Small Ruminant Research

and Development
in Africa

Proceedings of the Second

Biennial Conference of the African
Small Ruminant Research Network
AICC, Arusha, Tanzania
7-11 December 1992

September 1994

International Livestock Centre for Africa (ILCA)

Centre international pour l'élevage en Afrique (CIPEA)

Technical Centre for Agricultural and Rural Co-operation (ACP-EEC)

СТА Centre Technique de Coopération Agricole et Rurale (ACP-CEE)
 1

Small Ruminant Research

and Development
in Africa

Proceedings of the Second

Biennial Conference of the African
Small Ruminant Research Network
AICC, Arusha, Tanzania
7-11 December 1992

Edited by:

S.H.B. Lebbie
B. Rey
E.K. Irungu

September 1994

International Livestock Centre for Africa (ILCA)/

Technical Centre for Agricultural and Rural Co-operation (СТА)
P.O. Box 5689, Addis Ababa, Ethiopia
Thi s One

WCZX-LZJ-WA8A
The Technical Centre for Agricultural and Rural Co-operation
The Technical Centre for Agricultural and Rural Co-operation (СТА) operates under the Lomé Convention
between member States of the European Community and the African, Caribbean and Pacific ( ACP) States.
The aim of СТА is to collect, disseminate and facilitate the exchange of information on research, training
and innovations in the spheres of agricultural and rural development and extension for the benefit of the ACP
States.
Headquarters: "De Rietkampen", Galvanistraat 9, Ede, Netherlands
Postal address: Postbus 380, 6700 AJ Wageningen, Netherlands
Tel.: (31) (0) (8380) - 60400
Telex: (44) 30169 СТА NL >
Telefax: (31) (0) (8380) - 31052
Si

ISBN 92-9053-284-X

Correct citation: Lcbbie S H В, Rey В and Irungu E K. 1993. Small Ruminant Research and Development in Africa.
Proceedings of the Second Biennial Conference of the African Small Ruminant Research Network,
AICC, Arusha, Tanzania, 7-11 December 1992. ILCA (International Livestock Centre for Africa)/
CТА (Technical Centre for Agricultural and Rural Co-operation). ILCA, Addis Ababa, Ethiopia.
268 pp.
 Abstract
This volume contains papers and abstracts of the Second Biennial Conference of the African Small Ruminant
Network. In addition to the keynote address, there are nine papers on Small Ruminant Production Systems,
Economics and Policy, nine papers on Small Ruminant Performance and Reproduction, ten papers on Small
Ruminant Health and Reproductive Wastage, eight papers on Small Ruminant Feeds and Feeding Systems
and eight papers on Small Ruminant Genetic Resources and Breeding. Ten poster abstracts covering the
above topics add to the volume.

Keywords
/Sheep//goats//small ruminant production systems//productivity//economics//policy//weaning
practices//agropastoral//pastoral//production constraints//feeds//feed resources/Vfeeding systems/
/digestibility//nutrition//supplementation//diet//reproduction//reproductive wastage//performance
evaluation//health//peste des petits ruminants//infectious diseases//vaccines//breeds//breeding systems/
/breed improvement//Africa/

Résumé
Le présent volume rassemble les exposés et les résumés d 'articles présentés à la deuxième Conférence
biennale du Réseau africaine de recherche sur les petits ruminants. Outre le discours liminaire, il contient
neuf communications sur les systèmes de production, l'économie et les politiques, neuf autres sur les
performances et la production, dix sur la santé et les pertes en reproduction, huit sur l'alimentation et les
systèmes d'alimentation et huit autres enfin sur les ressources génétiques et la sélection. A quoi il convient
d 'ajouter dix séries d 'affiches présentées à cette conférence sur l 'ensemble des sujets énumeres ci-dessus.

Mots Clés
/Ovins//caprins//systèmes de production des petits ruminants//productivité//économie//politiques/
/modes de sevrage//agropastoral//pastoral//obstacles à la production/Zaliments du bétaiU/ressources
alimentaires//sy sternes d'alimentation//digestibilité//nutrition//complémentation//régime alimentaire/
/reproductionZ/pertes en reproduction/ /évaluation des performances//santé//peste des petits ruminants/
/maladies infectieuses//vaccins//espèces//systèmes d 'élevage/Zamélioration génétique//Afrique/

Small Ruminant Research Network Workshop

'

\
 Table of Contents
Preface ix

Constraints and prospects for small ruminant research and development in Africa
A.A. Ademosun 1

SESSION 1. SMALL RUMINANT PRODUCTION SYSTEMS, ECONOMICS AND POLICY

Socio-economic constraints to strategic sheep fattening: Evidence from the

Ethiopian highlands
B.I. Shapiro, M.A. Mohamed-Saleem and L. Reynolds 9

Socio-economic aspects of goat nutrition in Swaziland

B.H. Ogwang, D. Mavimbela, R. Vilakati and G.Z. Khumalo 15

The role of the West African Dwarf goat in the economy of the smallholder
arable farmer in the subhumid zone of Nigeria
O. A. Ikwuegbu, G. Tarawali and R. M. Njwe 19

Some socio-economic factors that affect decisions on small ruminant production by

small-scale farmers — A case study of west Laikipia
M. M. Mucuthi, K. Munei and К. C. Sharma 23

Priority setting for small ruminant research

L. Reynolds 29

Overcoming the constraints of dry matter intake on dual-purpose goat production by

feeding defoliated maize leaves
P.P. Semenye, M. Shisya and W.R. Getz 35

Goat development in Masvingo province, Zimbabwe: the farmers' perspective

C. Shumba 39

Goat production in a mixed cattle-goat system: Effect of stocking and substitution rate
on red soil thornveld stability
J.L.N. Sikosana and J. Gambiza 43

SESSION 2. SMALL RUMINANT PERFORMANCE AND REPRODUCTION

Evaluation de l'activité sexuelle pendant la saison de baisse de fertilité chez

la chèvre de race D'man
L. Derquaoui et O. El Khaledi 49

Etude du cycle oestral chez la brebis Djallonké

M.S. Hounzangbé-Adoté 53

Puberty and early mating performance in subtropical fat-tailed sheep and their crosses
Ferial Hassan, M.T. Mousa, A.M. Aboul-Naga, F. El-Hommosi and G. Abd El-Hafez 61

Preweaning performance of Yankasa sheep under semi-intensive management

O.A. Osinowo, B.Y. Abubakar, M.E. Olayemi, R.O. Balogun, O.S. Onifade,
A.A. Adewuyi, A.R. Trimnell and F.O. Dennar 65

Small Ruminant Research Network Workshop V

Table of Contents

Studies on factors affecting reproductive performance and mortality rates of

Small East African goats and their crosses
L.A. Mtenga, G. C. Kifaro and Berhanu Belay 69

Reproductive traits and disease incidence characteristics of Dorper, Dorper x Masai

and Masai ewes raised under semi-arid conditions in Kenya
W.O. Odenya 75

Impact de la prophylaxie sanitaire sur la productivité des petits ruminants dans

la zone semi-aride du Mali
Mamadou Doumbia 79

The performance of the Small East African goats and their Saanen crosses in Malawi
S.K. Karua andJ.W. Banda 81

The productivity of Small East African sheep and goats in Malawi

J.W. Banda 87

SESSION 3. SMALL RUMINANT HEALTH AND REPRODUCTIVE WASTAGE

Trypanosomiasis induced reproductive wastage in West African Dwarf sheep

O.O. Akinbamijo and L. Reynolds 95

Normes et variations de l'hématocrite et de la formule leucocytaire et particularités

histologiques leucocytaires chez la brebis Djallonké du sud du Sénégal
M. Ndiaye, M.B. Sow et T. Alogninouwa 103

Haemagglutination-inhibition technique for definitive diagnosis of peste des

petits ruminants virus specific antibody
L.O. Wosu andM.C.O. Ezeibe 109

Peste des petits ruminants (PPR) and rinderpest (RP) antibodies in clinically
normal small ruminants in the Cameroon and Nigeria
K.A. Majiyagbe, D. Shamaki, D. Kulu and Т.К. C. Udeani 113

Diseases of small ruminants in central Tanzania

L.E.G. Mboera and J.I. Kitalyi 117

Mortalité des chevreaux croisés Alpine au Burundi

M. Ba, B. Rey et Mesfin Shibre 121

Stocking rate of sheep and survival of helminth infective larvae on Brachiaria pastures
K.J.N. Ndamukong, S.E. Mbomi and S. Killanga 127

Effect of strategic anthelmintic treatment on growth performance and survival rate

of native lambs in the subhumid environment in Mali
S. Tembely, K. Bengaly and W. Berckmoes 133

Intensification of goat production in Uganda and associated health risk management

L. Siefert and J. Opuda-Asibo 137

vi Small Ruminant Research Network Workshop

 Table of Contents

Enquête écopathologique sur les pneumopathies des chèvres en saison fraîche

au Tchad: un outil au service du développement
R. Lancelot, M. Imadine, Y. Mopate et A. Idriss 143

SESSION 4. SMALL RUMINANT FEED AND FEEDING SYSTEMS

Selectivity in sheep and goats fed Albizia, Gliricidia, Leucaena and Tamarind
multipurpose trees
L.A. Mtenga, D.M. Komwihangilo and G.C. Kifaro 151
Influence of amount of feed offered on growth, intake and selectivity: observations
on sheep and goats
A.A.O. Aboud, E. Owen, J.D. Reed and A.N. Said 157
The effect of supplementation of cassava peel (CP) diets with graded levels of palm kernel
cake (PKC) on the performance of growing Djallonké sheep
A.K. Tuah, E.R. 0rskov, F.Y. Obese, D.B. Okai andJ.F.D. Greenhalgh 163
Valeur nutritive des résidus de récolte et des sous-produits agro-industriels offerts à des moutons
peuls du Sahel
G.P. Gongnet, M. Minguey et B.O. Brahint 1 69
Seasonal changes in body condition of the Senegalese Sahel goat: relationship to
reproductive performance
M. Cissé, M. M'Baye, I. Sané, A. Corréa and I. N'Diaye 1 75
Wet season supplementation of West African Dwarf goats raised under traditional
management in the subhumid zone of Nigeria
O.A. Ikwuegbu andS. Ofodile 179

Performance of weaner lambs fed maize leaves and napier grass

A.N. Abate and A. Abate 183
Rumen degradation of elephant grass supplemented with graded levels of
perennial peanut by West African Dwarf sheep
B. Tchinda, D. WegadandR. M. Njwe 187
Metabolisable energy requirements for maintenance and growth of castrate
Zimbabwean (Matabele) goats offered diets differing in roughage content either
ad libitum or restricted
P.R. Hatendi, L.R. Ndlovu, T. Smith and A.N. Said 191
Effect of energy level on milk production by Mossi ewes and Sahelian does
A.J. Nianogo and P. C. Ilboudo 197

SESSION 5. SMALL RUMINANT GENETIC RESOURCES AND BREEDING

Indigenous African small ruminants: a case for characterisation and improvement
J.E.O. Rege 205
Impact sur la productivité des ovins Djallonké de la sélection des béliers sur leur taille
A. Oya, C.V. Yapi et M. Ouattara 213

Small Ruminant Research Network Workshop vii

Table of Contents

Caractères phénotypiques de différenciation et de croissance des agneaux de

race pure Djallonké et croisés Sahélien x Djallonké
C.V. Yapi 215
Sheep production in Zambia: a review of sheep breeding
W.N.M. Mwenya 221
Prospects for breeding for resistance to endoparasites in small ruminants in Africa —
a new ILCA research programme
R. L. Baker, L. Reynolds, A. Lahlou Kassi, J.E.O. Rege, Tekelye Bekelye,
E. Mukassa-Mugerwa and B. Rey 223
Estimate of genetic and phenotypic parameters for the Dorper, Red Maasai and their crosses
P.M. Kiriro 229
Productivity of Blackhead Persian sheep in Tanzania
S.M. Das, J.E.O. Rege and Mesfin Shibre 235
Some production parameters of Blended goats in semi-arid regions of Tanzania
E.T.R. Lyatuu, S.M. Das and J.I. Mkonyi 241
Performances de reproduction des caprins dans les systèmes d'élevage traditionnels
des zones communales au Zimbabwe
F. Monicat, J. Keravec et P.M. Borne 247
Posters - Abstracts 249
List of Participants 255

Small Ruminant Research Network Workshop

 Preface
Two of the major constraints to small ruminant development in the developing world and, in particular, Africa
are limited access to information and limited opportunity for interaction among scientists working in the same
area of interest.
It is against this background that the African Small Ruminant Research Network (SRNET) has made
the promotion and improvement of information exchange on small ruminant research and development one
of its major priorities in addition to collaborative research and training. The network has set out to achieve
its goal through the organisation and sponsoring of a biennial scientific conference, amongst other channels.
This provides a forum for small ruminant scientists in Africa and elsewhere, developers, policy makers and
farmers to meet, exchange ideas and establish partnerships.
This proceedings reflects the oral and poster presentations at the Network's Second Biennial Conference
held in Arusha, Tanzania, from 7 to 11 December 1992. Some 100 participants, representing 30 African
countries, five European countries and various international and regional organisations (ILCA, OAU/IBAR,
NRI, ODA and GTZ) were present. Five scientific sessions were held, at which 48 oral and 10 poster
presentations were made and discussed. The sessions focused on production systems, economics and policy,
performance and reproduction, health and reproductive wastage, feeds and feeding systems and genetic
resources and breeding for both oral and poster presentations.
The Network is grateful to the European Economic Community (EEC) and ILCA for promoting and
supporting the activities of the Network including the holding of the Second Biennial Conference and the
publication of the proceedings. The Natural Resources Institute (NRI) of Great Britain and the Technical
Centre for Agriculture and Rural Co-operation (СТА) are recognised for their support in the collaborative
research and publication of this proceedings, respectively. A lot of thanks go to the ILCA Nairobi support
office for the usual superb logistical support.
Special thanks go to the Tanzanian Government for its support in holding the conference, and in
particular to the Ministry of Agriculture, Livestock Development and Cooperatives and the National Small
Ruminant Research Network for co-hosting the conference. The Network also appreciates the kind
willingness of Dr Ben Moshi, Principal Secretary, Ministry of Agriculture and Livestock Development, to
be the guest speaker. My final thanks and gratitude go to all those who made the publication of this proceedings
possible, particularly Mr Matthew Pritchard, for English editing, and the staff of the Publications Section of
the International Livestock Centre for Africa (ILCA).

Small Ruminant Research Network Workshop

 Constraints and prospects for small ruminant
research and development in Africa
A.A. Ademosun

P. O. Box 182, Ile-Oluji, Ondo State, Nigeria

Abstract ideal for families. Little capital investment in buildings

or other materials is required for their upkeep, and space
Small ruminants have a great potential to affect the and maintenance requirements are low. They are
socio-economic development of the majority of African suitable for family consumption in the absence of
rural [Link] paper discusses the constraints refrigeration for storage, or adequate transportation
on small ruminant production and the future prospects (Ademosun, 1988), and their reproductive efficiency is
for small ruminant production. high. However, the animals are given little attention.
Until recently, small ruminants were not a priority
in the research activities and development programmes
Contraintes etperspectives de la of African governments. This has changed with the
recherche-développement sur establishment of international research institutes under
the Consultative Group on International Agricultural
l'élevage des petits ruminants en Research, particularly ILCA (International Livestock
Centre for Africa). The broadening of the base of
Afrique livestock research in National Agricultural Research
Systems has also boosted small ruminant research,
Résumé which has taken its place besides research into cattle, pig
Les petits ruminants peuvent apporter une and poultry production.
contribution considérable au développement The new orientation of research can be attributed
socio-économique de la majorité des communautés to a number of factors. These include a new awareness
rurales de l'Afrique. Cet article examine les contraintes of the potential of the small ruminant, the need for
quipèsent sur l 'élevage de petits ruminants ainsi que les increased meat production, pressure on land,
perspectives d'avenir. urbanisation, disease problems with other traditional
livestock species and increasing feeding costs for
non-ruminants.
Introduction
Africa has a population of 205 million sheep and 174 Constraints
million goats representing approximately 17% and 31%
of the world total, respectively (FAO, 1990). Within
Africa, the distribution of these small ruminants varies Health
widely, with a higher concentration found in dry areas
than in humid areas. These animals serve primarily as A serious constraint on small ruminant production in
sources of meat, but also provide milk, skins and Africa has been the high prevalence of diseases and
manure. Sheep and goats produce only about 16% of the parasites, particularly in the more humid areas. This
world's meat, despite their high contribution to the total causes high mortality amongst kids and lambs,
world livestock population. Milk and skin production diminishing the benefits of their high reproductive
show similar figures. African small ruminants produce performance.
only 14% of the world's milk and 15% of the world's Preweaning mortality of up to 40% has been
skin. recorded with kids and lambs in Nigeria. However,
The ownership of small ruminants in Africa differs preweaning mortality under the extensive system of
from that of cattle. Only a small percentage of the management may be underestimated. Births which
population own cattle, and rear them mainly in the arid occur when animals are bush grazing and result in the
and subhumid zones. Most people in rural areas own death of the offspring usually go unnoticed. In addition,
small ruminants. further losses are caused by abortions and stillbirths.
The ownership of small ruminants is regarded as One of the most serious diseases of small
an investment. They are sold to meet compelling family ruminants, peste des petits ruminants (PPR), was first
financial obligations or slaughtered for consumption at recognised as a contagious "rinderpest-like" condition
home or at festivals. The size of the animals makes them in goats in Nigeria in 1930. It was first described in Côte

Small Ruminant Research Network Workshop 1

A. A. Ademosun

d'Ivoire in 1942 and later in the Benin Republic during High standards of sanitation and good
1944. The disease has been difficult to quantify due to management are essential to control disease. Table 1
the lack of precise statistical records. However, it has shows the incidence of diseases at the goat unit of the
been stated that it is the most destructive viral disease Obafemi Awolowo University in south west Nigeria.
among small ruminants (Bourdin, 1983). Goats are more When the flock was started in mid- 1981, animals were
susceptible to PPR than sheep and young animals are kept mainly on concrete floors spending four hours a day
more susceptible than adult ones. The use of on pasture. A change in management practices,
tissue-culture rinderpest vaccine (TCRV) is effective in including slatted floors, zero-grazing and once-a-year
the control of PPR. vaccinations against PPR with TCRV reduced cases of
Parasitism ranks high among the factors that limit disease. The goat population increased from 57 in 1982
the productivity of small ruminants although its effect is to 216 in 1985.
often underestimated. Akerejola et al (1979) have
estimated that the losses from parasitic diseases in Table 1. Incidence of disease among the goat flock at
Nigeria's subhumid zone are higher than those Obafemi Awolowo University Farm, Nigeria in
1982 and 1985.
attributable to PPR. Helminthiasis was the most
prevalent condition encountered. Assoku ( 1 980) studied No of cases
the helminths of sheep and goats on the Accra plains of Disease 1982 1985
Ghana. He found that 80% of the sheep and 88.3% of Helminthiasis 100 -
the goats were infected. The problem of parasitism is
Scabies 80 2
compounded by the fact that, under the traditional
system, livestock are usually reared extensively. This Tick borne diseases 41 -
increases infestation and makes control measures Lice 10 -
difficult. Young animals are the most affected and Ecthyma 12 18
parasitism could aggravate other conditions such as -
PPR 8
nutritional stress and susceptibility to disease.
Caseous lymphadenitis 10 5
About 46% of tropical Africa is infested by tsetse
flies, with the highest infestation in the humid (90%) and Diarrhoea 9 19
subhumid (68%) zones. It has generally been believed Crippling 3 -
that small ruminant breeds adapted to the humid zones Localised inflammation 10 -
are trypanotolerant. However, cases of trypanosomiasis
Abortions 7 14
have been recorded amongst indigenous breeds in these
Hoof problems - 4
zones. In a village survey of south west Nigeria,
trypanosomiasis was responsible for more of the Pneumonia - 1
diseases recorded than any other (ILCA, 1982). Kariuki Bloat - 1
and Jacobsen (1980) also reported trypanosomiasis in Retained placenta - 2
sheep in Kenya. Close attention needs to be paid to this
disease as it frequently occurs among animals under Others 3 2
stress, especially nutritional stress. Animals imported Total 293 68
for upgrading indigenous stock are also susceptible.
External parasites cause extensive losses among Nutrition
small ruminants, especially in humid areas. The two
most serious vectors are ticks and mites. The irritation Inadequate feeding is a major limiting factor to small
and dermatitis that accompany mange infestation can ruminant production in tropical Africa. Fodder is ofpoor
reduce the value of the skins for sale to the leather nutritional value for most of the year due to the rainfall
industry and the meat with skin for consumption as food. pattern. In the arid and semi-arid zones, rainfall is less
Haemoparasites transmitted by ticks are quite common. than 600 mm and between 600-1000 mm per year,
Assoku (1979) studied the incidence of bloodborne respectively. The management system is nomadic or
parasites of sheep and goats in southern Ghana. He transhumant pastoralism. Animals have to trek great
found moderate to heavy infestation of tick-borne distances in search of fodder and water. The quality of
parasites in both species. Extensively reared animals available forage is low and browse species which can
provide higher levels of proteins and carbohydrates are
showed greater infestation than semi-intensively kept
sparsely dispersed. In the humid and subhumid zones,
ones. Dwarf breeds were less infested than long-legged
up to six months of the year can be rainless, resulting in
breeds, but infestation increased with age. poor quality forages. The rapid buildup of cell-wall
Other diseases that have limited the productivity of materials and decline in crude protein (CP) content with
small ruminants in tropical Africa include pneumonia, maturity reduces the nutritional value of the forages.
coccidiosis, contagious caprine pleuropneumonia, Little is known about the nutritional value, distribution,
ecthyma, caseous lymphadenitis and brucellosis. palatability, seedling vigour and seasonal production of
Individually, these diseases might not constitute serious the forage species that characterise the natural grassland.
problems, but combinations of them or their occurrence This is particularly true of the arid, semi-arid and
under marginal conditions could result in serious losses. subhumid areas which contain 75% of the sheep and

Small Ruminant Research Network Workshop

 Constrainsts and prospects for small ruminant etc..

80% of the goats of tropical Africa and where the by-products, molasses etc). The types and quantities
rangeland is the most important source of food. available tend to be location and season specific.
A variety of grasses and legumes have been used The feeding quality of the rangelands in semi-arid
to improve pastures and high yields have been recorded. and subhumid areas can be improved by oversowing the
Under station conditions a yield of 20 tonnes of dry natural grassland with legumes. This has been
matter per hectare (DM/ha) has been obtained with successful in the subhumid zone of Nigeria (ILCA,
Pennisetum purpureum in the humid zone of Nigeria. 1 984). The legume of choice has been Stylosanthes. The
Under ranching conditions, with systematic clearing of ILCA work was with cattle but could be useful for small
woody vegetation, Jahnke (1982) reported a yield of 6 1 ruminants. Crop residues also provide good alternative
DM/ha with Cynodon dactylon. Results from West feeds in the dry season. Their feeding value and DM
Africa indicate annual DM yields of between 10-18 t/ha production can be improved if the crops are planted with
for Panicum maximum without fertiliser application. forage legumes. With the selection of the appropriate
Using fertilisers, appropriate cutting regimes and legume, grain yield need not decrease (Table 2).

Table 2. Grain and fodder yields of sorghum/forage legume mixtures in the subhumid zone of Nigeria.
Legume/crop
Crop mixture Grain yield residue DM (kg/ha) Total fodder Total CP
Sorghum alone 1296 4467 - 4467 126
Sorghum + S. hamata 313 1685 2778(11.4) 4463 363
Sorghum + S. guianensis cv Cook 388 1555 2063(12.8) 3618 326
Sorghum + M. atropurpureum 356 2111 1296(14.2) 3407 241
Sorghum + C. pascuorum 1019 2981 1204(14.9) 4185 259
Sorghum + A. vaginalis 1092 2519 926(10.8) 3445 168
Sorghum + M. lathyroides 1297 2761 1481 (16.5) 4222 316
Source: ILCA (1984).
Figures in parentheses are CP (%) of fodder crop.
S = Stylosanthes; M = Macroptilium; C = Centrosema; A = Alysicarpus.

suitable mixed improved pasture species, annual yields Breeds

of up to 30 t DM/ha are possible in humid areas.
However, high yields are not necessarily matched by Small ruminant breeds in tropical Africa are
high nutritional values. Tropical grasses are generally characterised by small birth weights, low milk
low in C P content. As they grow, the lignin content and production, slow growth rates and small mature weights.
other cell-wall materials build up rapidly, adversely The West African Dwarf goat has been variously called
affecting digestibility. Studies have shown that the use the Nigerian Dwarf, Cameroun Dwarf, Forest Djallon
of a grass such as Panicum maximum or Cynodon and Fouta Djallon, which indicate its wide geographical
nlemfuencis alone is not adequate to allow optimum spread. These animals weigh 20-45 kg at maturity, have
production in sheep and goats (Ademosun et al, 1985). short hair and attain sexual maturity at about five
The use of browse and other feed sources has proved months. Table 3 shows the production traits of the West
African Dwarf sheep and goats in south western Nigeria
satisfactory. Browses such as Gliricidia sepium and
(Mack, 1983).
Leucaena leucocephala have been successfully used for
small ruminant production systems in alley farms
(Sumberg, 1985) and intensive feed gardens Table 3. Production traits of village sheep and goats of the
(Ademosun, 1988). WAD breed in south west Nigeria.
Besides the use of browse, other strategies can be Trait Sheep Goat
employed to improve the feeding of animals. During the Litter size 1.23 1.52
dry season, the quality of available herbage is so low
that, unless the animals have access to supplementary Parturition interval (days) 322 267
feeds, they lose weight. These supplementary feeds can Birth weight (kg) 2.12 1.63
be obtained from agro-industrial by-products such as Preweaning liveweight %
residues of oil extracted from oil bearing seeds gain (g/day) 74 38
(groundnuts, coconut, palm kernels, cotton seed, Live weight at 12 months 19.7 9.6
soyabean etc), by-products of grain processing (maize,
rice, wheat, sorghum, millet etc), peelings of crops Survival index for
lambs/kids to 90 days 0.84 0.75
(yams, cassava, potatoes, plantains etc) and industrial
by-products (brewers' dried grains, fruit cannery Source: Mack (1983).

Small Ruminant Research Network Workshop

A. A. Ademosun

Ninety-nine per cent of the goats and most of the zones, cattle are reared with sheep and/or goats. In the
sheep in Botswana belong to indigenous breeds with humid zone, animals generally graze freely, with access
high variation in conformation, colour and type (Molefe, to household and kitchen wastes when available. These
1986). Small East African (SEA) goats are slightly are supplemented with bush grazing on low quality
bigger than their West African counterparts. According forages or browses. In some places, animals are tethered
to Mburu (1986), SEA goats are the most numerous in and fed kitchen wastes supplemented with zero-grazing.
Kenya. Okello (1983) indicated that SEA goats weigh Sheep and goat management under traditional
25 kg at maturity and reach sexual maturity at four systems has the following characteristics:
months when they weigh 14-16 kg. The local ecotype (a) stock owners are usually crop farmers (mostly
of the SEA goat in Burundi has recorded a litter size of arable crops in the arid and subhumid zones and tree
1 .75 under improved management. Primiparous females crops in the humid zone) for whom livestock
have a smaller litter size of 1 .3. The South African Bantu keeping is of secondary importance;
(SAB) goat is similar to the SEA. It is, however, bigger
and found in the southern parts of the zone. Like the (b) most households keep only a few sheep and goats,
other breeds, it is a meat animal. The Landim goat in also keeping other livestock such as pigs, horses,
Mozambique is similar to the SEA (Rocha, 1986). chickens and domestic animals such as cats and
Dlamini and Lebbie (1986) have stated that 99% of the dogs (Table 4);
goat and 81% of the sheep population in Swaziland (c) the flock structures do not reflect good breeding
belong to indigenous breeds. strategy;
Attempts to improve indigenous breeds by
crossing them with exotic breeds have been (d) veterinary and livestock improvement services are
unsuccessful due to the inability to sustain initial efforts minimal; and
and lack of coordination. The Black Headed Nungua (e) the management systems are not integrated with
sheep developed in Ghana and the Permer sheep (a cross crop production.
between the Persian and Merino) in Nigeria are These management practices are not ideal.
examples. Teeluck (1986) for Mauritius, Kassahun Mortality rates (particularly amongst the young) and
Awgichew and Getanen Hailu (1986) for Ethiopia and losses from accidents, theft and predators are high.
Dlamini and Lebbie (1986) for Swaziland, have given Research innovations and extension services have little
examples of such efforts. Crossbreeding should be impact on the production systems, and the benefits of an
preceded by rigorous selection among the adapted integrated cropAivestock production system are lost.
breeds for desirable characteristics and good
productivity. Meticulous performance records need to
be kept and accurate livestock census data are needed.
Each breed must be well defined. Using improved Research and development
breeds in the production process needs to be backed up
by a strong livestock extension service which, at present, The major problems of small ruminant production in
does not exist in many countries. Africa can all be tackled with the appropriate research.
In order to be effective, research must be targetted at the
smallholder farmer. It must also be systems oriented,
Management practical and adaptable. The most prevalent diseases in
each locality should be given priority. This can
Small ruminants in tropical Africa are kept under substantially improve animal production, as
traditional extensive systems. In the arid and subhumid demonstrated by ELCA in the control of PPR and mange

Table 4. Livestock species in households in the humid and subhumid zones of Nigeria.
2
Eruwa (У umid)*1 Kamerukaje (Subhumid)** Ganawuri (Subhumid)**3
Species No % No % No %
Sheep 84 10.9 1 2.6 5 12.2
Goats 521 67.8 39 100.0 38 92.7
Pigs 55 7.2 29 74.4 5 12.2
Chickens 635 82.6 33 84.4 37 90.2
Dogs 135 17.6 31 79.5 30 73.2
Cats 214 27.8 - - 8 19.5
* Source: Sellers et al ( 1976).
** Source: ILCA (1985).
1 Total number of households 769.
2 Total number of households 39.
3 Total number of households 4 1 .

Small Ruminant Research Network Workshop

 Constrainsts and prospects for small ruminant etc..

in some villages in the humid zone of Nigeria (ILCA, The FAO (Food and Agriculture Organization of
1984). the United Nations) has shown an interest in alley
The nutritional problem is a serious one, farming in some West African countries. It links crop
particularly during the dry season. This situation will be and livestock production. The legume browses provide
worsened if livestock productivity and population mulch which improves soil fertility, checks erosion and
increase. Research into providing good quality, shortens the fallow period. They also provide fodder for
year-round feed for the small ruminants is therefore very small ruminants on a cut-and-carry basis.
important. Adapted forage species can provide nutrients Management systems need to be improved in order
for small ruminants. The agronomic characteristics of to improve livestock performance.
these forages need to be explored in order to achieve The reproductive performance of small ruminants
maximum benefits from them. Fertiliser requirements, is high, therefore research into this area need not be a
frequency of cutting, cutting height, grass/legume high priority. Selecting for desirable traits could be used
mixtures and other factors affect the yield and nutritional to improve breeds. Cross-breeding should be kept
value of forages. These need to be studied. Browses between adapted breeds. Imported exotic breeds are
occupy an important place in small ruminant feeding, usually more susceptible to diseases, parasites and the
particularly during the dry season. This is because most harsh tropical environment. Selection and breeding
browse species are drought resistant and provide work is slow, expensive, painstaking, tedious and
proteins, vitamins and minerals which are lacking in the requires large populations. It should only be embarked
herbage species in the range. The use of browse in the upon when the resources are adequate, and must be
various locations needs to be investigated. Household systematic, consistent and aimed at developing breeds
and kitchen wastes and agro-industrial by-products that will succeed under field conditions.
provide valuable supplementary feed. Their feeding Institutional support for livestock development
value and rates of incorporation need to be determined. programmes is generally weak. Many African
If further processing is needed by foodstuffs before governments make minimal budgetary allocations to
livestock development. Existing government agencies
their use, it must be simple, adaptable and adoptable. For
may need to be reorganised to operate more effectively.
example, the use of sodium hydroxide on straw and other
Routine vaccination against PPR is cheap. Control of
poor quality roughages may not be attractive due to its
tsetse flies and other disease vectors should have a high
cost and the hazards involved in its handling. However,
priority in small ruminant development programmes.
the use of ash from various farm wastes may serve as an
The marketing of small ruminants and their products is
alkali source and achieve the same objective. The
largely unorganised and leads to exploitation of the
nutrient requirement tables of the ARC (Agricultural primary producers. Lack of credit has hampered small
Research Council) and NRC (National Research ruminant production in many places. Organising
Council) still serve as references for sheep and goat farmers into cooperatives and establishing fattening
requirements in tropical Africa, despite the fact that schemes run by smallholders will enhance productivity.
various workers have established energy and protein Conferences, workshops and seminars for sub-regional
requirements of tropical sheep and goats. There is a need groupings will establish and enhance communication
to collate this information and extend it to include channels among interested groups.
vitamins and minerals.
Research into the traditional husbandry system for
sheep and goats is a priority. Researchers must be aware
that small ruminants are usually reared with cattle in
Conclusion
pastoral systems, flock sizes are small and farmers in Tropical Africa has a high population of small ruminants
agro-pastoral and village systems see rearing small but productivity is low with low contribution to the total
ruminants as secondary to farming. Research into GDP. The ownership pattern of small ruminants is wide
integrating crop and animal production is essential, spread and the previously neglected animals are gradually
especially for the areas where crop production is the gaining a respectable place in research and development
farmers' primary preoccupation. In wetter areas, simple programmes. The constraints to production and strategies
and inexpensive structures can be designed to provide to enhance development are discussed.
shelter. This will help reduce high mortality rates
amongst kids and lambs, which are partly due to their
exposure to inclement weather conditions. Structures References
developed at Obafemi Awolowo University in Nigeria
Ademosun A A. 1988. Trends in small ruminant production
based on local materials are already being adopted by for the last two decades and its future in West and Central
farmers to provide protection for goats. Africa. Keynote address. In: Adeniji K O (ed),
ILCA's alley farming with the use oiLeucaena and Improvement ofsmall ruminants. OAU (Organization of
Gliricidia in the humid zone and fodder banks involving African Unity), Nairobi, Kenya.
oversowing the natural grassland with legume Ademosun A A, Jansen H J and van Houtert V. 1985. Goat
(Stylosanthes) in the subhumid zone are good adaptable management research at the University of Ife. In:
programmes. Sumberg J E and Cassaday K (eds), Sheep and goats in

Small Ruminant Research Network Workshop

A. A. Ademosun

humid West Africa. Proceedings of the Workshop on Small Ruminants in Eastern and Southern Africa.
Small Ruminant Production Systems in the Humid Zone Nairobi, Kenya, 18-22 August 1986. OAU
of West Africa held in Ibadan, Nigeria, 23-26 January (Organization of African Unity), Nairobi, Kenya. pp.
1 984. ILCA (International Livestock Centre for Africa), 269-274.
Addis Ababa, Ethiopia. pp. 34-37.
Mack S D. 1983. Evaluation of the productivities of West
Akerejola O O, va vee Schillhom T W and Njoku C О. 1979. African Dwarf sheep and goats in southwest Nigeria.
Ovine and caprine disease in Nigeria. A review of Humid Zone Programme Document 7. ILCA
economic losses. Bulletin of Animal Health and (International Livestock Centre for Africa), Ibadan,
Production in Africa 27:65-70. Nigeria. 39 pp.
Assoku R К G. 1979. A study of the incidence of blood borne Mburu J W S. 1986. Small ruminant production in Kenya. In:
parasites of livestock in southern Ghana. Bulletin of Adeniji К О and Kategile J A (eds), Proceedings of the
Animal Health and Production in Africa 27:29-39. Workshop on the Improvement of Small Ruminants in
Assoku R К G. 1980. Studies of parasitic helminths of sheep Eastern and Southern Africa. Nairobi, Kenya, 18-22
and goats in Ghana. Bulletin of Animal Health and August 1986. OAU (Organization of African Unity),
Production in Africa 29: 1-10. Nairobi, Kenya. pp. 275-289.
Bourdin P. 1983. History, epidemiology and economic Molefe D S. 1986. Sheep and goat production in Botswana. In:
significance of PPR in West Africa and Nigeria in Adeniji К О and Kategile J A (eds), Proceedings of the
particular. In: Hill D H (ed), Peste des petits ruminants Workshop on the Improvement of Small Ruminants in
(PPR) in sheep and goats. Proceedings of the Eastern and Southern Africa. Nairobi, Kenya, 18-22
International Workshop held at UTA, Ibadan, Nigeria, August 1986. OAU (Organization of African Unity),
24-26 September 1980. ILCA (International Livestock Nairobi, Kenya. pp. 235-243.
Centre Africa), Addis Ababa, Ethiopia. pp. 10-11. Okello К L. 1983. A review of goat production in Uganda. In:
Dlamini S and Lebbie S H В. 1986. Small ruminant production Joint IFS/ILCA Workshop on Small Ruminant Research
systems in Swaziland: An overview. In: Adeniji К О and in the Tropics. Addis Ababa, Ethiopia, 24-29 October
Kategile J A (eds), Proceedings of the Workshop on the 1983. IFS Provisional Report 14. IFS (International
Improvement of Small Ruminants in Eastern and Foundation for Science), Stockholm, Sweden, pp.
Southern Africa. Nairobi, Kenya, 18-22 August 1986. 51-58.
OAU (Organization of African Unity), Nairobi, Kenya. Rocha A. 1986. Production traits in small ruminants in
pp. 357-364. Mozambique. In: Adeniji К О and Kategile J A (eds),
FAO (Food and Agriculture Organization of the United Proceedings of the Workshop on the Improvement of
Nations). 1990. Production Yearbook. Volume 44. FAO, Small Ruminants in Eastern and Southern Africa.
Rome, Italy. 283 pp. Nairobi, Kenya, 18-22 August 1986. OAU
ILCA (International Livestock Centre for Africa). 1982. (Organization of African Unity), Nairobi, Kenya.
Humid zone programme. Annual Report l982. ILCA, pp. 315-322.
Addis Ababa, Ethiopia. 12 pp. Sellers К C, Dipeolu О О, Falade S and Babalola R. 1976.
ILCA (International Livestock Centre for Africa). 1984. Household livestock in the Eruwa district of Ibarapa
Humid zone programme. Annual Report 1984. ILCA, division, Oyo State. Nigerian Journal of Animal
Addis Ababa, Ethiopia. 21 pp. Production 3:33-37.
ILCA (International Livestock Centre for Africa). 1 985. Small Sumberg J E. 1985. Small ruminant feed production in a
Ruminant and Camel Group Newsletter 4. ILCA, Addis farming systems context. In: Sumberg J E and Cassaday
Ababa, Ethiopia. К (eds), Sheep and goats in humid West Africa.
Proceedings of the Workshop on Small Ruminant
Jahnke H E. 1982. Livestock production systems and livestock Production Systems in the Humid Zone of West Africa
development in tropical Africa. Kieler held in Ibadan, Nigeria, 23-26 January 1984. ILCA
Wissenschaftsverlag Vauk, Kiel, Germany. 253 pp. (International Livestock Centre for Africa), Addis
Kariuki D P and Jacobsen P. 1 980. Trypanosomiasis in sheep, Ababa, Ethiopia. pp. 34-37.
case report. Bulletin ofAnimal Health and Production in Teeluck J P. 1986. Small ruminant production in Mauritius.
Africa 28:7-9.
In: Adeniji К О and Kategile J A (eds), Proceedings of
Kassahun Awgichew and Getaneh Hailu. 1986. Review of the Workshop on the Improvement of Small Ruminants
small ruminant research and development activities in in Eastern and Southern Africa. Nairobi, Kenya, 18-22
Ethiopia. In: Adeniji К О and Kategile J A (eds), August 1986. OAU (Organization of African Unity),
Proceedings of the Workshop on the Improvement of Nairobi, Kenya. pp. 303-313.

Small Ruminant Research Network Workshop

 SESSION 1

Small Ruminant Production Systems,

Economies and Policy
B.I. Shapiro, M. A. Mohamed Saleem and L. Reynolds

Socio-economic constraints to strategic sheep fattening:

Evidence from the Ethiopian highlands

B.I. Shapiro, M.A. Mohamed-Saleem andL. Reynolds

International Livestock Centre for Africa

P.O. Box 5689, Addis Ababa, Ethiopia

Abstract Introduction
This study provides the economic analysis of on-farm Livestock productivity in both Ethiopia and sub
strategic feed trials carried out by ILCA (International Sanaran Africa is low. Recent increases in livestock
Livestock Centre for Africa) since 1985. There is the production have mainly been due to expansion of herds,
potential to increase farmers' incomes by strategic not improvements in productivity (Jahnke, 1982).
feeding to meet the increased demand created by According to Jahnke (1982), many areas have rapidly
holidays and other occasions. However, the results of increasing populations, resulting in small farm sizes and
the economic analysis show that current feed packages less pastureland. Intensified feed and livestock
are not economically viable and do not result in production may be one way to raise production per land
profitable strategic feeding. and livestock unit in a sustainable fashion. Livestock
This study also indicates that a number of also earn farmers cash, which can then be used to
socio-economic factors need to be considered in the increase the production of their food crops (Fitzhugh et
design of station trials. The potential benefits of al, 1992).
African farmers often fatten a few sheep to sell
including economic considerations at the beginning of
on-station research are highlighted. during feasts and holidays when prices are relatively
high. Traditionally, sheep are raised on native pastures
and crop residues. Animals which are to be sold during
peak demand periods may have some supplemental
Obstacles socio-économiques à feeding, but this traditionally includes only surplus
l'engraissement stratégique des grains and leftover household food. Forages, grains and
concentrates could be used to increase animals' weight
moutons dans les hauts plateaux and therefore the price they will sell for.
éthiopiens This study presents an economic analysis of
on-farm strategic sheep feeding trials carried out by
Résumé ILCA in the highlands (cool tropics) of Ethiopia since
1985. The purpose of these trials was to identify feeding
Cette étude présente l'analyse économique des strategies that could sustainably raise animal
essais d'alimentation raisonnée effectués en milieu productivity and farmer incomes, particularly where
paysan par le CIPEA (Centre international pour pasture land is becoming scarce due to crop growing.
l'élevage en Afrique) depuis 1985. En permettant au The objectives of the economic analysis presented here
paysan de satisfaire la demande croissante en mouton à are:
l'époque des fêtes et à d'autres occasions, l'ali
1. to demonstrate the returns that could be obtained
mentation raisonnée des ovins apparaît comme une
from strategic fattening using different feed types,
possibilité d'augmenter les revenus. Cependant, il
with and without veterinary care;
ressort des résultats de l'analyse que les stratégies
alimentaires utilisées ne sont pas économiquement 2. to provide nutritionists and animal production
viables et ne débouchent pas sur une spéculation specialists who are devising strategic sheep
économiquement intéressante. L'étude montre éga fattening methods with information on their
lement qu'un certain nombre de facteurs socio- potential for adoption, as well as constraints.
économiques doivent être examinés dès le stade de la Other fundamental research questions about
conception des essais en station. Les avantages qui on-farm research into intensified livestock production
pourraient découler de la prise en considération des practices and the role of economic analysis in livestock
facteurs économiques dès le début de la recherche en technology research are also addressed. Results of two
station sont présentés. indicative trials are presented in detail.

Small Ruminant Research Network Workshop

 Socio-economic constraints to strategic sheep etc...

Methodology rams and fatten them for sale as part of their normal
production activities. At Debre Zeit a high proportion of
farmers (80.4 %) fatten sheep for a period of 1-2 years
Data source (Lili, 1990). Sheep intended for fattening are usually
purchased at local markets or taken from the flock when
Production data from eight on-farm trials done by ILCA 1-2 years old (Lili, 1990).
were analysed with partial budgeting to assess the The ILCA on-farm site at Hidi/Debre Zeit is
financial viability of strategic sheep fattening. located 40 kilometres south of Addis Ababa and 13 km
east of Debre Zeit town. It is 1 850 metres above sea level
The partial budgeting methodology (masl). The temperature ranges from a mean minimum
of 10 C to a mean maximum of 22°C. Annual average
Partial budgeting is often used to evaluate the potential
profitability of new technologies and provides a rainfall is 800 mm and is bimodal. The main rains occur
criterion for choosing between alternatives. It involves between June and September (Gryseels and Anderson,
tabulating the costs and benefits of a small change in 1983). Most crop production takes place during these
farm practice. Only income and expenses which change rains. Average livestock holdings at Hidi/Debre Zeit are
are considered. The farmer's objective is assumed to be reported to be 6.2 sheep, 1.5 oxen, 1.9 cows, 1.7
the maximisation of profits. His other objectives can be donkeys, 0.1 mules and 1.1 horses (Asfaw et al, 1987).
accounted for in sensitivity analysis and qualifying the Debre Zeit is known for the quality of the teff that is
results (Boehlije and Eidman, 1984). grown there.
Partial budgeting measures profits (or losses) The ILCA on-farm site at Deneba/Inewari is
which are the net benefits or difference between gains located 50 kms west of Debre Berhan town, 1 80 kms
and losses for the proposed change. Gains include added north of Addis Ababa and is at an altitude of 2600 masl.
returns and reduced costs, losses include added costs and The temperature ranges from a mean minimum of 6°C
reduced returns. The method entails calculating net to a mean maximum of 18°C. Annual average rainfall is
income (N1), i.e. the amount of money left when total 880 mm and is bimodal, resulting in two crop seasons.
variable costs (TVC) are subtracted from total returns The main rains occur between July and October and the
(TR): short rains occur between February and April (de Leeuw
et al, 1992). The most important crops are wheat, teff,
NI = TR-TVC (1)
faba bean and chickpea.
Total variable costs include the costs of all inputs
that change due to the change in production technology. The supply of land is inelastic at both sites relative
The most important criterion in deciding whether or not to labour, but more so at Deneba/Inewari. Average land
to adopt a new technology is the change in net income holdings are approximately 2.5 hectares at Debre Zeit
(ANI). This amount is the difference between the change (Gryseels and Anderson, 1983), but only 1.8 ha
in total returns (ATR) and the change in total variable approximately at Deneba/Ineware (McCann, 1992).
Labour in both areas is normally provided by members
costs (ATVC):
of the farm household or neighbouring farmers.
ANI = ATR -ATVC (2) A survey conducted in the Debre Zeit area found
Assuming that capital is not a constraint, the that 50% of the farmers considered feed shortage to be
technology with the highest ANI is chosen. However, the major constraint to improved livestock production.
new technologies normally require investment, Disease was seen as the major constraint by 46% of the
therefore additional capital is necessary. When capital farmers (Asfaw et al, 1987). Traditionally, sheep
is limited, the extra (or marginal) costs should be fattening is based on grazing, aftermath grazing, crop
compared with the extra (or marginal) net benefits. The residues, and residues of local beer (tella). Some farmers
marginal rate of return (MRR) measures the increase in also feed the animals faba beans and chickpea grain
net income (ANI) associated with each additional unit of (Asfaw et al, 1987).
expenditure (ATVC):
MRR = A NI/ATVC (3) Trials and design
The MRR measures the effect of additional
investment in a new technology on additional net returns
(CIMMYT, 1988). Trials at Hidi/Debre Zeit, East Shoa Region

The study area Five fattening trials were carried out at Hidi/Debre Zeit.
The first two trials took place in 1987 and 1989 for a
Ethiopia has 490,000m2 or about 50% of all the highland duration of 24 and 34 weeks, respectively. The sheep
area of the cool tropics of Africa. In the central-eastern were sold during the Ethiopian Christmas in January of
highlands of Ethiopia the indigenous sheep are Menz, each year.
which are medium-sized and fat-tailed (Epstein, 1971) The treatment groups in Trial 1 were as follows:
and Arsi-Bale, which are thin-tailed. As in most of the
Ethiopian highlands, smallholder farmers at ILCA's • Treatment 1: Grazing only (ad libitum for 8 hours;
on-farm research sites in Debre Zeit and Inewari castrate 8 g salt per head/day);

10 Small Ruminant Research Network Workshop

B.I. Shapiro, MA. Mohamed Saleem and L. Reynolds

• Treatment 2: Pen-feeding (teff straw, 800 g;sesbania head/day oats hay plus maize stover, cowpea hay and
leaves, 375 g; salt, 8 g per head/day). wheat straw. Animals in Trial 5 (wet season) received
• Treatment groups for Trial 2 were as follows: oats hay only at 500 g per head/day. Animals in all
groups had unlimited access to urea and molasses block.
• Treatment 1 : Grazing (plus 400 g teff straw per head/
day due to poor pasture);
Trials at Deneba/Inewari, North Shoa Region
• Treatment 2: Grazing plus supplementation (200 g
sesbania, 50 g noug cake and 50 g
Three fattening trials were carried out in Denebe/
wheat bran per head/day); Inewari. Trial 1 compared supplementary feeds
• Treatment 3: Pen-feeding (700 g teff straw, 200 g normally grown on-farm; a high proportion of oats straw
sesbania leaves, 50 g noug cake and versus oats grain under zero-grazing, with and without
50 g wheat bran per head/day). parasite control. Zero-grazing was chosen due to the
shortage of grazing land in the area. Participating
As not enough sesbania was available for Trial 2, farmers purchased 40. Sheep the trial started in August
50 g noug cake and 50 g wheat bran were substituted for 1 988 and lasted 36 weeks. The animals were sold during
100 g sesbania in treatments 2 and 3. These treatments Easter 1989. Preventive health care included drenching
also included 8 g salt and 5 g bone meal per head/day. for nematodes and trematodes, dipping against
Each feed treatment group was further subdivided to test ectoparasites and vaccination against sheep pox and
pasteurellosis. The treatment groups were as follows:
the effect of veterinary prophylaxis.
For Trial 3, 95 sheep with an average live weight • Treatment 1 : High proportion of oats hay ( 1 000 g
of 23.6 kg were purchased in early January, 1990. 17 oats hay per head/day);
sheep (18%) died between January and March, 1990. • Treatment 2: High proportion of oats grain (450 g
The trial started with 78 sheep but 6 more (6%) died
oats grain and 100 g oats hay per head/
between May and June, 1990. The sheep had been
day).
vaccinated against anthrax, sheep pox and
All sheep were fed 100 g horse beans and 50 g
pasteurellosis. All had been drenched against internal
parasites. The cause of death could not be determined, rough peas per day.
therefore the surviving sheep were used for the Trial 2 took place during 1989 and lasted only 10
economic analysis. 72 sheep were sold after 135 days. weeks. Participating farmers purchased 40 sheep from
The treatment groups involved were as follows: the local market with a mean weight of 18.3 kg. The
same experimental design as Trial 1 was followed. By
• Treatment 1 : Pen-fed ( 1 50 g teff straw, 600 g wheat the end of the experiment four sheep had died. The
bran, 400 g cotton seed per head/day); surviving 36 sheep were sold by the farmers at Easter.
• Treatment 2: Grazing plus supplementation (300 g In Trial 3, participating farmers purchased 120
cotton seed, 150 g teff straw and 300 g sheep from surrounding markets in September 1 989 and
wheat bran per head/day); the sheep were sold during Easter 1990. Due to results
• Treatment 3: Grazing plus supplementation (200 g from the previous two trials at Deneba/Inewari, a 30
week fattening period was chosen. Changes were also
(200 g cottonseed cake, 1 50 g teff straw
made to the feed composition and amount. A high
and 200 g wheat bran per head/day).
proportion of oats grain and cottonseed cake were
compared with additional feed in both diets for the last
Trials 4 and 5 involved two batches of sheep which
five weeks, with and without health care. Health care
were fattened for 10 weeks each in 1991. The sheep in
included drenching for nematodes and trematodes,
Trial 4 were fattened during the long dry season (period
dipping against ectoparasites and vaccination against
1). The sheep in Trial 5 were fattened during the wet
sheep pox and pasteurellosis.
season (period 2). 147 and 120 sheep between 20-22 kg
live weight were bought from local markets for the two For the first 25 weeks the treatment groups were as
follows:
fattening periods respectively. After being vaccinated
against pasteurellosis and sheep pox they were treated • Treatment 1 : High proportion of oats grain (400 g of
for internal and external parasites. The treatment groups oats grain, 300 g wheat straw, 250 g
in each trial were as follows: cottonseed cake per head/day);
• Treatment 1: Grazing only; • Treatment 2: High proportion of cottonseed cake:
• Treatment 2: Grazing plus concentrates (300 g (400 g of cottonseed cake, 300 g wheat
wheat bran, 300 g cottonseed cake per straw, 200 g oats grain per head/day).
head/day); For the last five weeks the treatments were
• Treatment 3: Grazing plus supplementation with modified as follows:
forages. • Treatment 1 : High proportion of oats grain (650 g
Forage supplementation varied by season. oats grain, 300 g wheat straw, 350 g
Animals in Trial 4 (dry season) received 500 g per cottonseed cake per head/day);

Small Ruminant Research Network Workshop 11

 Socio-economic constraints to strategic sheep etc...

• Treatment 2: High proportion of cottonseed cake net benefit from supplementing grazing with
(550 g cottonseed cake, 300 g wheat concentrates was 23 Ethiopian Birr (EB) per animal with
straw, 325 g oats grain per head/day). a marginal rate of return of 505% compared to not
feeding concentrates. Veterinary intervention did not
The increased feed during the last five weeks was improve weight gains, therefore the additional costs
designed to supply an additional 73.2 and 82.4 g crude without additional benefits reduced profitability. The
protein (CP) and 1 .006 and 0.797 meal ME for sheep in sheep seemed to prefer sesbania to teff straw. However,
Treatments 1 and 2, respectively. to be an economically viable feed supplement, it would
have to be grown by the farmers at little or no oppor
tunity cost.
Results Trial 3. Average daily weight gains and body
weights for survivors were influenced by diets and initial
body weights (Gautsch, 1992). Pen-fed sheep on
Hidi/Debre Zeit Treatment 1 suffered severe diarrhoea that led to weight
loss. This was attributed to Gossypol toxicity related to
Trials 1 and 2. Sheep on Treatments 1 and 2 in Trial 1 the use of cotton seed and resulted in a negative return
showed similar growth performances (Gautsch, 1992)
of EB -51.85 per sheep. Treatments 2 and 3 resulted in
and commanded the same average sales price. The sheep
on the traditional grazing regime (Treatment 1) therefore positive net returns of about EB 2.30 and 4.70 per sheep,
returned a higher profit margin than the sheep that were respectively. The marginal rates of return for both 2 and
pen-fed (Treatment 2), due to the lower feed costs for 3 were negative. The additional profits from these
Treatment 1. treatments did not justify the additional cost. Diseased
The supplementation of grazing sheep in Trial 2 sheep may have been intentionally brought to the market
(Treatment 2) resulted in higher daily growth rates than and therefore bought by some participating farmers.
Treatments 1 (control) and 3 (pen-fed), by 16 and 7 g Farmers need to be encouraged to purchase sheep within
respectively. Table 1 shows that with or without their own locality, which could help reduce losses due
veterinary intervention, combining grazing with to disease.
concentrate supplementation is potentially more Trials 4 and 5. Mean daily weight gains of 33 g and
profitable than either grazing without concentrates or 70 g were observed during the dry and wet seasons,
pen-feeding of concentrates without grazing. For respectively. However, final body weights within
example, without veterinary intervention, the marginal seasons did not vary by feed treatments, perhaps due to

Table 1. Partial budget analysis for on-farm sheep fattening experiment at Hidi/Debre Zeit in the wet season, Trial 2 (1991)
(Ethiopian Birr/sheep).
Treatment group*
1 2 t
NVT VT NVT VT NVT VT
Revenue from sales (A) 73.50 97.34 154.11 131.26 120.48 114.48
Total variable costs (B) 44.26 71.65 101.50 90.50 90.00 87.00
Sheep purchase 28.50 29.10 29.40 28.65 29.65 28.85
Feed:
- Teff straw 5.76 5.76 - - 10.08 10.08
- Sesbania - - 6.19 6.19 6.19 6.19
- Noug cake - - 1.54 1.54 1.54 1.54
- Wheat bran - - 0.76 0.76 0.76 0.76
- Salt - - 0.76 0.76 0.76 0.76
- Bonemeal - - 0.24 0.24 0.24 0.24
Labour:
- Sheep purchase (per diem) 1.0 1.0 1.0 1.0 1.0 1.0
- Herding 8.0 8.0 8.0 8.0 8.0 8.0
Veterinary (Ranide) - 1.10 - 1.10 - 1.10
Transportation (sheep) 1.10 1.10 1.10 1.10 1.10 1.10
Net return (A-B) 29.24 25.69 52.61 40.76 30.48 27.48
Net return over control - - 23.37 15.07 1.24 1.79
NVT = no veterinary intervention
VT = veterinary intervention

12 Small Ruminant Research Network Workshop

B.I. Shapiro, M.A. Mohamed Saleem and L. Reynolds

the short fattening period (70 days). During the wet daily weight gain of 45 g for sheep on Treatment 2 was
season, grazing plus concentrates resulted in a 13% rise three times higher than for sheep on Treatment 1 . Health
in additional income per animal overgrazing alone. The care affected growth rates and feed conversion ratio
marginal rate of return was 28%. However, the marginal (feed intake/body weight gain) and hence feed costs per
net benefit was only EB 1.80 per head. During the dry kg live weight gain. All pen-feeding treatments resulted
season, grazing was the only viable option. Of the two in losses. Local market prices were high for oats straw,
feeding alternatives to grazing, only grazing plus oats grain, horse beans and rough peas. Thus the
concentrates appears to have more potential than opportunity values of oat straw as feed for cattle and for
grazing plus forages. Forage feeding was not found to horse beans, oats grain and rough peas as feed for
be economically viable in either period due to the humans are quite high.
income lost by growing forage instead of food crops. There was a significant difference in price for
different weight categories of animals. The average sales
Concluding remarks — Hidi/Debre Zeit price/kg live weight for animals on Treatment 1 was EB
2.19, while that for sheep on Treatment 2 was EB 3.09.
The economic analysis indicates that grazing plus Sheep on Treatment 2 reached a weight range where
concentrates has more potential than grazing plus consumers were willing to pay higher prices per kg.
forages to achieve sufficient weight gains for profitable
Trial 2. Although the effect of diet on weight was
short-term strategic fattening. Forage feeding was not
found to be viable due to the income lost by growing marginally positive, health care had no effect. The
fattening period was not long enough for sheep to reach
forage instead of food crops. Further testing of
supplementation with sesbania is indicated, therefore it the weight range that attracts higher prices per kg. The
would need to be grown on-farm to be a viable feed. low weights of the sheep when purchased ( 1 8.3 kg mean
Prophylactic health care did not improve weight gains, body weight) were also a contributing factor. Net returns
but added to production costs, thus reducing profits. for all treatments were negative (Gautsch, 1992).
In order to achieve an acceptable return, the Trial 3. Gautsch (1992) indicated that at the end of
following socio-economic factors may have to be taken the trial, animals on both dietary treatments attained a
into consideration when planning future trials: similar mean final body weight of 32.14 kg. Table 2
• sufficient time for fattening should be allowed so presents the partial budgeting results for this trial. The
average sales price per sheep was the same for both diets.
animals can gain sufficient weight to make a profit; Net profits for Treatment 2 without veterinary
• sheep need to be of sufficient size to achieve a intervention were EB 24.06 per head, about EB 9 per
profitable weight by the time of sale, and need to head higher than for Treatment 1. However, with
be purchased when prices are low and sold at times veterinary intervention, the net profits per animal for
of peak demand when prices are high; Treatment 1 were about EB 5 per head higher than
• health care measures for fattening animals over a Treatment 2. Cottonseed cake was cheaper than grain,
short period may not be needed. but the effect on weight gain was similar.

Concluding remarks — Deneba/Inewari

Deneba/Inewari
At Deneba/Inewari, zero-grazing was judged to be
Trial 1. Sheep on Treatment 2 were significantly appropriate. The high price of oats straw and grain was
(P<0.05) heavier than sheep on Treatment 1 after 16, 24 the major reason for negative returns. Similarly to the
and 36 weeks of fattening (Gautsch, 1992). The average trials at Hidi/Debre Zeit, the 10 weeks fattening period

Table 2. Partial budget analysis for on-farm sheep fattening experiment at Deneba/Inewari, Trial 3, 1989 (Ethiopian
Birr/sheep).
Treatment group*
!I 2
NVT VT NVT VT
Revenue from sales (A) 1 10.03 118.36 114.20 110.00
Total variable costs (B) 94.76 95.46 90.14 92.03
- Sheep purchase 50.63 50.73 51.60 51.50
- Labour 6.03 6.03 6.03 6.03
-Feed 33.13 33.66 27.71 29.58
- Barn depreciation 1.72 1.72 1.72 1.72
- Interest rate 3.25 3.32 3.08 3.20
Net return (A-B) 15.27 22.90 24.06 17.97
Change in net return - +7.63 +8.79 +2.70
*NVT = no veterinary intervention
VT = veterinary intervention

Small Ruminant Research Network Workshop 13

 Socio-economic constraints to strategic sheep etc..

proved inadequate for sheep to reach a body weight • the food preferences of farmers and relative market
which would have attracted the premium price. prices of alternative feeds; and
Consumers are willing to pay considerably higher prices • the relative land and labour requirements for crops
per kg live weight for sheep with high body weight. If a
pre-trial socio-economic survey had been done, these and forage production when considering the
factors could have been taken into account. fattening period.

This study showed that analysing socio-economic

factors is critical for achieving satisfactory results.
Conclusions and implications Socio-economists can play an important role in
designing effective strategic sheep fattening methods
Short-term strategic fattening programmes based on
and designing station and on-farm trials.
supplementing grazing are feasible even in areas of land
shortages, provided there is some land available for
grazing. Agro-industrial by-products such as oilseed
cakes and brans are good potential supplements to References
grazing. Urea and suitable forages and forage legumes
also have potential as good feed supplements. Research Asfaw Yimegnuhal, Nigussie Akalework and Jutzi S. 1987.
into integrating forage and food crops is needed Conventional sheep production and fattening in an
(intercropping, mixed cropping, etc.) and should be part intensively cropped Vertisol area of the central
of sheep fattening research. Pulses and grains should be Ethiopian highlands (Debre Zeit). Vertisol Management
avoided where their prices are high because they are part Project, ILCA (International Livestock Centre for
of human diets. Leguminous trees such as sesbania may Africa), Addis Ababa, Ethiopia. 9 pp.
also have potential as a supplement, but further on-farm Boehlije M D and Eidman V R. 1 984. Farm management. John
research is required. Wiley and Sons, New York, USA.
Conventional wisdom states that farmers' incomes
CIMMYT (International Maize and Wheat Improvement
can be increased through strategic feeding using the
Center). 1988. From agronomic data to farmer
zero-grazing pen-feeding method. However, economic
recommendations: An economics training manual.
analysis shows that feed packages developed for this CIMMYT, DF, Mexico.
method are not economically viable. Alternative feeds
tested have not resulted in much higher profitability Epstein H. 1971. The origin ofthe domestic animals ofAfrica.
Africana Publishing Corporation, New York, USA. 573 pp.
compared to traditional methods of strategic feeding.
Oilseed cakes such as noug and cotton do not seem to Fitzhugh H A, Ehui S К and Lahlou-Kassi A. 1992. Research
have any advantage over grains that are also used as strategies for development of animal agriculture. World
human food. Other factors such as availability of inputs Animal Review 72:9-19.
and financial risk have not been taken into consideration Gautsch К D. 1992. Consultancy report for period 1/9/91 -
in this partial budget analysis. Furthermore, the genetic 27/1/92. ILCA (International Livestock Centre for
potential of the sheep breeds to respond to better feeding Africa), Addis Ababa, Ethiopia.
may have to be studied. The low marginal returns Gryseels G and Anderson F M. 1983. Research on farm and
obtained for pen-feeding indicate that alternative livestock productivity in the central Ethiopian
feeding and crop production strategies should be tested highlands: Initial results, 1977-1980. ILCA Research
on-station before on-farm tests. Report 14. ILCA (International Livestock Centre for
Veterinary prophylaxis does not appear to affect Africa), Addis Ababa, Ethiopia. 52 pp.
weight gain under strategic short-term fattening Jahnke H. 1982. Livestock production systems and livestock
programmes. Reproduction is not a factor in fattening development in tropical Africa. Kieler,
for sale at times of peak demand. Wissenschaftsverlag Vauk, Kiel, Germany. 253 pp.
The results show a number of socio-economic de Leeuw P et al. 1992. Towards sustainablefarming systems
factors that can only be addressed at the farm level. in the Ethiopian highlands. Unpublished manuscript,
These also need to be addressed in the design of station ILCA (International Livestock Centre for Africa), Addis
trials. The following technical and socio-economic Ababa, Ethiopia.
factors may have to be considered when planning future Lili Века. 1990. Enquête sur le système d'élevage ovin dans
trials: la région de Debre Berhan (Ethiopie). Mémoire de
• the alternative economic activities of the farm Thèse, IEMVT (Institut d'élevage et de médecine
vétérinaire des pays tropicaux), Maisons Alfort, France.
household need to be considered when choosing
feeds; 46 pp.
McCann J. 1992. A rapid impact assessment of Vertisol
• the effect of times of purchase and sale on
technology. ILCA (International Livestock Centre for
profitability; Africa), Addis Ababa, Ethiopia.
• the factors affecting consumer preferences in Winrock International. 1992. Assessment of animal agri
animals such as desired size, and therefore price, culture in sub-Saharan Africa. Winrock International
in order to achieve maximum income; Institute for Agriculture, Morrilton, Arkansas, USA. 125 pp.

14 Small Ruminant Research Network Workshop

 Socio-economic aspects of goat nutrition in Swaziland
B.H. Ogwang, D. Mavimbela, R. Vilakati and G.Z. Khumalo

University ofSwaziland, P.O. Luyengo, Swaziland

Abstract total exports (CSO, 1987). In 1990 there were 715,741

head of cattle, 416,166 goats, 20,436 sheep and 23,802
Goats play an important role in Swazi custom and pigs (MOAC, 1991). The majority of these goats, cattle
culture. However, their performance is limited by poor and sheep are reared under a traditional system of
nutrition, especially during the dry winter months. management characterised by overstocking, communal
Improving goat nutrition could therefore enhance their grazing and uncontrolled herding. This has led to poor
productivity and utility. This study involved a survey of livestock performance in the form of low calving
goat production systems and an inventory of crop percentage, high young mortality rates and slow growth
by-products in the small-scale farming sector. The study rates. Lebbie and Manzini ( 1 989) reported the following
found cattle to be the dominant class of livestock, characteristics of the indigenous Nguni goats
(averages): age at first kidding 301 ± 55; litter size 1.18
followed by goats, although both contributed minimally
± 0.29; weight at birth 1.9 ± 0.14 kg and a growth rate
to the cash economy. The major factor affecting goat
to three months of 68.2 g/day. This performance is
production was identified as feed shortage during generally poorer than those reported for indigenous
winter. Several crop residues and agro-industrial goats elsewhere (Wilson, 1976; Khombe, 1985).
by-products were identified as potential supplements. Lebbie and Mastapha (1985) mainly attributed the
Their suitability for animal feeding is discussed. sub-standard performance to poor nutrition, especially
in winter when the quantity and quality of the feed from
the natural rangeland are very low. An improvement in
Caractéristiques socio- nutrition through strategic supplementation could
therefore enhance the productivity of the goats.
économiques de l'alimentation
des caprins au Swaziland
Résumé Methodology
Une large place est faite aux caprins dans les This study is the first phase of a research project being
conducted under the African Small Ruminant Research
coutumes et la culture du Swaziland. Toutefois, ces
Network on the use of crop residues and agro-industrial
animaux ont des performances faibles en raison de
by-products by smallholder farmers in Swaziland. It
l'inadéquation de leur alimentation, notamment au involved a survey of goat production systems and an
cours des mois secs de l'hiver. Pour que s'améliorent inventory of crop by-products, as well as the
leur productivité et leur utilité, il faut donc améliorer determination of their suitability for animal feeding.
leur alimentation. Dans le cadre de l 'étude présentée ici, A total of 121 small-scale farmers were
une enquête a été effectuée sur les systèmes d 'élevage interviewed using a semi-structured questionnaire.
caprin, et les sous-produits de récolte disponibles au Farmer distribution over three ecological zones of
niveau de la petite exploitation agricole ont été Swaziland is presented in Table 1.
inventoriés. Il ressort des résultats de cette enquête que
le cheptel national est essentiellement constitué par des
Table 1. Frequency distribution of the farmers interviewed
bovins, puis par des caprins, encore que la contribution across zones.
de chacune de ces deux espèces à l 'économie monétaire
Zone Number of farmers Percentage
soit insignifiante. Le déficitfourrager d'hiver constitue
le principal obstacle la production caprine. Plusieurs Highveld 30 24.8
résidus de récolte et sous-produits agricoles suscep Middleveld 45 37.2
tibles d'être utilisés comme compléments alimentaires Lowveld 40 38.0
ont été identifiés et sont examinés. Total 121 100.0

Introduction
Livestock have traditionally been an important Results and discussion
component of the agricultural industry in Swaziland. In
1986 their production accounted for about 21% of the The population of livestock in the homesteads surveyed
agricultural output, 5% of the GDP and nearly 3% of is presented in Table 2.

Small Ruminant Research Network Workshop 15

В. H. Ogwang, D. Mavimbela, R. Vilakati and G.Z. Khumalo

Table 2. The population of animals in the homesteads Table 4. Factors affecting goat production.
surveyed.
Percent of
Number of animals/ Number Percentage Mean factor farmers
Class of homestead of home- of home- Factor ranking* (N=113)
livestock (x ± SE) steads steads Feed shortage in winter 1.4 83.2
Diseases 1.9 83.2
Cattle 18 11 103 85.1
Labour shortage 2.1 82.3
Sheep 9 4 7 5.8 Land shortage 2.2 78.8
Availability of good bucks 2.3 77.9
Goats 15 9 86 71.1
Accidents 2.3 81.4
Pigs 3 1 30 24.8 Plant poisoning 2.6 77.9
SE = standard error. Feed shortage in summer 2.6 76.1
* Ranking of problems : 1 = serious, 2 = mild, 3 = none.
Table 2 demonstrates the dominance of cattle in the
rural Swazi farming system, followed by goats. Similar Accessibility of the various feed resources to
findings have been reported by de Vletter (1983). The livestock during the winter season is shown in Table 5.
commercial role of livestock at the household level is Natural pastures, maize stover, maize cobs and browse
indicated in Table 3. The main source of income was constitute the bulk of winter feed in most homesteads.
Apart from browse, the quality of these feed resources
from wage employment, followed by crop sales and
is known to be low and practical methods of improving
traditional medicine. Cattle and goats contributed little
their quality are being investigated. Identification and
to the household financially, confirming the general
quality determination of the browse species are also
observation that livestock ownership in Swaziland tends being undertaken. Besides the feed resources shown in
to emphasise numbers rather than productive capacity Table 5, some farmers were found to provide extra feed
and financial value. Goats are kept mainly for home to their livestock. Some of these supplementary feeds
consumption and to meet other socio-cultural needs. are listed in Table 6.
Only rarely do they contribute a useful source of cash.
Nonetheless, a fairly large proportion of the rural Table 5. Accessibility of livestock to various feed resources
community do keep goats (71.1%) and derive financial in winter.
benefits (40.7%) from them (Tables 2 and 3). Number of
homesteads
Table 3. Sources of income for the rural households. Feed resource (N=113) Percentage
Natural grazing 1 12 98.2
Percent of
Mean income households Maize stover 107 94.7
Income source ranking* (N=113) Maize bran 24 22.9
Wages 1.7 75.2 Maize cobs 52 47.3
Sorghum stover 2 2.1
Gross sales of crops 2.0 64.6
Bean haulms 5 5.3
Traditional medicine 2.0 14.2 Pumpkin residues 4 4.2
Handicrafts 2.2 36.3 Sweet potato vines 1 1.1
Sales of cattle 2.7 43.4 Browse 55 56.7
Others 2.7 23.9
Tableó. Agro-industrial by-product and supplement
Brewed beer 2.8 16.8 utilisation by livestock.
Sales of goats 2.9 40.7 By-product or Number of
* Highest income source scored 1 and lowest scored 5. supplement homesteads
utilised (N=110) Percentage
Farmers were asked to identify the major Sugar- cane tops 4 3.6
constraints to goat production. Their responses are Other by-products 4 3.6
shown in Table 4.
Common salt 5 4.6
The most common problems associated with goat
production were feed shortage in winter, followed by Rumevite mix 2 1.8
disease. However, more farmers reported winter feed Molasses 3 2.8
problems with cattle than with goats (55.9 vs 1 3%). This Urea 2 1.8
perception may be true considering that, unlike cattle,
Hay 2 1.8
goats are capable of browsing on shrubs and bushes
which normally remain green during the winter season. None of the above 88 75.5

16 Small Ruminant Research Network Workshop

 Socio-economic aspects of goat nutrition in Swaziland

Table 7. The potential of crop residues and agro-industrial by-products as feed for goats.
By-product
Maize Sugar-cane Pineapple Citrus
stover tops waste pulp

Where produced rural lowveld Malkerns Malkerns

homesteads sugar area factory
estates estates,
farms
When produced April to April to January to April to
May December November July
When available April to April to January to April to
July December December October
Annual quantities 154,800 38,000 24,000 3,000
produced (t)
Exploited no no no yes
commercially?
Alternative ploughed in feed base ploughed in no use
uses otherwise with or possibly
no use molasses burnt silage
Problem as storage removal removal none
livestock and from from except
feed social field and field and transport
factors distribution distribution
Source: Wickstead (1990) and Ogwang (1988).

Table 6 shows that the majority of farmers do not References

give their livestock any agro-industrial by-products as
feed. It also shows the diversity of supplements that are CSO (Central Statistical Office). 1987. Central Statistical
available in Swaziland but which are currently Office, Mbabane, Swaziland.
under-utilised. The potential of some of these feed Khombe C T. 1985. Environmental factors affecting the
resources for use in goat production is summarised in growth and viability of crossbred sheep and goats on
Table 7. range grazing in Zimbabwe. In: Wilson R T and Bourzat
The most critical period for grazing livestock in D (eds), Small ruminants in African agriculture.
Swaziland is between May and September when many Proceedings of a conference held at ILCA, Addis Ababa,
Ethiopia, 30 September-4 October 1985. ILCA
of the crop residues and agro-industrial by-products are
(International Livestock Centre for Africa), Addis
available. However, the practical problems mentioned Ababa, Ethiopia. pp. 46-52.
above need to be overcome and appropriate feeding
strategies devised before farmers can be encouraged to Lebbie S H В and Manzini A T. 1989. The productivity of
indigenous goats under traditional management in
use them.
Swaziland. In: Wilson R T and Azeb Melaku (eds),
African small ruminant research and development.
Proceedings of a conference held at Bamenda,
Cameroon, 18-28 January 1989. ILCA (International
Conclusion Livestock Centre for Africa), Addis Ababa, Ethiopia.
pp. 39-50.
Natural pastures, browse and maize stover are the most Lebbie S H В and Mastapha P R. 1985. Goat production in the
important component of goat diets in Swaziland. Goat Swaziland middleveld. In: Wilson R T and Bourzat D
owners generally invest minimally in livestock (eds), Small ruminants in African agriculture.
production. The animals are largely left to graze freely, Proceedings of a conference held at ILCA, Addis Ababa,
with any supplementation consisting of cheap green Ethiopia, 30 September-4 October 1985. ILCA
feeds and crop residues. However, there is an immense (International Livestock Centre for Africa), Addis
potential for improving goat nutrition and therefore Ababa, Ethiopia. pp. 224-234.
production by using crop by-products such as MOAC (Ministry of Agriculture and Cooperatives). 1991.
sugar-cane tops, pineapple waste and citrus pulp as feed Annual Report. Ministry of Agriculture and
supplements. Cooperatives, Mbabane, Swaziland.

Small Ruminant Research Network Workshop 17

В. H. Ogwang, D. Mavimbela, R. Vilakati and G.Z. Khumalo

Ogwang В H. 1988. Preliminary studies of crop residues as Wickstead M. 1990. Mission report to Swaziland. Livestock
livestock feed on Swazi Nation Land. UNISWA Section, ODNRI (Overseas Development Natural
Research Journal 1:14-21. University of Swaziland, Resources Institute), Chatham, UK.
Swaziland. Wilson R T. 1 976. Studies on the livestock of Southern Darfur,
de Vletter F. 1 983. The Swazi rural homestead. Social Science Sudan. IV. Production traits in goats. Tropical Animal
Research Unit, University of Swaziland, Kwaluseni, Health and Production 8:221-232.
Swaziland.

18 Small Ruminant Research Network Workshop

 The role of the West African Dwarf goat in the economy
of the smallholder arable farmer in the
subhumid zone of Nigeria
O. A. Ikwuegbu, G. Tarawali and R. M. Njwe

ILCA, P.M.B. 2248, Kaduna, Nigeria

Abstract importants dans la socio-économie du système

traditionnel de la petite exploitation.
This study analysed 875 records of West African Dwarf
(WAD) goats owned by 45 farmers and collected over
30 months. In the wet season WAD goats had access to
either fodder banks or natural pasture. After crop Introduction
harvest, animals roamed freely. Litter sizes were 1 .67 ±
0.08 and 1.56 ±0.06, respectively, on the two grazing
FAO (1991) estimated the goat population of Africa in
systems and were affected by parity (P<0.05). Births
1990 to be 171 million, out of which 22 million were
accounted for 87% of all entries while multiple births found in Nigeria. A recent livestock census, however,
accounted for about 68% of all kids. Castrates accounted puts the goat population in Nigeria at 34.5 million (RTM,
for about 60% of all mature males. Between 35 and 39% 1991).
of adult goats were sold for meat and offtake accounted The keeping of goats in any community is related
for between 51 and 53% of all exits. Sufficient cash was to the value attached to their production. Goats are
generated from goat offtake to purchase farm inputs and
mainly kept for meat (Williamson and Payne, 1978).
some household items. Goats play other important roles
Brinkmann and Adu (1977) estimated that goat meat
in the socio-economy of the traditional smallholder
accounted for about 20% of all meat eaten in Nigeria.
farming system.
This is probably an underestimate as goats slaughtered
in rural areas are seldom recorded.
Place de la chèvre naine In addition, goats play an important role in the
welfare of the smallholder arable farmer in Nigeria
d'Afrique occidentale dans through the income generated when sold. There is global
l'économie de la petite concern about alleviating poverty and improving the
welfare of the rural poor. The sale of goats could
exploitation agricole dans la improve the welfare of farmers, particularly women,
zone subhumide du Nigéria most of whom own goats and who, along with children,
are entrusted with the care of small ruminants.
Résumé About 85% of the smallholder farmers in the
Nigerian subhumid zone keep West African Dwarf
Cette étude analyse 875 relevés effectués sur des (WAD) goats (Bayer, 1986). Traditional methods of
chèvres naines d'Afrique occidentale détenues par 45 keeping goats in the wet season are tethering or
exploitants, et recueillis sur une période de 30 mois. Les free-range grazing on enclosed natural pasture. While
caprins séjournaient sur des banquesfourragères ou des this restriction of movement may prevent crop damage,
pâturages naturels durant la saison humide et étaient it induces feed stress. The introduction of herbaceous
laissés en pâture libre après la récolte. La taille des
legumes into enclosed fallow land is attractive to
portées était repectivement de 1,67 ± 0,08 et de 1,56 ±
farmers because it alleviates the feed constraint in the
0,06 pour ces deux systèmes de pâture et subissait
l'influence de la parité (P<0,05). Les naissances wet season. At the same time it fixes nitrogen from the
représentaient 87% de l'ensemble des entrées. 68% des atmosphere, helping to restore or increase soil fertility
chevreaux provenaient de naissances multiples. Les for subsequent cropping. There is, however, little
animaux castrés constituaient 60% des mâles adultes. information on WAD goat offtake in the subhumid zone
De 35 à 39% des animaux adultes étaient vendus comme (SHZ) of Nigeria.
animaux de boucherie. L'écoulement representait 51 à This paper documents a 30-month ILCA case
53% des sorties et procurait le revenu monétaire study of goat flocks in Ganawuri, in the subhumid zone
nécessaire à l'achat d'intrants agricoles et de certains of Nigeria, to assess the contribution of goat offtake to
articles de ménage. Les caprins jouaient d'autres rôles the economy of the smallholder farmer.

Small Ruminant Research Network Workshop 19

О. A. Ikwuegbu, G. Tarawali and R. M. Njwe

Materials and methods Data analysis

A total of 875 entries were made over the 30-month
Site period. The animals were divided into two groups based
on the differing wet-season feed regimes. These were
This study was carried out at Ganawuri, Plateau State of either enclosed herbaceous legume or natural pasture,
Nigeria, located at 9°N and 8°35'E. The average annual which included enclosure or tethering. Frequencies of
rainfall is about 1400 mm, 95% of which falls between exits from and entries to flocks and the reason for these
April and October. Cultivation density ranges from 40 occurrences were analysed in relation to wet season
to 70% (Bayer, 1986). grazing system and sex of the animal. General linear
models (SAS, 1987) were used to analyse litter size,
Animals using parity, birth type and sex as fixed effects.

Flocks of goats from 45 households were tagged and

recorded for 30 months to determine flock population Economic analysis
dynamics. Flock sizes ranged from 4 to 20 animals per
household and belonged to smallholder farmers.
Animals were weighed every two weeks using hanging In order to apportion monetary value to offtake, the
weight of animals sold or slaughtered for meat and the
spring scales. They were vaccinated annually against
peste des petits ruminants (PPR) and routinely treated average weight of a weaner were calculated. The cost in
Nigerian Naira (N) of fertiliser was taken as N 1/kg and
for ecto- and endoparasites. Birth weights, birth type and
labour as N 4/man-hour for 16 hr. The cost of
birth dates were recorded. When this study began in
establishing a small FB was taken as N 80.
August 1989, 18 farmers with 158 goats were included.
Eight had access to a communal fodder bank and 10
relied on natural pasture for wet-season grazing. In 1990
there were 40 farmers. Seventeen used fodder banks, 23
used natural pasture, and they had 324 animals. At the
beginning of August 1991 there were 45 farmers, 19 Results
using fodder banks while 23 used natural pasture. They
had 157 does, 15 bucks, 66 castrates, 57 young males Births accounted for 86.7% of the entries, with purchase
and 100 young females, making a total of 395 animals. and transfers accounting for the rest. Females accounted
There was a concomitant increase in the number of for 56.7% of all entries. Of the pioneer stock, 68% were
fodder banks from one in 1989 to 9 and 1 1 in 1990 and females and about 13% were breeding males (Table 1).
1991, respectively. Where farmers shared fodder Castrates accounted for 60% of the pioneer mature
banks, their animals were treated as single flocks for males.
analysis.
Table 1. Frequency distribution of entries of West African
Dwarf goats to flocks in Ganawuri by wet-season
Housing and feeding grazing system and sex over a 30-month period.
Goats were housed at night in small mud huts with Grazing system
thatched or metal roofs. In the dry season, animals were Fodder bank Natural pasture
allowed to graze freely from natural pasture (NP), which Parameter Cast. Male Female Cast. Male Female
included grasses and browses. In the wet season they Pioneer 21 19 83 37 22 127
were tethered on natural pasture or were allowed to 1 1 4 2 15
Purchase 3
graze freely on enclosed natural pasture. In addition,
Transfer 5 3 14 4 3 15
some flocks grazed freely from small plots or mini
fodder banks (FB) of about 0. 1 ha of herbaceous legume. Gift 1 0 1 0 0 3
After crop harvest, they roamed freely and grazed crop Birth 2 92 108 2 126 161
residues or were fed household wastes. At the beginning Total 30 115 210 45 154 321
of the next cropping season they were again confined. Cast. = castrate.
The period of greatest feed stress was usually August to
October.
Table 2 shows the distribution of exits of WAD
goats under the two wet-season grazing systems.
Pasture establishment Between 35 and 39% of all mature goats were sold for
meat. The NP and FB wet-season grazing systems gave
Stylosanthes pasture was established according to the overall percentage offtakes of 5 1 and 53%, respectively.
method of Otsyina et al (1987). The seed rate used was
12 kg/ha, and 150 kg of single superphosphate (SSP)/ha Overall percentage offtake was defined as per cent sold
was applied. Early-season grazing was used to control for meat + transfer + slaughtered for ill health. Deaths
grass growth. Pasture was top-dressed annually with 1 00 of kids up to one year of age accounted for between 12.8
kg SSP/ha. and 14.5% of all exits.

