, Florida. pp. 143-148.
Fiske, C.A. and Subbarow, Y. 1925. The colorimetric
determination of phosphorus. Journal of Biological
Chemistry 66:375-400.
Fleming, G.A. 1973. Mineral composition of herbage. In: G.W.
Butler and R.W. Bailey (eds) , Chemistry and biochemistry of
herbage. I. Academic Press, New York. pp. 529-566.
Fleming, G.A. and Murphy, W.E. 1968. The uptake of some major
and trace elements by grasses as affected by season and
stage of maturity. Journal of British Grassland Society
23:174-185.
Gomide, J. A. 1978. Mineral composition of grasses and tropical
leguminous forages. In: Latin American Symposium on Mineral
Nutrition Research with Grazing Ruminants. University of
Florida, Gainesville, Florida. pp. 32-40.
Gomide, J. A., Noller, C.H. , Mott, G.O., Conrad, J.H. and Hill,
D.L. 1969. Mineral composition of six tropical grasses as
influenced by plant age and nitrogen fertilization.
Agronomy Journal 61:120-123.
Hemingway, R.G. 1962. Copper, molybdenum, manganese and iron
contents of herbage as influenced by fertilizer treatments
over a 3-year period. Journal of British Grassland Society
17:182-187.
Jones, D.I.H. 1963. The mineral content of six grasses from a
Hyparrhenia-dominant grassland in Northern Rhodesia.
Rhodesia Journal of Agricultural Research 1:33-38.
McDowell, L.R., Conrad, J.H., Ellis, G.L. and Loosli, J.K. 1983.
Minerals for grazing ruminants in tropical regions.
University of Florida, Gainesville, Florida.
NRC (National Research Council). 1978. Nutrient requirements of
domestic animals. Nutrient requirements of dairy cattle.
Fifth edition. National Academy of Sciences, Washington,
D.C.
466
NRC. 1984. Nutrient requirements of domestic animals. No. 5.
Nutrient requirements of beef cattle. Fifth edition.
National Academy of Sciences, Washington, D.C.
NRC. 1985. Nutrient requirements of domestic animals. Nutrient
requirements of sheep. Sixth edition. National Academy of
Sciences, Washington, D.C.
Oyenuga, V.A. and Hill, D.H. 1966. Influence of fertilizer
application on the yield, efficiency and ash constituents of
meadow hay. Nigerian Agricultural Journal 3: 6-14.
Perdomo, J.T., Shirley, R.L. and Chicco, C.E. 1977. Availability
of nutrient minerals in four tropical forages fed freshly
chopped to sheep. Journal of Animal Science 45:1114-1119.
Steel, R.G.D. and Torrie, J.H. 1960. Principles and procedures
of statistics. McGraw Hill Book Co., New York.
Thomas, J.W. 1970. Metabolism of iron and manganese. Journal of
Dairy Science 53:1107-1123.
Thomas, B. , Thompson, A., Oyenuga, V.A. and Armstrong, R.H. 19 52.
The ash constituents of some herbage plants at different
stages of maturity. Experimental Agriculture 20:10-22.
Underwood, E.J. 1981. The mineral nutrition of livestock. 2nd
edition. Commonwealth Agricultural Bureau, London.
Whitehead, D.C. 1966. Nutrient minerals in grassland herbage.
Pub. 1/1966. Commonwealth Bureau of Pastures and Field
Crops, Commonwealth Agricultural Bureau, London.
467
RESEARCH NEEDS FOR FORAGE DEVELOPMENT IN ETHIOPIA:
THE FOURTH LIVESTOCK DEVELOPMENT PROJECT STRATEGIES
Alemayehu Mengistu and Alan Robertson
Fourth Livestock Development Project
Ministry of Agriculture, Addis Ababa, Ethiopia
Abstract
The Fourth Livestock Development Project will be promoting a
range of strategies for forage development, namely: establishment
of forage strips including bund-planted forages and tree legumes
in alley cropping systems; backyard forage development;
improvement of stock exclusion areas through the introduction of
legumes; the use of forage legumes in undersowing/relay cropping
systems; oversowing of grazing areas; and establishment of
perennial mixed pastures and annual fodder crops. This paper
details the strategies, ranks them in terms of the potential
impact within the project area and outlines the perceived
priority research programmes required to support these
strategies.
The approach to establishing a large forage seed production
capacity and the research needs to support this programme are
discussed.
The various strategies described have different labour
requirements, and the monitoring of labour inputs would be a
significant part of the applied research programme.
INTRODUCTION
The Fourth Livestock Development Project (FLDP) is a five-year
phased programme designed to build an efficient national animal
health service and simultaneously improve animal nutrition in the
peasant sector. The primary objective is to improve livestock
and agricultural production through increasing the efficiency of
resource utilisation at the farm level in both the government and
peasant sector cooperative support services. Increasing foreign
exchange earnings and decreasing potential ecological damage
through increasing live animal and hide and skins exports and
through import substitution (dairy produce) is also an important
objective. Institution-building within the Ministry of
Agriculture represents a major project objective.
The project consists of six main components:
1. Animal health development,
2. Animal feed and nutrition development,
3. Southern rangeland pilot development,
4. Agricultural credit,
5. Institutional development,
6. Live animal trade development study.
The major objectives of the animal feed and nutrition
development component of the Fourth Livestock Development Project
(FLDP) are:
468
a) To increase the supply of forage for ruminant livestock
in the peasant sector,
b) To conserve soil on arable land and catchment areas,
c) By increasing the quality and quantity of feed;
increase livestock meat and milk production,
d) Increase manure production,
e) Increase draft power for cropping,
f) Increase fuel wood and timber supplies.
This paper details with the various Fourth Livestock
Development Project strategies and research needs for forage
development which will be directed at improving forage
production, management and utilisation at farm level.
FACTORS IN DETERMINING THE STRATEGIES
Long-term sustainable production of livestock and cropping is
dependent on dramatic changes in livestock management systems.
The key components of these changes are a shift towards more
intensive feeding systems, with more emphasis on cut-and-carry
feeding, and a gradual shift away from uncontrolled grazing,
particularly on sloping areas.
The strategies would be integrated closely with cropping
practices, and in almost all cases, would avoid displacement of
arable crops; in fact, the strategies would generally be
complementary to arable cropping. There would be emphasis on the
high potential areas, partly because those areas have the
greatest potential for increased forage production under the
kinds of strategies promoted, and also because most extension
staff are being deployed in the high potential cropping areas.
Strong links will be maintained with soil conservation
programmes: physical soil conservation works can be
substantially improved through the incorporation of improved
forages; also, some forage development strategies can have a
direct benefit in terms of erosion control, even in the absence
of physical soil works. However, strategies would not be
dependent on "food for work programmes" as this is not feasible
in a programme of the size envisaged, and it would also reduce
the likelihood of widespread rapid and spontaneous adoption of
some strategies.
There should be strong links with animal health programmes
because of the mutual synergism not only in physical production
terms, but also in the extension context. The interface between
health and nutrition programmes would occur at the service
cooperative, which is the major unit for extension.
There would also be close links with the beef fattening and
dairy programmes; the widespread success of both these programmes
will be largely dependent on the incorporation of sound forage
development, and there is also a reverse benefit in that the
Project's forage development proposals would be more readily
adopted within the peasant sector if they are linked with such
programmes, thus offering the opportunity for conspicuous cash
benefits.
469
Legumes would be heavily emphasised because of their dual
roles in both animal nutrition and in the maintenance or
improvement of soil fertility and hence crop production.
It is imperative that most strategies have a low financial
cost if they are to be adopted within the peasant sector; in this
context the heavy use of fertilizers would generally be avoided,
and beef fattening and dairying programmes should not be
dependent on the construction of expensive animal housing or on
conventional post-and-wire fencing.
Some strategies at least should have sufficiently
conspicuous benefits to facilitate spontaneous and widespread
adoption. For instance, the use of tree legumes in some
strategies could fit into this category.
It is feasible to initiate a development programme
immediately with existing technologies; however, there would be a
need for continued adjustment in terms of genetic material and of
strategies in the light of newly gained experience from both
within the Project and other related institutions. It is
imperative that the Project maintain a maximum of flexibility to
accommodate these improvements.
The success of various Project interventions will be
markedly dependent on the agroecological zone and on other
factors including cropping patterns and prior experience with
more intensive feeding systems. Whilst the Project would
initiate some activites in some of the more difficult areas or
systems, it is important that emphasis initially be placed on
those areas or production systems affording the best chances of
success. Typically these areas would be in the low to medium
altitude areas i.e. below 2400 m.
THE STRATEGIES
The development strategies to be promoted by the project are:
a) Strip establishment of forages,
b) Backyard forages,
c) Improved forages in stock exclusion areas,
d) Legumes undersown in crops,
e) Legumes oversown on grazing areas,
f) Perennial mixed grass/legume pastures,
g) Annual fodder crops.
These strategies would generally be associated with an
extension programme to improve the utilisation of the forage.
Improved grazing management would be extremely difficult to
incorporate into common practice and would generally require
substantial reduction in stock numbers to be effective. It will
therefore play only a minor role in the development programme.