20 Small Ruminant Research Network Workshop

 The role of the West African Dwarf goat in the etc..

Table 2. Frequency distribution of exits from flocks and Table 3. Least squares means of litter sizes of WAD goats
reproductive performance of West African Dwarf in Ganawuri in the SHZ of Nigeria.
goats under two wet-season grazing systems.
Effect Level Parturition Mean ± SE
Grazing system
Fodder bank Natural pasture Treatment FB 121 1.67 ±0.082
Parameter No. % No. % NP 171 1.56 ± 0.064
Adult
Year 1989 24 1.64± 118
Sold for meat 70 39.1 94 35.1
Transfer 16 8.9 32 11.9 1990 115 1.59 ±0.069
Slaughtered for ill health 9 5.0 11 4.1
1991 154 1.61 ±0.053
Unknown 3 1.7 1 0.4
Parity ***
Gift 14 7.8 31 11.6
Died accident/sickness 7 3.9 11 4.1 1 79 1.31a ±0.075

2 45 1.65b ±0.009
Kids
Died within 1 year 23 12.8 39 14.5 3 68 1.67b± 0.073
Abortion or still birth 10 5.6 8 3.0
4 61 1.74bc± 0.079
Transferred 9 5.0 13 4.8
Slaughtered 9 5.0 10 3.7 >5 40 1.70bc± 0.102
Sold 3 1.7 9 3.4
Season
Lost or stolen 3 1.7 9 3.4
Accidental death 3 1.7 - Dec-Feb 40 1.67 ±0.097
Mar-May 86 1.62 ±0.074
Reproduction
Does 43 73 Jul -Aug 80 1.6 ±0.083
Parturitions 80 126
Sep-Nov 77 1.56 ±0.076
Parturitions/doe 1.86 1.73
Litter size L67 L56 FB = Fodder bank; NP = Natural pasture.
Means with different superscripts within each column are significantly
different (P<0.05).
Legume supplementation significantly (P<0.05)
reduced weight losses in non-pregnant does on FB when
compared with those on NP between August and
October. Viability of kids of up to one year of age lay Discussion and conclusions
between 86 and 89%. Table 3 shows that litter size was
The overall size of a WAD goat flock at any time is a
significantly (P<0.001) affected by parity, with first
function of entries and exits. Offtake was mainly to
kidders having smaller litters. Average litter sizes were secure cash to purchase household needs such as food
1.67 ± 0.08 and 1.56 ± 0.06 on the FB and NP feeding and farm inputs. In addition, goats could be slaughtered
regimes, respectively. Multiple births accounted for for meat. Bayer (1986), using a small data set, reported
about 68% of all kids in the study. Does that remained offtake at 23.4%. This was made up of all transfers out,
in the system throughout the period showed improved sales and slaughters. This study showed the offtake of
productivity on the fodder bank feeding system (1 .86 vs mature animals to account for between 51 and 53% of
1.73 parturitions/doe). Grazing of fodder banks is thus exits.
beneficial in this cropping system. Births accounted for approximately 87% of all
The average weight of animals sold or slaughtered entries. This suggests that flocks are able to maintain
for meat was 22 kg. A goat currently sells for about N their numbers through birth replacements, partly due to
10 per kg liveweight (US$ 1= N 18.5). The average the tendency of WAD goats to produce multiple births.
weight of a weaner at 150 days of age is 8 kg (un Of particular interest is the appreciable number of
animals given out as gifts (10% of exits) in contrast to
published data). This would give an average value of N
entries (0.9%). This suggests that the greater portion of
220 per adult goat and N 80 for each weaner when sold.
animals received as gifts were not added to the flocks
Figures based on numbers of animals sold, given hut were eaten or sold.
in Table 2, show that sales of adults (sold for meat, Pioneer flocks showed a preponderance of
transferred out or slaughtered because of ill health) from castrates over bucks. This suggests that the castrates
the two grazing systems amounted to N 51,040 (232 x were used as the "savings account" which could be
N 220). Kid sales amounted to N 960 (12x N 80). This drawn upon at short notice to meet family needs. Goats
sum was generated by 45 farmers over two years, which which were sold or slaughtered for meat also included
comes to N 580/ household per year. Actual sales from those that were needed for ceremonies such as weddings
13 households, each owning nine animals or more, gave or communal feasts (Bayer, 1986). These ceremonies
an income of N 633/household per annum. are part of the social system and help to sustain the

Small Ruminant Research Network Workshop 21

О. A. Ikwuegbu, G. Tarawali and R. M. Njwe

smallholder farming system. Most goat sales were S and Blench R (eds), Livestock systems research in
completed just before the cropping season. This allowed Nigeria's subhumid zone. Proceedings of the Second
farmers to reduce the labour requirement for tethering ILCA/NAPRI Symposium held in Kaduna, Nigeria, 29
during a period when labour was required for cultivating October - 2 November 1984. ILCA (International
Livestock Centre for Africa), Addis Ababa, Ethiopia. pp.
their crops.
141-166.
The low mortality rate contrasts with the higher
mortality of 40 to 50% reported in the humid zone Brinkmann W L and Adu I F. 1977. The problems of goat
(Reynolds et al, 1988). This may in part be due to the production in the savannah region of Nigeria. NAPRI
small veterinary package provided in this study which (National Animal Production Research Institute)
improved flock survival. (Manuscript). Ahmadu Bello University, Zaria, Nigeria.
Litter size and survival rate were improved among FAO (Food and Agriculture Organization of the United
goats on the fodder bank. The slight increase in litter Nations). 1991. Livestock and Livestock Products.
sizes and birth rate provides evidence of improved Quarterly Bulletin of Statistics 4(3):39. FAO, Rome,
reproductive efficiency for goats on the fodder bank Italy.
grazing regime. These figures also support
Mba A U. 1 979. Meat production in Nigeria. Prospects and
incorporating herbaceous legumes into this cropping
problems. In: Proceedings of the National Conference
system. on Food and Nutrition Policyfor Nigeria in the I980's.
The observed ratio of intact males to does of 1 : 10 Think Tank University of Ibadan, Nigeria. pp. 89-97.
in this study and 1 : 16 in another study in the same zone
(Bayer, 1986) indicates that more breeding males than Otsyina R M, von Kaufmann R R, Mohamed-Saleem M A and
necessary are kept within these flocks. However, the Suleiman H. 1987. Manual on fodder bank
establishment and management. Subhumid Zone
total absence of breeding males in some flocks is a cause
Programme, ILCA (International Livestock Centre for
for concern. The tethering of all animals in the day Africa), Kaduna, Nigeria. 27 pp.
during the wet season means that no breeding will occur
in flocks that do not keep breeding males. Reynolds L, Atta-Krah A N and Francis P A. 1988. A strategy
With money realised from the sale of a goat a for improving goat productivity under village production
smallholder farmer can purchase about 200 kg of systems in the humid tropics. In: Smith О В and
fertiliser for use during the cropping season. The average Bossman H G (eds), Goat production in the humid
tropics. Proceedings of a workshop held at the
farmer in Ganawuri cultivates about 2-3 ha. The
University of Ife, Ile-Ife, Nigeria, 20-24 July 1987.
fertiliser requirement could, therefore, be met by selling Pudoc (Centre for Agricultural Publishing and
three goats. In addition, manure could be collected from Documentation), Wageningen, The Netherlands.
goathouses and used as fertiliser. The fertiliser and pp. 29-37.
labour costs required for establishing an FB could be
further reduced by undersowing a cereal crop with stylo RIM (Resources Inventory and Management Ltd). 1991.
a year before the land is left fallow (Tarawali et al, 1 987). Nigeria National Livestock Survey. Federal Department
of Livestock and Pest Control Services, Abuja, Nigeria.
287 pp.

Acknowledgements SAS Institute Inc. SAS/STAT. 1987. Guide for personal

computers Version 6 Edition. SAS Institute, Cагу, North
The cooperation of smallholder farmers is appreciated. Carolina, USA. 1028 pp.
The partial financial support provided by the National
Livestock Projects Division of the Federal Department Tarawali G, Mohamed-Saleem M A and von Kaufmann R.
of Livestock and Pest Control Services, Abuja, Nigeria, 1987. Comparison of labour requirements for cropping
maize on land previously used for Stylosanthes fodder
is gratefully acknowledged.
banks and on adjacent fallows in the subhumid zone of
Nigeria. ILCA Bulletin 26:36-39. ILCA (International
Livestock Centre for Africa), Addis Ababa, Ethiopia.
References
Williamson G and Payne W J A. 1978. An introduction to
Bayer W. 1986. Traditional small ruminant production in the animal husbandry in the tropics. Longman, London,
subhumid zone of Nigeria. In: von Kaufmann R, Chater UK. pp. 463^183.

22 Small Ruminant Research Network Workshop

 Some socio-economic factors that affect decisions on
small ruminant production by small-scale farmers —
A case study of west Laikipia
M. M. Mucuthi , K. Munei and K. C. Sharma

1. Laikipia Research Programme, P.O. Box 144, Nanyuki, Kenya

2. Department ofAgricultural Economics, University ofNairobi
P.O. Box 30197, Nairobi

Abstract revenus de l'élevage représentent une part importante

du revenu agricole net. Des études antérieures ont
One of the main problems of small-scale farming in the montré que les petits ruminants offraient de nombreux
arid and semi-arid lands (AS AL) of west Laikipia is that avantages par rapport aux gros bétail. La présente étude
of low household incomes. A major contributor to net visait à identifier certains des facteurs socio-
farm income in this area is income from livestock économiques qui déterminent les décisions prises en
production. Previous studies show that small ruminants matière d'élevage ovin et caprin et à recommander les
have many advantages over the large ruminants (Mburu, moyens à mettre en oeuvre pour augmenter les revenus
1988; Ngategize, 1989; Andargachew Kebede, 1990). qu 'on peut en dégager.
The objective of this study was to identify some of the
socio-economic factors that affect decisions on small Un échantillon de 65 petits exploitants a été
ruminant production enterprises and recommend ways sélectionné dans deux des ranches subdivisés situés
of increasing income from them. dans les terres arides et semi-arides de West Laikipia.
A sample of 65 small-scale farmers was selected Il n'y avait pas de corrélation entre l'âge des enquêtés
from two of the subdivided ranches in the ASA lands of et la détention de petits ruminants. D'après les
west Laikipia. We found that the age of the respondents exploitants interrogés, les principaux obstacles à
and the ownership of small ruminants were not l'élevage de petits ruminants sont la pneumonie et la
correlated. Respondents identified pneumonia and péripneumonie contagieuse des caprins, et non pas
Contagious Caprine Pleuropneumonia (CCPP) as major l'absence de pâturage. Il a pourtant été constaté qu'à
constraints on small ruminant production, but not certains endroits de la zone d'étude le couvert herbeux
insufficient grazing. However, we observed that parts of était de médiocre qualité. Par ailleurs, il y avait une
the study area had only fair to poor grass cover. Lastly, corrélation négative entre la disponibilité d'un revenu
we found a negative correlation between the presence of non agricole et les revenus de l'élevage de petits
off-farm income and income from small ruminant ruminants, et il n 'y avaitpas de corrélation entre l 'accès
production, but none between contact with livestock à des services de vulgarisation de l 'élevage et le nombre
extension personnel and the number of respondents d'enquêtés pour lesquels la mortalité des agneaux
reporting small ruminant mortality as a major constraint. constituait un importantfacteur limitant.

Introduction
Facteurs socio-économiques It is estimated that 82% of the population in Kenya live
intervenant dans les décisions in rural areas, mainly as small-scale farmers. In Laikipia,
91% of the population live in rural areas (GOK, 1991).
prises par les petits exploitants From this, it can be deduced that the majority of the
de West Laikipia concernant population in the district are small-scale farmers. This is
confirmed by the fact that by 1987, land buying
l'élevage des petits ruminants — companies in the district had bought 74% of the former
European ranches and subdivided 63% of them (GOK,
étude de cas 1987). In the study area, nearly two-thirds (2/3) of the
small-scale farms have been settled (Table 1). This
Résumé makes small-scale farming in the district significant. It
is for this reason that the Laikipia Research Programme
La faiblesse des revenus des ménages constitue (LRP) has orientated its research towards this type of
l'un des principaux problèmes qui touche la petite farming. Most of the LRP research is focused on small-
exploitation agricole dans les terres arides et scale farming in the ASA lands which make up 80% of
semi-arides de West Laikipia. Dans cette région, les the district.

Small Ruminant Research Network Workshop 23

Martín M. Mucuthi, Kimpei Munei, Kailash С Sharma

Table 1. Information on the human settlement in the study west Laikipia. When using the FSR approach, there is a
area. time-lag in the recognition of the problem and finding
Frequency Per cent the relevant solution. Therefore, the approach is
inherently expensive. Nonetheless, it is consistent with
Number of settled farms 605 61
the development strategies of the Kenya District Focus
Number of unsettled farms 390 39 For Rural Development as it builds on, rather than
Total number of farms 995 100 destroys, the farmers' development techniques. This
study considered the age of the respondent, farm size,
frequency of contact with livestock extension personnel,
These small-scale farmers face many challenges in type and amount of income accruing from off-farm
generating household incomes. One way of helping income and constraints to small ruminant production.
them is to increase the income they receive from small
ruminant production. This study therefore focuses on
some of the socio-economic factors that affect decisions Results and discussion
on small ruminant production in west Laikipia. Some of
these factors are more economic than sociological, while
others are more sociological than economic. It is Age of respondents
difficult to separate the two, but this study focuses
primarily on the sociological factors. A discussion of all Only in 20% of the cases was the male head of the
socio-economic factors can be found in Mucuthi (1992). household involved in herding. Most of the herding was
done either by the wives (35%), children (25%), or hired
labour (20%).
Materials and methods Neither ownership nor herd size of small ruminants
was associated with age of respondent (P>0.05). Age
In 1 990 there were 28 former European ranches and five may not be a crucial factor in the adoption or non-
government settlement schemes in west Laikipia that adoption of projects on small ruminant production.
were in various stages of settlement and/or subdivision.
In eight, subdivision was completed and therefore maps
were available. For sampling convenience, we selected Farm sizes and small ruminant production
two ranches from this group. Each selected ranch was
subdivided into four equal subclusters and one In the study area, the mean farm size was 5.3 ha, with
subcluster was chosen from each ranch as the starting most of the respondents owning between 3.7 and 7.2 ha
point. The farms or sampling units were not sampled (Table 2). Thirty-six per cent of the land was used for
randomly within the subclusters but systematically from crop land while 4% was used as homesteads (Table 3);
a randomly selected corner of the subcluster. the rest of the land (60%) was available for grazing.
Using this method, 65 respondents who were Grazing land size, however, tends to be determined
defined as small-scale farmers were interviewed. They by the area reserved for crop production rather than vice
represented 9% of the potential respondents in each versa. This explains why land size per se was neither
ranch. Herd size (50 small ruminants and 10 cattle) significantly correlated with the herd size nor with the
instead of land size was used to classify small-scale presence or absence of the small ruminant enterprises
farmers, though recognising that land size is the standard (P<0.05). It also confirms our earlier assumption that
measure of farm classification. Herd size was used land size can be a deceptive criterion for identifying
because Köhler (1987) found that in west Laikipia most small-scale farmers for livestock-related studies and
small-scale farmers owned no more than 50 head of projects, especially where the farmers are employing
small ruminants and eight head of cattle. No prior extensive methods of livestock production. It was
appointments were made for visits, but efforts were observed that when small-scale farmers had extra
made to interview the husband or the wife, or in default, income to invest, they preferred to invest in livestock
the defacto head of the farm. Otherwise, the enumerator rather than in crop production (Mucuthi, 1992).
moved to the next farm. Each respondent was asked
general questions relating to small ruminant production. Table 2. Land ownership by farm sizes.
They were also asked specific questions relating to the
factors that affected decisions on small ruminant Farm in hectares Percentage of respondents
production.
Under 1.8 3
The Farming Systems Research (FSR) approach
was used in determining the factors considered to affect 1.8-3.6 9
decisions on small ruminant production. This approach 3.7^.8 40
was used because, among other attributes, it is
characterised by the "bottom-up" rather than the 4.9-7.2 37
traditional "top-down" approach to farmers' problems. Over 7.2 11
It also recognises the need to have research tailored to a
specific area, in this case the small-scale farming area of Mean 5.3 100

24 Small Ruminant Research Network Workshop

 Some socio-economic factors that affect decisions etc...

Table 3. Mean land sizes for various land uses in the study A comparative analysis of the levels of income from
area. small ruminant production for respondents with
Land size Per cent of off-farm income and those without, did not indicate any
Type of land use (ha) total significant difference (P>0.05) in the mean small
Mean grazing land 3.2 60 ruminant herd sizes. Therefore, any significant
Mean crop land 1.9 36 difference in mean incomes from small ruminant
production was from the number of small ruminants sold
Mean size of homestead 0.2 4
by the respondents.
Mean land holding 5.3 100 Table 4 shows that the mean income from small
ruminant production for respondents without off-farm
income was significantly higher (P<0.05) than for those
Availability of grazing for livestock with off-farm income. Though the sample sizes were
small ( 1 9 respondents) and the standard deviations high,
Most respondents (84%) utilised their own farms and the large variations across observations do occur in
unoccupied farms for grazing. They kept as many cross-section studies (Pindyck, 1981).
animals as they were able to, because there were no
grazing restrictions. The stocking rate, which was Table 4. Comparison of differences of mean income from
calculated by dividing the number of ha of the available small ruminant production, for respondents with
grazing land by the Tropical Livestock Unit (TLU=250 and without off-farm income.
kg) was 2.8 ha per TLU. This was less than the carrying Without off- With off-
capacity recommended to graziers of 3-5 ha per TLU farm income farm income
(Pratt and Gwynne, 1977). Mean income 1326a 523b
When the stocking rate was calculated with the (Kenyan shilling)
assumption that only one's own farm was available for
Standard deviation 1344 942
grazing, the stocking rate was even lower at 1 .4 ha/TLU.
Without feed supplementation, the available grazing Sample size 19 19
land in the study area could support only one cow or 10 Values within the same row with different superscripts are
small ruminants per respondent. Unfortunately, only significantly different (P<0.05).
27% of the respondents gave any form of supplementary
feed (maize stover and crop residue) to small ruminants. The high mean income from small ruminant
Available fodder was fed mainly to cattle. production for respondents without off-farm income
indicates the significant contribution of small ruminants
to net farm income. Indeed, small-scale farmers in the
ASA lands considered small ruminant production and
Effect of small-scale farmers' off-farm income as the most important sources of
off-farm occupations on small household income for development as well as daily
ruminant production expenses. Small ruminant production contributed 50 and
54% to income used for daily and development
Respondents engaged in off-farm occupations because expenses, respectively (Mucuthi, 1992).
food crop yields in the ASA lands are unreliable. Even
in good years, maize crop yields (a maximum of 20 bags
of 90 kg) are not enough to guarantee the respondents
sufficient food and income for one year. Effect of livestock extension services
When off-farm work was the main occupation, the
head of the household was engaged in permanent
on small ruminant production
off-farm work. However, we could not rule out the fact
The five Ministry of Livestock Development extension
that respondents engaged in farming to supplement personnel were either livestock production personnel
off-farm income. We found that 60% of the respondents (LPP), who advised the respondents on animal
depended solely on net farm income while 40% were husbandry, or veterinary personnel (VP), who were in
also engaged in off-farm work. charge of livestock treatment. While the LPP were
The proportion of respondents dependent on net expected to visit farmers on a regular basis, the VP were
farm income was higher than that reported by Köhler expected to stay at predetermined places where farmers
(1987). He found that in central Laikipia, only 33% of could reach them. However, advisory and treatment
the respondents depended on net farm income. The work was sometimes done by the same people.
lower dependence on off-farm work in the study area Table 5 shows the number of visits by the LPP as
compared to central Laikipia was because the reported by the 50 respondents who had small
respondents in west Laikipia were farther away from the ruminants. Of the respondents, 74% were never visited
main employment areas. at all while 20% were visited whenever the need arose.
Of the 43 respondents, 38 with positive household For the VP (Table 6), the majority of the respondents
income had small ruminants; only data from these were visited whenever the need arose and the rest were
respondents were used in the test of difference of means. never visited.

Small Ruminant Research Network Workshop 25

Martin M. Mucuthi, Kimpei Munei, Kailash C. Sharma

Table 5. Reported visits by the livestock production Insufficient grazing was not seen as a major
personnel. constraint by the respondents. Nevertheless it was
Response Frequency Per cent apparent, especially in the unoccupied farms. In
assessing the availability of grazing, grass cover was
Never visited 37 74
determined by the amount of bare soil observed by the
Whenever need arises 10 20 enumerator at the time of the interview. Less than 10%
Fortnightly 2 4 of bare soil cover was considered to be good grass cover,
11-30% was medium and 31-50% was bad. It was
Weekly 1 2
estimated that 59% of the occupied farms had good grass
Respondents interviewed 50 100 cover, 30% had medium grass cover and 11% had bad
grass cover. The unoccupied farms had medium soil
Table 6. Reported visits of the veterinary personnel. cover or worse.
To achieve the optimum carrying capacity, one
Response Frequency Per cent
needs to do any or all of these things. First, livestock
Whenever need arises 39 78 numbers must be reduced. However, where grazing is
Never visited 11 22 communal, as in the study area, this method has been
unpopular. Second, livestock feed can be improved and
Respondents interviewed 50 100
increased. Although this is easier to achieve than the first
option, lack of enough grazing for small ruminants was
It appears that the problem of infrequent visits by not seen as a significant constraint by the respondents.
the LPP was not due to lack of personnel, but due to the Feed availability could be increased by planting fodder
disproportionate time given by the extension personnel trees like Leucaena and Sesbania along the fences and
to the treatment service. Where the same extension on terraces. Third, the productivity per animal can be
personnel are expected to perform both advisory and enhanced through upgrading and improving
treatment work, it is obvious that the advisory work will management. This is the easiest of the three options to
be compromised because the personnel cannot visit implement. It is not expensive and would be agreeable
farms and wait for the farmers' visits at the same time. to the respondents.
Visits by Ministry of Livestock Development
extension personnel are expected to have an impact on
the small-scale farmers' management practices. It is
expected that through contact with extension personnel, Conclusion
small-scale farmers will improve livestock husbandry
practices. Thus it was surprising to find no correlation a) In west Laikipia the age of the respondent did not
between the frequency of visits by the LPP or the VP affect participation in small ruminant production
and the small ruminant mortality rate. enterprises.
b) Land size was not correlated with small ruminant
herd size or the ownership of small ruminants.
Land size can therefore be a deceptive criterion in
Constraints to small ruminant classifying small-scale farmers for livestock-
production related studies and projects.
Respondents were asked to indicate the most important c) There is a significant contribution (P<0.05) of
constraints on small ruminant production and to income from small ruminant production to net farm
prioritise them. They indicated that mortality through income for respondents without off-farm income.
disease (53%) and lack of enough water for livestock d) Respondents were satisfied with the services of the
(21%) were the most important constraints. This
VP but not with those of the LPP.
confirms what had been reported in other studies, that
diseases are among the most important constraints to e) Grazing land was determined by the area reserved
small ruminant production (GOK, 1990; Herren, 1990; for crop production but not vice versa.
Mucuthi, 1992). Respondents also perceived capital, rather than
The LPP and the VP are supposed to advise land, as the limiting factor in livestock production.
small-scale farmers on the best animal husbandry
practices. It was expected that more frequent visits by
these personnel would lower the significance of diseases
as a constraint. However, an insignificant correlation Recommendations
(P<0.05) was found between the frequency of their visits
a) For livestock-related studies and projects, land size
(both LPP and VP) and the number of respondents who
should not be used as a criterion for classifying
reported that diseases were the major constraint. This is
small-scale farmers.
because most respondents were satisfied with the
veterinary treatment but not the small ruminant b) Action needs to be taken to improve the optimum
production advisory services. livestock carrying capacity of the study area.

26 Small Ruminant Research Network Workshop

 Some socio-economic factors that affect decisions etc..

c) The question of poor extension services for small GOK (Government of Kenya). 1991. Economic Survey.
ruminant producers needs to be addressed. Some Government Printer, Nairobi, Kenya.
options would be to separate the roles of the Herren U J. 1990. Socio-economic stratification and small
personnel and/or charge a fee for their services. stock production in Mukogodo Division, Kenya.
Research in Economic Anthropology 12:1 1 1-148.
Köhler T. 1987. Land use in transition: Aspects and problems
ofsmall-scalefarming in new environment, the example
Acknowledgements of Laikipia District, Geographica Bernensia Vol A 5,
Bern, Switerland.
This research was carried out with funds provided by
Laikipia Research Programme and in collaboration with Mburu J W S. 1988. Reflections of Livestock Production
Nairobi University and the Ministry of Livestock Strategy Under District Focus for Rural Development.
Ministry of Livestock Development, Nairobi, Kenya.
Development. The authors are grateful to Dr Urs Herren
of Laikipia Research Programme for his comments, and Ngategize P K. 1989. Constraint identification and analysis in
his help in editing this paper. African small ruminant systems. In: Wilson R T and
Azeb Melaku (eds), African small ruminant research
and development. Proceedings of a conference held at
Bamenda, Cameroon, 18-25 January 1989. ILCA
Bibliography (International Livestock Centre for Africa), Addis
Ababa, Ethiopia. pp. 7-22.
Andargachew Kebede. 1990. Sheep marketing in the central Pindyck R and Daniel Rubinfeld. 1981. Econometric models
highlands of Ethiopia. Unpublished MSc thesis, and economic forecasts. Second edition. McGraw-Hill
Alemaya University of Agriculture, Alemaya, Ethiopia. Book Company, New York, USA.
117 pp. Pratt D J and Gwynne M D (eds), 1977. Rangeland
GOK (Government of Kenya). 1987. Subdivision of Group management and ecology in East Africa. Hodder and
Ranches in Laikipia District. District Livestock Stoughton, London, UK. 310 pp.
Development Office, Nanyuki, Kenya.
Mucuthi M M. 1992. Some economic aspects of small
GOK (Government of Kenya). 1990. Rumuruti Division ruminantproduction by small-scalefarmers in theASAL
Annual Reports, Rumuruti Divisional Veterinary Office — A case study of west Laikipia. Forthcoming MSc
(Laikipia District), Rumuruti, Kenya. thesis, Nairobi University, Nairobi, Kenya.

Small Ruminant Research Network Workshop 27

 Priority setting for small ruminant research
L. Reynolds

International Livestock Centrefor Africa

P.O. Box 80147, Mombasa, Kenya

Abstract been made in African agriculture. Donors, be they

national or international are more than ever before
The limited resources available to national agriculture concerned about impact. Technology developed from
research and extension systems in Africa will allow only agricultural research has not always been adopted in
top priority areas to be funded. Priority identification SSA over the past 20 years. Promises of technological
must take into account farmers' objectives and those of breakthroughs and of adoptable packages have not been
national planners. A farming systems approach is kept. Scepticism on the part of paymasters must be
essential to determine farmers' objectives and overcome, realistic objectives set and adoptable
constraints and to identify appropriate solutions. Such technology produced if financial support is to be
information is an essential prerequisite in the sustained, let alone increased. Sustainability as an issue
formulation of research programmes, which must be has been accepted as an objective for biological
backed up by a thorough review of the literature. Small research, but less attention has been paid to the
sustainablity of the research institutions and the
ruminants may or may not be a national priority.
accompanying infrastructural support required for
research and extension.
Populations in SSA are expected to expand
Définir les axes prioritaires de threefold by 2025 from the present level of 500 million.
recherche sur les petits Over the same period, urbanisation is expected to rise
from 31 to 57% of the population (Winrock
ruminants International, 1992). Urban consumers tend to require
more processed foods, high value cereals, livestock
Résumé products and vegetables. To meet this demand
L'insuffisance des ressources dont ils disposent agricultural production must increase and become more
oblige les services nationaux de recherche et de market oriented to supply urban areas. Market forces can
vulgarisation agricoles en Afrique de ne financer que stimulate production, as seen in the variations in cereal
les domaines revêtant une priorité absolue. Pour sales in Zimbabwe in response to changes in the level of
identifier ces priorités, les objectifs des petits guaranteed prices, but market forces alone will not be
exploitants et des planificateurs nationaux doivent être enough. More land will be cropped to supply the
pris en considération. L'adoption d'une approche requirements of higher populations, but uncultivated
systèmes agraires est essentielle si l'on veut déterminer productive land is already limited. Unit area
les objectifs des paysans et les contraintes auxquelles ils productivity must therefore increase. Storage,
se heurtent, et identifier des solutions appropriées. Ce processing and transport must become more efficient.
type d'information est un préalable indispensable à la Research, in collaboration with producers and advisory
formulation de programmes de recherche, qui devront services, has a critical role to play in these
être appuyés par une étude bibliographique developments.
approfondie. Les petits ruminants peuvent ou non Nations must decide on their priorities and ensure
figurer parmi les priorités nationales. that they are providing enough funds for essential
research. Donor-driven research programmes are
affected by the availability of funds. They may not
Introduction conform to national priorities, and may divert the limited
number of trained personnel away from more important
Africa is experiencing political and economic changes,
work. In most countries in SSA, very few activities can
and national research and extension organisations must
be funded solely from national resources. External funds
adapt to survive in the face of these changes. Across
sub-Saharan Africa (SSA) the level of funding from should therefore be directed towards supporting the
governments for agricultural research has fallen below priority research areas. Only after those areas have been
the level needed for active and productive use of trained adequately covered should other work be considered. If
manpower. While the present level of funding is a and when external support declines, high priority areas
reflection of the economic realities for the Governments will be less affected because core funding will already
of SSA, it also accounts for the limited progress that has have been assured from the national exchequer. This is

Small Ruminant Research Network Workshop 29

L. Reynolds

a risk-avoidance strategy against variations in donor and goat meat is the least important livestock item, after
countries' economic and political climates, similar to milk and beef.
that adopted by farmers in minimising annual Table 2 shows the output of meat and milk from
differences in the biophysical climate. However, this ruminants in the regions of SSA. Of the small ruminants
strategy will be effective only if focussed on the highest mutton is more important in East Africa, while goat meat
priorities. predominates in West Africa. From the limited data
The ratio between staffing and operational costs is available, small ruminant milk appears more important
9: 1 in many countries. This ratio is unsustainable and in East Africa than in West Africa, but much of the
must therefore be narrowed. A ratio nearer 3:2 would production is consumed by young stock.
produce sufficient operational funding to allow useful
research to continue. When operational funds are in Table 1. Gross value of annual production of the twenty
short supply staff still receive salaries, but cannot carry most important commodities in SSA, 1983-85
out any useful function. Unpalatable though it may (US$ million).
seem, closing research stations and reducing staff levels Fuel (non-coniferous) 12489
to allow those remaining to operate effectively is
essential, given that additional funding is unlikely Banana/plantain 8518
(Eicher, 1989).
Sawlogs (non-coniferous) 4844
Sweet potato 4406
Priority setting: the key issue Cassava 4377
Agricultural research has largely focussed on biological
Coffee 4008
problems, and there is a shortage of economists and
social scientists in NARS (National Agricultural Milk 2894
Research Systems). However, policy and
socio-economic considerations must be considered in Yam 2624
order to increase the probability of developing adoptable
Other industrial wood 2361
technology. Technology must be appropriate to the
prevailing policies and economic conditions of the Beef/buffalo meat 2548
country, or to those expected when the research matures.
A valid research objective would be to model the effect Cocoa 2361
of alternative policies on agricultural production, so that
Maize 2064
policy makers can make informed choices. It follows
that adaptive research, which must include on-farm Charcoal 1721
studies in collaboration with development/extension
agencies and farmers, must try to anticipate the national Sheep/goat meat 1648
environment at the time when the research comes to
Groundnut 1425
fruition. While it can be argued that "on-the-shelf"
technologies and production systems are a valid output Cotton 1342
from research, they may not fit the present or projected
economic and political environment. They may Sorghum 1648
therefore be a luxury when current needs are so pressing.
Millet 1147
Rice 1092
The overall research agenda Inland fish 1044
A research organisation must decide where to focus its
work and how to allocate resources between A summary ofmethods for priority setting has been
commodities, species and research areas, including presented by Norton and Pardy (1987), and Contant and
policy to fulfill its mandate. It must decide how Bottomley (1988). Checklists and scoring systems are
important small ruminants are, and where and for which the most practical methods for setting priorities in
section of society research could have the most impact.
agricultural research. However, there are practical
Numbers of animals, distribution and values of
difficulties in attempting to rank programmes dealing
production are required to quantify importance. The first
stage in deciding on the allocation of resources is to with aspects of the natural resource base (e.g. soil
consider the following elements: the estimated value of fertility) alongside those dealing with commodities (e.g.
production, adjusted by modifying factors to allow for small ruminant meat). Benefits from commodity
national research objectives; the probability of research research programmes can be expressed directly in
success; efficiency in the use of research resources; and monetary terms. Benefits from research on factors such
distribution of impact. as soil can only be estimated from assumptions
Table 1 shows the value of production of the most regarding increased productivity over a range of
important commodities in sub-Saharan Africa. Sheep commodities.

30 Small Ruminant Research Network Workshop

 Priority setting for small ruminant research

Table 2. Output ('000 mt) of meat and milk in the regions of SSA, 1987.

Meat Milk

Beef Sheep Goat Cow Sheep* Goat*

West Africa 587 135 240 1057 139 288
Central Africa 216 17 30 308 3 21
East Africa 860 265 199 4858 1052 1414
Southern Africa 349 10 26 630 NA 12
•Data not available from some countries in each region.

ISNAR (International Service for National of these issues toward the fulfillment of the goals of the
Agricultural Research) have assisted NARS in a number institution. Efficiency criteria should include: import
of countries to prioritise their research programmes. ance of the problem; adoption rate; probability of
These countries include the Dominican Republic, research success; environmental impact; research costs;
Ecuador, Uruguay, Peru, Gambia, Zimbabwe and comparative advantage; and compatability with national
Kenya (Cessay et al, 1989; Norton and Pardy, 1987; goals. Equity criteria will cover: farm scale; producer/
Norton, 1991). consumer benefits; gender; and agro-ecological zones.
Security issues would include production stability and
food security.
Choosing priorities Consideration must also be given to whether
research leads to an increase in the use of abundant
There are so many scientifically valid research activities resources and a saving of scarce resources, and also the
that some formal way is required to identify policies for number and severity of research problems. Other areas
technological and socio-economic research that will are non-duplication with transferable research from
have real impact. Selected research activities should outside, and the current emphases in the research
stimulate action by policy makers or development and programme. The process is consultative and iterative.
extension colleagues, and be capable of adoption by The criteria and rankings are produced in discussions
farmers. with policy makers, researchers, development/extension
Identifying problems worthy of research can be workers and producers.
started by researchers or by farmers. An example of the The priorities of other countries in the region and
first route is goal-oriented programme planning of IARCs (International Agricultural Research Centres)
(GOPPing). GOPPing is a component of the ISNAR such as ILCA, and their compatibility with the priorities
approach to long-term programme design as described of the NARS should also be considered. If, for example,
by Collion and Kissi (1991). It was used by ILCA,
we consider that priority should be given to improving
assisted by national scientists, when it produced its long-
the nutritional value of crop residues, we would need to
term strategy document (ILCA, 1987). A multi-
disciplinary group, including representatives from collaborate with crop breeding institutions and ensure
different research, development and extension groups that their priorities reflected our concerns. There is a vast
participate in the exercise. Farmers are usually not quantity of published information on most subjects,
included because it is assumed that the group already therefore any research process must start with a
knows the problems that limit on-farm production, and literature search. All involved in research planning
hence can identify research routes to improve output. should be aware of the results of past research and
The second approach involves researchers meeting development projects, so that we build on success and
farmers and extension staff to identify problems and avoid repeating the same mistakes. Researchers in
propose solutions (Chambers et al, 1989). Farmers are Africa often have only limited access to adequate library
the key players in this approach as they know the facilities and are unaware of published and ongoing
problems and will have to work with any proposed work inside their own country, let alone outside. The
solutions. ILCA library provides a literature search service for
However, merely identifying a problem says livestock researchers.
nothing about its importance in relation to other Most research activities undertaken by NARS in
problems or constraints. As indicated earlier, problems SSA will be applied or adaptive in nature, as that is
are many but funds are limited, therefore priorities must where their comparative advantage lies. Traditionally,
be set. Priorities across agro-ecological zones need to be researchers have set the agenda, but, given the lack of
considered, as these will affect the balance of work adoptable output, we need to explore other ways of
between regions. If a weighting method is used, issues planning research to ensure applicable results. Output
of efficiency, equity and security need to be considered. from applied and adaptive research that cannot be
Criteria are required to measure the contribution of each adopted by farmers is valueless. Strategic research that

Small Ruminant Research Network Workshop 31

L. Reynolds

is not based on problems identified on-farm and which been distributed to farms, but only 10% of the does are
does not have a target group of farmers in mind as milked. All the milk is used for household consumption
beneficiaries should not be started. Sumberg and Okali (Mbabu et al, 1 99 1 ). The main attraction of the crossbred
(1989) go so far as to propose that adoption by farmers animals to Kenyan farmers is reported to be their large
is sufficient validation of a technology, even if we are body frame. At both locations surplus male kids are
unable to identify or quantify the effects of the purchased by the project. Time will tell which of the two
technology. approaches will be sustainable when external funding
ceases, but so far the market-led scheme in Burundi has
had more impact. The scheme in Kenya provides no
clear-cut economic advantage to farmers that would
Principles to follow encourage their continued interest when external project
funding ends and staff are withdrawn.
All applied and adaptive research must have a strong Research into improved feeding must be linked to
on-farm base. For the humid and subhumid zones this a strong market demand for the animal product, and its
means that the low priority given to small ruminants by profitability demonstrated by ex-ante economic
farmers must be taken into account. In mixed analysis. Research into forage production has only
crop/livestock farming systems where crops are the
produced adoptable technology when linked to milk
more important, small ruminant research will be only
one component of multidisciplinary on-farm studies. production. Specialised short-term fattening schemes,
Monodisciplinary small ruminant research on-station designed to produce animals for festival peaks, may be
can be justified only when the need has arisen from viable when the feed sources are crop residues or
on-farm observations and the results can be taken back agro-industrial by-products. The economic viability of
for application on-farm. This is particularly relevant to meat production based on planted pasture is poor, as
universities where suitable component work for student shown by Doppler (1980) for beef cattle in Togo.
projects can be identified from on-farm studies. Over most of Africa, small ruminants are kept as
As small ruminants are of secondary importance to free roaming scavengers. Many flocks in the humid zone
small-scale mixed farmers, they will only become of West Africa have no adult male, relying on those in
prominent if justified by market demand and price. other flocks in the village. The more active males will
Projects in which small ruminant production is the sire the most offspring, from which replacement sires
central aspect have little chance of success if imposed will be drawn. At village level there is therefore the
on farmers unless their objectives have been danger of inbreeding. In addition the most rapidly
incorporated into the design. With small farmers, risk growing males are sold or slaughtered first, so the few
avoidance usually takes precedence over increased males that reach breeding status have been selected for
production. slow growth (Reynolds and Adediran, 1993). The
Little progress has been made in humid or selection, distribution and use of superior sires may be
subhumid Africa by research or development projects a valid research and development objective. However,
for increasing small ruminant meat production on maintaining improved genes in village herds will require
small-scale farms. In general, the disease risks infrastructural support superior to that which present
associated with raising youngstock are too high to attract governmental organisations have been able to maintain.
farmers. Youngstock mortality rates of 40% are Farmer demand for improved breeding stock will be
common. Thus for increased productivity the first generated if there is a market demand for the eventual
priority should be to reduce the death rate. product, as shown by smallholder dairy schemes with
Unfortunately, small farmers do not have ready access cattle in Kenya (Walshe et al, 1991).
to veterinary services and governments cannot afford the Much of the day to day management of small
cost of expanding the service. Improved management, ruminant herds is in the hands of women, even when the
including better health through improved feeding, can flock owner is male. Research is needed to determine
reduce susceptibility to disease and improve survival the flexibility that women have to alter flock
(Reynolds and Adediran, 1988; Reynolds and management, use different feedstuffs and to market
Ekwuruke, 1988). The integration of small ruminants animals at different times or in different ways. Will the
into alley farming is one such improved management woman benefit directly as a result of the changes, or will
option (Kang and Reynolds, 1989; Kang et al, 1990). the benefits accrue to the owner? The contribution to
Market importance can be illustrated by household income from small ruminants is less than 5%
comparing two dairy goat schemes, both started about of farm income, and probably half that figure for
10 years ago in East Africa, based on crossbred household income. However, their role as an investment
exotic-indigenous animals. In Burundi around 10,000 for savings may account for the wide ownership pattern
crossbreds have been accepted by farmers and the milk (ILCA, 1979). Improved animal productivity is
produced is purchased by a factory for processing into equivalent to a higher rate of interest on the savings
cheese, which is then sold to urban consumers (Rey, account.
personal communication). Regular income from the sale Small-scale farming is a high risk business, and
of milk is the key factor encouraging adoption. In with few capital reserves to fall back on in bad years
Western Kenya a much smaller number of animals have farmers minimise risks by undertaking a wide range of

32 Small Ruminant Research Network Workshop

 Priority setting for small ruminant research

activities. Lack of appreciation of this aspect of the information and policy implications. IFPRI Working
strategy of small farmers is one reason why the Papers on Commercialisation of Agriculture and
traditional biological research approach has failed. Nutrition 5. IFPRI (International Food Policy Research
Increasingly, farmers are looking to off-farm activities Institute), Washington, DC, USA. 206 pp.
and involvement in the monetary economy as risk Cessay S, Gilbert E, Mills B, Rowe J and Norton G W. 1989.
avoidance strategies, in addition to or in place of Analysis of agricultural research priorities in the
established farming practices (Downing, 1989). In much Gambia. Project Paper 16. ACIAR/ISNAR (Australian
of Africa small farm households derive over half of their Centre for International Agricultural Research/
International Service for National Agricultural
income from off-farm sources such as remittances (von
Research), The Hague, Netherlands.
Braun and Pandya-Lorch, 1991). The positive side of
this picture is that the extended family can generate Chambers R, Pacey A and Thrupp L A (eds). 1989. Farmer
funds off-farm for investment in farming. The negative first. Intermediate Technology Publications, London,
UK. 218 pp.
side is that the more innovative members of the
household move away from the farm unless the returns Collion M H and Kissi A. 1991. An approach to long-term
from farming appear attractive in comparison to program design. Working Paper 37. ISNAR
employment prospects in urban areas. (International Service for National Agricultural
Research), The Hague, Netherlands. 53 pp.
It is important that all small ruminant researchers,
biological, social and economic, combine to provide Contant R В and Bottomley A. 1988. Priority setting in
data and analyses to policy makers so that the agricultural research. Working Paper 10. ISNAR
(International Service for National Agricultural
consequences of alternative courses of action can be
Research), The Hague, Netherlands. 19 pp.
predicted. Modelling is the way this objective can be
achieved. Piecemeal research will not provide the depth Doppler W. 1980. The economics ofpasture improvement and
and breadth needed for modellers to construct and test beef production in semi-humid West Africa. GTZ
(German Agency for Technical Cooperation), Eschborn,
their models. Research and development groups must
Germany. 195 pp.
cooperate within and across both national and research
borders to prevent unnecessary duplication of effort. Downing T E. 1989. Climatic variability, food security and
smallholder agriculturists in six districts of central and
eastern Kenya. Disssertation, Clark University. 276 pp.
In: Dissertation Abstracts International. 1989, 49: 2347.
Conclusion Eicher C K. 1989. Sustainable institutions for African
In the past, livestock research has been ineffective in agricultural development. Working Paper 19. ISNAR
producing technologies that can be adopted by (International Service for National Agricultural
Research), The Hague, Netherlands. 38 pp.
small-scale farmers. NARS have insufficient resources
to stretch further than top priority areas. In any one ILCA (International Livestock Centre for Africa). 1 979. Small
country these may or may not include small ruminants. ruminant production in the humid tropics. ILCA
A top-down allocation of resources for small ruminant Systems Study 3. ILCA, Addis Ababa, Ethiopia. 122 pp.
research is not effective; appropriate research can only ILCA (International Livestock Centre for Africa). 1987.
be identified in consultation with farmers, developers ILCA 's strategy and long-term plan. ILCA, Addis
and extensionists. Ababa, Ethiopia. 99 pp.
For humid and subhumid zones integrating sheep Kang В T and Reynolds L. 1989. Alley farming in the humid
and goats more closely into the farming system should and subhumid tropics. Proceedings of an international
be a long-term objective, taking into account the workshop held at Ibadan, Nigeria, 10-14 March 1986.
emphasis placed by small-scale farmers on risk IDRC (International Development Research Centre),
avoidance. Market demand for the product and the Ottawa, Canada. 25 1 pp.
economics of production should be important Kang В T, Reynolds L and Atta-Krah A N. 1990. Alley
considerations in determining the priority of biological farming. Advances in Agronomy 43:315-359.
research. Another key factor is clarifying who will
Mbabu A N, Nolan M F and Gilles J. 1991. An evaluation of
implement the technology at farm level, and who will SR-CRSP's techpacks in western Kenya: a year-long
reap the benefits. study. In: Proceedings ofthe Ninth S R-CRSP Scientific
The need for multidisciplinary work at farm level Workshop. Held on 27-28 February 1991 at
is overwhelming, but the costs of such work can be high. Ketri-Muguga. KARI (Kenya Agricultural Research
Cooperation between research, development/extension Institute)/SR-CRSP (Small Ruminant Collaborative
and farmers, within and across national borders will be Research Support Programme), Nairobi, Kenya, pp.
necessary to achieve the conditions for success. 35-^8.
Networking can help to build these links. Norton G W. 1 99 1 . Methodsfor agricultural research priority
setting: Implications for research planning in
Zimbabwe. ISNAR (International Service for National
Agricultural Research), The Hague, Netherlands.
References
Norton G W and Pardey P G. 1987. Priority-setting
von Braun J and Pandya-Lorch R. 1991. Income sources of mechanisms for National Agricultural Research
malnourished people in rural areas: microlevel Systems: Present experience andfuture needs. Working

Small Ruminant Research Network Workshop 33

L. Reynolds

Paper 7. ISNAR (International Service for National sheep at two planes of nutrition. Small Ruminant
Agricultural Research), The Hague, Netherlands. 33 pp. Research 1:175-188.
Reynolds L and Adediran S. 1988. The effects of browse Sumberg J and Okali C. 1989. Farmers, on-farm research and
supplementation on the productivity of West African new technology. In: Chambers R, Pacey A and Thrupp
Dwarf sheep over two reproductive cycles. In: Bosman L A (eds), Farmer first. Intermediate Technology
H and Smith О В (eds), Goat production in the humid Publications, London, UK. pp. 109-1 14.
tropics. PUDOC (Centre for Agricultural Publishing and Walshe M J, Grindle J, Nell R and Bachmann M. 1991. Dairy
Documentation), Wageningen, The Netherlands.
development in sub-Saharan Africa. World Bank
pp. 83-91. Technical Paper 135. The World Bank, Washington DC,
Reynolds L and Adediran S. 1 993. Composition of village goat USA. 94 pp.
herds in southwest Nigeria. Small Ruminant Research Winrock International. 1992. Assessment of Animal
(in press).
Agriculture in sub-Saharan Africa. Winrock
Reynolds L and Ekwuruke J O. 1988. The effect of International Institute for Agricultural Research,
Trypanosoma vivax infection in West African Dwarf Morrilton, Arkansas, USA. 125 pp.

34 Small Ruminant Research Network Workshop

 Overcoming the constraints of dry matter intake
on dual-purpose goat production by feeding
defoliated maize leaves
P.P. Semenye, M. Shisya and W.R. Getz

SR-CRSP/Winrock International
P. O. Box 252, Maseno, Kenya

Abstract production de fourrages, il est recommandé d'aligner

stratégiquement l'alimentation et la conduite des
Feeding trials were conducted at Maseno, Kenya, based caprins sur la production de maïs et de synchroniser la
on defoliated maize leaves from three plots measuring période de lactation et l'époque où les feuilles de maïs
0.25 ha each. On average, each plot produced 359 kg sont disponibles. Par ailleurs, les chevreaux de 1 an
(SE 40 kg) dry matter of defoliated leaves as forage. devraient pouvoir bénéficier d'un régime à base de
Defoliation, which commenced at the tasselling stage at feuilles de maïs, ce type de ration permettant des
the rate of one leaf per week per maize plant, did not performances supérieures à celles obtenues avec un
affect the grain yield. The defoliated maize leaf forage régime à base d'un mélange d'herbe
had 12% crude protein and in vivo dry matter
digestibility of 60%.
In maize-growing areas where there are no Introduction
pastures or land for growing fodder, it is recommended
that feeding and management of goats should be Kenya' s population is expected to increase by more than
strategically matched with maize production. Lactation 50% by the end of the century. This forecast has serious
period should coincide with the availability of maize implications for livestock production in densely
leaves, while yearlings should take full advantage of a populated areas with small farms as is the case in western
maize leaf diet which gives better performance than a Kenya. In these areas human and livestock populations
grass mixture diet. are inversely correlated. As the human population
increases, more land is needed for food crop production.
This is acquired from livestock' s share ofland resources.
Utilisation de feuilles de maïs As land previously reserved for livestock is taken,
stocking pressure increases. This results in logistical
dans l'alimentation de chèvres à destocking and poor performance syndrome caused
primarily by low intake of dry matter. A sustainable
aptitudes mixtes afin de solution to this problem could be the integration of
supprimer les facteurs limitant livestock with crop production. For this reason research
for dual-purpose goats (DPG) in western Kenya focuses
la consommation de matière on integration of the sub-systems.
sèche
Résumé Materials and methods
Des essais d'alimentation ont été conduits à Three plots of a quarter hectare each were demarcated
Maseno (Kenya) avec des feuilles de plants de maïs from a plot of two hectares planted with maize hybrid
provenant de trois parcelles de 0,25 ha. Chaqueparcelle seed no. 62 1 . The number of maize plants at knee height
produisait en moyenne 359 kg (+ 40 kg) de matière sèche on each plot was counted. Plot 1 had 7875 maize plants,
sous forme de feuilles utilisées comme fourrage. La Plot 2 had 7536 and Plot 3 had 9278, giving a total of
défoliation des plants de maïs, commencée au stade de 24,689 and an average of 8230 plants per plot.
la formation des aigrettes à raison d'une feuille par Randomly, Plot 1 was allocated to two lactating does,
semaine et parplant, n 'a pas eu d 'effet sur le rendement Plot 3 to three yearlings and Plot 2 was the control.
en grains. Les feuilles de maïs servant de fourrage Defoliation began 79 days after planting. Grain yield
was determined by harvesting every fifth maize plant on
présentaient une teneur en protéines brutes de 12% et
a transect running longitudinally on each plot. Nine
une digestibilité in vivo de la matière sèche de 60%.
yearlings and two does were used, all crossbreeds (Fl)
Dans les régions productrices de maïs où l'on of Toggenburg with East African Goat. The yearlings,
manque de pâturages et de terres se prêtant à la with an average live weight of 21.2 kg (± 0.2 kg), were

Small Ruminant Research Network Workshop 35

P. P. Semenye, M. Shisya and W. R. Getz

Table 1 Maize grain and defoliated leaves production per quarter hectare.

Plot 1 Plot 2 Plot 3 Overall

Parameter (Does) (Control) (Yearlings) mean SD
Number of plants 7875 7536 9278 8230 924
Number of leaves 110250 105504 129892 115215 12930
Leaf dry matter (kg) 344 329 405 359 40
Grain yield (kg) 1362 1263 1387 1337 66
Grain yield/plant (g) 173 168 149 163 13
SD = standard deviation.

allocated randomly in groups of three to three feeding available than the goats could eat. Plot 1, which catered
treatments of grass mixture, maize leaves and for two does, produced 344 kg dry matter of green maize
concentrates (dairy meal). Their adaptation period to the forage, while the total dry matter intake (DMI) over the
feeding trial was determined by the yearlings' live trial period of 90 days was 333 kg. Plot 3 produced 405
weight trend. The adaptation period was five weeks, kg dry matter, while the total DMI of the three yearlings
which was the time the yearlings took to regain their was 213 kg. The yearlings' plot produced sufficient
initial live weight, following a weight loss on intro forage either to cater for two more yearlings over the
duction to the experiment. The experimental period was same duration, or to extend the three yearlings' feeding
determined by the availability of green maize leaves and duration from 90 to 171 days.
lasted eight weeks. Yearlings and does were treated Stripping maize leaves was an easier task than
against helminthiasis before the experiment started. All cutting mixed grass because, unlike cutting grass
the goats were placed in metabolic cages to determine mixture, it is a focused activity and requires no
dry matter intake (DMI) and in vivo dry matter implements.
digestibility (IVDMD). The animals were fed three Table 2 shows that the crude protein (CP) content
times a day at 0900 h, 1300 h and 1600 h. of maize leaves was above the threshold known to limit
Kjeldahl, and Tilley and Terry (1963) chemical DMI. Maize leaves had better in vitro dry matter
analyses were conducted. Energy calculations were digestibility (DMD) than grass leaves. As expected the
based on coefficients and formulas of MAFF ( 1 975) and concentrate diet had a higher CP content and DMD than
NRC (1981). SYSTAT (Wilkinson, 1990) was used for maize leaves and grass mixture.
data analysis.

Table 2. Chemical composition of the diets (%).

Grass mixture Maize leaves Concentrate
Parameter Mean ± SD Mean ± SD Mean ± SD
Dry matter 31a 34a 94b 0.9
3.2 3.6
Crude protein 12a 0.8 12a 18b 1.0
0.8
In vitro DMD 5T 2.1 64h 78c 1.9
1.8
In vivo DMD 61a 6.4 60a 4.5 68b 2.9
abe = means in the same row having different superscripts differ significantly (P¿0.05).
SD = standard deviation.

Results and discussion There was an inverse correlation between the age
of the maize and the quality of the defoliated leaves. CP
The grain yield was affected by plant population, but not dropped from 14 to 10% (r = -0.91) while DMD dropped
by defoliation (Table 1). Plot 3 had the highest from 66 to 56% (r= -0.74). Concentrate quality remained
population and the highest total grain yield, but the grain constant, while variations in the grass mixture were kept
yield per plant was the lowest. The mean grain yield of
minimal by selective cutting.
5349 kg per hectare or 59 bags of 90 kg each is above
the western Kenya average of 29 bags/ha (Ministry of The success of fodder production is ultimately
Agriculture, 1987). Stripping the maize leaves allows judged by livestock response. Table 3 shows that
for easy under sowing in the maize fields. Similar results yearlings responded positively to a diet of maize leaf.
from maize defoliation have been reported elsewhere. Kayongo-Male and Abate (1982) reported growth
During defoliation there were more green leaves responses of 95 and 98 g/day for sheep fed on maize
leaves, higher than that found in this study. Sheep

36 Small Ruminant Research Network Workshop

 Overcoming the constraints of dry matter intake on etc...

performance was better than that achieved by goats with lactation, but also due to the insufficiency of maize
concentrates. The difference between the species was leaves for sustaining milk production. The average DMI
probably due to ease of adaptation to a new feed rather for maize leaves was 1 .3 ± 0.3 kg per day/doe. Based on
than conversion efficiency. Sheep are more adaptable to an in vivo DMD of 62%, this supplied 10.9 MJ (ME)/day
a new feed than goats. per doe. This energy supply fell short of the daily energy
Bulkiness is known to fill the gut resulting in requirement of a doe by 1.3 MJ ME. An attempt by the
lowered dry matter and energy intake. Contrary to this does to meet this energy deficit may have resulted in the
known phenomenon, yearlings fed the grass mixture and observed average loss of 3.75 kg or 9% of body
maize leaves diets (Table 3) consumed more dry matter weight/doe during the 90 days' lactation period. Maize
on body weight basis (3.5%) than yearlings fed the leaves can promote milk production, but in order to
concentrate diet (3.1%). Notwithstanding their minimise mobilisation of body reserves, keeping the
relatively lower intake of dry matter, yearlings fed the does' milk production at 0.7 kg/day is recommended.
concentrate diet utilized their feed more efficiently and For western Kenya and other areas with two maize
grew better than yearlings fed the grass mixture and seasons per year, DPG production can be integrated with
maize leaves diets. The yearlings on the grass mixture maize production for the latter to provide forage (Table
were the least efficient in feed utilization and growth. 4). Except for the 0.1 ha plot size, defoliated maize
The lower protein:energy ratios of the grass mixture and leaves' potential production is in excess of a year's
demand for a 40 kg doe producing 0.7 kg milk per day.
maize leaves diets may have been largely responsible for
Excess forage can therefore be stored, used for raising
the lower response of yearlings fed these diets.
more DPGs or both. The nutritive value of defoliated
leaves is high for three months per season. With two
Table 3. Performance of DPG yearlings fed grass mixture,
defoliated maize leaves and concentrates. growing seasons there are six months of high quality
forage per year. Excess leaves could be sold and the
Grass Maize proceeds used to purchase farm inputs or to meet
Parameter mixture leaves Concentrates personal needs.
As-fed (kg/day) 3.0 3.5 1.5
DMI (g/day) 764 788 723
DMI (% body wt.) 3.5 3.5
Conclusion
3.1
Average daily gain 27 45 89 Kenyans grow maize year in year out because it is a
(g/day) preferred food crop. As long as maize maintains this
role, its production must be capitalised on for the
Protein: energy ratio 8 8 12 enhancement of sustainable agriculture. The importance
Feed conversion 28 18 8 of integrating maize production with other crops and
livestock production cannot be over emphasised. Also
the benefits derived from under sowing in defoliated
The two lactating does in the trial were in their third maize plots need to be assessed. An under sown crop in
and sixth months of lactation. They were previously on a plot of defoliated maize plants has more access to
a high plane of nutrition made up of a grass mixture as sunlight, and hence the opportunity for higher
the basal diet, and Leucaena leucocephala and photosynthetic output.
concentrates as supplementary feeds. For the nine
months' lactation on a high plane of nutrition, the does
produced an average milk yield of 1 .7 kg per day/doe. References
When the does switched to defoliated maize leaves, the
average milk yield dropped to 0.9 kg per day/doe for a Haenlein G F W. 1978. Dairy goats do well on free choice
six months' lactation. The drop in yield was partially due feeding. Hoard's Dairyman 123:1 194.
to the normal decline in milk production with advancing Kayongo-Male and Abate. 1982

Table 4. Production of defoliated maize leaves per plot size and DPG feeding days.
Potential feeding days for a 40 kg doe producing
Defoliated leaves 0.7 kg milk per day
Maize plot production potential Dry matter
size (ha) (number) yield
per season per year
1.0 460 860 1437 1105 2210

0.5 230 430 719 553 1106

0.25 11525 359 276 552

0.1 46 086 144 111 222

Small Ruminant Research Network Workshop 37

P. P. Semenye, M. Shisya and W. R. Getz

MAFF (Ministry of Agriculture, Fisheries and Food). 1975. Requirements of Domestic Animals Number 15.
Energy allowances and feeding systems for ruminants. National Academy of Sciences, Washington DC, USA.
Technical Bulletin 33. Her Majesty's Stationery Office, 91 pp.
London, UK. 79 pp. ТШеу j M A and R A Terry 1953 A tw0 stage technique for
Ministry of Agriculture, Kenya Government. 1987. Per w vitro digestion of forage crop. Journal of the British
hectare gross margins. Nairobi, Kenya. Grasslands Society 1 8: 1 04.
NRC (National Research Council). 1981. Nutrient Wilkinson L. 1990. SYSTAT. The System for Statistics,
requirements ofgoats: Angora, dairy and meat goats in SYSTAT, Inc. Evanston IL, USA. .
temperate and tropical countries. Nutrient

38 Small Ruminant Research Network Workshop

 Goat development in Masvingo province, Zimbabwe:
the farmers' perspective
С Shumba

AGRITEX, P.O. Box 354, Masvingo, Zimbabwe

Abstract of goat production within the communal areas of Gutu

and Zaka Districts of Masvingo Province. The survey
A great deal of goat research has been conducted in was carried out in Natural Regions 3, 4 and 5, both in
Zimbabwe since the 1980's. There is, however, a need and out of grazing schemes. The objectives were:
to link research to the farmers' needs and objectives. analysing production systems and constraints; assessing
This paper focuses on the farmers' perspective with socio-economic effects and constraints; and analysing
regard to goat development in Masvingo province. marketing trends, constraints and opportunities.
Information from farmers could indicate points of
intervention for technological innovations and help in
Le développement de l'élevage designing appropriate goat development strategies.
caprin dans la province de
Masvingo, au Zimbabwe: Methodology
le point de vue des paysans Informal and formal surveys were conducted at the same
time. A sample size of 120 respondents per district was
Résumé taken.

Les recherches sur les caprins effectuées au

Zimbabwe depuis les années 80 sont légion. Il reste Summary of survey findings
cependant que des liens doivent être établis entre la
recherche d'une part et les besoins et les objectifs des The goat enterprise in communal areas has more
paysans d'autre part. Cet article présente lepoint de vue opportunities than constraints. 27% of respondents had
des éleveurs sur le développement de l'élevage caprin no problems at all with their goat enterprises. Common
dans la province du Masvingo. problems are given below.

Production constraints
Introduction
Herding
Ngategize (1989) stresses the fact that, when making
recommendations for interventions in small ruminant The selective nature of feeding behaviour and their small
production, farmer goals and objectives have to be given size make goats difficult to herd even within grazing
serious consideration. What is the farmer trying to schemes that use fences. Goats move fast in a given
maximise or minimise in his production activities? Is it grazing area or paddock, they jump over fences into
profitable to the farmer to adopt such innovations and rested paddocks or they stray into fields during the
interventions? Are they going to compete with existing growing season. Fencing that can restrain cattle cannot
farm or non-farm enterprises? necessarily restrain goats. This means that there must
A social and financial cost-benefit analysis is always be a person looking after them. During the
necessary before implementing innovations. For cropping season this presents a labour constraint.
example, apparent production constraints such as high Evidence of labour constraint can be seen in the
mortality, long birth intervals and slow growth rate may restriction of goats to pens up to mid-afternoon during
not be as critical to the farmer as animal production summer.
scientists think. This explains why some technologically
sound interventions and innovations may not be adopted Kid mortality
in certain socio-economic and ecological circum Kid mortality was high at 33% compared to 18% at
stances. A multidisciplinary approach is necessary to research stations. About 55 and 30% of this mortality
understand fully the farmer's goals, objectives and his was due to diseases and predation respectively. 30% of
environment, and also to minimize disciplinary biases. respondents had no disease problems, but 50% reported
With this in mind, this study focused on current some problems. The commonest diseases mentioned by
goat management practices and socio-economic aspects farmers included heartwater (a tick borne disease),

Small Ruminant Research Network Workshop 39

С. Shumba

pneumonia (caused by exposure to cold, wind and rain), regarded as capital); and a highly popular and acceptable
foot rot (caused by muddy conditions in goat houses) meat among different peoples and religions.
and worms and infestation.
Managerial advantages
Housing and kraaling
Apart from herding problems, goats are not difficult
15% of respondents had poor goat houses, which limited
animals to manage. Very few respondents reported
production. Such houses had no protection against wind,
management problems. Family labour can adequately
cold or rain, and muddy conditions prevailed during the
manage goats except during the growing season when
rainy season. Lack of knowledge and financial resources
labour demands are high.
were cited as causes for such poor goat houses. Housing
improvement is required, especially in flooring.
Biological advantages
Lack of knowledge
These include: versatility and ability to adapt to various
Extension staff and farmers have received no training in ecological environments; high resistance to disease and
goat production, which is a serious constraint to dehydration; and complementarity of goats with cattle,
production. Most of the production constraints resulting in efficient resource utilization as goats use
mentioned above are attributed to lack of knowledge and those resources which cattle do not.
poor managerial skills.

Socio-economic constraints The proposed strategy for goat

development
Goat marketing system
From the survey findings a goat development strategy
The viability of a goat enterprise depends not only on for Masvingo Province can be proposed. The major
technical and biological efficiency, but also on a well feature of the strategy is the proposal that the different
organised marketing system. facets of goat development work should be integrated
At present, goat markets are characterised by through research, extension training, production and
poorly managed and unrealistic carcass grading and marketing techniques.
pricing systems, inadequate promotion of goat meat and If any of these areas is overlooked in the overall
an inadequate and inefficient transport system. These goat development strategy, problems can be expected
problems will have to be corrected if goat production is even before a project begins. Each element in the
to be improved. proposed strategy will be looked at in turn.

Attitudes to goats Research

The farmer regards cattle as being more important than
goats. In the case of limited resources, cattle always take With on-station research information as the guideline,
priority, putting goats at a disadvantage. This is not to more on-farm adaptive research is necessary. On-farm
say that goats are unimportant to the farmer, but the adaptive research is essential because it assesses the
multipurpose uses of cattle vis-à-vis goats produce a bias interrelationships between cattle and goats within the
in favour of cattle. context of the communal farming system. Such research
should take into account the limited grazing resources
available; the prevailing land tenure system; and the
Opportunities for goat production financial status and management systems in the
in the districts communal set-up. Specific areas for adaptive research
include: determining goat and cattle populations in
The environment relation to available grazing resources; determining
differences between vegetation types; analysing
Masvingo Province lies in the semi-arid region of vegetation composition in different natural regions or
Zimbabwe and is basically a livestock area. The two areas to assess the availability of palatable species for
districts surveyed lie in the low rainfall areas of the goats; and determining the extent to which goats control
country where browsable species are abundant. bush in the grazing system of a communal area.
Palatable browse species for goats are location-
Socio-economic advantages specific, and research on them cannot be done on-station
as findings will be limited to that locality. Carrying
Leaving aside the investment needs and risks of goat capacities and stocking rates are also location-specific.
rearing compared with cattle, the socio-economic Variations occur even within one natural region due to
advantages of goat production include: high profitability climatic, soil and topographic factors. Research
and fast turnover due to earlier maturity and shorter institutions should give guidelines for the assessment of
generation interval; suitable meat quantities for rural carrying capacity, stocking rates and available browse
families; easy salability (unlike cattle which are for goats in conjunction with on-farm adaptive research.

40 Small Ruminant Research Network Workshop

 Goat development in Masvingo province, Zimbabwe etc..

Extension/Training The marketing policy should therefore be to

improve prices, improve transport facilities for live
Devendra (1982) observed that the goat has for a long goats from communal areas and design goat meat
time been considered a veld degrader and neglected in promotion and advertising techniques.
both research and extension institutions. This bias
against goats, coupled with the low priority goats are
given and other negative attitudes, has resulted in goat
development being allocated limited financial Conclusions
resources. Agricultural institutions have concentrated
their training on beef and dairy development. Such Farmers in different localities have different needs and
attitudes should change and more personnel should be objectives relating to goat production. Within the two
trained specifically in goat management at both higher districts surveyed there are variations in the
and lower levels. The training of extension staff is a sure environment, in the availability of grazing area
way to have communal area farmers realise and exploit resources and in browse species. Herd sizes vary from
the potential of goat production. Extension staff should 8-10 head and goats are kept mainly for home
be trained at the worker, officer and specialist levels. consumption. Such variations mean that it is difficult to
Once staff are trained, dissemination of information to make blanket recommendations about goats. This also
farmers should be the next step. Farmer training highlights the need for baseline surveys before
programmes, the provision of reading material and implementing goat projects in communal areas. The
demonstration units for goat management practices can important thing is that any goat development strategy
all be used to educate the farmers. designed should address the farmers' needs and
problems directly.
Goat production
Goat production should move from an extensive (low
production level) system to a semi-intensive (higher References
production level) system, but at costs which are within
the reach of farmers. This part of the strategy should Blackie M J. 1982. A time to listen: A perspective on
agriculture policy in Zimbabwe. Zimbabwe Agricultural
incorporate improved nutrition, better kid management Journal 79(5):151-156.
and housing, reduced disease, reduced predation and
better breed selection from within existing local stocks. Devendra C and Mcleroy G В. 1982. Goat and sheep
production in the tropics. Longman, Harlow, UK. 27 1
Marketing pp.
Mutetwa C. 1 987. Developing a marketing strategyforpoultry
The viability of goat production depends not only on in Masvingo Province—Zimbabwe. Msc thesis
technical and biological efficiency, but also on market (unpublished).
factors (Mutetwe, 1987). The goat market is a neglected Ngategize P K. 1989. Constraint identification and analysis in
area and limits goat development in communal areas. African small ruminant systems. In: Wilson R T and
Blackie (1982) suggests that an effective marketing Azeb Melaku (eds), African small ruminant research
system for goats (and sheep) would likely increase and development. Proceedings of a conference held at
communal area meat production and peasant incomes Bamenda, Cameroon, 18-25 January 1989. ILCA
and improve veld (range) conditions more than would a (International Livestock Centre for Africa), Addis
substantial rise in the price of beef. Ababa, Ethiopia. pp. 7-22.

Small Ruminant Research Network Workshop 41

 Goat production in a mixed cattle-goat system:
Effect of stocking and substitution rate on redsoil
thornveld stability
1 2
J.L.N. Sikosana and J. Gambiza

Matopos Research Station, Private Bag K5137, Bulawayo, Zimbabwe

Makoholi Experiment Station, Private Bag 9150, Masvingo, Zimbabwe

Summary because of the encroachment of dense bush. Donaldson

(1979) reported that it is usually impractical or un
The effects on animal performance and vegetation economical to thin or remove the woody components.
changes of cattle and goats grazing together at five It has been suggested that the full production potential
different stocking rates in a continuous grazing system of bushveld areas can be realized by integrating goats
are being studied at Matopos Research Station. In the into cattle fanning systems (Aucamp, 1976).
long term, the trial is expected to provide information Mixed species grazing has received little research
on the impact of climatic cycles. Preliminary results for attention, despite the fact that most communal and
the first three years are presented. Year effect was small-scale commercial farmers in Zimbabwe graze
significant (P<0.01). The lightest stocking rate (3.3 cattle and goats together. Goats seem to survive
ha/LU) for both goats and cattle had higher (P<0.01) droughts better than cattle. This has led to the belief that
performance levels compared with other treatments. goats degrade rangelands more than cattle. However, it
There was no effect of treatments on density and basal is argued that stocking rate is the most important
cover of herbaceous plants. variable that affects rangeland condition rather than
animal species per se (Gambiza and Sikosana, 1989). A
multidisciplinary study on the interaction of goats and
Elevage caprin en système mixte cattle grazing on redsoil thornveld was started in 1989.
The objectives were to study the effects on rangelands
bovin-caprin: effet des taux de of different goat and cattle stocking and substitution
charge et de substitution sur la rates and the productivity of goats in a mixed goat and
cattle system.
stabilité des formations à A continuous grazing system was used.
épineux sur sol rouge Preliminary results on the standing biomass of
herbaceous plants and animal performance for the three
Résumé seasons are presented.

Les effets sur les performances zootechniques et Materials and methods

sur l'évolution de la végétation de la mise au parcours
de bovins avec des caprins à raison de cinq taux de The study is being conducted at the Matopos Research
charge différents ont été étudiés à la station de Station in the south west of Zimbabwe. The site is at an
recherche de Matopos (Zimbabwe). A long terme, altitude of 1340 m with a mean annual rainfall of 600
mm which falls between October and April. Marked
l 'essai devraitfournir des informations sur l 'impact des
fluctuation in rainfall brings about corresponding
cycles climatiques. Les premiers résultats des trois
fluctuations in herbage production from year to year.
premières années sont présentés. L 'effet de l 'année était October is the hottest month (mean maximum of 29.4°C)
significatif(P<0,01 ). Le niveau de performance obtenu and June the coolest (mean maximum of 20.9°). The land
avec le taux de charge le plus faible (3,3 ha/UBT) pour is characterised by clay soil (red and black), thomy
les caprins et les bovins était significativement supérieur acacia species and other shrubs in the browse layer,
(P<0,01 ) à celui obtenu avec les autres traitements. Les while the field layer is composed of perennial grasses
traitements n'avaient pas d'effet sur la densité et la with occasional annuals. A brief description of the area
surface terrière pastorale des herbacées. and its dominant vegetation has been documented
(Wardetal, 1979).
Initially the site was subjected to a uniform grazing
Introduction regime prior to the introduction of treatments. Ten
paddocks of 20 hectares each were grazed continuously
Many grazing areas of Southern Africa are badly at a medium stocking rate of 2.8 ha per livestock unit
degraded or are in the process of becoming unproductive (LU = 500 kg liveweight). In June 1989 the planned

Small Ruminant Research Network Workshop 43

J. L. N. Sikosana and J. Gambiza

stocking regimes were applied (Table 1). Goats were Figure 1. Annual rainfall at the site ofthe cattle -goat mixed
substituted for cattle on the basis of metabolic grazing system.
liveweight. The conversions to LUs assume 1.67:1 for
Rainfall (mm)
cattle and 10: 1 for goats. Additivity is then assumed for
the treatments although this is likely to give high 1000
estimates of the true rates and is to be the subject of study
in the trial. A total of 60 steers were used in the trial.
They were maintained in two age groups, weaners and 800
long weaners in equal proportions. The long weaners
were slaughtered each year at the age of two and a half Long term mean
years and replaced by weaners. Monthly weights were 600
recorded, and carcasses were assessed as described by
Moyo (1991). Slaughter data is used to assess cattle
productivity from each treatment. 400

Table 1. Cattle and goat numbers and stocking density in

a mixed species stocking rate trial on redsoil
thornveld. 200

Treatment Cattle Goat Ha/LU* LU/20ha

1 . Light - cattle 8 12 3.3 6.0
i ' 1 1 i 1 1 I—T—i r
2. Heavy - cattle 8 36 2.1 8.1 1982 83 84 85 86 87 88 89 90 91 92
Year
3. Light -goat 4 36 3.3 6.0
4. Heavy - goat 4 60 2.4 8.1 Statistical analysis
5. Medium: cattle Data were analysed by the analysis of variance
-goat 6 36 2.8 7.2 (ANOVA) using a General Linear Model (SAS, 1982)
LU = 500 kg liveweight. The animal species contributing the higher fitting replication, treatments and their interactions.
proportion of livestock units has been quoted. Animals that became sick or died in the course of
the experiment were replaced with animals of similar
A total of 360 indigenous (Matabele goat) breeding weights. This was done to maintain the prescribed
does of stratified age groups were used. AH goats were stocking pressure. Data from these animals and their
mated in May and June to kid once a year during the start replacements were not included in this preliminary
of the wet season (October and November). The analysis.
offspring from the goats were weaned at an average age
of five months and details of birth, deaths and weaning
were recorded. Reproductive performance of the does
was assessed. Results
Herbaceous and woody plant dynamics (species Livestock data for different stocking rates are presented
composition, basal cover, density of above-ground grass in Table 2 (production traits of goats), Table 3 (steer
biomass) and soil variables were measured in permanent productivity), Figure 2 (total weaner production from
plots. goats, kg per 20 ha) and Figure 3 (the overall doe
The annual rainfall at the experimental site for the productivity).
duration of the trial is shown in Figure 1. Superimposed Goat performance under the lightest stocking rate
on Figure 1 is the long term mean rainfall at the site. was higher (P<0.01) than other treatments. Figure 2

Table 2. Production traits of goats in a mixed cattle and goat grazing system (1989/90 to 1991/92).
1 2 3 4 5 SE
Stocking rate (ha/LU) 3.3 2.4 3.3 2.4 2.8
No. of goats:steers 12:8 36:8 36:4 60:4 36:6
Total no. of does kidded 51 136 119 157 136
Prolificacy 1.6 1.4 1.4 1.3 1.4 0.05
Kid birth weight (kg) 2.0 1.8 1.9 1.8 1.8 0.14
Kid weaning weight (kg) 15.9 13.5 13.6 20.8 13.0 0.30
Pre-weaning gain (kg) of kids 13.0 11.0 11.0 8.6 10.7 0.37
Kidding %2 127.2 97.5 95.0 71.9 104.9 5.31
Prolificacy (litter size)= number of kids born/number of does kidding
Kidding % = number of kids bom/number of does mated X100

44 Small Ruminant Research Network Workshop

 Goat production in a mixed cattle-goat system etc...

Table 3. Weight gain and carcass characteristics of steers in a mixed cattle and goat grazing system (1989/90 to 91/92).
Treatment 1 2 3 4 5 SE
Stocking rate (ha/LU) 3.3 2.4 3.3 2.4 2.8
No. of goats: steers 12:8 36:8 36:4 60:4 36:6
Total no. of cattle slaughtered 16 16 8 8 12
Initial weight at 8 months of age (kg) 191 192 197 197 189 4.14
Weight gain per head (kg) 111 115 126 122 109 7.19
Cold carcass weight (kg) 117 115 126 120 112 3.29
Carcass length (cm) 118 117 118 118 115 1.26
Eye muscle area (cm ) 38.4 40.6 43.9 40.5 37.2 1.8

Figure 2. Total weanerproductionfrom goats (kg per 20 ha) Figure 3. Overall doe productivity (number ofkids weaned
in a cattle - goat mixed grazing system. XAverage weaning weight ofkids/Number ofdoes
mated.
Total weaner production from goats
Overall Doe Productivity
(kg per 20 ha)
20n
400 i | Treatment 1
U Treatment 2
[] Treatment3

300 0 Treatment 4
/.
[] Treatments

200

100

1990

| Treatment 1 Ц Treatment 2 Щ Treatments

1992
] Treatment 4 ] Treatment 5

shows that weaner production from goats increased in conjunction with grazers (cattle), is potentially
when total and goat stocking rate was low (P<0.01). acceptable. This agrees with Aucamp et al (1984). A
Other treatments resulted in inconsistent productivity prerequisite for implementing an integrated system is
over the three years. This is also the trend in the overall determining an appropriate mix of cattle and goats in
doe productivity (Figure 3). order to utilise efficiently both the herbaceous and
It is evident that the annual liveweight production woody vegetation. Discussion of mixed grazing at this
of steers (Table 3) was largely dependent on total stage of the trial is restricted due to lack of adequate
biomass (Table 4). Steers in treatment 3 had the highest
information. More research is needed to explain the
(P<0.05) weight gains over the three years.
biological processes involved (Nolan, 1979; Shuster,
1986).
These results show that livestock productivity
Discussion and conclusion varied across treatments. This could be due to seasonal
variations and differences in forage availability during
Stocking rate is now widely accepted as the major factor the study, as well as stocking rate.
associated with livestock output. If productivity is to be
maintained, the stocking rate has to be appropriate to the Livestock equivalents for cattle and goats need
output goal and adjusted to the carrying capacity of the further investigation. These may have to vary from
ecosystem. This should also reduce abuse of grazing region to region depending on the species composition
land, essential in semi-arid areas. of the grazing areas. Sound animal and veld
Results from this study suggest that the pro management practices, together with the correct ratios
ductivity of indigenous goats, when managed correctly of cattle and goats, are essential for success.

Small Ruminant Research Network Workshop 45

J. L. N. Sikosana and J Gambiza

Table 4. Grass basal cover (%) and peak biomass (tonnes dry matter per ha) in a mixed cattle and goat grazing system.
Treatment 1 2 3 4 5
Stocking rate (ha/LU) 3.3 2.4 3.3 2.4 2.8
No. of goats:steers 12:8 36:8 36:4 60:4 36:6
Basal cover (1990) 2.9 3.5 4.0 3.4 3.3
(1991) 2.9 3.5 4.4 3.0 3.4
(1992) 2.2 1.8 3.0 1.6 1.6
Biomass (1990) 1.2 1.3 1.6 1.3 1.3
(1991) 0.6 0.5 0.9 0.5 0.6
(1992) 0.12 0.16 0.08 0.01 0.01

Gambiza J and Sikosana J L N. 1989. Goat production in a

Acknowledgements mixed cattle and goats system. Journal ofthe Grassland
The technical assistance of the Smallstock and Veld and Society ofSouthern Africa 1 :29-32.
Pasture Teams of Matopos Research Station is gratefully Moyo S. 1991. Carcass characteristics of crossbred progeny
acknowledged. Mr Mario Beffa is thanked for assistance from beef breeds in Zimbabwe. Proceedings of a
with statistical analyses (livestock data). Dr Smith is Regional Workshop of Animal Production Societies of
thanked for reading and commenting on the original Eastern and Southern Africa held in Harare, Zimbabwe,
manuscript. August, 1991.
Nolan T. 1979. Grazing grounds for cattle, sheep and goats.
Advantages and disadvantages of joined pasturing.
References Fourteenth International Symposium of Zootechny.
Milan, 15-17 April, 1979.
Aucamp A U. 1976. The role of browse in bushveld of the
Eastern Cape. Proceedings of the Grassland Society of SAS. (Statistical Analysis System). 1982. SAS users guide
Southern Africa 11:135-138. SAS: Statistical Analytical Institute Inc., Cary, North
Carolina, USA.
Aucamp A U, Danckwerts J E and Venter J J. 1984. The
production of an Acacia karro community utilised by Schuster J I. 1986. Multispecies grazing research strategies.
cattle and goats. Journal of the Grassland Society of Inter African Animal Resources Information Leaflet
Southern Africa 1:29-32. Vol. 34. Number 36.
Donaldson С H. 1979. Goat and/or cattle on Mopane veld. Ward H К, Richardson F D, Denny R P and Dye P T. 1979.
Proceedings ofthe Grassland Society ofSouthern Africa Matopos Research Station — a perspective. Rhodesia
14:119-123. Agricultural Journal 76:5-1 8.

46 Small Ruminant Research Network Workshop

 SESSION 2

Small Ruminant Performance and

Reproduction
 Evaluation de l'activité sexuelle pendant la saison de
baisse de fertilité chez la chèvre de race D'man
1 2
Derquaoui L. et El Khaledi O.

Département de reproduction animale et d'insémination artificielle

Institut agronomique et vétérinaire Hassan II, Rabat (Maroc)
Clinique vétérinaire, Assila (Maroc)

Résumé simultaneous decline of ovarian activity and oestrus

expression.
Une baisse de la fertilité a été observée chez la
chèvre D'man pendant les mois de février, mars et avril.
Les activités ovarienne et oestrale entre mars et juin de Introduction
20 chèvres adultes de race D'man ont été étudiées, la
première par détection biquotidienne (8 heures et 17 Au Maroc, l'élevage caprin représente une activité
heures) des chaleurs et la seconde par observation agricole très importante, surtout dans les régions les plus
directe des ovaires (endoscopie) 4 à 10 jours après la défavorisées telles que les montagnes et les parcours
manifestation du comportement oestral. L'incidence des dégradés. En 1 987, le cheptel caprin national était estimé
ovulations silencieuses a été estimée par la mesure du à 5,8 millions de têtes (MARA, 1990). La quasi-totalité
taux de progestérone plasmatique. est élevée en système agropastoral ou pastoral extensif,
à l'exception de la chèvre laitière de race D'man, dont
La proportion des chèvres en chaleur était de 70%, l'effectif est estimé à 20 000 têtes, soit 0,34% de
45%, 55% et 90% pour les mois de mars, avril, mai et l'effectif total. Cette race est élevée en système très
juin respectivement; celle des chèvres ayant ovulé était intensif et en troupeaux de petite taille de 2 à 17 têtes
respectivement de 85%, 60%, 65% et 95%. La dans les oasis de la vallée du Draa (Hachi, 1990).
proportion des cycles courts (39%) et l'incidence des L'analyse des données de mise-bas enregistrées pendant
ovulations silencieuses (15 à 25%) étaient relativement sept années à la station de Tinzouline (centre de la vallée
élevées. La baisse de la fertilité de la chèvre D'man au du Draa) (Ezzahiri et Benlakhal, 1989) et chez les
printemps est donc principalement due à une baisse éleveurs (Hachi, 1990) a montré que les chevrottages
simultanée de l'activité ovarienne et de l'extériorisation s'étalent sur toute l'année. Toutefois, une faible
des chaleurs. incidence des naissances (7%) a été observée en été
(juillet, août et septembre), ce qui correspond à des
saillies de printemps (février, mars et avril). L'objectif
Evaluation of breeding activity de cette étude est d'évaluer les activités ovarienne et
in D'man goats during lower oestrale entre les mois de mars et de mai chez la chèvre
D'man élevée en station en dehors de son berceau de
fertility period race.