A significant reduction in livestock numbers could
dramatically affect livestock production and land degradation;
its potential economic impact is therefore large. However,
470
within current production systems, such a reduction
will not occur in the foreseeable future.
in numbers
Table 1 illustrates the perceived significance of the
various strategies. The potential economic impact, with a score
of five representing the maximum impact, is assessed on a basis
of increased forage production and benefits in terms of fuelwood
supply, erosion control and contributions to the maintenance of
soil fertility. The technical possibilities are assessed for low
and high altitudes (say, below and above 2 2 00 m) , with an
indication of acceptance of the strategy amongst farmers or
cooperatives. Major research needs in support of the strategies
are listed. These are discussed in more detail later.
Table 1. FLDP forage development strategies.
Strategy Potential Technical Farmer Research Comments
economic possibilities accep needs
impact Low High tance
Forage
alt alt
strips
(bunds,
alleys, 5 Good Good Variable Species, Stock
etc.) Mgmt,
labour
control,
erosion
Backyard
control
forage 3 Good Good Good Labour Supplemen
tation
Undersown 2-4 Good 7 Variable Species Maize/
legumes timing sorghum
Oversown 0-1 Fair Low Good Species To 2000 m
legumes no control
Exclusion
area 4 Good - Good Species Cut 4 carry
improvement fattening
link
Conventional <1 Good Good Low - Dairy only
pastures
Fodder crops <1 Good Good Fair — Dairy,
fattening
Grazing 2? Good Good V. low - -
management
Reduced
numbers 5?? - - V. low - -
(current
systems)
471
Strip Establishment of Forages
The strip establishment of forages includes:
a) Forages planted on physical soil structures including
bunds and terrace walls within arable and stock
exclusion areas, but with primary emphasis on the
arable areas.
b) Forages planted on contour strips within arable areas,
without physical soil structures.
c) Tree or shrub legumes established in parallel strips
within crop areas (i.e. in "alley cropping" systems).
d) Forages established in strips around crop plots as
shelter belts.
The forage strips would have multiple roles, including the
provision of forage for cut-and-carry management, fuelwood
supply, soil erosion control and stabilisation of associated crop
yields through contributions of organic matter and nitrogen, and
through the shelter belt effects.
Tree legumes would form the basis of the forage strips. At
higher altitudes grasses (particularly Phalaris spp.) would
initially predominate in the bund plantings until more reliable
tree legumes are available. However, where grasses are used in
arable areas, it is important that they are not creeping types to
avoid invasion of adjacent crop areas; herbaceous legumes would
be incorporated with the grasses where no suitable tree legume is
available.
Close spacing of the tree legumes is important for erosion
control, but is feasible only with the use of either cuttings or
direct seeding; seedlings are too expensive to be used at higher
densities. Even at the spacing of 0.5-1.0 metre between plants,
costs are substantial, and techniques for improved direct seeding
establishment demand immediate investigation.
Management, and particularly the close control of stock
access, is crucial to the success of the strategy. Wherever
possible, the strips should be reserved for cut-and-carry
utilisation; this would greatly enhance persistence and
productivity, and the programme should, therefore, be restricted
initially to areas where such control is assured. Excess
leguminous forage from the strips would be used as a mulch on
adjacent crop areas.
The strategy will play an extremely important role in the
longer term, but successful demonstration is central to its
widespread adoption; initial sites should therefore be selected
with good prospects of success.
Farmers have demonstrated fair interest in the strategy,
primarily on the basis of increased forage supply; reservations
included possible competition effects with adjacent crops and
increased difficulty of subsequent cultivation. The latter
concern is valid, and the importance of establishing parallel
472
rows wherever possible should be stressed. The importance of
establishing a significant visual impact in the first season was
clearly demonstrated. Producer cooperatives were shown to be
generally much more receptive to the strategy and, where
possible, should be used to provide the initial focus for such
development.
Backyard Forages
Backyard forage development involves the establishment of plots
or hedges of forage in the immediate vicinity of the house.
Typically, plots would comprise perennial grass with herbaceous
legumes included where possible, or with the grasses interplanted
with tree legumes; hedges around the house plots would be of tree
legumes or occasionally of tall growing grasses. The inherently
higher soil fertility and good control of livestock adjacent to
the house enables the establishment of highly productive species
with close to maximal yields.
The backyard areas would be used solely for the production
of high quality forage for the supplementation of lower quality
forage for the supplementation of lower quality feedstuffs
including crop residues, particularly in backyard fattening or
dairying operations; hence, even small areas of 50-100 m can be
significant.