Abstract
A decline infertility was observed in D 'man goats in the
Matériel et méthodes
months of February, March and April. Ovarian and Cette étude a été réalisée à la ferme d'application
oestrus activity of 20 adult D'man goats was observed de l'Institut agronomique et vétérinaire Hassan II, située
during March and June, the first by oestrus detection à la périphérie du périmètre irrigué du Tadla. Cette
twice daily (8 am and 5 pm) and the latter by direct région à climat méditerranéen de type aride et sous
observation ofthe ovaries (endoscopy) 4 to 10 days after influence continentale est caractérisée par des hivers
the first signs of oestrus behaviour. Silent oestrus was froids et peu pluvieux et des étés chauds et secs.
estimated by measuring plasma progesterone rate.
Seventy per cent, 45%, 55% and 90% of the goats Les animaux
were in heat in March, April, May and June,
respectively, while ovulation was observed in 85%, Le troupeau expérimental était constitué
60%, 65% and 95% goats, respectively. The proportion d'animaux achetés dans la vallée du Draa en février
ofshort cycles (39%) and the incidence ofsilent oestrus 1988. Après une période d'adaptation de deux ans, 20
(15 to 25%) were relatively high. The decline offertility chèvres adultes âgées de 2 à 7 ans ont été utilisées. Les
in D 'man goats during spring is therefore mainly due to animaux étaient maintenus en stabulation libre et

Small Ruminant Research Network Workshop 49

Derquaoui L. et El Khaledi О.

recevaient une alimentation à base de foin de luzerne et présentaient une activité ovarienne était toujours
de concentrés (orge grain et pulpe sèche de betterave). supérieure à celle des sujets qui manifestaient des signes
L'ensemble du troupeau caprin a été traité contre les des chaleurs. Cette caractéristique témoigne de
parasites internes avant le démarrage de l'étude. l'existence d'ovulations non accompagnées de chaleurs.
L'incidence des ovulations silencieuses était très élevée
Etude de l'activité oestrale aux mois de mars (17,6%), d'avril (25%) et de mai
( 1 5,4%), alors qu'elle était faible au mois dejuin (4,2%).
L'activité oestrale a été évaluée par une détection Par ailleurs, sur l'ensemble des animaux, 1 2 femelles sur
minutieuse des chaleurs suivant une exposition 20 (soit 60%) avaient une activité ovarienne continue
biquotidienne (8 heures et 17 heures) d'une demi-heure pendant toute la période de l'étude et 25% présentaient
des femelles à un mâle vasectomisé. Une femelle était un arrêt momentané ou continu de cette activité de mars
considérée en chaleurs quand elle devenait réceptive au à mai. Le nombre maximum d'ovulations enregistré
bouc, s'immobilisait et acceptait le chevauchement. dans les conditions de cette étude était de 2. Le taux
moyen d'ovulations doubles (46,5%) était légèrement
Etude de l'activité ovarienne inférieur à celui des ovulations simples (53,5%). Le
taux moyen d'ovulation ne variait pas en fonction du
On a procédé à l'observation directe des ovaires 4 mois (P>0,05).
à 10 jours après la manifestation du comportement des
chaleurs, afin de confirmer l'ovulation et d'en déduire
le taux. L'incidence des ovulations silencieuses a été
estimée à partir de la progestérone plasmatique dosée à
Discussion
l'aide d'un kit commercial (Diagnostic Laboratory, Los Au cours de cette étude, il a été observé une baisse
Angeles, Californie, E.-U.) dans des échantillons de de l'activité oestrale et ovarienne chez des chèvres
sang prélevés à 5 jours d'intervalle. La limite de D'man maintenues vides dans des conditions
sensibilité du dosage était de 0,06 ng/ml et les expérimentales en dehors de leur berceau de race entre
coefficients de variation intra- et interdosage étaient mars et juin. Ainsi, sur l'ensemble de la période d'étude,
respectivement de 6,0% et 8,2%. Une ovulation était 70% des femelles ont ovulé au moins une fois par mois
confirmée quand le taux de progestérone plasmatique mais 55% seulement ont manifesté un comportement
était inférieur ou égal à 1 ng/ml. oestral à ce rythme. Cela correspond à un taux moyen
de 21,4% des ovulations non accompagnées de signes
de chaleurs. Dans une étude antérieure, Hachi (1990)
Résultats rapporte des taux d'ovulations de 8,3 à 87% pour une
saison similaire. Ces écarts pourraient en partie
Activité oestrale s'expliquer par les conditions environnementales telles
que le niveau alimentaire ou les variations
Les résultats obtenus montrent que la saison influe interannuelles. Au demeurant, l'une des raisons à
sur l'activité oestrale de la chèvre D'man maintenue l'origine du faible taux de chevrottage observé chez la
dans des conditions expérimentales en dehors de son chèvre D'man en été est la diminution de l'activité
berceau de race. En effet, l'expression de l'oestrus, 70% sexuelle au printemps (Ezzahiri et Benlakhal, 1989;
au mois de mars, était minimum (45%) au mois d'avril. Hachi, 1990).
Des signes de chaleurs apparaissaient chez 55% des Il ressort également de cette étude que certaines
animaux à partir du mois de mai, et chez 92% des ovulations ne sont pas accompagnées de signes de
animaux au mois de juin. La durée du cycle oestral, chaleurs. La fréquence de ce type d' ovulations diminuait
définie comme l'intervalle de temps entre deux oestrus au fur et à mesure qu'approchait la saison sexuelle des
consécutifs, variait de 5 à 27 jours. La distribution des races dites saisonnées. Par ailleurs, les ovulations
cycles en fonction de leur durée fait ressortir deux types silencieuses constituent un phénomène physiologique
de cycle: l'un à 10 et l'autre à 20 jours. Le premier normal au début de la saison sexuelle chez les chèvres
correspond aux cycles courts dont la durée moyenne est alpine (Thibier et ai, 1981) et nubienne (Camp et ai,
de 10,5 ± 3,45 jours (5 à 16 jours); le second correspond 1982). Chez la brebis, elles sont généralement associées
aux cycles longs d'une durée moyenne de 20,96 ± 2,24 à la reprise de l'activité sexuelle post-partum (Chami,
jours (17 à 26 jours). 1981; Khaldi, 1984; Boukhliq, 1986; Lahlou-Kassi et
al., 1989) ou saisonnière (Lahlou-Kassi et Marie, 1985)
Activité ovarienne et à la période de passage de la vie prépubertaire à la vie
postpubertaire (Foster et Ryan, 1979; Derquaoui, 1992).
L'activité ovarienne suit les mêmes variations que La durée moyenne du cycle oestral est de 10,5 ±
l'activité oestrale. La proportion des chèvres ayant ovulé 3,45 jours pour les cycles courts et de 20,96 ± 2,84 jours
était de 85% au mois de mars et de 60% au mois d'avril. pour les cycles normaux chez la chèvre D'man. Cette
A partir du mois de mai, la proportion des chèvres à durée est comparable aux valeurs rapportées pour un
ovaires actifs s'élevait à 65% pour atteindre 95% au certain nombre de races caprines tropicales et revues par
mois de juin. Par ailleurs, la proportion des chèvres qui Garcia et Gall (1981). Il semblerait que l'incidence

50 Small Ruminant Research Network Workshop

 Evaluation de l'activité sexuelle pendant la ...

relativement élevée des cycles courts soit une and ovarian activity. Reproduction nutrition
caractéristique de l'espèce caprine. En effet, 86%, 32% développement. 236-104.
et 11% des cycles sont courts respectivement chez la Chemineau, P. et Thimonier, J. (1986). Methods for evaluation
chèvre nubienne (Camp et al. , 1 982), créole (Chemineau of reproduction and growth rate performance in sheep
et Thimonier, 1986) et chez la chèvre locale du and goat. World Review of Animal Production, 22
Vénézuela (Carlos et Madrid-Burry, 1982. L'origine et (n°4): 28-32.
l'étiologie des cycles courts des petits ruminants ne sont Derquaoui L. 1992. Onset of puberty and development of
pas complètement élucidées, mais il semble que les reproductive capacity in D'man and Sardi breeds of
corpsjaunes des cycles courts soient de mauvaise qualité sheep and their crossed. Thèse de doctorat ès-sciences.
et que leur durée de vie secrétoire soit limitée Institut agronomique et vétérinaire Hassan II, Rabat
(Chemineau et Thimonier, 1986) et fortement (Maroc).
influencée par le niveau alimentaire (Oldham et Martin, Ezzahiri A. et Benlakhal M. 1989. La chèvre D'man.
1979). Séminaire sur l'élevage caprin: problématique et
possibilité de développement. IXX* Journées de
l'ANPA (Association nationale pour la production
animale) Ouarzazat (Maroc).
Conclusion
Foster D.L. et Ryan K.D. 1979. Mechanisms governing onset
La baisse du taux de chevrottage observée chez la of ovarian cyclicity at puberty in the lamb. Annales de
chèvre D'man en été est principalement due aux baisses biologie animale, biochimie, biophysique 19:
simultanées de l'activité ovarienne et de 1369-1380.
l'extériorisation des chaleurs pendant le printemps, la Garcia O. et Gall C. 1981. Goats in the dry tropics. In: Goat
mortalité embryonnaire demeurant à priori faible. Par production. Publié sous la direction de C. Gall.
ailleurs, 60% des femelles présentaient des signes d'une Academic Press Inc., New York (E.-U.). p. 515 à 551.
activité sexuelle continue. Du fait que la chèvre D'man Hachi A. 1990. La chèvre D'man: Contribution à l'étude des
soit élevée en troupeaux de petite taille et dans des caractéristiques de reproduction. Thèse de doctorat
conditions très intensives, la sélection, à travers les vétérinaire. Institut agronomique et vétérinaire Hassan
générations, a probablement résulté en une perte de la II, Rabat (Maroc).
saisonnalité de la reproduction chez certains individus Khaldi G. 1984. Variations saisonnières de l'activité
de la population. ovarienne, du comportement d'oestrus et de la durée de
l'anoestrus post-partum de la race barbarme: Influence
du niveau alimentaire et de la présence du mâle. Thèse
de doctorat ès-sciences, Université de Montpellier II,
Références Montpellier (France).
Boukhliq R. 1986. Variations saisonnières de l'âge à la Lahlou-Kassi A. et Marie M. 1985. Sexual and ovarian
puberté, de la cyclicité sexuelle et de l'anoestrus function of the D'man ewe. In: Genetics ofreproduction
post-partum chez des brebis de race D'man, Sardi et leurs in sheep. Publié sous la direction de R.B. Land et D.W.
produits de croisement. Thèse de doctorat vétérinaire, Robinson. Butterworths, Londres (R.-U.). p. 245 à 260.
Institut agronomique et vétérinaire Hassan II, Rabat Lahlou-Kassi A., Berger Y.M., Bradford G.E., Boukhliq R.,
(Maroc). Tibary A., Derqaoui L. et Boujenane I. 1989.
Camp J.C., Wildt D.E., Howard P.K., Stuwart L. et Performance of D'man and Sardi sheep on accelerated
Charkaborty G. 1982. Behaviour, endocrine and lambing. I. Fertility, litter size, postpartum anoestrus and
ovarian relationships during the estrous cycle ofNubian puberty. Small Ruminant Research 2:225-239.
Goat. 3rd International Conference on Goat Production MARA (Ministère de l'agriculture et de la réforme agraire,
and Disease. Tucson, Arizona (E.-U.).
Maroc). 1990. L'élevage au Maroc. Symposium
Carlos Gonzalez, S. et Madrid-Burry, M. (1982). Sexual international sur l'élevage dans les systèmes céréaliers
season and estrous cycle of native goats in a tropical zone méditerranéens. Rabat, 7-10 octobre 1990. Direction de
of Venezuela. Third Int. Conf. on Goat Prod, and Disease. l'élevage, Rabat (Maroc).
10-15 Janvier, Tuscon, Arizona, USA.
Oldham C.M. et Martin G.B. 1979. Stimulation of seasonally
Chami M. 1981. Etude de la physiologie de l'anoestrus post anovular Merino ewes by rams. II. Premature regression
partum chez deux races ovines, D'man et Sardi, et leurs of ram induced corpora lutea. Animal Reproduction
produits de croisement. Thèse de doctorat vétérinaire, Science 1:291-295.
Institut agronomique et vétérinaire Hassan II, Rabat
(Maroc).
Thibier M., Pothelet D., Jean Guyot N. et De Montiguy G.
1981. Estrous behavior, progesterone in peripheral
Chemineau P. 1986. Sexual behaviour and gonadal activity plasma and milk in dairy goats at onset of breeding
during the year in the tropic. I. Female estrous behaviour season. Journal ofDairy Science 64: 513-519.

Small Ruminant Research Network Workshop 51

 Etude du cycle oestral chez la brebis Djallonké
M.S. Hounzangbé-Adoté

B.P. 1170, Cotonou (Bénin)

Résumé institutions de recherche. Compte tenu de l'importance

de la durée du cycle oestral dans les efforts visant à
Le cycle oestral de brebis Djallonké allaitantes et améliorer et à accroître la production de la brebis, il
de brebis non allaitantes a été étudié pendant une période s'avère nécessaire de recourir à d'autres techniques
d'un an. Les chaleurs ont été détectées par exposition à fiables mais plus abordables.
des béliers et par examen des frottis vaginaux. Qui plus est, selon Doney et Gunn (1981) et
La plupart des cycles oestraux déterminés par Haresing (1984), la persistance de plusieurs parasites
détection des chaleurs avaient une durée de seize à gastro-intestinaux et la restriction alimentaire pendant la
dix-neufjours et se répartissaient sur toute l'année. Les période de croissance, caractéristiques des systèmes
chaleurs silencieuses étaient plus fréquentes en saison extensifs d'élevage dans les régions humides de
pluvieuse qu'en saison sèche. l'Afrique de l'Ouest, peuvent avoir un effet inhibiteur
permanent sur les capacités de reproduction.
Au cours du cycle oestral, les variations des
Le présent article est consacré à l'étude de la durée
proportions des cellules et des polynucléaires de
du cycle oestral par détection des chaleurs et examen des
l'épithélium vaginal étaient maximum au cours du
frottis vaginaux chez la brebis Djallonké au Bénin.
dioestrus. L'examen de l'aspect des frottis vaginaux a
permis de déterminer la phase du cycle oestral.

The oestrus cycle of the West 2. Matériel et méthode

African Dwarfsheep
Le troupeau étudié dans le cadre de ces travaux est
Abstract gardé dans une ferme de l'Université nationale du Bénin
à Abomey-Calavi, localité située à une vingtaine de km
The oestrus cycle of lactating and non-lactating West de Cotonou (Bénin). Les animaux, des moutons de race
African Dwarfewes was monitored over a period ofone Djallonké, correspondent à la sous-race décrite par
year. Oestrus was detected by exposure to rams and Rombaut et Van Vlandaeren (1976) pour le sud de la
vaginal smears examination. Côte d'Ivoire.
The length ofmost oestral cycles determined by oestrus L'étude du cycle oestral a porté sur 4 brebis non
detection ranged from 16 to 19 days and were spread allaitantes et 12 brebis allaitantes âgées de 18 à 32 mois
over the whole year. Silent oestrus were more frequent et pesant 17 à 23 kg, et sur 4 agnelles âgées de 6 à 16
in the rainy season. mois dont les poids variaient entre 10 et 16 kg.
During the oestrus cycle, variability ofthe proportion of Au début de l'expérience, les animaux ont été
cells andpolynuclears ofthe vaginal epithelium reached vaccinés contre la peste des petits ruminants et le
a maximum at dioestrus. The stage of the oestral cycle tétanos, et vermifugés (helminthes et coccïdies). Des
was determined by examining the appearance of the traitements prophylactiques contre les parasites externes
vaginal smears. (tiques et gales) ont été réalisés plus fréquemment.
Toutes les brebis étaient alimentées sur pâturage naturel
et recevaient en outre des compléments de sous-produits
agro-industriels.
1. Introduction
Au cours des travaux, les animaux étaient pesés, et
Le cycle oestral chez la brebis est défini comme des frottis vaginaux étaient prélevés une fois par
l'intervalle de temps qui sépare deux oestrus (Berger et semaine. Des observations journalières (matin, midi,
Ginisty, 1980). Les données dont on dispose soir) du comportement sexuel des brebis en présence
actuellement sur ce paramètre sont essentiellement d'un bélier muni d'un tablier ont été effectuées. Les
basées sur des observations (Fall et coll., 1983; frottis vaginaux avaient été colorés par la technique de
Vallerand et Branckaert, 1975) ou sur des techniques Papanicolaou (1942). Ces expériences ont été réalisées
plus fiables telles que les dosages d'hormones au cours de la saison des pluies (mai-juillet) et des deux
(Yenikoye, 1984; Thibier, 1976). Très onéreuses, ces saisons sèches (février-avril et août-octobre) de la
dernières ne sont pas toujours accessibles aux petites même année.

Small Ruminant Research Network Workshop 53

M.S. Hounzangbé-Adoté

3. Résultats à l'autre (P>0,3) (tableau 1). Cependant, les écarts

entre les valeurs extrêmes de durée de cycle étaient
plus importants en saison sèche (P<0,01) qu'en saison
3.1 Durée du cycle oestral estimée par la des pluies. Aussi la comparaison des variances de ces
détection des chaleurs moyennes par le test F a-t-il révélé des différences
significatives (P<0,001) dans les fluctuations des
Les 90 cycles détectés sur les 112 cycles observés valeurs autour des moyennes selon la saison et l'âge.
se répartissaient en 57 cycles chez les brebis non La courbe de répartition de la fréquence des durées
allaitantes contre 33 cycles chez les brebis allaitantes. des cycles était normale. Ces fréquences, comprises en
L'intervalle entre deux oestrus était de 17,98 ± majorité entre 16 et 19 juin (figure 1), étaient un peu plus
0,66 jours avec des extrêmes allant de 6 à 28 jours; les étalées chez les brebis âgées de plus de 20 mois.
cycles se déroulaient sans interruption pendant toute Chez l'ensemble des brebis allaitantes, et quel que
l'année. soit leur âge, la durée moyenne des cycles oestraux était
Le test de chi-carré n'a pas fait apparaître d'effet de 17,75 ± 0,88 jours, une valeur comparable à celle de
significatif de l'état physiologique, de l'âge et des la durée (18,12 ±0,92 jours) des cycles des brebis non
saisons sur la durée des cycles oestraux (P>0,05) allaitantes. Par ailleurs, les chaleurs silencieuses étaient
(tableaux 1 et 2). moins fréquentes chez les brebis allaitantes (11%)
La comparaison des durées des cycles oestraux par que chez les brebis non allaitantes (26%) et plus
le test t de Student a révélé que les valeurs moyennes des fré-quentes en saison pluvieuse (53%) qu'en saison
durées du cycle oestral ne variaient pas d'une saison sèche (11%).

Tableau 1. Effet de la saison sur la durée du cycle oestral chez des brebis Djallonké

Durée moyenne Taux de chaleurs

Nombre de cycles (écart type) silencieuses
détectés (jours) Extrêmes (%)
Saison sèche Ensemble 78 18,22 6-28 11
des brebis 50 0,73
Brebis 50 18,22 6-28 20
cyclées 1,00
Brebis 28 17,9 6-20 0
allaitantes 1,2
Saison des pluies Ensemble 12 17,92 14-24 53
des brebis 1,38
Brebis 7 18,43 14-24 58
cyclées 2,11
Brebis 5 17,2 16-19 43
allaitantes 1,14

Tableau 2. Effet de l 'âge sur la durée du cycle oestral chez des brebis Djallonké
Taux de chaleurs
Nombre de cycles Durée moyenne silencieuses
Age (écart type) (jours) Extrêmes (%)
Plus de 20 mois Ensemble 46 17,91 6-28 18
des brebis 1,1
Brebis 29 18,41 6-28 20
cyclées 1,5
Brebis 27 17,1 6-20 11
allaitantes 1,5
Moins de 20 mois Ensemble 44 18,07 15-20 22
des brebis 0,66
Brebis 7 18,82 16-20 26
cyclées 0,97
Brebis 5 18,5 15-20 12
allaitantes 0,69

54 Small Ruminant Research Network Workshop

 Etude du cycle oestral chez la brebis Djallonké

Figure 1. Répartition des fréquences de durée de cycle cellules parabasales et polynucléaires (en
oestral selon l 'âge surabondance). On notait la présence de cellules
superficielles de type squame colorées en vert et
Fréquence regroupées en amas avec les autres types cellulaires.
14 Au cours des trois jours suivants, le frottis
| Age < 20 mois
présentait (figure 2E) les cellules colorées en vert ou
Ц Age > 20 mois
12 violet isolées les unes des autres avec des contours
réguliers. La différenciation des cellules à ce stade
n'était pas toujours aisée.
10
Phase 5
8
Pendant les deux ou trois jours précédant l'oestrus
il visible, le frottis présentait les mêmes images
6
cytologiques, mais était beaucoup plus pauvre en
cellules (figure 2F).

I ■ ■ L
Durée
à ■ ■ ■ ■
0 13 14 15 16 17 18 19 20 21 22 23 24 25
3.2.2 Pourcentages des cellules de répithélium
vaginal
Les cellules superficielles étaient plus ou moins
nombreuses (40 %) à l'apparition des premiers signes de
chaleur. Ce taux atteignait rapidement 50 à 60% deux à
cinq jours plus tard, pour chuter 24 heures après
l'obtention du pic (figure 3).
3.2 Modification des frottis vaginaux au Les pourcentages les plus faibles de cellules
superficielles (10-20%) s'observaient lors de la
cours du cycle oestral deuxième moitié du cycle et peu avant l'oestrus suivant.
Il y avait une forte corrélation (r = +0,83; [Link])
3.2.1 Aspect des frottis au cours du cycle entre l'indice karyopcnotique (IK) et le pourcentage de
cellules superficielles.
L'examen de l'aspect des frottis a permis de faire
les observations suivantes: La durée du cycle (19,67 ± 1,33 jours) déterminée
à partir de la cytologie vaginale semblait plus longue que
Phase 1 celle (17,958 ± 0,66 jours) déterminée par la détection
des chaleurs (P<0,01).
Le jour de l'apparition des chaleurs, les cellules
étaient nombreuses (parabasales ou intermédiaires) et
isolées les unes des autres. Les cellules intermédiaires
4. Discussion
(vertes ou violettes) étaient de grande taille et aussi Selon Vallerand et Brankaert (1975), le cycle
nombreuses que les cellules superficielles (rose pâle). oestral chez la brebis Djallonké dure de 16 à 17 jours.
Les polynucléaires étaient plus ou moins abondants Berger (1979) estime la moyenne à 17,4 jours avec des
(figure 2A). extrêmes allant de 16 à 19 jours. Les valeurs moyennes
observées ici correspondent à celles indiquées par ces
Phase 2 auteurs. En effet, dans 71% des cas, elles se situaient
dans l'intervalle de 16 à 19 jours.
72 heures après le début des chaleurs, le frottis
contenait beaucoup de grandes cellules superficielles; Les fluctuations de la durée du cycle oestral chez
certaines étaient étalées, d'autres (25-50%) kératinisées la brebis âgée de plus de 20 mois peuvent être attribuées
ou plicaturées. Les polynucléaires étaient beaucoup moins à l'âge de l'animal d'une part et à la fréquence des
abondants qu'au premier jour de l'oestrus (figure 2B). mises-bas d'autre part. En effet, ces femelles étaient
saillies tôt (première mise-bas à 16 mois) et avaient un
Phase 3 rythme de deux agnelages par an pendant environ trois
ans, ce qui a entraîné une fatigue physiologique. A
Au cours des trois jours suivants, l'éosinophilie celle-ci s'ajoute l'insuffisance alimentaire à laquelle
augmentait (figure 2C) et la quasi-totalité des spasmes étaient exposées les brebis avant le début des
colorés étaient cornifiés. La kératinisation était bien expériences, la restriction alimentaire pendant la période
visible et le nombre de polynucléaires avait augmenté. de croissance ayant un effet dépressif sur les capacités
de reproduction des animaux.
Phase 4
Les écarts entre les valeurs extrêmes de la durée du
Comme l'indique la figure 2D, les cellules cycle oestral étaient plus importants en saison sèche
parabasales et intermédiaires basophiles ont réapparu. qu'en saison des pluies. Au cours de cette saison là, les
Pendant environ cinq jours, le frottis était riche en conditions plutôt défavorables d'élevage (manque d'eau

Small Ruminant Research Network Workshop 55

M.S. Hounzangbé-Adoté

Figure 2 A.

III- ft,

V
•Л - - .v !»Л

* D——
*
r. •* .t

-C ' *
|H m
Figure 2B.

С—

ч
(2)

et d'herbe fraîche) ont pu perturber la biologie des lumière se fait par l'intermédiaire de l'hypophyse via
reproductrices. Yenikoye (1984) a montré qu'entre l'hypothalamus. Ce complexe hypotalamo-hypo-
décembre et avril (saison sèche relativement fraîche), la physaire contrôle l'apparition de l'oestrus. Thimonieret
durée des cycles était de 31 jours, avec des ovulations Mauleon ( 1 969) ont confirmé que la diminution des taux
silencieuses. de gonadotrophines pendant les jours courts et à faible
La fréquence élevée des chaleurs silencieuses en ensoleillement pouvait bloquer ou diminuer les activités
saison des pluies pourrait s'expliquer par la variation de ovariennes chez la brebis, entraînant des anoestrus vrais
l'éclairement : bien que faible dans les pays tropicaux, ou des chaleurs silencieuses.
celle-ci diffère néanmoins d'une saison sèche à l'autre. Dans les pays tropicaux à climat humide, le degré
Il est établi que l'activité des gonades induite par la hygrométrique varie entre la saison sèche et la saison

56 Small Ruminant Research Network Workshop

 Etude du cycle oestral chez la brebis Djallonké

Figure 2C.
•

f
e

(3)

Figure 2D.

humide. L'humidité au sol en saison des pluies peut L'étude de la transformation périodique de
également influencer indirectement la physiologie de la l'épithélium vaginal en fonction du cycle oestral a fait
brebis. apparaître des modifications assez diffuses mais
Chez les brebis allaitantes, le nombre de chaleurs néanmoins sensibles.
silencieuses diminue considérablement après la reprise Comme dans tout cycle oestral, les cellules
de l'activité sexuelle. L'anoestrus post-partum étant un superficielles sont sensibles aux oestrogènes, ce qui se
repos sexuel, la brusque décharge des facteurs de traduit par une éosinophilic intense de ces cellules,
libération de l'hypothalamus augmenterait l'activité tandis que l'abondance des polynucléaires correspond
oestrale et rendrait les brebis plus aptes à l'acceptation souvent à la mise en place d'une phase lutéale. La
du bélier. disparition partielle des polynucléaires s'explique par le

Small Ruminant Research Network Workshop 57

M.S. Hounzangbé-Adoté

Figure 2E.
I

(5)
уТ-

Figure 2F.

d—

$9 ^•^

•*
•**

§ ¡ei

faible taux de progestérone au moment du pic des Les résultats relatifs aux frottis sont conformes aux
cellules superficielles. descriptions cytologiques faites par Sanger et coll.
Sur la base de ces données et des signes cyto- (1958) et Thibault (1971) pour la brebis.
logiques (indices d'éosinophilie, indice karyopcnotique Chez la brebis, il semblerait que les premiers
et indice des polynucléaires), on peut déduire que les signes de chaleur, qui durent de 24 à 72 heures,
quatre phases observées correspondent aux quatre apparaissent avec le début du proestrus, c'est-à-dire au
phases du cycle oestral définies par Galet (1980), à moment de la réapparition des oestrogènes consécutive
savoir la phase 1 ou proestrus (durée 4-5 jours), la phase à la croissance folliculaire. L'oestrus proprement dit
2 c'est-à-dire oestrus et le postoestrus (72 heures), la s'étale sur une période de 72 heures au cours de laquelle
phase 3 ou métoestrus (durée 8 jours) et la phase 4 ou a lieu l'ovulation. Celle-ci se produit dès la fin des signes
dioestrus (durée 3-4 jours). apparents de chaleurs.

58 Small Ruminant Research Network Workshop

 Etude du cycle oestral chez la brebis Djallonké

Figure 3. Variations des pourcentages moyens des cellules Doney J.M. et Gunn R.G. 1981. Nutritional and other factors
superficielles (CS) (les traits verticaux indiquent in breeding performance of ewes. In: Environmental
l 'erreur standard de la moyenne; le jour D est factors in mammalian reproduction. Publié sous la
celui du début des chaleurs). direction de D. Gilmore et В. Cook. London McMillan
Pourcentage de cellules vaginales Pub. Ltd, Londres (R.-U.). p. 169 à 172.
Fall A., Diop M., Sandford J., Gueye E., Wissocq Y.J., Durkin
J. et Trail J.C.M. 1983. Etude sur la productivité de
moutons Djallonké au Centre de recherches
zootechniques de Kolda, au Sénégal. 2. Poids corporels,
productivité des brebis et du troupeau. Revue d'élevage
et de médecine vétérinaire des pays tropicaux
36:283-289.
Galet M. 1980. La maîtrise des chaleurs par la méthode des
synchronisations de l'oestrus chez les ovins. Rapport
annuel 80-81. Document multicopié. Centre national
d'insémination artificielle, Bingerville (Côte d'Ivoire).
Haresign W. 1984. Underfeeding and reproduction:
physiological mechanism. In: Reproduction des
ruminants en zone tropicale. Publié sous la direction de
P. Chemineau, D. Gauthier et J. Thimonier. Colloque
international sur la reproduction des ruminants en zone
tropicale, Pointe-à-Pitre (Guadeloupe), 8-10 juin 1983.
Les Colloques de l'INRA 20:339-365. INRA (Institut
012 345678 9 10111213141516 0 national de la recherche agronomique), Paris (France).
Jours Mauleon P. 1972. Cours approfondi d'amélioration génétique
Robinson et Moore (1956) cités par Mauleon des animaux domestiques: Physiologie de la
reproduction. Station de physiologie de la reproduction
(1972) signalent que les polynucléaires sont peu
de l'INRA/Tours, Nouzilly (France).
nombreux les jours qui suivent l'oestrus mais restent
présents pendant tout le cycle chez la brebis. Papanicolaou G.N. 1 942. A new procedure for staining vaginal
smears. Science 95: 438-439.
Rombaut D. et Van Vlandaeren G. 1976. Le mouton Djalonké
Conclusion de Côte-d'Ivoire en milieu villageois: comportement et
alimentation. Revue d'élevage et de médecine
Les valeurs moyennes de la durée du cycle oestral vétérinaire des pays tropicaux 29(2):157-172.
déterminées par la détection des chaleurs au cours de nos Sanger V. L., Engle P. N. et Bell D.S. 1958. Cytologie vaginale
travaux correspondent à celles indiquées par la de la brebis. American Journal of Veterinary Research
bibliographie. 19:282-289.
Malgré quelques différences en ce qui concerne les
Thibault C. 1971. Cours approfondi de Physiologie de la
modifications de l'épithélium, les résultats observés sur
reproduction : 1 . Analyse des contenus hypophysäres et
les frottis vaginaux étaient cohérents et conformes à plasmiques en hormones gonadotrope. 2. Variation
ceux enregistrés par d'autres chercheurs avec des cyclique de l'épithélium vaginal. Notes de cours.
dosages hormonaux. De fait, il suffit d'un seul frottis Université Paris VI, Paris (France).
pour savoir si la brebis se trouve en période pré- ou post
Thibier M. 1 976. Le cycle sexuel de mammifères domestiques.
ovulatoire. Des travaux supplémentaires sont
II. Etude de la régulation du cycle sexuel. Revue
néanmoins nécessaires pour déterminer le degré de d'économie et de médecine animale 17:135-137.
fiabilité de cette méthode, notamment en période
ovulatoire. Thimonier J. et Mauleon P. 1969. Variation saisonnière du
comportement d'oestrus et des activités ovariennes et
hypophysaires chez les ovins. Annales de biologie
animale, biochimie, biophysique. 9:233-250.
Références Vallerand et Branckaert, 1975. La race ovine Djallonké au
Berger Y. 1979. Sélection et amélioration des ovins de Côte Cameroun. Potentialités zootechniques, conditions
d'Ivoire. Rapport annuel n° 5. Centre de recherches d'élevage ovin. Revue d'élevage et de médecine
zootechniques, Bouaké (Côte d'Ivoire). vétérinaitre des pays tropicaux, 28:523-545.
Berger Y. et Ginisty L. 1980. Bilan de 4 années d'étude de la Yenikoye A. 1984. Variations annuelles du comportement
race ovine Djallonké en Côte d'Ivoire. Revue d'élevage d'oestrus du taux et des possibilités d'ovulation chez la
et de médecine vétérinaire des pays tropicaux brebis peulh du Niger. Reproduction nutrition
33(l):71-78. développement. 24:1 1-19.

Small Ruminant Research Network Workshop 59

 Puberty and early mating performance in subtropical
fat-tailed sheep and their crosses
11 2
Ferial Hassan , M.T. Mousa , A.M. Aboul-Naga ,
F. El-Hommosi and G. Abd El-Hafez

Animal Production Research Institute, Ministry ofAgriculture, Dokki, Cairo, Egypt

Agricultural Counsellor, Embassy ofEgypt in Rome, Via Salaria 267, 00199 Rome, Italy
Faculty ofAgriculture, University ofAssiut, Assiut, Egypt

Abstract Introduction
In a crossbreeding trial of both Ossimi (O) and Awassi The Mediterranean subtropical Chios (C) sheep are
(A) with the highly prolific Chios (C) subtropical known for their high prolificacy, early sexual maturity
fat-tailed sheep, crossbred ewe lambs exhibited their and good milk production. Their usefulness for
first oestrus at a significantly (P<0.05) younger age than improving sheep production in the subtropics where
the purebred parents. They also had a higher (P<0.05) prolific temperate breeds performed poorly as purebreds
conception rate at yearling. C ewe lambs exhibited their has been reported (Aboul-Ela and Aboul-Naga, 1987).
first oestrus at the youngest age (230 days) and had the A crossbreeding programme was carried out by the
highest conception rate at one year of age. Season of Ministry of Agriculture in Egypt between the imported
lambing had little effect on age at puberty of either C sheep from Cyprus and both Ossimi (O) and Awassi
purebred or crossbred ewe lambs. (A) sheep. Production performance of these breeds and
Crossbred ram lambs performed better in pubertal their crosses has been described by Mousa (1991). The
age than purebred parents. Semen quality at first present study reports on the puberty and early mating
performance of the crosses and compares it with that of
ejaculation had poor quality, while it was good at 80%
their purebred parents.
semen motility score.

Puberté et premières Materials and methods

performances de reproduction An experiment was conducted at Mallawi Research
chez des moutons à queue Farm, Ministry of Agriculture, Upper Egypt between
January and September 1987. 0 and A ewes were mated
grasse des régions subtropicales to C rams to produce the Fi CO and CA lambs together
et chez les produits de leur with their parental groups. Lambs were kept with their
dams up to weaning at eight weeks of age. There was a
croisement gradual introduction of pelleted concentrate mixture
plus Berseem hay into their diets. From December till
Résumé May, the flock grazed on Egyptian clover (Trifolium
alexandrinum). In the summer the flock grazed on crop
Dans des essais de croisement entre des moutons residues and green fodder when available. Grazing was
de race Ossimi (O) et Awassi (A) avec des moutons à supplemented with pelleted concentrate mixture.
queue grasse très prolifiques de race Chios (С), les Puberty performance was studied in both ewe and
agnelles métisses avaient leurs premiers oestrus à un ram lambs born in the lambing season of May/June 1987
âge significativement moins avancé (P<0,05) que leurs and in ewe lambs born in January/February 1988,
parents de race pure, et un taux de conception à l'âge starting from four months of age. Behaviourial oestrus
de 1 an également plus élevé. Les agnelles Chios was detected by active rams twice daily at 800 and 1600
affichaient l 'âge au premier oestrus le plus précoce et hours. The first incidence of heat symptoms was
le taux de conception à l 'âge de 1 an le plus élevé. La recorded and the lambs were weighed. Conception rate
saison d'agnelage avait très peu d'effet sur l'âge à la among yearlings was tested by putting a fertile ram with
puberté des agnelles de race pure ou croisées. each group of ewe lambs for a period of 35 days.
Les performances à la puberté des agneaux mâles The ram lamb's ability to ejaculate in an artificial
croisés étaient supérieures à celles de leurs parents de vagina was considered as a puberty criterion. The date
race pure. Le premier éjaculat était de qualité médiocre of first ejaculation was recorded and the lambs were
mais présentait une bonne note sur le plan de la weighed. Semen samples from the ram lambs in each
motricité du sperme (80%). breed group were used to evaluate the physical

Small Ruminant Research Network Workshop 61

Ferial Hassan et al

properties of semen at first ejaculation. Semen motility or A ewe lambs (P<0.01). The crossbred yearling ewes
was examined weekly until it reached a score of 80% of had significantly (P<0.05) higher conception rates than
the age of the ram lambs' and other physical properties О and A, but lower conception rates than their parent
of the semen were also recorded. (C). The difference between С and CO crosses was not
statistically significant.

Table 2. Conception rate and body weight of yearling ewe

Results lambs in the breed groups.
No. of Body weight
Ewe Iambs Breed ewe Conception
group lambs rate% x(kg) ±SE
Puberty performance 57.1b
Ossimi (0) 28 37.5 1.6
The breed of the ewe lambs had a highly significant 44.0c
Awassi (A) 25 34.7 1.7
effect on age and weight at puberty (P<0.01). О ewe
lambs reached puberty at younger ages and heavier Chios (C) 27 74.1a 38.2 1.4
weights (P<0.05) than A ewes (Table 1). C ewe lambs, 65.4a
however, reached puberty at a younger age than both О CO 26 39.2 1.6
and A ewe lambs (P<0.01). CO ewe lambs exhibited CA 32 59.3b 37.4 1.5
their first oestrus significantly earlier than the О ewes
(P<0.01), but at a similar weight. The CA ewe lambs Values within the same column with different superscripts differ
reached puberty significantly younger than A ewes significantly (P<0.05).
(P<0.01) and at slightly heavier weights.
Single and triplet ewes lambed at a slightly older
age than twins, but the differences were not statistically Ram lambs
significant (Table 1).
Ewe lambs born in May/June reached puberty a First ejaculation
few days later than those born in January/February, and
were slightly heavier (Table 1). The two CO and CA crossbred ram lambs performed
their first ejaculation at a slightly older age and heavier
Yearling fertility weight than their A and О parents (P<0.05). They
showed better performances in pubertal weight and age
Conception rate was highest amongst yearlings, lowest than C ram lambs (Table 3). C ram lambs reached their
amongst A, and О yearlings were intermediate (Table first ejaculation at the oldest age and lightest weight of
2). C ewe lambs had a better performance than either О all the groups studied, possibly due to slow growth rates.

Table 1. Mean age in days and body weight in kg at first oestrus for ewe lambs in the breed groups.
Age Body weight
Classification No.
Cons. (day) ±SE Cons, (kg) ±SE
Overall means 138 275.3 17.2 32.0 1.3
Breed groups
Ossimi (0) 28 12.9 0.4d 1.3 0.0
Awassi (A) 25 40.2 2.0b 1.4 0.2
Chios (C) 27 -45.0 3.8a -1.8 0.7
CO 26 -7.8 1.3е 1.2 0.1
CA 32 -0.0 0.1" 1.0 0.0
Type of birth
117 5.1 12.3a 1.1 0.9
Singles
Twins 19 -10.8 1.2b -1.0 0.1
19.8a -0.0 1.4
Triplets 2 -5.8
Seasons
May-June 75 19.4 1.9 1.0 0.1
Jan-Feb 63 -6.6 2.4 -1.0 0.2
Values within the same column with different superscripts differ significantly (P<0.05).

62 Small Ruminant Research Network Workshop

 Puberty and early mating performance in subtropical etc..

Variations in pubertal age within each group were high, Performance at 80% semen motility
except for C ram lambs.
As the age of the ram advanced, semen motility
increased to 80%. The crossbred ram lambs reached
Table 3. Age and body weight of ram lambs at first 80% semen motility at a younger age than purebred
ejaculation.
parents (Table 5 and Figure 1) and were also heavier.
Age at puberty Weight at puberty Most of the semen samples were creamy or whole milk
Breed in appearance. The pH values at first ejaculation were
group No. high, and gradually declined in alkalinity. Differences
(day) + SD mean (kg) + SE
Ossimi (0) 14 296.8 35.4al 41.1 6.6a in percentage of dead sperms between breeds were not
48.5a 38.7 5.9a significant. Variations in the percentage of abnormal
Awassi (A) 14 284.6 spermatozoa were generally insignificant, except for CA
Chios (C) 17 334.9 45.3b 35.4 1.9b
ram lambs (P>0.05). Sperm cell concentration per
CO 15 318.5 39.7ab 51.7 5.3d ejaculation was high in C, CA and CO ram lambs (Table
24.4a 44.7 5.5d 5), probably due to their high semen volumes.
CA 16 286.9
Estimates followed by the same letter in any column do not differ
significantly (P>0.05).
Discussion and conclusions
Semen physical properties Most subtropical breeds reach puberty late (Osman,
1985), possibly due to low growth rates (Aboul-Ela and
Table 4 shows the analysis of semen samples from first Chemineau, 1989). However, C ewe lambs attain their
ejaculation at puberty. C rams showed the lowest puberty at about 225 days (Fox et al, 1 97 1 ), which agrees
motility score while CA rams had the highest. The other with the present study. The results of this study show
breeds had intermediate scores. Most of the other that crossbreeds between C and О and C and A reached
properties followed the same trend as the motility scores. puberty earlier than purebred О and A sheep. Similar
Both CO and CA crossbreeds had high values for semen reports have showed that heterosis may contribute to
volume, percentage of live sperm and total number of earlier sexual development (Land, 1978; McMillan and
sperms per ejaculation. C ram lambs had the poorest McDonald, 1983; Lahlou-Kassi and Marie, 1985).
semen quality. However, season of birth seems to have less influence

Table 4. Physical properties of semen at first ejaculation.

Breed Volume Motility pH Dead Abnormality Sperm Con.

group No. (ml) % (%) sperm % x 106/mm3
Ossimi (0) 5 50a 7.5a 28.0a 11.6a 2.0
0.6
Awassi (A) 5 0.8b 50a 7.3a 34.0a 12.8a 1.5
Chios (C) 5 0.8b 45a 7.5a 24.0a 9.8a 1.3
CO 5 0.8ab 50a 7.2a 25.0a 8.0a 1.9
CA 5 0.9b 55a 6.9a 21.0a 7.5a 1.9
ab = Estimates followed by the same letter in any column do not differ significantly (P>0.05).

Table 5. Semen physical properties, age and body weight at 80% semen motility score.

Dead Sperm Age Body 'weight

Breed Volume sperm Abnormality Con. Sperm Con.
group No. (ml) рн % % xl06/mm3 xl07ejac
x(day) ± SE x(day) + SE
Ossimi (O) 5 1.5" 6.8a 16.6a 7.4a 3.2 4.6a 414.2 19.2a1 51.4 4.2a

Awassi (A) 5 1.2a 6.8a 17.2a 6.8a 4.2a 401.7 19.6a 45.8 4.3a
3.5
Chios (C) 5 2.0a 6.7a 18.8a 6.2a 2.7 5.3b 425.6 18.7a 48.8 4.1a

CO 5 1.9a 6.7a 14.6a 6.4a 2.5 4.8b 393.3 20.2b 55.9 4.6a

CA 5 2.0a 6.8a 13.5a 4.3b 2.7 5.3b 398.6 20.5ab 51.1 4.5a

Overall
1.7a 6.8a
mean 16.4 6.2 2.9 4.8

ab = Estimates followed by the same letter in any column do not differ significantly (P>0.05).

Small Ruminant Research Network Workshop 63

Ferial Hassan et al

Figure 1. Semen motility changes from puberty up to score References

of 80% for different breed groups.
Aboul-EIa M В and Aboul-Naga A M. 1987. Improving
Semen motility % fecundity in subtropical fat-tailed sheep. In: Fayez I,
100 Marai M and Owen J В (eds), New techniques in sheep
production. Butterworths, London, UK. pp. 163-171.
90
Aboul-EIa M В and Chemineau P. 1989. Seasonality of
80 reproductive activity in native sheep and goat breeds and
their crosses with introduced breeds. In: Aboul-Naga A
70 H M (ed), Small ruminants research and development in
the Near East. Proceedings of a workshop held in Egypt,
60 Cairo, 2-4 November 1988. IDRC (International
Development Research Centre), Ottawa, Canada. pp.
50 74-87.
О
40 Aboul-Naga A M, Ashmawy G and Shalaby T H. 1987. Early
С
breeding of subtropical ewe lambs. International Goat
30 CA and Sheep Research. (In press)
А
20 Fox C W, Chaaban R, Hardison W A, Badawi S M, Choueri
CO E and Maalouf M. 1971. Selection and breeding
10 programs of Awassi and Chios Sheep, production
characteristics of Awasssi and Chios yearling ewes.
1 I I I I I I I I I I I I Г Magon, Lebanon. 1971.
0 1 2 3 4 5 6 7 8 9 10 11 12 13
Lahlou-Kassi A and Marie M. 1985. Sexual and ovarian
function of the D'man ewe. In: Land R В and Robinson
D W (eds), Genetics of reproduction in sheep.
on age at puberty of subtropical breeds, compared to its Butterworths, London, UK. pp. 245-260.
effect on their body weight. Results obtained by Younis Land R B. 1978. Reproduction in young sheep: some genetic
et al (1978) support these findings. and environmental sources of variation. Journal of
Early breeding of subtropical sheep can result in Reproduction and Ferttility 52:427^136.
lower conception and twinning rates and higher lamb
losses (Younis et al, 1978; Aboul-Naga et al, 1987). Lysandrides P. 1981. The Chios sheep in Cyprus. World
This showed that C, CO and CA yearling ewes had better Animal Review 39: 1 2-16.
early breeding performances than the О and A yearling Mann T. 1964. In the biochemistry of semen and of the male
ewes (Table 2). C sheep are expected to infuse the reproductive tract. Methuen and O. hid., London; John
advantage of early breeding to their crosses. Willy and Sons Inc., New York.
The two crossbred ram lambs reached puberty at McMillan W H and McDonald M F. 1983. Reproduction in
younger ages and heavier weights than С ram lambs. C ewe lambs and its effect on 2-year-old performance.
ram lambs seem to be adversely affected by the New Zealand Journal of Agricultural Research
environment in Upper Egypt, resulting in slower growth 26:437^42.
rates than those reported in their home country
Mousa M T. 1991 . Effect ofcrossing Ossimi, Awassi and Chios
(Lysandrides, 1981). The evidence does not show
sheep on some productive traits. PhD thesis, Assuit
crossbreeds having better semen properties at puberty. University, Assuit, Egypt.
However, by the time the sperm had reached 80%
motility, ejaculation volumes in the two crossbreeds and Osman H A. 1985. Near East sheep breeding and
C ram lambs were significantly higher (P<0.05) than in improvement. World Animal Review 54:2-15.
both О and A ram lambs. These differences may be due Younis A A, El-Gaboory I A, El-Tawil E A and El-Shobokshy
to the variation in testicular size (Mann, 1964). This A S. 1978. Age at puberty and possibility of early
study shows that crossbreeding with the highly fertile breeding in Awassi ewes. Journal of Agricultural
Chios subtropical breed can considerably improve the Science (Cambridge) 90:225-260.
early breeding performances.

64 Small Ruminant Research Network Workshop

 Preweaning performance of Yankasa sheep under
semi-intensive management
O.A. Osinowo, B.Y. Abubakar, M.E. Olayemi, R.O. Balogun, O.S.
Onifade, A.A. Adewuyi, A.R. Trimnell and F.O. Dennar

National Animal Production Research Institute (NAPRI)

Ahmadu Bello University, Zaria, Nigeria

Summary Introduction
Preweaning growth traits and weaning rate of Yankasa
The preweaning traits of sheep which influence flock
sheep under a semi-intensive management system with productivity include birth weight, weaning weight,
twice-yearly Iambings were studied for nine years at average daily gain to weaning and weaning rate. These
Shika, in the subhumid zone of Nigeria. traits are in turn influenced by genetic and
Least squares means (± SE) for birth weight (BW), environmental factors. For a given breed, influencing
90-day weaning weight (WW), average daily gain to environmental factors may include dam parity and type
weaning (ADG) and weaning rate (WR) were 2.51 ± of birth and sex, month and year of birth of young
0.01 kg, 10.87 ± 0.08 kg, 91.86 ± 0.91 g/day and 77.8 ± (Terrill, 1966).
1.0%, respectively. BW, WW and ADG were The Yankasa sheep is the most numerous breed of
significantly (P<0.001) affected by dam parity and the sheep in Nigeria and also has the widest distribution,
lamb's litter size, sex, month and year of birth. WR was being found throughout the subhumid and semi-arid
significantly (P<0.01) affected by parity, litter size, zones (FDLPCS, 1991). It is estimated to constitute
month and year of birth but not by sex of lamb (P>0.05). about 60% of the national sheep population of 22.1
Preweaning growth performance was generally better in million (Osinowo, 1992). Preweaning traits have been
the wet seasons than in the dry seasons while the reverse studied in Yankasa sheep (Adu et al, 1979; Adu and
was true for weaning rate. Buvanendran, 1982; Taiwo and Buvanendran, 1985;
Hassan, 1987). However, these earlier studies were
carried out with smaller data sets of less than four
Performances présevrage du hundred records.
This study re-examined preweaning traits of
mouton Yankasa élevé en Yankasa sheep under semi-intensive management at
système semi-intensif Shika in the subhumid zone of Nigeria, using a much
larger data set collected over a nine-year period.
Résumé
Les paramètres de croissance présevrage et les
taux de sevrage du mouton Yankasa élevé selon un mode Materials and methods
semi-intensif, avec deux agnelages par an, ont été
étudiés sur une période de neufans à Shika, dans la zone The Yankasa sheep flock from which the data were
subhumide du Nigéria. obtained was located at the National Animal Production
Les moyennes des moindres carrés (± erreur type) Research Institute (NAPRI), Ahmadu Bello University,
du poids à la naissance, du poids au sevrage à 90jours, Shika, Nigeria, latitude 1 1°12'N, longitude 7°33'E and
610 metres above sea level (masl). The climate is
du gain moyen quotidien (GMQ)jusqu 'au sevrage et du
subhumid and the vegetation zone is Northern Guinea
taux de sevrage étaient respectivement de 2,51 ± 0,01
Savanna. Mean annual rainfall and temperature
kg, 10,87 ± 0,08 kg, 91,86 ± 0,91 g/jour et 77,8 ± 1,0%. recorded at Samaru, 10 km from Shika, were 1 107 mm
La parité de la mère, la taille de la portée, le sexe des and 24.4°C, respectively. The seasonal distribution of
jeunes, et le mois et l 'année de naissance avaient un effet rainfall is approximately 0.1% in the late dry season
significatif (P<0,001) sur les poids à la naissance et au (January - March), 25.8% in the early wet season (April
sevrage et sur le GMQ. A l'exception du sexe des -June), 69.6% in the late wet season (July - September)
agneaux (P>0,05), ces mêmes paramètres avaient aussi and 4.5% in the early dry season (October- December).
un effet significatif(P<0,01) sur le taux de sevrage. Les Fodder is generally scarce and of low quality in the late
performances de croissance avant le sevrage étaient dry and early wet seasons. It is in abundant supply and
généralement meilleures en saison des pluies alors que of relatively good quality in the late wet and early dry
l'inverse était vrai pour le taux de sevrage. seasons.

Small Ruminant Research Network Workshop 65

O.A. Osinowo et al

The origin and management of the Yankasa sheep The animals were housed in well ventilated pens
flock studied have been previously described (Osinowo, overnight. The ewes were routinely bred to lamb twice
1982; Osinowo and Ekpe, 1985). In 1987, the flock was a year under an accelerated lambing programme. The
enlarged from 100 to 300 ewes by additional purchases lambs were weaned at about 90 days of age.
from livestock markets in the same North Western Data on birth weight (BW), weaning weight (WW)
region of Nigeria. The sheep were managed under a and weaning rate (WR) were collected over nine years,
semi-intensive system. This involved grazing on from 1983 to 1991. Preweaning average daily gain
improved and sown pastures for 6 - 8 hours daily, with (ADG) was calculated for each lamb after age-
0.3 - 0.5 kg/day of a 15-20% crude protein concentrate correction of WW to 90-day weight. Factors included in
supplement throughout the year, depending on the the study were parity of dam (1-6) and lamb litter size
animal's physiological status (pregnant, nursing or dry). (single vs twin), sex (male vs female), month of birth

Table 1. Effects of some production and environmental parameters on the preweaning performance of Yankasa sheep.

Birth weight (BW) Weaning weight (WW) Average daily gain

(kg) (kg) to weaning Weaning
TRAITS (ADG) (g/d) rate
LSM ± SEM 2.507 ±0.011 10.874 ±0.084 91.861 ±0.911 0.779± 0.010
No. of observations 2020 1203 1203 1576
Least-squares constants
Variable No. No. No. No.
Parity 1 589 -0.179 320 -1.194 320 -11.159 440 -0.045
2 461 -0.048 260 -0.529 260 -5.253 359 -0.002
3 369 0.042 247 0.077 247 1.012 307 0.053
4 237 0.033 157 0.565 157 5.918 185 0.074
5 155 -0.002 103 0.363 103 4.174 119 0.097
6 87 0.124 56 0.012 56 -1.506 73 -0.018
7 59 0.050 26 -0.254 26 -4.716 39 -0.083
8 38 0.005 19 0.437 19 4.759 29 -0.010
9 30 -0.025 15 0.523 15 6.771 29 -0.066
Litter size
Single 1385 0.207 889 1.160 889 10.618 1120 0.066
Twin 635 -0.207 314 -1.160 314 -10.618 456 -0.066
Sex Male 1004 0.058 604 0.523 604 5.199 781 0.007
Female 1016 -0.058 599 -0.523 599 -5.199 795 -0.007
Month Jan 218 -0.032 154 -0.970 154 -10.812 190 0.026
Feb 123 -0.249 100 -1.743 100 -16.445 110 0.033
Mar 200 -0.304 145 -0.236 145 0.068 176 0.007
Apr 161 -0.089 103 0.165 103 2.538 132 0.017
May 234 -0.032 123 0.947 123 11.404 170 0.079
Jun 215 0.111 78 -0.285 78 -5.398 149 -0.132
Jul 192 0.260 97 1.197 97 10.934 130 -0.025
Aug 146 0.150 70 0.997 70 9.308 130 -0.114
Sep 99 0.138 66 0.934 66 8.731 86 -0.035
Oct 100 0.014 68 0.048 68 0.277 81 -0.033
Nov 129 0.070 86 -0.655 86 -6.715 96 0.159
Dec 203 -0.037 113 -0.393 113 -3.890 163 0.180
Year 1 983 86 -0.077 41 -0.160 41 -1.490 48 0.132
1984 197 -0.078 158 0.228 158 3.587 167 0.123
1985 201 0.068 154 -0.024 154 -0.096 176 0.143
1986 128 0.087 95 -0.440 95 -5.622 104 0.088
1987 210 -0.072 86 1.440 86 15.809 108 -0.034
1988 453 -0.064 236 -1.139 236 -12.355 376 -0.166
1989 249 -0.173 138 -0.656 138 -6.088 196 -0.125
1990 308 -0.062 167 -0.403 167 -3.295 231 -0.078
1991 188 0.371 128 1.154 128 9.550 170 -0.083

66 Small Ruminant Research Network Workshop

 Preweaning performance of Yankasa sheep under etc..

Table 2. Comparison of estimates of preweaning and weaning performance of Yankasa sheep by different investigators.
Trait No. of observations Mean ± SE Investigators
Birth weight (kg) 214 2.80 ±0.19 Aduetal(1979)
43 2.54 Adu and Buvanendran (1982)
357 2.51 ±0.03 Hassan (1987)
2020 2.51 ±0.01 Current study
Weaning weight (kg) 32 9.14. Adu and Buvanendran (1982)
298 10.90 ±0.19 Hassan (1987)
1203 10.87 ±0.08 Current study
Preweaning ADG (g/day) 214? 107.8 Adu et al (1979)
32 75.8 Adu and Buvanendran (1982)
1203 91.86±0.91 Current study
Weaning rate (%) 43 76 Adu and Buvanendran (1982)
362 85 Taiwo and Buvanendran (1983)
1576 77.9 ±1.0 Current study

(January - December) and year of birth (1983 - 1991) Weaning weight (Table 1)
of the lambs. After editing, 2020 BW, 1203 WW, 1203
ADG and 1576 WR records were included in the The 90-day WW for Yankasa sheep in this study agreed
analyses. closely with that reported by Hassan (1987) (Table 2)
Environmental effects on the preweaning traits but both are higher than the earlier estimate by Adu and
were examined by least squares analyses (SYSTAT, Buvanendran (1982). The effects of parity, litter size,
1989) according to the linear model : sex, month and year of birth were highly significant
(P<0.001).
Yijklmn = u + Pi + Lj + Sk + Ml + Ym + Eijklmn
WW increased consistently from first to fifth
where Yijklmn was the preweaning trait (BW, WW, parity, and both singles and male lambs had a higher
ADG or WR) of an individual lamb, u was the overall WW than twins and female lambs. WW of lambs born
mean, Pi was the fixed effect of the ith parity, Lj that of between November and March and also in June were
the jth litter size, Sk that of the kth sex, Ml that of the 1th below average, with lambs born in July having the
month of birth, Ym that of the mth year of birth and highest WW. This pattern reflects the seasonal
Eijklmn was the random error term associated with each availability of forage, except for June. The low WW of
record. lambs born in June could be due to a high disease
incidence during this month; June also had the highest
preweaning mortality rate. This observation needs
Results and discussion closer clinical investigation to determine the cause.
Yearly variations followed no particular pattern except
for the consistent improvement in WW in the last four
Birth weight years of the study.
The least squares mean for BW (Table 1) agreed closely
with earlier reports (Adu and Buvanendran, 1982;
Hassan, 1987) but is lower than the estimate by Adu et Preweaning ADG (Table 1)
al (1979), as shown in Table 2.
The effects of dam parity and litter size, sex, month The preweaning ADG obtained in this study is
and year of birth of lambs on BW were all highly intermediate between values reported by Adu et al
significant (P<0.001). BW tended to increase with ( 1 979) and Adu and Buvanendran (1982). The effects of
parity, with the highest BW occurring at the 6th parity. parity, litter size, sex, month and year of birth on ADG
As expected, singles and male lambs weighed more at were highly significant (P< 0.001). ADG improved
birth than their twins and female counterparts. BW was consistently from first to fourth parity. Singles and male
higher in June-November than in Dec-May. BW lambs had a higher ADG than twins and female lambs.
followed the seasonal fluctuation in forage availability. Below average ADG were obtained in June and between
Yearly variations in BW did not follow any particular November and February. Yearly variations in ADG
pattern except for a consistent improvement over the last followed no pattern except for a consistent increase over
three years of the study. the last four years of the study.

Small Ruminant Research Network Workshop 67

O.A. Osinowo et al

Weaning rate (Table 1) References

The WR obtained in the present study compares Adu I F and Buvanendran V. 1982. Preweaning performance
favourably with that of Adu and Buvanendran (1985) of lambs from pure and crossbred matings among
but is lower than the figure reported by Taiwo and Nigerian breeds of sheep. World Review of Animal
Buvanendran (1985) (Table 2). WR was significantly Production 18:73-77.
affected by litter size, month and year of birth (all
Adu I F, Brinckman W L and Kuteyi I S. 1979. Reproductive
P<0.001) and parity (P<0.01), but not by sex (P>0.05). performance of indigenous sheep and their crosses.
WR increased consistently from first to fifth parity, with Nigerian Journal ofAnimal Production 6:38-40.
higher parities giving below average values. Single
lambs had a higher WR than twin lambs. Lower WR FDLPCS (Federal Department of Livestock and Pest Control
values were obtained for lambs born between November Services). 1991. Nigerian national livestock survey.
and May. There is a close parallel between WR and Volume 2. FDLPCS, Abuja, Nigeria. 289 pp.
rainfall pattern; lambs survive better if born in the dry Hassan W B. 1987. Genetic and environmental effects on the
rather than the wet months. growth rate of Yankasa lambs. MSc thesis, Ahmadu
Bello University, Zaria, Nigeria.
Conclusion Osinowo О A. 1982. Oestrus synchronization, artificial
insemination and early rebreeding in Yankasa sheep.
This study shows that there may be no advantage in Nigerian Journal ofAnimal Production 9: 1 07-1 1 1 .
keeping Yankasa breeding ewes under semi-intensive
management beyond the sixth parity in terms of Osinowo О A. 1992. Problems and prospects for the
pre weaning traits, especially WR which consistently fell development of small ruminants in Nigeria. In: The
below average in the latter group. However, the high Nigerian livestock industry: Problems and prospects.
parities had a relatively low number of observations and Proceedings of a workshop held at Abuja, Nigeria,
therefore need further study when larger data sets 26-27 February 1992, FDLPCS (Federal Department
become available. The reasons for the peculiarly bad of Livestock and Pest Control Services), Abuja, Nigeria.
growth and weaning rate performances of lambs born in Osinowo О A and Ekpe G A. 1985. Post-partum intervals to
June also need further investigation. Seasonal effects on oestrus and conception in Yankasa sheep. Journal of
preweaning traits show that growth traits were better for Agricultural Science 1 04:253-255.
lambs born in the wet seasons rather than the dry
seasons, while the reverse was true of the weaning rate. SYSTAT. 1989. SYSTAT Version 4.1. SYSTAT, Inc.,
Evanston, Illinois, USA.

Acknowledgements Taiwo В В A and Buvanendran V. 1983. Breed and

environmentalfactors that influence lamb loss in Shika.
The authors are grateful to field staff of the Small NAPRI (National Animal Production Research Institute)
Seminars 5: 1-13. NAPRI, Zaria, Nigeria
Ruminant Research Programme, NAPRI, for the general
assistance with this project and to the Federal Terrill C E. 1966. Genetic improvement in sheep and goats. In:
Directorate of Food, Roads and Rural Infrastructure for Cole H H (ed), Introduction to livestock production. 2nd
funding support. The permission of the Director of edition. W H Freeman and Co., San Francisco, USA. pp.
NAPRI to publish this paper is gratefully acknowledged. 301-309.

68 Small Ruminant Research Network Workshop

 Studies on factors affecting reproductive performance and
mortality rates of Small East African goats and their crosses
11 2
L.A. Mtenga , G. С. Kifaro and Berhanu Belay

Department ofAnimal Science and Production

Sokoine University ofAgriculture, P. O. Box 3004, Chuo Kikuu, Morogoro, Tanzania
Ministry ofAgriculture, P. O. Box 30579, Addis Ababa, Ethiopia

l'Université agricole de Sokoine (Tanzanie) ont été

Abstract utilisées pour comparer les facteurs génétiques et les
This study compares genetic and non-genetic factors facteurs non génétiques qui agissent sur les
affecting the reproductive performance and mortality performances de reproduction et les taux de mortalité
rates of Kamorai x Small East African (SEA), Boer x des caprins Kamorai x petite chèvre d'Afrique de l'Est,
SEA and SEA goats using records kept at the Depart Boer x petite chèvre d'Afrique de l'Est, et petite chèvre
ment ofAnimal Science and Production, Sokoine University d'Afrique de l'Est.
of Agriculture, Tanzania, between 1972 and 1989.
L 'âge au premier vêlage était compris entre 638 et
Age at first kidding and kidding interval ranged
984 jours, et l'intervalle de vêlage entre 293 et 419
from 638 to 984 days and 293 to 419 days, respectively. jours. La période de mise-bas avait un effet significatif
Age at first kidding was significantly (P<0.001) (P<0,01) sur l'âge au premier vêlage. La période et la
influenced by period of kidding. Period of kidding and saison des mises-bas précédentes avaient un effet
season of previous kidding affected kidding interval significatif (respectivement P<0,01 et P<0,05) sur
significantly (P<0.01 and P<0.05, respectively). l'intervalle de mise-bas.
Overall average mortality rate was 40.6% and
25.7% for preweaning and postweaning periods, La mortalité générale moyenne était de 40,6%
respectively. Animals with birth weights of less than avant le sevrage, et de 25,7% après le sevrage. Les
1.5 kg had the highest preweaning mortality rate animaux pesant moins de 1,5 kg à la naissance
(57.9%). The lowest preweaning mortality rate occurred présentaient le taux de mortalité présevrage le plus
in animals with a birth weight of greater than 2.6 kg élevé (57,9%) et ceux pesant plus de 2,6 kg, le taux de
(29.8%). Twins exhibited a higher preweaning mortality mortalité présevrage le plus bas (29,8%). La mortalité
rate than singles (48.3% vs 38.5%). Preweaning présevrage des chevreaux issus de naissances
mortality was lower in period 1 (1972-1974) than in gémellaires était supérieure (48,3%) à celle des
later years. agneaux nés simples (38,5%). La mortalité présevrage
Non-genetic factors, especially period and season au cours de la période 1 (1972-1974) était inférieure à
of kidding, were the main sources of variation in celle enregistrée les années suivantes.
reproductive performance and mortality rate. Crossing Les facteurs non génétiques, en particulier la
SEA goats with Kamorai and Boer showed little période et la saison de mise-bas, constituaient les
improvement in these factors. Poor feeding management principales sources de variation des performances de
and disease control probably prevented the crosses from reproduction et de la mortalité. Le croisement de la
expressing their genetic potential. petite chèvre de l'Afrique de l'Est avec des animaux de
race Kamorai et Boer n 'a pas apporté d'améliorations
en ce qui concerne ces facteurs. La piètre conduite de
Etude des facteurs influant sur l'alimentation et de la prophylaxie sanitaire a
les performances de probablement empêché les sujets de croisement
d'extérioriser leur potentiel génétique.
reproduction et la mortalité de
la petite chèvre d'Afrique de Introduction
l'Est et de ses produits de Poor reproductive efficiency of tropical goats compared
with temperate breeds has been documented in the
croisement literature (Wilson, 1989). However, tropical goats are
able to adapt and reproduce under harsh environmental
Résumé conditions (Kiwuwa, 1992) and have good fitness traits
(Olayiwole and Adu, 1989; Kiwuwa, 1992). Rearing
Les données enregistrées entre 1972 et 1989 au exotic goats under tropical conditions has disadvantages
Département de science et de production animales de of high mortality and poor adaptation (Olayiwole and

Small Ruminant Research Network Workshop 69

L.A. Mtenga, G. С. Kifaro and Berhanu Belay

Adu, 1989; Kiwuwa, 1992). Tropical goats could be j = type of birth (1 or 2)

genetically improved by selecting for growth,
к = season of birth (1-4)
reproduction and survivability. This has not been
seriously undertaken because the process is 1 = period of birth (1-4)
time-consuming and expensive (Mchau, 1979). Good
selection is also impossible in flocks with inadequate or m=doe
non-existent records, as is the case in many developing \i = overall mean
countries (Olayiwole and Adu, 1989). Wilson (1989)
and Kiwuwa (1992) suggest that selection should В = the effect of genetic group
improve reproduction and growth while conserving the S = the effect of type of birth
characteristics that adapt tropical goats to their
M = the effect of season of birth
environment. On the other hand it has been well
documented (Kyomo, 1978; Mchau, 1979; Senyatso, R = the effect of period of birth
1986) that crossbreeding tropical goats with exotic
'improver' breeds could increase meat and milk e = the error term
production. This advantage has led to the introduction The model used for analysing KI was:
of crossbreeding programmes in Tanzania. Yijklm = Ц + Bi + Pj + Mk + Rl + eijkim
Data on the reproductive traits and mortality rates
of the exotic Boer and indigenous Tanzanian goats and where:
their crosses were collected at the Sokoine University of Y = kidding interval (days)
Agriculture between 1972 and 1989. Prior to this study
the data had not been systematically analysed. The purpose m =doe
of this study was to analyse the performance of the cross i = genetic group (1-3)
bred goats before their large-scale spread to the villages.
j = parity (1-3)
к = season of previous birth (1—4)
Materials and methods l = periodofbirth(l-4)
u, = overall mean
Animals and management В = effect of genetic group
The genetic groups involved in this study were the Small P = effect of parity
East African (SEA), Boer x SEA and Kamorai x SEA
goats. The SEA goats were the dam line. This breed has M = effect of season of previous birth
been characterised by Mason and Maule ( 1 960). The sire R = effect of period of birth
lines were the Boer and Kamorai, which have been
e = error term
characterised by Mason (1981). The goats were
regularly provided concentrates between 1972 and Seasons were grouped by weather pattern: season
1974. Thereafter the regular provision of concentrates 1, February to April (late rains); season 2, May to July
was hampered by their scarcity. (cool dry season); season 3, August to October (hot dry
season) and season 4, November to January (dry
season). As there were only a small number of
Data collection and analyses observations across the years, the data were divided into
Data on reproductive traits were extracted from record periods in order to assess effects over time: period 1
cards for individual animals. Records on mortality rates (1972-1974); period 2 (1975-1977); period 3
were collected from liveweight records and yearly ( 1 978- 1 980) and period 4 (> 1 98 1 ).
inventory books. Mortality rates (MR) were determined for three
Reproductive traits studied were age at first periods: birth to weaning (16 weeks); 16 to 52 weeks of
kidding (AFK) and kidding interval (KI). For AFK the age; and birth to 52 weeks age. MR for each period was
effects of genetic group, birth type of the first-kidding determined as the proportion of all kids born in period
doe, season of birth and period of birth were studied. For that have died.
KI the effects of genetic group, parity, season of Mortality rates were calculated by genetic group,
previous kidding and period of kidding were studied. sex, birth type, season of birth, period of birth and birth
AFK and KI were analysed using the GLM weight. These factors had further subclasses: genetic
procedure in SAS (1988). The model used for the group (Kamorai x SEA, Boer x SEA and SEA); sex
analysis of AFK was: (male and female); birth type (singles and twins); and
birth weight (1 to 1.5, 1.6 to 2.0, 2.1 to 2.5, 2.6 to 3.0
Yijidm = \i + Bi + Tj + Mk + Ri + eijkim, and 3.1 to 3.5 kg). Season and period of birth were
where: classified in the same way as for reproductive traits.
The effects of the above factors on mortality rates
Y = Age at first kidding (days)
were analysed using the chi-square test (X ) in a 2 x С
i = genetic group (1-3) contingency table (Snedecor and Cochran, 1989).

70 Small Ruminant Research Network Workshop

 Studies on factors affecting reproductive etc..

Table 2. Least squares means and standard errors (days)

Results for kidding interval.
Kidding interval (days)
Age at first kidding Effect Mean SE
Overall mean 336 355.89 63.3
The overall mean age at first kidding during the whole
Genetic group
period of the study was 810.78 days (SE = 172.73)
(Table 1). Of the factors studied, only period of birth KamoraixSEA 59 350.09 8.9
significantly (P<0.001) affected age at first kidding. Boer x SEA 68 355.86 8.0
Figures given in Table 1 show that does kidded at SEA 209 350.77 5.67
younger ages in periods 1 and 4 than in other periods.
Although age at first kidding was not significantly Parity
affected (P>0.05) by birth type, female kids born as 1 163 349.41 5.75
twins were on average 50.88 days older at first kidding 2 91 352.68 7.76
than those born as singles.
3 82 354.63 8.27
Season of previous kidding
Table 1. Least squares means and standard errors (days) 1 44 331.31a 10.16
for age at first kidding.
2 79 359.69b 7.78
Age at first kidding (days) 3 85 362.08b 8.16
Effect Mean SE 4 128 355.88b 6.54
Overall mean 291 810.78 172.73
Period of kidding
Genetic group 1 32 369.75a 12.66
KamoraixSEA 50 828.05 28.00 2 166 365.83a 6.02
Boer x SEA 61 806.34 26.31 3 69 342.92a 8.12
SEA 180 790.44 18.47 4 69 325.45b 8.33
Birth type
SE = standard error.
Single 210 782.83 15.83
ab = means within the same column that have different superscripts
Twins 81 833.71 27.11 are significantly different (P<0.05).
Season of birth
1 49 833.36 28.63
2 75 832.99 25.33 Mortality
3 88 798.08 23.34
The chi-square test indicated that differences in
4 79 768.67 24.15 mortality rates among genetic groups were not
Period of kidding significant (P>0.05). The Boer x SEA group had the
1 90 762.54b 16.70 highest mortality rates, SEA had the lowest and the
2 75 883.0a 27.08 Kamorai x SEA was intermediate (Table 3).
3 78 860.20a 26.47 There was a significant influence (P<0.05) of birth
type on the mortality rate of kids from birth to one year
4 48 728.26Ь 37.45
of age (Table 3). There was no significant difference
SE = Standard error. (P>0.05) in mortality rate between singles and twins
from 0 to 16 and 16 to 52 weeks of age. However, there
ab = means within a column that have different superscripts are
significantly different (P<0.03). was a tendency for the mortality rate to be higher for
twins than for singles in all age groups (Table 3). Season
of birth had a significant effect (P<0.05) on the mortality
Kidding interval rate of kids from birth to 16 weeks and birth to one year.
Kids born in season 1 had the highest mortality rate,
followed by those born in season 2. Sex had no effect on
Season of previous kidding and period of kidding had
mortality rate (P>0.05).
significant (P<0.05 and P<0.01 , respectively) effects on
Birth weight had a significant effect on the
the intervals between successive kiddings (Table 2).
mortality rate of kids from 0 to 16 weeks (P<0.01) and
Does that kidded in season 1 had shorter subsequent
kidding intervals than those kidding in other seasons. from 0 to 52 weeks of age (P<0.05). Mortality rate
Does that kidded in period 4 had shorter subsequent decreased with increasing birth weight (Table 3).
kidding intervals than does that kidded in other periods. Preweaning mortality rate in kids was also
The differences in kidding intervals due to period of significantly (P<0.05) affected by period of birth of the
kidding were not significant (P>0.05) among periods 1, kids. Mortality rates were highest for kids born in period
2 and 3. 4 and were lowest in kids born in period 1 (Table 3).

Small Ruminant Research Network Workshop 71

L.A. Mtenga, G. С. Kifaro and Berhanu Belay

Table 3. Mortality rate of goat kids in different age groups by genetic group, sex, birth type, period of birth, season of birth
and birth weight.
Birth to weaning Weaning to 1 year Birth tc) 1 year
Factor Born Died (%) At risk Died (%) Born Die< 1 %
Genetic group
SEA 345 138 (40.0) 207 51 (24.6) 345 189 54.8
Boer x SEA 30 14 (46.7) 16 6 (37.5) 30 20 66.6
Kamorai x SEA 24 10 (41.7) 4 4 (28.6) 24 14 58.3
Total 399 162 (40.6) 237 61 (25.7) 399 223 55.9
Sex
Males 212 87 (41.0) 125 32 (25.6) 212 119 56.1
Females 187 75 (40.1) 112 29 (25.7) 187 104 55.6
Birth type
Single 312 120 (38.5) 192 45 (24.4) 312 165 52.9b
Twins 87 42 (48.3) 45 16 (29.9) 87 58 66.7a
Period of birth
1 1972-74 15 4 (26.7) 11 3 (27.3) 15 7 46.6
2 1975-77 81 25 (30.9) 56 12 (21.4) 81 37 45.7
3 1978-80 135 54 (40.0) 81 20 (24.7) 135 74 54.8
4 1981- 168 79 (47.0) 89 26 (29.2) 168 105 62.5
Season of birth
1 Nov - Jan 108 51 (47.2) 57 18 (31.6) 108 69 63.9
2 Feb - April 131 62 (47.3) 69 17 (24.6) 131 79 60.3
3 May - July 78 28 (35.9) 50 11 (22.0) 78 39 50.0
4 Aug - Oct 82 21 (25.6) 61 15 (24.6) 82 36 43.9
Birthweight (kg)
1.0-1.5 57 33 (57.9) 24 7 (29.2) 57 40 70.2
1.6-2.0 116 56 (48.3) 60 16 (26.6) 116 72 62.2
2.1-2.5 96 34 (35.4) 62 17 (27.4) 96 51 53.1
2.6-3.0 94 28 (29.8) 66 15 (22.7) 94 43 45.7
3.1-3.5 36 11 (30.5) 25 6 (24.0) 36 17 47.2
ab = means within the same column that have different superscripts are significantly different.

Discussion have also been reported in the literature. Raja and

Mukundan (1974) showed that purebred Malabari goats
were younger at first kidding than were Jamunapari x
Age at first kidding Malabari goats. Wilson and Murayi (1988) also found
that SEA goats were younger at first kidding than were
The average age at first kidding was 810.78 days Anglonubian x SEA goats. However, Devendra and
(SE=172.73). This value is close to those reported by Burns (1970) reported little variation in age at first
Kyomo (1978) and Wilson and Murayi (1988), which kidding between Indian native and crossbred goats, and
range from 600 to 780 days for on-station studies. The Kyomo (1978) found no difference between SEA and
values reported in this study are higher than those
Kamorai x SEA goats.
reported in studies of the traditional sector of
sub-Saharan Africa, which range from 301 to 431 days Does born single tended to kid at a younger age
(Wilson, 1988; Lebbie and Manzini, 1989). This is than twins, which agrees with other findings (Wilson
because, in this study, goats were mated for the first time and Murayi, 1988; Wilson et al, 1989). Female twins
at 72 weeks of age. It is also possible that feeding and were probably delayed in reaching puberty due to a
general husbandry practices were substandard under lower-than-average growth rate (Wilson and Murayi,
station conditions, contributing to the greater age at first 1988; Wilson et al, 1989). For twins and singles to kid
kidding in this study. at the same age, twins may need supplements to their
SEA goats tended to be younger than Kamorai x diet to encourage growth. This is especially true for
SEA and Boer x SEA at first kidding. Younger ages at twins from dams with low milk yields. Period of birth
first kidding for tropical goats compared with crosses influenced age at first kidding, also reported by Wilson

72 Small Ruminant Research Network Workshop

 Studies on factors affecting reproductive etc..

and Murayi (1988) and Wilson et al (1989). The effect both physical and biological agents due to their lack of
of period of birth on age at first kidding was mainly a immunity. This makes them more susceptible to enteric
reflection of changed management practices (Wilson and respiratory infections (Ndamukong, 1985). Losses
and Murayi, 1988), which was probably true for this in the preweaning period can be minimised by providing
study. proper shelter, prophylaxis and better nutrition.
This study showed that twins were more likely to
Kidding Interval die than were singles. This agrees with observations
made by Sarmah et al (1981) and Wilson and Murayi
The overall mean kidding interval (Table 2) was close (1988). Twins tended to have lower birth weights. Curtis
to values reported in studies carried out on experimental (1969) concluded that animals with low birth weights
and multiplication stations in the tropics (Wilson and had lower energy reserves and were therefore less able
Murayi, 1988; Wilson et al, 1989). It is, however, longer to withstand harsh environments; also if the dam has a
than values found in the traditional sector, which ranged
poor milk yield, she may be unable to provide adequate
from 238 to 265 days (Wilson and Durkin, 1 988; Lebbie
and Manzini, 1989). The longer kidding intervals nutrition for twins, thus increasing their susceptibility to
recorded on stations is the result of a controlled breeding disease.
policy imposed to achieve the best breeding season. Kids born during the wet season had a higher
Normally there is no imposition of seasonal breeding in mortality rate than those born in the dry season. Similar
the traditional sector and animals of all ages and sex run trends have been reported elsewhere (Mazumdar et al,
together day and night. In this study, although there was 1980; Sarmah et al, 1981; Chawla et al, 1982; Wilson,
no well-defined seasonal breeding policy, mating was 1988). The higher mortality rate in the rainy season was
taking place only at night (between 1800 and 0800 associated with high rainfall and high relative humidity
hours), resulting in less exposure of the does for mating (Mazumdar et al, 1980), both of which are known to
than in the traditional sector. promote disease and parasitic infection.
This study shows that the season of previous birth The birth weight of the kid had an influence on pre-
and period of birth were the only main sources of and post-weaning mortality rates. This confirms earlier
variation contributing to differences in kidding findings in tropical goats (Mazumdar et al, 1980; Singh
intervals. Does that had their previous kid in season 1 et al, 1990). Mazumdar et al (1980) reported mortality
(late rains) showed shorter kidding intervals than does rates of 1 00% for kids weighing 1 .0 to 1 .5 kg and 32.2%
that gave birth in other seasons. This confirms the for kids weighing from 2.0 to 2.5 kg. This study shows
findings of Wilson and Murayi (1988), who reported a lower mortality rate for kids weighing 1.0 to 1.5 kg
shorter kidding intervals for dams that had their previous than Mazumdar et al (1980) (Table 3).
kid during the rainy season. Wilson and Durkin (1988) The effect of period of birth on viability of kids
also reported a similar seasonal effect. This is possibly agrees with results published by Chawla et al ( 1 982) and
related to the availability of feed from pasture, which Wilson and Murayi (1988). Table 3 shows that kids born
may have influenced ovulation rate and fertility. between 1972 and 1974 had lower pre-weaning and
The effect of period (year) of kidding on kidding post-weaning mortality rates than those born in other
interval has also been demonstrated by Adeoye (1985), periods. During this period disease prevention and feed
Wilson and Durkin (1988) and Wilson and Murayi provision were good. Thereafter, inconsistent
(1988). In this study the variation might have been management contributed to higher mortality rates. The
associated with inconsistent management on the standard of flock hygiene and disease monitoring needs
station. improvement in order to lower mortality rates.

Mortality
The effect of genetic group on mortality rate was small Conclusion
and non-significant. However, Boer x SEA tended to Crosses were expected to perform better in reproductive
show higher mortality rates (Table 3). This is in traits and survivability of kids compared to SEA goats
agreement with other findings (Mchau, 1979; Senyatso,
but did not. This may be due to poor adaptation of the
1986). This may be due to the poor adaptation of Boer crosses and poor nutrition and disease control from 1 975
goats to harsh environmental conditions and increased
to 1989.
susceptibility to disease. Similar conclusions have been
drawn for crosses between Boer and indigenous tropical
goats by Mchau (1979) and Senyatso (1986). In this
study the major causes of mortality were not fully References
recorded. Nonetheless, from the limited records
available, pneumonia seems to have been a major cause Adeoye SAO. 1985. Reproductive performance of West
of death in kids. African Dwarf goats in south-western Nigeria. In:
Wilson R T and Bourzat D (eds). Small ruminants in
Mortality was high in the preweaning period African agriculture. Proceedings of a conference held at
(Table 3). Similar results were reported by Kyomo ILCA, Addis Ababa, Ethiopia, 30 September-4 October
(1978), Mchau (1979) and Sarmah et al (1981). 1985. ILCA (International Livestock Centre for Africa),
Younger animals seem less able to withstand attack by Addis Ababa, Ethiopia. pp. 19-24.