Backyard areas also provide a very convenient point of entry
for new species or species types prior to their subsequent use on
a wider scale e.g. within the various forage strip programmes.
Under the close control offered, farmers can gain an appreciation
of both the production potential and of the management
requirements of the various species.
Backyard forage is the most readily accepted of all
strategies and could be very cost-effective in that it requires
relatively little material or extension input. The strategy has
already proved to be acceptable and logistically simple in all
project areas.
Improved Forages in Stock Exclusion Areas
This strategy involves the establishment of a range of forage
species within stock exclusion areas to improve both the
productivity of the areas and the feeding value of the forages on
offer. Such improvements would greatly enhance the acceptance of
the exclusion strategy locally. Major emphasis would be on the
herbaceous legumes, because of their contribution to forage
quality and to total production, in some cases even under low
soil fertility, and because of their relative ease of
establishment even on unprepared seedbeds. The grasses typically
require much better seedbed preparation for successful
establishment, and their use would be largely restricted to sites
with recent soil disturbance (e.g. where soil bunds have been
constructed within the exclusion areas) . Also, the productivity
of most exotic grasses would initially be very low in such
severely degraded environments; substantial fertility accretion
would be expected after several years of active legume growth,
whereupon improved grasses could be gradually introduced. The
tree legumes would generally have only a minor role within the
473
Legumes Oversown in Grazing Areas
Grazing areas could be improved if it were possible to introduce
more productive or better quality species. Such species would
need to be adapted to prevailing grazing pressures and soil
fertility and to establishment without any land preparation.
Grasses are generally unsuitable because the establishment rates
of the oversown grass seed on compacted grazing areas are
generally extremely low; legumes are generally more suited to
such establishment.
The higher altitude areas are typically served by a wide
range of useful native legumes, with the Trifolium and Medicago
spp. dominating; other useful species may be identified for this
niche, but it is not a priority programme.
At lower altitudes there is a relative dearth of productive
herbaceous legumes in the grazing areas, with most of the
trailing types having disappeared under the heavy grazing
pressures. Also, there are some exotic species which are capable
of filling this niche. The genus with the best prospects is
Stylosanthes , which is typified by ready establishment on poor
seedbeds and very low fertility soils, and in many cases by
tolerance of extreme grazing pressure. S. hamata and S. scabra
are the most promising species at this stage. Some less
palatable climbing species including Calopogonium mucunoides will
also be usefully oversown in some higher rainfall areas of the
western slopes.
Because of the low productive potential, the strategy must
be based on very low inputs; in particular, seeding rates should
be extremely low.
A modification of the oversowing strategy is the sowing of
seed on road verges. This can be done very quickly, is simply
monitored, providing a long transect for evaluation and will
often enable maximal spread of seed to surrounding areas.
Because of the relatively low establishment rates from
oversowing, even with legumes, the programme would be undertaken
only on a pilot scale in the first couple of years. Although in
suitable areas it can be highly cost-effective.
Oversowing requires minimal inputs of local labour and no
management inputs; therefore, there will be no difficulties in
promoting the strategy. During the first two seasons the limited
areas sown under the programme would be closely monitored to more
closely define the site parameters indicating suitability for
oversowing.
Establishment of Mixed Grass/Legume Pastures
This strategy involves the establishment of perennial mixed
pastures on prepared seed beds and with management including
grazing control and fertilizer inputs as required to maintain
high productivity. Good arable soils are required for such
productivity, and the strategy typically represents a diversion
of land from arable cropping. Hence, there is a considerable
opportunity cost. Other costs include fertilizer, seed costs
476
Initial emphasis would be on the low to medium altitude
areas for reasons similar to those given for the stock exclusion
strategy i.e. a relative dearth of legumes in these areas and a
range of suitable exotic legumes available. Most work will be
undertaken initially on the pasture/ forage type legumes.
Desirable characteristics for undersown legumes include deep
rooting/drought tolerance to enable the legume to continue growth
after the annual crop has been harvested; retention of leaf and
hence of feeding value during the dry periods, to maintain a
higher feeding value of the stubble; ease of establishment and
heavy early seeding, to allow adequate regeneration in subsequent
cropping cycles; and a proportion of hard seed, so that in
subsequent crop cycles there is adequate seed remaining in the
soil for establishment after the final weeding of the annual
crop.
The maize and sorghum systems provide the best opportunities
for the strategy, and development can be initiated in such
systems without further evaluation. Promising species include
the free seeding Stylosanthes species including S. hamata and the
Graham cultivar of S. guianensis ; the climbing legumes include
Macroptilium atropurpureum, Macrotyloma spp, and Desmodium spp. ,
and the dual purpose types such as Lablab purpureas.