Small Ruminant Research Network Workshop 72

L.A. Mtenga, G. С. Kifaro and Berhanu Belay

Chawla D S, Bhatnagar D S and Mishra R R. 1982. Factors Olayiwole M В and Adu I F. 1989. Past and present research
affecting kid mortality in dairy goats. Indian Journal of on sheep and goats breeding in Nigeria. In: Adeniji К О
Animal Sciences 52:166-171. (ed), Improvement of small ruminants in West and
central Africa. Proceedings of a workshop held at
Curtis H J. 1969. Aging. In: Hafez E S E and Dyer I A (eds),
Ibadan, Nigeria, 21-25 November 1988. OAU
Animal growth and nutrition. Lea and Fibger,
(Organization of African Unity), Nairobi, Kenya. pp.
Philadelphia, USA. pp. 165-174.
61-69.
Devendra C and Burns M. 1970. Goat production in the
Raja C A and Mukundan G. 1 974. Age at first kidding, kidding
tropics. Bureau of Animal Breeding and Genetics,
Technical Communication 19. Commonwealth rate and kidding interval in Malabari and
Majunapari-Malabari crosses. Karala Journal of
Agricultural Bureaux, Farnham Royal, UK. 184 pp.
Veterinary Science 4: 1 65- 1 69.
Kiwuwa G H. 1992. Breeding strategies for small ruminant
productivity in Africa. In: Rey В, Lebbie S H В and Sarmah P C, Thakuria K, Sarma H К, Borah H P, Mohan M
and Pant KP. 1 98 1 . Note on kid mortality in Assam local
Reynolds L (eds), Small ruminant research and
development in Africa. Proceedings of the First Biennial breed. Indian Journal ofAnimal Sciences 5 1 :248-249.
Conference of the African Small Ruminant Research SAS (Statistical Analysis System). 1988. Statistical analysis
Network, ILRAD (International Laboratory for system user's guide. Statistical Analysis System
Research on Animal Diseases), Nairobi, Kenya, 10-14 Institute, North Carolina, USA. 1028 pp.
December 1990. ILCA (International Livestock Centre
for Africa), Nairobi, Kenya. pp. 423-434. Senyatso E K. 1986. Performance of sheep and goats in
Botswana comparative studies. II. In: Adeniji К О and
Kyomo M L. 1978 Meat from goats in Tanzania. PhD thesis, Kategile J A (eds), Improvement of small ruminants in
University of Dar es Salaam, Tanzania. 292 pp. eastern and southern Africa. Proceedings of a workshop
Lebbie S H В and Manzini A T. 1989. The productivity of held at Nairobi, Kenya, 10-14 December 1986. IB AR
indigenous goats under traditional management in (Inter African Bureau for Animal Resources), Nairobi,
Swaziland. In: Wilson R T and Azeb Melaku (eds), Kenya. pp. 61-66.
African small ruminant research and development. Singh D K, Mishra H R and Singh C S P. 1990. Genetic and
Proceedings of a conference held at Bamenda, non-genetic factors affecting pre-weaning survivability
Cameroon, 18-25 January 1989. ILCA (International in kids. Animal Production 5 1 :504-559.
Livestock Centre for Africa), Addis Ababa, Ethiopia. pp.
39-50. Snedecor G W and Cochran W G. 1989. Statistical methods.
8th ed. Iowa State University Press, Ames, Iowa, USA.
Mason I L. 1981. Breeds. In: Gall G (ed), Goat production.
503 pp.
Academic Press, London, UK. pp. 57-1 10.
Wilson R T. 1988. The productivity of Sahel goats and sheep
Mason I L and Maule J P. 1960. The indigenous livestock of under transhumant management in northern Burkina
Eastern and Southern Africa. Bureau of Animal Faso. Bulletin ofAnimal Health and Production in Africa
Breeding and Genetics, Technical Communication 14. 36:348-355.
Commonwealth Agricultural Bureaux, Farnham Royal,
UK. 151 pp. Wilson R T. 1989. Reproductive performance of African
indigenous small ruminants under various management
Mazumdar N H, Mazumdar A and Goswami К К. 1980. systems: a review. Animal Reproduction Science
Studies on some factors affecting mortality and survival 20:265-286.
rates in Pashmina kids. Indian Journal of Animal
Sciences 50:251-255. Wilson R T and Durkin J. 1988. Livestock production in
central Mali: Reproductive components in traditionally
Mchau К W. 1979. Influence of Boer goats for crossing with
managed sheep and goats. Livestock Production Science
Tanzanian goats. MSc thesis, University of Dar es 19:523-529.
Salaam, Tanzania. 92 pp.
Wilson R T and Muray i T. 1 988. Productivity of the Small East
Ndamukong К J N. 1985. Effects of management system on
African goat and its crosses with Anglonubian and the
mortality of small ruminants in Bamenda, Cameroon.
Alpine in Rwanda. Tropical Animal Health and
In: Wilson R T and Bourzat D (eds), Small ruminants in
Production 20:219-228.
African agriculture. Proceedings of a conference held at
ILCA, Addis Ababa, Ethiopia, 30 September-4 October Wilson R T, Murayi T and Rocha R. 1 989. Indigenous African
1985. ILCA (International Livestock Centre for Africa), small ruminant strains with potentially high reproductive
Addis Ababa, Ethiopia. pp. 108-1 16. performance. Small Ruminant Research 2:107-1 17.

74 Small Ruminant Research Network Workshop

 Reproductive traits and disease incidence characteristics
of Dorper, Dorper x Masai and Masai ewes raised under
semi-arid conditions in Kenya
W.O. Odenya

Egerton University, Animal Science Department

P.O. Box 536, Njoro, Kenya

Abstract parentales (D et M). Les différences entre types

génétiques maintenus dans des conditions environ
Reproductive and disease incidence characteristics of nementales identiques montrent que la prédisposition
Dorper (D), Fi Dorper x Masai (DM) and Masai (M) ewes aux maladies repose sur une base génétique.
were compared under semi-arid conditions in Kenya.
Percentage of lambs born alive were 87. 1 , 90. 1 and
98.2 for D, DM and M ewes respectively. Gestation
Introduction
lengths were 149.1, 148.2 and 146.7 days respectively; Hair sheep form a major component of the livestock
birth weights were 3.1, 3.5 and 2.7 kg; ages at first sector of agriculture in Kenya. Of a total of
service were 12.2, 11.8 and 12 months and weights at approximately 7 million sheep in Kenya (De Boer,
first service were 30, 32.2 and 26.2 kg respectively. 1981) about 95% are hair sheep (Odenya, 1984) which
Significant differences between breed groups were produce an estimated 15 to 20% of the country's red
observed in reproductive traits and disease frequencies. meat (Allonby, 1975). The predominant indigenous hair
The DM ewes were younger and heavier at first service sheep breeds are Masai and Blackhead Persian.
However, Dorper has emerged as the most important
than the parental breeds (D and M). The differences
sire breed for use in crossbreeding for mutton production
between breed groups kept under the same
in the semi-arid areas of Kenya. The indigenous hair
environmental conditions indicate that disease breeds have not been adequately evaluated against
susceptibility has a genetic basis. Dorper under similar conditions to ascertain genetic
differences between them. It is only after genetic
evaluation of individual, maternal and paternal traits that
Caractéristiques de la appropriate breeding plans can be formulated to
reproduction et de l'incidence optimise production in a given environment.
Reproductive traits and disease incidence within
des maladies chez les brebis de and between breeds are important traits that are taken
race Dorper, Dorper x Massai into account in selecting sheep for meat production and
culling. De Boer (1981) reported that the reproductive
et Massai élevées en milieu rate of hair sheep in semi-arid areas is low. Allonby
semi-aride au Kenya (1975) reported prevalence of disease in hair sheep
breeds in the same environment. Low reproductive rates
Résumé and a higher disease incidence would ensure a low
selection differential resulting in reduced genetic gains.
Les caractéristiques de la reproduction et de The objective of the research reported here was to
l 'incidence des maladies chez des brebis Dorper (D), F¡ examine flock records and determine reproductive traits
Dorper x Massaï (DM) et Massaï élevées en milieu (percentage of lambs born alive; gestation length; birth
semi-aride au Kenya ont été comparées. Les weight; age and weight at first service), disease
pourcentages d'agneaux nés vivants étaient incidence and reasons for culling in Dorper (D), Dorper
respectivement de 87,1, 90,1 et 98,2 pour les brebis D, x Masai (DM) and Masai (D) breed groups.
DM et M. Les durées respectives de gestation étaient de
149,1, 148,2 et 146,7jours et les poids à la naissance de Materials and methods
3,1, 3,5 et 2,7 kg. L'âge à la première saillie était
respectivement de 12,2, 1 1,8 et 12 mois et les poids à la Animals and traits
première saillie de 30, 32,2 et 26,2 kg. On observe des
différences significatives entre les types génétiques pour Reproductive parameters were determined for 330 D,
ce qui concerne les caractéristiques de la reproduction 260 Fi DM and 100 M ewes between 1982 and 1984.
et lafréquence des maladies. Les brebis DM étaient plus Parameters included percentage of lambs born alive in
jeunes et plus lourdes à la première saillie que les races a lambing season, gestation length (days), birth weight

Small Ruminant Research Network Workshop 75

W.O. Odenya

of lamb (kg) as a trait of the dam, age of ewe at first Table 1. Reproductive parameters for Dorper, F\ Dorper x
service (months) and weight of ewe at first service (kg). Masai, Masai breed groups.
Data on the number of animals treated or diagnosed for Fi Dorpers
dystocia, mastitis, foot-rot, pneumonia, haemonchosis Dorper X Masai Masai
and other undefined problems and diseases were Parameters (N=330) (N=260) (N=100)
collected from 3110 ewes (860 D, 1200 DM and 1050 87.1a 90.1ь 98.2е
M) between 1974 and 1984. Similarly, the reasons for % Lambs bom alive
culling were determined from 428 ewes (200 D, 1 50 DM
Gestation length 148.1a
and 78 M) from 1979 through to 1984. The reasons for (days) 149.2a 146.7b
culling were the same as those for disease incidence,
plus low milk yield. 3.1a 3.5b 2.7е
Birth weight (kg)
The animals were located at Ol'Magogo farm in
Naivasha, Kenya at the UNDP/FAO (United Nations Age at first service 11.8a
(months) 12.2a 12a
Development Programme/Food and Agriculture
Organization of the United Nations) Sheep and Goat Weight at first
Development Project. 30b 32.2b 26.2a
service (kg)
Means with different superscripts within the same row are statistically
Animals management different, P<0.05.

All the animals were managed in the same flock. They Studies by Dickerson and Laster (1975) and
were maintained on natural pastures that consisted Jakubec (1977) showed that heterosis may contribute to
primarily of Naivasha star-grass. Once a month the earlier sexual development in sheep. Their observations
animals were supplied with mineral supplements. were confirmed in this study (Table 1), although the
Water was supplied freely. The animals were mated in differences were not significant (P>0.05).
December (after the short rains) to lamb in April-May
M ewes had the shortest gestation length of the
(after the long rains). Flocks of purebred D and M were
maintained. Some of the D breed sires were used for three breed groups (Table 1). Studies by Ricordeau
crossbreeding M ewes to obtain Fi DM animals. All (1981) have shown that larger animals have longer
animals were eartagged at birth. gestation length than smaller ones. In this study, M ewes
were significantly (P<0.05) lighter than DM and D ewes
at first service (Table 1). This may have accounted for
Analysis of data the shorter gestation length in the M ewes.
The data were analysed with a one-way (breed group)
analysis of variance. Pairwise comparisons of means for Disease incidence
reproductive traits were done using Duncan's multiple
range test (Steel and Torrie, 1960). The Friedman test Table 2 shows the number and frequency of different
(Steel and Torrie, 1960) was used to find significant diseases by breed group. The DM breed group had the
differences between breed groups for disease incidence. highest incidence of dystocia while M had the lowest.
The DM breed group also had the highest birth weight
while M had the lowest (Table 1). Breed groups with
Results and discussion higher incidence of dystocia also tended to have longer
gestation lengths (Tables 1 and 2). This concurs with
Reproductive parameters results by Philipsson (1976) who reported that gestation
length is positively correlated with dystocia. The results
Results of reproductive parameters for D, DM and M therefore suggest a possible positive phenotypic
breed groups are shown in Table 1. M ewes had the correlation between dystocia, birth weight and gestation
highest percentage of lambs born alive, followed by DM length.
and then D. The increased reproductive performance of The D breed group had the highest incidence of
the crossbred ewes relative to that of D ewes is probably foot-rot, pneumonia and haemonchosis (Table 2). The
due to the increased oestrons and gonadal activities M breed group had the lowest incidence of all specified
coupled with improved uterine and peri- and post-natal diseases but also recorded the highest incidence of
conditions for foetal and lamb survival. One important unspecified diseases (Table 2).
determinant of the number of lambs born is ovulation Differences in the frequency of diseases between
rate. Ovulation rate, however, shows little or no breed groups kept under the same environmental
heterosis (Nitter, 1978). The little heterosis observed in conditions indicate that disease susceptibility has a
this study in the crossbreds could be due to this reason. genetic basis.
Another possible reason for the low heterosis shown in
the crossbreds could be that there was little genetic
variance in the ovulation rate of their purebred parents. Reasons for culling
Also, lack of heterosis may have been possible if both
parental breeds (D and M) had been well adapted to the Table 3 shows the frequency distribution (%) of reasons
prevailing environmental conditions. for culling among ewes of the three breed groups. D

76 Small Ruminant Research Network Workshop

 Reproductive traits and disease incidence etc..

Table 2. Number (No.) and frequency (%) per breed group of different diseases in ewes.
Dorper Fi Dorper x Masai Masai
Diseases No. % No. % No. %
Dystocia 86 10.0b 145 12.1b 21 2.0a
Udder (mastitis) 172 20.0b 286 23.8C 66 6.3a
21.5e 153 12.8b 5.3a
Feet (foot-rot) 185 56
Respiratory (pneumonia) 133 15.5e 155 12.9* 126 12.0a
20.5e 180 15.0b 107 10.2a
Digestive (haemonchosis) 176
Others 108 12.6* 281 23.4b 674 64.2e
Total 860 100 1200 100 1050 100
Friedman test: Significant differences between breed groups, P<0.05.

ewes were culled most frequently for udder problems References

and low milk yield, while the M breed group had the
lowest culling rate in all categories of known reasons for Allonby E.W. 1975. Investigations of smallstock diseases in
culling. Kenya. UNDP/FAO (United Nations Development
Programme/Food and Agriculture Organization of the
United Nations) Sheep and Goat Development Project.
Table 3. Frequency distribution (%) within breed groups Project Report, Ministry of Agriculture and Livestock
for reasons for culling ewes. Development, Nairobi, Kenya.
Fi Dorper De Boer A J. 1981. Production economics, UNDP/FAO
Reasons for culling Dorper x Masai Masai (United Nations Development Programme/Food and
12e 10b 4a Agriculture Organization of the United Nations) Sheep
Dystocia problems and Goat Development Project. Project Report.
40e 30ь 4a Mininstry of Agriculture and Livestock Development,
Udder problems
9ь 8b 5a Nairobi, Kenya.
Feet problems
Dickerson G E and Laster D B. 1975. Breed, heterosis and
Respiratory problems 5ab 4a environmental influences on growth and puberty in ewe
7*
23e 20b 10a lambs. Journal ofAnimal Science 41 : 1-9.
Low milk yield
Jakubec V. 1977. Productivity of crosses based on prolific
Unknown problems 9a 27b 73е
breeds of sheep. Livestock Production Science
4:379-392.
Total 100 100 100
Nitter G. 1978. Breed utilisation for meat production in sheep.
Means with different superscripts within the same row are different Animal Breeding Abstract 46: 1 3 1-143.
(P<0.05).
Odenya W O. 1984. Sheep breeds and types; for the Kenyan
farmer. Proceedings of the Animal Production Society
ofKenya. Animal Production Society of Kenya, Kabete,
Kenya.
Implications and conclusion
Philipsson J. 1 976. Studies on calving difficulty, stillbirths and
The difference in reproductive traits, the known culling associated factors in Swedish cattle breeds. III. Genetic
parameters. Acta Agriculturae (Scandinavica)
reasons and the known disease frequencies in Dorper,
26:21 1-220.
Dorper-Masai and Masai breed groups may have a
genetic basis, probably due to differences in adaptation Ricordeau G. 1981. Genetics: Breeding plans. In: Gall C (ed).
to their environment. In disease epidemic areas it may Goat production. Academic Press, London, UK. pp.
be cost-effective to rear Masai rather than Dorper sheep 111-169.
as Masai seem to be the most appropriate breed for Steel R D G and Tome J H. 1960. Principles and Procedures
farmers in semi-arid areas, followed by Fi Dorper- of Statistics. Mcgraw-Hill Book Company, New York,
Masai crosses. USA. 481 pp.

Small Ruminant Research Network Workshop 77

 Impact de la prophylaxie sanitaire sur la productivité des
petits ruminants dans la zone semi-aride du Mali
Mamadou Doumbia

Station de recherche zootechnique, Sotuba

B.P. 262, Bamako (Mali)

Résumé Impact ofdisease control on

L'effet de la prophylaxie sanitaire sur les small ruminant productivity in
performances de croissance et de reproduction en milieu
paysan a été étudié sur 500 ovins et 700 caprins the semi-arid zone of Mali
appartenant à 34 unités de production réparties entre 8
villages du Cercle de Banamba (Mali). Les animaux ont Abstract
été répartis en trois lots: un lot témoin; un lot d'animaux
vaccinés contre la pasteurellose, le charbon bactéridien A study was conducted to determine the effect of
et la peste des petits ruminants; et un lot d'animaux disease control on on-farm growth and reproductive
vaccinés contre les trois maladies précitées et performance of 500 sheep and 700 goats belonging to
déparasités aux anthelminthiques (Thibenzol, Panacur, 34 production units distributed among eight villages
Exhelm et Bolumisole ND). around Banamba in Mali. The animals were divided in
Il ressort des résultats obtenus que l' infestation est three groups: a control group, a group of animals
beaucoup plus importante chez les adultes que chez les vaccinated against pasteurellosis, anthrax and PPR
jeunes. En juillet, la moyenne de l'OPG (oeufs par (peste des petits ruminants), and a group of animals
gramme de fèces) des ovins était de 146,25 chez les vaccinated against these three diseases and treated with
adultes et de 108,75 chez les jeunes. Chez les caprins anthelmintics (Thibenzol, Panacur, Exhelm and
elle était respectivement de 24,07 et de 10,32. Après le Bolumisole ND).
déparasitage, l'OPG avait baissé dans le lot déparasité Results show that infestation was much higher in
(26,05) mais restait élevé dans les lots non déparasités adults. In July, average EPG (eggs/g in faeces) was
(X = 168, 1 8). La différence entre les moyennes des OPG 146.25 and 108.75 in adult and young sheep,
des lots déparasité et non déparasités était significative. respectively, and 24.07 and 10.32, respectively in goats.
Les meilleurs gains pondéraux étaient réalisés par les After anthelmintic treatment, EPG declined in the
animaux déparasités au Panacur et à l'Exhelm.
treated group but remained high in the two untreated
La moyenne des GMQ des animaux traités au groups (X=168.18). Average EPG of treated and
Panacur était la plus élevée tant chez les ovins que chez untreated groups differed significantly. Animals treated
les caprins, tandis que celle du lot témoin était supérieure with Panacur and Exhelm showed the highest weight
en certains cas à la moyenne obtenue pour les autres gains. Mean daily weight gain was highest in animals
traitements. Toutefois, l'analyse n'avait pu être treated with Panacur, both in sheep and goats and in
effectuée que sur 68 chèvres et 43 ovins, en raison des some cases was higher in the control group than the
nombreuses sorties et de la gestation.
mean obtainedfrom other treatments. However, only 68
Le taux de naissance était de 108% et 116% goats and 43 sheep were included in the analysis, due to
respectivement chez les ovins des lots non déparasités numerous disposals and pregnancy.
et déparasité, et de 126% et 140% chez les caprins. Les Birth rate was 108% and 116%, respectively, for
taux d'avortement chez les ovins étaient de 2,9 et 3,2% sheep in the untreated and treated groups, and 126%
dans les lots non déparasités et déparasité, et de 6,25 et
and 140% for goats. Abortion rate in sheep was 2.9%
20% chez les caprins.
and 3.2% in untreated and treated groups, and 6.25 and
20% in goats.

Small Ruminant Research Network Workshop 79

 The performance of the Small East African goats and their
Saanen crosses in Malawi
S.K. Kama andJ.W. Banda
Department ofAnimal Science, University ofMalawi, Bunda College ofAgriculture
P.O. Box 219, Lilongwe, Malawi

Abstract Ces résultats montrent que le croisement de la

petite chèvre d 'Afrique de l 'Est avec les Saanen permet
The performance of randomly selected Malawian local d'accroître la production laitière et la croissance des
goats and their Saanen crosses was compared. The métis et partant, la production de viande.
average litter size at kidding was 1.35 and 1.52,
respectively. Birth weights, 12-week weights and
weaning weights at 17 weeks were respectively 2.2 kg
(SE= 0.12 kg), 8.2 kg (SE=0.41 kg) and 10.7 kg (SE = Introduction
0.62 kg) for the local goats and 2.31 kg (SE=0.40 kg),
The rural people of Malawi depend on the zebu cow for
1 1.4 kg (SE = 1.59 kg) and 19.2 kg (SE = 2.39 kg) for
their milk supply. Malawi has one million cattle. Of the
the Saanen crosses (P<0.001). The growth rate up to
eight million people, 89% live in rural areas. The
weaning was 71 g (SE = 4.8 g/day) and 139 g (SE=18.5 population of people is increasing at a rate of 3.7% per
g/day) for the local goats and their Saanen crosses year, but that of cattle is decreasing (NSO, 1988). With
respectively (P<0.01). Saanen crosses produced more a rural population density of 85 people/km (NSO,
milk (83 kg, SE 4 kg) during a 12-week period 1988) and per capita land availability of 0.97ha
(P<0.001) than the local does (37 kg, SE 5 kg). (Munthali, 1987), the prospects for meeting the rising
The results indicate that crossing Small East demand for milk from the zebu alone do not seem
African goats with Saanens would improve not only promising.
milk production, but also the growth rate of their kids, Some 28% of the rural community keep goats
contributing to improved meat production. (NSSA, 1981/82). Thus there are considerable
opportunities for milk production from goats.
The performance of the Small East African (SEA)
Performances de la petite chèvre goat of Malawi has been extensively investigated
(Karua, 1989;Kasowanjeteetal, 1987). Results on milk
d'Afrique de l'Est et des yield indicate that the Malawi goat is a poor yielder
(Mwenifumbo and Phoya, 1982; Mchiela, 1989; Banda,
produits de son croisement avec 1 989; Banda, 1 992). Crossbreeding may be a more rapid
la Saanen au Malawi way of improving their milk yield. The choice of the
most efficient breed/genotype requires estimates of the
Résumé relative performance of the genotypes. The objective of
this study was to evaluate the milk production and
Les performances de chèvres locales du Malawi reproduction performance of local goats and their
choisies au hasard ont été comparées à celles des Saanen crosses. Crossbreeding with Saanen dairy goats
produits de leur croisement avec la Saanen. La taille was started in 1988. This paper reports the results of the
moyenne des portées de ces deux types génétiques était study.
respectivement de 1,35 et de 1,52; les poids à la
naissance et à 12 semaines et le poids au sevrage à 17
semaines étaient de 2,2 kg (±0,12 kg), 8,2kg(±0,41 kg) Material and methods
et I0,7 kg (± 0,62 kg) respectivement chez les chèvres
locales, et de 2,31 kg (± 0,40 kg), 11,4 kg (± 1,59 kg) et The SEA goat indigenous to Malawi (local) and Saanen
19,2 kg (± 2,39 kg) respectivement chez les croisés x local halfbreds were randomly mated to either local or
Saanen (P<0,001). La croissancejusqu'au sevrage des Saanen bucks. The four kid genotypes produced were
chèvres locales était de 71 kg (± 4,8 g/jour), celle des pure local, 1/4 Saanen, 1/2 Saanen and 3/4 Saanen and
croisés Saanen de 139 kg (± 18,5 g/jour) (P<0,01). Sur were compared for birth weight. The 1/4 and 3/4 Saanen
une période de 12 semaines, la production laitière des genotypes were not included in the comparisons for
croisées Saanen était supérieure (83 kg ±4 kg) à celle daily weight gains due to their small numbers. The dams
des chèvres locales (37 kg ±5 kg) (P<0,001). were compared for milk yields and litter size.

Small Ruminant Research Network Workshop 81

S. К. Каша and J.W. Banda

Both the dams and kids were fed indoors on parturition weight than the local dams and dams kidding
stargrass (cynodon nlemfriensis) and Napier grass in the dry season had significantly higher (P<0.05)
(pannisetum purpureum) for the first month after parturition weights. Year of parturition did not affect the
parturition. Thereafter they were herded on natural dam parturition weight.
pasture with the rest of the flock. Supplementary feeding
comprising 30% Leucaena (leucaena leucocephala) Litter size
was given at 1 kg/goat during milking times. The kids
were supplemented with the same concentrates at 1/2 kg Results indicate that the breed of dam, season of birth,
per kid/day until weaning. The animals were milked or year of birth did not affect litter size. Saanen crosses
once daily in the morning. Kids were left with their dams had slightly higher litter sizes. Dams kidding in the dry
during grazing hours but separated in the evening and season also had higher litter sizes, but the differences
weaned at 12 weeks of age. were not statistically significant (Table 3).
Data collected included: birth weight; sex of kid;
weekly bodyweight gains; litter sizes; dam postpartum
Milk yields
weight; and milk yields. The milk yield was multiplied
by two, assuming equal intake by the kids. The data was Results show that the total weekly milk yield was
analysed using Harvey's (1977) Least Square analysis affected by breed and year (Table 4). The Saanen
and the Statistical Package for Social Sciences (SPSS) crossed dams produced significantly more milk
programmes available at Bunda College. (P<0.001). There were significant differences in milk
yield in 1988, 1989 and 1990 (P<0.01). Season did not
Results influence milk yield. Although lactation stage did not
significantly affect total weekly milk yields, indications
were that peak yield was reached in the fifth week of
Birth weight lactation.
Table 5 shows the total milk yield in 12 weeks. The
Birth weight was examined in four genotypes. Results
overall milk yield was 69.6 ± 5.70 Kg. The total milk
show that the birth weight increased (P<0.001) with the
yield in 12 weeks was affected by breed; Saanen crosses
level of crossing with Saanen (Table 1).
produced more milk than the locals (P<0.001). Neither
the season of kidding nor the year of kidding affected
Table 1. Least square means for birth weight of kids (kg). milk yields.
Source of No. of
evaluation observations Mean SE
2.025a Growth rate to weaning
Pure local 120 0.146
1/4 Saanen 7 2.285b 0.191 Table 6 shows the daily weight growth of the kids.
1/2 bred Saanen 128 2.406° 0.044 Saanen crosses grew faster (P<0.0 1 ) than pure local, and
2.733d growth was higher in the dry season (P<0.05) than in the
3/4 Saanen 15 0.130 rainy season for both genotypes. Daily weight gains
Means bearing different superscripts differ significantly (P<0.05). were higher in 1988 than in the subsequent years.
Weaning weight also followed the same pattern
Dam parturition weight with Saanen crosses having a higher weaning weight
than the local. Those weaned in the dry season and in
Parturition weight was affected by breed and season of 1988 also had higher weaning weights (P<0.01). Male
kidding (Table 2). Saanen crosses had a higher (P<0.01 ) kids were heavier (P<0.05) than female kids.

Table 2. Least square means for parturition weight of dam (kg).

Main factor Sub-class No. of observations Mean SE
Breed local 23 27.64a
1.57
Saanen x local 10 37.7b 3.67
Season rainy 15 29.6a'
1.81
dry 18 35.7b
3.10
Year 1988 10 32.3 3.21
1989 10 35.5 2.73
1990 13 30.27 2.60
Regression litter size
linear 3.35 2.90
lactation no.
linear 3.70 1.26
Means bearing different superscripts differ significantly (P<0.05).

82 Small Ruminant Research Network Workshop

 The performance of the Small East African goats etc...

Table 3. Least square means for litter size.

Source of variation Sub-Class No. of observations Mean SE
Breed local 23 1.35 0.10
Saanen 10 1.52 0.23
Season rainy 15 1.38 0.12
dry 18 1.48 0.19
Year 1988 10 1.42 0.20
1989 10 1.61 0.15
1990 13 1.27 0.15
Lactation 1 14 1.21 0.13
2 8 1.78 0.17
3 9 1.50 0.14
4 2 1.24 0.27

Table 4. Least square estimates of total weekly milk yields (kg/wk).

Source of variation Sub-class No. of observations Mean SE
Breed local 276 3.08a 0.14
8.50Ь
Saanen x local 120 0.33
Season rainy 180 5.65 0.17
dry 216 5.94 0.28
Year 1988 120 6.88" 0.29
1989 120 4.74е 0.25
1990 156 5.97b 0.24
Lactation week 1 33 4.80 0.35
2 33 5.47 0.35
3 33 5.41 0.35
4 33 5.97 0.35
5 33 6.35 0.35
6 33 6.11 0.35
7 33 5.67 0.35
8 33 6.06 0.35
9 33 6.08 0.35
10 33 5.94 0.35
11 33 5.69 0.35
12 33 5.96 0.35
Regression litter size 0.26 0.26
lactation no. 0.48 0.11
Means bearing different superscripts differ significantly (P<0.05).

Table 5. Least square means for total milk yield in 12 weeks (kg).
Source of variation Sub-class No. of observations Mean SE
37.1a 4.79
Breed local 23
10 102.0b 11.21
Saanen
Season rainy 15 67.7 5.52
dry 18 71.4 9.47
Year 1988 10 80.3 9.79
1989 10 56.8 8.32
1990 13 71.5 7.93
Regression Sibsta linear 3.2 8.85
lactation no. linear 5.70 3.85
Means bearing different superscripts differ significantly (P<0.05).

Small Ruminant Research Network Workshop 83

S.K. Karua and J.W. Banda

Table 6. Least square means for daily weight gains (g).

Source of variation Sub-class No. of observations Mean SE

Breed local 19 70.78a 4.81

Saanen x local 8 139.22b 18.49

dry 14 115.82b 10.78

Year 1988 11 132.45c 14.88

1989 12 85.07a 8.70

1990 4 97.48b 16.73
Regression weaning weight 10.53 10.88

lactation no. -5.16 5.54

Means bearing different superscripts differ significantly (P<0.05).

Discussion and conclusion References

Kids born to both Saanen crosses and local goats grew Banda J W. 1 989. Milk production potential ofgoats and sheep
at similar rates. Results indicate that Saanen crosses are in Malawi. Phase 1 . Genotype and seasonal influence on
superior to locals in terms of milk yields and kid birth milk yield and composition. Final Technical Research
weight. Survival rate was the same for both Saanen and Report. Contract Research Committee, Ministry of
the locals. The higher daily weight gain of the Saanen Agriculture, Lilongwe, Malawi.
crosses indicates that they can be used for meat as well Banda J W. 1992. Genotype and seasonal influence on milk
as milk production. yield and milk composition of sheep and goats in
The higher birth weights of kids sired by Saanen Malawi. PhD thesis. University of Gissein, Germany.
bucks reported in this study confirms the findings of 227 pp.
Karua and Banda (1992). Litter sizes were similar
Eldestein M. 1988. Health problems associated with
irrespective of the breed, season and year of birth, intensification of livestock industry. Proceedings of the
showing an equally high reproductive capacity in both first workshop on livestock production in Malawi,
genotypes. Zomba, Malawi, 3-9 January 1988.
Results show that crossbred goats produced 275% Harvey W R. 1977. User's guide for LSML76. Mixed model
more milk than did the locals. Similar results have been least-squares maximum likelihood computer
obtained in India using the Saanen to improve milk programme. Ohio State University, Columbus, Ohio,
production in native goats (Sahni and Chawla, 1982). USA. (Mimeograph).
These authors obtained increases of between 65 and Karua S К. 1989. A study of some of the phenotypic
130%. They suggest that the improvement in milk characteristics of the indigenous goats in Malawi. MSc
production is greater when the native breed has a lower thesis. University of Malawi, Malawi. 130 pp.
potential milk production than the well established dairy
Karua S К and Banda J W. 1992. Dairy goat breeding in
breeds. The current results are encouraging, showing Malawi: Gestation length, birth weights and growth of
that Saanen sires could be used on Malawi local goats the indigenous Malawi goats and their Saanen crosses.
to improve milk production. In: Rey В, Lebbie S H В and Reynolds L (eds), Small
The results also indicate that the goats perform ruminant research and development in Africa.
better in the dry season than in the rainy season. Similar Proceedings of the First Biennial Conference of the
results were reported by Karua (1989) among village African Small Ruminant Research Network, ILRAD,
Nairobi, Kenya, 10-14 December 1990. ILCA
goats. Tethering was given as the major reason for loss (International Livestock Centre for Africa), Nairobi,
of condition. The likely reason for difference in Kenya. pp. 453-459.
performance could be a higher worm burden with the
Kasowanjete D, Stotz D and Zerfas H P. 1987. Goat
onset of rains (Eldestein, 1988; Senganimalunje, 1992;
development programme in Malawi. The Small
Banda, 1992). Ruminant and Camel Group Newsletter 8: 1 5-2 1 . ILCA
The better performance seen in 1988 could be (International Livestock Centre for Africa), Addis
management or environment related. Production is Ababa, Ethiopia.
likely to fall when there is drought or too much rain, both
Mchiela A F. 1989. Milk production and growth rates ofgoats
of which reduce grazing periods and intakes. Unless the maintained under free range, tethered and confined
animals' feed supply is supplemented, productivity is management systems. MSc thesis. University of Malawi,
likely to be affected. Malawi. 1 19 pp.

84 Small Ruminant Research Network Workshop

 The performance of the Small East African goats etc...

Munthali J T. 1987. Cattle fattening on basal diets of maize NSSA (National Sample Survey of Agriculture). 1981/82.
stover and groundnut tops in Malawi. In: Little D A and Department ofNational Statistics. Government Printer,
Said A N (eds), Utilisation of agricultural by-products Zomba, Malawi. April 1984.
as livestock feeds in Africa. Proceedings of a workshop Sahni K L and Chawla D S. 1982. Crossbreeding of dairy goats
held at Ryall's Hotel, Blantyre, Malawi, September for milk production. In: Proceedings of the Third
1986. ILCA (International Livestock Centre for Africa), International Conference on Goat Production and
Addis Ababa, Ethiopia. pp. 45-52.
Disease, 10-15 January 1982, Tucson, Arizona, USA.
Mwenifumbo A L and Phoya R K D. 1982. Composition and pp. 575-583.
milk yield of Malawian local goats. Tropical Animal Senganimalunje T C. 1992. The effect ofdry season treatment
Production 7:71. on helminth infestation in goats. BSc Dissertation,
University of Malawi, Malawi.
NSO (National Statistical Office). 1988. Malawi population
and housing census 1987. National Statistical Office,
Zomba, Malawi. 28 pp.

Small Ruminant Research Network Workshop 85

 The productivity of Small East African sheep and goats
in Malawi
J.W. Banda

University ofMalawi, Bunda College ofAgriculture

P. 0. Box 219, Lilongwe, Malawi

Abstract La production de lait sur une période de 12

semaines était en moyenne de 79,0 kg (± 2,0 kg) chez la
The total and additional biological productivity of local chèvre et de 54,5 kg (±1,3 kg) chez la brebis. La
(LL) and Boer (BB) goats and their crosses (BL) and of différence était significative (P<0,001). Les chèvres LL
the local (LL) and Dorper (DD) sheep and their crosses et BB produisaient du lait en quantités à peu près égales
(DL) were estimated over two kidding seasons at the (73,7 ±2,5 kg et 78,3 ±3,8 kg respectivement) mais
Lifidzi Goat Breeding Centre, Salima, Central Malawi. inférieures aux quantités produites par les mères BL
The average 12-weeks milk production was 79.0 (85,1 ± 2,3 kg). Les brebis LL, DL et DD produisaient
(+ 2.0 kg) in goats and 54.5 (± 1 .3 kg) in sheep. The respectivement 41,4 ± 1,8, 53,5 ± 1,6 et 68,7 ± 2,1 kg
difference was significant (P<0.001). The yields of LL de lait et les différences entre génotypes étaient
(73.7± 2.5 kg) and BB (78.3 ± 3.8 kg) were similar, but significatives (P<0,001).
lower (P<0.001) than those of BL does (85.1 ± 2.3 kg).
Le gain moyen quotidien de la naissance à l'âge
LL, DL and DD ewes produced 41.4 ± 1.8, 53.5 ± 1.6
de 17 semaines était de 87,0 (± 3, 5 g) chez les chevreaux
and 68.7 ±2.1 kg milk respectively and the differences
et de 128,9 (±2,8 g) chez les agneaux. Les taux de
among these genotypes were significant (P<0.001).
conversion étaient respectivement de 15,6 et 11,9 kg de
The average daily gain (ADG) from birth to 17
lait/kg de poids vif. La mortalité des chevreaux était en
weeks of age was 87.0 (± 3.5 g) for kids and 128.9 (±
moyenne de 18,9%, celle des agneaux de 5,3%.
2.8 g) for lambs. Milk conversion was 15.6 and 1 1.9 kg
milk/kg liveweight gain in kids and lambs, respectively. Les indices I, II et 111, qui représentent la
The average mortality was 18.9 and 5.3% in kids and productivité biologique apparente totale, étaient
lambs respectively. respectivement de 16,6,0,47 et 1, 15 kg pour les chèvres
The total apparent biological productivity indices et de 24,9, 0,80 et 1,89 kg pour les moutons. Les indices
represented by indices I, II and III were, respectively, IV, V et VI, qui représentent la productivité biologique
16.6, 0.47 and 1.15 kg for goats. For sheep, the apparente complémentaire, étaient respectivement de
corresponding values were 24.9, 0.80 and 1.89 kg. 7,08, 0,204et0,495kgpourleschèvresetde6,13, 0,197
Additional apparent biological productivity represented et 0, 464 kg pour les moutons. Il découle de ces résultats
by indices IV, V and VI were 7.08, 0.204 and 0.495 kg que les deux espèces diffèrent sur le plan de la
respectively for goats and 6.13, 0.197 and 0.464 kg for production de lait mais que cette différence n 'est pas
sheep. These results suggest that although the two importante.
species differ in milk production, the difference in
production is not substantial.

Introduction
Productivité des moutons et des
chèvres de l'Afrique de l'Est au From 20-60% of the variation in weaning weight of kids
and lambs in the tropics and sub-tropics is accounted for
Malawi by the variation in the amount of milk produced by the
dams or the amount suckled by lambs and kids (Peart,
Résumé 1982). Genotypes differ in this ability. Mortality prior to
weaning, lambing/kidding interval, weight and milk
Les productivités biologiques totale et production of the dams are major factors influencing the
complémentaire de chèvres de races locale (LL) et Boer productivity of sheep and goats. These individual
(BB) et des produits de leur croisement (BL), ainsi que variables can be combined into indices termed either
de moutons locaux (LL) et Dorper (DD) et des produits total apparent biological productivity or additional
de leur croisement (DL) ont été déterminées sur deux apparent biological productivity. These indices can then
saisons de mise-bas au Centre d'élevage caprin de be used to compare the reproductive efficiency in
Lifidzi à Salima (centre du Malawi). species and/or populations which differ in mature size.

Small Ruminant Research Network Workshop 87

J. W. Banda

In Malawi, there is little information on the pro Method 1

ductivity of sheep and goats. The objectives of this study
were to evaluate the present and potential milk Index I = Weight of weaned lamb or kid produced
production capacity of sheep and goats in Malawi and per dam per year (kg)
calculate and compare their apparent biological
productivity. = litter weight at 120 days (weaning age) X
365/subsequent lambing or kidding
interval

Materials and methods Index II = Weight of lamb or kid/liveweight of dam

per year (kg)
The study was carried out at the Lifidzi Goat Breeding = Index I/dam post-partum weight (kg)
Centre near Salima in Central Malawi at an altitude of
600 metres above sea level. Index III = Weight of lamb or kid produced/dam
Data were collected from 1 1 local (LL), four Boer metabolic weight per year (kg)
(BB) and lOcrossbred (BL) does and from 12 local (LL), = Index I/dam post-partum weight '
15 Dorper (DD) and 17 crossbred (DL) ewes in the dry
season. Almost 80% of the experimental animals were Method 2
used again in the subsequent rainy season. Milk yield
was estimated by the suckling, handmilking and Index IV = Additional lamb or kid weight
oxytocin + handmilking techniques on three different produced/dam per year (kg)
days in each of the 12 weeks after parturition. The
= Milk yield X milk conversion rate X
oxytocin dose was 10 i.u. (1 ml) per animal. All these
litter size
factors were arranged ina2x3x2x3 factorial design.
For the dry season trials, does and ewes were mated Index V = Additional lamb or kid weight produced/
to bucks and rams of their own genotype between kg liveweight of the dam
December 1987 and February 1988. Milking
commenced in May 1988. For the rainy season trials, the = Index IV/dam post-partum weight (kg)
same animals were mated to a Boer buck and Dorper
Index VI = Additional lamb or kid weight
ram between September and October 1988. Milking
produced/dam's metabolic weight (kg)
started in January 1989. All animals grazed natural
pastures. During milking only, animals on the trial = Index IV/dam post-partum weight '
received a concentrate mixture composed of 9 1 % maize
bran, 8% dried leucaena leaves and 1% common salt Dependent variables were milk yields, liveweights and
(NaCl) at a rate of 300 g per animal/day. The concentrate ADG of kids and lambs. Fixed least-squares models
mixture contained 12.7% crude protein and 19.7 MJ described by Harvey ( 1 977) for use on data with unequal
gross energy (GE) or about 13 MJ metabolisable energy subclass numbers were used for analysing the milk
(ME)/kg dry matter (DM). yields of dams and the liveweights and ADG of kids and
lambs. No statistical analysis was conducted on the
The milk yield of each animal was measured every
productivity indices calculated.
week for the 1 2 weeks of lactation. On test days, kids
were separated from their dams for two four-hour
periods. For the suckling method, the young were
weighed before and after suckling. The difference in Results and discussion
weight was taken as the milk production during the
separation period. For the hand-milking technique, Milk yield
animals were milked out at 1 100 and 1500 h. The daily
milk yield for both methods was calculated by summing The total milk production of goats was 45% higher than
up the two four-hourly yields and multiplying the results that of sheep (Table 1). Based on mean liveweight, goats
by three. The oxytocin technique (oxytocin + produced 26.0 g milk/day per kg body weight or 63.0
g/day per kg ' . Sheep produced only 19.0 g milk/dav
hand-milking) was done by milking the animals after
per kg body weight or 45.0 g milk/day per kg
injection of 10 i.u. of oxytocin at 1 100 h. The amount of
Economides (1986) compared Damascus goats and
milk obtained during this single separation period was Chios sheep kept under the same feeding and
multiplied by six to obtain daily milk production. management practices and obtained 1 1-42% more milk
Liveweights of the kids and lambs were taken at from goats than from sheep. Previous reviews and
birth and at weekly intervals until weaning at 17 weeks. reports have compared goats, sheep and other ruminants
Records of litter sizes at birth and at weaning, mortalities for milk production. It has been concluded that goats
of kids and lambs, kidding/lambing interval and dam produce more milk for the following reasons: better feed
weights were kept and used in the calculation of the utilisation efficiency; higher lactation persistency;
biological productivity indices of the dams. The indices mammary tissue comprising a greater proportion of the
in the two species were calculated in two ways as body weight; and a more pronounced milk ejection
indicated below: reflex (Devendra, 1975; Morand-Fehr and Sauvant,

88 Small Ruminant Research Network Workshop

 The productivity of Small East African sheep and etc..

Table 1. Least squares means (kg). standard errors (se) and t-values for total milk yields of sheep and goats.
Main effect Sheep Goats t-value
Sub class No. obs. Mean ± se No. obs Mean ± se Sheep vs Goats
Overall 250 54.5 1.30 163 79.0 2.00 -12.24*
Genotype
Local 84 41.1" 1.80 65 73.7" 2.50 -10.48***
Crosses 102 53.5b 1.60 76 85.1b 2.30 -11.28***
Exotic 64 68.7C 2.10 22 78.3a 3.80 -2.21**
abc = means within variable groups bearing different superscripts differ significantly (P<0.05).
*** P< 0.001 ; ** P<0.05. No. obs. = number of observations.

1978; Devendra, 1980; Devendra, 1981; Economides, From local to exotic sheep, there was a linear
1986; Jenness, 1986; Zometa et al, 1987). increase in milk yield while from local to exotic goats,
In this study, the higher initial milk yield (56.8%) there was a quadratic response with Boer x local
and higher persistency (6.3%) in goats may have led to crossbreds producing the highest yields, indicating 12%
higher milk production than sheep. Although not heterosis. Indications are that crossbreeding improves
measured, the size of the udders of goats in relation to milk production in the local small ruminant species.
their body weight as explained by Jenness (1986) may
also have contributed to this difference. The absence of Weight gains and milk conversion rates
total feed intake data limits a discussion on the
efficiency of milk production with respect to nutrition. A summary of the liveweights, liveweight gains and
This implies that there is a need to conduct more milk conversion efficiency of kids and lambs is shown
research in order to ascertain whether the differences in Table 2. It is quite clear that lambs weighed more and
could also have been partly due to feed utilisation grew faster than kids, irrespective of the stage of growth.
efficiency. This has been demonstrated previously (Economides,
The milk yields of sheep and goats used in this 1986; Wilson, 1986). Consequently, kids required 15.6
study are much lower than those recorded for these litres of milk for each additional kg of weight gained
while lambs required only 12.0 litres.
species in temperate, Indian and Mediterranean areas for
Economides (1986) further observed that lambs
recognised dairy, meat and general-purpose breeds. The consumed more solid feed and started doing so at an
concentration of constituents in their milk is, however, earlier age than kids. It is not clear whether the
higher than that for the temperate breeds, but differences observed in the present study are due to
comparable to that for other tropical/subtropical breeds differences in solid food intake, as no data were
as a consequence of lower milk yield (Devendra and collected on this aspect. Generally, the results suggest
Burns, 1983; Gatenby, 1986). that it would be more efficient to rear lambs than kids

Table 2. Live weights and liveweight gains of lambs and kids from birth to six months of age (two months post-weaning) and
their pre-weaning milk conversion efficiency.
Lambs (L) Kids (K) t-value
Mean ± se Mean ± se Lambs vs Kids
No. of observations 236 152
Birth weight (kg) 3.19 0.05 2.77 0.08 4.45***
12-week weight (kg) 14.07 0.25 10.07 0.32 9.85***
17 week weight (kg) 18.52 0.32 12.99 0.41 10.63***
Gain (g/day)
Birth-12 weeks 129.4 2.8 87.9 3.2 9.76***
12-17 weeks 128.5 3.5 87.0 3.6 8.27***
Overall 128.9 2.8 87.0 3.5 9.35***
Milk conversion (kg gain/litre milk)
1-12 weeks 0.084 0.008 0.064 0.011 !.47 ns
Weeks 1-4 0.098 0.008 0.049 0.012 1.66**
Weeks 5-8 0.076 0.009 0.068 0.014 0.30 ns
Weeks 9-12 0.058 0.012 0.059 0.012 0.06 ns
*** P<0.001; ns = not significant; ** P<0.05.

Small Ruminant Research Network Workshop

J. W. Banda

Table 3. The apparent total biological productivity of sheep the milk yield level of the does and the larger litter size
and goats. per annum (Table 4). When the metabolic body weight
Goat vs. of the dam was considered, the difference between the
Sheep two species narrowed. This was due to the lower weight
Parameters Goats Sheep (%) of the sheep.
Offspring weaning weight (kg) 12.99 18.52 -30.0 Thus, although the two species differ in milk
Litter size at weaning 1.05 0.98 7.1 production and annual litter size, the difference in
Total weaning weight (kg) 12.41 18.15 -31.6 overall production is not very great.
Kidding/lambing interval (days) 275.6 266.5 3.4
Dam weight (kg) 34.7 31.1 11.6
Dam weight (kg ) 14.3 13.2 8.4 References
Productivity indices (kg)
Devendra С. 1975. Biological efficiency of milk production in
Index I 16.4 24.9 -34.2 dairy goats. World Review of Animal Production
Index II 0.47 0.80 -51.2 11:46-53.
Index m 1.15 1.89 -39.2 Devendra С. 1980. Milk production in goats compared to
buffalo and cattle in the humid tropics. Journal ofDairy
Science 63:1755-1767.
for meat production, especially during the first four
Devendra C. 1981. Potential of sheep and goats in less
weeks. However, valid conclusions can only be reached
developed countries. Journal of Animal Science
when other factors, e.g. solid feed intake, costs of 51:461^173.
treatments, mortality and reproduction are considered.
Devendra C and Burns M. 1983. Goat production in the
tropics. 2nd ed. Commonwealth Agricultural Bureaux,
Farnham Royal, UK. pp. 64-73.
Dam productivity Economides S. 1986. Comparative studies of sheep and goats:
milk yield and composition and growth rate of lambs and
The measurements of biological productivity as total kids. Journal of Agricultural Science (Cambridge)
weight of offspring at weaning per female per year 106:477-484.
(Index I), per kg dam weight per year (Index II) and per
kg dam weight ' per year (Index II) in both sheep and Gatenby R M. 1986. Sheep production in the tropics and
goats are given in Table 3. Productivity was highest in subtropics. lsted. Longman, London, UK. pp. 202-225.
LL does and lowest in BB does, with that in BL does Harvey W R. 1977. User's guide for LSML76. Mixed model
being closer to that of LL does. least-squares maximum likelihood computer program.
The LL ewes outperformed the DL and DD ewes, Ohio State University, Columbus, Ohio, US.
with the last showing the lowest performance. The ewes (Mimeograph).
were superior to does overall. Jenness R. 1986. Lactational performance of various
A summary of the additional apparent productivity mammalian species. Symposium: Species Variation in
of the two species based on milk production and milk Mammary Gland Function. Journal of Dairy Science
conversion efficiency by the offspring as well as weight 69:869-885.
of the dams is presented in Table 4. Morand-Fehr P and Sauvant D. 1978. Nutrition and optimum
The additional productivity per female was higher performance of dairy goats. Livestock Production
in goats than in sheep. The difference was largely due to Science 5:203-213.

Table 4. The apparent additional biological productivity of sheep and goats based on milk conversion.
Goats vs Sheep
Parameters Sheep ± se Goats ± se (%)
No. of observations 236 152
Milk yield (kg) 54.5 1.3 79.0 0.2 45.0
Milk conversion rate (lambs & kids) 0.084 0.008 0.064 0.011 -23.8
Dam weight (kg) 31.1 0.2 34.7 0.3 11.6
Dam weight (kg0 75) 13.2 14.3 8.3
Litter per annum 1.34 1.40 4.5
Productivity indices (kg)
per female 6.13 7.08 15.5
per kg dam weight 0.197 0.204 3.6
per kg dam weight ' 0.464 0.495 6.7

90 Small Ruminant Research Network Workshop

 The productivity of Small East African sheep and etc...

Peart J N. 1982. Lactation of suckling ewes and does. In: Coop Zometa C A, Cunha M G G, Shelton M, Sousa W H and Leite
I E (ed), Sheep and goat production. World Animal P R M. 1987. Comparative lactation performance on
Science С 1. Elsevier, Amsterdam, The Netherlands, native goats and cows in semi-arid northeastern Brazil.
pp. 119-134. In: Santana О P, Gabriel da Silva H and Foote W C (eds).
Proceedings of the IV International Conference on
Wilson R T. 1986. Livestock production in Central Mali:
Goats. Volume II. Departamento de Defusäo de
Long-term studies on cattle and small ruminants in the
Technologia-DDT, Brasilia, Brazil, 8-13 March 1987.
agropastoral system. ILCA Research Report 14. ILCA
(International Livestock Centre for Africa), Addis p. 1547. (Abstract no. 346).
Ababa, Ethiopia. pp. 82-95.

Small Ruminant Research Network Workshop 91

 SESSION 3

Small Ruminant Health and

Reproductive Wastage
 Trypanosomiasis induced reproductive wastage in
West African Dwarf sheep
O. O. Akinbamijo and L. Reynolds

Agricultural University, Wageningen, The Netherlands

International Livestock Centre for Africa
P.O. Box 80147, Mombasa, Kenya

Abstract la gestation et de la lactation, la quantité de matière

organique digestible ingérée était inférieure (P<0,05)
This study reports the results of investigations on the chez les brebis infectées. Il n 'y avait pas de différence
effects of Trypanosoma vivax infection in pregnant and entre l'accroissement pondéral des animaux infectés et
lactating West African Dwarf ewes on digestible non infectés pendant quatorze semaines après
organic matter intake (DOMI), live weight changes, l'agnelage (huit semaines après l'infection). A la
milk yield pattern, lamb birth weight and growth rate. parturition, les brebis infectées affichaient un poids et
During pregnancy and lactation, DOMI was lower un état corporel médiocres, associés à de faibles
(P<0.05) in infected ewes. Rate of weight gain did not (P<0,01) poids à la naissance et à de faibles taux de
differ between infected and control animals before 14 survie des agneaux. Le taux d'avortement était de 15%
weeks post-breeding (eight weeks post-infection). chez les brebis infectées, avec un taux de mortalité des
Infected ewes had low maternal weights and poor body agneaux pouvant atteindre 85%. Le poids au sevrage
condition at parturition, which were associated with low des agneaux nés de mères infectées était inférieur à celui
(P<0.01) lamb birth weights and survival. Abortion rate des agneaux nés des brebis témoins.
was 15% in the infected ewes with a lamb mortality rate
Pendant la première phase de la lactation,
of up to 85%. Lambs born by infected dams had lower
l 'infection n 'influait ni sur le rendement laitier ni sur la
weaning weights compared with lambs from control
composition du lait. En fin de lactation toutefois, les
ewes.
brebis infectées produisaient moins (P<0,05) de lait et
In lactating dams, neither milk yield nor
perdaient plus de poids que les brebis allaitantes non
composition was affected by infection during the first
infectées. Au cours de la période présevrage, le taux de
half of lactation. However, during late lactation infected
croissance des agneaux allaités par les brebis témoins
ewes produced less (P<0.05) milk and lost more weight
était le même que celui des agneaux allaités par les
compared to uninfected lactating ewes. Growth rates of
brebis infectées. Il ressort des résultats que l 'infection à
lambs nursed by control and infected dams did not differ
Trypanosoma vivax était la cause des avortements
in the pre-weaning [Link] findings suggest that
observés, de la mortalité des brebis, du faible gain de
Trypanosoma vivax infection was responsible for the
poids des brebis gravides et dufaible poids des agneaux
observed abortions, ewe mortality, reduced ewe live
à la naissance et au sevrage. Les auteurs constatent
weight gain and lamb birth and weaning weight in
également que les performances des agneaux, la
pregnant ewes. It is also concluded that infection ofearly
production laitière et la composition du lait n'étaient
lactating dams did not affect lamb performance, milk
pas affectées par l'infection des mères au début de la
yield and composition although significant differences
lactation, encore que des différences significatives de
were encountered in milk yield during late lactation.
production laitière soient apparues enfin de lactation.

Pertes en reproduction dues à la

trypanosomiase chez le mouton Introduction
Djallonké The role of abortion and perinatal losses in causing
heavy losses in animal production has been discussed by
Résumé Sellers and Leech (1955). Regrettably, studies on
trypanosomiasis as a cause of reproductive losses in the
Cet article présente les résultats d'une étude sur West African Dwarf (WAD) sheep have been rare. In a
les effets de l'infection à Trypanosoma vivax sur livestock improvement scheme, an interruption of the
l'ingestion de matière organique digestible, l'évolution reproductive process can result in death of offspring,
pondérale, le rendement laitier, et le poids à la loss of a reproductive season for the dam and a reduction
naissance et les taux de croissance des agneaux chez des of profit to the producer. Most of the studies with
brebis Djallonké gestantes ou allaitantes. Au cours de trypanosomiasis in the perinatal period were undertaken

Small Ruminant Research Network Workshop 95

O. O. Akinbamijo and L. Reynolds

with cattle and very little information is available on the weaned 12 weeks after birth. Lamb performance was
incidence of trypanosome-induced reproductive measured by the estimated pre-weaning live weight
wastage in West African Dwarf sheep and goats. gain. Survival rate of the lambs was calculated as a ratio
The subject of perinatal losses and trypanosome- of weaned lambs to lambs born.
induced reproductive disorders is well reviewed by
Gunn (1983) and Ikede et al (1988). Owen (1976)
reported that the main difference in lamb growth in the
Housing and management
first half of lactation is due to differences in the dam milk All experimental ewes were individually penned in two
yield. This view was also expressed in the studies of separate, well ventilated, and naturally illuminated
Reynolds and Ekwuruke (1988), Robinson et al (1969), housing units. The infected ewes were kept in a
Peart ( 1 967) and Adu et al ( 1 974). fly-screened housing unit while the uninfected ewes
The present study was aimed at investigating the were housed in an open barn. Routine station health
effect of trypanosomiasis on pregnant and lactating management was performed as described by Reynolds
WAD ewes. Factors investigated were pre- and and Ekwuruke (1988).
post-natal ewe weight changes, lamb birth weight and
lamb performance following T. vivax infection at two
stages of pregnancy. The effect of the infection on milk
yield and composition following infection post-partum Experiment 1
was also studied. This experiment used 38 West African Dwarf ewes aged
24—36 months. The sheep were randomly allotted into
infected (n = 20) and control (n = 18) groups. The ewes
Materials and methods of the infected group were subdivided into groups 1 A
(infected, mid pregnancy; n=10) and IB (infected, late
The experiment was conducted at the International pregnancy; n=10). The uninfected control ewes were
Livestock Centre for Africa (ILCA) Humid Zone station divided into groups 1C (control, restricted diet; n= 8)
in Ibadan, Nigeria. and ID (control, ad libitum diet; n=10). Animals in
All experimental animals were screened for T. groups 1A and IB were inoculated in the seventh and
vivax antibodies pre-infection (Katende et al, 1987) and 15 week post-breeding to simulate infection during
for trypanosomes (Murray et al, 1983) at the start of the mid and late pregnancy respectively.
experiment and were confirmed to have no previous All animals were offered a basal diet of 4 kg
history of trypanosomiasis. Panicum maximum except for group 1С which received
The pathogen used was T. vivax prepared from the 2 kg Panicum maximum from weeks 7-21 of pregnancy.
Kaban strain isolated in 1984 from an ox in Kaban Browse supplement of 550 g/day of Gliricidia sepium
village, Plateau State of Nigeria. Each inoculated ewe and Leucaena leucocephala as well as 50 g/day dried
was infected artificially by intravenous needle challenge cassava peels were also offered to all animals. The feed
using 4 X 10 trypanosomes in heparinised medium. compositions are presented in Table 1 . After parturition,
Infection was confirmed by thin/thick blood smears and all animals were offered similar rations ad lib. Drinking
water and mineral licks were offered ad lib throughout
by the phase contrast buffy coat technique (Murray et al,
the experiments.
1983).
All dams were given therapeutic doses of Berenil
Table 1. Mean composition (g/I00g) offeed offered on dry-
(Diminazine aceturate 3.5 mg/kg body weight) at the matter basis.
end of their respective studies, when abortion occurred,
or when packed cell volume (PCV) was less than 12% Dry Crude
Feed stuff matter nitrogen Ash
for two consecutive weeks.
Panicum maximum 17.42 1.70 14.91

Measurements Dried cassava peels 70.37 0.734 11.59

Gliricidia sepium 31.94 3.69 9.88
Blood samples were taken twice weekly to estimate
Leucaena leucocephala 30.67 3.68 9.21
PCV and parasitaemia. The mean value of the
determinations was taken to be the animals' average
PCV for that week. Rectal temperature (RT) was
measured daily in all experimental animals between Experiment 2
0800h and 0900h before feeding. Intake of digestible
organic matter was measured during the eighth week of This experiment used twenty WAD ewes in their second
gestation (Experiment I) and during the third and ninth and third parity (age 24-36 months), born and raised on
weeks post partum (Experiment 2). station. Experimental animals were all oestrus-
All dams were weighed once a week before synchronised using 2 ml analogue of prostaglandin
feeding through the gestation and lactation period. All F2-alpha to facilitate synchronised parturition. All
lambs were weighed within 24 hours of birth and live animals received identical veterinary treatment before
weights were monitored weekly thereafter. Lambs were and during gestation.

96 Small Ruminant Research Network Workshop

 Trypanosomiasis induced reproductive wastage in etc...

Experimental animals were randomly allocated Figure 1. Group average packed cell volume in T. Vivax
into three groups shortly after parturition: infected infected and uninfected pregnant West African
dwarfewes.
(n=10); control and fed restricted diet (n=5); and control
fed ad lib (n=5). These constituted groups 2A, 2B and
2C respectively. Infected animals were inoculated one %PCV
week post-partum and treated at the end of the trial with 30 -i
Berenil.
Animals in groups 2A and 2C were offered 3
kg/day of Gliricidia sepium, 1.5 kg/day of Leucaena
leucocephala and 50 g/day dried cassava peels on as fed
bastSvOroup 2B ewes received 50% of the ration offered 25
groups 2A and 2C. Composition of feed offered is
presented in Table 1. All animals had free access to
drinking water and mineral licks.
Data were collected for 12 weeks after parturition.
From the seventh day of lactation the milk yield of each 20 -
ewe was estimated weekly for 1 1 successive weeks.
Milk production over 24 hours was measured in each
ewe using the oxytocin method (Coombe et al, 1960;
Owen, 1957).
15

13 I I I I I I I I I I I r^
Metabolism study 5 IM 10 IL 20

Three sets of feed intake trials (one in Experiment 1 and Weeks of gestation
two in Experiment 2) were conducted. Each metabolism
study lasted for three weeks, the first two being an
adaptation period. In the pregnant ewes, a feed intake
trial was conducted during the eighth week of pregnancy
using five ewes each from groups 1 A, 1C and ID. Results
During Experiment 2, all animals were used in the
metabolism studies conducted at two phases of lactation
— early and late. Experiment 1
Effect of T. vivax on pregnant WAD sheep.
Chemical analysis Packed cell volume declined rapidly following
infection (Figure 1 ) with PCV in infected ewes dropping
Milk samples were analysed for fat by the Gerber by 40% from pre-infection values. All infected dams
method (BSI, 1955). Crude nitrogen, ash and total milk remained febrile throughout the experiment. Mean (±
solids were determined by the methods of AOAC ( 1 975) SD) rectal temperature was 40.4±0.95°C compared with
while non-fat solids were determined by difference.
38.6 ± 0.39'C in the controls.
Six of the experimental animals did not carry their
pregnancy to term and were excluded from the live-
Statistical analysis weight studies. Three ewes (two from group 1 A and one
Data in treatment groups are summarised as mean ± from group IB) aborted while one each from groups 1 A
standard error (SE). Liveweight data collected in and 1 В died from the infection. One ewe (group 1 A) was
Experiment 1 were analysed in four periods, early removed from the experiment due to anaemia. All
pregnancy, mid pregnancy, late pregnancy and 1 0 weeks control ewes carried their pregnancy to term.
post natal. Linear regression models of weight on
gestation period and of weight on time during the first
10 weeks post partum were fitted for each animal. The Feed intake and digestibility during
estimated regression coefficients representing the rate of pregnancy
live weight change for each ewe were further analysed
using analysis of co-variance for comparing the different
treatment groups during pregnancy and post-natal Mean values of intake and digestibility data are
phases. Corrections were made for weight at breeding presented in Table 2. While dry matter intake (DMI),
and for the number of lambs born and nursed as digestible organic matter intake (DOMI) and nitrogen
co-variates. Milk yield and lamb performance data were intake (N1) were lower (P<0.05) in infected ewes (Group
analysed on an individual animal basis from the seventh 1A) compared to control ewes fed ad lib (Group ID),
day of lactation. Milk yield and dam and lamb weight there were no significant differences measured between
gains were analysed in three lactation periods, early (1-6 groups 1A and 1C in all parameters. The digestibility
weeks), late (7-12 weeks) and entire (1-12 weeks). coefficients did not differ between groups.

Small Ruminant Research Network Workshop 97

О. О. Akinbamijo and L. Reynolds

Table 2. Mean intake and digestibility performance of Table 4. Growth performance (kg) of Iambs produced by T.
WADS during mid-pregnancy under different vivax infected and control ewes.
dietary treatments.
Lamb growth rate (g/day)
Groups Birth
weight Early Late Weaning
Variables 1A 1C ID SE lactation lactation (kg)
Groups (kg)
n 5 5 5
88.0a 26.0a 5.0a
50.7a 51.2a 62.4b 1A 1.4°
DMI 1.75
44.4a 44.5a 54.5b IB 1.0a ND ND ND
OMI 1.50
26.83a 25.8a 32.7b 1.9е 76.0a 39.0a 7.1b
DOMI 1.02 1C
0.75a 0.83b 0.85b 2.0е 77.0a 33.0a 7.1b
N1 0.04 ID
62.5a 60.8a 62.6a Means within rows with different superscripts are different from each
DMD 1.02
60.3a 58.1a 60.1a other (P<0.05).
OMD 1.06
ND = not determined.
Means within rows with different superscripts are different from each
other (P<0.05).
DMI = Dry matter intake. DOMI = Digestive organic matter intake.
Experiment 2
OMD = Organic matter digestibility. OMI = Organic matter intake.
N1= Nitrogen intake. DMD = Dry matter digestibility. Feed intake

Liveweight changes Feed intake data for early and late lactation periods are
presented in Table 5. All measured parameters differed
Table 3 shows the rates of liveweight changes. In the (P<0.05) across the groups, except nutrient digestibility,
early pregnancy phase, all groups' estimated rates of which was not affected by the infection.
liveweight gains did not differ. Group 1A ewes lost
weight post-inoculation during mid-pregnancy. The Milk yield and composition
differences between the treatment groups, however,
were not statistically significant. During the late
Figure 2 shows the lactation pattern based on the mean
pregnancy period, infected and restrictedly-fed ewes
milk yield of the ewes. Rate of milk production in the
lost weight (P<0.05) compared to the controls fed ad lib.
first half of lactation was not affected by infection. Mean
In the post-natal phase, all ewes lost weight until lambs
daily milk yields on a group basis are presented in Table
were weaned.
6. Differences in mean daily milk production became
significant only during late lactation. Mean daily yields
Table 3. Rate of liveweight changes (g/day) of T. vivax
infected and control pregnant WAD ewes.
during both phases of lactation were contrasted and the
differences subjected to analysis of variance. The drop
Early Mid Late 10 weeks in daily milk production was highest in group 2A while
Group pregnancy pregnancy pregnancy post natal 2B and 2C did not differ from each other.
37.1a 15.7a -27.1a -38.6b
1A The difference in milk constituents became
IB 37.1a 32.9a -21.4a -25.7a significant during the late lactation period with infected
37.1a 37.1a 1.4b -52.9е ewes producing less (P<0.05) milk compared with
1С
37.1a 32.9a -75.7d control ewes.
ID 30.0° Mean component concentrations of the milk are
P * ** presented in Table 7. Milk total solids increased with
Means within rows with different superscripts are different from each lactation period. All other components except milk ash
other. followed this trend. The differences in these components
* = P<0.05; ** = P<0.01 between treatment groups were not statistically
significant.
Lamb performance
The mean birth weights are presented in Table 4. Ewe liveweight changes and lamb
Uninfected dams had heavier (P<0.05) lambs compared performance
with their infected counterparts. Only 15% of the lambs
in group IB survived longer than 48 hours post-partum, Rates of live weight changes over time are presented in
therefore the group was not included in the statistical Table 8. Infected ewes showed a rapid decline in weight
analysis of lamb performance. Analysis of variance gain immediately post-infection. This pattern was
showed that dam treatment and sex of lambs had no maintained until the end of lactation. Control ewes fed
effect on lamb growth rate. Estimated pre-weaning restricted diets lost weight during early lactation. This
growth rates are presented in Table 4. Lambs nursed by trend was reversed during late lactation when the ewes
uninfected dams were heavier (P<0.05) at weaning increased their rate of gain. Control ewes fed ad lib
compared to lambs nursed by infected dams. consistently gained weight throughout the study.

98 Small Ruminant Research Network Workshop

 Trypanosomiasis induced reproductive wastage in etc..

Figure 2. Average daily milk yield ofT. vivax infected and balance (Zwart et al, 1991; Verstegen et al, 1991),
control WAD ewes. increased basal metabolic rate (Zwart et al, 1991 and
Stephen, 1986), increased catabolism of tissue reserves
Milk yield (g/d) (Akinbamijo et al, 1992) and possibly uptake of host
nutrients by the parasites. Infected ewes were losing
500 weight and in a poor body condition towards end of
pregnancy. Compared with the findings of Adu et al
450 ( 1 974) and Osuagwuh and Aire ( 1 990), control ewes did
not differ in the liveweight pattern expected of pregnant
400 - WAD ewes. Weight changes in control ewes were
positive and consistent through pregnancy. The
350 - reduction in feed intake from mid-pregnancy onwards
was possibly masked by the effect of the previous level
300 of nutrition. The limited feed allowance may also have
been high enough to meet maintenance requirement and
250 support weight gains.
Infected Lamb birth weight and growth rate have been
200 control (restricted) described as functions of pregnancy nutrition and
maternal weight at parturition (Peart, 1967; Treacher,
150 control (ad lib) 1970; Adu and Olaloku, 1979; Gibband Treacher, 1980
and 1982). They are also strongly correlated with dams'
100 -1 1 1 1 1 1 1 1 1 г mothering ability. The findings of this study support
2 3 4 5 6 7 8 9 10 11 12
these views. All infected ewes ended pregnancy with
Weeks of lactation low maternal weights and a depressed feed intake, hence
lambs were born with lower birth weights. Control ewes
on the other hand had higher maternal weights at
parturition and hence heavier lambs.
Discussion Of the lambs born to ewes infected at late
Infection was confirmed by the presence of parasites in pregnancy, 85% died within 48 hours of birth. They
the buffy coat area (Murray et al, 1983) eight days after were all born weak as their dams were probably denied
infection in both trials. Inoculated animals were febrile adequate nutrition during pregnancy due to the
and anaemic. These two symptoms have been widely infection. Both dam and foetal nutrition have been
reported as hallmarks in the diagnosis of trypano described as critical during late pregnancy (Osuagwuh
somiasis. All animals gained weight during early and Aire, 1990). Although the ewes were treated after
pregnancy. The weight changes observed in the control lambing, weaning weights of lambs differed between
ewes fed ad lib were similar to those reported by Orji
groups mainly because of differences in birth weights.
and Steinbach (1980) in pregnant WAD ewes. During
the post-infection phase of this study, infected ewes In a complementary study (Experiment 2) with
(groups 1A and IB) were anorexic, anaemic and lost ewes infected immediately post-partum with T. vivax
weight. Weight loss in pregnant ewes was reported by (same strain as in experiment 1), all ewes had adequate
Reynolds and Ekwuruke ( 1 988) due to trypanosomiasis. nutrition during pregnancy. There were sufficient body
This is associated with reduction in dry matter intake reserves to meet the daily milk yield demand during
(Akinbamijo et al, 1990), reduced nitrogen and energy early lactation. Lactation is generally associated with an

Table 5. Mean intake and digestibility performance of WADS during early and late lactation under different dietary treatments.
Early lactation (Mean ± SE) Late lactation (Mean ± SE)
Variables 2A 2B 2C p 2A 2B 2C P
n 10 5 5 10 5 5 -
DMI 81.6±2.98b 61.5±4.21a 96.0±4.21c 0.001 76.4±2.98b 63.8±4.21a 89.6±4.21c 0.0001

OMI 74.4±2.74b 55.7+3.87* 7.9±3.87c 0.001 69.1±2.74b 57.5+3.87* 81.2±3.87c 0.0001

DOM I 47.2±2.23b 33.9±3.16a 54.0±3.16c 0.001 41.8±2.23b 35.7±3.16a 49.5±3.16c 0.0002

N1 2.53±0.11b 1.8±0.6* 3.01±0.16c 0.001 2.36±0.11b 1.91 ±0.1 6* 2.83±0.16b 0.0001

DMD 56.9±1.55 54.1+2.19 54.8±2.19 0.500 52.6+1.55 54.2+2.19 53.2±2.19 0.5004

OMD 63.0±1.22 60.8il.73 60.8±1.73 0.73 60.3±1.22 62.0+1.73 60.8±1.73 0.7288
DMI = Dry matter ¡make. DOMI = Digestive organic matter intake.
OMD = Organic matter digestibility. OMI = Organic matter intake.
N1 = Nitrogen intake. DMD = Dry matter digestibility.

Small Ruminant Research Network Workshop 99

0.0. Akinbamijo and L. Reynolds

Table 6. Mean daily milk production of T. vivax infected and control WAD ewes.
Groups (Mean ± SE)
Variables 2A 2B 2C P
Early lactation 426.4± 18.52" 417.6 ±26.2" 399.4± 26.2"
Late lactation 181.7±12.0" 238.0± 17.0b 240.1 ± 17.0b

Entire lactation 295.3 ±12.8" 318.2± 18.1" 312.0±18.1"

Mean difference 244.8 ± 17.4" 179.6±24.6b 159.3 ±24.6b

Means within rows with different superscripts are different from each other.
* = P<0.05;** = P<0.01.

Table 7. Milk composition (g/lOOg milk) of control and infected dams during early and late lactation.
Early Lactation Late Lactation
Group 2A Group 2B Group 2C Group 2A Group 2B Group 2C
Total milk solids 15.49 15.62 15.42 17.92 17.78 18.72
(0.74) (1.06) (1.04) (0.46) (0.32) (0.46)
Milk fat 5.77 5.49 5.62 7.13 6.43 7.05
(0.21) (0.30) (0.30) (0.20) (0.29) (0.29)
Milk solids non fat 9.72 10.13 9.8 10.79 11.35 11.67
(0.50) (0.72) (0.70) (0.30) (0.42) (0.42)
Milk protein 5.04 4.91 5.04 6.12 5.61 6.00
(0.04) (0.05) (0.05) (0.02) (0.03) (0.03)
Milk ash 0.97 0.95 0.98 0.92 0.90 0.91
(0.05) (0.07) 0.07) (0.02) (0.07) (0.02)
Figures in brackets are SE.

Table 8. Live weight changes (SE) of T. vivax infected and control lactating WAD ewes and their lambs.

Ewe ¡group Lamb group

Variables
2A 2B 2C P 2A 2B 2C P

Initial weight (kg) 23.4 21.6 20.2 ** 2.13 1.94 1.96 NS

(0.64)a (0.91)ab (0.9 l)b (0.10) (0.14) (0.14)

Early lactation (g/day) -37.2 -7.3 30.86 ** 75.1 71.4 77.3 NS

(10.8)a ls^* 15.27b 5.1 7.1 7.1

Late lactation (g/day) -52.0 6.24 12.28 *** 34.9 37.6 35.1 NS
(6.28)a 8.89b 8.89b 4.5 6.4 6.4

Entire lactation (g/day) -45.0 -5.4 12.08 *** 59.25 53.00 55.53 NS

(5.66)" 8.00b 3.2 4.5 4.5

8.00b
Final weight (kg) 19.35 21.1 21.3 * 7.15 6.9 6.9 NS
(0.38)a 0.54b 0.54b 0.32 0.45 0.45

Means within rows with different superscripts are different from each other.
* = P<0.05 ** = P<0.01 ; *** = P<0.001; NS = not significant
Figures in brackets are SE

100 Small Ruminant Research Network Workshop

 Trypanosomiasis induced reproductive wastage in etc...

increase in feed intake. It is not surprising, therefore, that References

infected lactating dams were consuming more than 1 .5
times their maintenance requirement per unit metabolic Adu I F and Olaloku E A. 1979. A note on nutrition during late
size per day. Although the infected dams had a reduced pregnancy in West African Dwarf sheep. Animal
nutrition intake, lactation normally continues for some Production 28:123-126.
time even in the event of anorexia. The animal is then Adu I F, Olaloku E A and Oyenuga V A. 1974. The effects of
forced to catabolise body tissues which is associated energy intake during late pregnancy on lamb birth
with increased heat production. In bioenergetic terms, weights and lactation of Nigerian Dwarf sheep. Journal
Langland et al (1963) equate lg DOMI to 3.64 kcal ofNigerian Animal Production 1(2): 15 1-161 .
metabolisable energy (ME). Kearl (1982) suggested 105
Akinbamijo O O, Ademosun A A, Brouwer В O, Tolkamp В
kcal ME per unit metabolic size as the maintenance J and Zwart D. 1990. Effect of Trypanosoma brucei
requirement for healthy lactating ewes and about 130 infection on live weight, organic matter intake,
kcal ME for febrile lactating dams. The nutrient drain digestibility and N-balance in West African Dwarfgoats.
caused by an increased maintenance requirement further Trop. Vet. 8:140-148.
decreases the net energy available for production. In this Akinbamijo O O, Hamminga В J, Wensing Th, Brouwer В O,
study, 2A, 2B and 2C ewes had an excess of 67, 18.4 Tolkamp В J and Zwart D. 1992. The effect of T. vivax
and 128 kcal ME. In relation to the liveweight pattern, infection in West African Dwarf goats on ~nergy and
it could be adduced that ewes in groups 2A and 2B nitrogen metabolism. The Veterinary Quarterly (The
catabolised body reserves to supplement dietary Netherlands) 14(3):95-100.
nutrients required for milk production (Gibbs and AOAC (Association of Official Analytical Chemists). 1975.
Treacher, 1982). During the late lactation phase, the Official methods of analysis. 12th edition. AOAC,
milk yield differed significantly, with group 2A Washington DC, USA.
producing less milk than 2B and 2C.
Lamb birth weights were similar to those reported BSI (British Standard Institution). 1955. Gerber method for
the determination of fat in milk and milk products.
by Orji and Steinbach (1980). There was no difference British Standard Institution Report 6%. Part 2. p. 7.
between groups as all ewes were identically treated
during the pre-natal period. According to Owen (1976), Coombe J D, Wardrop I D and Tribe D E. 1960. A study of
lamb growth rate is strongly correlated with the dam's milk production of the grazing ewe with emphasis on
experimental technique employed. Journal of
milk yield during early lactation. This relationship, Agricultural Science (Cambridge) 54:353-359.
however, weakens with time as lambs begin to depend
less on milk as the sole nutrient supply. As milk yield Gibb M J and Treacher T T. 1980. The effect of ewe body
did not differ in early lactation, growth rate was also not condition at lambing on the performance of ewes and
different between groups. The difference in milk yield their lambs at pasture. Journal ofAgricultural Science
(Cambridge) 95:63 1-640.
during late lactation therefore had less effect on growth
rates, because lambs supplemented their milk intake Gibb M J and Treacher T T. 1 982 . The effect of body condition
from their dam's diet. A biological implication on and nutrition during late pregnancy on the performance
reproductive rate, however, is that loss of weight and the of grazing ewes during lactation. Animal Production
resulting poor body condition might lead to a delayed 34:123-129.
return to oestrus. In its extreme it could cause a cessation Gunn R G. 1983. The influence ofnutrition on the reproductive
of oestrus and hence increased parturition intervals. As performance of ewes. In: Haresign W (ed), Sheep
ewes infected during late pregnancy showed the highest production. Butterworths, London, UK. pp. 99-1 10.
level of lamb wastage, a prophylactic treatment is Ikede В O, Elhassan E and Akpavie S 0. 1988. Reproductive
suggested during late pregnancy to prevent the disorders in African trypanosomiasis. A review. Acta
incidence of perinatal losses. Tropica 45:5-10.
Ratende J M, Musoke A J, Nantulya V M and Goddereis В M.
1987. A new method for fixation and preservation of
Conclusion trypanosomal antigens for use in the indirect
immunofluorescence antibody test for diagnosis of
bovine trypanosomiasis. Tropical Medicine and
1. Trypanosomiasis induces loss of body condition in
Parasitology 38:41-44.
pregnant and lactating dams.
Murray M, Trail J C M, Turner D A and Wissocq Y (eds). 1983.
2. Animals infected during gestation had lambs with Livestock productivity and trypanotolerance: Network
low birth weights and poor performance. training manual. ILCA (International Livestock Centre
3. Milk composition was not affected by the disease for Africa), Addis Ababa, Ethiopia. 198 pp.
but yield was reduced during late lactation. Orji В I and Steinbach J. 1980. Gestation period and weight
changes during pregnancy in the Nigerian Dwarf sheep.
4. The performance of lambs nursed by infected dams Bulletin of Animal Health and Production in Africa
was not affected by trypanosomiasis. 28:356-364.
5. Trypanosomiasis caused foetal and neonatal losses Osuagwuh A I A and Aire T A. 1990. Intrauterine growth rates
in WAD sheep. of the West African Dwarf goat and some foetal organs

Small Ruminant Research Network Workshop 101

O.O. Akinbamijo and L. Reynolds

in relation to strategic feed supplementation during Sellers К С and Leech F В. 1955. Survey of losses associated
pregnancy. Journal of Veterinary Medicine 37: 198-204. with pregnancy and parturition in Yorkshire sheep.
Journal ofAgricultural Science (Cambridge) 46:90-96.
Owen J B. 1957. A study of the lactation and growth of hill
sheep in their native environment and under lowland Stephen L E. 1986. Trypanosomiasis — A veterinary
conditions. Journal of Agricultural Science perspective. First edition. Pergamon press, Oxford, UK.
(Cambridge) 48:287-311. 551 pp.
Owen J B. 1976. Sheep production. Balliere Tyndall, London, Treacher T T. 1970. Effects of nutrition in late pregnancy on
UK. 436 pp. subsequent milk production in ewes. Animal Production
12:23-36.
Peart J N. 1967. The effect of different levels of nutrition
during late pregnancy on subsequent milk production of Verstegen M W A, Zwart D, van der Hel W, Brouwer В О and
Blackface ewes and on growth of their lambs. Journal Wensing T. 1991. Effect of Trypanosoma vivax
ofAgricultural Science (Cambridge) 68:365-371. infection on energy and nitrogen metabolism of West
African Dwarf goats. Journal of Animal Science
Reynolds L and Ekwuruke J O. 1988. The effect of 69:1667-1677.
Trypanosoma vivax on West African Dwarf sheep at two
Zwart D, Brouwer В О, van der Hel W, van den Akker H N
planes of nutrition. Small Ruminant Research
and Verstegen M W A. 1991. Effect of Trypanosoma
1:175-188.
vivax infection on body temperature, feed intake and
Robinson J J, Foster W H and Forbes T J. 1 969. The estimation metabolic rate of West African Dwarf goats. Journal of
of milk yield of ewe from body weight data of suckling Animal Science 69:3780-3788.
lambs. Journal of Agricultural Science (Cambridge)
72-108.

102 Small Ruminant Research Network Workshop

 Normes et variations de l'hématocrite et de la formule
leucocytaire et particularités histologiques leucocytaires
chez la brebis Djallonké du sud du Sénégal
M. Ndiaye, M.B. Sow et T. Alogninouwa

Institut sénégalais de recherches agricoles/Laboratoire national d'élevage et de recherches vétérinaires

B.P. 2057, Dakar (Sénégal)

Résumé Ewes with no apparent clinical symptoms or

injuries and no bloodparasites detected in blood smears
Cet article est consacré à l'étude de l'hématocrite had an average PCV of 30.69%, with minimum and
et de la formule leucocytaire chez des brebis Djallonké maximum values of 30.02 and 31.34, respectively. The
adultes de la zone de Kolda, au sud du Sénégal. Les average differential leukocyte count was as follows:
données obtenues sur un total de 355 prélèvements neutrophils 33%, esinophils 9%, lymphocytes 55% and
effectués sur des animaux âgés de plus de 12 mois monocytes 2-3%.
permettent une bonne interprétation sémiologique des Variation in these values was estimated according
résultats de prospections cliniques faisant appel à la to infestation or non-infestation by blood parasites,
valeur de l'hématocrite et à la formule leucocytaire. body condition, season and management. Histologic
characteristics of leucocytes of West African Dwarf
Les brebis, dont l'examen clinique n'avait révélé sheep are specified.
ni symptôme ni lésion et dont l' examen du frottis de sang
réalisé simultanément n'avait pas non plus permis de
détecter des hémoparasites, avaient un hématocrite
moyen de 30,68% avec respectivement 30,02 et 3 1 ,34% 1. Introduction
comme valeurs minimum et maximum. Leur formule
Dans les zones du Sénégal exposées à la
leucocytaire moyenne se présentait comme suit:
trypanosomiase, la brebis Djallonké, réputée
neutrophiles 33%, éosinophiles 9%, lymphocytes 55%
trypanotolérante, constitue la base démographique et
et monocytes 2 à 3%.
génétique des troupeaux ovins des élevages extensifs
Les variations de ces constantes hématologiques traditionnels. Aussi est-il urgent de développer la
ont été appréciées en fonction de l' infestation ou non par recherche en matière de sémiologie médicale sur ce
des hémoparasites, de l'état corporel, de la saison et du groupe qui, de par son importance démographique et
mode de conduite. Par ailleurs, les particularités zootechnique, sera vraisemblablement la cible
histologiques des leucocytes des Djallonké ont été principale de toutes les actions futures d'amélioration
précisées. sanitaire. La définition des normes de l'hématocrite et
de la formule leucocytaire de la brebis Djallonké du
Sénégal présentée dans cette étude va dans ce sens.
PCVand differential leukocyte
count standards and variations 2. Matériel et méthodes
and histologic leukocyte
characteristics in West African 2.1 Les effectifs
Dwarf sheep in southern Senegal Au Sénégal, les femelles Djallonké constituent
l'essentiel des troupeaux d'élevage extensiftraditionnel,
Abstract dont elles assurent la base génétique et démographique.
Elles ne sont exploitées que de façon exceptionnelle,
This article presents a study of PCV and differential contrairement aux mâles dont l'exploitation est
leukocyte count in adult West African Dwarf sheep in systématique dès le sevrage et principalement au
the Kolda area, in southern Senegal. Data from 355 moment de la fête du mouton. Le renouvellement
samples taken on animals over 12 months old make it génétique des troupeaux se fait par l'achat de femelles,
possible to diagnose properly symptoms based on the puisqu'il n'existe aucun programme de reproduction
results of clinical investigations using PCV values and centré sur des géniteurs sélectionnés. De par leur
differential leukocyte count. fonction de génitrice, les femelles assurent la pérennité

Small Ruminant Research Network Workshop 103

M. Ndiaye, M. В. Sow et T. Alogninouwa

et la dynamique de production des troupeaux. Aussi Note 3: apophyses épineuses ne peuvent plus être
cette étude se limite-t-elle aux brebis adultes, pincées entre le pouce et l'index mais leur
c'est-à-dire âgées de 12 mois et plus. pointe encore individualisable à la palpation;
Des prélèvements de sang ont été effectués sur 159 transverses couvertes par les masses
femelles durant la saison des pluies, et sur 193 femelles musculaires mais espaces entre deux apophyses
durant la saison sèche 1990/91. Bien qu'entretenues transverses encore différenciés sous la pression
selon un mode extensif traditionnel, ces brebis avaient des doigts; arc vertébral et apophyses
la particularité d'appartenir à un réseau de troupeaux mamillaires non palpables.
surveillés dans le cadre d'un programme de recherche.
A ce titre, elles bénéficiaient d'un léger encadrement 2.4 Analyse des données
zootechnique et sanitaire destiné à fidéliser les
propriétaires, et avaient été vaccinées contre la pasteu- Les paramètres statistiques de référence utilisés
rellose et la peste des petits ruminants en août 1990. sont la moyenne arithmétique des valeurs d'hématocrite,
En même temps que les prélèvements, on a procédé les fréquences moyennes relatives des types cellulaires
à l'examen clinique des animaux et à l'attribution d'une et les fréquences absolues d'animaux porteurs d'un type
note de pointage reflétant l'état d'entretien des brebis. d'hémoparasites ou d'une association d' hémoparasites.
Le mode de conduite des animaux au pâturage (pâture Les comparaisons de moyennes de groupe ont été
libre ou avec gardiennage, ou mise au piquet) a effectuées par la méthode dite de la variable réduite Z
également été enregistré. de distribution normale pour les échantillons de grande
taille, et par la variable t de distribution de Student pour
2.2 Réalisation et lecture du les groupes de moins de 30 individus.
microhématocrite et du frottis
Suivant les individus, la prise de sang a été réalisée
tantôt par ponction de sang veineux à la jugulaire, tantôt
3. Résultats et discussion
par ponction des capillaires de l'oreille. Des tubes et
microtubes héparinés ont été utilisés ainsi que des lames 3.1 L'hématocrite
dégraissées à l'alcool.
Selon Murray et coll. (1983), les résultats de Le tableau 1 présente les valeurs moyennes et les
l'hématocrite et de la formule leucocytaire sont en effet écarts types de l'hématocrite pour les différents facteurs
aussi fiables à partir de sang périphérique qu'à partir de de variation considérés indépendamment.
sang veineux. La recherche de parasites sanguins à partir
du sang périphérique est cependant plus sensible de 20% 3.1.1 Hématocrite et saison
que celle effectuée à partir de sang veineux.
Pour établir la formule leucocytaire et L'hématocrite varie de 14 à 36% en saison des
l'hématoscopie parasitaire, les frottis ont été colorés par pluies et de 19 à 47% en saison sèche. La valeur de la
la méthode RAL 555. Cette méthode de coloration variable réduite Z (11,7) révèle que la supériorité de
panoptique fait intervenir trois solutions, à savoir le l'hématocrite de saison sèche est statistiquement
fixateur, l'éosine et le bleu de méthylène. Le frottis est significative (P<0,05).
trempé pendant cinq secondes dans chaque solution et La baisse de l'hématocrite lors de la saison des
la lame mise à égoutter verticalement pendant cinq pluies peut s'expliquer par l'action concomitante de
secondes entre chaque solution. Après trempage dans la divers facteurs:
troisième solution, elle est lavée à l'eau déminéralisée. - le parasitisme gastro-intestinal tel que constaté par
La lecture se fait au microscope optique à faible divers examens coproscopiques;
grossissement (x 100). - le parasitisme externe attesté par la présence dans
le sang de parasites transmis par des insectes
hématophages (glossines, tiques).
2.3 Détermination de la note d'état
En saison sèche par contre, les animaux ont un
La note d'état des femelles Djallonké a été environnement alimentaire et sanitaire tout autre:
déterminée d'après l'échelle de Russel et coll. citée par - la divagation leur permet d'accéder à diverses
Richard (1990). ressources alimentaires (champs post-culturaux et
pâturages naturels, ordures ménagères);
Note 1 : apophyses épineuses et transverses saillantes; la complémentation est souvent pratiquée en fin de
arc vertébral et apophyses mamillaries saison sèche par les éleveurs qui disposent de
aisément palpées du fait de masses musculaires réserves fourragères (fanes d'arachide, tiges de
peu abondantes; mil) (Merlin et coll., 1990);
Note 2: arc vertébral moins net et apophyses transverses - le parasitisme digestif va en s'amenuisant et perd
moins pointues, voire même légèrement de sa gravité clinique; la sécheresse du milieu
arrondies; apophyses épineuses et mamillaires ambiant limite les réinfestations (Vassiliades,
toujours aussi faciles à différencier; 1978);

104 Small Ruminant Research Network Workshop

 Normes et variations de l'hematocrite et de la etc..

- le parasitisme externe évolue de même, du fait de 3.1.3 Hématocrite et présence d'hémoparasites

la grande mobilité des animaux qui ne sont plus
La présence d'hémoparasites dans les frottis a été
attachés ni soumis à la promiscuité des abris de
observée en saison des pluies uniquement. 31 brebis se
saison des pluies.
sont révélées porteuses d'hémoparasites, avec une
prévalence estimée comme suit:
3.1.2 Hématocrite et note d'état
Anaplasma centrale 12,55%
D'après les résultats des tests de différence de
moyennes, il apparaît que: Theileria sp. 1,
- au cours d'une même saison, les hématocrites
d'animaux affectés de notes d'état différentes Trypanosoma brucei et vivax 1.Í
(tableau 1) sont toujours significativement Anaplasma centrale et Theileria sp. 3,14%
différents au niveau de 0,05.
L'hématocrite varie de manière significative
la supériorité de l'hématocrite de saison sèche sur (P<0,05) avec la présence (animal porteur) ou non
celui de saison des pluies se confirme pour les (animal non porteur) d'hémoparasites dans le frottis
animaux présentant des notes d'état identiques, (tableau 1). Signalons que Toure (1982) a constaté une
sauf en ce qui concerne la note 3 pour laquelle les baisse significative de l'hématocrite chez des moutons
hématocrites ne sont plus statistiquement Djallonké expérimentalement infestés par Trypanosoma
différents au niveau de 0,05 (tableau 2). congolense.

Tableau 1. Valeurs moyennes de l'hématocrite des brebis Djallonké selon la saison, la note d'état corporel et le statut
hémoparasitaire
Moyenne Ecart
Saison Saison des pluies 159 24,71 4,86
Saison sèche 193 30,68 4,61
Interaction saison x note
Saison sèche Note 1 79 21,59 3,32
Note 2 54 27,35 3,57
Note3 9 33,00 2,00

Saison des pluies Note 1 40 26,82 3,73

Note 2 91 30,42 3,66
Note3 57 33,91 3,88
Présence d'hémoparasites
Animaux porteurs 31 22,51 5,15
Animaux non porteurs 125 25,08 4,55
Mode de conduite
Piquet 40 23,45 4,84
Gardiennage 31 28,06 4,54

L'aggravation de la baisse de l'hématocrite, déjà

nette et statistiquement significative en saison des
Tableau 2. Valeurs de Zoudet pour une comparaison inter pluies, que provoque la simple présence d'hémo
saison des moyennes d'hématocrite par notes parasites laisse perplexe quant à la prétendue rusticité
d'état
du Djallonké.
Saison des pluies

Saison sèche Note 1 Note 2 Note3 3.1.4. Hématocrite et conduite du troupeau

Note 1 7,47 S 0,69 NS 4,7 S Le tableau 1 donne la valeur moyenne de

l'hématocrite mesuré sur deux troupeaux, dont l'un était
Note 2 16.66S 4,95 S 2,06 S maintenu au piquet durant toute la saison des pluies et
l'autre géré en pâture libre. Le second mode de conduite
Note3 19,55 S 13.66S 0,68 NS
est associé à une plus forte valeur de l'hématocrite
S = significatif; NS = non significatif (au niveau de 0,05). (P<0,05).

Small Ruminant Research Network Workshop 105

M. Ndiaye, M. В. Sow et T. Alogninouwa

3.2 Formule leucocytaire ont été observées dans certains cas. Ce phénomène
pourrait être lié au colorant utilisé.
3.2.1 Formule leucocytaire et saison Les monocytes se présentent comme des cellules
de grande taille et de morphologie variable. Ce sont en
Le tableau 3 donne les taux moyens des différents général des cellules rondes avec un cytoplasme
types de leucocytes pour la saison sèche et la saison légèrement basophile. Le noyau n'a pas de structure bien
humide. La formule leucocytaire ne diffère pas définie; il peut être en cloche, en fer à cheval ou bilobé.
significativement d'une saison à l'autre. Le rapport nucléocytoplasmique est élevé. La
distinction entre monocytes et lymphocytes est souvent
3.2.2 Formule leucocytaire et statut difficile à établir.
hémoparasitaire Un type cellulaire à morphologie tenant à la fois
du monocyte et du lymphocyte a souvent été observé. Il
Contrairement à l'hématocrite, la formule existe en outre un type cellulaire binucléé à cytoplasme
leucocytaire n'est pas modifiée par la présence plus neutrophile que basophile.
d' hémoparasites (tableau 4). Les examens n'ont pas révélé de polynucléaires
Il ressort de ces données que l' infestation basophiles.
hémoparasitaire réduit d'abord l'hématocrite; lorsque la
parasitémie s'aggrave, on assiste à l'apparition des
signes cliniques de l' infestation.

3.3 Caractéristiques morphologiques 4. Conclusion

et histologiques des leucocytes
L'hématocrite des brebis Djallonké subit des
Les neutrophiles matures présentent un noyau variations significatives en fonction de la saison, de
lobé, mais la morphologie des lobes varie d'une cellule l'état du sujet face aux hémoparasitoses et du mode de
à l'autre. Dans la majorité des cas, il s'agit de plaques conduite, étant entendu par ailleurs que le mode de
concentriques réunies entre elles par un filament de conduite et la saison ont un effet déterminant sur le
chromatine. Il arrive cependant que le noyau ne soit pas régime alimentaire des animaux et sur leur susceptibilité
segmenté, auquel cas il se présente en une bande de aux infestations par les helminthes et les hémoparasites.
chromatine dense. Des valeurs références de l'hématocrite en
Les lymphocytes (grands et petits) se différencient fonction de la saison et du mode de conduite peuvent
facilement, contrairement à ce que rapporte la être dégagées de cette étude. La note d'état permet une
littérature. Le rapport nucléocytoplasmique varie d'une bonne prédiction de l'hématocrite. La formule
cellule à l'autre. Le noyau présente une morphologie leucocytaire ne subit pas de variations saisonnières et
variable (ovale, en cloche ou en raquette de tennis). Le n'est pas modifiée par la simple présence d'hémo-
cytoplasme est basophile, avec parfois des granules parasites.
éosinophiliques de taille et de nombre variables. Sur le plan histologique, la signification sémio-
Les éosinophiles sont conformes à la description logique de la plus grande fréquence de granulations
classique qui en est faite. Cependant, des éosinophiles éosinophiliques en saison des pluies et de la rareté des
dont les granules sont plus basophiles qu'éosinophiles polynucléaires basophiles reste à élucider.

Tableau 3. Formule leucocytaire par saison

Saison des pluies (n= 159) Saison sèche (n = 193)
Neutro Eosino Lympho Mono Neutro Eosino Lympho Mono
Moyenne 32,84 9,16 55,61 2,37 33,26 5,95 57,31 3,52
Ecart type 6,77 6,22 8,45 1,31 5,80 3,79 6,26 2,02

Tableau 4. Formules leucocytaires et statut hémoparasitaire

Animaux porteurs (n = 3 1 ) Animaux non porteurs (n = 125)
Taux (%) Ecart type Taux (%) Ecart type
Neutrophiles 32,61 6,40 32,8 6,85
Eosinophiles 7,19 3,64 9,73 6,63
Lymphocytes 57,61 7,29 55,13 8,73
Monocytes 2,5 1,23 2,33 1,34

106 Small Ruminant Research Network Workshop

 Normes et variations de l'hématocrite et de la etc..

Références Richard D. 1990. Les notes d'état des moutons Waralé et

Peul-peul. Document multicopié. ISRA (Institut
sénégalais de recherches agricolesVLNERV (Labo
Merlin P., Ndiaye M. et Faugère O. 1990. Facteurs de risque ratoire national de l'élevage et de recherches
des pneumopathies des petits ruminants au Sénégal. I. vétérinaires), Dakar-Hann (Sénégal). 4 p.
Zone soudanienne, saison sèche 1988-1989. Document
de travail du programme PPR (Pathologie et productivité Toure S. M. 1982. Caractérisation de la trypanotolérance et
des petits ruminants) n' 19. LNERV (Laboratoire comparaison des races bovines et ovines. Référence n°
national de l'élevage et de recherches vétérinaires)/ISRA 61 /Parasitologic LNERV (Laboratoire national de
(Institut sénégalais de recherches agricoles), l'élevage et de recherches vétérinairesyiSRA (Institut
Dakar-Hann (Sénégal). 21 p. + 14 tableaux. sénégalais de recherches agricoles), Dakar-Hann
(Sénégal). 12 p.
Murray M., Trail J.C.M., Turner D.A. et Wissocq Y.J. 1983.
Productivité animale et trypanotoléranee. Manuel de Vassiliades G. 1978. Notes sur le parasitisme gastro-intestinal
formation pour les activités du réseau. CIPEA (Centre des moutons du Sénégal. In: 46ème session de VOIE
international pour l'élevage en Afrique), Addis-Abeba (Office international des épizooties), OIE, Paris
(Ethiopie). 221 p. (France).

Small Ruminant Research Network Workshop 107

 Haemagglutination-inhibition technique for definitive
diagnosis of peste des petits ruminants
virus specific antibody
L.O. Wosu and M. CO. Ezeibe

Department of Veterinary Medicine, University ofNigeria, Nsukka, Nigeria

Abstract Even the serum neutralisation test (SNT) currently

being used as the confirmatory diagnosis for PPR and
A simple and rapid serological method for definitive rinderpest also shows cross-reactivity (Mornet et al,
identification for Kata virus or peste des petits ruminants 1956; Taylor, 1979; Obi, 1984). Wosu (1985) first
(PPR) virus specific antibody is described. The demonstrated the haemagglutinin or PPR homogenate
technique is based on adsorbing out the cross-reacting antigen, to porcine erythrocytes.
antibodies to rinderpest antigen from a PPR serum and Wosu ( 1 977) employed an adsorption technique to
leaving the specific antibody to PPR which is make a definitive differentiation between very closely
determined by haemagglutination-inhibition test. related and cross-reacting strains of the protozoan
Trichomonasfoetus var. brisbane, var. belfast, and var.
manley. It was therefore thought desirable to try the
Utilisation de la réaction adsorption technique in this virological system to see if
it could eliminate the cross-reactivity between PPR virus
d'inhibition de (Kata virus) and rinderpest virus (RV) and make a
^hémagglutination pour la definitive diagnosis of the PPR specific antibody in a
single test. This is of special interest in West Africa
détection de l'anticorps contre le where both rinderpest and PPR are endemic.
virus de la peste des petits This report details the technique used in
eliminating the cross-reactivity between PPR virus and
ruminants rinderpest virus serologically.

Résumé
Cette étude décrit une méthode sérologique simple Materials and methods
et rapide utilisée pour l'identification définitive de A known lymph node homogenae PPR antigen
l'anticorps contre le virus Kata ou contre le virus de la (Tambo), a known rinderpest cultured antigen and a
peste des petits ruminants. La technique est basée sur known PPR immune serum were obtained from the
I 'adsorption des anticorps réagissant à l 'antigène de la National Veterinary Research Institute, Vom, Nigeria.
peste bovine dans un sérum de peste des petits Porcine red blood cells (RBC) collected in Alsevers
ruminants. L 'anticorps spécifique de la peste des petits anticoagulant were used. The porcine red blood cell
ruminants qui reste dans ce sérum est ensuite déterminé suspension (0.6%) was prepared as described by Wosu
au moyen de la réaction d'inhibition de (1984).
I 'hémagglutination. The diluent was phosphate buffered saline (PBS)
at pH 6.8. The adsorption technique described by
Johnson (1967) and Wosu (1977) was used as the basis
for the removal of antibodies cross-reacting with
Introduction rinderpest virus from known PPR serum.
The causative viral agents of both peste des petits A known PPR immune serum was divided into two
ruminants (PPR) disease and rinderpest disease belong portions, A and B. Complement was removed from A
to the same Morbillivirus subgroup of the genus and В by heating at 56°C for 30 min. In addition, natural
Paramyxovirus and show a great deal ofcross-reactivity. agglutinins and non-specific serum inhibitors were
They cross-react in the immunodiffusion and removed from portion В by adsorption with porcine
complement fixation tests (Johnson and Ritchie, 1968; erythrocytes and kaolin respectively, as described by
Ihemelandu et al, 1976; Nawathe, 1983), immuno- Joo (1976). The treatment of the PPR immune sera A
osmo-precipitation test (Majiyagbe et al, 1984) and and В involved adding 0.5 ml of known rinderpest
indirect haemagglutination-inhibition test using measles antigen (Tambo) to 0.5 ml of each of the sera and 0.5 ml
virus and monkey red blood cells (Nawathe, 1983). of each of the known PPR antigens to another 0.5 ml of

Small Ruminant Research Network Workshop 109

L.O. Wosu and M.C.O. Ezeibe

each of the sera. Another 0.5 ml of each sera was The treatment of serum A alone is quite adequate
untreated and used as a control. for a rapid qualitative survey or epidemiological work,
All the portions were incubated at 37°C for two but for quantitative work treatment of serum В would be
hours and centrifuged for 10 minutes at 3000 rpm. The necessary. Since the result was obtained, the technique
supernatants were used for the haemagglu- has been used on many PPR and rinderpest serum
tination-inhibition (HI) test as described by Wosu samples with the same outcome.
(1985), involving V-bottomed microtitre plates and Definitive tests for antibodies due to either antigen
microtitre pipettes. Four haemagglutinating units have generally been based on carrying out at least two
(HAU) of the PPR homogenate antigen were used. tests with the two antigens by SNT. It was evident that
The procedure for the HI test involved adding 0.05 the antibody was due to the virus which gave the higher
ml of PBS (pH 6.8) to each well of the V-bottomed titre due to cross-reactivity between the two antigens by
microtitre plate. Serial double dilutions of 0.05 ml of SNT (Mornet et al, 1956; Taylor, 1979; Obi, 1984).
each of the test sera were made on the same plate. Next However this procedure was rather laborious. More
0.05 ml of the PPR homogenate antigen containing four importantly, when the results of the two tests were close,
HAU was added to each well. This was incubated at it was very difficult to ascertain the correct antigen.
room temperature (25°C) for one hour. Then 0.05 ml of Moreover, it was frequently necessary to know whether
0.6 % porcine RBC suspension (the same porcine RBC a known serum would neutralise and consequently
sample used in determining the four HAU) was added identify an unknown antigen or vice-versa.
to each well. Controls for the four HAU of the PPR virus The technique described in this paper appears to
and the porcine erythrocytes were included in the have elucidated the problem of cross-reactivity which
protocol. The plates were incubated at 4°C for six hours. exists in the diagnosis of PPR and rinderpest. Indeed,
Five repeats of the HI test were carried out immediately this HI technique was used in a virological system. An
and the treated sera A and В stored at -20°C for immune serum was made more specific by adsorption
preservation. After a month, HI tests were carried out on with a heterologous virus of the same group. Adsorbed
the preserved serum portions. serum tested for haemagglutination-inhibition (HI)
activity reacted only with homologous and not with
Results and discussion heterologous strains of arboviruses, thereby facilitating
the identification of strains (Jawetze et al, 1978). It is
The HI titre of the PPR immune serum adsorbed with
suggested that this treatment of serum samples be
rinderpest antigen was less than the non-adsorbed
applied to the other tests in order to eliminate
portion and the titre of the serum adsorbed with PPR
antigen was zero. Thus adsorption with rinderpest cross-reactivity between PPR and RV. A comparative
antigen neutralised the common antibodies leaving only study could be made to determine the optimal test for
antibodies specific to PPR antigen. This was confirmed identifying the specific antibody to PPR antigen.
by the fact that the HI titre of a portion of the same serum The disappearance of differences in HI titres after
adsorbed with PPR was zero. The PPR antigen therefore the tested sera were stored at -20°C for one month was
neutralised both the antibodies it shared with rinderpest a chance finding (Table 2). While the cause of the
and those which were specific to the PPR virus. differences was not investigated, it may be attributed to
Differences between treatments were maintained viral elution from antibodies (Nathalie et al, 1973). An
even after the removal of serum natural haemagglutinins investigation is needed to determine how long the
and non-specific serum inhibitors as shown in Table 1. treated sera can be stored at -20°C without elution.
This eliminated the possibility that the difference in titre
could have arisen from serum non-specific Table 2. HI titres of PPR serum portions as in Table I after
haemagglutinins or non-specific inhibitors. storage at -20 °C for one month.

Table 1. HI titres of PPR immune serum. Titre

Serum treatment Serum A Serum В Remarks
Titre
Control
Serum treatment Serum A Serum В Remarks for four
HAU and
Adsorption with RBC
Adsorption with 1:1024 1:512 Control rinderpest antigen 1:40% 1:2048 were good
rinderpest antigen for four
HAU and Adsorption with PPR —
RBC antigen 1:4096 1:2048
were good -
No adsorption 1:4096 1:2048
Adsorption with PPR Zero Zero -
antigen The HI test, especially in less sophisticated
laboratories, is a very useful, quick and accurate method
No adsorption 1:4096 1:2048 for determining the antibody levels in a given immune
Repeats of the HI test, carried out on the same treated serum samples serum (Johnson, 1971). Where there are closely related
immediately after the adsorption, gave the same titres as above. antigens which cross-react, the test cannot readily be a

110 Small Ruminant Research Network Workshop

 Haemagglutination-inhibition technique for etc...

definitive diagnosis of a specific antibody unless it is Joo H S. 1976. Studies on the porcine parvoviruses. PhD
possible to eliminate the cross-reactivity due to the other thesis, University of North Queensland, Townsville,
related antigens. Australia.
It is possible to make a definitive serological Majiyagbe K A, Nawathe D R and Abegunde A. 1984. Rapid
diagnosis of PPR specific antibodies by haema diagnosis of peste des petits ruminants (PPR) infection,
gglutination-inhibition test using the adsorption application of immunoelectroosmophoresis (IEOP)
technique. Revue d'élevage et de médecine vétérinaire
technique.
des pays tropicaux 37: 1 1-15.
Mornet P, Orue J, Gilbert Y, Thiery G and Sow Mamadou.
1956. La peste des petits ruminants en Afrique
Acknowledgements occidentale française ses rapports avec la peste bovine.
Revue d'élevage et de médecine vétérinaire des pays
We appreciate the co-operation we received from the tropicaux 9:3 1 3-342.
National Veterinary Research Laboratory (NVRI),
Nathalie J, Schnudt T, Edwin H and Lenette R. 1973.
Vom, Nigeria. We are grateful to the University of Advances in the serodiagnosis of viral infections.
Nigeria Senate Research Grant Committee for funding Progress in Medical Virology 15:244-308.
the work. We also thank the secretarial and other staff
of the Department of Veterinary Medicine for their help. Nawathe D R. 1983. Some observations in the diagnosis of
peste des petits ruminants in Nigeria. In: Hill D H (ed),
Peste de petits ruminants (PPR) in sheep and goats.
Proceedings of the international workshop held at IITA
References (International Institute of Tropical Medicine), Ibadan,
Nigeria, 24-26 September 1980. ILCA (International
Ihemelandu E C, Hamdy F M, Dardiri A M, Nduaka O and Livestock Centre for Africa), Addis Ababa, Ethiopia. pp.
Breese S S Jr. 1976. Etiology of the stomatitis 34-39.
pneumoenteritis complex in Nigerian Dwarf goats. Obi J. 1984. Personal Communication.
Canadian Journal of Medicine 40:276-284.
Taylor W P. 1979. Serological studies with the virus of peste
Jawetze, Melnick J I and Adelberg E A. 1978. Review of des petits ruminants in Nigeria. Research in Veterinary
medical microbiology. Thirteenth edition. Lange Science 26:236-242.
Medical Publications, pp. 361-362.
Wosu L O. 1977. Trichomonas infection in a bull — An
Johnson A E. 1967. Comparison of six strains of Trichomonas apparent change in serotype of infecting organisms.
foetus. II. Gel diffusion precipitation studies of soluble Australian Veterinary Journal 53:340-341.
antigens. Experimental parasitology 21:336-343.
Wosu L O. 1984. Standardisation of erythrocyte
Johnson RH. 1971. Serological procedures for the study of concentrations for haemagglutination tests and removal
feline panleukopenia. Journal of the American of natural agglutinins. Nigerian Veterinary Journal
Veterinary Medical Association 158:876-884. 13:39^2.
Johnson R H and Ritchie J S D. 1968. A virus associated with Wosu L 0. 1985. Agglutination of red blood cells by peste des
pseudorinderpest in Nigerian Dwarf goats. Bulletin of petits ruminants (PPR) virus. Nigerian Veterinary
Epizootic Diseases ofAfrica 16:41 1-417. Journal 14:56-58.

Small Ruminant Research Network Workshop 111

 Peste des petits ruminants (PPR) and rinderpest (RP)
antibodies in clinically normal small ruminants
in the Cameroon and Nigeria
К.A. Majiyagbe, D. Shamaki, D. Kulu and T.K.C. Udeani

National Veterinary Research Institute

Vom near JOS, Plateau State, Nigeria

Abstract afin d'évaluer la prévalence des anticorps contre la

peste des petits ruminants et contre la peste bovine chez
An investigation was conducted in the small ruminant des ovins et des caprins bien portant en apparence et
producing areas of the northern part of Nigeria and the non vaccinés. Cette étude avait pour objectif d'aider à
Adamawa Province of the Cameroon (Garoua and its mieux comprendre I 'épidémiologic de la peste des petits
suburb) to assess the prevalence of peste des petits ruminants par rapport à une infection cliniquement
ruminants (PPR) and rinderpest (RP) antibodies in semblable, la peste bovine.
apparently normal and non-vaccinated sheep and goats.
La technique immuno-enzymatique basée sur les
This study was aimed at providing a better under
anticorps monoclonaux du virus a été utilisée pour
standing of the epidemiology of PPR in relation to a
analyser 702 échantillons de sérum prélevés dans la
clinically similar disease, RP.
zone d'étude. Les résultats révèlent que, dans
The virus-specific monoclonal antibody (Mab) l 'ensemble, 46, 7% des sujets abritaient des anticorps
based competitive ELISA (enzyme-linked contre la peste des petits ruminants et 18,2% des
immunosorbent assay) technique was used to analyse anticorps contre la peste bovine. Ces chiffres étaient
702 serum samples from the surveyed area for PPR and respectivement de 35% et de 3,1% au Cameroun (n =
RP. 320), et de 56,5% et 30,9% au Nigéria (n = 382).
The results show that overall, 46.7% (328/702) had
PPR antibodies while 18.2% had RP antibodies. For En comparant entre eux les spectres des profils des
Cameroon, (N = 320), 35% and 3.1% had PPR and RP anticorps, on a noté que certains animaux réagissaient
antibodies respectively, while for Nigeria (N= 382), the uniquement au virus de la peste des petits ruminants ou
corresponding values were 56.5 and 30.9%. à celui de la peste bovine, alors que d 'autrespossédaient
Comparison of the spectrum of antibody profiles des anticorps contre chacun de ces deux virus. Cette
showed that some animals reacted solely to either PPR observation a des conséquences importantes sur le
or RP virus infection, while others simultaneously programme de la Campagne panafricaine contre la
possessed antibodies for both viruses. This observation peste bovine qui s'emploie à faire disparaître cette
has serious implications for the on-going Pan-African maladie du continent africain.
Rinderpest Campaign (PARC) program to eradicate
rinderpest from the African continent.
Introduction
Présence d'anticorps contre la Livestock are very important for both the subsistence
and economic development of the African continent.
peste des petits ruminants et la They provide a flow of essential food products
peste bovine chez des petits throughout the year. In some countries they are a major
source of government revenue and export earnings.
ruminants du Cameroun et du They also sustain the employment and income of
Nigéria ne présentant pas de millions of people in rural areas, contribute draught
energy and manure for crop production and arc the only
signes cliniques de maladie food and cash security available to many Africans
(Brumby, 1990). In Nigeria, as in other African
Résumé countries, small ruminants (sheep and goats) contribute
a substantial proportion of the nation's meat supply.
Une enquête a été réalisée dans les zones Jahnke (1982) stated that 21.5% of the nearly 104
d'élevage de petits ruminants de la région million sheep and 25.4% of the 125 million goats in
septentrionale du Nigéria et de la province de tropical Africa are located in the humid and subhumid
l'Adamaoua au Cameroun (Garoua et sa périphérie) zone. Some 13.6% of the sheep and 16.2% of the goats

Small Ruminant Research Network Workshop 113

К. A. Majiyagbe, D. Shamaki, D. Kulu and T.K.C. Udeani

are located in the humid zone, while 7.9% of the sheep the Artiodactyla group, are also susceptible to rinderpest
and 9.2% of the goats are found in the subhumid zone. virus and are hence occasionally infected with bovine
There are an estimated 22 million sheep and 34.5 million rinderpest. However, the PPR virus causes a non-clinical
goats in Nigeria (FDLPCS, 1991). The indigenous disease in bovines; infected cattle sero-convert and
breeds of sheep in order of importance are Yankassa become immune to RP and PPR viruses. This will
(60%), West African Dwarf (WAD, 20%), Uda (10%) interfere with successful TCRV vaccination and has
and Balami (10%). For the goat breeds, the order is Red serious implications when cattle are sero-surveyed to
Sokoto goat (50%), West Africa Dwarf (45%) and Sahel determine the level of protection against rinderpest.
(5%) (Osinowo, 1992). To provide a better understanding of the
In sheep and goats, PPR is considered the most epidemiology of PPR and RP in small ruminants, an
important single cause of morbidity and mortality. It is investigation was conducted in small ruminant
a highly contagious rinderpest-like viral disease that producing areas of the northern part of Nigeria and the
particularly affects goats, but also affects sheep. The Adamawa Province of Cameroon (Garoua and its
disease is characterised by fever, oculonasal discharge, suburbs). Its purpose was to assess the prevalence of
necrotic stomatitis, bronchopneumonia and PPR and RP antibodies in apparently normal and
gastro-enteritis. The disease is caused by the PPR virus, non-vaccinated sheep and goats.
a member of the Morbillivirus genus which includes
rinderpest, canine distemper and measles. PPR is
endemic in the humid zone and to a lesser extent in the Materials and methods
subhumid zone of the West African subregion. Its
economic importance in these areas is enormous. It can
quickly decimate whole flocks either by itself or in Sera samples
combination with other diseases, such as pneumonia or
gastro-intestinal parasitism (Majiyagbe, 1985). In A total of 702 sheep and goat sera samples were
susceptible flocks, morbidity may be 100% and collected from 352 survey regions in Nigeria and 320
survey regions in Garoua and its environs in the
mortality greater than 90%, especially amongst animals
Adamawa Province of Cameroon. Serum was
under six months of age. This has caused severe aseptically separated from the blood clot and kept
economic hardship to many small ruminant producers. chilled while being transported to the laboratory for
Clinically, PPR occurs as an acute disease in goats, analysis. Details of samples collected and species
while in sheep it is generally benign. Surviving animals distribution are shown in Table 1.
usually develop a dual immunity to PPR and RP viruses.
In practice, tissue culture rinderpest virus (TCRV)
vaccine is used in immunoprophylaxis against PPR in Serum testing for PPR and RP antibodies
the absence of a homologous vaccine. Control of PPR is A competitive ELISA (C-Elisa) for the detection of
therefore paramount to efficient small ruminant antibodies to RP and PPR viruses was used for serum
production in the West African subregion. analysis. This is a monoclonal antibody (Mab) based
Currently, the PARC (Pan African Rinderpest test. Two types of Mab are used, one specific against the
Campaign) to control and subsequently eradicate hemagglutinin (HA) protein of RP virus and the other
rinderpest from the African continent is being carried specific against the H-protein of the PPR virus. Unlike
out in over 22 countries. The target animals for the the indirect ELISA and the virus neutralisation test
campaign are cattle. Small ruminants, being animals of (VNT), the RP C-Elisa detects only antibodies to RP

Table 1. PPR and RP antibodies in clinically normal small ruminants from Nigeria and Cameroon (%).

PPR+pos. PPR+ RPV+ PPR++ PPR-

only(l) PPRneg RPV+(2) RPVneg RPV-<3) PPR- RPV++ RP-

Global summary
(n=702) 46.7 53.3 18.2 81.8 34.9 6.8 11.4 46.44
(Cameroon + Nigeria)
Cameroon (n=320) 35 65 3.1 96.9 32.8 1.6 1.6 63.4
Nigeria (n=382) 56.5 43.5 30.9 69.1 36.6 11.3 19.6 32.2
Nigerian control TCRV 62 38 60 39.1 22.4 21.9 39.1 16.2
vaccinates (n= 192)
Sheep (n = 303) 50.5 49.5 1.3 98.7 50.2 0.9 0.3 45.2
Goats (n =399) 43.9 56.1 31.1 68.9 23.8 11.3 19.8 38.6
Cameroon - samples from Boklé (IRZ, LANAVET), Lagdo, Ngong and Garoua Abbatoir.
Nigeria - samples from Bauchi State (Soro and Alkaleri) and Plateau State (Vom, Jos and Eviron).
Vaccinated control samples from Ife and Shika Research Farms.

114 Small Ruminant Research Network Workshop

 Peste des Petis Ruminants (PPR) and Rinderpest (RP) etc..

virus and gives no cross-reactivity with antibodies to showed that very few goats and sheep exhibited
PPR virus. evidence of infection (Majiyagbe, 1985). This was
The Mab used in the PPR C-Elisa also gives a high probably due to prior PPR exposure or infection with a
level of specificity to the PPR virus. However, some mild RP virus strain. A small ruminant adapted strain of
anti-sera to rinderpest do compete against this PPR RP virus is know in India (Dardiri et al, 1976; Rao et al,
specific Mab. The test was performed as outlined in the 1984) and there is evidence that this strain circulates
C-Elisa kit manual (Majiyagbe et al, 1991; Anderson et
within cattle, sheep and goats. Experimental studies
al, 1991). The ELISA test reagent kits were supplied in
(Anderson et al, 1991) have shown that cattle may be
two parts, Buffers + Conjugate, from the joint
FAO/IAEA (Food and Agriculture Organization of the infected by contact with the PPR virus, but will not
United Nations/International Atomic Energy Agency) exhibit any symptoms and will resist RP. Similarly
Agricultural Laboratory in Seibersdorf, Austria and RP Anderson et al (1991) have shown that goats with
and PPR antigens kits from the Pirbright Laboratory, circulating antibodies to PPR virus will not respond to
Pirbright, Surrey, UK. rinderpest vaccination.
In the absence of vaccination, the presence of PPR
and RP virus antibodies indicates that PPR and RP are
Results circulating in the population, even if only at low levels.
Table 1 shows that just under half of the samples had This observation has serious implications for the
PPR antibodies, while one fifth had RP antibodies. The on-going PARC seromonitoring program, particularly
rest were negative to both viruses. in the assessment of immunity levels to RP after
Because of the discriminating nature of the vaccination. In a situation where the PPR virus infects
monoclonal antibodies used in the test, it was possible cattle before RP vaccination, the animal would not
to distinguish between specific PPR and RP virus respond to the vaccination and would be considered an
induced antibodies in these animals' serum samples. animal at risk and test negative in the RP C-Elisa test
Comparison of the spectrum of antibody profiles used in evaluating sero-conversion. Our own results
showed that some animals reacted solely to either PPR show that a similar situation occurs in small ruminants
or RP virus infection. and this could influence the results of seromonitoring in
the current PARC program.
The goal of the current PARC program is to
Discussion and conclusion eradicate RP from the African continent. For a country
to be able to declare itself free from disease, a
Serum samples collected from small ruminant
sufficiently high level of immunity (>95%) must be
producing areas in Nigeria and the Cameroon showed
that some of the animals possessed antibodies to PPR achieved in the national herd and vaccinations against
and RP viruses. None of the animals were less than one RP must cease.
and a half years of age and they had not been vaccinated The contribution of PPR virus immunity to
with either TCRV or homologous PPR vaccine. The animals' resistance to RP must be properly quantified.
source of these antibodies can therefore only be from an Appropriate tests such as the C-Elisa must be used for a
unapparent infection or from immunity gained through meaningful evaluation of survey results.
recovery from a field outbreak. Animals that recover
from field outbreaks posses a durable immunity, as an
infected animal either dies or survives. Antibodies can
therefore be found in the blood of surviving animals.
The West African subregion is considered an Acknowledgement
endemic zone for PPR. The disease is considered the
single most important cause of morbidity and mortality
in small ruminants, especially goats. The most effective We would like to thank Dr Zyam and Dr Abdoulaye
form of protection against PPR is through vaccination. Nakano, the Director, LANAVET (Laboratoire
Since the causal agent of PPR is a morbillivirus related National Vétérinaire), Boukle, Garoua for their support
and provision of laboratory facilities during sample
to rinderpest virus, the use of TCRV vaccine has been
collection. The technical assistance of Mr S Igbabab and
shown to be effective in offering protection against PPR;
Mr O Ochai is appreciated.
a single vaccination confers immunity for at least 12 The permission of the Director, National
months (Majiyagbe et al, 1991; Rao et al, 1984).
Vaccinated animals are therefore able to withstand the Veterinary Research Institute, Vom, to publish this
virulent PPR virus in the field. report is also acknowledged.
Field experience, however, shows that these The work was supported by the FAO/IAEA/SIDA
animals are rarely vaccinated. Their survival therefore (Food and Agriculture Organization of the United
depends on being able to overcome a field outbreak of Nations/International Atomic Energy Agency/Swedish
PPR or infection with the closely related RP virus. International Development Agency) Research contract
Observations from the last RP outbreak in Nigeria No. 6559/SD.

Small Ruminant Research Network Workshop 115

K.A. Majiyagbe. D. Shamaki, D. Kulu and T.K.C. Udeanl

Jahnke H E. 1982. Livestock production systems and livestock

References development in tropical Africa. Kieler
Anderson J, McKay J A and Butcher R N. 1991. The use of Wissenschaftverlag Vauk, Kiel, Germany. 253 pp.
monoclonal antibodies in competitive Elisa for detection Majiyagbe К A. 1985. Health management in small ruminants
of antibodies to rinderpest and peste des petits ruminants with emphasis on peste des petits ruminants (PPR). In:
viruses. In: Seromonitoring of rinderpest throughout Adu I F, Osinowo O A, Tai wo В В A and Alhassan W
Africa, Phase one. Proceedings of final Research S (eds). Small ruminant production in Nigeria.
Coordinating Meeting of the FAO/ Proceedings of National Conference on Small Ruminant
IAEA/SIDA/OAU/IBAR/PARC coordinated Research Production held in Zaria, Nigeria, 6-10 October 1985.
Programme, Bingerville, Côte d'Ivoire, 19-23 NAPRI (National Animal Production Research
November 1990. IAEA-TECDOC - 623. pp. 43-53. Institute), Shika, Zaria, Nigeria, pp. 220-230.
Brumby P. 1990. Systems and livestock development in Majiyagbe К A, Ayeni A, Idowu So and Shamaki. 1991.
tropical Africa. In: Dauda К Jawara (ed), Animal disease Effect of age on immune response of West African
as a factor limiting economic development in Africa. Dwarf (WAD) sheep to tissue culture rinderpest virus
Cornell Vet. 80:17-25. (TCRV) vaccine. (Submitted to Nigerian Journal of
Dardiri A H, de Boer C J and Hamdy F M. 1976. Response of Immunology).
American goats and cattle to PPR virus. Proceedings of Osinowo О A. 1992. Problems and prospects for development
the 1<? Annual Meeting, The American Association of of small ruminants in Nigeria. In: The Nigerian livestock
Veterinary Laboratory Diagnosticians, November 7-9 industry: Problems nd prospects. Proceedings of a
1976. P. 337. workshop held at Abuja, Nigeria. 26-27 February 1992,
FDLPCS (Federal Department of Livestock and Pest
FDLPCS (Federal Department of Livestock and Pest Control
Control Service), Abuja, Nigeria.
Services). 1991. Nigerian livestock survey. Federal
Department of Livestock and Pest Control Services, Rao M, Indra Devi T, Ramachandran S and Scott G R. 1984.
Ministry of Agriculture, Abuja, Nigeria. Volume 2. 289 Rinderpest in sheep in Andhra Pradesh and its control by
pp. vaccination. Indian Veterinary Journal 51:439-450.

116 Small Ruminant Research Network Workshop

 Diseases of small ruminants in central Tanzania
LE.G. Mboera1 and J.I. Kitalyi3

Livestock Training Institute

P.O. Box 51, Mpwapwa, Tanzania
Veterinary investigation Centre
P.O. Box 159, Mpwapwa, Tanzania

Abstract morbidity. Several reports have been written on

livestock diseases in the country, but with very little
Information on diseases affecting small ruminants in attention to small ruminant diseases. Nyange (1984)
Tanzania has not been extensively documented. More mentioned the incidences of diseases such as caprine
than 14 years' worth of data on disease cases in sheep pleuropneumonia and sheep pox in northern Tanzania.
and goats was analysed from records kept at the while Mbise et al (1984) reported incidences of anthrax
Veterinary Investigation Centre (VIC) and Livestock and brucellosis. Matovelo et al (1987) reported that
Training Institute (LTT), Mpwapwa. Results showed that parasitic gastroenteritis, pneumonia and diarrhoea were
the majority of the disease cases were parasitic the dominant problems in small ruminants in the eastern
infections (53.1%), followed by bacterial infections part of the country.
(23.6%), other infectious diseases (12.6%) and Control of animal diseases is essential for realising
non-infectious conditions (10.7%). the potential of small ruminant production. For proper
This study shows that the diseases affecting sheep control to be instituted, however, diseases and their
and goats in Tanzania, particularly the central zone, need dynamics must be known. This paper reports disease
to be diagnosed and documented in order to implement findings for cases brought to the Veterinary
more effective disease control measures. Investigation Centre (VIC) and Livestock Training
Institute (LTI) laboratories at Mpwapwa for routine
diagnosis between 1978 and 1991.
Pathologie des petits ruminants
en Tanzanie centrale
Résumé Materials and methods
Il existe peu d'études sur la pathologie des petits The area covered in the study is the central zone of
ruminants en Tanzanie. Des données sur les cas de Tanzania. The zone lies roughly between latitudes 4°
maladies relevés pendant plus de 14 ans sur des ovins and 8°S and longitudes 32* and 39°E. It is predominantly
et des caprins au Centre de recherche vétérinaire et à a livestock area occupied by sedentary pastoralists.
l'Institut deformation en élevage de Mpwapwa ont été There are approximately half a million sheep and 1.2
analysées. Les maladies les plus répandues étaient les million goats in this zone, representing about 16 and
parasitoses (53,1%), suivies des maladies bactériennes 18%, respectively, of the total small ruminant
(23,6%) et d'autres maladies infectieuses (12,6%) et population in the country. The sheep and goats are
non infectieuses (10, 7%). predominantly local breeds which are usually kept
extensively.
Cette étude fait ressortir la nécessité de
The data was collected from records of cases
diagnostiquer et de répertorier les maladies qui
iiagnosed at the VIC and LTI laboratories. Sheep and
frappent les petits ruminants en Tanzanie,
particulièrement dans la zone centrale, en vue de la mise goats have been grouped together. The data reported are
either confirmed diagnoses or conclusions. The
en place de mesures efficaces de prophylaxie.
methods used in diagnosis were clinical examination,
post-mortem findings and laboratory tests such as
bacterial cultures and serological tests.
Introduction
Among the many constraints which limit productivity in
livestock populations, disease and parasites are of major
importance. In Africa, however, little is known about Results
small ruminant diseases and parasites.
In Tanzania there is a dearth of information on Table 1 shows the major diseases identified and their
specific causes of sheep and goat mortality and prevalence.

Small Ruminant Research Network Workshop 117

LE. G. Mboera and J.l. Kitatyi

Table 1. Diseases diagnosed in small ruminants classified disease reported in this study. Dermatomycosis was the
by major aestological groups. only fungal disease reported in the area.
Per cent of
Number of total Table 4. Other infectious diseases diagnosed in small
Disease group diagnoses diagnoses ruminants.
Parasitic infections 278 53.1 Number of
Bacterial infections 124 23.6 Disease condition diagnoses Per cent
Other infectious diseases 66 12.6 Caprine ophthalmia 28 42.4
Non-infectious diseases 56 10.7 Contagious ecthyma 21 31.9
Total 524 100.0
Enteritis 9 13.6
Dermatomycosis 6 9.1
Of the parasitic infections, helminthiasis was the
most common (Table 2). Coccidiosis, flea Peritonitis 2 3.0
(Ctenocephalides canis) infestation and Oestrus ovis
larvae infestation were other parasitic infections Table 5. Non-infectious disease conditions diagnosed in
small ruminants.
diagnosed in high incidences. Mange and anaplasmosis,
though in lower incidences, were also reported. Number of
Bacterial infections ranked second in the diseases Disease condition diagnoses Percent
affecting small ruminants (Table 1). There were high Fracture 15 26.7
incidences of foot rot, abscess, pneumonia and Wounds 10 17.8
dermatophilosis (Table 3). Other bacterial diseases Poisoning 9 16.1
occasionally diagnosed included anthrax, pasteurellosis, Dystocia 5 8.9
streptococcosis, mastitis, caseous lymphadenitis,
Retained placenta 4 7.1
tetanus, colibacillosis and brucellosis.
Stillbirth 3 5.4
Senility 3 5.4
Table 2. Parasitic infections diagnosed in small ruminants.
Starvation 2 3.6
Number of
Disease condition diagnoses Per cent Hernia 2 3.6
Helminthiasis 278 64.4 Abortion 2 3.6
Coccidiosis 52 18.7 Rectal perforation 1 1.8
Flea infestation 27 9.7
Oestrus ovis 12 4.3 Fractures, wounds and poisoning were the
larvae infestation commonest non-infectious conditions (Table S).
Mange 5 1.8 Reproductive problems such as abortion, dystocia,
Anaplasmosis 1.1 stillbirth and retained placenta were recorded. Hernia,
3
senility and rectal perforation were also observed.
Table 3. Bacterial diseases diagnosed in small ruminants.
Number of
Disease condition diagnoses Per cent Discussion
Foot rot 47 37.9
In small ruminants, the majority of diseases reported in
Abscess 26 21.0 central Tanzania were parasitic infections with
Pneumonia 24 19.4 helminthiasis and coccidiosis being the most prevalent
Dermatophilosis 11 8.9 M atovelo et al (1987) observed that the diagnosed
Anthrax 4 3.2 disease conditions affecting sheep and goats in the
Pasteurellosis 3 2.4 eastern part of the country were mainly parasitic
gastroenteritis.
Streptococcosis 2 1.6
The economic impact of helminthiasis is
Mastitis 2 1.6
enormous. Herlich ( 1 978) estimated that the total animal
Caseous lymphadenitis 2 1.6
loss due to helminthiasis worldwide is equivalent to 30
Tetanus 1 0.8 million goats and sheep. Constraints to animal
Colibacillosis 1 0.8 production attributable to helminthiasis include
Brucellosis 1 0.8 inefficient food conversion, poor growth and reduced
fertility.
Table 4 shows other infectious diseases, including Protozoan diseases of small ruminants have
viral, fungal and systemic diseases for which the received very little attention in Tanzania. Coccidiosis
etiological agent could not be determined. In this group, has been reported as the major protozoan disease of
incidences of caprine ophthalmia had the highest small ruminants in the region. Other protozoan diseases
frequency. Contagious ecthyma (orf) was the only viral of small ruminants have been reported in the country.

118 Small Ruminant Research Network Workshop

 Diseases of small ruminants in central Tanzania

Toxoplasmiosis has been diagnosed serologically in reported from the viewpoint of human health, the
southern Tanzania (Connor et al, 1 985 as cited by Otaru, disease is important because the causative organism
1985), whereas trypanosomiasis has been reported in the causes human brucellosis when fresh milk from infected
eastern part of the country (Matovelo et al, 1987). goats or sheep is consumed or when direct contact is
Ectoparasitism accounted for about 15.8% of the made with infected placenta following abortion.
parasitic diseases reported in this study. Of the Caseous lymphadenitis, colibacillosis, mastitis,
ectoparasites, Ctenocephalides canis appears to be the streptococcosis and tetanus are also important bacterial
most important in the region. Fleas have been reported diseases of sheep and goats. Caseous lymphadenitis is
to occur in sheep and goats in other areas of the country. usually clinically unapparent and is observed during
Kilonzo and Khama (1989) observed that goats and post-mortem examination. It causes high economic
sheep in north-east Tanzania were heavily infested with losses through condemnation of affected carcasses and
the cat flea Ctenocephalidesfelis. General infestation of organs at slaughter and through losses in production and
sheep and goats by fleas produces debilitating effects on reproduction (Kuria and Ngatia, 1990). Few cases of
their hosts. Heavily infested goats and sheep are weak, mastitis in small ruminants have been reported in other
anaemic and emaciated. Heavy flea infestation of lambs areas of the country (Madsen and Mtenga, 1989).
and kids may cause death. Of the other infectious diseases, caprine
Oestrus ovis larvae infestation was the second ophthalmia and contagious ecthyma have been reported
most prevalent ectoparasitism in small ruminants. O. with higher frequencies. Caprine ophthalmia causes
ovis larvae infestation has also been recorded elsewhere ocular pain and poor vision, so affected goats do not eat
in Tanzania (Matovelo et al, 1987). The larvae at the properly and fail to thrive. Where treatment is delayed,
nostrils of sheep or goats annoy the animals and cause affected goats become blind. Blind goats are difficult to
them to avoid the fly and seek shelter. Snorting, move and are often discriminated against when they are
stamping of the feet and shaking of the head are sold at auction. Contagious ecthyma causes some
common. Such actions interfere with normal economic losses. It appears to be endemic in the region
grazing/browsing. The spiny surface of Oestrus ovis as many fanners have complained of several outbreaks
larvae causes irritation of the nasal mucosa resulting in in their flocks.
sneezing, mucopurulent discharge and difficult, snoring In the other infectious diseases group, enteritis,
respiration. Oestrus ovis may deposit larvae in the eye, dermatomycosis and peritonitis were also reported.
nose or on the lips of man and is thus an important Enteritis is in most cases manifested by diarrhoea.
myiasis in some countries (Blood et al, 1983). Although enteritis was reported in few cases in this
study, reports from eastern regions of the country have
Mange is another important ectoparasitic infection
shown that diarrhoea is one of the dominant problems
diagnosed in sheep and goats in central Tanzania. The
for small ruminants (Matovelo et al, 1987).
disease has been reported to occur in other areas of the
Dermatomycosis of animal origin may affect humans
country (Matovelo et al, 1987). Mange can be a very
and its diagnosis and treatment are often difficult (Blood
stubborn condition, particularly in goats, causing much et al, 1983).
damage to the skin.
There are many non-infectious diseases of sheep
Bacterial diseases were the second most prevalent and goats which result in the animal's death and cause
diseases of small ruminants in central Tanzania. Of the great economic losses, but have not received sufficient
specific bacterial diseases, foot rot was the most attention. Fractures, wounds, poisoning and dystocia
prevalent. Most outbreaks occur during the warm and represent the largest proportion of conditions in this
wet months of the year. In serious outbreaks, lameness group. Reproductive disorders such as abortion,
may be so severe that many animals are forced to walk dystocia, retained placenta and stillbirth contribute
on their knees. highly to flock infertility. Stillbirth and dystocia appear
Although pneumonia and abscess have been to be among the primary causes of pre- and perinatal
reported with higher incidences, in many cases they lamb and kid losses in the area. The economic losses
have been due to non-specific bacteria. Pasteurella attributable to injuries such as fractures, wounds and
species, however, were isolated in three cases of rectal perforation are difficult to assess owing to the
pneumonia. Several investigations have reported limited data available on their incidence, prevalence,
Pasteurella organisms as causing some types of morbidity and mortality rates. However, expenses
pneumonia in goats (Ngatia et al, 1989). charged in treatment and nursing may result in large
Dermatophilosis was another prevalent disease of small economic losses.
ruminants. Outbreaks have frequently been reported in
the region (TVA, 1991).
Though figures of anthrax reported in this study are
not very significant, outbreaks killing several hundred
Conclusions
cattle, sheep and goats have been reported in northern It can therefore be concluded that helminthiasis,
Tanzania (Mbise et al, 1984). Brucellosis is a cause of coccidiosis, foot rot, caprine ophthalmia, flea infestation
abortion and infertility in small ruminants (Tekelye and and pneumonia are the major specific diseases affecting
Kasali, 1990). Although only one case of brucellosis was small ruminants in central Tanzania. Although similar

Small Ruminant Research Network Workshop 119

L.E.G. Mboera and J.I. Kitalyi

diseases have been reported in other parts of the country, Centre, lst-3rd December 1987. Tanzania Veterinary
their relative frequency and economic importance have Association/Sokoine University of Agriculture, Faculty
not been assessed. In the absence of reliable of Veterinary Medicine, Morogoro, Tanzania. 5:
epidemiological and economic data, it is impossible to pp. 151-163.
quantify precisely the losses incurred as a result of small Mbise A N, Nyange J F С and Mbasha EMS. 1984. An
ruminant diseases in Tanzania. Although these have outbreak of anthrax in wildlife in Lake Manyara
been recorded at low incidences, the stock owners do not National Park, Tanzania. In: Proceedings of the 2nd
recognise early clinical signs. Consequently they are late Tanzania Veterinary Association Scientific Conference
in seeking veterinary advice and suffer losses through held at Arusha International Conference Centre, 4th-6th
lack of timely treatment. December 1984. Tanzania Veterinary Association/
Sokoine University of Agriculture, Faculty of Veterinary
Medicine, Morogoro, Tanzania. 2: pp. 126-139.

References Ngatia T A, Kimberling С V, Johnson L W, Whiteman С E

and Lauenman L H. 1989. Experimental pasteurellosis in
Blood D C, Radostitis О M and Henderson J A. 1983. goats. Bulletin of Animal Health and Production in
Veterinary medecine. 6th ed. Bailleurs Tindall, London, Africa 37:13-19.
UK. 1310 pp.
Nyange J F С 1984. Animal health trends in northern
Herlich H. 1978. The importance of helminth infections in Tanzania. In: Proceedings of the 1 1th Scientific
ruminants. World Animal Review 26:26-29. Conference held at Arusha, Tanzania, 23-26 October
Kilonzo В S and Khama L R S. 1989. The effects of goats' 1984. Tanzania Society of Animal Production,
{Capra hircus) age and sex on flea infestation in Morogoro, Tanzania. p. 212.
Tanzania. Bulletin ofAnimal Health and Production in Otara МММ. 1985. Potential zoonotic diseases in the
Africa 37:61-66. southern zone of Tanzania. In: Proceedings of the 3rd
Kuria J К N and Ngatia T A. 1990. Caseous lymphadenitis of Tanzania Veterinary Association Scientific Conference
sheep and goats in Kenya. Bulletin ofAnimal Health and held at Arusha International Conference Centre, 3rd-5th
Production in Africa 38:15-18. December 1985. Tanzania Veterinary Association/
Sokoine University of Agriculture, Faculty of Veterinary
Madsen A and Mtenga L A. 1989. Dairy goats successfully Medicine, Morogoro, Tanzania. 3: pp. 367-398.
introduced in Tanzania. ILCA Newsletter 8(4):9. ILCA
(International Livestock Centre for Africa), Addis Tekelye Bekele and Kasali О B. 1 990. Brucellosis in sheep and
Ababa, Ethiopia. goats in central Ethiopia. Bulletin ofAnimal Health and
Production in Africa 38:23-25.
Matovelo J A, Semuguruka W D, Mella P N P and
Mwamengele G L M. 1987. Some indications on the TVA (Tanzania Veterinary Association). 1991. Professional
animal disease status in Tanzania. In: Proceedings of the reports. Tanzania Veterinary Association Central Zone
5th Tanzania Veterinary Association Scientific Meeting, Singida, October 25 1991. TVA, Morogoro,
Conference held at Arusha International Conference Tanzania.

120 Small Ruminant Research Network Workshop

 Mortalité des chevreaux croisés Alpine au Burundi
M. Ba, B. Rey2, Mesfin Shibre2

Projet caprin Ngozi, B.P. 1041, Bujumbura (Burundi)

2CIPEA, B.P. 5689, Addis-Abeba (Ethiopie)

Résumé In 1991, kid mortality was higher (P<0.05) than in 1991

for four months out of six. Mortality probability was
Les données recueillies en 1990 et 1991 sur 200 higher in triplets compared to twins or singles each
petits élevages de chèvres suivis dans le cadre du Projet month. Total effect of breed group was significant in
caprin de Ngozi ont mis en évidence un pic de mortalité June and July only, but the difference between survival
entre mars et juillet chez des chevreaux âgés de moins simulation curves showed that local kids performed
de 6 mois. Afin de simuler la survie de cohortes de better than the most common crossbreds (50% and 75%
chevreaux indépendamment des ventes et d'autres Alpine blood) with similar survival rates. Back crosses
raisons de sorties, le taux de mortalité des chevreaux a with local rams had an intermediary survival simulation
été estimé chaque mois de février à juillet par l'analyse curve. Each month, the most affected age groups were
de ressemblance maximum. Les facteurs de pondération the kids aged 60 to 120 days, which is roughly the
considérés sont l'année, l'âge du chevreau au premier weaning period, while the lowest mortality was
jour du mois étudié, le génotype, la taille de la portée et observed in very young kids. Cohort simulation showed
le sexe du chevreau. that 63% of the kids born in January or February will
En 1991, la mortalité des chevreaux a été quatre reach the age ofsix months.
mois sur six plus forte (P<0,05) qu'en 1990. La
probabilité de mortalité était chaque mois plus élevée Introduction
chez les triplets que chez les chevreaux issus de Située dans la région du Buyenzi, la province de
naissances simples ou gémellaires. L'effet global du Ngozi fait partie de la région des Hauts Plateaux du
génotype n'était significatif qu'en juin et juillet, mais la Burundi. Elle est caractérisée par une forte densité de
divergence des courbes simulées de survie indiquait une population (plus de 350 habitants/km ), une altitude
meilleure performance des chevreaux de race locale par moyenne de 1 800 m et une pluviométrie annuelle
rapport à ceux issus des croisements les plus fréquents moyenne de 1 400 mm répartie selon un régime bimodal.
(50% et 75% de sang alpin) et dont les taux de survie Depuis 1980 le Projet caprin de Ngozi développe
étaient similaires. Les rétrocroisements avec des boucs un programme de croisement des chèvres de race locale
de race locale présentaient une courbe simulée de survie avec des boucs de race alpine (8 000 saillies en 1989;
intermédiaire. Les classes d'âge les plus affectées Schmidt, 1991) ainsi que la production de lait (70 000
chaque mois correspondaient aux chevreaux âgés de 60 litres en 1991 contre 24 000 litres livrés à la laiterie en
à 120 jours, soit approximativement l'époque du 1986 (Michel et Mukankusi, 1992).
sevrage, alors que la mortalité des très jeunes chevreaux Les données recueillies en 1990 et 1991 dans la
était la plus faible. Par simulation de cohortes, on a zone d'intervention du projet ont mis en évidence un pic
montré que 63% des chevreaux nés en janvier ou en de mortalité des chevreaux âgés de moins de 6 mois
février atteindront l'âge de 6 mois. entre les mois de mars et de juillet.
La présente étude a pour objectifs de déterminer
l'influence d'un certain nombre de paramètres sur la
Mortality ofAlpine cross-bred mortalité des chevreaux entre février et juillet, et de
kids in Burundi comparer la survie des jeunes au cours de cette période
en fonction de leur âge et de leur type génétique.
Abstract
Data collected in 1990 and 1991 on 200 small-scale 2. Matériel et méthodes
goatfarms monitored by the Ngozi Goat Project showed
a peak in mortality between March and July in kids 2.1 Matériel
below six months old. The monthly mortality rate of the
kids was estimated using a maximum likelihood analysis Les données utilisées provenaient du suivi d'un
to simulate kid cohort survival irrespective ofsales and échantillon de 200 élevages de chèvres établis dans la
other reasons ofdisposal. Weighing factors were: year, zone d'intervention du Projet caprin de Ngozi. Les
age of kids on the first day of the month, breed group, exploitations ont été visitées tous les 15 jours par 8
litter size and sex of kids. enquêteurs, qui ont collecté les informations relatives à

Small Ruminant Research Network Workshop 121

M. Ba, В. Rey, Mesfin Shibre

la démographie et à la productivité du troupeau: variation considérés (année, genre, mode de naissance,

naissances, mortalités, entrées et sorties, évolution génotype et âge).
pondérale, production laitière. • Année
Le taux de mortalité variait en fonction de l'année.
2.2 Méthodes Le taux de mortalité en 1991 était significanvement plus
élevé en avril et mai (P<0,01), et en février et juillet
Le taux mensuel de mortalité des chevreaux a été (P<0,05).
utilisé pour étudier la mortalité des chevreaux de moins • Genre
de 6 mois durant la période considérée (début février à
Le sexe des chevreaux a peu influé sur le taux de
fin juillet). Il est défini comme le rapport entre le nombre
mortalité, qui n'était significativement plus élevé
de chevreaux morts durant le mois considéré et le
(P<0,05) qu'au mois de mai chez les chevreaux mâles.
nombre de chevreaux vivants au début de ce mois. Les
Pour les autres mois, il n'y avait pas de différence
chevreaux sortis pour d'autres raisons (ventes, dons,
significative entre les mortalités des chevreaux quel que
abattage familial) durant un mois donné ont été
soit leur sexe.
arbitrairement considérés comme vivants au dernierjour
du mois, mais ne sont pas entrés dans l'analyse du mois • Taille de la portée
suivant. La taille de la portée avait un effet significatif sur
L'analyse de vraisemblance maximum (SAS, le taux mensuel de mortalité aux mois de mars (P<0,01)
1987) utilisée pour l'étude des taux mensuels de et d'avril (P<0,05). Le taux de mortalité des chevreaux
mortalité a été effectuée en tenant compte des effets de portée double n'était pas différent de celui des
fixes suivants: chevreaux nés simples (sauf en avril). Celui des
- l'année; chevreaux nés de portées triples était significativement
- l'âge du chevreau au premier jour du mois étudié; plus élevé que celui des chevreaux nés simples sauf en
les chevreaux nés durant le mois sont exclus de mai, et en général plus élevé que celui des jeunes nés
l'analyse pour le mois considéré; doubles. Le taux de mortalité des triplets était en
- le génotype: chevreaux de race locale (petite moyenne deux fois plus élevé que ceux des autres types
chèvre d'Afrique de l'Est); chevreaux croisés (race d'agneaux, mais son coefficient de variation était
locale x race Alpine): FI (50% de sang alpin), RI également chaque fois plus important.
(75% de sang alpin), R2 et R3 (75% et 87% de sang Génotype
alpin respectivement); autres produits croisés issus Le génotype avait un effet significatif sur le taux
de la monte de chèvres croisées par un bouc de race de mortalité mensuel aux mois de juin (P<0,01) et juillet
locale; (P<0,05). En juin, la mortalité des chevreaux de race
- la taille de la portée ou le mode de naissance: locale et des croisements divers était similaire ou
simple, double ou triple; inférieure à celle des chevreaux FI, RI, R2 et R3. En
le sexe du chevreau. juillet, elle était plus élevée chez les chevreaux R2 et R3.
Les valeurs probabilistiques estimées des taux de Pour les autres mois, les seules moyennes n'ont fait
mortalité pour chaque classe d'âge et l'erreur type de apparaître aucune tendance constante.
l'estimation ont été obtenues par la procédure • Age des chevreaux
LOGMLVAR (Rege, 1992).
L'âge des chevreaux au premier jour du mois avait
Le taux mensuel de survie (Qs) a été défini comme
un effet significatif sur le taux de mortalité aux mois
étant l'opposé du taux de mortalité (Qs = 100 - Qm). Afin
d'avril (P<0,01), juin et juillet (P<0,05). Pour ces trois
d'étudier le comportement des chevreaux de moins de 6
mois, la mortalité mensuelle des chevreaux âgés de
mois au cours de la période considérée, on a simulé leur
moins d'un mois était la plus faible (toujours inférieure
survie en cumulant les taux de survie:
à 5%). La mortalité la plus élevée était enregistrée en
SURVIEmois2 = Qsl*Qs2 avril et juin chez la classe d'âge des 2 mois et en juin et
juillet chez celles de 3 et 4 mois. Une tendance similaire
Le taux de survie a ainsi une valeur probabilistique s'observait durant les autres mois, mais les différences
pour une cohorte donnée de chevreaux, n'étaient pas significatives.
indépendamment du système d'exploitation dans lequel
ils sont maintenus. 3.2 Survie simulée de février à juillet
Le tableau 2 présente les taux simulés de survie sur
six mois pour 100 chevreaux vivants et âgés de moins
3. Résultats de 6 mois au 1er février. Les facteurs de variation sont
les mêmes que ceux présentés dans la section 3.1.
3.1 Taux de mortalité mensuel • Année
Le tableau 1 présente les moyennes des taux de L'importante variation entre les taux mensuels de
mortalité mensuel des chevreaux de moins de 6 mois mortalité de 1990 et de 1991 se répercute sur le taux de
entre février et juillet en fonction des divers facteurs de survie cumulé au 3 1 juillet; ce dernier était supérieur de

122 Small Ruminant Research Network Workshop

 Mortalité des chevreaux croisés Alpine au Burundi

Tableau 1. Taux de mortalité mensuel de chevreaux de 1 à 6 mois entrefévrier etjuillet

Facteurs Février Mars Avril Mai Juin Juillet
Valeur réelle 2,35 3,77 5,30 7,47 7,60 5,30
Moyenne estimée 2,21 5,75 5,33 7,18 8,32 5,87
Année
(Prob) 0,0305 0,62% 0,0072 0,0006 0.6886 0,0405
1990 1,30 5,32 3,76 4,84 8.70 4,39
1991 3,73 6.20 7,51 10,55 7,95 7,81
Sexe
(Prob) 0,9116 0,4950 0,8242 0,0194 0,8984 0,7630
Femelle 2,16 6,36 5,47 5,59 8.21 6,11
Mâle 2,26 5,19 5,19 9,19 8,43 5,64
Mode de naissance
(Prob) 0,2231 0,0001 0,0240 0,2354 0,1049 0,1021
Simple 3,53 3,53 2,97 5,35 5,88 3,58
Double 2,94 3,00 5,33 6,46 6,90 4,86
Triple 0a 16,67 9,40 10,64 13,90 11,35
Génotype
(Prob) 0,2029 0,1905 0,0867 0,5430 0.0018 0,0484
Croisé 4,00 6,13 4,92 7,22 3,88 5,96
Fi 2,74 7,47 7,61 6,70 13,20 3,73
Ri 3,50 8,34 6.73 9,97 10,52 4,34
R2 et R3 1,27 4,94 5,68 5,57 13,64 13,69
Race locale 1,06 3,27 2,97 7,08 5.13 5,07
Age au 1er du mois
(Prob) 0,3244 0,1253 0,0003 0,1364 0,0156 0,0299
0-30 jours 0,81 3,59 1,93 2,84 4,23 2,05
31 -60 jours 2,53 5,94 8,28 6,23 6,58 3,70
6 1-90 jours 1,89 6,39 11,80 8,74 16,54 5.97
9 1-1 20 jours 3,72 14,20 3,37 9.17 10,80 11,84
121-150 jours 2,57 2,10 4,25 11,59 9,82 11,75
151-180 jours 3,08 8,25 7,98 8,05 6,44 6,07
0?rob) = valeur probabilistique
0* est une valeur exacte non incluse dans les analyses.

Tableau 2. Taux de survie de cohortes de chevreaux entre février etjuillet, Ngozi (Burundi)

Facteurs Février Mars Avril Mai Juin Juillet

Valeur réelle 97,65 93,97 88,99 82,34 76,08 72,05
Moyenne estimée 97,79 92,17 87,25 80,99 74,25 69,89
Année
1990 98,70 93,45 89,93 85,58 78,14 74,71
1991 96,27 90,30 83,52 74,71 68,77 63,40
Sexe
Femelle 97,84 91,62 86,60 81,76 75.05 70,46
Mâle 97,74 92,67 87,86 79,78 73,06 68,94
Mode de naissance
Simple 98,35 94,88 92,06 87,13 82,01 79,08
Double 97,06 94,15 89,13 83,37 77,62 73,85
Triple 0a 83,33 75,50 67,46 58,09 51,49
0* est une valeur exacte non incluse dans les analyses.

Small Ruminant Research Network Workshop 123

M. Ba, В. Rey, Mesfin Shibre

1 1 points (soit 18%) en 1990 par rapport à 1991 (74% La courbe de survie des chevreaux issus d'autres
de survivants contre 63%). croisements avait un comportement intermédiaire: au 30
• Sexe des chevreaux juillet, elle se situait à 8 points en dessous de celle des
chevreaux de race locale et à 9 points au-dessus de celle
En valeur cumulée fin juillet, la survie des femelles
des produits de croisements d'absorption.
était très légèrement ( 1 ,5%) supérieure à celle des mâles,
ce qui reflète l'absence de différences significatives • Mois de naissance
entre les taux de mortalité pour ce facteur. La figure 2 présente la courbe de survie simulée
pour la période considérée, pour des chevreaux nés à des
• Mode de naissance
mois différents. Jusqu'à l'âge de 4 mois, les jeunes nés
Sans être toujours significative, la différence entre en janvier ont une meilleure chance de survie que ceux
les taux de mortalité des chevreaux nés simples et de nés en février (3 points) et en mars (10 points). En
ceux nés doubles se traduisait par une probabilité de supposant une faible mortalité après l'âge de 180 jours
survie cumulée des premiers supérieure de 5,3 points ainsi qu'aux mois de décembre et janvier, et en tenant
(soit 7% de différence) à celle des seconds. compte de l'erreur inhérente à la simulation, il apparaît
Les chevreaux nés triples avaient une probabilité que la période de naissance la moins favorable se situe
de survie très inférieure à celles des chevreaux nés aux mois de janvier et de février (63% de survie espérée
simples ou nés doubles (28 points par rapport aux fin juillet).
premiers et 22 points par rapport aux seconds). La moitié
seulement des chevreaux nés triples survivait jusqu'à 6
mois pendant la période considérée. Figure [Link] simulée de chevreaux nés différents mois.
• Génotype Nombre vivant
L'évolution mensuelle entre début février et fin UU-]
>;C"'"'-«.~
juillet du taux de survie simulé de chevreaux appartenant ^ч^ \C"--,.
à différents génotypes et âgés de moins de 6 mois est
Vs- \
>o N N
illustrée à la figure 1 . 'X, S N
90-
^v4 \ 0 •.'•\
*\ ч S V \
Figure 1. Survie de cohortes de chevreaux de moins de \\ S V
6 mois de differents génotypes. •. \ \ N

Nombre vivant
80- ч'ОчЛ
** ^^^v V
106 >. -,.._ \ v\
\ \ ^v
F1 ,\ \ X
70- Dl '*•.*«-..., N s
m
90 - R2&R3 :-V.
Race locale лч
60- Ц . .T .... T , ,
1 Févr.28 Févr.31 Mars30 Avril 31 Mal 30 Juin 31 Jull.
80
Mois

Novembre
70- Décembre
Janvier
Février
Discussion
Mars Le taux de mortalité des chevreaux variait en
60 i 1 1 1 r~ fonction de l'année, du mois, du type de naissance et du
Févr. Mar. Avril Mal Juin 31 Juin. génotype. Pour les six mois considérés, il n'y avait pas
Mois
de différences entre les taux de mortalité des chevreaux
femelles et des chevreaux mâles, bien que la littérature
indique généralement une meilleure survie des femelles.
La comparaison des taux de mortalité aux précipitations
La courbe fait apparaître de faibles différences mensuelles relevées à la station du Projet (Projet caprin
entre les pourcentages d'animaux croisés (FI, RI, R2 et de Ngozi, 1991) et à leur somme pour les deux années
R3) survivants: 1 point entre chaque groupe, avec un ne permet pas de corroborer l'hypothèse selon laquelle
taux de survie croissant en fonction inverse du taux de une plus forte pluviométrie serait directement liée à un
croisement. Par rapport à ces chevreaux, la probabilité taux de mortalité plus élevé: la saison des pluies est à
de survie jusqu'à 6 mois des jeunes de race locale était son maximum en février-mars, se termine fin mai, et les
supérieure de 14,5 points. mois de juin et juillet sont parmi les plus secs.

124 Small Ruminant Research Network Workshop

 Mortalité des chevreaux croisés Alpine au Burundi

La mortalité supérieure des chevreaux nés triples 1990 et 1991 fait ressortir les effets de l'année, du mode
pourrait s'expliquer par les difficultés de leurs mères à de naissance et du génotype des chevreaux.
les allaiter. Parmi les causes possibles de mortalité, les
La race locale est généralement supposée plus verminoses joueraient un rôle important. Pour
rustique et mieux adaptée aux conditions tropicales que déterminer leur part dans le pic de mortalité, un
les produits de croisement. Les taux de mortalité des protocole expérimental est actuellement mis en place
chevreaux locaux étaient plus faibles aux mois de dans les exploitations couvertes par l'étude. Des
février, mars, avril et juin que ceux des chevreaux mesures prophylactiques efficaces seront ensuite
croisés, mais la différence n'était significative qu'au déterminées.
mois d'avril. Le taux de survie, estimé par simulation,
est cependant plus élevé pour la race locale. La
simulation a aussi montré que l'effet cumulé de la Références
mortalité des chevreaux était inversement proportionnel
au degré de croisement alpin (croisements d'absorption) Michel K. et Mukankusi F. 1992. Analyse de la livraison du
des animaux. lait et de la vente de fromage entre 1986 et 1992. VII.
Les classes d'âge affichant la mortalité la plus Rapport section suivi et évaluation. Projet caprin de
élevée au cours des mois considérés étaient celle des Ngozi, Ngozi (Burundi).
60-120 jours, soit approximativement la période du Projet caprin de Ngozi. 1991. Rapport annuel. Ministère de
sevrage. Ceci pourrait s'expliquer par le fait qu'à cette l'agriculture du Burundi. GTZ (Office allemand de la
période de leur vie, les chevreaux ont besoin d'une coopération technique), Ngozi (Burundi). 57 p.
attention particulière en raison des problèmes liés au Rege J.E.0. 1992. LOGMLVAR: A computer programme for
changement d'alimentation et aux risques d'infestation estimating predicted probabilities from maximum
parasitaire. Les principales causes de mortalité dans la likelihood estimates in a logit response function. CIPEA
région du Buyenzi seraient en effet d'origine parasitaire (Centre international pour l'élevage en Afrique),
et alimentaire (Shulze-Althoff et al, 1986). Addis-Abeba (Ethiopie).
Les taux de mortalité périnatale très faibles, voire Schmidt U. 1991. Ziegenzucht in Burundi : Ein Beitrag zur
nuls, enregistrés pour les chevreaux âgés de 0 à 30 jours Tierproduktion bei Kleinbauern. Entwicklung +
résultent certainement d'une erreur d'enregistrement Ländlicher Raum 4/9 1 .
des données. Schulze-Althoff K.J., Ntahimpeye C., Ngendakuma C.,
Nduvoimana A. et Karundere G. 1986. Guide pratique
pour éleveur de chèvres. Projet caprin de Ngozi, Ngozi
(Burundi).
Conclusion
SAS. 1987'. SAS/STAT Guideforpersonal computers, Version
L'analyse des facteurs influençant la mortalité des 6. SAS Institute Inc. Cary, Caroline du Nord (E.-U.).
chevreaux âgés de moins de 6 mois entre février etjuillet 1028 p.

Small Ruminant Research Network Workshop 125

V
 Stocking rate of sheep and survival of helminth infective
larvae on Brachiaria pastures
K.J.N. Ndamukong, S.E. Mbomi and S. Killanga

Institute ofAnimal Research, Mankon Station

P.O. Box 125, Bamenda, Cameroon

Abstract Les résultatsfont apparaître un accroissement du taux

d'helminthiase et de la contamination des pâturages
A study was undertaken at Mankon, Cameroon, to avec l'augmentation du taux de charge. La
determine a pasture grazing system for sheep that would contamination des pâturages qui portaient 29 animaux
produce optimum animal performance and herbage à l'hectare demeurait insignifiante jusqu'après la 12e
yield as well as limit the hazards of [Link] semaine de pâture. Le taux de charge n'avait
optimum stocking rate for Cameroon Grassland Dwarf presqu 'aucun effet sur le poids vif. Les résultats de la
ewes was determined by comparing animal régression de la production de matières vertes sur le
performance, herbage yield, pasture contamination and temps (P<0,01) variaient enfonction du taux de charge.
infectivity of the animals. The animals were grazed on Les larves d'helminthes survivaient sur les pâturages à
predominantly Brachiaria pastures at three stocking Brachiaria pendant 11 semaines environ au cours de la
rates: 29, 39 and 47 animals/ha. The results indicated saison des pluies. Il ressort de l 'étude que les meilleures
that helminth infestation rate and pasture contamination performances zootechniques et la production optimale
rose as the stocking rate increased. Pasture de matières vertes s'obtiennent avec un taux de charge
contamination at 29 animals/ha remained insignificant de 29 animaux par hectare, à condition de déplacer les
till after 12 weeks of grazing. Live weight was little animaux après trois mois afin de permettre la
affected by stocking [Link] results of the regression of décontamination du pâturage.
herbage yield over time differed (P<0.01) with the
stocking rate. Helminth larvae survived on Brachiaria
pastures for about 1 1 weeks during the wet season. This Introduction
study indicates that optimum animal performance and
herbage yield can be achieved at a stocking rate of 29 Stocking rate is the most important management factor
animals/ha, provided animals are moved after three influencing the output of animal products on pasture, the
months to allow for the previously grazed pasture to be stability and persistence of the pasture components and
decontaminated. also the financial returns which the farmer receives
(Humphreys, 1978).
Overstocking reduces the opportunity for selective
grazing. This may result in overgrazing, poor nutrition
Taux de charge ovine et survie of the animals and an increased risk of helminthiasis,
des larves infestantes with consequent reduction in the condition of the
animals. There is little information in Cameroon on the
d'helminthes surpâturages relation between flock management practices, the level
à Brachiaria of parasitism and the production of herbage. This study
was therefore undertaken to determine a pasture grazing
Résumé system for sheep that would produce optimum animal
performance and pasture yield as well as limit the
Une étude a été entreprise à Mankon (Cameroun) hazards of helminthiasis.
afin d'établir un système d'alimentation en vert destiné
à des ovins et qui permette d'obtenir des performances
zootechniques et une production de matières vertes Materials and methods
optimales, tout en limitant les risques d'helminthiase.
Pour déterminer le taux de charge optimum pour des Experiment 1. Studies on stocking rate
brebis Djallonké du Cameroun, on a procédé à une
comparaison des performances zootechniques, de la Recently bred Cameroon Grassland Dwarf ewes were
production de matières vertes, de la contamination des maintained at three stocking rates (29, 39, 47 ewes/ha)
pâturages et de l'infectiosité des animaux. Ceux-ci on paddocks of predominantly Brachiaria pastures for
étaient conduits sur des pâturages à dominante 14 weeks during the rainy season (July to September).
Brachiaria à raison de 29, 39 et 47 animauxpar hectare. The rates chosen were based on previous experience of

Small Ruminant Research Network Workshop 127

K.J.N. Ndamukong, S.E. Mbomi and S. KNIanga

grazing sheep and goats at the research station and were survival period on pasture for infective larvae of
similar to those used by Cameron and Gibbs (1966). The Haemonchus contortus and Trichostrongylus were
animals were dewormed with Ivermectin at 200 g/kg of assessed by Mann Whitney u test (Snedecor and
body weight prior to the start of the study. Cochran, 1967). The correlation between rainfall and
Supplementary feeding was not practised. However, relative transmission potential was assessed by
water and salt were provided ad libitum. Spearman's rank correlation coefficient (Snedecor and
Data collected included measurements of live Cochran, 1967).
weight, faecal strongyle egg counts, pasture larval
counts, feed availability measured at ground level
(forage yield/ha) and grazing pressure (leaf-stem ratio).
Results
Experiment 2. Survival of strongylid Liveweight changes
infective larvae on pasture
The data on liveweight changes under the three stocking
rates are shown in Table 1. The weights adjusted for
The method used was similar to that described by Okon lambing were not significantly affected by stocking rate
and Enyenihi (1977). The study was carried out on a 1/8 (P>0.05). The regressions had similar slopes and did not
hectare fenced area of level pasture land consisting differ in magnitude. Parturition and lactation, on the
predominantly of Brachiaria ruziziensis. Twelve other hand, generally resulted in a progressive decrease
experimental plots were used, each of 2 m and with in the live weight of the ewe for at least 10 weeks, the
boundaries marked by 20-cm wide shallow furrows. The decrease already being significant (P<0.001) by two
grass on each plot was maintained at a uniform height weeks post parturition.
of 6 cm by periodic clipping, the clippings being allowed
to remain on the plot. During the first three days of each
month throughout the calendar year, fresh goat faeces Parasitology
containing a known number of strongyle eggs were
The parasitological data are presented in Tables 2 and 3.
spread uniformly on one of the 1 2 plots. A sample of the
The infection rate generally rose as the stocking rate
faeces spread each month was cultured for two weeks at
increased from 29 animals/ha to 39 animals/ha, although
ambient temperature to determine the proportion ofeggs
the difference in egg count was not significant. The
of each species in the sample. One week after the last pasture stocked at 29 animals/ha remained free of
faeces were spread on the pasture, a 1 00 g grass sample contamination for 12 weeks from the start of grazing
was collected at random from the plot and infective
(Table 2). The animals grazing on this pasture
larvae were extracted and counted using a modification
maintained a median egg count of less than 500 eggs per
of the technique of Lancaster (1970). The collection of
g of faeces during this period. By contrast, the median
grass samples was continued weekly until no larvae
egg count increased to more than 600 eggs per g of
could be extracted for two successive weeks. The faeces between the 6 and 12 weeks in the ewes
procedure was repeated for each plot until all 12 plots stocked at 39 and 47 animals/ha. Egg counts and pasture
had been examined. Daily records of contamination were generally lower at 47 animals/ha
minimum/maximum temperature and rainfall were kept than at 39 animals/ha (Table 2). Ewes stocked at 29 and
throughout the study period. 39 animals/ha had lambing rates of 67 and 71%
Statistical analyses were carried out according to respectively. The post-parturient rise in faecal egg
the methods described by Steel and Torrie (1960). counts was more pronounced at 39 animals/ha. At 47
Differences in faecal egg counts between animals on the animals/ha only two ewes lambed, the first at 10 weeks
three stocking rates were determined by the and the second at one week before the end of the
Kruskal-Wallis test (Neter and Wasserman, 1974). experiment. The post-parturient rise in egg count was
Changes in liveweight with time within stocking rates evident in the ewe that lambed earlier, where the egg
were subjected to the paired t-test. The regression slopes count rose from 100 eggs to 3250 eggs per g of faeces
for forage yield were compared using multiple within 2 weeks of lambing. Haemonchus contortus was
regression (least squares regression). The differences in the predominant helminth species in the animals of the

Table 1. Mean initial liveweight and mean weight changes in Cameroon Grassland Dwarf ewes at three stocking rates on
predominantly Brachiaria pastures.
No. of Initial weight Liveweight change
Stocking rate (animals/ha) Class of animal animals Ott)
29 Ewes that lambed 7 32.3 -2.4
Ewes that did not lamb 2 30.5 4.5
39 Ewes that lambed 5 32.6 -3.4
Ewes that did not lamb 2 32.0 4.5
47 Ewes that lambed 2 35.5 2.0
Ewes that did not lamb 4 31.8 4.2

128 Small Ruminant Research Network Workshop

 Stocking rate of sheep and survival of helminth etc...

three groups, accounting for 9 1 % of the strongyle eggs Production of herbage

in the faeces. Trichostrongylus sp and Oesopha-
gostomum eggs together comprised only 9% (Table 3). The leaf-stem ratios measured from the eighth week of
the experiment are shown in Table 4. The leaf-stem ratio
Table 2. Larval contamination of pasture grazed by was lower at 39 and 47 animals/ha than at 29 animals/ha,
Cameroon Grassland Dwarf ewes at three stocking but the difference was not statistically tested. The
rates. regressions of herbage yield on time differed according
Stocking Infective larval counts per kg dry + to the stocking rate (P<0.01). The regression slopes were
rate matter (DM ) of pasture (weeks) -0.14, -0.17 and -0.21 at 29, 39 and 47 animals/ha,
(animals/ha) 2 4 6 8 10 12 14 respectively, but these were not significantly different
29
from each other.
0 0 0 0 0 0 146
39 0 0 0 0 278 250 268
47 0 0 0 20 105 86 124
Survival of infective larvae on pasture
Period after the start of study. The duration of survival of infective larvae on pasture
in relation to the month of deposition of the faeces

Table 3. Proportion of each strongyle egg type in faeces from sheep grazing at three stocking rates at Mankon, Cameroon.