An opportunity also exists in some of the higher altitude
areas for the introduction of legumes with a longer growing
season than the native species. Trlfolium repens and Trifolium
subterraneum are likely prospects. The best opportunity appears
to be in the fcelg-grown barley system.
The initial sowing would take place after the final weeding,
by which stage it is unlikely that the growth of most legumes
would offer substantial competition to the companion crop.
Sowing rates would be typically quite low, unless the seed is to
be harvested for human food or as a cash crop. The strategy does
present opportunities for seed collection for some species
including Lablab.
Potentially useful food grain species which will be
incorporated in the programme at least for further evaluation
include Phaseolus acutifolius, Lablab, and cowpeas. The use of
the biennial or short-lived perennial Cajanus cajan (pigeon pea) ,
which could be interplanted with a range of crops in an alley
cropping type system, offers similar benefits, and the Project is
expected to participate vigorously in the wider promotion of this
species.
Undersowing was attempted on a range of sites during the
1986 season. It is evident now that the strategy will be quite
well accepted by farmers. The strategy is relevant to both the
well managed row planted crops characteristic of Harerge, and
also to the very poorly managed, weedy crops e.g. in parts of the
Rift Valley. In the former case, attempts would be made to link
undersowing to contract seed production where possible, at least
initially; in the latter case emphasis would be on hardier, self
regenerating species.
475
(seed inputs would generally be higher than with most other
strategies) , stock control costs whether for fencing or
additional herding labour and weed control. Also, successful
management of mixed pastures for high productivity of grazing
stock requires considerable skills. The strategy would generally
be confined to the dairy programme, but in some cases could
encompass the rehabilitation of crop areas through the use of a
moderately long-term pasture phase. There are no serious
technical constraints in any altitude zone.
The strategy presents an opportunity for occasional seed
production, particularly for the grasses, because areas
established would generally be of sufficient size to warrant the
necessary close supervision of harvest.
Establishment of Short-Term Fodder Crops
The establishment of fodder crops such as alfalfa, oats, vetch
and fodder beet will be technically feasible in most altitude
zones. The strategy provides a convenient, simple introduction
to the concept of actually growing forage for livestock feeding.
Yields may be high in all altitude zones. However, the
strategy would not be heavily emphasised in the project for
several reasons: The project aims at integrating forage
development with erosion control, but excessive use of short-term
forage crops may actually aggravate soil erosion through repeated
cultivations. Farmers cultivating an area of land would
typically establish a subsistence or cash crop and would be
reluctant to regularly invest such labour in short-term fodder
cropping. The requirement for repeated inputs of seed is a
further disincentive. The annual fodder crops may meet short-term
needs, but often the perennial species (including some grasses
and tree legumes) provide forage over a longer period.
The establishment of short-term forage crops would be
confined mostly to the dairy programme. There will be some use
of the strategy also in backyard areas, especially with very
productive species including alfalfa and fodder beet. Oats could
become more widely grown particularly because of its broad
adaptation and its dual-purpose nature; it would generally be
grown in association with vetch.
SEED PRODUCTION
General Comments
The seed production programme outlined for the Project is large
scale by any standards; by Year 4, the Project will need to
produce more than 150 tonnes of herbaceous legume seed annually.
Also, the programme encompasses a much wider range of species
than is usual within a single project. The forage development
programme is entirely dependent on the seed production
activities. The strategies for production are crucial, and,
clearly, strategies previously employed are entirely inadequate
for a programme of this scale. The only feasible approach is to
produce seed in contract systems.
477
A programme has already been initiated to collect seed at
contract prices from some areas previously established to various
forages, including Leucaena, Sesbania, Lablab, and vetch. Some
indication of the suitability of the various contract prices will
be afforded by this initial programme. To obviate the need for
importation of large quantities of seed, some irrigated seed
production areas have been established in collaboration with the
Ministry of State Farms, Institute of Agricultural Research, ILCA
and various service cooperatives; however, most production would
be under rainfed conditions to facilitate cheaper seed.
The Contract Seed Production Programme
The great majority of seed must be produced under contract
arrangements (with service cooperatives, producer cooperatives
and individual farmers) , if the overall programme is to succeed.
Contract systems, wherein producers are paid contract rates per
quantity of clean seed produced, will result in much greater
volumes of seed at significantly lower cost per kg, higher
production per unit area, much greater efficiency of production,
and after the first couple of years, quite reasonable
predictability of total seed output.