Proportion of egg type in faeces (% )

Stocking rate (animals/ ha) Haemonchus Trichostrongylus Oesophagostomum
29 90.5 7.5 2
39 93.5 1.5 5
47 90.0 4.0 6
* Mean values at the 12 and 14 weeks from the start of the study.

Table 4. Leaf-stem ratio for pastures grazed by Cameroon containing the eggs is shown in Table 5. Infective larvae
Grassland Dwarf ewes at three stocking rates. developed from eggs deposited on pasture between
Leaf-stem ratio April and October (rainy season). None developed from
Weeks from the 29 animals/ 39 animals/ 47 animals/ eggs deposited between November and March (dry
start of grazing ha ha ha season) except in December when there were
unseasonal rains. The shortest survival time during this
8 1.0 0.5 0.3
period was for larvae hatched from eggs deposited at the
10 1.0 0.4 0.3 beginning of April. Infective larvae that hatched during
12 0.9 0.3 0.3 the rainy season survived on pasture for an average of
14 0.6 0.2 0.2 10 weeks, with a range of 6 to 13 weeks. Larvae of
H. contortus survived longer than those of

Table 5. Survival of infective larvae of H. contortus and Trichostrongylus on pasture at Mankon.

Month of
deposition Duration of survival of infective Number of infective larvae Relative transmission potential
of eggs on larvae on pasture (weeks) recovered (No. of days x % eggs)
pasture
1988-89 HC тс HC ТС HC тс
April 7 6 430 2884 0.98 38.64
May 13 11 6570 17455 16.38 154.77
June 10 10 4087 4214 32.20 51.10
July 11 9 4411 3564 25.41 15.12
August 10 9 1747 9057 45.50 103.95
September 12 12 27664 15561 10.44 39.48
October 13 9 7968 9504 29.12 97.65
November 0 0 0 0
December 6 6 338 56 0.84 0.42
January 0 0 0 0 0 0
February 0 0 0 0 0 0
March 0 0 0 0 0 0

Small Ruminant Research Network Workshop 129

K.J.N. Ndamukong, S.E. Mbomi and S. Killanga

Table 6. Seasonal variation in population of H. contortus (HC) and Trichostrongylus (TC) on pasture at Mankon.
Period (month No. of infective larvae on pasture plots % of infective larvae of each species on pasture
in 1988-89) HC TC Total HC TC
April 376 2704 3084 12.2 87.8
May 5248 16308 21556 24.3 75.7
June 4487 5167 9654 46.5 53.7
July 3068 2966 6034 50.8 49.2
August 3073 8548 11621 26.4 73.6
September 14496 12092 26528 54.6 45.4
October 18196 12816 31012 58.7 41.3
November 3178 1662 4840 65.7 34.3
December 981 62 1043 94.1 5.9
January 112 26 138 81.2 18.8
February 0 0 0 0 0
March 0 0 0 0 0

Trichostrongylus sp (1 1 and 9 weeks, respectively), but or were so few that they could not be recovered by the
the difference was not significant. extraction method used. The change in level of pasture
Table 5 also shows the number of infective larvae contamination and infestation rate with stocking rate
recovered from each experimental plot as a percentage was clearly evident as the stocking rate rose from 29 to
of the number of eggs spread and gives the relative 39 animals/ha. There was no evidence of an increase in
transmission potential of these larvae. A significantly pasture contamination or nematode burden with the
(P<0.05) greater proportion of eggs oí Trichostrongylus increase in stocking rate from 39 to 47 animals/ha.
sp successfully developed into infective larvae during Cameron and Gibbs (1966) obtained lower counts
the rainy season than did eggs of H. contortus. The of H. contortus, Ostertagia circumcincta and Tricho
relative transmission potential was highest in August for strongylus sp in lambs stocked at 48 animals/ha than in
H. contortus and in May for Trichostrongylus sp. The those stocked at 41 or 35 animals/ha. Michel (1964)
correlation between rainfall and transmission potential observed that at a heavier stocking rate the grass is
was significant (P<0.01) for both species (H. contortus: grazed low resulting in a smaller daily intake of herbage
r = 0.8; Trichostrongylus sp: r = 0.73). per head. While this would result in the ingestion of
Table 6 gives data on the proportion of each larval fewer larvae, a lower nutrition status resulting from low
type present on the plots during each month of the year. herbage intake may reduce the animal's resistance to
The data revealed that the proportion of infective larvae worm infection and to its harmful effects.
of H. contortus on pasture steadily increased as the rainy This probably explains why a stocking rate of 39
season progressed. The proportion of Trichostrongylus animals/ha resulted in lower faecal egg counts and lower
sp larvae on the pasture was highest at the beginning of pasture larval counts than a stocking rate of 47
the rains and decreased progressively towards the end of animals/ha. However, it is also probable that the higher
the rainy season.
lambing rate of the animals grazing at 39 animals/ha
(71%) could have resulted in a more rapid rise in faecal
egg counts (post-parturient rise) and pasture
Discussion contamination than occurred in the group stocked at 47
animals/ha. In this group a 33% lambing rate was
The results of this study show that the best performance
recorded only during the last four weeks of the
of Cameroon Grassland Dwarf ewes grazing on
experiment.
predominantly Brachiaria pastures can be obtained at a
stocking rate of 29 animals/ha. The changes in body The stocking rate of 29 animals/ha resulted in
weight adjusted for parturition were not different patch grazing with uneven decline in pasture yield since
between the three stocking rates. However, there was an there was an unlimited opportunity for selective grazing.
indication that grazing for longer than 12 weeks would At 39 and 47 animals/ha, the grazing pressure was heavy
depress body weight at stocking rates higher than 29 and the pastures were grazed more evenly. Overgrazing
animals/ha. was clearly evident at 47 animals/ha. The high stocking
More obvious was the effect of stocking rate on rate resulted in a rapid decline in herbage production and
helminth infestation. The greater mass of herbage at the less feed being available to the animals.
beginning of the study plus a washing effect by the rain Haemonchus contortus and Trichostrongylys sp
probably tended to reduce the concentration of larvae on are the predominant helminth species affecting sheep
the herbage. The larval count of zero during the early and goats in the high- and mid-altitude zone of
period of grazing meant that larvae were either absent Cameroon (Ndamukong et al, 1986). One of the most

130 Small Ruminant Research Network Workshop

 Stocking rate of sheep and survival of helminth etc...

interesting features of the trichostrongylids is their Acknowledgements

dependence on climate for development and survival
(Dunn, 1978). Rainfall appears to be the basic We appreciate the technical assistance of Oscar Ndi and
requirement for development and survival of eggs and Rene Degrace Tiako. We are grateful to the Cameroon
larvae of H. contortus and Trichostrongylus sp on Government for financing this study.
pasture. Shorb (1944) noted that a lowering of the
moisture content of the faeces raised the death rate of
the eggs. Temperature, on the other hand, did not appear References
to be a limiting factor for larval development and
Andersen F L, Wang G T and Levine N D. 1966. Effect of
survival. The mean minimum temperature (14.8 ± temperature on survival of the free-living stages of T.
1.8°C) recorded during the study was well above the colubriformis. Journal ofParasitology 52:71 3-72 1 .
minimum (10°C) at which infective larvae of H.
Cameron C D T and Gibbs H C. 1966. Effects of stocking rate
contortus and T. colubriformis can develop (Wang, and flock management on internal parasitism in lambs.
1967; Ndamukong, 1987). Canadian Journal ofAnimal Science 46: 1 2 1 - 1 24.
The results presented here reveal that a signi Crofton H D. 1957. Nematode parasite populations in sheep
ficantly greater proportion of eggs of Trichostrongylus on lowland farms. III. The seasonal incidence of species.
than of H. contortus developed into infective larvae Parasitology 47:304-3 1 8.
during the rainy season (Table 5). However, the high Dunn A M. 1978. Veterinary helminthology. 2nd edition.
fecundity constant of H. contortus (Crofton, 1957) William Heinemann Medical Books Limited, London,
enabled the latter to leave more surviving infective UK. 323 pp.
larvae on pasture at the end of the rains. Although eggs Humphreys L R. 1978. Tropical pastures andfodder crops.
placed on pasture during the dry season did not develop, Intermediate Tropical Agricultural Series. Longman
larvae that had developed during the last month of the Group Ltd, London, UK. 135 pp.
rains survived until January. This is in agreement with Lancaster M В. 1970. The recovery of infective nematode
the finding of Andersen et al (1966) that infective larvae samples. Journal ofHelminthology 44(2):219-230.
are more resistant to the influence of external conditions
Michel J F. 1964. Helminthiasis and grazing management.
than are eggs and pre-infective larvae. Agriculture London 7 1 (2):80-83.
Ndamukong K J N. 1987. Sheep and goat production in the
Conclusions North West Province of Cameroon with special
reference to parasitic gastro-enteritis. PhD thesis. The
These studies indicate that infective larvae of University of Edinburgh, Edinburgh, UK. 405 pp.
Haemonchus and Trichostrongylus on Brachiaria
Ndamukong К J N, Sewell M M H and Asanji M F. 1986.
pasture die after about 12 weeks on the pasture. The
Parasitic gastro-enteritis in sheep and goats at Bamenda,
great majority of eggs or larvae probably die within a N.W. Province of Cameroon. 1. Productivity
week after arrival on pasture, but enough survive for epidemiology. Revue Science et Technique. Serie
9-11 weeks to produce severe infestations in susceptible Sciences Zootechniques 2(4):101-117.
sheep and goats. Under a rotational system of grazing
management, optimum animal performance and Netter J and Wasserman W. 1974. Applied linear statistical
herbage yield can be achieved at a stocking rate of 29 models. R. Irwin Inc., Homewood, Illinois, USA. 842 pp.
animals/ha. The risk of parasitic gastro-enteritis at this Okon E D and Enyenihi U K. 1977. Development and survival
stocking level is minimal. However, animals should not of Haemonchus contortus larvae on pastures in Ibadan.
be allowed to remain on the same pasture for more than Tropical Animal Health and Production 9:7-10.
three months, as pasture larval contamination starts Shorb D A. 1944. Survival on grass plots of eggs and larvae
escalating after this time. Safe pastures can be produced of stomach worm H. contortus. Journal ofAgricultural
from contaminated pastures by resting them for at least Research 68:317-324.
three months. Since pastures are virtually free of
Snedecor G W and Cochran W G. 1967. Statistical methods.
contamination during the last two months of the dry
The Iowa State University Press, Ames, Iowa, USA. 593 pp.
season (mid January - mid March), the duration of
grazing on a pasture during this period could depend on Steel R G D and Torrie J H. 1960. Principles and procedure
herbage availability rather than on parasitic ofstatistics. McGraw-Hill, London, UK. 481 pp.
considerations. These findings can be used in grazing Wang G T. 1967. Effect of temperature and culture methods
pasture management to limit the intake of infective on the development of the free-living stages of
larvae from pasture and ensure adequate nutrition for the Trichostrongylus colubriformis. American Journal of
animals. Veterinary Research 28:1085-1090.

Small Ruminant Research Network Workshop 131

 Effect of strategie anthelmintic treatment on growth
performance and survival rate of native lambs in the
subhumid environment in Mali
12 3
S. Tembely , K. Bengaly and W. Berckmoes

Laboratoire central vétérinaire, P.O. Box 2295, Bamako, Mali

2Departement de la recherche sur les systèmes de productions rurales, B.P. 186 Sikasso, Malt
ZKoren Bloemstraat 5, 3911 ZG Rhenen, The Netherlands

Abstract été effectuée dans la zone subhumide du Mali de mai

1989 à avril 1990. Les gains moyens quotidiens (GMQ)
A study on the effect of strategic anthelmintic treatment sur six périodes de deux mois ont été analysés. Le GMQ
on the growth performance and survival rate of native des agneaux traités était significativement supérieur
lambs was carried out in a subhumid environment in (P<0,05) à celui des témoins pendant la saison des
Mali from May 1989 to April 1990. Daily weight gain pluies (de septembre à décembre), ce qui permettait
was analysed for six observation periods of two months d'obtenir un revenu net de 206 FCFA (1 $US = 250
each over the year. The daily weight gain of treated FCFA) par agneau. Cependant, cet avantage pondéral
lambs was significantly higher (P<0.05) than that of the n'a pas été maintenu pendant le reste de l'année en
control lambs during the rainy season (September to raison du déficit quantitatif et qualitatif des fourrages
December) with a net return of 206 CFA (1US$=250 au cours de la saison sèche. Dans l 'ensemble, il n 'y avait
cfa) per lamb. However, this weight gain advantage pas de différence significative entre les agneaux des lots
could not be maintained for the rest of the year due to 1 et 2 et le traitement anthelminthique n 'a pas influé sur
the shortage and poor quality of fodder during the dry le taux de mortalité des agneaux (P>0,10). Ce dernier
season. Overall, there was no significant difference in s'élevait à 19% chez les agneaux de l'expérience.
the daily weight gain of Groups 1 and 2 lambs and
anthelmintic treatment did not affect lamb mortality Pour maximiser laproductivitédes agneaux, ilfaut
(P>0. 10). The average mortality rate in the experimental leur administrer dufenbendazole seul deuxfois par an
lambs was 19%. (au milieu et à lafin de la saison despluies) et leur offrir
To maximise lamb productivity fenbendazole des aliments complémentaires pendant la saison sèche
alone should be given to lambs twice a year (in the suivante.
middle and at the end of the rainy season) along with Les résultats de cette étude font également
feed supplementation in the following dry season. apparaître que, dans la zone subhumide du Mali, les
The results of this study also suggest that gastro nématodoses gastro-intestinales n'ont pas d'effet
intestinal nematode infection has no cumulative dépressif sur les performances de croissance des
negative effect on lamb growth performance in the agneaux soumis à un faible niveau d'alimentation. Les
Malian subhumid environment with a low plane of principaux facteurs de la mortalité des agneaux élevés
nutrition. Primary causes of lamb mortality in this dans ce type de milieu devront être recherchés, la
environment should be further investigated as nematode mortalité due aux nématodoses se révélant peu
infections do not appear to cause many deaths. importante.

Effet de traitements anthel-

minthiques stratégiques sur les Introduction
performances de croissance et la In southern Mali, small ruminants (mainly composed of
viabilité des agneaux indigènes West African Dwarf sheep and goats) are owned by a
large proportion of the rural community on account of
dans la zone subhumide du Mali their ability to tolerate trypanosomiasis which is
endemic in mat area. Small ruminants serve as the main
Résumé source of animal protein for the rural community and
also provide a valuable income source for smallholder
Une étude de l'effet des traitements farmers, including women. These small animals are
anthelminthiques stratégiques sur les performances de hardy, easy to handle and highly prolific. The main feed
croissance et le taux de survie d'agneaux indigènes a supply is the native grassland adjacent to the villages,

Small Ruminant Research Network Workshop 133

S. Tembely, К. Bengaly and W. Berckmoes

sometimes supplemented by cereal grain stubble, straw Temperature and rainfall data were obtained from the
or crop residues. local climatology station.
Because small ruminants are kept by mixed
crop-livestock farmers as a sideline to their crop Statistics
oriented activities, they are very often neglected. They
are herded by village children during the cropping Average daily gain (ADG) analysis was performed only
season, but during the dry season are allowed to browse on animals which had been weighed at least twice. Each
and scavenge freely over the compounds and along the age group was analysed by one-way analysis of variance
roadsides. The animals sometimes stay in the bush for (ANOVA) using the SPSS package. Animals over 365
many days without any guard before returning home. days old were not included in the analysis.
Due to poor health management and inadequate
nutrition, the productivity of these animals remains low Results and discussion
(Wilson, 1982). An important factor responsible for the
mediocrity of small ruminant production in tropical Faecal egg counts
Africa is disease, particularly parasitism (Allonby and
Urquhart, 1975; Malaki, 1976; Vercruysse, 1985; No strongyle eggs were found in the faecal samples of
Shavulimo, 1986; Pandey, 1991; Waller, 1991). the treated lambs three days after drenching, indicating
The objective of this study was to evaluate the a good treatment response to fenbendazole. However,
effect of strategic anthelmintic treatment on growth three weeks later massive re-infestation was observed in
performance and survival rate of native lambs in a treated animals resulting in higher egg counts in some
subhumid environment. cases than in the control animals. This is indicative of
loss of host resistance following the total absence of
worms within the host as described by Georgi (1974).
Materials and methods Faecal egg count of the control lambs and the rainfall
pattern are shown in Figure 1 . Lambs started excreting
Area of study eggs in July with a peak in September (mean of 1700
eggs per gram of faeces). Few eggs could be detected
The production system is agro-pastoral in which small from November to April, which corresponded to the dry
ruminants are kept indoors at night and herded by village season.
children during the day as described above. The main
vegetation is Sudanese type with large broad-leaved
deciduous trees, tall grasses and shrubs. The climate is Figure 1. Faecal egg countsfor the control lambs and mean
subhumid, with rainfall ranging between 900 and 1 100 monthly rainfall.
mm per annum. There are two distinct seasons, the rainy Mean monthly epg Mean monthly rainfall (mm)
season from May to October and the dry season from
November to April. Average temperature is around 20°C 2000 -i 400
EPG count
with a maximum reaching 35°C in March/April and the
minimum 10°C in December/January. Rainfall - 350

1500 - 300
Experimental design
Between May 1989 and April 1990, 552 Djallonké - 250
lambs were systematically allocated at birth to three
treatment groups. Lambs ofgroups one and two received 1000 - 200
fenbendazole at 10 mg/kg" birth weight at three weeks
of age and every month thereafter until October 1989. 150
Lambs of group two received amprolium at the same
dose during the same period of treatment. A single 500 - 100
treatment was given to lambs born from October
onwards. Lambs of group three served as the control - 50
group and were left untreated.
T —1 I T^ 1 - 0
May 89 July Sept Nov Fab 90
Measurements
Month of the year
Faecal samples were taken directly from the rectum of
each animal every month from May to November 1989, Growth performance
then every two months from December 1 989 to April 90.
They were examined by the modified McMaster The average daily gain was significantly (P<0.05)
technique for faecal egg counts (FEC). Liveweight affected by period of year in all groups combined
change as a measurement of growth performance was (Tables 1). Differences of 7 to 15 g/day between the
monitored by lift-weighing individual animals every 45 treated lambs (Group 1) and the untreated lambs (Group
days. Mortality was recorded as it occurred. 3) were observed in October/November but could not be

134 Small Ruminant Research Network Workshop

 Effect of strategic anthelmintic treatment on etc...

Table 1. Average daily gain of experimental lambs for different age groups.
Period of the year

Treatment group May-June July-Aug Sept-Oct Nov-Dec Jan-Feb Mar-Apr Mean

0-183 days
All groups combined 94.1 66.5 22.4 49 52.7 60 57.45
Group 1 88.6 71.5 33.3 52.1 54.9 64.0 59.8
Group 2 97.0 66.3 29.0 48.1 52.4 54.7 54.0
Group 3 96.3 64.8 13.3 48.1 50.1 64.0 54.5
184-385 days
All groups combined 42 24. 1 10.0 18.3 22.8 40.5 20.18
Group 1 50.9 22.4 -7.5 25.8 18.1 30.3 20.6
Group 2 39.8 22.1 0.08 20.9 32.9 44.6 24.2
Group 3 (control) 39.4 25.7 16.8 4.8 18.7 51.6 17.3
0-365 davs
All groups combined 71.9 43.6 5.2 31.3 34.8 35.9 37.17
Group 1 75.4 41.1 9.2b 37.4a 35.8 29.8 36.2
Group 2 68.4 46.1 15a 35a 41.7 39.4 38.2
Group 3 (control) 71.5 43.3 -2.6e 22. lb 27.4 40.1 32.2
abc =Means within a column that have different superscripts are significantly different (P<0.05)

maintained for the rest of the year. Shortage of forage Cost-benefit analysis
and its poor quality may account for this. A study of the
availability and quality of forage in the region indicated In an attempt to estimate the influence of anthelmintic
that quality was low between November and the onset treatment on liveweight change under field conditions,
of the rainy season, i.e. May-June (Berkmoes and a cost-benefit analysis was performed. Labour costs
Diabate, 1988). were omitted from the analysis because the opportunity
costs were negligible. Variations in growth rates (Table
Rate of mortalities 1) had a negative impact on returns. Table 3 shows the
net profit for lambs of Group 1 at the end of the rainy
Table 2 shows that mortality rates varied according to season, but this advantage could not be maintained for
the period of the year. Mortality rates were similar for the rest of the year.
all groups during the pre-weaning period at four to five
months of age. Peaks were observed in June/July and Table 3. Cost-benefit analysis of anthelmintic treatment at
October/November, corresponding to the beginning and the end of the rainy season (Oct. 89) for lambs of
the end of the rainy season respectively. Group 1.
These observations agree with reports by Traore
Gross Anthelmintic Net benefit
and Wilson (1988), Bourzat (1984) and Wilson (1982) ADG Period Gain return cost of treatment
in the agro-pastoral system in West Africa. Causes of (gm) (day) (kg) (CFA) (CFA) (CFA)
mortality were not fully investigated, but apparently
were not associated with parasite burdens. Earlier 13.5 120 1.62 406 200 206
studies in the zone by Berkmoes et al (1988) indicated 1 US$ = 250 CFA
that deaths of lambs were more likely to be due to
malnutrition, disease and management (losses,
predators).
Conclusion
Table 2. Rate of mortality (0-12 month old lambs).
Although fenbendazole at the dose of 10 mg/kg"-1 body
Rate of weight was found to be highly effective against
Number mortality
Date of birth of births Mortalities % gastro-intestinal nematodes, the strategy adopted in the
experiment could not maintain a low level of eggs. It
May/June 45 11 24.0 eliminated the existing worm population but the animals
June/July 24 8 33.3 were quickly reinfested. Unless treatment is continued
July/August 16 2 12.5 it is unlikely to reduce the worm burden. However,
continuing treatment also precipitates drug resistance,
August/September 12 1 8.3 as pointed out by Nansen ( 1 99 1 ) and Waller ( 1 99 1 ). The
September/October 38 9 23.7 most effective approach is a combination of good
Nov 89-May 90 management, veterinary care and an improved plane of
50 5 10.0
nutrition. However, anthelmintic and other drugs must
Total 185 36 19.0 be available to the farmers at an acceptable price.

Small Ruminant Research Network Workshop 135

S. Tembely, К. Bengaly and W. Berckmoes

Pandey V. 1991. Strategies for development of basic

Acknowledgment applicable and sustainable helminthological research in
The authors gratefully acknowledge the contribution developing countries. Expert Consultation on Helminth
made to this report by R H Bosman and extend Infections of Livestock in Developing Countries. FAO
(Food and Agriculture Organization of the United
appreciation to Y Waigalo, P A Diarra, Z M Yattara and
Nations), Rome, Italy.
D Diabate for their valuable technical assistance.
Shavulimo R S. 1986. Control measures of helminths in small
ruminants: A review. In: Adeniji К О and Kategile J A
(eds), Proceedings of the workshop on the improvement
References of small ruminants in Eastern and Southern Africa.
Nairobi, Kenya, 18-22 August 1986. OAU/IDRC
Allonby E W and Urquhart G M. 1 975. The epidemiology and (Organization of African Unity/International
pathogenic significance of haemonchosis in a Merino Development Research Centre), Nairobi, Kenya.
flock in East Africa. Veterinary Parasitology pp. 217-226.
1:129-143. Traore A and Wilson R T. 1988. Livestock production in
Berkmoes W and Diabate D. 1 988. Rapport d'activité. DRSPR central Mali: Environmental and pathological factors
(Division de recherches sur les systèmes de production affecting morbidity and mortality of ruminants in the
rurale). VF Sikasso. 1ER (Institut d'économie rurale), agropastoral system. Preventive Veterinary Medicine
Mali. 6:63-75.
Bourzat D. 1 984. Synthèse des essais d'alimentation menés au Vercruysse J. 1985. The seasonal prevalence of inhibited
centre d'appui zootechnique de Ouaguiya (Bourkina development of Haemonchus contortus in sheep in
Fasso). IEMVT (Institut d'élevage et de médecine Senegal. Veterinary Parasitology 1 7(2): 159—163.
vétérinaire des pays tropicaux), Maisons-Alfort, France. Waller P J. 1991. The status of anthelmintic resistance in the
Georgi J R. 1974. Parasitologyfor veterinarians. 2nd edition. world, its impact on parasite control and animal
W В Saunders Co., Philadelphia, USA. production. Expert Consultation on Helminth Infection
of Livestock in Developing Countries. FAO (Food and
Malaki A. 1 976. A check-list of helminth parasites of domestic Agriculture Organization of the United Nations), Rome,
animals in Nigeria. Bulletin of Animal Health and Italy.
Production in Africa 3:301-305.
Wilson R T 1982. Husbandry, nutrition and productivity of
Nansen P. 1991. Strategies for the development and goats and sheep in tropical Africa. In: Gatenby R M and
establishment of cost-efficient and sustainable control Trail J C M (eds), Small ruminant breed productivity in
programmes for helminth infections in the developing Africa. Proceedings of a seminar held at ILCA, Addis
countries. Expert Consultation on Helminth Infections Ababa, Ethiopia, October 1982. ILCA (International
in Developing Countries. FAO (Food and Agriculture Livestock Centre for Africa), Addis Ababa, Ethiopia.
Organization of the United Nations), Rome, Italy. pp. 61-75.

136 Small Ruminant Research Network Workshop

 Intensification of goat production in Uganda and associated
health risk management
L. Siefert and J. Opuda-Asibo

Department of Veterinary Public Health and Preventive Medicine, Faculty of Veterinary Medicine
Makerere University
P.O. Box 7062, Kampala, Uganda

Abstract phenotypically. Stiff natural selection over the years,

however, has produced a population that could be better
Indigenous goats and traditional goat husbandry characterised by two phenotypes: the Small East African
systems in Uganda are described. Intensive management Goat which is a small and low weight type, and the
is suggested for the benefit of certain social target Mubende which is a fairly heavy and tall type. Their
groups. Corresponding husbandry, nutritional and spatial distribution is obviously not distinct and stunted
health requirements and risks, particularly peracute growth resulting from marginalisation could obscure an
clostridial diseases associated with high planes of otherwise better phenotypic performance. This is chiefly
nutrition and their management, are presented. due to opportunistic breeding, feeding and husbandry
Chemopro- and meta-phylaxis and vaccination methods. In 1963 the goat population was estimated to
programmes carried out in the isolation and zero-grazing be 2.5 million compared to 0.8 million sheep and 3.2
million cattle (Jameson, 1970). Devendra and Burns
units can substantially reduce reinfection and
(1970) compiled the following data on Ugandan goats:
corresponding drug exposure of hosts and parasites, and average birth weight of 2.1-2.4 kg; sexual maturity at
eventual drug resistance of parasites. These could also four months with an average body weight of 14-15 kg;
reduce costs and drug residues in the host. no seasonal breeding; three kiddings in two years with
10-30% twinning; and adult live body weight of 35 kg
(range: 20-50 kg). Over 1.5 million goats and sheep
Intensification de la production were slaughtered in 1960 for human consumption and
1.4 million skins were exported. Jameson (1970) and
caprine en Ouganda et gestion Okello and Obwolo (1985) stressed the significance of
des risques sanitaires associés goats as producers of meat, as it is generally preferred
over other meats. Mubende goat skins are also of
Résumé outstanding quality, but there is a prejudice against
consumption of goat milk outside Karamoja.
Cet article fournit une description des chèvres Socio-economically, goats play a traditional role as part
indigènes et des systèmes traditionnels de production of a dowry and, more recently, as a financial asset.
caprine de l'Ouganda. Un mode intensif d'élevage Amann (1972) reported that a pound of goat meat
profitable à certaines catégories sociales cibles est fetched 50 cents more than beef and highlighted the
proposé. Les besoins correspondants en matière de suitability of a smaller carcass within a rural context.
conduite des effectifs, de nutrition et de santé et les The average herd size ranges between 5-10 according
to Sherman (1991).
risques liés à ce type d'élevage, notamment les
clostridioses suraigiies associées à la conduite d'une Generally, traditional goat husbandry systems are
alimentation hautement nutritive, sont ensuite exposés. determined by land availability, intensity of crop
Les traitements chimioprophylactiques et cultivation and availability of young family members or
métaphylactiques ainsi que les programmes de underpaid herdsmen to attend to the goats. Where the
vaccination réalisés dans des conditions d'isolement et area of fallow and marginal land is comparatively small,
de zéro-pâturage engendrent une diminution tethering is the predominant practice, rather than
considérable des réinfections, donc de l'exposition des extensive grazing and browsing on free range together
hôtes et des parasites aux médicaments, soit, à terme, de with cattle. The characteristics of goat production
la chimiorésistance chez les parasites. Ils peuvent par systems are presented in Table 1.
ailleurs contribuer à réduire les frais ainsi que les Because of pressure on land due to an estimated
résidus de médicaments chez l'hôte. 3% annual human population growth and reduced
available child labour due to free primary education in
Introduction the near future, extensive production methods will have
Goats in Uganda have not been artificially selected to be replaced with intensive ones. This will require
along breed characteristics to distinguish them intensive research into potentials and limitations in order

Small Ruminant Network Workshop 137

L. Siefert and J. Opuda-Asibo

Table 1. Characteristics of traditional goat husbandry systems practiced in Uganda.

Comparison
Risks/requirements Tethering Zero-grazing
Labour inputs low high о very high
Capital inputs negligible modest о very high
Land inputs marginal favourable
Gains low high о very high
Controlled breeding unfavourable favourable
Theft more likely unlikely
Strangulation likely unlikely
Dog attacks very likely unlikely
Nutrition poor <> fair excellent
Forage losses more likely less likely
Plant poisoning possible unlikely
Snake bites likely unlikely
Management poor excellent
Crop-owner conflict likely unlikely
Environment damage likely unlikely
Intensification risky controlled & low risk
Health conditional good <> excellent
Rabies (risk of infection) high low/nil
Ticks and TBD-risks high low/nil
Helminth infection risks high low/nil
Reinfection very high very low
Fly attacks few <> many few
Disease resistance promoted lost
Drug exposure (H/P)** negligible low
Drug resistance (P)*** negligible low
Drug residues negligible low
TBD = tick-borne diseases **H/P = host/parasite ***P = Parasite

to utilise efficiently the abundant small ruminant It is advisable to vaccinate the goats twice within
resources in Uganda. a fortnight against enterotoxaemia and other clostridial
toxico-infections. Enterotoxaemia in particular has been
a significant killer of goats. Any considerable intake of
Considerations for the concentrates could precipitate this toxico-infection with
intensification of goat production only half an hour between the onset of clinical signs and
peracute death, leaving no time for successful therapy.
Covexin 8 multivalent vaccine (Cooper) conferred at
Health risks management
least partial protection against the aetiological agents
Only clinically healthy individuals without any types A and D of Clostridium perfringens. The
behavioural abnormalities should be selected. All goats incidence of the disease was reduced and the few
kept under intensive regimes must be free from affected suffered a more chronic course with more time
tuberculosis, brucellosis and certain other zoonotic for charcoal soda drench/intraruminal injection therapy
diseases. Standard criteria for the interpretation of (Thedford, 1990). However, many goats reacted
intradermal comparative tuberculin-test after visibly, forming sterile abscesses at the cervical
inoculation of bovine and avian PPD, using the
hypodermic inoculation site of the vaccine.
Hauptner test kit and reading results 72 hours later,
are to date not available. Similarly, no conclusive The contagious ecthyma (Orf) epidemic proved to
evidence is available to verify the prevalence of brucella be another killer disease, predominantly of kids.
melitensis in Uganda (Nakavuma, personal Attenuated tissue-culture vaccine, non-pathogenic to
communication, 1991). human, is not yet available. Vaccination with this

138 Small Ruminant Research Network Workshop

 Intensification ot goat production in Uganda etc...

Table 2. Combined/alternative essential isolation measures.

Risks Identification Measures
Ticks and tick-borne diseases (TBD) collection Ivermectin TM SC or
ante/post mortem Deltamethrine percutaneous

TBD parasitological chemo-vaccination

blood/serum tests antibiotic metaphylaxis

Tsetse/biting flies improved and treated traps reducing population and Deltamethrine
collection percutaneous

Trypanosomes parasitological drug availability? chemoprophylaxis/

bloodAserum tests prophylaxis/metaphylaxis?

Mycoplasmoses clinical/serology LA-tetracyclines

Clostridial toxicoinfection bacteriologically polyvalent clostridial
GC VFA-pattern toxoid vaccine 2 3 ml
Enterotoxaemia 6-week interval (and/2weeks ante partum)
Tetanus, etc.
Contagious ecthyma (Orf) clinical/serology attenuated TC-vaccine
no human pathogenicity

vaccine should be given priority since enterotoxaemia environmental control. Most parasites or their vectors
can be better controlled by delayed concentrate requiring a suitable microenvironment can be
provision. Inoculation of fully virulent vaccine in a more successfully controlled by altering the environment to
suitable site could become a risky alternative because of the parasites' disadvantage. Zero-grazing units offer a
the zoonotic properties of the Orf virus. controlled environment. Smooth walls and clean
Both ectoparasitosis (including nose-bots and concrete floors or deep litter that absorbs moisture and
tick-borne diseases) and roundworm endoparasitosis covers infective stages are not conducive to parasite
can be controlled by the injection of avermectine. survival and propagation. Once the animals have been
However, it is not always available and costs may be cleared of parasites in the isolation unit they can be
prohibitive even when it is available. Ticks and flies maintained on low challenge, providing the
(including tsetse) can be controlled by spot-on cut-and-carry fodder originates from an enclosed
(Cooper) application and a suitable broad spectrum controlled area.
anthelmintic such as Fenbendazole or, where still active,
Thiabendazole. Long acting tetracyclines can control
subclinical infections, especially subclinical Improved housing
mycoplasmoses. Sufficient time should be allowed
between individual measures and withdrawal periods Two types of housing are proposed: a high cost, more
must be observed. durable type for high income groups and a low cost, less
durable type for low income groups.
Non-endemic diseases can attain epidemic
proportions (in case of infections); this requires
diagnostic monitoring and corresponding modification Low cost version
of isolation periods. Presently the period covers at least
Roofing is similar to traditional grass thatched huts with
three weeks which seems insufficient for Orf. But since
an additional polythene sheath between the upper and
the viability of the virus is extremely high and of long
lower grass or papyrus-mat layers at the lower edge of
duration, any extension of the isolation period becomes the roof. The plastic sheath can be folded and suspended
an economic issue. In Table 2, alternative isolation to make a gutter for rain-water collection. The enclosure
measures are presented in view of drug costs, ease of is made of scrap timber ensuring that the lower 50 cms
application and side effects. of a 1 .50 m fence is keeping the deep litter coffee husk
Once the goats have been carefully selected and layer of at least 30 cm inside. The animals are kept in
isolated, certain prophylactic and metaphylactic pens (same height) and fed from troughs; water is
measures have to be taken into account: essentially, provided ad libitum. No additional walking area is

Small Ruminant Network Workshop 139

L. Siefert and J. Opuda-Asibo

provided. Bedding must be changed when faeces and are not systematically exploited. Merely importing
urine visibly accumulate; it provides excellent manure exotic breeds is not an effective exploitation measure.
for crops and forage. The few exotic goats which have found their way into
Uganda as an adjunct to NGO (Non Governmental
Organisation) sponsored zero-grazing cattle systems
High cost version have been introduced without due consideration of the
Roofing is made from corrugated iron sheets. The inputs required to meet expected outputs. Within
surface is cement, and a cement walking area is opportunistic husbandry systems, the exotics (British
provided. Resting places are roofed and have slatted Alpine, Toggenburger and Angora) cannot be expected
floors. Feeding is from troughs and mangers. The unit to display superior performance.
should be cleaned twice daily and the sewage processed The systematic exploitation of indigenous
in a biogas plant before being used to fertilise crops and resources should serve as a good starting point. Selection
forage. of well adapted indigenous goats is imperative so that
their adaptation to parasitic, nutritional and climatic
stress can be fully exploited. This should be followed by
Improved feeding education on scientific husbandry methods.
Most of the provisions which proved useful in After the farmers have received training on the
zero-grazing cattle systems have been experimentally advantages of scientific husbandry methods, suitable
adopted and are now under trial. Improved Pennisetum exotic breeds should be selected along their adaptive and
purpureum (Kawanda variety) cut as roughage at 1 m productive potential within each distinct environment
height, chopped to a length of approximately 5 cms and e.g. tropical montane, dry savannah, moist savannah,
offered ad libitum is supplemented by not more than 1/3 rain forest and islands. They should then be
Leucaena leucocephala to prevent mimosin performance-tested in controlled crossbreeding trials.
intoxication. This browsing species is drought-resistant Goats, like other small domesticated stock species
and apparently very palatable. A legume mixture of (rabbits, poultry, fish), are excellent experimental
Desmodium intortum and Uncinatwn, Centrosema material for studying indigenous resource potentials
pubescens, Macroptilium atropurpureum, Dolichos and utilisation alternatives. They could offer relevant
lablab and Stylosanthes spp. is readily taken as an and exciting exemplary learning and teaching
enriched supplement. Keeping a feeding calendar would alternatives in formal and adult education of farmers.
improve feeding management and cost effectiveness. Since women's ownership of goats is empirically much
Farm-produced concentrates provide fresh, high less contested in Uganda than the ownership of cattle,
quality meals vis-a-vis substandard commercial gender issues could be defused in a productive way in
products. Like any new ration component, concentrates this vital socio- economic sector.
must not be introduced abruptly otherwise bloat (in case Intensified goat production could provide socio
of overeating) or enterotoxaemia could cause death. economic care and vocational training opportunities for
Some dairy cow farmers still find it wasteful to over 1 .5 million war- and AIDS-orphaned young people
provide mineral supplements to their cows, but unlike (equivalent to >10% of the total population).
game, the cows are confined or restricted in their
Feedlot methods in or around cane plantations can
movements. Therefore they are less able to seek mineral
utilise cane-tops and other by-products to promote
rich soils from which they can obtain their mineral
offtake in seasonally overstocked areas. This makes
requirements. This situation is worse with small
better use of available biomass, which is often burned
ruminants and only free-range husbandry could allow
and does not need high capital investments.
goats to seek mineral-rich soils actively. However,
geophagia increases the risk of lesions in the oral and
gastro-intestinal tract linings and toxico-infection by
clostridial pathogens. Kids born with signs of bone
deformities which respond favourably to injectable Acknowledgement
mineral glucose depots or the provision of salt licks are
an indication of this condition. Intensification must GC-diagnostic typing was carried out in the Institute of
therefore provide for mineral supplementation either Applied Biotechnology in the Tropics at the
mixed in the ration or fed separately. Georg-August-University Goettingen, Federal Republic
Multipurpose indigenous goat breeds respond well of Germany; technical assistance of Prof Dr Seifert and
to the provision of crude rock salt blocks and, in Dr Habil Böhnel is gratefully acknowledged.
particular, of compound MEN (M = mineral, E = energy,
N = non-protein-nitrogen) salt licks.

Opportunities for and benefits of Editors' note

intensified goat production
Traditional or opportunistic husbandry systems produce Readers interested in the composition of MEN salt licks
hardy animals by natural selection; the potentials of this should contact the authors.

140 Small Ruminant Research Network Workshop

 Intensification of goat production in Uganda etc...

Okello K L and Obwolo M J. 1 985. Review of the potentialities

References of goat production. World Animal Review 5:27-32.
Amann V F (ed). 1972. Nutrition and food in an African Sherman D M.1991. Improved goat production in Uganda.
economy. Volume 1. Department Rural Economy, Opportunities for the private and public sector. A
Makerere University, Kampala, Uganda. pp. 193, 205. concept paper. Tufts University School of Veterinary
Medicine. North Grafton, Massachusetts 01536, USA.
Devendrá C and Burns M. 1970. Goat production in the
tropics. Commonwealth Agricultural Bureau, Farnham Thedford T R. 1990. Goat health handbook. A Winrock
Royal, Bucks, England, UK. pp. 13, 32, 35, 75. International information services publication. Winrock
International. University of Arkansas, USA. pp. 28-29.
Jameson J D (ed).1970. Agriculture in Uganda. Oxford
University Press, London, UK. pp. 114, 135, 318, 323,
342, 373.

Small Ruminant Network Workshop 141

 Enquête écopathologique sur les pneumopathies des chèvres
en saison fraîche au Tchad: un outil au service du
développement
R. Lancelot, M. Imadine, Y. Mopate et A. Idriss

Laboratoire de recherches vétérinaires et zootechniques

de Farcha, B.P. 433, N'Djamena (Tchad)

Résumé développement - Département d'élevage et de médecine

vétérinaire) ont donc décidé de réaliser une enquête pour
Cet article décrit la méthodologie d'une enquête tenter d'élucider ce problème, en s'imposant comme
écopathologique effectuée par le Laboratoire de Farcha contraintes de tenir compte de tous les phénomènes
pour déterminer les facteurs de risque intervenant dans susceptibles d'influer sur la pathologie respiratoire et
la pathologie respiratoire et la mortalité des caprins au d'obtenir rapidement des résultats vulgarisables.
Tchad. Des prémodèles conceptuels d'analyse à Pour ce faire, le Laboratoire a réalisé une enquête
l'échelle de l'individu, du troupeau de concession et du écopathologique dont la méthodologie est présentée ici
troupeau de jour ont été constitués afin d'assurer la mais dont les résultats seront publiés ultérieurement.
pertinence et l'exhaustivité des informations
recueillies. 2. Méthodologie
Surveying the ecology of lung
diseases in goats during the cold 2.1 L'écopathologie
season in Chad: A tool for Le concept d'écopathologie est apparu, en France,
development à la fin des années 70 devant l'inaptitude du modèle
pasteurien classique (un germe, une maladie) à
expliquer certaines pathologies de groupe. Considéré
Abstract sous l'angle de l'écopathologie, le niveau d'équilibre
This article describes the methodology ofan ecological entre performances zootechniques et sanitaires est la
survey of diseases used by the Farcha Laboratory to résultante des caractéristiques du milieu d'élevage, de la
define risk factors of lung diseases and mortality in pression des contaminants, des caractéristiques du
Chadian goats. Conceptual analytical preliminary peuplement animal et des pratiques d'élevage (Landais,
models at the individual animal, individual herd and 1991). Branche de l'épidémiologie, l'écopathologie a
collective herd levels have been designed to ensure pour objectifs de déterminer parmi ces éléments les
relevance and completeness of collected data. facteurs de risque des pathologies multifactorielles
(Madec et Fourichon, 1990) ayant une importance
économique et de proposer des solutions rationnelles
Introduction propres à rétablir un équilibre plus favorable à l'éleveur.
En saison fraîche, les pneumopathies des petits Jusqu'à présent, la démarche écopathologique a été
ruminants sont une préoccupation majeure des éleveurs appliquée aux élevages intensifs (Ganière et al, 1991).
tchadiens. De nombreuses recherches ont été effectuées Elle paraît intéressante pour les systèmes extensifs des
sur ce problème au Laboratoire de Farcha, recherches pays tropicaux car, par son approche globale, elle est
dont la synthèse a été faite par Lefèvre (1977). Celui-ci susceptible d'offrir une alternative aux programmes
estime que les pneumopathies spécifiques dues à un seul sanitaires classiques basés sur des prophylaxies
agent infectieux responsable de l'ensemble des médicales. De sérieux aménagements méthodologiques
symptômes et des lésions observées sont rares. En sont cependant nécessaires pour pallier les contraintes
revanche, les pneumopathies à étiologie multiple sont de liées à l'environnement tropical. Plusieurs tentatives
loin la cause essentielle de la mortalité par maladie sont en cours: étude des facteurs de risque d'avortements
infectieuse. Cette notion de pathologie multifactorielle chez la chèvre au Brésil (Faye et Quirin, 1991), de la
ne s' arrête pas aux seuls agents infectieux: les conditions mortalité des chevreaux au Zimbabwe dans le cadre du
de vie des animaux jouent également un rôle très Projet vétérinaire caprin, et des pneumopathies des
important (Le Jan et al., 1987). Le Laboratoire de Farcha petits ruminants au Sénégal dans le cadre du programme
et le CIRAD-EMVT (Centre de coopération Pathologie et productivité des petits ruminants (Merlin
internationale en recherche agronomique pour le etal., 1990).

SmaH Ruminant Research Network Workshop 143

R. Lancelot, M. Imadine, Y. Mopate et A. Idriss

2.2 Matériel ans, plus de 4 ans. Chacune des variables caractérisant

la pathologie respiratoire et la mortalité a été calculée
En termes d'épidémiologie, l'écopathologie sur les sous-cohortes. Les quotients ont ensuite été
s'appuie sur des enquêtes longitudinales de type suivi de établis pour chaque troupeau, en pondérant Г importance
cohorte. Il était donc nécessaire de trouver des éleveurs relative de chaque sous-cohorte des animaux des 59
coopératifs et facilement accessibles. Une typologie des élevages.
élevages de petits ruminants a été réalisée (Lancelot et
Mopaté, 1991), de manière à définir des critères
2.6 Les paramètres utilisés
objectifs de choix: éleveurs ne possédant pas de bovins
et dont le troupeau caprin s'élevait à 20 têtes au moins. Trois paramètres ont été utilisés pour décrire la
59 éleveurs ont été choisis dans 13 villages au nord de pathologie respiratoire: la fréquence, les récidives et la
N'Djamena, le long d'un axe goudronné, à un maximum gravité. La fréquence est le nombre total d'animaux
de 55 km de la capitale. malades observés par troupeau au cours des passages
successifs, rapporté à l'effectif du troupeau (cohorte).
2.3 Dispositif d'enquête Le quotient de récidive est le pourcentage d'animaux
chez lesquels des signes de maladie ont été observés plus
Un dispositif de suivi individuel a été établi pour d'une fois. Une échelle de gravité a été construite pour
déterminer les indicateurs des niveaux de productivité calculer une note individuelle de gravité, permettant
et de santé. Une fréquence bimensuelle des visites a été d'établir une note moyenne de troupeau (figure 1). La
instaurée. Les informations ont été stockées sur mortalité a été déterminée en rapportant le nombre des
micro-ordinateur, grâce au progiciel PIKBEU animaux morts (toutes causes confondues) pendant la
(Planchenault et Sahut, 1989) qui permet la saisie et le période de l'enquête à l'effectif de la cohorte.
contrôle des données. Il permet également de calculer
les indicateurs démographiques et de croissance usuels 2.7 Les échelles d'observation
et de réaliser des extractions de fichiers pour analyses
ultérieures. Les animaux ont été identifiés par des Les phénomènes à expliquer étant précisés, trois
boucles auriculaires plastifiées: au 1er juillet 1991,
échelles d'observation ont été choisies: l'individu
environ 3 500 petits ruminants étaient répertoriés, dont (caprin), le troupeau de concession (animaux
85% de chèvres. appartenant à un éleveur et passant la nuit dans la même
concession) et le troupeau de jour (animaux conduits au
2.4 Méthodologie d'enquête pâturage par un même berger). Un prémodèle
conceptuel d'analyse a alors été bâti pour chacune des
L'enquête s'est déroulée du 15 décembre 1991 au échelles d'observation. Ces schémas (figures 2, 3 et 4)
31 mars 1992. Entre juillet et décembre 1991, les résument toutes les interactions présumées entre les
éleveurs et les enquêteurs ont été formés au suivi. Les hypothèses de facteurs de risque et les phénomènes à
neuf mois allant de début juillet 1991 à fin 1992 expliquer. Leur élaboration est indispensable pour
représentent environ un cycle de production et l'on a s'assurer de la pertinence et de l'exhaustivité des
donc pu comparer les données entre élevages. informations recueillies. Les prémodèles conceptuels
Il était nécessaire de rendre compte de la d'analyse permettent la constitution d'une base de
complexité du système de production et de récolter des données rationnelle et facilitent le dépouillement de
informations couvrant toutes les facettes des élevages l'enquête.
suivis, notamment les aspects socio-économiques, Les fiches d'enquête ont ensuite été réalisées et
zootechniques, agropastoraux et pathologiques. Une testées. L'informatisation de la saisie a été préparée
équipe pluridisciplinaire a donc été constituée. Ce noyau avant le début de l'enquête, pour permettre
de chercheurs a réalisé des enquêtes auprès des éleveurs l'enregistrement et le contrôle des données en temps réel.
et consulté la bibliographie afin d'établir une liste
pertinente et exhaustive des hypothèses relatives aux 2.8 Analyse statistique
facteurs de risque. Il a également défini les phénomènes
à expliquer. Deux groupes d'indicateurs ont été retenus: A ce stade, les grandes lignes de la stratégie
le premier caractérisant la pathologie respiratoire et le statistique du dépouillement ont été tracées. Aucun
second, la mortalité. En effet, la pathologie respiratoire résultat préalable n'autorisait la définition d'un modèle
recouvre l'essentiel des symptômes rencontrés en saison pertinent d'interaction entre variables explicatives et
fraîche et il est vraisemblable qu'une part importante de variables à expliquer. Les techniques exploratoires ont
la mortalité observée lui soit imputable. donc été privilégiées (analyse des correspondances,
classification). Ces méthodes permettent de décrire la
2.5 L'échantillon structure des tableaux de données (Benzécri, 1973; Faye
et Lescourret, 1991), pour élaborer éventuellement un
La population étudiée était la cohorte des caprins modèle adéquat (Ducrot et Climarosti, 1991). L'analyse
présents au jour J0 de l'enquête. Cette cohorte a été a été effectuée dans deux sens, à savoir le sens ascendant
subdivisée en 4 sous-cohortes: 0-3 mois, 3-12 mois, 1-4 (c'est-à-dire de l'individu vers le troupeau de jour) et le

144 Small Ruminant Research Network Workshop

 Enquête écopathologique sur les pneumopathies .

Figure 1. Echelle de gravité des symptômes respiratoires

Animal Facteur de confusion Facteur de confusion

sain possible avec possible avec
l'oestrose pathologie intercurrente

LA MOYENNE DES NOTES RAPPORTEE A L'EFFECTIF TOTAL DU TROUPEAU

DONNE L'INDICE DE GRAVITE DES SYMPTOMES RESPIRATOIRES

Figure 2. Prémodèle conceptuel d 'analyse à l 'échelle de l 'individu

PATHOLOGIES PRATIQUES
MARQUEURS SANITAIRES
Age / Sexe Traitement

Small Ruminant Research Network Workshop 145

R. Lancelot, M. Imadine, Y. Mopate et A. Idriss

Figure 3. Prémodèle conceptuel d'analyse à l'échelle du troupeau de concession

GROUPE DE VARIABLES QUALIFIANT LE TROUPEAU CP/OV GROUPE DE VARIABLES QUALIFIANT

L 'ELEVEUR ET SES ACTIVTTES
FLUX Appartenance au
troupeau de village
ENTREE SORTIE
Taux d'animaux Taux d'exploitation
d'origine externe Taux de mortalité Nombre d'actifs dans
Taux de naissance et abattage maladies la famille
Taux de fécondité * <3 mois Effectif du
* 3à 12 mois
troupeau Autres activities que l'agro-élevage
* >12 mois
к
1
Ressources principales
* Bétail/Lait
Profils démographiques * Autres
PNEUMOPATHIES
% d'ovins Indice moyen de gravité
Type de logement
MORTALITE Entraves
Revenus de l'éleveur Quotients de mortalité
Densité animale
Prix de vente / Effectif Hygiène des locaux
Cohabitation

SEROLOGIE COMPLEMENTATION
(P13, mycopl.,PPR) * Fourrages
* Minéraux
Indicateur du * Sous-produits
„ microbisme à INDICATEUR DU
tropisme respiratoire STATUT NUTRITIONNEL
(seroconversions) (profil des notes)

COPROSCOPIE t
paramètre(s) discriminant(s)
TRAITE
Prélèv. de lait
ALLAITEMENT
Colostrum
par classes d'âge
par les éleveurs Adoption
Indicateur du
parasitisme
BIOCHIMIE GROUPE DE VARIABLES QUALIFIANT
NOTE D'ETAT CORPOREL GLU, AGL, Urée LE LOGEMENT ET LES PRA TIQUES
Ca, Mg
GROUPE DE VARIABLES INDICATRICES

Figure 4. Prémodèle conceptuel d'analyse à l'échelle du troupeau de jour

TROUPEAU DE JOUR
Effectif
Nombre de troupeaux
de concession
% d'ovins

BERGER
Age
Statut
Expérience
Indicateur d'attention

% d'animaux morts,
voles, disparus...
PNEUMOPATHIES
indice moyen de gravité
MORTAUTE
Taux moyen de mortalité

Typologie des parcours

CONDUITE AU PATURAGE
Rythme et temps consacré
au pâturage
Fréquence d'abreuvement
RATION AU PATURAGE Nature de l'esu

146 Small Ruminant Research Network Workshop

 Enquête écopathologique sur les pneumopathies ...

sens descendant (c'est-à-dire du troupeau de jour vers Il faut la conjonction de plusieurs facteurs dans un
l'individu). A chaque échelle, les facteurs associés à la même élevage pour entraîner une incidence accrue ou
mortalité ou à la pathologie respiratoire ont été une plus grande gravité de la pathologie. Dans les
recherchés puis reportés en éléments supplémentaires à programmes de vulgarisation, il sera donc nécessaire de
l'échelle supérieure (sens ascendant) ou inférieure (sens personnaliser les conseils aux éleveurs en fonction d'un
descendant). De cette façon, il a été possible de détecter diagnostic préliminaire (recherche des facteurs de risque
et de corriger les facteurs de confusion. dans chaque élevage).
A la fin du dépouillement, le groupe de travail s'est L'accent a été mis sur la recherche des facteurs de
réuni pour procéder à un examen critique des résultats risque relevant des pratiques des éleveurs: logement,
obtenus. Les thèmes améliorateurs ont été choisis et mis complémentation, traite, sevrage, conduite au pâturage.
en forme en vue de leur vulgarisation auprès des Les thèmes améliorateurs correspondront donc à des
éleveurs et des services de développement. éléments concrets et bien connus des éleveurs, ce qui
devrait contribuer au succès des programmes de
Discussion et conclusion vulgarisation qui découleront de ce travail.
Le dépouillement des résultats de l'enquête permet Au demeurant, l'écopathologie est un outil
de déterminer les facteurs associés à la pathologie performant adapté à la mise au point d'une politique
respiratoire et à la mortalité. En toute rigueur, on ne peut d'amélioration des productions et de la santé animales,
pas parler à ce stade de facteur de risque. Il s' agit en effet car elle appréhende globalement et dans sa complexité
d'observations et non pas d'expérimentations (problème le fonctionnement du système d'élevage.
de l'inférence causale). Cependant, à la lumière de
connaissances antérieures, des présomptions fortes
peuvent suffire à préconiser certains thèmes améliorateurs.
Les facteurs de risque rencontrés isolément dans Remerciements
un élevage n'ont guère de poids et peuvent même donner
Les auteurs remercient le Fonds d'aide et de
lieu à des interprétations contradictoires, comme le
coopération (France) et la Banque mondiale pour le
démontre bien l'exemple suivant tiré de l'enquête
financement de cette enquête. Ils remercient également
(figure 5). Ainsi, la seule présence de ténias ou de virus
le docteur B. Faye, directeur du Laboratoire
de la peste des petits ruminants est associée à un taux de
d'écopathologie de l'INRA-Theix (France), qui a veillé
mortalité assez proche de celui observé en l'absence de
à la rigueur scientifique de ce travail et les a aidés dans
ces deux facteurs. En revanche, dans les élevages où la
l'élaboration des prémodèles conceptuels d'analyse.
peste des petits ruminants et le ténia coexistent, le taux
de mortalité augmentait considérablement.

Figure 5. Relaliona entre la circulation du virus de la peste Références

des petits ruminants, l'infestation par le ténias et
le taux de mortalité des chèvres Benzécri J.P. 1973. La place de l'a priori. Encyclopaedia
universalis 17:1 1-24.
Quotient de mortalité (%) 11,2
Ducrot C. et Climarosti I. 1991. Complémentarité du modèle
12 (11) logistique et de l'analyse des correspondances pour la
Pas do circulation du virus PPR recherche des facteurs de risque en pathologie animale:
|| Circulation du virus PPR application à l'étude des facteurs de risque de l'ecthyma
10
5,1 Quotient de mortalités des ovins. Epidémiologie et santé animale 20: 1 26- 131.

(11) Nombre d'élevages concernés Faye B. et Lescourret F. 1991. Stratégie statistique du

Laboratoire d'écopathologie. Epidémiologie et santé
animale 20:103-116. Faye B. et Quirin R.1991. Etude
i 5,1 des facteurs de risque des avortements chez la chèvre.
Mise en place du pré-modèle conceptuel d'analyse dans
(11)
le cadre d'une enquête écopathologique réalisée dans le
district de Massaroca, Etat de Bahia (Brésil). INRA
3,8 (Institut national de la recherche agronomique), centre
(19) de recherches de Clermont-Ferrand/Theix, Leyrat
(France). 20 p.
1,5 Ganière J.P., André-Fontaine G., Drouin P., Faye В., Madec
F.,RosnerG.,FourichonC.,[Link].P. 1991.
(6) L'écopathologie: une méthode d'approche de la santé en
élevage. INRA Production animale 4(3):247-256.
Taenia - Taenia +
Lancelot R. et Mopaté Y. 1991. Typologie opérationnelle des
Infestation par les taenias élevages de petits ruminants en zone sahélienne du
Tchad. Rapport technique. Laboratoire de Farcha,
(Données analysées sur 47 troupeaux) N'Djamena (Tchad). 11 p.

Small Ruminant Research Network Workshop 147

R. Lancelot, M. Imadine, Y. Mopate et A. Idriss

Landais E. 1991. Ecopathologie et systémique. Etudes et Madec F. et Fourichon C. 1990. Les facteurs de risque en
recherches sur les systèmes agraires et le développement épidémiologic animale. Epidémiologie et santé animale
21:5-11. 18:31-43.
Lefèvre P.C. 1977. Pathologie infectieuse. In: Etude sur Merlin P., Ndiaye M. et Faugère O. 1990. Facteurs de risque
l'élevage des petits ruminants au Tchad. Publié sous la des pneumopathies des petits ruminants au Sénégal. I.
direction de R. Dumas. BIRD (Banque internationale Zone soudanienne, saison sèche 1988-1989. Document
pour la reconstruction et le développement/IEMVT de travail du programme PPR (Pathologie et productivité
(Institut d'élevage et de médecine vétérinaire des pays des petits ruminants) n° 19. LNERV (Laboratoire
tropicaux), Maisons-Alfort (France). p. 271 à 282. national de l'élevage et de recherches vétérinaires)/ISRA
Le Jan С, Sow A.D., Thiemoko С, François J.L. et Diouara (Institut sénégalais de recherches agricoles, Dakar-Hann
A. 1987. Pneumopathies enzootiques des petits (Sénégal). 21 p. + 14 tableaux.
ruminants en Mauritanie: situation d'ensemble et
approche expérimentale. Revue d'élevage et de Planchenault D. et Sahut С 1989. PIKBEU: manuel
médecine vétérinaire despays tropicaux 40(2): 1 03- 112. d'utilisation. IEMVT (Institut d'élevage et de médecine
vétérinaire des pays tropicaux), Maisons-Alfort
(France). 85 p.

148 Small Ruminant Research Network Workshop

 SESSION 4

Small Ruminant Feeds and Feeding

Systems
 Selectivity in sheep and goats fed Albizia, etc..

Selectivity in sheep and goats fed Albizia, Gliricidia,

Leucaena and Tamarind multipurpose trees
/ 2 1
L.A. Mtenga , D.M. Komwihangilo and G.C. Kifaro

Department ofAnimal Science and Production

Sokoine University ofAgriculture, P.O. Box 3004, Chuo Kikuu, Morogoro, Tanzania
Livestock Production Research Institute
P.O. Box 202, Mpwapwa, Tanzania

Abstract agricole de Sokoine, en Tanzanie, pour déterminer les

choix alimentaires de moutons Blackhead Persian et de
Selectivity of Blackhead Persian sheep and Small East petites chèvres d Afrique de l'Est recevant du fourrage
African goats fed Albizia lebbek, Gliricidia sepium, d 'Albizia lebbek, Gliricidia sepium, Leucaena leuco
Leucaena leucocephala and Tamarindus indica was cephala et Tamarindus indica Les 4 moutons utilisés
determined through behavioural and intake studies étaient âgés de 1 à 2 ans, et les 4 chèvres, de 4 à 5 ans.
conducted at the Sokoine University of Agriculture, Le poids des animaux variait de 20 à 29 kg. Pour l 'étude
Tanzania between May and July 1991. The four sheep de comportement, les animaux étaient enfermés dans
used were 1-2 years old. The four goats were 4—5 years des box de 170 x 160 cm. Le temps consacré par chaque
old. The weight of the animals was between 20 and 29 animal à l'alimentation, à la rumination et au repos a
kg. During the behavioural study, animals were confined été observé entre 9 et 1 7 heures pendant quatrejournées
into pens of 170 x 160 cm. Time spent by each animal on consécutives; la quantité de matière sèche des espèces
feeding, ruminating and idling activities was monitored ligneuses consommée par chaque animal a été relevée
from 0900 hours to 1700 hours for four consecutive sur deux journées consécutives.
days. The amount of dry matter (DM) of tree species
eaten by animals was recorded for two consecutive days. Leucaena s'est révélé l'espèce ligneuse la plus
appréciée. Sur une période de huit heures, les ovins et
The behavioural and intake studies showed that les caprins consacraient respectivement 155 et 160 min
leucaena was the most preferred tree species. Sheep and au broutage de cette espèce, dont ils consommaient
goats spent 155 and 160 min per eight hours respectively respectivement 40,2 ± 15 et 46,50 ± 11,32 g de MS/kg
eating leucaena and consumed 40.2 ± 15 and 46.50 ± de P parjour. L'espèce la moins recherchée tant par
1 1 .32 g DM/kgW per day. Gliricidia sepium was the les ovins que les caprins était Gliricidia sepium; le temps
tree least preferred by both animal species and time consacré à son broutage et les quantités consommées
spent on it and amount consumed was negligible. There étaient négligeables.
was no difference (P<0.05) between sheep and goats in
utilising these tree species during the behavioural study, L'étude de comportement n 'a pas fait apparaître
de différence (P<0,05) entre les caprins et les ovins pour
but there was a difference (P<0.05) between animal
ce qui est de l'utilisation de ces espèces ligneuses, mais
species during the intake study. Albizia lebbek was more
une différence (P<0,05) entre espèces animales a été
favoured by sheep (P<0.05) than by goats. Sheep and
observée dans l'étude de consommation. Albizia lebbek
goats did not differ (P>0.05) in utilising Tamarindus
était plus appréciépar les moutons (P<0,05) que par les
indica. The need for further investigation is highlighted
chèvres. Aucune différence (P>0,05) n 'a été observée
in this study.
entre les moutons et les chèvres pour ce qui est de
l'utilisation de Tamarindus indica. L'article souligne la
nécessité de poursuivre cette étude.
Choix alimentaires d'ovins et de
caprins recevant du fourrage
úPAlbizia, Gliricidia, Leucaena Introduction
et Tamarindus, espèces à The use of multipurpose trees (MPTs) has been
advocated in the tropics for several reasons. These
usages multiples include: supply of fodder to livestock; wind breaks;
providing protection and supplying nutrients to the soil
Résumé and therefore to plants; supplying fuel for cooking and
heating; acting as live fences; and providing shade to
De mai àjuillet 1991, des études de comportement both man and livestock. In Tanzania, the cut-and-carry
et de consommation ont été effectuées à l'Université system of using MPTs is common in many areas,

Small Ruminant Research Network Workshop 151

L.A. Mtenga, D.M. Komwihangilo and G.C. Kifaro

particularly Kilimanjaro, Arusha, Mbeya and Tanga. were ground to pass through a 1.25 mm screen in a
The MPTs play a special role in supplying the animals, Christy and Norris 20 cm laboratory hammer mill.
particularly sheep and goats, with protein and energy Chemical components such as dry matter (DM), Ash,
during the dry season. ether extract (ЕЕ), crude fibre (CF), calcium (Ca) and
There has been little study however of the physical phosphorus (P) were determined by standard methods
and chemical composition of the MPTs commonly used (АОAC, 1990). Nitrogen-free extract (NFE) was
by farmers. There is also limited information on the calculated by using the formula:
behavioural aspect of sheep and goats when fed several NFE = 100 - (Moisture + CP% + CF% + EE% + Ash%)
cut MPTs simultaneously. The aim of this preliminary
study was to determine the physical and chemical
composition of the cut MPTs and to assess the eating
Statistical analysis
preferences and selectivity of sheep and goats. Data on time spent by both sheep and goats on different
tree species were subjected to analysis of variance as
Materials and methods outlined by Snedecor and Cochran (1989). Parameters
considered during the analysis included means, standard
Location, animals and feed deviations and standard errors or the difference between
means. These were calculated according to the
The experiments were conducted at the Department of following model:-
Animal Science and Production, Sokoine University of
Agriculture, Tanzania between May and July, 1991. Yijk = Ц+ T + Aj + Ek + (TA)ij + (TE)ik + AEjk +
Four female mature Blackhead Persian (ВНР) (TAE)ijk + Rijki
sheep and four female mature Small East African goats where:
were used. The sheep were 1-2 years old and weighed
between 20-25 kg. The goats were 4—5 years old and д = general mean
weighed between 21-29 kg. Ti - effect of tree species
The tree species used in the study were Albizia
lebbek, Gliricidia sepium, Leucaena leucocephala and Aj = effect of animals species
Tamarindus indica Chloris gayana hay was provided as
a basal ration to the animals. Ek = effect of the day of test
(TA)ij = interaction between animal and tree
Experiment I: behavioural study species
Two sheep and two goats were confined in separate
(TE)ik = interaction between tree species and
groups in pens of size 170 x 165 cm inside the main
day of test
sheep/goat house. Branches of the tree species were tied
separately and placed at different heights for maximum (AE)jk = interaction between animal species and
exploitation of the grazing/browsing behaviour and for day of test
ease of observation during the experiment. Chloris
gayana hay and drinking water were also available ad (TAE)ijki = interaction between tree species, animal
libitum. The pens were cleaned and the previous day's species and day of test
residues removed before the feed or water was replaced.
There was a preliminary adjustment period of Rijid = residual
seven days followed by the trial period of four
consecutive days. The animals were observed from 0900 Experiment 2: intake study
hours to 1700 hours every day. The animals' behaviour
actions towards fodder were observed as described by The eight animals used in Experiment 1 were put in
Lehner (1987). The way an animal utilised the four individual pens and provided with the four tree species,
randomly placed tree species and Chloris gayana hay hay and water for a preliminary period of seven days.
was monitored. Records kept included: time in minutes Samples of the branches of trees fed to animals were
spent on each tree species; time spent eating hay; time taken before feeding and fractionated separately into
spent drinking water; time spent walking; time spent leaves (including tender branches), pods, bark and
wood. The samples were weighed when fresh, dried in
lying and ruminating; time spent standing and
an oven at 60 - 62°C for 24 hours, then weighed again.
ruminating; and time spent idling. The plant parts
The measurement period was two consecutive
browsed/grazed by the animals were also noted. At the
end of every observation day animals were transferred days for each of the tree species provided per animal.
To ensure that every animal had a continuous access to
to other pens where they were housed at night without
the feeds, every fodder tree was provided to allow for
any feed.
20% refusals. Clean water was available all the time.
Refusals were collected, fractionated and dried in
Chemical analysis an oven at 60-65° C for 24 hours for dry matter (DM)
Dried samples of favoured plant parts i.e. leaves, pods determination. The quantity of leaf materials eaten by
and bark were taken for chemical analysis. Materials an individual animal was calculated by the difference

152 Small Ruminant Research Network Workshop

 Selectivity in sheep and goats fed Albizia, etc...

between weight of leaves given (gDM) and the weight higher values of CF were obtained in this study (28.3%
of leaves refused (gDM). vs 26.5%). The variations in the chemical composition
between different parts are due to the physiological
Chemical analysis fractions of the parts such as photosynthesis in the leaves
and storage in the pods. Variations are also caused by
Chemical analysis of Albizia lebbek, Gliricidia sepium, the stage of growth of the plant parts (Brewbarker and
Leucaena leucocephala, Tamarindus indica, and Hulton, 1979).
Chloris gayana hay was as described in Experiment 1 . The chemical composition of Gliricidia leaves and
Leucaena leaves was similar to values reported by
Data processing and statistical analysis Atta-Krah and Reynolds (1989). For Gliricidia leaves,
however, slightly higher values of protein were obtained
Data on leaf, bark and wood fractions eaten or refused
when compared to the value obtained by Semenye et al
of each tree species were expressed simply as
(1986). The crude protein and crude fibre contents of
percentages of total amount of each tree plant material
offered. Leucaena leaves were within the ranges reported by
Data on the amount of leaf materials eaten were Atta-Krah and Reynolds (1989). The ash value is similar
subjected to the same ANOVA (Snedecor and Cochran, to that of 6.95% obtained by Hulman et al (1978). Le
1989) model used for experiment 1. Houérou (1980) reported average crude protein and
crude fibre contents of tamarind leaves to be 10.9% and
14.8%, respectively. Results from Tamarindus indica
for crude protein were similar to those reported by Le
Results and discussion Houérou, but in this study the crude fibre content was
The physical and proximate composition of MTPs and higher.
hay fed to the sheep and goats are presented in Tables 1 The stage of regrowth of different forages were
and 2, respectively. reported to cause differences in nutritive values
(Brewbaker and Hulton, 1979). This could account for
Table 1. Physical composition of samples of multipurpose variations in the chemical composition reported by
trees fed to sheep and goats (%). different workers. The plant materials in this study were
Parts collected from different locations and were at different
Trees species stages of growth. Variation in chemical composition of
Leaves Bark Pod Wood different parts of the same plant species could also be
Albizia lebbek 38 16 13 33 due to the functional role of the plant part and its stage
of growth in relation to the other parts.
Gliricidia sepium 40 16 - 44

Leucaena leucocephala 39 16 - 45

Tamarindus indica 22 14 36 28 Behavioural and intake studies

In the behavioural study (Tables 3, 4 and 5), sheep and
goats showed no difference (P<0.05) in utilising
Table 2. Proximate composition of leaves of the multi Gliricidia sepium and Leucaena leucocephala.
purpose trees and hay fed to sheep and goats (%). However, these animals differ (P<0.05) in the way they
utilise Albizia lebbek and Tamarindus indica. The
Tree species DM CP СF Ash behavioural study showed that sheep find Albizia lebbek
Albizia lebbek' 38.7 21.9 28.3 7.3 more palatable than do goats. Similarly, the intake study
showed that there was a difference (P<0.05) in the
Gliricidia sepium 23.7 25.2 17.9 11.1 amounts of Albizia lebbek consumed by the goats and
sheep. Goats consumed an average of 1 1 2 g DM/day of
Leucaena leucocephala 33.0 24.0 22.2 6.7 Albizia leaves, compared to sheep whose daily
consumption was 289 g DM/day.
Tamarindux indica2 46.5 11.5 20.8 9.1
The study showed also that pods of Albizia lebbek
Chloris gayana hay 90.8 6.6 44.4 8.2 were not eaten by either sheep or goats. This is contrary
' Pods of A. lebbek had 20.3% CP. to what was reported by Singh et al (1989). Palatability
2 Pods of T. indica had 8.7% CP. and selectivity are determined partly by the animal's
3 Barks of L leucocephala had 1 1.8% CP. prevailing nutritional and physiological status and partly
by the environmental circumstances. Environmental
circumstances include the relative availability of various
The chemical composition ofAlbizia lebbek leaves parts from which the choice is made (French, 1970).
(Table 2) compared well with values reported by NAS These factors may contribute to the differences observed
(1971) and FAO (1981) in terms of CP but slightly in this study.

Small Ruminant Research Network Workshop 153

L.A. Mtenga, D.M. Komwihangilo and G.C. Kifaro

Table 3. Time in minutes spent by sheep and goats on Gliricidia. Chadhokar and Kantharaju (1980), however,
eating the multipurpose trees and hay on free
reported that Gliricidia maculata was found palatable
choice basis for eight hours.
by both cattle and sheep even when fed in very large
Feed Sheep Goat
20.5a 5.1b
quantities over a long period.
A. lebbek Although the difference in intake by sheep and
(5.67) (1.47)
goats of wilted and fresh Gliricidia has not been well
G. sepium 0.3 1.3 documented, wilting is believed to increase the
(0.63) (0.18) acceptability of the forage leaves.
L. leucocephala 155.1 160.1 The ranking of tree species eaten is shown in
(42.15) (45.90) Table 5. Semenye et al (1986) ranked Gliricidia last and
53.7b 100.3a
T. indica Leucaena leucocephala as number one among the six
(14.59) (28.78) protein feed resources studied in their experiment.
138.5a 81.8b
С. gayana hay Meulen et al (1979) reported the presence of secondary
(36.91) (23.7) plant metabolites such as beta - carotene and
Figures in brackets represent percentage time. xanthophyll in Leucaena. These may make Leucaena
ab = Values within rows bearing different superscripts are significantly leucocephala the most preferred of the four tree species.
different (P<O05).
The results of this study are contrary to the report
by Wilson (1957) that goats are not very selective
Table 4. Time in minutes spent by sheep and goats on
regarding plant species. Selectivity by goats and sheep
different activities during the eight hour study was shown to be related to the species of the plant
period. materials offered, especially in the cut-and-carry system
Activity Sheep Goats (Semenye, 1991). Knowledge of the palatability of
plants fed to sheep or goats may be used as a tool to
MPTs 229.4 267.1
increase dry matter intake (DMI) of animals by offering
Eating hay 138.5 82.0 the least palatable first and the most palatable last
Ruminating 68.3 33.8 (Semenye, 1991). Palatability ranking could have been
Drinking water 31.2 85.2 clearer if more forage tree species had been included.
Walking 1.3 1.4
Idling 11.1 9.7

This study shows that the animals did not like Conclusion
Gliricidia sepium. Both species spent very little time
eating this plant. Studies in India also showed that cattle The higher crude protein and lower crude fibre contents
did not like this tree species. The intake study showed in leaves and pods of MPTs suggests that they may be
that goats ate only 45 g DM/day (4.6 g DM/W0'75). used as supplements to low crude protein feeds such as
However, Onwuka (1986) reported an intake of 176.83 Chloris gayana hay particularly during the dry season.
g/day and Mba et al ( 1 982) reported an intake of between However, acceptability of the different fractions of
233.80 to 294 g/day by goats. Onwuka (1986) used dry MPTs differ between sheep and goats. The chemical
Gliricidia leaves to feed his goats while Mba et al (1982) composition of a plant or plant part may not determine
used fresh materials. In both studies periods of up to six its acceptability to the animal. Techniques of sampling
months were used. It is probable that longer periods of and measuring yield and nutritive value of edible parts
exposure are required before the animals can accept (leaves, bark, pods) of MPTs also need to be developed.

Table 5. Palatability ranking and dry matter intake (g DM/day) by sheep and goats.
Sheep Goats
Feed Intake Rank Intake Rank
/.. leucocephala 432 ±16 1 453 ± 25 1
(40.2 ± 15.6) (46.5 ±11.3)
T. indica 229 ± 75 3 251 ± 87 2
(21.3 ±6.9) (25.8 ±14.1)
A. lebbek 289 ±99 2 112±60 3
(26.9 ± 90) (11.4 ±6.2)
G. sepium 18 ±4.0 4 45 ±22 4
(1.7 ± 1.0) (4.6 ±2.2)
Figures in brackets show intake in g/kgW075

154 Small Ruminant Research Network Workshop

 Selectivity in sheep and goats fed Albizia, etc...

Mba A V, Mangui S A and Awah A A. 1982. The influence of

References concentrated supplementation with browse plant (G.
AOAC (Association of Official Analytical Chemists). 1990. sepium) on nutrient utilization and growth of West
Official methods ofanalysis ofthe Association ofOfficial African Dwarf (Fonta Djalon) kids in Niger. Journal of
Analytical Chemists. 15th edition. AOAC, Arlington, Animal Production 9(2):63-73.
Virginia, USA. 2 Volumes.
Meulen U S, Struck E A and Harith E G 1979. A review of
Atta-Krah A N and Reynolds L. 1989. Utilisation of pasture nutritive value and toxic aspects of Leucaena leuco
and fodder shrubs in the nutrition of sheep and goats in cephala. Tropical Animal Production 4: 1 1 3-1 26.
the humid tropics of West Africa. In: Timon V M and
Baber R P (eds), Sheep and goat meat production in the NAS (National Academy of Science). 1971. Leucaena: a
humid tropics of West Africa. FAO Animal Production promising forage and tree crop for the tropics. NAS,
and Health paper 70:68-91. FAO (Food and Agriculture Washington DC, USA.
Organization of the United Nations), Rome, Italy.
Onwuka C F I. 1986. Gliricidia sepium as dry season feed for
Brewbaker J L and Hulton E M. 1979. Leucaena. In: Ritchie goat production in Nigeria. In: Haque I, Jutzi S and Neate
G A (ed), New agricultural crops. AAAS (American P J H (eds), Potentials offorage legumes in farming
Association for the Advancement of Science) Selected systems of sub-Saharan Africa. Proceedings of a
Symposium 38. Westview Press, Boulder, Colorado, workshop held at ILCA, Addis Ababa, Ethiopia, 16-19
USA. pp. 207-259. September 1985. ILCA (International Livestock Centre
Chadhokar P A and Kantharaju H R. 1 980. Effect of Gliricidia for Africa), Addis Ababa, Ethiopia. pp. 533-539.
macúlala on growth of Bannur ewes. Tropical
Semenye P P. 1991. Nutrition and management for dual-
Grasslands 14:78-79.
purpose goats: Research highlights. Winrock Inter
FAO (Food and Agriculture Organization of the United national Institute for Agricultural Development,
Nations). 1981. Tropical feeds. Feed information Morrilton, Arkansas, USA. 53 pp.
summaries and nutrition values. FAO Animal Pro
duction and Health Series 12. FAO, Rome, Italy. 529 pp. Semenye P P, Musalia L, Khainga S and Fitzhugh H A. 1986.
Small ruminants nutrition in small farms in Kenya. In:
French M H. 1970. Observations on the goat. FAO Agri Adeniji K O and Kategile J A (eds), Proceedings of the
cultural Studies 80. FAO (Food and Agriculture Organ Workshop on the Improvement of Small Ruminants in
ization of the United Nations), Rome, Italy. 204 pp. Eastern and Southern Africa. OAU/IDRC (Organization
Hulman В E, Owen E and Preston T R. 1978. Comparison of of African Unityflnternational Development Research
Leucaena leucocephala and groundnut cake as protein Centre), Nairobi, Kenya. pp. 127-133.
sources for beef cattle fed ad libitum molasses. Tropical
Singh N, Sharma K and Ogra J L. 1 989. Chemical composition
Animal Production 3:1-8.
and nutritive value of siris (Albizia lebbek) and sababul
Le Houérou H N. 1980. Chemical composition and nutritive (Leucaena leucocephala) pods in goats. Indian Journal
value of browse in tropical West Africa. In: Le Houérou ofAnimal Nutrition 6(3):259-261.
H N (ed), Browse in Africa - the current state of
knowledge. Proceedings of an International Symposium Snedecor G W and Cochran W G. 1989. Statistical methods.
on Browse in Africa, Addis Ababa, 8-12 April 1980. 8th edition. Iowa State University Press, Ames, Iowa,
ILCA (International Livestock Centre for Africa), Addis USA. pp. 297-329.
Ababa, Ethiopia. pp. 261-289.
Wilson P N. 1957. Studies of the browsing and reproductive
Lehner P N. 1987. Design and execution of animal behaviour behaviour of the East African Dwarf goat. East African
research - An overview. Journal of Animal Science 65: Agricultural Journal 23:138-147.
1212-1219.

Small Ruminant Research Network Workshop 155

 Influence of amount of feed offered on growth, intake and
selectivity: observations on sheep and goats
A.A.O. Aboud1, E. Owen2, J.D. Reed3 and A.N. Said4

Department ofAnimal Science, Sokoine University ofAgriculture, P.O. Box 3004, Morogoro, Tanzania
Department ofAgriculture, University ofReading, Early Gate, RG 6 2AT, Reading, UK
University of Wisconsin, Department ofAnimal Science, USA
International Livestock Centre for Africa, P.O. Box 5689, Addis Ababa, Ethiopia

Abstract La consommation de paille augmentait (P<0,00I)

avec l'augmentation du niveau de rationnement (22, 31
Comparisons were made between 24 Ethiopian rams et 32 g de MS/kg de poids vif/j chez le mouton et 19, 26
and 24 Galla goats for voluntary intake, feed selectivity et 29 g de MS/kg de poids vif/j chez la chèvre). Toutefois,
and growth performance when offered three levels of aucune augmentation substantielle de l'ingestion n'a
sorghum stover. The levels observed were 25, 50 and 75 été obtenue lorsque le niveau de rationnement était
g/kg live weight per day. Measurements of intake, amené à plus de 50 g/kg. Il n 'y avait pas de différence
growth and feed selectivity were taken over a period of (P>0, 05) entre les choix alimentaires des chèvres et des
10 weeks. moutons. Les deux espèces consommaient de préférence
Both sheep and goats consumed more stover les feuilles et les gaines foliaires dans la mesure où les
(P<0.001) as rate of offer was increased (22, 31, 32 apports alimentaires augmentaient.
gDM/kg live weight per day; 19, 26, 29 g/kg live weight
Les moutons réalisaient de meilleurs (P<0,001 )
per day, respectively, for sheep and goats). Increasing
gains de poids que les chèvres, quels qu 'aient été les
the level of offer beyond 50g/kg live weight per day did
niveaux de rationnement (28,0 contre 9,0, 54,0 contre
not, however, promote a substantial increase in intake.
23,0, et 62,0 contre 31,6, au bénéfice des ovins). Les
No difference (P>0.05) was observed in the pattern of
gains de poids étaient plus rapides chez les animaux
feed selectivity between sheep and goats. Both species
soumis à un niveau d'alimentation élevé que chez ceux
showed a similar preference for leaf and sheath matter
soumis à un niveau d'alimentation modéré.
as the level of offer was increased.
Sheep gained weight faster (P<0.00 1 ) than goats at En conclusion, il apparaît que les deux espèces
all rates of offer, (28.0 vs 9.0; 54.0 vs 23.0; 62.0 vs 31.6 sont susceptibles de tirer parti d'un apport alimentaire
in favour of sheep). Those on a higher level of offer généreux et, qu'à en juger d'après l'amélioration de la
showed a faster rate of gain than those on a lower level croissance et de l'ingestion, les avantages retirés
of offer. s 'expliquent par la sélection des aliments.
It was concluded that both species can benefit from
generous offers and that the benefit, judged by improved
intake and growth, is due to selective feeding.
Introduction
Previous studies (Aboud et al, 1992) clearly indicate the
Effet de l'offre alimentaire sur value of selective feeding. Sheep were shown to perform
la croissance, la consommation better and consume more digestible matter through
selective intake of leaf and sheath fractions of sorghum
et les choix alimentaires chez les stover as a result of liberal offers. This observation is of
practical value for small ruminants in the tropics because
ovins et les caprins the traditional feeding systems depend largely on crop
residues. However, there is little evidence in the
Résumé literature that the approach adopted by Aboud et al
(1992) would be as effective for goats as it was for sheep.
L 'ingestion volontaire, les choix alimentaires et les Results from comparative studies between sheep and
performances de croissance ont été comparés chez 24 goats of selective efficiency are generally conflicting
béliers éthiopiens et 24 chèvres Galla auxquels de la (Huston, 1978; Devendra and Burns, 1983). Goats in the
paille de sorgho avait été servie selon trois niveaux de tropics are more able than sheep to consume feed
rationnement (25, 50 et 75 g/kg de poids vif/j). Les fractions of higher nutritional quality when offered the
mesures de l'ingestion, de la croissance et de la opportunity to eat selectively (Devendra and Burns,
sélectivité à l'égard des aliments ont été réalisées sur 1983; Hoppe et al, 1977; Huston, 1978). This
une période de 10 semaines. observation does not agree with studies carried out in

Small Ruminant Research Network Workshop 157

A.A.O. Aboud, E. Owen, J.D. Reed and A.N. Said

temperate countries, where sheep have been shown to about 1700 metres above sea level, 45 km south-east of
perform better than goats under most practical feeding Addis Ababa, Ethiopia.
systems. A comparative feeding and performance study A total of 48 animals were used in this study. Half
on Small East African goats and the Red Maasai sheep were Ethiopian lowlands hair-type rams and half were
in Tanzania (Shoo, 1986) was inconclusive, although Galla goats. The animals were between 1 1-13 months
sheep appeared to perform better than goats when old and weighed between 14-16 kg. They were divided
offered a basal diet of Chloris gayana hay with leucaena into three groups of 16 animals, eight rams and eight
as a supplementary forage. However, observations in goats in each group. The three groups were allocated to
Shoo' s (1986) study did not include an assessment of three different levels of offer, which were 25, 50 and 75
selective feeding ability between the two species. The g stover DM/kg ' daily.
inclusion of a supplementary forage may have
Sorghum stover (cv. Seredo: a bird resistant
confounded the results, as it is known that goats would
variety) was used as the basal diet. In addition, the
preferentially consume browse species (Lu, 1 988). Most
animals also received 150 g/day of solvent extracted
of the comparative studies in the tropics were made
under grazing conditions, whereas those in the cottonseed cake (Table 1) and mineral lick (Table 2).
temperate countries involved comparisons under The animals were fed twice a day. On each occasion half
stall-feeding conditions in which feed was offered at of the daily ration was provided after collection of
restricted levels, usually at 35 g dry matter M(D)/kg live refusals. The experimental period lasted 10 weeks after
weight per day or to achieve 10-20% rate of refusal. an initial adaptation period of 21 days.

Table 1. Composition offeedstuff's utilised during the adaptation and trial periods (g/kg).
Cottonseed cake
Composition* Chloris gayana Wheat bran Seredo stover Sample 1 Sample 2
DM 870 876 913 923 932
CP 37 169 46 427 286
CF 37 113 - - -
Ash 87 64 9.4 67 70
NFE 444 616 - - -
NDF - - 697 776 452
ADF - - 452 293 376
Lignin - - 444 88 92
* DM = dry matter; CP = crude protein; CF = crude fibre; NFE = nitrogen-free extract;
NDF = neutral detergent fibre; ADF = acid detergent fibre.

Comparative studies under grazing or in stalls at Table 2. Composition of mineral block.

restricted levels of offer may give misleading results
(Demment and Van Soest, 1983). Restricted levels of DM (g/kg) 958
offer do not usually provide sufficient opportunity for Phosphorus 43
selective feeding (Aboud et al, 1993). This is
particularly serious where low-quality crop residues are Potassium 11
used in comparisons. Calcium 71
No performance comparison has been made
Magnesium 4
between sheep and goats that have been fed untreated
crop residues at levels that would allow selective intake. Sulphur 7
Alimon (1989) only measured the digestible organic Iron Mg/kg 74.89
matter (OM) intake of goats offered increasing levels of
feed, but did not measure any related growth Manganese 79.31
performance. Aboud et al (1993) showed that certain Copper 197.40
levels of offer encouraged sheep to eat selectively. By
offering goats and sheep the same type of sorghum Zinc 148.77
stover at these levels both species may perform
similarly.
Sampling and measurement of feed intake

Material and methods Before each meal, the refusals from previous meals were
removed from troughs and weighed. The difference
The experiment was conducted at the Debre Zeit between amount offered and amount refused was taken
Experimental Station, an ILCA sub-station located as the quantity consumed.

158 Small Ruminant Research Network Workshop

 Influence of amount of feed offered on growth, etc..

For each animal, a handful of the stover refused digestibility (DMD) were done in duplicate on all
during the morning and evening feedings was collected samples collected. The methods of Goering and Van
to make a single daily sample. Each daily sample was Soest (1975) were used for fibre analysis and for the in
added to the previous day's sample for one week. At the vitro digestibility. Results from the in vitro digestion
end of each week, the combined sample was divided into were expressed as the residual amount of NDF
two parts, one for chemical analysis and the other for remaining after 48 hours incubation. N was determined
botanical fractionation. The parts retained for chemical using the macro-Kjeldahl procedure.
analysis from each individual were then mixed by
treatment group and sub-sampled to form a single Results
sample per week from each treatment group. Altogether
10 such samples were collected from each treatment Intake of stover
group during the experiment.
Botanical fractionation was based on the weekly Table 3 shows the main effect means for intake of stover
pooled samples for each animal. Proportions of by sheep and goats. Analysis of variance showed no
botanical components (leaf, sheath and stem) were significant interaction between the main effects (species
expressed in g/kg of sample and averaged for each and level of offer) on stover intake. Effects of species
treatment group per week. This provided 10 such and level of offer were highly significant (P<0.001).
averaged values for the experimental period, one per Sheep consumed significantly (P<0.001) more stover
week based on means from eight animals per treatment. than goats. Both species improved their DM intake as
Botanical fractions were partitioned using the traditional the level of offer was raised from 25 to 75 g/kg live
'winnowing' process. This involved shaking the sample weight per day. The improvement was particularly large
in a gentle circular motion using a 'safed', a reed woven between animals offered 25 and 50 g/kg live weight per
mat about 1 m across. This trapped the heavier material day. Increasing the offer to 75 g/kg live weight per day
(stem and sheath) in grooves while the lighter material resulted in a small increase in intake.
(leaf) was shaken towards the edge. The fractions were
then picked by hand. Rate of refusals and selectivity
Effects of level of offer and of differences between the
Growth performance and measurements species were highly significant (P<0.001). Figure 1
shows a comparison between the species and effect of
Once a week the animals were weighed, on each level of offer on the rate of stover refusal. Rate of refusal
occasion at 0700 hours. Prior to weighing, water and was expressed as rate of utilisation (1 - rate of refusal)
feed were withdrawn for at least six hours. A canvas to show the fall in the proportion of stover utilised as the
sling hung on a spring balance was used for weighing. amount offered was raised. Goats have lower utilisation
values because of their higher rates of refusals.
Chemical analyses Proportionately more stover was refused as the level of
offer was raised.
Chemical analyses for dry matter (DM), acid detergent Summaries of analysis of variance for the
fibre (ADF), neutral detergent fibre (NDF), nitrogen comparison of the stover leaf, sheath and stem
(N), ash, silica, ADF-ash and in vitro dry matter proportions (g/kg) in the refusals are given in Table 4.

Table 3. Main effects means for stover intake, rate of stover refusal and growth rate (g/d).
Level of offer
Species (g/kg live weight per day)
Goats Sheep SE 25 50 75 SE Species Level Species x Level
No. animals 24 24 - 16 16 16 - - -
Initial live weight (kg) 14.5 17.0 - - - - - - - -
Stover
Offered 731 752 - 352 757 1143 - - - -
Refused (gDM/kgDM) 384 298 9.9 101 373 549 12.1 *** *** NS
DM intake
(g/d) 404 456 13.2 316 468 506 16.2 *** *** NS

(g/kg per day) 25 29 0.4 21 29 31 0.5 *** *** NS

(g/kg075) 50 57 1.0 41 58 62 1.3 *** *** NS

Rate of gain (g/d) 28 53 4.4 22 46 52 5.6 *** *** NS

NS = not significant.
*** = significant (P<0.001).

Small Ruminant Research Network Workshop 159

A.A.O. Aboud, E. Owen, J.D. Reed and A.N. Said

Figure 1. Effect of level ofoffer on stover intake and rate of Figure 2. Effect oflevel ofoffer on botanical composition of
utilisation by sheep and goats. DMI by sheep and goats.
DMI (g/d) and rate of utilisation Intake of components (gDM/kgDMI)
1,000 1,000
DM I (g/d) goat
с DMI (g/d) »heap
Utlllsatlon/goat ; ] Stem (g DM/kg DMI.d)
BOO 800 -
Utlllsatlon/sheep
[7] Sheath (g DM/kg DMI.d)
[J Leaf (gDM/kgDMI.d)
600
600

400
400

200
200

n
\г
25 50 75
25g/kg 50g/kg 75g/kg
Level of offer (g/kgM.d)
Laval of offer (g/kgM.d)

For leaf matter proportions, no significant interaction Discussion

between main effects was observed. Differences
between species and effects of level of offer were both
significant (P<0.01). The stover refused by goats had Effect of level of offer on stover intake
more leaf (P<0.05) than that left by sheep. In both and stover selectivity
species stover refused from higher rates of offer
contained more (P<0.001) leaf than that refused at the The results show that sheep and goats respond in almost
25 g/kg live weight per day level. the same manner when offered increasing quantities of
Significant interactions (P<0.001) between main sorghum stover.
effects were observed in the proportion of sheath and Similarly to previous findings (Owen and Aboud,
stem in the stover refused. Effect of level of offer and 1 988), the response to increasing level of offer appeared
differences between species were highly significant to diminish in both species as the level of daily offer was
raised beyond 50 g/kg live weight per day. It also
(P<0.001). In goats, the proportion of sheath and stem
appears that both species optimised their DMI and
in refused stover increased with increasing level of offer. growth performance at the same level of offer. The
Sheep had no consistent pattern. Table 5 shows the main efficiency with which this was achieved was different
effects means of intakes of the three botanical fractions between the species. Sheep showed a better utilisation
(leaf, sheath and stem). Sheep and goats consumed about rate than goats (P<0.01). The DMI achieved by both
the same amount of leaf and sheath fractions. Sheep had species at the 50g/kg live weight per day level of offer
a considerably higher intake of stem fractions than was close to the levels reported by Devendra and Burns
goats. More leaf and sheath were consumed by both (1983) for Baribari goats in India where the animals
species as the level of offer was raised (Figure 2). The were fed forages ad libitum.
amount of stem consumed, however, decreased with Most reports in the literature suggested that goats
increasing levels of offer, although differences in intake are more able than sheep to eat selectively (Devendra
at the higher levels were about the same. and Burns, 1983; Lu, 1988). These observations are
largely based on grazing animals, most of them on
natural rangelands (Lu, 1988). The findings in this study
Rate of growth with sorghum stover showed that in general both species
are able to eat selectively. The results also suggest that
Sheep gained nearly twice as much as goats (Table 3). the differences in selection between the sheep and goats
Increasing the level of offer significantly (P<0.00l) used were dependent on the level of feed on offer. The
improved the rate of gain in both species, particularly botanical and chemical composition of the stover
when the offer was raised from 25 g to 50g/kg live refused showed that sheep tended not to select when
weight per day. offered stover at 25g/kg live weight per day, whereas

160 Small Ruminant Research Network Workshop

 Influence of amount of feed offered on growth, etc..

Table 4. Analysis of variance for comparison of stover refusals by sheep and goals.
Dependent variable: leaf fraction (g/kg):
Source DF Type 1 SS F Value PR>F DF Type III SS F Value PR>F
Species (S) 1 21.59801198 5.21 0.0264 1 21.59801198 5.21 0.0264
Level of offer (L) 2 47.29336599 5.71 0.0056 2 47.29336599 5.71 0.0056
SxL 2 17.21797981 2.08 0.1351 2 17.21797981 2.08 0.1351
Dependent variable: sheath fraction (g/kg):
Source DF Type I SS F Value PR>F DF Type III SS F Value PR>F
Species (S) 1 7987.878489 14.24 0.0004 1 7987.87848945 14.24 0.0004
Level of offer (L) 2 66984.34385595 59.71 0.0001 2 66984.34385595 59.71 0.0001
SxL 2 10367.34709726 9.24 0.0004 2 10367.34709726 9.24 0.0004
Dependent variable: stem fraction (g/kg):
Source DF Type I SS F Value PR>F DF Type III SS F Value PR>F
Species (S) 1 18633.39211586 13.11 0.0006 1 18633.39211586 13.11 0.0006
Level of offer (L) 2 57777.98010960 20.33 0.0001 2 57777.98010960 20.33 0.0001
SxL 2 22624.55784522 7.96 0.0009 2 22624.55784522 7.96 0.0009

Table 5. Main effects means for intake of stover fractions by sheep and goats.
Species Level of offer (g/kg live weight per day)
Goats Sheep 25 50 75
Stover offered (g/d):
Total amount 731 752 352 757 1143
leaf 61 63 29 63 95
sheath 173 178 83 179 271
stem 451 463 217 467 704
Stover refused (g/d):
Total amount 281 224 36 282 627
leaf 0.8 0.3 0.04 0.5 2.1
sheath 21.0 22.2 2.1 18.6 82.8
stem 237 197 31 255 519
Intake of components (g/d):
leaf 60 63 29 63 93
sheath 152 156 81 161 188
stem 214 266 187 212 185
Component proportion (g/kg
stover intake):
leaf 133 112 9 132 180
sheath 338 295 256 338 364
stem 475 503 591 446 358

goats were selective at all levels of offer. At 50g/kg live Effect of level of offer on growth
weight per day and 75g/kg live weight per day there performance
were no substantial differences between the two species
in selective feeding, both in ability or efficiency. The For both sheep and goats, growth performance improved
pattern of intake of botanical components was similar in significantly as the rate of offer was increased. The
the two species. Both showed a preference for leaf and improvement in performance was remarkable when the
sheath against stem. Figure 2 shows that the intake of rate of offer was raised from 25g/kg live weight daily to
leaf and sheath was increasing even at the highest level 50g/kg live weight daily. Feed conversion efficiency
of offer. The two components were constituting an also improved, particularly in sheep.
increasing proportion of the DMI as the level of offer The difference in rate of gain between the two
was increased. species cannot be explained by differences in their

Small Ruminant Research Network Workshop 161

A.A.O. Aboud, E. Owen, J.D. Reed and A.N. Said

ability to eat selectively. Because both species were September 1991. BSAP Occasional Publication 16.
compared for their absolute gain rather than their BSAP (British Society of Animal Production),
relative gain, this may have favoured sheep due to their Edinburgh, UK. pp. 202-203.
higher initial weight. It would have been preferable to Alimón A R. 1989. An examination of strategies to maximize
express the rate of gain as a proportion of the expected dietary selectionfor improving the intake and utilization
mature weight. of barley straw by goats. PhD thesis, University of
Reading, Reading, UK. 348 pp.
Conclusion Demment M W and Van Soest P J. 1983. Body size, digestive
capacity andfeeding strategies of herbivores. Winrock
The findings in this study clearly support earlier International Institute of Livestock Research and
observations that generous offers of feed can promote Training Center, Morrilton, Arkansas, USA. 66 pp.
intake both qualitatively and quantitatively. This Devendra C and Bums M. 1983. Goat production in the
possibility should offer opportunities for modifying tropics. CAB (Commonwealth Agricultural Bureaux),
current practices of supplementary feeding. Where Farnham House, Slough, UK. 1 83 pp.
restricted feeding is practised there is always the
likelihood of underestimating the potential nutritive Hoppe P P, Qvortrup S A and Woodford M H. 1977. Rumen
fermentation and food selection in East African sheep,
quality of forages, especially crop residues.
goats, Thompson gazelle. Grant gazelle and Impala.
Journal ofAgricultural Science 89: 129-135.
Huston J E. 1978. Forage utilization and nutrient requirements
References of the goat. Journal of Dairy Science 6 1 :988-993.

Aboud A A O, Owen E, Reed J D and Said A N. 1992. Lu С D. 1988. Grazing behaviour and diet selection of goats.
Influence of stover variety and level of offer on voluntary Small Ruminant Research 1:205-216.
intake, feed selectivity and growth performance of Owen E and Aboud A A 0. 1 988. Practical problems of feeding
Ethiopian rams: Perspectives on practical problems. A crop residues. In: Reed J D, Capper В S and Neate P J H
paper presented at the 19 Scientific Conference of (eds), Plant breeding and the nutritive value of crop
Tanzania Society ofAnimal Production. residues. Proceedings of a Workshop held at ILCA,
Aboud A A O, Owen E, Reed J D, Said A N and McAIlan A Addis Ababa, Ethiopia, 7-10 December 1987. ILCA
B. 1993. Feeding sorghum stover to Ethiopian goats and (International Livestock Centre for Africa), Addis
Ababa, Ethiopia. pp. 133-155.
sheep: effect of amount offered on intake, selection and
performance. In: Gill M, Owen E, Pollott G E and Shoo R A. 1986. Effect of supplementation with leucaena to
Lawrence T L J (eds), Animal production in developing sheep and goats on intake and growth performance. MSc
countries. Proceedings of a symposium organized by Dissertation, Sokoine University of Agriculture,
The British Society of Animal Production, held at Wye, Morogoro, Tanzania.

162 Small Ruminant Research Network Workshop

 The effect of supplementation of cassava peel (CP) diets
with graded levels of palm kernel cake (PKC) on the
performance of growing Djallonké sheep
A.K. Tuah1, E.R. 0rskov2, FY. Obese1,
D.B. Okai1 and J. F.D. Greenhalgh3

Department ofAnimal Science, Faculty ofAgriculture

University ofScience and Technology, Kumasi, Ghana
Rowett Research Institute, Aberdeen, Scotland, U.K.
3Gilfach, Rowen, Conwy, Gwynedd, LL32 8TS, U.K.

Abstract animaux n'avaient pas d'effet significatif (P>0,05) sur

la consommation journalière moyenne des pelures de
Djallonké sheep were fed an ad libitum basal diet of manioc ou des pelures de manioc complétées par du
cassava peel (CP) supplemented with graded levels (0, tourteau de palmiste, ni sur la quantité moyenne d'eau
50, 1 00, 1 50 and 200 g/head per day) of palm kernel cake ingérée par jour. La consommation du tourteau de
(PKC). This study reports the effects of this dietary palmiste augmentait (P<0,01) avec l'accroissement du
supplementation on feed intake, bodyweight changes, niveau de complémentation mais demeuraitfaible d 'une
water intake and mortality. The supplementary levels manière générale, probablement en raison de sa
corresponded to dietary treatments 1-5, respectively. mauvaise qualité, elle-même imputable à une
The mortality rates of the animals were 20, 20, 40, 50 préparation défectueuse. Une autre étude a porté sur
and 0% for diets 1, 2, 3, 4 and 5, respectively. Mean daily l'examen des caractéristiques de la dégradation de ces
intakes of CP, CP and PKC supplements and water were deux aliments au moyen de la technique des sachets en
not significantly (P>0.05) affected by dietary treatments nylon. La portion insoluble maisfaiblement dégradable
or sex. There was an increase (P<0.01) in the intake of du tourteau de palmiste était peu importante (près de
PKC as the level of supplementation increased. 16%).
Generally the intake of PKC was low. This might be
attributable to poor quality resulting from its
preparation. In another study the degradation
characteristics of the two feedstuffs were studied using Introduction
the nylon bag technique. The insoluble but slowly
degradable portion of the PKC was low (about 16%). It is the policy of the Government of Ghana to increase
the production of small ruminants to meet the animal
protein requirements of the population. The Ministry of
Agriculture (1988) envisaged an increase in the
Effet de la complémentation de population of goats from about 1.9 million in 1987 to
rations à base de pelures de about 9.2 million by the year 2000; the population of
sheep was expected to increase from about 2 million in
manioc avec du tourteau de 1987 to about 6.4 million by the year 2000. The main
palmiste sur les performances producers of small ruminants in Ghana are small-scale
rural farmers. It is mainly these farmers who will be
du mouton Djallonké expected to increase their production. They, however,
have serious problems feeding their animals. Increased
Résumé pressure on the land has forced people to farm near their
villages within walking distances of animals. During the
Cette étude présente les effets de rations à base de cropping season animals are therefore kept in pens for
pelures de manioc complémentées par du tourteau de most of the day. They are allowed to graze later in the
palmiste (0, 50, 100, 150 et 200 g/tête/jour) sur afternoon, when they are less likely to travel outside the
l'ingestion, l'évolution pondérale, la quantité d'eau villages and damage the crops.
ingérée et la mortalité de moutons Djallonké nourris ad In the dry season the quality of the sparsely
libitum. Les niveaux de complémentation scattered forages is also poor, especially in the northern
correspondaient respectivement aux traitements savanna zones of the country. The rural farmers, unlike
alimentaires 1, 2, 3, 4 et 5. La mortalité des animaux the urban backyard farmers, are not able to purchase
était de 20, 20, 40, 50 et 0% pour les traitements 1, 2. 3, completely mixed diets or feed ingredients to feed their
4 et 5, respectivement. Les traitements et le sexe des animals. They have to rely on crop residues generated

Small Ruminant Research Network Workshop 163

AK. Tuah et al

from their own farms and also foliage of multi-purpose allowed two weeks to adjust to their respective diets
trees, shrubs and grasses which they harvest from their before data began to be collected. Feed and water intake
farms or roadsides. The commonest crop residue fed to and bodyweight changes were monitored. Leftover feed
animals by farmers in both the south and north of the and water were measured every morning before fresh
country is cassava peel (CP), although peels of plantain, feed and water were provided. Samples of feed and oats
yam and cocoyam are also used by farmers in the south. or leftovers were taken once a week and kept for
About 3,327,000 metric tons of cassava is analysis. The animals were weighed every two weeks
produced in Ghana per year (FAO, 1990) and with an after 18 hours without food throughout the 12-week
extraction rate of about 20% (Kossila, 1984), about study.
665,400 metric tons of CP are produced in the country
in a year. CP, however, has low nitrogen (Tuah, 1990) Experiment 2
and mineral (Adegbola et al, 1988) contents. The
objective of this study was therefore to investigate the In sacco degradation of CP and PKC
effects of supplementing CP diets with graded levels of
PKC as a protein source on the performance of young This study was undertaken at the Rowett Institute,
growing Djallonké sheep. PKC is available in most Aberdeen, Scotland. Samples of CP and PKC weighing
about 3 g each were put in nylon bags. They were then
villages in the southern forest zone of the country but is
incubated in the rumens of three fistulated wethers to
not used at present as animal feed. measure degradability of dry matter at various
incubation periods. The wethers had a mean body
weight of 58 kg. The incubation periods were 4, 8, 16,
Materials and methods 24, 48, 72 and 96 hours. The type of bags and the
incubation procedures were the same as described by
Tuah et al (1986). The sheep were fed 700 g/head per
Experiment 1 day at 0900 h and 1600 h of a diet containing (per kg):
500 g hay; 300 g barley; 100 g molasses; 91 g fishmeal;
Supplements 5 g dicalcium phosphate; 3 g NaCl; and 1 g
mineral-vitamin premix. The mineral-vitamin premix
The PKC used in the trial was obtained from a palm oil contained (per kg): 185 g calcium; 104 g magnesium;
production factory at Ahodwo, a suburb of Kumasi, near 2250 mg cobalt; 44,000 mg manganese; 36,400 mg zinc;
the University of Science and Technology, Kumasi, 1300 mg iodine; 100 mg selenium; 10,000,000 iu
Ghana. To extract the oil from the kernels, they were retinol; 2,000,000 iu vitamin D3; and 40,000 iu
first dried, then roasted and ground through a alphatocopherol.
hammermill (about 1mm sieve). The ground materials CP and PKC were ground in a 3 mm sieve and
were boiled with water and the supernatant oil skimmed
triplicate samples of each were soaked in warm water
off. The residue kernel cake was dried in the sun.
(about 39-39°) for one hour. The samples were
The CP were obtained from a "garri" processing
subsequently washed in a washing machine as was done
factory in the city of Kumasi. They were peels of mixed
for the CP and PKC samples incubated in the rumens of
varieties of cassava. The peels sometimes had some
the sheep. The washing losses were determined after
cassava tuber pieces in their linings. They were sun-
drying the samples in an oven at about 70°C for 48 hours.
dried before being fed to the animals.
The equation of 0rskov and McDonald (1979), p = a +
b( 1 -ec ), was used to describe the course of degradation
Animals, housing, management and feeding of the feeds.
A growth rate study of twenty three young Djallonké
sheep of both sexes was carried out at the Department Calculation of energy intake, maintenance energy
of Animal Science, University of Science and and rumen degradable nitrogen requirements
Technology, Kumasi, Ghana. The animals had a mean
live weight of approximately 12 kg and were dipped and It was assumed that the PKC supplied negligible
dewormed at the start of the experiment. They were amounts of metabolisable energy (ME). Thus the
weighed after 18 hours without food. The animals were calculation of ME intakes was based on the energy from
kept in individual pens measuring 1 x 3 m on wooden the CP only. It was also assumed that all the nitrogen in
slatted floors in a building described by Tuah et al the CP was degradable in the rumen and that nitrogen in
(1985). They were randomly allocated to five dietary PKC was only 16% degradable in the rumen.
treatments. Treatment groups were balanced for sex of The fractional outflow rate of the particles of CP
animals. Dietary treatments 1-5 corresponded to five was assumed to be 0.04/h. The apparent digestibility
graded levels of PKC supplements to the basal CP diet coefficient of dry matter of the CP was then estimated
(0, 50, 100, 150 and 200 g/head per day). The animals to be 0.58, based on the in sacco degradation
were fed daily at 0800 from wooden troughs described characteristics of the peels. The ME content of the CP
by Tuah et al (1985). The troughs for those animals was estimated using the formula MJ ME/kg dry matter
which received the PKC supplements were partitioned = 0. 1 56 x DOM, given by Roy et al ( 1 977), where DOM
into two compartments with pieces of wood. Water was is the digestive organic matter per kg of dry matter. The
offered ad libitum at 0800 daily. The animals were estimate of the ME content of CP (9 MJ ME/kg dry

164 Small Ruminant Research Network Workshop

 The effect of supplementation of cassava peel (CP) etc..

matter) may have been overestimated because instead of 7 and 6.5 g/kg Vr' per day for males and females,
using the in vivo DOM content of dry matter, digestible respectively. The mean daily water intakes were 269 and
dry matter (DDM) content was used. The mean ME 284 ml/kg W e for the males and females, respectively.
intakes of the animals were then calculated from their The changes in the body weights of the animals which
mean daily dry matter intakes. The rumen degradable survived throughout the 12-week study period did not
nitrogen (RDN) requirements of the animals were follow any consistent trend. Two of the animals on Diet
calculated using the requirement value of 1.33 g N/MJ 1 and one each of the animals fed Diets 3, 4 and 5
ME (ARC, 1984). The maintenance energy requirement maintained their body weights. Two of the animals fed
for the animals was assumed to be similar to the Diet 5 and one of those fed Diet 1 lost weight. All the
requirement of the temperate breeds for which 0rskov other animals gained weight (range 0.5 to 2.0 kg/84
and McDonald (1970) reported a value of 420 KJ/kg days).
W075/day.
Figure [Link] vitro dry matter degradability values of cassava
Chemical analysis peels and palm kernel cake at various incubation
periods.
The dry matter contents of the CP and PKC and the fat
content of PKC were determined according to the Degradation (%)
methods of AOAC (1980). The nitrogen contents of the 100 -,
two feedstuffs were determined by the Kjedahl method
(AOAC, 1980).
B0
Statistical analysis
The intake of CP, PKC, CP + PKC and water were
subjected to statistical analysis using the 2-way analysis 60 -
of variance (Steel and Torrie, 1980).

40
Results Cassava peels
Palm kernel cake
Chemical composition of the feedstuffs 20

The dry matter contents of the CP and PKC were 84.49

and 86.67%, respectively. The nitrogen contents of the
CP and PKC were 5.44 and 33.53 g per kg dry matter, ~i—i—i—i—i—i—i—г ~[—i—i—i—r ~i—i—
04 8 16 24 48 72 96
respectively. The PKC contained 10.2 g fat per kg dry
Incubation time (h)
matter.

Degradation characteristics of the feedstuffs

Two animals each from Diets 3 and 4 died while
The degradation characteristics of the feedstuffs are one animal each died from those fed Diets 1 and 2.
shown in Figure 1. The rate and level of degradation of
Table 2 shows the calculated energy intake,
CP were higher than PKC after 16 hours of incubation.
maintenance energy requirements and rumen
Before then, the apparent degradability of PKC was
degradable nitrogen requirements of the animals. The
better than that of CP. This may be attributed to particle
size differences. total nitrogen supplied by the PKC for the levels of
intake registered is also shown in Table 2.
Feed and water intake, bodyweight changes and
general performance of the animals
The performance of the animals was generally poor
Discussion
resulting in the deaths of some animals as shown in
Generally the performance of the animals was poor
Table 1.
because of rumen degradable nitrogen deficiency in
The mean intake of water, combined CP and palm their diets. This deficiency was not alleviated by
kernel cake and CP alone were not significantly supplementing the CP with PKC as the intake of PKC
(P>0.05) affected by dietary treatments or sex. The supplements was generally very low. This low intake
intake of the palm kernel supplement was, however, could not be attributed to the fat content of PKC which
significantly (P<0.01) affected by dietary treatment. As was also low (10.2 g/kg dry matter), but could be
the level of supplementation increased, the intake of the attributed to the low palatability of PKC (McDonald et
PKC increased. The mean intake of CP alone was 63 al, 1988) Yussof (1985) reported high intakes of diets
g/kg W per day for males and 61 g/kg W0175 per day containing PKC by cattle, but the cake was completely
for females. For the PKC, the mean intake values were mixed with the other ingredients. The objective of this

Small Ruminant Research Network Workshop 165

A.K. Tuahetal

Table 1. Mean weight performance, feed and water intakes and mortality rates of Djallonké sheep.
Diets
Parameter 1 2 3 4 5
Initial body weight (kg) 11.25 12.38 12.00 11.50 13.38
Final body weight (kg) 11.00 13.25 12.33 12.50 12.88
Changes in body weight (kg/84 -0.25 +0.87 +0.33 +1.00 -0.50
days)
Daily dry matter intake of:
CP(g/kgW°75) 59 60 68 58 64
PKC (g/kgW° 75) - 3c 7b 9a 10a

CP and PKC (g/kgW075) 59 63 75 67 74

Daily water intake (ml/kgW075) 320 267 263 263 256
Mortality rate (%) 20 20 40 50 0
abc = Values in a row with different superscripts are significantly different (P<0.05).

Table 2. Mean performance characteristics of Djallonké sheep.

Diets
1 2 3 4 5
Initial body weight (kg) 11.25 12.38 12.00 11.50 13.38
Metabolic body weight (kg W075) 6.14 6.60 6.45 6.24 6.70
Daily intake of CP(g) 362 396 439 362 429
Daily energy intake from CP (MJ/ME) 3.28 3.564 3.951 3.258 3.861
Maintenance energy requirements (MJ ME/day) 2.5788 2.772 2.709 2.6208 2.814
Energy intake (from CP) as a proportion of maintenance 1.26 1.28 1.46 1.24 1.32
energy requirement (x maintenance)
Rumen degradable nitrogen requirement for the level of energy 4056 4429 4908 4043 4788
intake (mg/day)
Quantity of nitrogen supplied by CP (mg/day) 1969 2154 2388 1969 2334
Mean quantity of nitrogen contained in the PKC eaten (mg) 671 1510 1880 2250
Deficiency of rumen degradable nitrogen (mg/day) 2087 2275 2520 2074 2454

study would have been defeated if the two ingredients rumen degradable nitrogen supplied by CP in this study
had been ground and mixed, as rural farmers would not did not meet the calculated RDN requirement. The
have adopted the feed package because of the cost of deficits ranged from 1969 to 2388 mg/day. These
grinding. The PKC used by Yussof (1985) was not a deficits in the requirements of RDN could not have been
product of heat treated kernels as was the case with the met by nitrogen supplied by the PKC at the levels of
solvent-extracted or screw-pressed kernels in this study. intake recorded, even if all the nitrogen it contained was
The application of heat in the presence of water might rumen degradable.
have caused the Milliard reaction, which could have
Notwithstanding the deficiency of RDN, about
rendered the nitrogen unavailable to both the rumen
74% of the animals survived on the diets for three
microbes and the host animal. This could also partly
months. The intakes of the diets (59 to 68 g/kg W°'75
explain the low intakes of PKC.
per day) were also remarkable, though lower than the
The zero hour dry matter degradability values for
values (87 to 108 g/kg W per day) reported by
PKC and CP shown in Figure 1 may be misleading. The
Adegbola et al ( 1 988) when they fed CP diets containing
values seem to be high whereas the actual degradability adequate amounts of nitrogen to the same breed of
values may not have been. This could be because the sheep.
PKC was pulverised through a 3 mm sieve into very fine
powder, most of which could have leaked out of the
nylon bag and therefore not degraded. The filter paper
soluble fraction was 17%. The calculated mean energy
Acknowledgements
intakes from CP ranged from 1 .24 to 1 .46 x maintenance The funding of this study by the European Community
energy requirements of the animals. The calculated (EC) is acknowledged.

166 Small Ruminant Research Network Workshop

 The effect of supplementation of cassava peel (CP) etc..

References Wenk G (eds), Energy metabolism in farm animals.

EAAP Publication 13. Iuris, Druck and Verlag, Zurich,
Adegbola A A, Smith О В and Okeudo N J. 1988. Responses Switzerland. pp. 121-125.
of West African Dwarf sheep fed cassava peel and 0rskov E R and McDonald I. 1979. The estimation of protein
poultry manure based diets. In: PANESA/ARNAB degradability in the rumen from incubation
(Pastures Network for Eastern and Southern measurements weighted according to rate of passage.
Africa/African Research Network for Agricultural Journal of Agricultural Science (Cambridge) 92:
By-products). Utilisation of research results on forage 499-503.
and agricultural by-product materials as animal feed
resources in Africa. Proceedings of the first joint Roy J H В, Balch С C, Miller E L, 0rskov E R and Smith R
workshop held in Lilongwe, Malawi, 5-9 December H. 1977. Calculation of nitrogen requirements from
1988. PANESA/ARNAB, Addis Ababa, Ethiopia. pp. nitrogen metabolism. In: Tamminga S (ed), Protein
357-366. metabolism and nutrition. Centre for Agricultural
Publishing and Documentation, Wageningen,
AOAC (Association of Official Analytical Chemists). 1980. Netherlands. pp. 126-129.
Official methods of analysis. 13th edition. AOAC,
Steel R G D and Torrie J H. 1980. Principles and procedures
Washington DC, USA. 1081 pp.
of statistics: A biometrical approach. 2nd edition.
ARC (Agricultural Research Council). 1984. The nutrient McGraw-Hill, New York, USA. 633 pp.
requirements of ruminant livestock. Supplement 1. Tuah A K. 1990. Utilisation of agricultural by-products for
Commonwealth Agricultural Bureaux, Slough, UK. village and commercial production of sheep rations in
FAO (Food and Agriculture Organisation of the United Ghana. In: PANESA/ARNAB (Pastures Network for
Nations). 1990. FAO production yearbook 43. FAO Eastern and Southern Africa/African Research Network
Statistics Series 94. FAO, Rome, Italy. for Agricultural By-products). Utilisation of research
results on forage and agricultural by-product materials
Kossila V L. 1984. Location and potential feed use. In: as animal feed resources in Africa. Proceedings of the
Sundstol F and Owen E (eds), Straw and other fibrous first joint workshop held in Lilongwe, Malawi, 5-9
by-products as feeds. Developments in Animal and December 1988. PANESA/ARNAB, Addis Ababa,
Veterinary Sciences 14. Elsevier Science Publishers, Ethiopia. pp. 57-69.
Amsterdam, Netherlands. pp. 4-24.
Tuah A K, Adomako D and Dzoagbe S. 1985. Evaluation of
McDonald P, Edwards R A and Greenhalgh J F D. 1988. cocoa-pod husk as feed ingredientfor sheep in Ghana.
Animal nutrition. 4th edition. Longman Scientific and Proceedings of the 9th International Cocoa Research
Technical Division, Harlow, Essex, UK. pp. 462-463. Conference, Lomé, Togo. pp. 505-509.
Ministry of Agriculture, Accra, Ghana. 1988. Medium-term Tuah A K, Lufadeju E, 0rskov E R and Blackett G A. 1986.
agricultural development programme. Working Paper4. Rumen degradation of straw. 1. Untreated and
Livestock subsector. ammonia-treated barley, oat and wheat straw varieties
Orskov E R and McDonald I. 1970. The utilisation of dietary and triticale straw. Animal Production 43:261-269.
energy for maintenance and for protein and fat Yussof S M. 1985. Feeding value of palm kernel cake for
deposition in young growing sheep. In: Schurch A and growing heifers. Kajiani Veterinary 17(l):49-54.

Small Ruminant Research Network Workshop 167

 Valeur nutritive des résidus de récolte et des sous-produits
agro-industriels offerts à des moutons peuls du Sahel
G. P. Gongnet / , M. Minguey2 et B.O. Brahim
i 2

Ecole inter-Etats des sciences et médecine vétérinaires

B.P. 5077, Dakar (Sénégal)
Laboratoire de recherches vétérinaires et zootechniques
B.P. 433, N'Djamena (Tchad)

Résumé cottonseed cake in the ration was positively correlated

(R=0.96) with rice straw consumption by the sheep.
L'effet de rations à base de paille de riz Adding cottonseed cake to the ration increased
complémentée avec des drèches artisanales, du tourteau lipid level and resulted in the consumption of reduced
de coton et de la poudre d'os sur les performances organic matter, crude fibre and nitrogen-protein. The
d'engraissement du mouton peul du Sahel a été étudié type of ration had a significant effect on mean daily
au Tchad sur 32 moutons âgés de 1 ,5 an et pesant en weight gains (P<0.01).
moyenne 24,25 kg. Ces rations ont eu un effet Three two-year old Fulani sheep were fistulated
statistiquement significatif sur la consommation de and used to determine dry matter fermentability of
matière sèche (P<0,01). Une corrélation positive (R = cowpea straw and pods, groundnut straw and dry
0,96) a été mise en évidence entre la proportion de brewer's grains after 0, 6, 12, 24, 48 and 72 hours
tourteaux de coton dans les rations et la consommation incubation. The breakdown level was maximum after 24
de paille de riz par les moutons. hours incubation for groundnut straw (73.62%), 48
L'augmentation du taux de lipides favorisée par hours for cowpea straw (73.48%) and 72 hours for
l'addition de tourteaux de coton dans la ration a entraîné cowpea pods (54%) and brewer's grains (47.05%).
une diminution de la consommation de matière
organique, de cellulose brute, d'extractif non azoté et
d'énergie brute, mais non de celle de matières azotées 1. Introduction
totales. Le type de ration avait un effet significatif sur
les gains moyens quotidiens (P<0,01). En zones tropicale et subtropicale, et plus
Trois moutons peuls du Sahel âgés de 2 ans ont été particulièrement au Sahel, les ruminants sont entretenus
fistulisés et utilisés pour déterminer la dégradabilité de toute l'année sur des pâturages naturels constitués en
la matière sèche des fanes et des coques de niébé, des grande partie de graminées, et dont la valeur nutritive
fanes d'arachide et des drèches artisanales sèches après diminue très rapidement en saison sèche. Leurs besoins
des temps d'incubation de 0, 6, 12, 24, 48 et 72 heures. n'étant alors plus couverts, les animaux doivent
Le taux de dégradation des fanes d'arachide atteignait quotidiennement parcourir de grandes distances à la
son maximum (73,62%) après 24 heures d'incubation, recherche d'eau et d'aliments. Cette situation entraîne
contre 48 heures pour celles de niébé (73,48%) et 72 généralement des pertes considérables de production:
heures pour les coques de niébé (54%) et les drèches chutes de poids et taux de mortalité et de morbidité
artisanales (47,05%). élevés chez les jeunes et les mères gestantes ou
allaitantes.
Dans les zones agropastorales du Tchad, les
Nutritive value of crop residues sous-produits agricoles tels que le tourteau de coton, les
fanes et les coques des légumineuses (arachides, niébés)
and agro-industrial by-products sont généralement utilisés seuls, à grande échelle,
fed to Fulani sheep surtout en saison sèche. Les drèches de sorgho ne sont
pas utilisées judicieusement dans l'alimentation
Abstract animale.
Le but de la présente étude est d'une part de
The effect of rations based on rice straw supplemented présenter des rations de complémentation composées de
with brewer 's grains, cottonseed cake and bone meal on drèches artisanales sèches de sorgho, de tourteaux de
fattening performance of Fulani sheep was determined coton, de poudre d'os et de sel destinées à l'alimentation
in Chad on 32 1.5 year old sheep with an average weight du mouton du Sahel en saison sèche, et d'autre part
of 24.25 kg. The rations had a significant effect d'étudier la dégradabilité de la matière sèche, des
(P<0.01) on dry matter consumption. The proportion of drèches artisanales sèches et de certains sous-produits

Small Ruminant Research Network Workshop 169

G. P. Gongnet, M. Minguey et B.O. Brahim

de légumineuses. Les meilleures rations pourraient être Tableau 1. Composition des concentrés
vulgarisées au niveau des éleveurs sahéliens afin de 1-ots
permettre aux animaux d'éviter des pertes de poids et
Ingrédients (%) I II III IV
d'améliorer leur productivité en saison sèche.
Drèches sèches 97,00 72,00 47,00 22,00
Tourteaux de coton 0,00 25,00 50,00 75,00
2. Matériel et méthodes
Poudre d'os 2,00 2,00 2,00 2,00

2.1 Essais d'alimentation Sel (NaCl) 1,00 1,00 1,00 1,00

Total 100,00 100,00 100,00 100,00
2.1.1 Animaux
32 moutons mâles peuls du Sahel âgés de 1,5 an Les refus de paille et de concentrés étaient
environ et pesant en moyenne 24,25 kg ont été utilisés récupérés séparément chaque jour et pesés chaque
dans cette expérimentation, qui s'est déroulée d'octobre semaine, ce qui a permis de calculer les quantités
à décembre 1 989 (65 jours) au Laboratoire de recherches d'aliments, de nutriments et d'énergie brute
vétérinaires et zootechniques de Farcha au Tchad. effectivement consommées. L'eau d'abreuvement était
Après trois semaines d'adaptation, ces animaux, fournie à volonté.
achetés sur le marché de N'Djamena, ont été déparasités
au Tigal (ND) et au Vadephen (ND). Ils ont également 2.1.3 Prises de sang et analyses
reçu une injection intramusculaire de Terramycine
A la fin des expériences, des prises de sang ont été
longue action (ND).
effectuées par ponction de la veine jugulaire sur tous les
Les animaux, identifiés individuellement par des animaux. Le sang était recueilli dans des tubes secs en
boucles auriculaires, ont été répartis en 4 lots de 8 vue de la récolte de sérum après centrifugation. Le sérum
moutons chacun. Au cours des expériences, ils étaient a été conservé à -20 °Cjusqu'au moment de son analyse.
maintenus en stabulation libre et pesés tous les 14 jours.
Après décongélation du sérum, le calcium (Ca), le
phosphore (P), le magnésium (Mg) et les protéines
2.1.2 Aliments et alimentation totales ont été dosés à l'aide du Technicon R.A. 1000,
L'aliment de base était constitué de paille de riz un auto-analyseur, selon les méthodes décrites par
distribuée à raison de 2 kg par lot et par jour. Ces Skeggs et Hochstrasser ( 1 964) et Audege et coll. ( 1 983).
quantités étaient présentées en deux fois aux animaux (à
8 heures et à 16 heures). Des rations de complé 2.2 Digestibilité in sacco
mentation, composées de drèches artisanales sèches, de
tourteaux de coton, de poudre d'os et de sel tableau 1) Trois moutons peuls âgés en moyenne de 2 ans et
étaient distribuées à raison de 50 g/kg P ' /j en une fistulisés ont été utilisés pour l'étude de la dégradabilité
seule fois à 1 1 heures, c'est-à-dire 3 heures après la de la matière sèche des drèches artisanales sèches et des
distribution de la première portion de l'aliment de base. fanes d'arachides et de niébé.
Après chaque pesée des moutons, les quantités de 2 à 3 g d'aliments finement broyés (1 mm) ont été
concentrés à fournir étaient calculées en fonction de pesés dans des sachets (8 x 14 cm) en tissu Tripette de
l'évolution pondérale des animaux. 46 nm de pores. Six sachets numérotés ont été prévus

Tableau 2. Composition chimique (en % de MS) des concentrés et de la paille de riz utilisés dans les rations
Concentrés
rame ae riz
I II III IV

Mat. sèche 96,32 95,89 95,69 92,69 92,67

Mat. organique 73,00 75,50 78,16 79,74 78,16

Mat. azotée tot. 20,31 26,18 30,65 31,51 4,20

Cellulose brute 6,57 7,22 7,57 5,51 31,32

Mat. grasse 7,54 9,99 11,22 13,40 1,43

Mat. minérale 23,32 20,39 17,53 15,95 14,51

Ca 0,74 0,88 0,58 0,66 0,09

P 6,74 0,88 0,91 1,02 0,30

170 Small Ruminant Research Network Workshop

 Valeur nutritive des résidus de récoltes et des..

par mouton et par échantillon d'aliment. Cinq sachets des expériences, un taux de mortalité de 25% a été
contenant des échantillons d'aliments broyés, enregistré dans chacun des lots I et IV. Les causes de
préalablement séchés à l'étuve et pesés, ont été cette mortalité n'ont pas été déterminées.
introduits dans la panse de chaque mouton. Après 6, 1 2,
24, 48 et 72 heures d'incubation, ces sachets ont été
3.1 .3 Performance d'engraissement
retirés et lavés à l'eau du robinetjusqu'à obtention d'une
couleur claire. Les résidus contenus dans les sachets ont
Les consommations de matière sèche étaient
été séchés pendant 48 heures à l'étuve à 70 °C. La statistiquement différentes selon le type de ration
différence entre les poids avant incubation et après (P<0,01). Le tableau 3 fait apparaître la corrélation
incubation donne les quantités de matière sèche positive observée entre l'augmentation de la consom
disparues lors de l'incubation. mation de paille de riz et la proportion de tourteaux de
coton dans les différentes rations (r = 0,96).
2.3 Analyse statistique L'accroissement du taux de tourteaux de coton
dans la ration a entraîné une augmentation des quantités
Une analyse de variance avec test F de Fisher (Essl,
de lipides ingérées par les moutons de chaque lot. On
1987) a été effectuée pour comparer les gains moyens
note une diminution de l'ingestion de cellulose brute, de
quotidiens (GMQ), la consommation de la paille de riz
l'extractif non azoté et de l'énergie brute dans les lots
et des concentrés et les constituants sanguins (Ca, P, Mg
III et IV par rapport au lot II. La quantité de matières
et protéines totales).
azotées totales la plus importante est consommée par les
animaux du lot III suivi de ceux du lot II (tableau 4).
Les GMQ du lot I étaient statistiquement différents
3. Résultats (P<0,01) de ceux des autres lots. Des chutes du poids vif
ont été observées au cours des quatre premières
3.1 Essais d'alimentation semaines dans les lots I et IV, dont les rations
contenaient une forte proportion de drèches et de
tourteaux de coton respectivement. Après la 4e semaine,
3.1.1 Composition chimique des aliments
les animaux de ces lots ont progressivement repris du
La composition chimique des aliments est donnée poids, pour atteindre l'état d'équilibre vers le 30e jour
au tableau 2. La composition organique et minérale des pour les animaux du lot IV et le 42e jour pour ceux du
quatre rations était acceptable et permettait de couvrir lot I. Après cette période, aucune différence de poids n'a
les besoins des ruminants tropicaux. été notée entre les lots II, ГII et IV.

3.1.2 Etat de santé et taux de mortalité

3.1.4 Constituants sanguins
Après la première semaine d'expérimentation, la
plupart des animaux des lots I, III et IV ont eu des A la fin de l'expérimentation, les concentrations
diarrhées et certains présentaient du larmoiement. des constituants sanguins (Ca, P, Mg et protéines totales)
Plusieurs moutons du lot IV ont présenté des dépilations des moutons de chaque lot étaient significativement
au niveau de l'encolure, sur le dos et les flancs. A la fin différentes (P<0,01) d'un lot à l'autre (tableau 5).

Tableau 3. Consommation de rations complémentées avec des drèches artisanales et des tourteaux de coton, et croissance
pondérale
Lots
I II III IV
Poids moyen (kg) 25,00 23,50 24,54 23,94
(±3,56) (± 3,96) (± 2,85) (± 1.21)
Consommation partielle
Paille de riz (g de MS/tête/j) 146,46 177,00 181,35 204,42
Concentrés (g de MS/tête/j) 508,00 607,42 552,00 486,00
(gdeMS/lOOkgdePV) 2 031,48 2 584,76 2 245,71 2 029,42
(g de MS/kg de P075) 45,42 56,91 49,99 44,88
Consommation totale
Paille de riz (g de MS/tête/j) 654,00 784,00 733,00 690,00
(g de MS/kg de P0,75) 58,5 73,5 66,5 63,8
Croissance pondérale
Gain total (kg) 1,00 3,83 4,07 4,33
GMQ (g) 15,38 61,53 62,63 61,54

Small Ruminant Research Network Workshop 171

G. P. Gongnet, M. Minguey et B.O. Brahim

Tableau 4. Consommation de nutriments (g/tête/jour) et d'énergie brute par type de concentré

Lots
II III IV
Mat. azotées totales 103,15 159,02 169,12 153,02
Cellulose brute 33,36 43,85 41,77 26,76
Mat. grasses 38,29 60,68 61,91 65,07
Extractif non azoté 195,94 195,05 158,46 142,40
Energie (en UFV)a 1,06 1,38 1,33 1,22
aUFV = unité fourragère viande.

Tableau 5. Constituants sanguins des moutons du Sahel enfin d'essai d'alimentation

Lots
I (n = 7) II (n = 7) III (n = 8) IV(n = 4)
Ca (mg/I) 64,57 61,43 64,75 80,00
(± 6,21) (± 7,94) (± 10,38) (± 12,31)

P(mg/1) 63,33 60,71 75,50 77,00

(± 12,83) (± 23,31) (± 30,06) (± 18,45)

Mg (mg/1) 12,14 13,00 21,25 21,00

(± 2,75) (± 2,87) (± 5,09) <± 4,52)
Protéines totales (g/1)
56,00 48,71 52,25 65,00
(± 9,64) (± 11,94) (± 9,94) (± 11,49)
n = nombre de prélèvements.

3.1.5 Digestibilité in sacco

Figure 1. Dégradabilité de la matière sèche des différents
Avant l'incubation (0 heure), la digestibilité de la aliments
matière sèche des fanes de niébé et d'arachide était
nettement supérieure à celle des coques de niébé et des Dégradabilité MS (%)
drèches. Les coefficients de dégradabilité des fanes 80-,
étaient les plus élevés, quelle que soit la durée
d'incubation (figure 1). 70
Après 24 heures d'incubation, la dégradabilité
moyenne de la matière sèche des fanes d'arachide et de 60
niébé était respectivement de 73,62% et de 58,97%,
contre 35,75% pour les coques de niébé et 39,86% pour 50
les drèches. Les fanes de niébé n'atteignaient leur
dégradation maximale (73,48%) qu'après 48 heures, 40
durée d'incubation après laquelle la dégradation des
coques de niébé n'était que de 52%, avant d'atteindre Fane d'arachide
30
54% au bout de 72 heures. Enfin, la dégradabilité de la Fane de niébé
matière sèche des drèches artisanales n'était que de Coque de niébé
20
47,05% après 72 heures d'incubation.
Drèche de bière artisanale
10 i i i r~i i i i i i i rn i i i i r~i r~r
4. Discussion 0 20 40 M 80
Temps d'Incubation (heures)
4.1 Essais d'alimentation
Les performances d'engraissement des moutons de relativement âgés (18 mois environ). Des GMQ
cette expérience étaient faibles ou moyennes, avec des semblables ont cependant été enregistrés par Calvet et
GMQ allant de 1 5,32 à 62,63 g/j. Ces résultats pourraient Denis (1974) sur des moutons peul-peul sénégalais âgés
s'expliquer par le fait que les animaux utilisés étaient de 8 à 10 mois.

172 Small Ruminant Research Network Workshop

 Valeur nutritive des résidus de récoltes et des...

Des valeurs plus faibles ( 1 8,50 g et 20,60 g) ont été niébé et les drèches artisanales utilisées dans cette étude.
obtenues par Fall (1989) sur des moutons peuls du Par ailleurs les taux de dégradabilité des fanes
Sénégal âgés de 1 an en moyenne, avec des rations d'arachide étaient supérieurs aux chiffres de 55,10% et
composées de paille de riz, de tourteaux d'arachide et de 66,00% rapportés par Parigi-Bini et coli .( 1 99 1 ) pour des
poudre d'os. fanes d'arachide du Sénégal après 12 et 24 heures
En revanche, Bourzat et coll. ( 1 987) rapportent des d'incubation. Avec les aliments provenant du Tchad, les
valeurs de 86 g/j à 117 g/j chez des moutons peuls âgés taux de dégradabilité de la matière sèche étaient ici de
de 4 à 5 mois, respectivement non traités et traités aux 62,99% et de 73,62% respectivement pour les mêmes
anthelminthiques dans le nord du Burkina Faso, avec des durées d'incubation.
rations composées de foin de Stylosanthes et de
concentrés composés de 45% de son de blé, 35% de
brisures de riz, 15% de tourteaux de coton et 3% de 5. Conclusion
complexe minéral vitaminé. Avec des rations de
complémentation en saison sèche, ces mêmes auteurs L'amélioration de l'alimentation de saison sèche
ont obtenu un gain moyen quotidien de 126 g/j sur des des petits ruminants au Sahel passe par l'utilisation
moutons peuls âgés de 8 mois. Ces résultats démontrent rationnelle des résidus de récolte et des sous-produits
les bonnes aptitudes bouchères des moutons peuls du agro-industriels locaux. Ainsi, en formulant des rations
Sahel, déjà mentionnées par Charray et coll. (1980). composées de tourteaux de coton, de drèches artisanales
Les GMQ de 61,53 g à 62,63 g obtenus avec les sèches et de poudre d'os, on s'est aperçu que les chiffres
rations II, III et IV peuvent être considérés comme les plus élevés de matière sèche consommée et de gain
satisfaisants, vu l'âge des animaux et la courte durée de moyen quotidien étaient respectivement associés à des
la période d'observation (65 jours), l'embouche ovine taux de tourteaux de coton de 25% et de 50% dans les
requérant une période d'observation d'au moins 90 rations.
jours, temps utilisé par la plupart des auteurs cités Les meilleurs taux de dégradabilité de la matière
ci-dessus. sèche étaient obtenus avec les fanes d'arachide et de
La diminution des quantités de cellulose brute, niébé, lesquelles pourraient être utilisées pour la
d'extractif non azoté et d'énergie brute ingérées pourrait complémentation des pâturages naturels pauvres et/ou
s'expliquer par l'augmentation de la teneur en lipides des rations à base de paille de graminée utilisées pour
des concentrés. l'alimentation des ruminants en saison sèche au Sahel.
Les taux de calcium et de phosphore sont
identiques à ceux rapportés pour les brebis peules du
Sahel par Brahim et Gongnet ( 1 987) au Tchad et Assane Remerciements
et coll. (1989) au Sénégal. Les valeurs de calcémie
obtenues par Assane et coll. (1989) de 103,71 mg/1 à Les auteurs tiennent à remercier très sincèrement
1 1 1,86 mg/1 chez la brebis et de 61,43 mg/1 à 80,00 mg/1 la Fondation internationale pour la science (FIS) qui a
chez les moutons mâles non castrés sont nettement bien voulu financer ce projet.
supérieures à celles observées ici.
La calcémie de la brebis peule du Sénégal ne
semble subir aucun effet significatif de la gestation Références
(Sawadogo et coll., 1988), mais il est probable qu'elle
soit positivement influencée par l'alimentation. Audege C.I., Dupont G. et Zonszain F. 1983. Principes des
méthodes d'analyses biochimiques. Tome 1. Doin
Lorsque la ration est composée soit uniquement de France. 122 p.
drèches artisanales sèches (lot I), soit de 25% de
tourteaux de coton (lot II), la phosphorémie de la brebis Assane M., Sere A., Djimrao S., Sow R. I. Ba A.C. et Gaye O.
peule (45,95 mg/1 à 66,64 mg/1) (Assane et coli, 1989) 1989. Variations physiologiques de la calcémie et de la
phosphorémie chez la brebis Peulh du Sahel en gestation.
est identique à celle des moutons mâles de la présente
Sciences vétérinaires-Médecine comparée 9 1 :263-27 1 .
étude.
La concentration en protéines totales augmente Brahim B.0. 1987. Etude de la protéinémie des caprins et ovins
du Tchad en fonction de l'espèce, de la race et de l'état
considérablement lorsque le taux du tourteau de coton
de santé. In: Rapport Annuel 1986-1987, Laboratoire de
des rations est de 75% (lot IV). Les valeurs des protéines recherches vétérinaires et zootechniques de Farcha,
totales observées ici sont semblables à celles obtenues N'Djamena (Tchad), p. 183 à 184.
par Brahim (1987) chez les brebis peules du Sahel au
Brahim B.O. et Gongnet G. P. 1 987. Hématologie et biochimie
Tchad.
des petits ruminants du Laboratoire de Farcha et de la
Faculté des sciences exactes et appliquées de
4.2 Etude de digestibilité in sacco l'Université du Tchad. In: Rapport Annuel 1986-1987,
LRVZ (Laboratoire de recherches vétérinaires et
Selon Preston ( 1 986), la dégradabilité de la matière zootechniques de Farcha), N'Djamena (Tchad), p. 1 87 à
sèche d'un aliment de bonne valeur nutritive doit être 190.
égale ou supérieure à 50% après 24 heures d'incubation Bourzat D., Bonkoungou E., Richard D. et Sanfo R. 1987.
dans le rumen. Ce ne fut pas le cas pour les coques de Essais d' intensification de la production animale en zone

Small Ruminant Research Network Workshop 173

G. P. Gongnet, M. Minguey et BO. Brahim

sahélo-soudanienne: alimentation intensive de jeunes international pour l'élevage en Afrique), Addis-Abeba

ovins dans le Nord du Burkina. Revue d'élevage et de (Ethiopie). p. 155 à 166.
médecine vétérinaire des pays tropicaux 40(2): 151-156.
Parigi-Bini R., Xiccato G., Garazzolo A., Guzzinati R. et
Calvet H. et Denis J.P. 1974. Embouche intensive du mouton Cinetto M. 1991. Composizione e degradabilità
Peulh-Peulh sénégalais. LNERV (Laboratoire national ruminale de alimenti zootecnici de provenienza
de l'élevage et de recherches vétérinaires), Dakar-Hann tropicale (Sénégal). Actes du IXe Congrès national de
(Sénégal). 54 p. l'ASPA (Associazione scientifica di produzione
Charray J., Coulomb J., Haumesser J.B., Planchenault D. et animale) tenu à Rome (Italie), 3-7 juin 1 99 1 . p. 35 à 46.
Pugliese P.L. 1980. Les petits ruminants d'Afrique
Preston T.R. 1986. Better utilization of crop residues and
centrale et d'Afrique de l'Ouest: Synthèse des
by-products in animal feeding: research guidelines. 2. A
connaissances actuelles. IEMVT (Institut d'élevage et
practical manual for research workers. FAO Animal
de médecine vétérinaire des pays tropicaux),
Production and Health Paper: No 50/2. FAO
Maisons-Alfort (France); Ministère de la coopération,
(Organisation des Nations Unies pour l'alimentation et
Paris (France). 295 p.
l'agriculture), Rome (Italie). 154 p.
Essl A. 1987. Statistiche Methoden in der Tierproduktion.
Oeter. Agrarverlag, Vienne (Autriche). 316 p. Sawadogo J., Abiola F., Abassa P.K. et Sow R. 1988.
Principaux constituants minéraux sériques chez les
Fall S. T. 1989. Utilisation d'Acacia albida et de Calotropis moutons Peulh du Sénégal. Communication présentée
procera pour améliorer les rations des petits ruminants aux XIIe Journées médicales et pharmaceutiques, 16-23
au Sénégal. In: African Small Ruminant Research and janvier 1988. Dakar (Sénégal). 7 p.
Development. Publié sous la direction de R.T. Wilson et
Azeb Melaku. Actes de la conférence tenue à Bamenda Skeggs L.T. et Hochstrasser H. 1964. Multiple automatic
(Cameroun), 18-25 janvier 1989. C1PEA (Centre sequential analysis. Clinical Chemistry 10:918-936.

174 Small Ruminant Research Network Workshop

 Seasonal changes in body condition of the Senegalese Sahel
goat: relationship to reproductive performance
M. Cissé, M. M'Baye, I. Sané, A. Corréa and I. N'Diaye

Service de nutrition, Laboratoire national de l'élevage et des

recherches vétérinaires (ISRA), B.P. 2057, Dakar, Sénégal

Abstract with the Sahel goat and to investigate relationships

between body condition and reproductive performance.
The level of feeding was found to be an important
determinant of goat body condition and correlated
positively with the conception rate. Goats lost body Material and methods
condition with the progressive deterioration of pasture
in the dry season. This condition can be improved with The 49 female and six male Sahel goats were on average
a sufficient level of concentrate supplementation. The two years old. They were managed under the same
lumbar area score was more closely related to live conditions during the dry season in the peri-urban zone,
weight than the sternum score. with eight hours a day of pasture grazing. After two
months of daily supplementation with 300 g of
groundnut cake per animal, oestrus was synchronised
Rapport entre les variations with a vaginal sponge containing 45 mg of fluorogestone
acetate (Chronogest, Intervet, France). The objective
saisonnières de l'état corporel et was to encourage kidding at the beginning of the raining
season. From March to May, goats were daily offered
les performances de 200 g of concentrate (66% maize, 30% groundnut cake,
reproduction de la chèvre du 4% CMV) per animal. From June to July, each
non-pregnant goat received 400 g of concentrate daily.
Sahel au Sénégal Pregnant goats were offered 500 g per animal.
Pregnancy was tested by progesterone analysis in blood
Résumé sampled at 10-day intervals six weeks after heat began.
Goats were weighed every 14 days and body condition
Cette étude fait apparaître que le niveau scored early in the morning the day before the vaginal
d'alimentation est un facteur déterminant de l'état sponge was withdrawn. Thereafter they were weighed
corporel des caprins et qu 'il est positivement corrélé once a month according to the method of Santucci
avec le taux de conception. La détérioration progressive ( 1 984), using a 1-5 scale of scoring. The last scoring was
des pâturages en saison sèche a entraîné une done in July (at 16.43 ± 3.1 days before kidding). Three
déperdition de l'état corporel des caprins, qu'une scores (ocular, sternum and lumbar) were assigned after
supplémentation suffisante avec des aliments observation and palpation on the lumbar and sternum
concentrés permet cependant de relever. La note de areas. Goats' grazing behaviour and plant species
selected from pasture were studied twice a month. The
pointage obtenue à la région lombaire était plus
Student t-test was used to compare the performance of
étroitement liée au poids vif que celle du sternum.
pregnant and non-pregnant goats. In each group,
seasonal changes in performance were calculated with
a paired t-test.
Introduction
Lipids are the main form of body energy storage in goats, Results and discussion
as in others mammals (Chilliard, 1 987). Lipid deposition
and mobilisation cycles are essential for reproduction Conception rate was 41% and correlated positively (r=
and lactation performance and for ecological adaptation. 0.3) with condition score on the day before vaginal
Numerous reliable methods have been developed to sponge withdrawal. Does that did not conceive had a
predict in vivo changes in body reserves. Morand-Fehr lower condition score at mating (Table 1), as already
et al (1987) extensively reviewed the use of these reported for ewes by Gunn et al (1969). Figure 1(a)
methods in goats. With the exception of body condition shows the mean condition score performance as based
scoring, most of them are very expensive. Body in Morand-Fehr ( 1 992), taking into account sternum and
condition scoring is however easier to use although it is lumbar scores. Pregnant does recorded higher (P<0.01)
subjective. The purpose of this study was to calibrate it scores than the non-pregnant does. All groups showed a

Small Ruminant Research Network Workshop 175

M. Cissé, M. M'Baye, I. Sané, A. Correa and I. N'Diaye

Table 1. Seasonal changes in goat body condition scores2 of Sahel goats.

OCS STS LCS
Month P NP P NP P NP

February3 2.9 ± 0.5 2.3 ± 0.8

April 3.1 ±0.9 2.6 ±0.9 3.0a ±1.1 2.3b± 1.1 3.1 ±0.9 2.7 ±0.8
3.1a±0.7 2.5b±0.7 3.2a ±0.6 2.6b ± 0.7 3.9a ± 0.5 3.3b±0.6
May
June 2.5a ±0.4 2.1b±0.4 2.6a ±0.6 2.1b±0.5 3.1a±0.5 2.7b±0.5

July 3.3 ±0.6 3.0 ±0.8 3.1 ±0.8 2.8 ±0.9 3.9 ±0.4 3.6 ±0.7
Means (and standard deviation) in pregnant (P, n = 20) and non-pregnant (NP, n= 29) goats.
OCS = ocular condition score, SCS = sternum condition score, LCS = lumbar condition score.
21 February= the day before vaginal sponge withdrawal (i.e. at mating).
All means with different superscripts are different (P<0.01).

Figure 1. Seasonal change in mean body condition score and liveweight ofgoats.

Condition »core Liveweight (kg)

40
«) 40

At mating
Increase In level of b)
supplementation
30 30

20- 20

10 Pregnant 10
Non-pregnant Pregnant

Non-pregnant

Apr May Jun Jul Months Feb Apr May Jun Jul Months

rapid decline in body condition between May and June. contents (Morand-Fehr et al, 1987). Computations on
This corresponded to a general decline in the quantity 196 pooled data from all goats for the period between
and quality of the herbaceous pasture that goats depend April and July showed approximately 3.2 kg liveweight
on for the major supply of their nutrients. However, change for one unit of mean condition score (Table 2).
pregnant does lost more condition (P<0.01) than the However, the prediction was more precise with the
others (-0.63 ± 0.71 vs -0.59 ± 0.43). From June to July lumbar score (RSD = 3.9) than that of change in sternum
the increase in the level of concentrate supplementation
(RSD = 4.1) or ocular score (RSD = 4.2). Generally,
improved body condition in all does in agreement with
the findings of Santucci (1984) and Branca (1987). there was more correlation between live weight and
Pregnant does, however, recovered slightly less condition score in non-pregnant than in pregnant does
(P<0.01) than non-pregnant does (0.7 ±0.54 vs 0.85 (Table 2). Bias due to calibrating on animals differing
± 0.52). This may be related to increased energy needs in age and physiological stage or overestimating
in gravid does due to foetal growth which could lead to pregnant goats, body condition must be taken into
less lipid deposition or an increase in lipomobilisation. account. Correlations between condition scores and
Liveweight changes did not always parallel body liveweight in this study (Table 2) agree with similar
condition scores (Figure lb). This is not surprising relationships observed by Branca ( 1 987, R = 0.6) and by
because it is well established that body condition reflects Morand-Fehr et al (1987, R = 0.57 for sternum score and
body lipids more than live weight because of gut R = 0.44 for lumbar score). However, we found sternum

176 Small Rumiant Research Network Workshop

 Seasonal changes in body condition of the etc..

Table 2. Linear regressions (Y = aX + b) between live-wtright (LW) and body condition scores of the Sahel goat.
Y X2 a b R RSD a' V R' RSD'
LW(kg) LCS 3.1 15.5 0.52 3.9 3.5 12.9 0.63 3.3
ses 2.4 19.1 0.47 4.1 3.0 16.5 0.61 3.4
oes 2.3 19.4 0.39 4.2 2.3 17.3 0.47 3.8
MCS 3.2 16.2 0.53 3.9 3.7 13.6 0.67 3.2
a, b, R, RSD are calculated from all goats (n= 196) from April to July; a', b', R', RSD' were obtained from non-pregnant goats only (n =1 16).
1 See Table 1 plus MCS = mean condition score = (LCS+SCS/2).

and lumbar scores were less related (R = 0.73) than in Kenya). We are grateful to Drs Arona Guèye and Mei'ssa
the study reported by Morand-Fehr et al (1987, R N'Diaye for assistance during this study and to S Souaré
= 0.87). These differences are perhaps due to for herding the animals.
peculiarities in the adipose deposits of local breeds
and/or to the frequent hardening of the skin on the
sternum area which disturbed the palpation. This study
clearly showed a seasonal change of body condition
References
scores, but absolute values could be expected to be Branca A. 1987. Evolution de l'état corporel et du niveau des
smaller if a 0-5 scale scoring (Morand-Fehr, 1992) was dépôts adipeux chez la chèvre sarde. Symposium
used. We found good repeatability of the method (R = Philoekios sur "L'évaluation des ovins et des caprins
0.9) between two scores from the same observer but méditerranéens", Santarem (Portugal), 23-25
reproducibility was somewhat low (R = 0.62) between septembre, 1987.
scores from two observers, partly because of the Chilliard Y. 1 987. Variations quantitatives et métabolisme des
inexperience of one observer. lipides dans les tissus adipeux et le foie au cours du cycle
gestation-lactation. Reproduction Nutrition
Development 27:327-347.
Conclusion Gunn R G, Donney J M and Rüssel A J F. 1969. Fertility in
Scottish Blackface ewes as influenced by nutrition and
Body condition scoring is of great interest in nutritional body condition at mating. Journal of Agricultural
management and is useful not only for research Science (Cambridge) 73:289-294.
scientists but also for farmers and development
planners. Its validity will be enhanced by complete Morand-Fehr P, Branca A, Santucci P and Napoleone M. 1 987.
dissection of the goats and body chemical composition Méthodes d'estimation de l'état corporel des chèvres
reproductrices. Symposium Philoekios sur "L'éva
analysis in order to establish relationships between the
luation des ovins et des caprins méditerranéens",
scores and body lipids and to improve methods for
Santarem (Portugal), 23-25 septembre, 1987.
evaluating body condition of the sahel goat.
Morand-Fehr P. 1992. Intérêt d'évaluer l'état corporel des
chèvres dans les milieux peu maîtrisés. Capricorne
5:2(9-14).
Acknowledgements
Santucci P. 1984. Essai de mise au point d'une méthode
This work received financial support from the African d'estimation de l'état d'engraissement des chèvres
Small Ruminant Research Network (International corses. Séminaire FAO sur la nutrition et l'alimentation
Livestock Center for Africa, P.O. Box 46847, Nairobi, des caprins, Grangeneuve (Suisse), 16- 18 octobre 1984.

Small Ruminant Research Network Workshop 177

 Wet season supplementation of West African Dwarf goats
raised under traditional management in the
subhumid zone of Nigeria
O.A. Ikwuegbu and S. Ofodile

International Livestock Centre for Africa

P.M.B. 2248, Kaduna, Nigeria

Abstract pour les animaux nourris sur les banquesfourragères et

de 1,5 ± 0,5 pour ceux nourris sur les pâturages
Analyses were conducted on the performance and naturels. Les intervalles de mises-bas étaient
productivity of West African Dwarf (WAD) goats kept respectivement de 259 ± 19,8 et 269 ± 12,4 jours et ne
under village conditions in Ganawuri of Plateau State. différaient pas de manière significative. L'accès aux
The analyses covered 420 kids from 264 kiddings over banques fourragères engendrait une réduction
a two-year period (August, 1989 - August, 1991). significative (P<0,05) des pertes pondérales chez les
During the wet season one group of flocks grazed forage chèvres non gravides pendant la saison des pluies. Le
legume while another group grazed natural pasture. The type et la saison de naissance avaient un effet significatif
average litter sizes were 1 .6 + 0.60 and 1 .5 ± 0.5 on the (P<0,05) sur la mortalité cumuléejusqu 'à 1 an. L 'accès
fodder banks (FB) and natural pasture (NP), aux banques fourragères améliorait significativement
respectively, and were significantly (P<0.01) affected (P<0,05) la viabilité des jeunesjusqu'à 1 an. La saison
by parity. The kidding intervals were 259 ± 19.8 and de naissance, le type de naissance et le sexe avaient un
269 ± 1 2.4 days, respectively, and were not significantly effet significatif (P<0,05) sur le poids des jeunes mais
different. Access to the fodder bank significantly ces différences dispa- raissaient après 90 jours. Les
(P<0.05) reduced weight losses in non-pregnant does in innovations visant à soutenir les activités de culture et
the wet season. Cumulative mortality to one year was d'élevage dans des régions à prédominance culturale
significantly affected (p<0.05) by birth type and season sont examinées.
of birth. Access to FB significantly (P<0.05) improved
viability of kids to one [Link] of kids were
significantly affected (P<0.05) by season of birth, birth
type and sex but these differences had disappeared by Introduction
90 days. Innovations to sustain crop/livestock activities
in a predominantly cropping situation are discussed.
Small ruminants, particularly goats, play an important
role in the economic life of the smallholder farmer,
converting low-cost inputs to high-value products
Complémentation alimentaire de (meat, milk and skins). Indeed the goat has been
saison des pluies de la petite described as "the poor man's cow" (Mba, 1979) and is
well adapted to a relatively harsh environment. The
chèvre d'Afrique occidentale West African Dwarf (WAD) goat is kept by crop farmers
as a secondary enterprise in the subhumid zone, defined
traditionnelle élevée dans la ecologically as lying between 900 and 1500 mm
zone subhumide du Nigéria isohyets and having 180 to 270 plant-growing days
(Jahnke, 1982). Goats mature fast and are prolific.
Résumé Despite high kid mortality rates, goat-keeping could be
a profitable enterprise at village level.
Les performances et le niveau de production de la In an area where cropping is the major activity, the
petite chèvre d'Afrique occidentale élevée en milieu traditional method of keeping goats during the wet
villageois à Ganawuri dans l'Etat du Plateau, ont été season is by tethering or free-ranging on enclosed fallow
analysés. L'analyse a porté sur 420 chevreaux issus de land to prevent damage to crops. Often the grazing orbit
264 mises-bas survenues au cours d 'une période de deux of the goat is determined by the length of the tether. This
ans (août 1989 - août 1991). Pendant la saison des restriction does not permit free selection of feed from
pluies, un lot d'animaux pâturaient des légumineuses natural pasture which matures rapidly and has a low N
fourragères et un autre séjournait sur des pâturages and mineral content (Kapu, 1975; Mohammed Saleem
naturels. La parité avait un effet significatif (P<0,05) and Otsyina, 1986). Goats therefore experience feed
sur la taille moyenne des portées, qui était de 1,6 ± 0,60 constraints in the wet season resulting in weight losses

Small Ruminant Research Network Workshop 179

O.A. Ikwuegbu and S. Ofodile

in adult goats and high kid mortality rates, partly due to Least Squares Models (SAS, 1987). Models for
malnutrition, disease and internal parasites. mortality and growth to one year included the fixed
This study reports on the differences in effects of year of birth, season, birth type and parity.
productivity between goats allowed to graze small plots There were two wet season feed regimes, natural pasture
of forage legume (mini fodder banks) in the wet season and fodder bank (FB) grazing. Goats on natural pasture
and those which grazed natural pasture under village were either free-grazing in enclosures where they could
conditions. eat selectively or were tethered. The seasons were
divided into four: late dry (December - February); early
wet (March - May); peak wet (June - August); and early
dry (September - November). The year 1990 was
Materials and methods defined as Aug 1989 to July 1990 and the year 1991 as
Aug 1990 to July 1991. Kid mortality included only
Site deaths occurring within one year of birth.

This study was carried out in the Ganawuri district of

Plateau State. It is located 9° N and 8° 35' E with an Results
average annual rainfall of about 1400 mm, 90% of which
falls between April and October. Figure 1 shows the weight changes in adult non
pregnant does during a typical period of restricted
Animals movement in the wet season. A similar pattern was
observed during subsequent wet seasons. As the wet
Flocks of goats belonging to smallholder farmers were season progressed, weight losses occurred when feed
used. Their histories were obtained and they were tagged material decreased. Significant differences (P<0.05)
at the beginning of the study. They were vaccinated occurred between the two groups at the end of the wet
annually against peste des petits ruminants (PPR) and season. The kidding intervals were 259 ± 19.8 and 269
dipped fortnightly to control lice and mange. Births, ± 12.4days and were not significantly different
birth types, sales and deaths were recorded as and when (P<0.05). The litter sizes were significantly affected
they occurred. The animals roamed freely in the village (P<0.01) by parity. About 68% of all kids resulted from
during the dry season consuming browses, grasses, crop multiple births. Kidding was all year round but
residues and household wastes. In the wet season, as the cumulative mortality (Table 1) was significantly
crops were being planted, goats were tethered alongside (P<0.05) affected by season up to one year of age. The
the road, near the homestead or allowed to graze on unadjusted cumulative mortality to one year was 17.5
enclosed fallow land. In addition, some flocks were and 35.8% in 1990 and 1991, respectively. Viability,
allowed to graze fodder banks of Stylosanthes. Animals improved by access to FB, became significant at one
are restrained between August and October. The first year of age.
crop is maize; after it is harvested, the animals are The weights of kids to one year (Table 2) were
allowed more freedom but are kept away from significantly affected (P<0.05) by season of birth and
late-maturing crops such as millet. At the start of the
experiment, there were eight farmers who shared a Figure 1. Weight changes of adult WAD female goats in
communal fodder bank where stylo was used for land Ganawuri in the wet season (August-October).
reclamation. Natural pasture was used by 10 farmers. By
1991 there were 1 1 fodder banks while 26 farmers used Weight changes (g)
natural pasture. The animals increased from 157 (56 1000
does) in August 1989 to 324 (136 does) and 395 (157
does) in August 1990 and 1991, respectively.
500
Pasture development
Stylosanthes pasture was established on small plots at a
seed rate of 1 2 kg/ha with 1 50 kg single superphosphate
fertilizer/ha. The size of these plots varied and were -500
determined by a technique described by Lawrence
(1991). Random quadrants (lm ) of pasture were LSD (0.05)
clipped, weighed and sorted into stylo, grasses and forbs -1000
for botanical composition and dry matter yield Fodder bank N=67
determinations. Tethered N = 81
-1500
Description of data and statistical analysis
-2000 -4
Births, birth types, weights, weighing dates and parity 15 30 45 60 75 90
of does were recorded. Kidding intervals were Trial period (days)
computed. Data analysis was done using the General

180 Small Ruminant Research Network Workshop

 Wet season supplementation of West African Dwarf etc..

Table 1. Least squares ineans ± SE of cumulative mortality (%) of West African Dwarf goat kids from birth to 365 days in
Ganawuri with access to fodder bank or natural pasture
Age (Days)
Variables No 30 days 90 days 150 days 365 days
Overall 414 4.8 ±1.92 11.3 ±3.71 15.3 ±3.41 24.7 ±4.09
Year of birth
1990 68 5.42 ± 2.94 12.9 ±4.48 16.0 ±5.72 9.5 ± 8.80
1991 346 4.2 ±1.82 9.6 ±2.80 14.6 ±3.71 39.8 ± 6.70
Season of birth
Dec-Feb 60 1.3 ±3.21 3.7 ± 4.97 2.5 ±6.32 1.71 ±0.53
Mar-May 122 1.4 ±2.38 5.4 ±3.65 11.5 ±4.71 17.9 ±8.55
Jun-Aug 122 9.6 ± 2.48 19.3 ±3.83 22.1 ±4.92 42.2 ±8.72
Sep-Nov 110 6.9 ± 2.65 16.7 ±4.03 22.0 ± 6.32 36.9 ± 8.50
Sex
F 208 3.5 ±2. 16 11.0 ±3.29 15.2 ±4.36 21.4 ±6.98
M 206 6.0 ±2.25 11.4 ±3.46 15.5 ±4.49 27.9 ± 6.72
Birth type
Single 132 3.4 ±2.04 8.8±3.16 15.1 ±4.07 15.6 ±6.12
Twin 262 4.2 ±1.61 10.5 ±2.45 14.1 ±3.17 26.8 ±4.65
Triplet 20 6.7 ±4.81 14.4 ±7.29 16.8 ±9.90 31.6 ±15.78
Treatment
FB 192 3.2 ± 2.35 9.3 ±3.61 11.5 ±3.21 16.2 ±7.34
NP 222 6.4 ± 2.07 13.2 + 3.15 17.6 ±3.83 33.1 ±6.48
FB = Fodder bank; NP=Natural pasture; F = Female; M=Male; *P<0.05.

birth type. Prior to introducing the animals to grazing in accompanied by abortions in many does and could be a
August, the mini fodder banks produced, on average, 2.6 case of vaccine failure. The overall mortality in
± 0.87 tonnes dry matter (DM)/ha while the natural Ganawuri of 24% is comparable to that found in Abet
vegetation produced 2.4 ±0.31 tonnes (DM)/h. The (Bayer, 1986b). Bayer (1986a) estimated that the
botanical composition of pasture showed 30 - 50% intensity of land cultivation in the zone ranged from 15
stylo, 40 - 60% grasses while the rest were forbs. to 60%, with the highest intensity in the Ganawuri area.
Comparisons between treatments were not possible as The farmers' practice is to allow periods of fallow to
the flock sizes varied between five and 30. Plots of encourage nutrient recycling and restore soil fertility.
pasture varied between 0.04 and 0.36 ha and the stylo With increasing pressure on land, fallow periods cannot
content of forage also differed. be extended thus soil fertility declines and the
production system deteriorates. Goat production
involving the use of mini fodder banks in the wet season
can easily be integrated into the existing agricultural
Discussion system for increased productivity.
In the subhumid zone, ILCA scientists showed that
To make goat-keeping profitable, offtake from a flock fallow land sown to legume pasture regained soil
should be high. There was a prevalence of multiple fertility much more quickly (Mohammed-Saleem and
births which could be exploited through management to Otsyina, 1986). There were also increased grain yields
ensure greater survival to weaning. The results of this when fodder banks were cropped three to four years after
study suggest that supplementation of the does in the wet establishment (Mohammed Saleem and Otsyina, 1986)
season leads to reduced weight losses. This could reduce which made the concept attractive to crop farmers.
kid mortality in the wet season through improved Farmers have exploited the benefit of incorporating
nutrition. However, the wet season is characterized by forage legume into their cropping system by putting
the presence of diseases and a high helminth burden
goats in these mini fodder banks during the wet season.
(Smith et al, 1986). Drugs could reduce kid mortality.
The least squares mean mortality of 25% recorded in this This is borne out by the number of farmers adopting the
study is lower than the 40 ± 50% recorded in the humid fodder bank concept: 8 (1989), 17 (1990) and 19 (1991).
zone (Reynolds et al, 1987). The almost two-fold This study clearly shows that the use of fodder banks
increase in unadjusted mortality means between 1990 reduces weight losses in non-pregnant mature does.
and 1991 was attributed to the annual vaccination Weight losses in adult females are known to influence
against PPR between June and July 1991. This was reproductive performance.