The contract prices will be set annually though for many
species there will be minimal or no change from one season to the
next; these prices must be standardised through the Project area
and must be valid for at least one production season. The prices
will reflect the seed yield potential, labour requirements and
potential returns from alternative crops on the same areas. The
prices should encourage close attention to management and high
recovery of available seed. If the scheme is to succeed within
the given time frame, the prices will offer substantially higher
returns than for most competing crops. However, because of the
relatively small area involved to meet the Project's demands,
there would be no disruption of crop production. There are
essentially three broad systems which might be used within the
contract programme.
For Some Grasses
The Fourth Livestock Development Project (FLDP) would assist
dairy cooperatives to establish significant units of grass-based
pasture with the understanding that a portion of the area would
be closed for a period for seed production, with the cooperative
providing harvesting labour under supervision and receiving
contract prices. The system is convenient because it offers the
chance for larger areas to be harvested, thus facilitating the
supervision which is necessary for the harvesting of most of the
grasses and overcoming the organisational problems associated
with the typically lower seed yields of the grasses.
Seed Collection from Stock Exclusion Areas
Some species will seed quite well on relatively poor sites even
though their dry-matter production may be quite low. Therefore,
the stock exclusion areas provide a good opportunity for
collecting substantial quantities of seed from herbaceous and
tree legumes. Most labour would be provided by women and
children.
478
Seed Production on Specialised Plots
Species with high yield potential, and those with high labour and
management requirements, may often be best grown on specialised
seed production plots. Some of these would be small plots
operated by individual farmers. Species suited to this general
category include Stylosanthes spp. , Lablab and other climbing
legumes grown in trellis systems.
FLDP RESEARCH NEEDS
General
Pasture/ forage research programmes require substantial
modification to adequately meet the demands of FLDP and similar
projects. In particular, there could be less emphasis on
evaluating species for conventional pasture development, and more
emphasis on species and establishment requirements for specific
strategies, and for some of the more difficult but very important
systems including stock exclusion. The continued analysis of
nutritive value of introduced commercial cultivars, and of the
material already comprehensively analysed elsewhere, represents a
serious misallocation of research resources.
Species Screening
The Project is servicing a very wide range of environments, and
information on species performance from just a few sites may be
quite misleading.
There is still inadequate information on the agroecological
range of the species which would be used initially within the
project. In particular, knowledge on the altitude limits is
imprecise. The Project will adopt the approach of using the most
promising species/cultivars in the development programmes, with
continuous monitoring and refinement; additionally, new material
will be continually introduced for rapid screening. Evaluation
will be introduced for rapid screening. Evaluation will be
undertaken on sites representative of target areas. For example,
those species expected to be used in the improvement of stock
exclusion areas would typically be assessed on degraded soils of
low phosphate status.
Persistence of species must become one of the major criteria
in selection. Given the scale of the Project, seed production
potential should be granted more emphasis.
There is a particular need to define a wider range of tree
legumes suited to intensive cut and-carry-management, in
particular for use in higher altitude areas where there is a
relative dearth of such species.
Site Assessment
It is crucial that field extension staff be in a position to
categorise sites according to suitability for various species and
strategies. During the first two years of implementation, some
150-200 service cooperatives will be incorporated, covering a
479
very wide agroecological range. By the end of that period, rapid
assessment formulae should have been developed, utilising the
key determinants of production including altitude, rainfall and
soil factors such as drainage, depth, pH and phosphate status.
Fertilizers will definitely not be used extensively for
forage production within the peasant sector in the foreseeable
future, and currently there is excessive research on the use of
fertilizers for forage production. It is more valid to
concentrate on defining site criteria which will determine the
success strategic of species. Some research is justified on the
use of phosphate (and other nutrients) in seed pelleting as an
aid to the Project distributed pelleted seed.
Inoculation
Inoculation requirements for the key species demand urgent
attention, and the most convenient media for local use should be
assessed. Substantial quantities of inoculant will be required
for at least 5-10 species, and it must be determined whether a
local production capacity is justified.
Seed Production
It is crucial to collect information on production, harvesting
and post-harvest handling for commercial quantities of seed.
This will be done by close monitoring of the Project's seed
production activities rather than in specific research
programmes. It is particularly important to determine likely
yields and labour inputs to define optimal prices for the
contract seed production programmes.
Propagation of Tree Legumes
The development programmes are heavily dependent on the use of
tree legumes, particularly for the more intensive cut-and-carry
systems. Current propagation techniques within nurseries are
excessively expensive, and there is an urgent need to define
systems appropriate to the use of direct seeding cuttings and
bare root/bare stem seedlings, enabling the use of small backyard
nursery plots.
Economic Aspects
There is an urgent need to undertake economic evaluation of the
use of fertilizers within various production systems. This can
be done immediately with the wealth of nutrient response data
already available. After the first couple of years of the
Project, there will be sufficient data for a comprehensive
economic analysis of the various strategies. This is
particularly important in the case of those strategies with high
labour inputs, such as backyard forage and alley cropping.
Also, it is crucial to assess the economics of backyard
fattening and dairying using various feed inputs. It is expected
that labour requirements for forage collection and forage yields
will be central to the economic viability of the strategies; FLDP
will allocate adequate resources for determining these parameters
during the first two years.
480
Studies should be undertaken immediately on the feasibility
of inducing a shift away from grazing and towards more intensive
cut-and-carry systems involving total stock control in the
context of peasant associations, producer cooperatives, service
cooperatives and larger administrative units.
REFERENCES
ARDP. 1970-1980. Annual report on crop and pasture research.
Arsi Rural Development Project (ex-CADU) , Ethiopia.
FAO. 1970-1982. Various reports on livestock study, Ethiopia.
IAR. 1967-1980. Annual reports on livestock range and pasture
research. Institute of Agriculture Research, Ethiopia.
ILCA 1981-1985. Various reports from highland programme and
Forage Agronomy Group (FLAG) , Ethiopia.
Ministry of Agriculture. 1984. Dairy Project, Animal & Fisheries
Resources Development, Main Department, Min. Agric. , Addis
Ababa, Ethiopia.
Ministry of Agriculture. 1984. Feed and Forage Project, Animal &
Fisheries Resources Development, Main Department, Min.
Agric. , Addis Ababa, Ethiopia.
Ministry of Agriculture. 1984. Livestock Sub-sector Study,
Animal & Fisheries Resources Development, Main Department,
Min. Agric, Addis Ababa, Ethiopia.
481

\LIST OF PARTICIPANTS
BOTSWANA
ETHIOPIA
Mr. Doug R. Chandler
PANESA Working Committee Member
Animal Production Research Unit
P.O. Box 69
Lobatse
Mr. Gubungano Tacheba
Animal Production Research Unit
Private Bag 0033
Gabarone
Mr. Arabia Moyo
Animal Production Research Unit
Private Bag 0033
Gabarone
Ato Daniel Keftasa
South-East Agric. Development Office
Asela-Kalumsa
Dr. J.C. Tothill
PANESA Working Committee Member
Division of Plant Sciences
ILCA
P.O. Box 5689
Addis Ababa
Mr. Johnson J. Hakiza
Forage Agronomy Group
Division of Plant Sciences
ILCA
P.O. Box 5689
Addis Ababa
Dr. James Kahurananga
Forage Agronomy Group
Division of Plant Sciences
ILCA
P.O. Box 5689
Addis Ababa
Dr. Jean Hanson
Forage Agronomy Group
Division of Plant Sciences
ILCA
P.O. BOX 5689
Addis Ababa
Dr. J.R. Lazier
Forage Agronomy Group
Division of Plant Sciences
ILCA
P.O. Box 5689
Addis Ababa
483
r
Dr. E. Kabaija
Division of Animal Sciences
ILCA
P.O. Box 5689
Addis Ababa
Mr. P.N. de Leeuw
ILCA-Kenya
P.O. Box 46847
Nairobi
Dr. Ben H. Dzowela
PANESA Coordinator
ILCA-Kenya
P.O. Box 46847
Nairobi
Prof. J. A. Kategile
PANESA Working Committee Member
IDRC
P.O. Box 62084
Nairobi
Mr. Moses Ooroo Olang
Tana & Athi Rivers Development Authority
P.O. Box 47309
Nairobi
Dr. Aggrey Abate
Department of Animal Production
University of Nairobi
Kabete Campus
P.O. Box 29053
Nairobi
Dr. W.W. Mukhebi
SR-CRSP
P.O. Box 252
Maseno
Mr. P.P. Semenye
SR-CRSP
P.O. Box 252
Maseno
Dr. Shankute Tessema
FAO/UNDP
Dryland Farming Research & Development Project
Katumani
P.O. BOX 340
Maseno
Dr. J.F.M. Onim
SR-CRSP
P.O. Box 252
Maseno
484
Mr. L.K. Menin
NDFRS-Katumani
P.O. Box 340
Machakos
Mr. G. Njarui
NDFRS-Katvunani
P.O. Box 340
Machakos
Mr. E.K. Maranga
National Range Research Station
Kiboko
P.O. Box 12
Makindu
Dr. H.L. Potter
Animal Production Research Department
KARI
P.O. Box 21
Kikuyu
Dr. Moses Karachi
NDFRS-Katumani
P.O. BOX 340
Machakos
Mr. F. Attere
Field Officer for Eastern & Southern Africa
IBPGR
ILRAD Campus
P.O. BOX 30709
Nairobi
Mr. F. Bedini
Undugu Society of Kenya
Dr. V. Cagnolati
Undugu Society of Kenya
Terra Nuova
Italian Volunteer Services
P.O. Box 40417
Nairobi
Mr. K.S.A. Buigutt
Technical Editor
KREMU
P.O. BOX 47146
Nairobi
MADAGASCAR Dr. J.H. Rasambainarivo
Departement de Recherche Zootechnique
et Veterinaires
Division Alimentation Animale
B.P.4, Tananarive (101)
485
MALAWI
MAURITIUS
MOZAMBIQUE
NIGERIA
SOMALIA
Ravelson Ainjara
NORAD
110 Antsirade
Mr. Bjorn Lilleeng
NORAD
110 Antsirade
Mr. H.D.C. Msiska
Ministry of Agriculture
Lunyangwa Research Statioi
P.O. Box 59
Mzuzu
Dr. J.T. Munthali
Ministry of Agriculture
Chitedze Research Station
P.O. Box 158
Lilongwe
Dr. D. Jotee
School of Agriculture
University of Mauritius
Reduit
Dr. Jonathan Timberlake
FAO/UNDPT
C.P. 4595
Maputo
Mr. Celaine da silva
IRIMA
C.P. 1410
Maputo
Mrs. A. D'Hondt
INIA
C.P.U. 595
Maputo
Dr. G. Tarawali
ILCA-Subhumid Programme
21 Ahman Pategi Road
P.M.B. 2248
Kaduna
Dr. Alex Dickie
USAID/RONCO
P.O. Box 574
Mogadishu
Mr. Mohammed Aye
Ministry of Livestock, Forestry & Range
P.O. Box 3567
Mogadishu
486
TANZANIA Prof. A.B. Lwoga
Chairman, PANESA Working Committee
Deputy Vice-Chancellor
Sokoine University of Agriculture
P.O. Box 3004
Morogoro
Mr. M.M.S. Lugenja
Livestock Research Centre
P.O. Box 5016
Tanga
Mr. N. Mukurasi
Uyole Agricultural Centre
P.O. Box 400
Mbeya
Mr. E.W. Kapinga
Livestock Training Institute
Morogoro
Mr. Martin L. Kusekwa
Livestock Production Research Station
P/Bag Mpwapwa
Mrs. A.J. Kitalyi
Livestock Production Research Station
P/Bag Mpwapwa
Mrs. M.A. Kabatange
Livestock Production Research Station
P/Bag Mpwapwa
Mr. R.N. Mero
Pasture Research Centre
P.O. Box 5
Kongwa
Mr. Reinhard Pfeiffer
TIRDEP-SECAP
P.O. Box 72
Lushoto
Mr. S.V. Sarwatt
Animal Sciences S Production Department
Sokoine University of Agriculture
P.O. Box 3004
Morogoro
Mr. P.X. Kapinga
TALIRO
P.O. Box 3021
Malya
Dr. Ghamunga Sudi
Tengeru Livestock Training Institute
P.O. Box 3101
Arusha
487
Mr. I.M. Issae
Ministry of Agriculture & Livestock Development
P.O. Box 9152
Dar es Salaam
Mr . T.J. Myoya
Uyole Agricultural Centre
P.O. Box 4 00
Mbeya
Mr. R.S. Kidunda
Sokoine University of Agriculture
P.O. Box 3004
Morogoro
Mr. A. Kajuni
Agriconsult (T) Ltd.
P.O. Box 446
Tukuyu
Dr. A.M. Macha
Director of TALIRO
P.O. Box 6910
Dar es Salaam
Mr. Ben O. Jalli
P.O. Box 1548
Horogoro
UGANDA Mr. E.N. Sabiiti
Department of Crop Science
Makerere University
P.O. Box 7062
Kampala
Mr. S.G. Byenkya
Animal Production Research Section
P.O. Box 24
Entebbe
ZAMBIA Dr. Jaroslav Kulich
Mochipapa Pasture/Animal Research Station
P.O. Box 630090
Choma
Mr. C.K. Kaonga
Mochipapa Pasture/Animal Research Station
P.O. Box 630090
Choma
ZIMBABWE Mr. C. Madakadze
Henderson Research Station
Private Bag 2004
Mazowe
488

 Printed at ILCA, Addis Ababa, Ethiopia