Small Ruminant Research Network Workshop 181

O.A. Ikwuegbu and S. Ofodile

Table 2. Least squares means ± SE of the weight (kg) of West African Dwarf kids in Ganawuri with access to fodder or
natural pasture.
Effect No Birth 30 days 90 days 150 days 365 days
Overall 420 1.6 + 0.13 3.1+0.74 5.6 ±1.25 7.7 ±1.58 12.0 ±2.14
Year of birth
1990 68 1.8 + 0.05 3.1 ±0.14 5.7 ±0.24 8.1 ±0.42 12.4 ±0.98
1991 352 1.5 ±0.03 3.1 ±0.08 5.5 ±0.15 7.2 ± 0.29 11.6 ±0.74
Season of birth
Dec-Feb 60 1.8 ±0.06 3.4 ±0.15 6.3 ±0.26 8.1 ±0.48 11.6 ±0.98
Mar-May 120 1.6 ±0.04 3.3 ±0.10 5.9 ±0.19 7.2 ±0.34 13.0 ±0.94
Jun-Aug 126 1.6 + 0.04 2.8 ±0.11 5.1 ±0.21 7.3 ± 0.34 12.0 ±0.98
Sep-Nov 112 1.6 ±0.05 2.8 ±0.12 4.9 ± 0.43 7.8 ±0.43 11.3 ±0.99
Sex
F 213 1.6 ±0.04 3.1 ±0.9 5.5 ±0.17 7.6 + 0.33 12.5 ±0.70
M 207 1.7 ±0.04 3.1 ±0.10 5.6 ±0.18 7.6 + 0.34 11.4 ±0.66
Birth type
Single 134 1.8 ±0.03 3.7 ± 0.09 6.0 ±0.18 8.1 ±0.31 12.1 ±0.60
Twin 265 1.6 ±0.03 3.1 ±0.08 5.3 ±0.14 7.4 ± 0.23 11.4 ±0.48
Triplet 21 1.5 ±0.08 2.5 ± 0.21 5.3 ± 0.38 7.3 ± 0.76 12.4 ±1.61
Treatment
FB 195 1.6 ±0.04 3.1 ±0.11 5.4 ±0.19 7.6 ±0.32 11.9 ±0.72
NP 225 1.6 ±0.04 3.1 ±0.09 5.8 ±0.17 7.9 ± 0.30 12.0 ±0.68
FB = Fodder bank; NP = Natural pasture; F = Female; M = Male; *p<0.0.5.

To make the overall production system sus Jahnke H E. 1982. Livestock production systems and livestock
tainable, livestock-crop activities have to be integrated. development in tropical Africa. Kieler
A trade-off in terms of reduced crop yield and Wissenchaftsverlag Vauk, Kiel, Germany. 253 pp.
availability of high quality forage for feeding goats is Kapu M M. 1975. The natural forages of northern Nigeria. I.
inevitable. In this study the number of goats per flock, Nitrogen and mineral composition of grasses and
the sizes of the fodder banks and the quality of pasture browses from Northern Guinea Savannah and standing
all varied. Further investigation to determine the hays from different savannah zones. Nigerian Journal of
Animal Production 2:235-246.
optimum number of goats per mini fodder bank would
be worthwhile in order to sustain this agricultural Lawrence PR. 1991. How big is that field? ILCA Newsletter
system. 10(4):7-8. ILCA (International Livestock Centre for
Africa), Addis Ababa, Ethiopia.
Mba A U. 1979. Meat production in Nigeria. Prospects and
Acknowledgement problems. National Conference on Food and Nutrition
Policy for Nigeria in the 1980's. Proceedings, Think
The authors acknowledge the diligence of the field staff Tank/University of Tbadan, Nigeria. pp. 89-97.
and the co-operation of the smallholder farmers. This
Mohamed Saleem M A and Otsyina R M. 1986. Grain yields
project was primarily funded by the National Livestock of maize and nitrogen contribution following
Projects Division Kaduna, Nigeria and this support is Stylosanthes pasture in the Nigeria subhumid zone.
highly appreciated. Experimental Agriculture 22:207-214.
Reynolds L, Atta-Krah A N and Francis P A. 1987. A strategy
References for improving goat productivity under village production
systems in the humid tropics. In: Smith О В and Bosman
Bayer W. 1986a. Agropastoral herding practices and grazing H G (eds), Goat production in the humid tropics.
behaviour of cattle in the subhumid zone of Nigeria. Proceedings of a workshop at the University of Ife,
ILCA Bulletin 24:8-13. ILCA (International Livestock Ile-Ife, Nigeria, 20-24 July 1987. Pudoc (Centre for
Centre for Africa), Addis Ababa, Ethiopia. Agricultural Publishing and Documentation),
Wageningen, Netherlands. pp. 29-37.
Bayer W. 1986b. Traditional small ruminant production in the
SAS (Statistical Analysis System). 1987. Guide for personal
subhumid zone of Nigeria. In: von Kaufmann R, Charter
computers Version 6 edition. SAS Institute, Cary, North
S and Blench R (eds), Livestock systems research in
California, USA. 1028 pp.
Nigeria's subhumid zone. Proceedings of the Second
ILCA/NAPRI Symposium held in Kaduna, Nigeria, 29 Smith О В, Ayeni A O, Onawunmi J and Ademosun A A. 1 986.
October-2 November 1984. ILCA (International Helminths and protozoan parasites of dwarf goats in
Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. south-western Nigeria. Bulletin of Animal Health and
141-166. Production in Africa 34:67-70.

182 Small Ruminant Research Network Workshop

 Performance of weaner lambs fed maize leaves
and napier grass
1 • 2
A.N. Abate and A. Abate

Kenya Agricultural Research Institute, P.O. Box 57811, Nairobi, Kenya

2
Department ofAnimal Production, University ofNairobi
P.O. Box 29053, Nairobi, Kenya

Abstract In Kenya napier grass has been identified as being

productive, persistent and easy to harvest (Mwakha,
Maize leaves were compared to napier grass in 1972; Boonman, 1977; Karanja, 1984). It is also
experiments designed to measure their nutritive value palatable and gives reasonable growth rates and milk
and their effect on the growth of weaner lambs under yields in cattle (Abate et al, 1992). However, under
stall feeding conditions. In the first trial, one leaf per smallholder conditions, the growing of pure fodder may
plant per week was systematically plucked about three be limited by the farmer's need to meet his own food
months post-emergence; in the second, green leaves requirements. Defoliated maize leaves for fodder have
were defoliated from the bottom of the plant about 4—5 been integrated into crop and animal production trials
months after germination. In both cases, animals fed the with satisfactory results (Abate 1981, 1982, 1983). A
maize leaves gained significantly more weight (P<0.01) study was undertaken to record the effects of maize leaf
than those fed napier grass. The effect of leaf defoliation and napier grass diets on Corriedale sheep. This paper
on forage dry matter (DM) productivity and grain yield reports the findings.
is discussed.

Performances des agneaux Materials and methods

sevrés nourris avec des feuilles Weaner lambs of the Corriedale breed were allocated to
de maïs et de l'herbe à éléphant two groups of four. The lambs were approximately five
months old and had an average weight of 21 kg. They
Résumé were dewormed prior to the experiment. One group was
the treatment group and received maize leaves ad
Deux expériences ont été mises en place pour libitum. The control group was offered napier grass ad
mesurer la valeur nutritive des feuilles de maïs et de libitum. The trial lasted for 10 weeks. Maize leaves were
l'herbe à éléphant et comparer leurs effets sur la defoliated from a plot of hybrid maize plants about three
croissance d'agneaux sevrés alimentés en bergerie. months post-emergence at a frequency of one leaf per
Dans le premier essai, une feuille par plante était plant per week. A second plot remained undefoliated as
a control. Both the maize leaves and napier grass were
cueillie toutes les semaines trois mois environ après la
chopped before being fed to the animals. Group dry
levée; dans le second essai, lesfeuilles vertes du bas de
matter (DM) intake was recorded daily and each sheep
la plante ont été cueillies 4 à 5 mois après la was weighed once a week. Clean water and mineral licks
germination. Dans les deux cas, la croissance pondérale
were freely available. Feeds offered and refusals were
des animaux nourris avec les feuilles de maïs était sampled weekly, dried in a draught oven and milled
supérieure (P<0,01) à celles des animaux nourris avec through a 2 mm sieve for laboratory analysis.
l'herbe à éléphant. L'effet de la défoliation sur le In an experiment carried out the following year, the
rendement en matière sèche fourragère et en grains de digestibility of maize leaves defoliated about 4-5
maïs est examiné. months after germination was compared to that of napier
grass. Maize leaves were obtained by picking green
leaves from the bottom of the plant upwards, leaving
Introduction only the four nearest the cob. The maize leaves and
napier grass were chopped before being fed ad libitum
Grazing lands in the agriculturally rich highlands and to the animals. The wethers were fed for a preliminary
coastal subhumid zones of sub-Saharan Africa are period of approximately three weeks, then there was a
diminishing, mainly due to the increase of human collection period of seven days. Composite samples
settlements. This is increasingly causing a scarcity of collected over the seven days were oven-dried at 60°C
fodder for small ruminants. and ground for analyses. Individual animal weights were

Small Ruminant Research Network Workshop 183

A.N. Abate and A. Abate

recorded at the beginning and end of the trial. Data from Table 2 shows the results of the digestibility
both experiments were subjected to statistical analysis. experiment. Both defoliated maize leaves and napier
grass were similar in chemical characteristics, although
the latter had more fibre. DM digestibility was also
similar. However, intake of DM was significantly
Results (P<0.001) higher in maize leaves than in napier grass.
All wethers fed on maize leaves gained weight while
Results of the performance trial are given in Tables 1 those fed napier grass lost weight; the differences in
and 2 and in Figure 1. The maize leaves were similar in weight changes were significant (P<0.05).
DM content to the napier grass but higher in crude
protein (CP) levels (Table 1 ). The concentration of CP Table 2. Nutritional characteristics of defoliated maize
in both forages declined with age (Figure 1 ) at the rate leaves and napier grass when fed to wethers.
of 1.1% and 0.4% per week for defoliated maize leaves
and napier grass respectively. Defoliated maize leaves Defoliated Napier
maize leaves grass
gave a higher DM intake and a daily gain in weight that
was roughly three times higher than napier grass. Dry matter, % 34.5 33.9
Efficiency of feed conversion was more than two times Crude protein, % 9.6 8.2
higher in the lambs fed maize leaves than in those fed
napier grass. Maize defoliation produced approximately Neutral detergent fibre, % 66.7 73.0
1.5 tonnes of leaf DM per hectare without significantly Dry matter digestibility, % 54.6 56.1
(P>0.05) reducing grain yield (4.90 vs 5.31 1161.0a 662.0b
tons/hectare). Dry matter intake (g/day)
Average daily gain, g +67.0a ^*3.0b
Table 1. Chemical composition and effect on weaner lamb ab = Values along a row bearing different letters are significantly
performance of defoliated maize leaves and napier different (P<0.001 ) or (P<0.05).
grass.
Defoliated Napier
maize leaves grass
Dry matter, % 22.4 21.3 Discussion and conclusion
Crude protein, % 16.6 8.6
The results of this experiment have shown that
Dry matter intake (g/day) 887.0 683.0 defoliated maize leaves can be a source of quality feed
Average daily gain (g/day) 136.0a 48.0b for ruminants. The leaves proved to be more acceptable
and of higher nutritional value than napier grass. This is
Feed conversion efficiency 6.5 14.1 significant because for a given weight of DM more meat
ab = Values in a row bearing different superscripts are significantly can be produced from animals fed maize leaves than
different (P<0.05). from those fed napier grass. Earlier, Kayongo-Male and
Abate (1982) had shown that fresh maize leaves were
nutritionally superior to hay. Stripping one leaf per plant
Figure 1. Changes in crude protein content of defoliated per week spread the use of the leaves over a period of
maize leaves and napier grass. 10 weeks although at the cost of quality. Plucking more
leaves per plant per week produces better quality feed
but for a shorter duration (Abate, 1983). The effect of
Crude protein
defoliation on grain yield is important as maize forms
25
the staple diet of most countries in East Africa. A
Napier grass reduction in grain yield of about 8% is considered small
Maize leaves and the liveweight gain would compensate for the loss.
20- Lower reductions in grain yield of 2.6% have been
reported before (Abate, 1 98 1 ).

15-

References
10 Abate A N. 1981. Performance of weaner lambs fed on
systematically defoliated maize leaves. Annual Report,
Animal Production Research Department, KARI (Kenya
5 Agricultural Research Institute), Muguga, Kenya.
Abate A N. 1982. Performance of Corriedale lambs fed on
maize leaves defoliated at different stages ofgrowth and
н r —г at different frequencies. Annual Report, Animal
5 6 10 Production Research Department, KARI (Kenya
Weeks Agricultural Research Institute), Muguga, Kenya.

184 Small Ruminant Research Network Workshop

 Perform, of weaner lambs fed maize leaves, Napier grass

Abate AN. 1983. Voluntary dry matter intake and digestibility tropical pastures in East Africa (Part II). Ministry of
of maize leaves and napier grass by young wethers. Agriculture, Nairobi, Kenya.
Annual Report, Animal Production Research
Department, KARI (Kenya Agricultural Research Karanja G M. 1984. The effects offrequency and height of
Institute), Muguga, Kenya. cutting on tillering, herbage yield and nutritive value of
four varieties of napier grass. MSc thesis, University of
Abate A, Dzowela В H and Kategile J A. 1992. Intensive Nairobi, Nairobi, Kenya.
animal feeding practices for optimum feed utilisation. In:
Kategile J A and Mubi S (eds), Future of livestock Kayongo-Male H and Abate A N. 1982. Role of systematic
industries in East and southern Africa. Proceedings of a maize (corn) defoliation in small ruminant production.
workshop held at Kadoma Ranch Hotel, Zimbabwe, In: Proceedings of the small ruminant-CRSP Workshop,
20-23 July 1992. ILCA (International Livestock Centre March 15th, 1982, Nairobi, Kenya. pp. 145-150
for Africa), Addis Ababa, Ethiopia. pp. 9-19.
Mwakha E. 1972. Effect of cutting frequency on productivity
Boonman J G. 1977. Fodder farming in East Africa. A of napier and guatemala grasses in western Kenya. East
preliminary report on the status of the development of African Agricultural and Forestry Journal 37:206-2 1 0.

Small Ruminant Research Network Workshop 185

 Rumen degradation ol elephant grass supplemented with ...

Rumen degradation of elephant grass supplemented with

graded levels of perennial peanut by West African Dwarf sheep
B. Tchinda, D. Wegad and R. M. Njwe

Department ofAnimal Production

INADER, Dechang University Centre, Cameroon

Abstract dégradabilité dans le rumen a été étudié dans une

expérience factorielle avec quatre traitements, cinq
The effect of 0, 10, 20 and 30 per cent supplementation périodes d'incubation, trois animaux et deux
of elephant grass (Pennisetum purpureum) with échantillons dans des sacs en plastique par animal et
perennial peanut (Arachis glabrata) on rumen par période. Trois brebis de la race naine d'Afrique
degradability of dry matter was investigated in a occidentale dont le rumen a étéfistulé, âgées de 24 mois
factorial trial involving 4 treatments, 5 incubation et pesant en moyenne 18 kg ont été utilisées. Les
periods, 3 animals and 2 nylon-bag samples per animal échantillons ont été incubés dans leur rumen pendant 6,
per incubation period. Three rumen-fistulated West 12, 24, 48, et 72 heures.
African Dwarf ewes of 24 months of age and weighing La dégradation de la matière sèche et de la matière
an average of 1 8 kg were used. Samples were incubated
organique augmentait significativement (P<0,05) avec
in the rumen of sheep for 6, 12, 24, 48 and 72 hours. l 'accroissement de la proportion de légumineuse de la
Dry and organic-matter degradation (DM and OM) ration de 0 à 20%. L'augmentation de cette proportion
increased significantly (P<0.05) when the legume à 30% n 'entraînait aucun accroissement significatifde
fraction in the ration was raised from 0 to 20 per cent. la dégradabilité de la matière sèche et de la matière
Further increase in legume fraction to 30% of the ration organique. Toutefois, celle-ci augmentait significative
did not result in significant increases in DM and OM ment (P<0,05) dans toutes les rations lorsque le temps
degradability. There was also a significant (P<0.05) d'incubation passait de 6 à 48 heures. Après 48 heures
increase in DM and OM degradability of all rations d'incubation, la dégradabilité de la matière sèche était
when incubation time in the rumen was increased from de 34% pour l 'herbe à éléphant seule puis de 42, 43 et
6 to 48 hours. Degradability of DM was 34% for elephant 46% pour les rations contenant respectivement 10, 20 et
grass alone and 42, 43 and 46%, respectively, for rations 30% de légumineuse. Après 72 heures d'incubation, il n 'y
containing 1 0, 20 and 30 per cent legume after 48 hours' avait pas d'augmentation significative de la dégrada
incubation. After 72 hours' incubation, there was no bilité de la matière sèche et de la matière organique.
significant increase in DM and OM degradibility.
The estimates of the immediate soluble fraction (a) Le taux de la fraction immédiatement soluble (a)
and the rate of degradability (c) increased with increase et la dégradabilité (c) augmentaient avec l'accrois
in the level of legume in the ration. The lag time sement de la proportion de légumineuse dans la ration.
necessary for the microbial population to invade the feed Enfin, le temps nécessaire pourpermettre à laflore micro
also reduced as the level of legume in the diet was raised. bienne d 'envahir les aliments diminuait avec l 'augmen
tation de la proportion de légumineuse dans la ration.

Dégradation de l'herbe à
éléphant complémentée avec des Introduction
niveaux croissants de fourrage The most widely available low-cost feeds for ruminants
in most tropical countries are usually native pasture,
d'arachide pérenne dans le crop residues and, to a lesser extent, agro-industrial
by-products. Legumes in the farming system may
rumen de moutons de la race constitute an important feed resource of fermentable
naine d'Afrique de l'Ouest nitrogen and of by-pass protein. When fed to ruminants,
legumes tend to increase the efficiency of utilisation of
the basal diet, which is usually a low nitrogen pasture or
Résumé a crop residue, at low levels of supplementation.
Legume supplements given at levels below 30 per cent
L'effet de la complémentation de 0, 10, 20, ou 30% of dietary dry matter have a catalytic effect on feed
d'herbe à éléphant (Pennisetum purpureum) avec du utilisation. They ensure a more efficient environment
fourrage d'arachide pérenne (Arachis glabrata) sur la for digestion of cell-wall carbohydrates, perhaps by

Small Ruminant Research Network Workshop 187

Tchinda В., Wegad D. and Njwe R. M.

providing micronutrients such as peptides, amino acids, in cold water for 25 minutes and dried to constant weight
minerals and vitamins which increase fungal biomass at 60°C in a forced air laboratory oven and weighed. Six
and/or the rate of bacterial colonisation of the fibre. The nylon bags containing samples of each ration were
potential of the diet to meet the requirements of the soaked in water for 30 minutes and washed in the same
animal for amino acids, glucogenic precursors and long manner and dried to constant weight.
chain fatty acid depends on the pattern of fermentation The difference in weight of dry matter in nylon bag
and on dietary protein lipids and starch that escape before and after rumen incubation represented material
fermentation and are digested in the intestines. The
degraded in the rumen. Similarly, the difference in
extent to which the protein in a supplement escapes the
organic matter weight before and after incubation was
rumen is partly a function of its rate of degradation in
equivalent to OM degraded in the rumen.
the rumen (Preston and Leng, 1987). The objective of
this trial was to study the effect of graded levels of The degradation characteristics of each ration were
perennial peanut (PP) supplement fed with elephant described using the following exponential equation
grass (EG) on the degradability of dry and organic (Orskov and McDonald, 1979),
matter by West African Dwarf sheep. Р = а + Ь(1-е~°')
where a, b and с are constants
a - intercept or immediate soluble fraction
Materials and methods b = insoluble but potentially degradable material
at time t
Three female West African Dwarf sheep fitted with
с = rate of degradation per hour
rumen cannulae were used for the study. Their average
age and body weight were 24 months and 18 kg, P = level of degradation at time t
respectively. The animals were fed four experimental
rations consisting of fresh EG (Pennisetum purpureum) Statistical analysis
only, or supplemented with PP (Arachis glabrata,
ex-Florida) at 10, 20 and 30 per cent. The experiment was a4x5x3x2 factorial design (4
rations, 5 incubation intervals, 3 sheep and 2 nylon bags
Table 1. Composition (%) of dried forages used in per animal per incubation period). Statistical analysis
constituting rations incubated in the rumen. was carried out according to the procedures of Snedecor
and Cochran (1976) for analysis of variance. The
Proximate components PP EG Duncan's multiple range test was used to test significant
Dry matter 92.03 92.10 differences between treatments.
Organic matter 86.18 86.25
Crude protein 17.80 12.50 Results and discussions
Crude fibre 22.13 32.50
Ash 3.82 13.75 The degradation of dry matter of the rations (Table 3)
increased significantly (P<0.05) as a result of
supplementation of EG with PP. Total dry matter
Table 2. Composition of experimental diets incubated in the degraded increased from 56.75% in the ration consisting
rumen of sheep. of EG only to 69.47% in the 30% PP ration.
These results confirm findings by Mosi and
Proximate EG:PP ratio
Butterworth (1983) that the addition of 30% legume
components 100:00 90:10 80:20 70:30
forage to a grass basal diet increased total dry matter
Dry matter 92.10 92.06 92.04 92.05 degradability from 40 - 50% to 60%.
Organic matter 86.25 86.20 86.19 86.20 A comparison of DM degradation of the three
Crude protein 12.50 13.03 13.57 14.11 rations containing legume supplement after 6 and 12
Crude fibre 32.50 31.35 30.10 29.22 hours' incubation in the rumen indicated a significant
Ash 13.75 13.80 13.79 13.80 difference (P<0.05) between the rations containing 10
and 20% legume; but no significant difference between
Fresh EG and PP were dried in a forced air those having 20 and 30% legume. DM disappearance
laboratory oven at 60°C for 48 hours, milled separately was similar for all rations having legume supplement at
to pass through a 2 mm screen, and mixed in the 24, 48 and 72 hours of incubation.
following gramineae/legume proportions, 100:0, 90: 10, After 72 hours of rumen incubation more than 57
80:20 and 70:30. Four grams of ration was placed in a per cent of potentially degradable amount of DM in the
nylon bag (11x7 cm, 20 micron porosity) and sealed. control ration had disappeared, whereas for the rations
The composition of EG, PP and their combinations containing legume more than 70 per cent had dis
incubated in the rumen are shown in Tables 1 and 2, appeared.
respectively. The method of incubation in the rumen of OM degradation in all rations is shown in Table 4.
sheep was as described by Orskov et al (1980). Nylon After 12, 24, 48 and 72 hours of rumen incubation all
bags were incubated in the rumen for 6, 12, 24, 48 and rations with legume supplement had significantly
72 hours after which they were withdrawn and washed (P<0.05) higher OM degradation than EG alone.

188 Small Ruminant Research Network Workshop

 ' с 3a 3 a. CD CO Ш Г1 5' 3 Ф CD CD 01
CO
01
0 0 C/0% Ï

-e'c
NB:
alevel
b
P=a+b(
equation
the
iwin
I
)
dP
of
time
intercept
is
fieibut
soluble
b
pthe
here
cmdrat
gt;
oti0
сaa,
nndearorstat
acgst;
oreedtrlaniuatbiodslnaeo;lbnyle -e'u)
NB:
aiin
b P=a+b(
equation
the
0
P
dwhere
of
level
time
intercept
is
fsithe
iebut
b pctime
dmrot;
at
nсaa,
gndesalat
rortt;
agocresuedlratnbiuatlbdsoiteanbo;lnye
5
79 0.00 0.00 5.75 0.09 0.00 0.00
Ti0 lag 00 Ti0 lag

a+b 53.03 56.78 55.03 55.88 a+b 60.05 66.94 64.00 65.30

Residual Residual
0an 08.76 9.46 3.00 9.08 0an 05.83 00.87 0.84 9.40
sq. sq.

0.0080 0.060 0.0080 0.0086 0.0085 0.0086 0.0308

0.053
с с

50.05 50.04 40.56 38.57 59.60 60.05 48.39 48.30

b b

0.06 4.64 07.30 0.43 5.89 05.63 07.00

a 00.0 a

30.46 39.50 40.70 40.07 35.68 46.30 46.83 46.90

0 4.
Table
sthe
level
of
Effect
loss
ousheep
in(g/100g
DM).
rpgalonrumen
neimce-nmtatioenr 0
3.
Table
level
of
Effect
ssheep
loss
dry-matter
the
in
(g/lOOg
uDM).
p lonrumenementation

09.09 36.63 37.06 38.08 34.08 40.08 40.77 45.53

04
00
48 00
04
48
Iperiod
(hours)
ncubation is
dthe
of
while
hour
fma+b
(erarate
gsaper
сxrciyatmdpiautomnbitoel)ne. Iperiod
(hours)
ncubation the
is
of
dwhile
hour
ma+b
fe(rate
argсper
sxrcaiytdmaiputominbtolen).

9.83 05.30 03.76 00.65 03.65 08.75 07.85

0.0

3.75 8.36 04.97 0.88 4.43 9.80 08.60 00.07

0.08 6.08 00.00 00.54 0.48 7.07 03.44 03.46

6 6

loss—
Washing loss-
Washing
(g/lO gDM) (g/lO gDM)
7.90 8.00 00.00 8.46 9.33 00.09 00.75
7.0

Supratele0ntation Supratele0ntation

Treat0nt
PP 0% 00% 30% Treat0nt
PP 0% 00% 30%
0% 0%

ел 3 R о о
Tchinda В., Wegad D. and Njwe R. M.

OM degradation attained a maximum increased Conclusion

value of 1 3 per cent especially with the ration containing
30% PP. Similar results were reported by Dixon et al. It may be concluded from this study that inclusion of PP
(1979) when supplements of Leucaena leucocephala in the diet of sheep fed EG up to a level of 20 per cent
and rice polishing were fed with sisal pulp to cattle. This results in a significant improvement in DM and OM
was associated with increased microbial synthesis. degradation in the rumen.
There was no significant difference between the
OM degradation of the control ration and the one
containing 10 per cent legume after 6 hours of rumen
incubation. Similarly, there was no significant
References
improvement in OM disappearance in the rumen when Dixon R M, Priego A, Eliot R and Preston T R. 1979. Studies
the level of legume supplement in the ration was raised on the digestion in the forestomachs of cattle of a diet
from 20 to 30 per cent. A significant improvement in based on sisal pulp supplemented with Leucaena
OM degradation was evident after 24 and 48 hours leucocephala and rice polishings. Tropical Animal
incubation of feed samples in the rumen. Extending Production 4(3):281-286.
incubation time to 72 hours did not result in a significant Mosi А К and Butterworth M H. 1 983. Utilisation de Trifolium
increase in OM degradation. tembense pour compléter les rations des ruminants à base
des résidus agricoles. CIPEA actualités 2(4):5-6.
It was also observed that estimated maximum
CIPEA (Centre international pour l'élevage en Afrique),
potential degradable OM in the rations increased with Addis-Abeba (Ethiopie).
increase in the level of legume supplement in the ration.
After 72 hours of rumen incubation 58% of maximum 0rskov E R and McDonald I. 1979. The estimation of protein
degradability in the rumen from incubation
degradable amount of EG organic matter in the control measurements weighted according to rate of passage.
ration had disappeared whereas for the other treatments Journal of Agricultural Science (Cambridge)
more than 69% had been degraded. 92:499-503.
In the present study supplementation of EG with 0rskov E R, Hovell F D de В and Mould F. 1980. The use of
graded levels of legume resulted in an increase in the the nylon bag technique for the evaluation of feedstuffs.
rate of DM and OM degradation in the rumen. For DM Tropical Animal Production 5: 1 95-2 1 3.
it increased from 0.0182 to 0.0286 per cent per hour as
Preston T R and Leng R A. 1987. Matching ruminant
the level of legume in the ration was raised from 0 to 30 production systems with available resources in the
per cent, whereas for OM it increased from 0.0185 to tropics and sub-tropics. Penambul Books, Armidale,
0.0308 per cent per hour. Also, it was observed that as Australia. 245 pp.
the level of legume in the diet was raised there was a Snedecor E W and Cochran W G. 1976. Statistical methods.
reduction in the lag time necessary for the microbial Sixth edition. Iowa State University Press, Ames, Iowa,
population to attack the feed. USA.

190 Small Ruminant Research Network Workshop

 Metabolisable energy requirements for maintenance and
growth of castrate Zimbabwean (Matabele) goats offered
diets differing in roughage content either ad libitum
or restricted
P.R. Hatendi1, L.R. Ndlovu, T. Smith3 and A.N. Said4

Grasslands Research Station, P. Bag 3701, Marondera, Zimbabwe

Department ofAnimal Science, University ofZimbabwe
P.O. Box MP 167, Mount Pleasant, Zimbabwe
3
Matopos Research Station, [Link] K5137, Bulawayo, Zimbabwe
International Livestock Centre for Africa
P.O. Box 5689, Addis Ababa, Ethiopia

Abstract calculées étaient respectivement de 482 kJ/kg 0,75 par

jour et de 34,2 kJ/g.
Feed intake and liveweight changes of 53 stall-fed
Matabele goats were monitored for 94 days. They were
offered diets containing 9.9, 11.3, 12.1 or 12.5 MJ
metabolisable energy (ME) per kilogram dry matter,
Introduction
either ad libitum or restricted. Prediction of an animal's response to nutrients is central
ME intake (ME¡) was regressed against liveweight to the science and practical application of nutrition.
change and liveweight change was regressed against Prediction implies extrapolation (AFRC, 1991), the
ME¡. Calculated estimates of ME requirements for inherent assumption being that the animals used in
maintenance (MEm) and liveweight change (MEg) from determining response are representative of the general
the regression of MEi and liveweight change were 519 population to which the results will be applied (Morton
kJ/kg ' per day and 25.2 kJ/g, respectively. For the andRidgeman, 1977).
reversed regression, the estimates for MEm) and MEg) There are few published reports on the energy
were 482 kJ/kg ' per day and 34.2 kJ/g, respectively. requirements of goats. Recommendations used have
often been extrapolated from sheep without validation
(NRC, 1 98 1 ; Prieto et al, 1 990). It is notable that in those
Besoins en énergie méiabolisable experiments where direct comparisons have been made
between goats and sheep, maintenance energy (MEm)
pour l'entretien et la croissance values differ. Mohammed and Owen (1980) reported
de caprins castrés du Zimbabwe MEm values for mature and castrate sheep and goats of
434 v 301 kJ ME/kg075 (P>0.01), respectively.
(Matabele) recevant des rations Conversely, Alam et al (1991) reported MEm values of
à base de fourrage 440 v 370 kJ ME/kg075 (P>0.05) for growing, castrated
kids and lambs, respectively. Such results underscore
Résumé species differences and the need for efficient and
effective diet formulation to generate information
La consommation alimentaire et l'évolution specific to goats.
pondérale de 53 caprins Matabele alimentés en This paper reports estimates of the metabolisable
chèvrerie ont été observées sur une période de 94jours. energy (ME) requirements for maintenance and growth
Les rations, offertes à volonté ou rationnées, of castrate, indigenous Zimbabwean (Matabele) goats
présentaient des teneurs de 9,9, 11,3, 12,1 et 12,5 MJ fed cereal-based diets in confinement.
d'énergie métabolisable par kilo de matière sèche.
L 'énergie métabolisable ingérée a été ramenéepar
régression à l'évolution pondérale, et vice versa. La Materials and methods
valeur calculée de l 'énergie métabolisable requise pour
l'entretien et la croissance que l'on a obtenue à partir 64 castrate Matabele kids, initially about eight months
de la première régression était de 519 kJ/kg ' parjour old and weighing approximately 10 kg, were used in an
pour l'entretien et de 25,2 kJ/g pour la croissance. En experiment of 94 days duration. Animals were housed
ce qui concerne la régression inverse, les valeurs in wire mesh pens in an open-sided, asbestos-roofed,

Small Ruminant Research Network Workshop 191

P.R. Hatendi, L.R. Ndlovu, T. Smith and A.N. Said

Table 1. Treatment diet formulation (g/kg) and dry-matter composition (g/kg DM).

Diet
1 2 3 4
Formulation
Veld hay* 500 330 220 100
Cottonseed meal 142 118 102 84
Maize meal 238 432 558 696
Molasses 80 80 80 80
Urea 20 20 20 20
Salt 7 7 7 7
Limestone flour 5 5 5 5
Gypsum 5 5 5 5
Vitamins and minerals 3 3 3 3
Composition
Dry matter (g/kg) 902 903 902 905
Ash 106 71 63 54
Crude protein 200 187 194 182
Crude fibre 183 126 93 66
Ether extract 37 39 43 40
Metabolisable energy MJ/kg DM 37 39 43 40
For general description of mixed veld hay see Topps and Oliver (1978).
1 kg = 3x10 i.u. retinol, 35 g manganese, 35 g zinc, 0.6 g cobalt, 0.1 g selection and 0.12 g iodine.
***MAFF(1984).

concrete-floored barn. No bedding was provided and Table 2. Initial numbers and live weights of goats in the
accumulated manure was removed once a week. Each eight treatment groups.
pen had separate feed and water troughs. (Live weight (kg)
At the beginning of the experimental period
animals were allocated to one of four experimental diets Feeding level Diet n Mean SE
(Table 1) offered either ad libitum (A) or restricted (R). Ad libitum (A) 1 10 10.2 0.54
Initial numbers and live weights of goats in the eight 2 6 10.3 0.64
treatment groups are shown in Table 2.
3 6 10.1 0.83
Animals fed ad libitum were offered feed daily at
a level of 20% above the preceding day's intake. For 4 10 10.1 0.45
animals on the restricted feeding regimes, those in group Restricted (R) 1 10 10.1 0.59
Rl were offered air-dry feed daily at the rate of 6% of 2 6 10.5 0.73
metabolic body weight. Goats in groups R2, R3 and R4
3 6 9.9 0.80
were offered their respective diets at a level
iso-energetic to that offered to animals of the same 4 10 10.2 0.54
weight in treatment Rl. Feeding levels of animals on
restricted diets were adjusted once a week. Daily ME Metabolisable energy requirements for
intake was calculated by the difference between the ME maintenance and growth of the goats were estimated by
content of feed offered and that of feed refused. Water regression analysis of the ME intake, growth and mean
was freely available to animals at all times. Animals metabolic weight data.
were weighed at seven-day intervals and liveweight Two equations were fitted; equation 1, as
suggested by Abate (1989), and its converse, equation
changes subsequently determined using polynomial
2, as suggested by Zemmelink et al (1991):
regression analysis (Allen et al, 1983). Initial liveweight
and linear regression coefficients of each animal were ME intake (MEi) (MJ/kg 0.75 per day)
used to estimate live weight (LW) at the mid-point of = a + (b*ADG) Equation 1
discrete time periods. Metabolic weight (LW 0.75 ) was
derived from these estimates of LW. where;

192 Small Ruminant Research Network Workshop

 Metabolisable energy requirements for maintenance etc...

a = ME requirement for maintenance (MEm) Results

(MJ/kg per day),
Two animals on ad libitum feeding and nine animals on
b = ME requirement for growth (MEg) (MJ/kg) restricted feeding were removed from the experiment
because of refusal to eat the diets or because they died.
Final numbers in each treatment sub-group are shown in
ADG = average daily gain (g/kg per day)
Table 3. Initial live weight of animals which completed
the feeding trial in each treatment sub-group was
ADG = a + (b*ME,) Equation 2 different from that of the total number of animals
initially allocated to their sub-groups (Tables 2 and 3).
where; However, differences between treatment sub-groups
were not significant (P>0.05).
a = MEm = -ab" Performance of the goats tended to improve with
time (Table 3). Liveweight gains in this experiment had
b = MEg = b"' a high coefficient of variation (70%). Overall mean
liveweight gains of feeding levels R and A were 5 and
Animal performance data were analysed using the 47 g/day (P<0.001), respectively. Though energy
General Linear Model procedure of the SAS (1990) intakes of animals on restricted feeding were similar
statistical package. Model variance was apportioned to: (P>0.05), diets influenced liveweight gains (P>0.05),
Diet: 3 df, Contrast 1 indicating differences in nutrient utilisation. Gross
Feeding Level: 1 df, Contrast 2 conversion of energy to liveweight gain (LWG) for
Diet* Level: 3 df feeding level A ranked A2>A3>A 1 >A4 (82, 92, 1 22 and
Error: (n-l)-7)df 130 MJ ME/kg LWG, respectively) from days 68 to 94.
Total: n-1 df The diet x level interactions were not significant
Least squares means are presented for relevant (P>0.05) for all variables.
treatment comparisons. Standard errors were calculated Coefficients of energy intake regressed on
using the error mean square as an estimate of residual liveweight change (equation 1) and its converse
variance. Where appropriate, means were contrasted as (equation 2) for the period 33-68 days are shown in
described in SAS (1990). Regression analysis was Tables 4 and 5, respectively. Derived estimates of MEm
carried out using the SAS (1990) programme. and MEg for this period are shown in Table 6. Estimates

Table 3. Performance of goats offered diets differing in roughage content either restricted or ad libitum.

Restricted fed Ad libitum fed Contrast

Rl R2 R3 R4 Al A2 A3 A4 SE* 1 2
Number of goats 7 5 4 7 10 5 6 9
Initial live weight (kg) 10.9 11.0 10.2 10.8 10.2 10.7 10.3 10.1 0.24 ns ns
Overall liveweight (kg) change 0.91 -0.17 1.03 0.15 3.38 3.99 5.78 4.20 0.25 ns [Link]
Day 0-33
Liveweight change (gd" ) 10 -5 4 -14 43 27 53 26 5.4 ns P<0.001
DMI** (gd"1) 245 223 209 194 368 317 381 334 11.4 ns P<0.001
DMKgkg"07^"1) 40.2 37.5 36.3 33.3 60.2 51.6 60.5 56.1 1.28 ns P<0.001
MJMEiíkg^^d')ílO1) 3.8 4.0 4.3 4.1 5.7 5.5 7.0 6.8 0.13 P<0.05 P<0.001
Day 33-68
Liveweight change (gd" ) 6 -8 6 8 29 35 71 52 4.0 ns P<0.001
DMI (gd1) 314 259 246 231 429 427 440 445 12.5 ns P<0.001
DMI (gkg075 d"1) 51.8 45.3 43.0 40.5 66.3 65.2 62.6 67.0 1.51 ns P<0.001
MMEiikg^VHlO1) 4.9 4.9 5.0 5.0 6.3 7.0 7.3 8.2 0.16 ns P<0.001
Day 68-94
Liveweight change (gd" ) 33 26 36 23 47 67 68 40 3.6 ns P<0.01
DMI(gd') 313 258 264 237 447 469 484 389 10.0 ns P<0.001
DMI (gkg-0-75 d"1) 50.2 44.3 44.1 40.1 65.5 64.9 63.1 53.5 1.07 P<0.01 P<0.001
MJMEifkg-'^VHlO11 4.8 4.8 5.2 4.9 6.2 7.0 7.3 6.5 0.12 ns P<0.001
* Standard error ** dry matter intake
Contrasts 1 and 2 = diet and feeding level respectively

Small Ruminant Research Network Workshop 193

P.R. Hatendi, LR. Ndlovu, T. Smith and A.N. Said

Table 4. Linear regression coefficients of daily energy intake (MJ ME/kg ' per day) on liveweight change (g/kg per day)
(Equation 1), for the period 33-68 days, of Matabele goats offered diets differing in roughage concentration.
Diet df a sea b seb Syx r2 Significance
1 16 0.508 0.0169 0.225 0.00284 0.061 0.808 P<0.001
2 9 0.551 0.0299 0.0231 0.00494 0.090 0.732 P<0.0016
3 9 0.471 0.0388 0.0251 0.00464 0.074 0.785 P<0.0001
4 15 0.518 0.0328 0.0311 0.00433 0.097 0.7 87 P<0.0001
All 53 0.518 0.0144 0.0252 0.00207 0.085 0.744 P<0.0001
Syx = regression error mean square.

Table 5. Linear regression coefficients of liveweight change (g/kg per day) on daily energy intake (MJ ME/kg' ' per day)
(Equation 2), for the period 33-68 days, of Matabele goats offered diets differing in roughage concentration.
Diet df a sea b seb Syx r2 Significance
1 16 -17.628 2.665 35.869 4.520 2.403 0.808 P<0.0001
2 9 -16.969 4.162 31.680 6.783 3.336 0.732 P<0.0016

3 9 -13.308 3.798 31.303 5.801 2.596 0.785 P<0.0001

4 15 -12.004 2.480 25.271 3.519 2.755 0.787 P<0.0001
All 53 -14.257 1.562 29.560 2.429 2.909 0.744 P<0.0001

Table 6. Estimates of metabolisable energy requirements for maintenance (MEm) (KJ ME/kg' ' per day) and liveweight gain
(MEg) (KJ/g) for the period 33-68 days, of castrate Matabele goa s offered diets differing in roughage content.

MEm Meg
Diet Equation 1 Equation 2 Equation 1 Equation 2
1 508 481 22.5 27.9
2 551 536 23.1 31.9
3 471 425 25.1 31.9
4 518 475 31.1 39.6
All 519 482 25.2 34.2

of MEm and MEg from the two estimation methods stasis. This definition is not synonymous with zero
followed a similar trend between diets. energy retention as found from calorimetric or
comparative slaughter studies (Armstrong and Blaxter,
1984). Use of the MEm estimates based on liveweight
stasis is arguably more relevant in the practical feeding
Discussion of livestock for meat production, where liveweight
The differences in dry matter intake (DMI) and the changes can be measured with relative ease and
similarity in energy intake among animals on restricted accuracy.
feeding are artifacts of the planned treatments. This study compared two forms of regression, used
Differences in the rate of liveweight gain among these interchangeably in literature (Zemmelink et al, 1991).
animals may indicate differences in absorption, The first regressed energy intake (y) on liveweight
partitioning and efficiency of utilisation of nutrients. change (x) (equation 1). In the second, liveweight
Animal performance on the ad libitum feeding regime
change (y) was regressed on energy intake (x) (equation
followed trends similar to those reported previously for
mature castrates fed similar diets (Hatendi et al, 1992). 2). The question of which variable should be used as the
Treatments in this experiment resulted in a wide dependent variable has been comprehensively discussed
range of growth rates. The data set was thus amenable by Zemmelink et al (1991), who argue that liveweight
to estimation of energy requirements for maintenance gain is the observed response to energy intake and
(MEm) and liveweight gain (MEg) by regression therefore should be considered the dependent variable.
analysis. MEm is defined here as the level of This approach is similar to that used in the ARC (1980)
metabolisable energy required to maintain liveweight feeding standards and is central to the theories of animal

194 Small Ruminant Research Network Workshop

 Metabolisable energy requirements for maintenance etc..

response to nutrients as developed by AFRC (1991). AFRC (Agricultural [Link] Food Research Council). 1991.
Our acceptance of these arguments, and therefore AFRC Technical Committee on responses to nutrients
equation 2, represents a change from our earlier number 7. Theory of response to nutrients by farm
animals. Nutrition Abstract and ie<evirws (Series B)
publication on this subject (Hatendi et al, 1990), in
61:683-722.
which equation 1 was used to apportion energy
utilisation. Alam M R, Poppi D P and Sykes AR. 1991. Comparative
The overall estimates of MEm in the period 33 to energy and protein utilisation in kids and lambs. Journal
68 days were 519 and 482 KJ ME/kg0'75 per day for ofAgricultural Science (Cambridge) i 17:12 1- 1 27.
equations 1 and 2 respectively. These results are Allen О В, Burton J H and Holt J D. 1 983. Analysis of repeated
consistent with estimates presented in the literature measurements from animal experiments using
(Abate, 1989; Hatendi et al, 1990; Zemmelink et al, polynomial regression. Journal of Animal Science
1 99 1 ; Sauvant et al, 1 99 1 ). Similarly, the MEg estimates 57:765-770.
of25.2and34.2kJ/gliveweightgainforequations 1 and ARC (Agricultural Research Council). 1980. The nutrient
2, respectively, are consistent with the findings of other requirements of ruminant livestock. Commonwealth
workers (Abate, 1989; Hatendi et al, 1990; Zemmelink Agricultural Bureaux, Slough, UK. 351 pp.
et al, 1991). Armstrong D G and Blaxter K L. 1984. Maintenance
If the inverse of MEg is considered as an estimate requirement: implications for its use in feed evaluation
of efficiency of energy use for liveweight gain (Abate, systems. In: Gilchrist F M C and Mackie R I (eds),
1989), then the efficiency of gain decreased with Herbivore nutrition in the sub-tropics and tropics. The
increasing dietary energy content and duration of Science Press (Pty.) Ltd, Craighall, South Africa.
feeding. Efficiencies in this experiment are similar to pp. 631-647.
those reported for kids by Alam et al (1991). Estimated Hatendi P R, Smith T, Ndlovu L R and Mutisi С 1990.
feeding levels in this experiment, based on equation 2 Estimated metabolisable energy requirements for
estimates of MEm, were 1.03 and 1.49 times maintenance and growth of indigenous Zimbabwe goats.
maintenance for the restricted and ad libitum feeding Zimbabwe Journal ofAgricultural Research 28:65-70.
regimes, respectively. Hatendi P R, Smith T, Ndlovu L R and Mutisi С 1992.
Fattening mature indigenous (Matabele) goats: Effects
on animal performance, oody and carcass composition.
In: Rey В, Lebbie S H В and Reynolds L (eds), Small
Conclusion ruminant research and development in Africa.
The estimates of MEm and MEg presented in this paper Proceedings of the First Biennial Conference of the
are of direct application to confined castrate Matabele African Small Ruminant Research Network, ILRAD,
goats offered similar diets. However, they may be Nairobi, Kenya, 10-14 December 1990. ILCA
(International Livestock Centre for Africa), Nairobi,
judiciously applied to related breeds of eastern and
Kenya. pp. 355-365.
southern African goats under similar husbandry
systems. Similarly, the increments for medium and high MAFF (Ministry of Agriculture, Fisheries and Food). 1984.
activity proposed by NRC (1981) serve as guidelines Energy allowances and feeding systems for ruminants.
and may, as appropriate, be added when considering the Reference Book 433. HMSO (Her Majesty's Stationery
energy requirements of goats in other systems such as Office), London, UK. 85 pp.
free-range grazing. Mohammed H H and Owen E. 1980. Comparison of the
maintenance requirement of sheep and goats. Animal
Production 30:479.
Acknowledgements Morton J R and Ridgeman W J. 1977. Problems in the
statistical design and analysis of feeding trials.
We thank the officers and staff of Matopos Research Proceedings of the Nutrition Society 36: 173-177.
Station, where this work was undertaken, for their
NRC (National Research Council). 1981. Nutrient
assistance and encouragement. Thanks are also due to
requirements ofgoats: Angora, dairy, and meat goats in
Mr J. Sherington and Ato Solomon Zewdie of the
temperate and tropical countries. National Academy
International Livestock Centre for Africa (ILCA) for
Press, Washington DC, USA. 91 pp.
biometrical assistance.
The senior author is grateful to the Rio Tinto Prieto C, Aguilera J F, Lara Land Fonolla J. 1990. Protein and
Foundation (Zimbabwe) and ILCA for grants of post energy requirements for maintenance of indigenous
graduate and research fellowships, respectively, to facili Granadina goats. British Journal of Nutrition 63:
155-163.
tate, in part, the conduct and write-up of this work.
SAS (Statistical Analysis System). 1990. Statistical Analytic
System. Statistical Analytical Institute Inc., North
Carolina, USA.
References
Sauvant D, Morand-Fchr P and Giger-Reverdin S. 1991. Dry
Abate A. 1989. Metabolizable energy requirements for matter intake of adult goats. In: Morand-Fehr P (ed),
maintenance of Kenyan goats. Small Ruminant Research Goat nutrition. EAAP (European Association for
2:299-306. Animal Production) Publication 46. Pudoc (Centre for

Small Ruminant Research Network Workshop 195

P.R. Hatendi, L.R. Ndlovu, T. Smith and A.N. Said

Agricultural Publishing and Documentation), Zemmelink G, Tolkamp В J and Ogink N W M. 1991. Energy
Wageningen, The Netherlands. pp. 25-36. requirements for maintenance and gain of West African
t
Topps„ Jt и
H „andj n«
Oliver J.t imo л 0 i feeds
1978. Animal t j oft central
. lit-
Africa. Dwarf goats.
ь Small Ruminant Research 5:205-215.
Rhodesia Agricultural Journal, Technical handbook 1.
Government Printers, Harare, Zimbabwe.

1 96 Small Ruminant Research Network Workshop

 Effect of energy level on milk production by Mossi ewes
and Sahelian does
A.J. Nianogo and P.C. Ilboudo

Département de zootechnie - Université de Ouagadougou

B.P. 7021, Ouagadougou, Burkina Faso

Abstract digestibles totaux (NDT) (régime la), 73,4% de NDT

(régimes 2a et 3a) et 83,9% de NDT (régime 4a). Dans
Two trials were conducted to evaluate the effect of le second essai, 18 brebis Mossi réparties en trois lots
energy level on milk production and composition and to recevaient une ration iso-azotée (17,4% de PB)
establish the relationship between milk production and apportant 62,1% de NDT (régime 1b), 73% de NDT
growth rate of suckling kids. In the first trial, 20 Sahelian (régime 2b) ou 83,1% de NDT (régime 3b). Ces régimes
does were divided into four groups to receive one of four avaient été formulés de manière à fournir des apports
isonitrogenous (18% crude protein (CP)) diets providing en énergie inférieurs (la et lb), conformes (2a, 2b et 3a)
42.2% total digestive nutrients (TDN) (diet la), 73.4% et supérieurs (3b et 4a) à la normale. Les quantités
TDN (diets 2a and 3a) and 83.9% TDN (diet 4a). In the servies étaient constamment remaniées en fonction des
second trial, 18 Mossi ewes were divided into three quantités de lait produites par les animaux. En outre, le
groups to receive one of three isonitrogenous (17.4% régime de base des animaux couvrait la totalité de leurs
CP) diets providing 62.1% TDN (diet lb), 73% TDN besoins d'entretien. La production laitière a été estimée
(diet 2b) and 83.1% TDN (diet 3b). These production par la méthode de la double pesée, avec quatre pesées
diets were formulated to provide subnormal (la & lb), quotidiennes espacées les unes des autres par des
normal (2a, 2b and 3a), or high (3b and 4a) levels of intervalles de durée identique. Dans les deux essais, le
energy; the amounts fed were constantly adjusted to régime n'avait pas d'effet significatifsur l'ingestion de
milk yield. Additionally, animals were fed a basal diet matière sèche, ni sur la composition et la production de
satisfying all maintenance requirements. Milk lait. Il apparaît ainsi que de faibles fluctuations dans
production was estimated by the double weighing l'apport énergétique ne modifient pas la production de
method, with four equally spaced weighing periods per lait chez la chèvre du Sahel et le mouton Mossi. Il y avait
day. In the two trials, daily dry matter intake (DMI), une corrélation entre la croissance des jeunes et la
milk composition and production were not significantly consommation de lait. Dans les régimes des petits
affected by diet. It was concluded that small fluctuations ruminants en lactation, on peut remplacer les grains de
in energy intake do not affect milk production in sorgho par des graines de dolique lablab.
Sahelian goats or in Mossi sheep. Kid growth is
correlated with milk consumption. Sorghum grain may
be replaced by Dolichos lablab seeds in diets for
lactating small ruminants. Introduction
Limited information is available on the milk
Effet de l'apport énergétique sur productivity of small ruminants in Burkina Faso. Milk
from Sahelian goats is consumed by Fulani and other
la production laitière chez la Sahelian livestock farmers. Milk production data is also
useful for selection purposes; it is well established that
brebis Mossi et la chèvre du females with high milk output promote faster growth
Sahel and better survivability of their young.
Available data indicates that Mossi (IEMVT,
Résumé 1980; Yoni, 1989) and Djallonké (Amegee, 1984a)ewes
have a low milk production potential. However,
Deux essais ont été conduits pour estimer l 'effet de preliminary studies conducted by Ouédraogo (1990)
l'apport énergétique sur la production de lait et sa indicate that Sahelian does may produce considerable
composition, et pour déterminer les relations qui amounts of milk.
existent entre la production laitière et le taux de This study evaluated the effects of graded levels of
croissance de jeunes allaités au pis. Dans le premier dietary energy on dry matter intake (DMI), milk yield
essai, 20 chèvres du Sahel ont été réparties en quatre and milk composition. It was also intended to establish
lots. Chaque lot recevait une ration iso-azotée (18% de the relationship between milk production and the growth
protéines brutes [PB]) apportant 42,2% de nutriments of suckling kids.

Small Ruminant Research Network Workshop 197

A.J. Nianogo and P.C. Ilboudo

Materials and methods per 24 hours, one day a week as described by Coombe
et al (1960). Milk was sampled during weeks 4, 8, and
A group of 20 Sahelian does and another of 18 Mossi 12 and preserved at 4°C with 1% of a 35% solution of
ewes were paired, based on body weight and age. Each formaldehyde until chemical analyses were performed.
animal entered the study 48 hours after kidding. A The milk was analysed for total solids, crude
two-week preliminary phase was observed, during protein (CP), casein and ash using methods described by
which all animals received the same diet to establish АОAC ( 1 984). Milk fat was determined by the Babcock
initial production levels. After the preliminary phase, a method (АОAC, 1984). Lactose was determined by
separate 12-day trial was conducted for each species difference.
using different dietary treatments. Data was statistically analysed using the SAS
General Linear Models (GLM) procedures (SAS, 1982).
Trial 1 Means were separated using Scheffe's Multiple
Comparisons Test.
Paired does were randomly assigned to one of four
production diets (Table 1). Three energy (E) levels were
used: diet la (low E); diets 2a and 3a (medium E); and
diet 4a (high E). Additionally, a maintenance (NRC,
Results and discussions
1981) diet composed of 77% sorghum straw, 5%
cottonseed meal and 18% wheat bran was fed to all Intake of dry matter
groups. The amount of basal diet served daily was based
on body weight. The amount of production diet served Dry matter intake (DMI) by goats increased with the
varied with milk yield. increase in energy levels of the diets (Table 3). With
Paired ewes were randomly assigned to one of ewes, DMI was depressed by the high energy diet (Table
three production diets (Table 2): diet lb (low E); diet 4). Goats appeared to consume more DM than ewes.
2b (medium E); and diet 3b (high E). The ewes also
received a basal diet composed of 83.9% chopped Yield and composition of milk
sorghum straw and 16.2% cottonseed meal (dry matter
basis). Data on the yield and composition of milk are shown in
In each trial, kids were offered an ad libitum supply Tables 5 (does) and 6 (ewes). The milk yield of does on
of sorghum straw and their mother's production diet. diet 4a was significantly higher (P<0.05) than that ofdiet
Body weights were monitored once a week. Feed intake la. Peak milk yields were 1091.67, 1062.5, 1079.2 and
of mothers was monitored daily during weeks 4, 6, 8, 10 1275g/d for diets la, 2a, 3a and 4a respectively. Diets
and 12. Milk production was estimated by weighing the 2a and 3a performed similarly, indicating that Dolichos
kids immediately before and after suckling, four times I. seeds may be a good substitute for sorghum grain.

Table 1. Composition of production diets for Trial 1 (goats).

Diet
Item Ia 2» 3" 4»
Ingredient (%):
Cottonseed meal 17.00 4.00 0.00 22.00
Commercial mix* 17.60 24.00 2.00 0.00
Dolichos I. seeds 5.00 5.00 42.20 0.00
Cottonseed 0.00 0.00 8.00 0.00
Wheat bran 14.50 28.50 23.00 0.00
Sorghum grain 17.50 36.00 10.00 72.50
Ground oyster shells 1.40 2.50 2.30 01.50
Molasses 0.00 0.00 5.00 0.00
Corn cobs 27.00 0.00 8.00 0.00
Cotton oil 0.00 0.00 0.00 4.00
Total 100.00 100.00 100.00 100.00
Expected chemical
composition (%):
TDN 62.16 73.38 73.42 83.87
crude protein (CP) 18.02 18.06 18.01 18.03
Ca 0.62 1.01 0.63 0.62
P 0.66 0.78 0.52 0.56
TDN = total digestible nutrients.
Cottonseed meal with 47% cottonseed hulls added.

198 Small Ruminant Research Network Workshop

 Effect of energy level on milk production etc...

Table 2. Composition of production diets for Trial 2 (ewes).

Diet
Item lb 2b 3b
Ingredient (%):
Cottonseed meal 32.50 12.00 00.00
Commercial mix 1 17.00 8.00 0.00
Dolichos L. seeds 5.00 42.00 49.50
cottonseed 0.00 3.00 7.00
Wheat bran 10.00 19.00 19.60
Sorghum grain 11.50 7.00 15.00
Ground oyster shells 2.50 3.00 1.70
Molasses 5.00 0.00 0.00
Corn cobs 8.00 27.00 0.00
Cotton oil 0.00 0.00 6.20
Urea 0.00 0.00 1.00
Total 100.00 100.00 100.00
Expected chemical
composition (%):
TDN2 62.09 73.03 83.13
crude protein (CP) 17.47 17.43 17.40
Ca 1.04 1.68 0.70
P 0.78 0.65 0.50
Cottonseed meal with 47% cottonseed hulls added.
TDN: total digestible nutrients.

Table 3. Effect of diet on intake of dry matter (DMI) and average daily gain (ADG) of Sahelian goats.
Diet
Parameter la 2a 3a 4a SE
Does
Number 6 4 6 4 -
DMI, g/day 975.24 1122.91 1116.07 1178.47 65.63
g/kg™ 84.88 92.57 96.13 101.16 2.96
Body weight, week:
0 25.25 29.12 25.05 24.50 0.99
2 26.03 28.45 25.10 24.35 0.98
ADG, week 3-14 1.98a 17.56ab 10.71ab 39.29b 4.65
Kids
Number 8 5 8 7 -
Body weight, week:
0 2.15 2.26 2.09 2.01 0.04
14 8.64ab 9.78a 8.5 lab 7.37b 0.39
ADG week:
3-14 65.92 70.24 67.26 55.44 3.76
0-14 66.20 76.74 65.56 54.67 4.22
Means in the same row with different superscripts differ significantly (P<0.05).

Average composition of the milk from does was higher than those reported by Ouédraogo (1990) with
not affected by diet (Table 5). However, diet 3a gave a hand-milked Sahelian does. However, Ouédraogo
higher (P<0.05) CP concentration than all other diets (1990) reported higher concentrations of solids
during week six. Milk yields obtained in this study were (16.42%) and fats (6.03%).

Small Ruminant Research Network Workshop 199

A.J. Nianogo and P.C. Ilboudo

Table 4. Effect of diet on intake of dry matter (DMI) and average daily gain (ADG) of Mossi ewes.
Diet
Parameter lb 2b 3b SE
Ewes
Number 6 6 6
DMI, g/day 1051.50 1052.94 1016.00 23.20
g/Kg0.75 85.95 87.33 79.58 1.95
Body weight. week:
0 27.60 26.78 28.97 4.16
2 27.47 26.57 28.97 0.94
ADG, week 3-14 5.16 17.26 12.50 3.48
Lambs
Number 6 6 7
Body weight, week:
0 2.76 2.33 2.41 0.43
14 10.33 10.53 8.16 0.64
ADG week:
3-14 76.19 80.95 54.59 6.17
0-14 77.2 lab 83.67a 58.45i 6.29
Means within the same row with different subscripts differ (P<0.05).

Table 5. Effect of diet on milk yield and milk composition of Sahelian does.
Diet
Parameter 2a 3a 4a SE
Yield (g/d):
week 3-14 582a 667ab 631ab 713" 45.72
week 1-14 632a 700ab 675ab 735b 47.98
Solids: % 13.67 12.31 13.38 13.21 0.25
total (g/d) 102.10 106.72 1 10.22 120.28 8.43
Fat: % 5.22 3.81 4.76 4.47 0.23
total (g/d) 38.66 32.46 38.53 36.95 2.49
Crude protein: % 3.83 4.05 3.76 3.62 0.07
total (g/d) 29.06 34.% 31.12 33.20 2.52
Casein, % 3.08 3.31 3.20 3.05 0.05
Lactose, % 3.84 3.63 4.09 4.33 0.15
Minerals, % 0.77 0.78 0.81 0.77 0.01
Means in the same row with different superscript differ (P<0.05).
Body weights and ADG of all animals are shown in Tables 3 (goats) and 4 (sheep).

Milk yield (MY), milk composition and pro Bodyweight changes

duction of milk constituents by Mossi ewes (Table 6)
were not affected by diet. Weight gains were low throughout both trials. Does had
The MY recorded here were lower than those higher weight gain (P<0.05) on diet 4a than those on diet la.
reported by Amegee (1984a) on Djallonké ewes for
mothers of singles (547g/day) and for mothers of twins Growth of kids and correlation with milk
(772g/day). Amegee also reported higher con yield
centrations of total solids (16.47%) and CP (6.02%). On average, lambs reached a higher body weight than
Goats produced more milk than ewes and goat milk goat kids at 14 weeks old. There was no effect of dietary
had a higher concentration of total protein, casein and treatments on average daily gain (ADG) or body weight
minerals. of kids.

200 Small Ruminant Research Network Workshop

 Effect of energy level on milk production etc...

Table 6. Effect of diet on milk yield and milk composition of Mossi ewes.
Diet
Parameter jb 2b 3b SE
Yield (g/d):
week 3-14 395.31 387.50 396.35 27.50
week 1-14 444.20 449.29 442.71 26.73
Solids: % 13.06 13.06 12.90 0.31
total (g/d) 89.40 81.55 80.11 7.39
Fat: % 4.94 5.33 4.65 0.21
total (g/d) 27.78 27.78 23.18 1.80
Crude protein: % 5.38 5.05 4.95 0.14
total (g/d) 33.16 29.01 28.40 2.69
Casein, % 4.69 4.16 4.27 0.12
Lactose, % 4.12 3.70 4.18 0.25
Minerals, % 0.86 0.84 0.82 0.01

Correlation analyses indicated that milk con References

sumption (MC) by a suckling kid is best estimated when
the weights of the kid at birth (W0) and at six weeks of Amegee Y. 1984a. Etude de la production laitière de la brebis
age (Wo) are included in the equation (MC in g, W0 and Djallonké en relation avec la croissance des agneaux.
W6inKg): Revue d'élevage et de médecine vétérinaire des pays
tropicaux 37(3):33 1-335.
e.g. For kids (R2 = 0.9375): Amegee Y. 1984b. Le mouton de Vogan (croisé Djallonké x
MC = -327.1092 Wo + 38.3492 W6 + 53.9617 sahélien) au Togo. La production lactée et ses relations
avec la croissance des agneaux. Revue d'élevage et de
W0xW6+ 1039.4727;
médecine vétérinaire des pays tropicaux 37( 1 ):82-90.
For lambs (R2 = 0.8619):
AOAC (Association of Official Analytical Chemists). 1984.
MC -366.92 - 58.40 W0 + 37.96 Wo + 13.70 Official methods ofanalysis ofthe Association ofOfficial
WoxW6. Analytical Chemists. 14th ed. AOAC, Washington DC,
Good correlations were also obtained between USA.
average daily MC in the first 13 weeks (91 days) of Coombe J B, Wardrop I D and Tribe D E. 1960. A study of
lactation, W0 and body weight at 1 3 weeks of age (Wi 3); milk production of the grazing ewe, with emphasis on
(MC in g, Wo and W13 in kg): the experimental technique employed. Journal of
Agricultural Science 54:353-359.
For goats (R2 = 0.897):
IEMVT (Institut d'élevage et de médecine vétérinaire des pays
MC= -51.8991 Wo + 4.7041 W13+ 14.3734 tropicaux). 1980. Les petits ruminants d'Afrique
W0xW13 + 469.2035; centrale et d'Afrique de l'Ouest. Synthèse des
For lambs (R2 = 0.7934): connaissances actuelles. IEMVT, Maisons-Alfort,
France. 295 pp.
MC: -118.8236 Wo + 7.7736 W13+ 11.3062 NRC (National Research Council). 1981. Nutrient
W0xW13 + 503.0907. requirements of goats. National Academy Press,
Multiple factor correlations provide a better Washington DC, USA. 91 pp.
estimate of milk production than those established Ouédraogo Z. 1 990. Aptitudes de la chèvre du Sahel burkinabé
between milk consumption and a single parameter à la production laitière. Influence du rang de mise-bas,
(ADG, birth weight or body weight at 30 or 40 days) by de l'alimentation et de la saison. Mémoire de fin
Amegee (1984a,b). d'études. 1DR (Institut du développement rural),
Université de Ouagadougou, Burkina Faso. 136 pp.
SAS (Statistical Analysis System). 1982. SAS User's guide:
Statistics. SAS Inst. Inc., Cary, North Carolina, USA.
Conclusions YoniT. 1989. Influence du taux de concentré sur la production
laitière des brebis mossi et la croissance des agneaux
Minor fluctuations in energy intake do not affect milk allaités. Mémoire de fin d'études. IDR (Institut du
production in Sahelian goats or in Mossi ewes. Kid développement rural), Université de Ouagadougou,
growth is highly correlated with milk consumption. Burkina Faso. 80 pp.

Small Ruminant Research Network Workshop 201

 SESSION 5

Small Ruminant Genetic Resources

and Breeding
 Indigenous African small ruminants: a case for
characterisation and improvement
J.E.O. Rege

International Livestock Centre for Africa

[Link] 5689, Addis Ababa, Ethiopia

Abstract à la sélection naturelle. Cet article décrit les systèmes

de production pratiqués en Afrique et fait ressortir le
The accelerating demands of a growing human rôle des petits ruminants dans les systèmes de
population and the pressures of economic development subsistance et le bien-être de la société. Les difficultés
are affecting the security and survival of many qui viennent contrarier l'utilisation extensive et
indigenous African breeds. Until now, these breeds have l'amélioration des ressources génétiques animales
been a stable part of their particular ecosystems for africaines doivent être identifiées. Des programmes
hundreds of years. There is an increasing tendency to d'amélioration génétique doivent être mis en place. Les
introduce exotic germplasm and/or to concentrate on a races déjà menacées doivent être protégées d'urgence,
narrow range of supposedly more profitable breeds. As même si leur valeur économique n'est pas encore
a result, indigenous breeds are threatened, even though apparente. De même, des informations sur les
they have been naturally selected for the local performances biologiques et les caractéristiques
environments and are therefore best adapted. This paper d'adaptation de ces populations doivent être
describes the production systems in Africa and rassemblées afin de contribuer à l'élaboration de
highlights the contribution of sheep and goats to programmes rationnels d'utilisation et de conservation
subsistence farming systems and the general welfare of de ces ressources génétiques animales.
society. Problems preventing extensive utilisation and
improvement of African animal genetic resources need
to be identified. Genetic improvement programmes need Introduction
to be developed. Breeds already endangered need to be
conserved as a matter of urgency even if their economic Indigenous sheep and goat breeds constitute over 95%
value is not presently apparent. Concurrently, of the small ruminant population of Africa. They are
information should be compiled on biological owned by the majority of smallholder rural farmers for
performance and adaptive characteristics of these AGR whom this resource is critical for nutrition and income.
(animal genetic resource) populations to aid the They are also an important and secure form of
development of rational utilisation and conservation investment. They provide the only practical means of
programmes. using vast areas of natural grasslands in regions where
crop production is impracticable. Low genetic potential
is often quoted as a major constraint to meat and milk
Nécessité de caractériser et production in sub-Saharan Africa (SSA). Con
sequently, most livestock improvement programmes in
d'améliorer les races de petits the region have resorted to crossbreeding with imported
exotic breeds or directly replacing the indigenous
ruminants africains indigènes genotypes. This trend continues as the human
population and the demand for animal products
Résumé increases. As previously isolated communities
intermingle, indigenous animal populations can
Les demandes d 'une population en expansion et les interbreed. This, combined with increasing animal
contraintes imposées par le développement économique movements through trade, presents additional pressures
compromettent la sécurité et la survie de nombreuses on these populations. In some regions of Africa the
races indigènes africaines. Jusqu'à peu, ces races effects of drought and famine, compounded by
constituaient un élément stable des écosystèmes dans prolonged civil wars, have also taken their toll.
lesquels elles évoluent depuis des centaines d'années. Consequently, indigenous African breeds are at risk.
Toutefois, on constate aujourd'hui une tendance à However, sustainable livestock improvement cannot be
introduire du matériel génétique exotique et à investir guaranteed for some environments without the adaptive
dans un petit nombre de races censées être plus traits of these genetic resources. The rate of
rentables. Les races indigènes sont de cefait menacées, developments in biotechnology also suggests that it will
même si elles sont bien adaptées au milieu dans lequel soon be possible to identify and manipulate genes,
elles évoluent, milieu dans lequel elles ont été soumises including those which confer disease resistance and

Small Ruminant Research Network Workshop 205

J.E.O. Rege

Table 1. Characteristics of agroecological zones of sub-Saharan Africa.

Proportionate Proportionate Proportion
Agro- Characteristics Proportionate human TRLU of cattle Predominant livestock
ecological of the area population population population Breed production system
zone zone a % % % % typec

Arid 1.0- 500 35 12 29.8 20.7 Indigenous Traditional nomadic, and

2. <75 days transhumant based on
communal grazing

Semi-arid 20 26 27. 30.6 Indigenous, Nomadic, transhumant;

1.500-1000 exotic (+) Smallholder crop-
2. (75 -180 days) livestock

Subhumid 1. 1000-1500 20 27 19.6 22.7 Indigenous, Smallholder crop-

2. 180 -270 days) exotic (++) livestock; agropastoral

Humid 1. >1500 20 20 6.1 6.1 Indigenous Small crop-livestock

2. >270 days

Highland (<20°C)b 15 17.4 19.9 Indigenous, Smallholder

exotic crop-livestock; animal
(+++) traction important
Total НИ) 100 100 100
Line 1 = rainfall, mm; 2 = length of growing period, days.
Highlands are defined as all those areas in the semi-arid, subhumid and humid zones of SSA where, because of altitude, mean daily temperature
during the growing periods is <20'C.
Use of exotic ("improved") breeds: +++ = very important; ++ = moderately important; + = some importance; blank = not important.
TRLU = Tropical ruminant livestock unit.

physiological adaptation to other environmental production options and hence production systems
stresses. The potential global contribution of the genetic (Table 1). In addition to the physical environment,
resources of indigenous small ruminants is tremendous. characterising livestock production in these zones
It would be tragic if unique genetic resources resulting consists of assessing the key products and functions of
from centuries of natural and artificial selection were livestock. Accordingly, the major livestock production
lost. The rate of erosion of indigenous animal genetic systems in Africa (Table 1) are: pastoral range systems,
resources therefore threatens prospects of providing the usually found in the arid and semi-arid zones;
livelihood of present and future human generations. agro-pastoralism in the semi-arid and subhumid zones;
African policy makers and other development agencies mixed crop-livestock systems found in the lowlands of
need to recognize the role these resources play in today ' s subhumid zones and high rainfall areas of the highlands;
production systems. Programmes should be set up to and ranching range systems and introduced systems
conserve and develop these resources in step with the such as modern intensive dairy, beef, poultry and pig
evolving agricultural systems. production which are increasingly common in
This paper discusses the role of indigenous small peri-urban areas. Major characteristics of these systems
ruminants in African production systems and highlights have been described (e.g. Pratt and Gwynne, 1977;
the key attributes and functions which justify their Jahnke et al, 1988).
conservation and improvement. ILCA (1987) has classified SSA into five principal
agroecological zones. These are the arid, semi-arid,
subhumid, humid and highland zones. The arid zone
covers over a third of the region. The semi-arid,
Agroecological zones and subhumid and humid zones each occupy about a fifth
production systems while the highland areas make up only 5% (Winrock
International, 1992) of the total land area. East Africa
Africa is a continent of enormous environmental contains 70% of the highlands and central Africa has
diversity. This is mainly due to variations in rainfall,
about 75% of the humid zone (ILCA, 1987). On the
altitude and soils. In general, there are close
other hand, arid, semi-arid and subhumid zones are
relationships between climate, land use and human
population densities. These relationships are important in all subregions of SSA except Central
particularly strong in systems where there is little or no Africa, where the arid and semi-arid zones are marginal.
external input into the production system. Thus, the The close association between agroecological
ecological zones, which are categorised mainly on zones and production systems is apparent from Table 1 .
climatic data, are quite closely linked with the Both pastoral and ranching systems are largely based on

206 Small Ruminant Research Network Workshop

 Indigenous African small ruminants: a case for etc...

range, primarily utilising natural vegetation. However, Due to their short generation lengths and high
the major outputs and their importance differ between reproductive rates, small ruminants have a high
these systems. Pure pastoralism is practised in the arid production efficiency (Winrock International, 1976). In
to very arid ecological zones, defined as those with an addition, the energetic efficiency of milk production
annual rainfall of less than 400 mm. The major product may be higher in dairy goats than for other dairy animals
from these systems is milk; the major livestock role is (Winrock, 1976). Certain breeds of sheep (e.g.
for subsistence and cultural or social functions (Jahnke, Djallonké) and goat (e.g. West African Dwarf) are
1982). The major functions of livestock in the ranching thought to be tolerant of trypanosomiasis and other
systems are meat and cash incomes. The contribution of diseases, allowing them to be grazed on land not
livestock in a mixed crop-livestock system is measured available to other domestic livestock.
in terms of the increase in overall farm productivity Small ruminants are useful to humans during
attributable to the presence of livestock. periods of cyclical and unpredictable food shortages.
They also help balance the energy and protein supply
during normal variations between seasons and years.
The small size and early maturity of sheep and goats give
Distribution of small ruminants by them several distinct economic advantages in
agroecological zones smallholder situations, such as: they can efficiently
utilise marginal and small plots of land; the risk on
The population of sheep in SSA is estimated at 127
investment is reduced by smaller individual size,
million head, while that of goats is estimated at 147
allowing more production units per unit of investment;
million (Winrock International, 1992). The arid and
semi-arid zones together hold the majority of sheep and there is a faster turnover of capital because they are
(57%) and goat (64%) populations of sub-Saharan sexually mature early and are younger at slaughter.
Africa. However, the arid holds 11% more sheep and Smaller carcasses are also easier to market and can be
12% more goats than the semi-arid, which contains 23 consumed in a short period of time. This is important as
and 26% of the sheep and goats, respectively. More most rural areas lack proper storage facilities. Their
sheep are found in the highland zone than in the humid strong flocking instinct makes herding by younger and
zone (21 and 8% respectively), but the humid and the older members of the family possible, allowing labour
highland zones account for about equal proportions of to be used more efficiently.
the population of goats (9 and 10% respectively). The Small ruminants produce lower absolute quantities
population of sheep and goats in the subhumid zone are of milk than cattle. However, when their body weight is
14 and 17%, respectively. Thus, the arid and semi-arid taken into account, their milk yield is higher than other
zones account for the majority of small ruminants in the species with the possible exception of camels (Wilson,
region. However, these species are an important 1991). During difficult periods of the year, these minor
livestock component in all agroecological zones. In levels of output become significant (Coppock et al,
addition, small ruminants are found in all production 1982). It has been estimated that up to 40 years may be
systems ranging from pastoral and agropastoral systems needed for cattle to attain the population and production
to ranching range systems to smallholder mixed levels existing prior to a drought (Wilson, 1991).
crop-livestock systems. Because of their shorter generation length and higher
reproductive rate, small ruminants have a much shorter
recovery period. For example, following a severe
Special attributes of small ruminants drought, goats conceive as soon as there is sufficient
moisture for leaves and browse plants to grow. They kid
Sheep and goats are highly adaptable to a broad range five months later and consequently produce milk for
of environments. They can utilise a wide variety of plant human consumption at a very early phase of the recovery
species and are thus complementary to cattle and cycle (Wilson, 1991).
camels. They generally do not compete directly with Although regional and breed variations exist, small
these species for feed. For example, a mixture of animal ruminants appear to withstand drought better than cattle
species on semi-arid rangelands makes it possible to (Campbell, 1978). The droughts of the early 1980s
change the stocking rate from 26 ha per tropical
which affected Ethiopia and the Sahcl, including Sudan,
ruminant livestock unit (TRLU) (250 kg liveweight
equivalent) for cattle alone to 13 ha per TRLU when resulted in cattle losses of 80% or more. Small ruminant
cattle and goats are reared together and to 10ha per losses did not exceed 50% (Wilson, 1991). Shafie
TRLU when camels arc included (Schwartz, 1983). It (1992) has summarised the physical and physiological
is the complementarity of the ruminant species that characteristics which enable sheep and goats to survive
maintains the high animal biomass that characterises in arid and saline environments. They include: body
pastoral production systems. Goats are more effective at conformation (slim trunks and slender limbs) which
grazing selectively than any other domestic livestock provide a large proportional surface area, helping excess
species (Winrock International, 1976) and they utilise body heat dissipate through non-evaporative cooling;
poor quality forage and browse better than sheep. Goats and a respiratory system with a larger proportion of dead
also walk longer distances in search of food than do space to respiratory space, a structure which facilitates
other domestic livestock (Wilson, 1991). heat dissipation via respiratory water vaporisation

Small Ruminant Research Network Workshop 207

J.E.O. Rege

Table 2 Sub-Saharan Africa 's estimated livestock populations and meat and milk output by agroecological zones (1988).
Meat output Milk output
Quantity
Species and Inventory3 Off take Carcass ('000 Inventory Lactation Quantity
agroecological zone (millions) wt(kg) tonnes) (millions) (kg) ('000 tonnes)
(%)
Cattle (total) 162.5 10.4 137 2315 25 330 8200
Arid/semi-arid 84 10 140 1176 13 250 3300
Subhumid/humid 46 10 130 598 7 200 1400
Highlands 32 12 140 538 5 700 3500
Sheep and goats 269 25 14 942 2128e

Sub total (red meat) 3257

Poultry 630 110 1.3 900
Pigs 11 79 40 348
Sub total (white meat) 1248
Total 4505 10,328
Source: Adapted from Winrock (1992).
Total population.
Milked population only.
c) Calculated from 1985 (FAO, 1987) country statistics.

without the need for severe panting. Due to their inappropriate when almost all parts of the animal are
adaptation to the environment (including feeding consumed. Thus, actual meat outputs from these species
behaviour), small ruminants, particularly goats, are in traditional production systems are expected to be
often the last to be affected by catastrophes. This causes higher than estimates in Table 2. Most importantly,
them to be blamed for desertification. production from these species makes a valuable
contribution to food resources in areas where it is most
needed.
Role of small ruminants
Small ruminants as producers of manure
Small ruminants as source of food Few farmers have access to commercial fertilizers.
Manure and urine from domestic animals are therefore
Human diets will not contain necessary amounts and essential components for maintaining agricultural
kinds of amino acids unless they include protein from production. However, only limited quantitative data are
either animal products or an unusually well designed available on the contribution of manure in traditional
combination of foods from plants (FAO, 1 983). production systems. While some data are available on
It is difficult to determine accurately the the contribution of cattle to manure production (e.g.
contribution of small ruminants to human food supply Rege and Agyemang, 1992), data on the contribution of
and general welfare. Available statistics are misleading small ruminants is scanty.
because much of the production from these species does
not enter formal trade channels and is therefore not
reported. Estimates indicate that ruminants contribute
Asset and security functions of small
80% of the total food production from livestock in ruminants
tropical Africa (Jahnke, 1982). Of this, small ruminants
account for about 22%. It is estimated that ruminants One of the reasons usually given for low live animal
offtake from traditional herds is the holding of livestock
supply over 3.2 million tonnes of meat per year,
as a status symbol or sign of wealth. Livestock owners
representing over 72% of total meat production (Table
in rural areas do not usually have access to banking
2). Small ruminant meat accounts for about 30% of the facilities and have come to rely on investment in their
total red meat production and over 20% of the total meat stock. While investment in livestock does carry its own
output of SSA. Small ruminants account for about 21% risks, it provides security against crop failure and
of the total milk produced in SSA. This is about 26% of currency fluctuations. An analysis of income accruing
the milk output from cattle in the region (Table 2). The from crops and livestock in a mixed farming situation
figures on meat production have been calculated from was carried out in Abet, northern Nigeria. It revealed
carcass weights which, in turn, were estimated from that, among farm items sold, livestock and livestock
assumed dressing percentages. In the African context, products accounted for 56% in value terms (Ingawa,
and indeed in most developing countries, the 1986). Crop sales were limited to certain seasons, but
conventional concept of dressing percentage is the sale of livestock and milk was less season

208 Small Ruminant Research Network Workshop

 Indigenous African small ruminants: a case for etc..

dependent. It therefore represents a dependable asset for Botswana, Swaziland, and Zimbabwe which export beef
emergencies. In general, livestock contribute to the and mutton and Kenya which exports meat and
stability of farm income because they can be bought indigenous animal germplasm (both live animals and
following good crop performance and sold following semen). Several countries (e.g. Ethiopia) also export
crop failures. Thus, livestock helps to cushion farm hides and skins. However, products from small
enterprises against unstable commodity prices during ruminants offer local alternatives to imported products
unfavourable years. Livestock also serves as a thereby saving foreign currency. The highest per capita
"currency" in which social obligations are expressed. consumption of animal products outside pastoral areas
occurs in urban areas. If there were no local supply of
Small ruminants facilitate utilisation of these products, the demand would have to be met
marginal lands through importation. Even where small ruminants do not
actually produce exportable products, they provide for
About 13-16 million km , or nearly half of the continent the local demand, allowing the export of prime beef and
south of the Sahara, is desert or arid grassland and mutton from limited high-input systems. In this way they
savanna. Crop production is a high risk enterprise contribute indirectly to export earnings. They also
(Brown, 1971). Nearly all livestock in these zones are contribute indirectly to the employment of urban labour
indigenous African breeds raised under traditional in the packing industry.
nomadic and transhumant pastoral systems based upon
communal grazing. They are well adapted to the Small ruminant genetic resources
characteristically sharp annual and seasonal variations of Africa
in rainfall, requiring mobility in search of forage. These
land use systems efficiently utilise available vegetative The majority of sheep and goats in tropical Africa are
resources and provide a reliable source of food for the "indigenous" types, although both species were
population (Cossins, 1983). Frequent reference to low originally domesticated in Asia (Epstein, 1971). In the
land productivity of the pastoral range livestock last 150 years or so exotic breeds have been imposed,
production systems ignores the fact that livestock mainly for modern commercial operations (Wilson,
farming is the only way to support any human life at all 1991). The Republic of South Africa has been in the
in much of this zone. Wilson and Light (1986) have forefront of this movement. There are large populations
reported that as many as 95% of agropastoral families in of Merino sheep in Kenya and Zimbabwe, and Karakul
Mali own goats and that the average flock size per family sheep in Namibia. Merino are also important in the
is over 30 head. smallholder sector of Lesotho, as are Angora goats
(Wilson, 1991). The latter are also gaining prominence
under commercial systems in Kenya. Karakuls are
Overall contribution of small ruminants present in Angola, Botswana and Rwanda, but in smaller
to national economies numbers. A composite breed, the Dorper, was bred in
the Republic of South Africa from the Dorset Horn and
A rough estimate of the importance of the livestock Blackhead Persian. It is rapidly gaining ground for meat
sector to the national economy is its contribution to the production under local conditions, notably in
Gross Domestic Product (GDP). The share of Zimbabwe, Kenya and, more recently, Botswana. Other
agriculture to the GDP is important in Africa, but varies countries have also imported exotic breeds but on a
from 2.2. and 2.3% for Gabon and Côte d'Ivoire lesser scale than in the southern African region and
respectively, to 35.7% for Zimbabwe, 38.7% for Chad Kenya.
and 81.6 and 86.3%, respectively for Somalia and In general, pressures of exotic breeds on
Mauritania. The average for the continent is 17.4% indigenous types is much less for small ruminants than
(Jahnke et al, 1988). The share of livestock in for cattle, where the contribution of AI to crossbreeding
agricultural GDP has been estimated at 25% but this and breed substitution has been considerable. However,
increases to 35% when traction and manure are as is the case with indigenous cattle, breeds are not
considered (Winrock International, 1992). However, distinct from one another. There are hardly any
these averages hide great differences between countries, indigenous small ruminant breeds which have been
ecological zones and subregions. For example, the share adequately characterised. Even so, it is widely
of livestock in agricultural GDP's ranges from 5% for recognised that because these genetic resources have
Ghana and Côte d'Ivoire to 82, 84 and 88% for Namibia, evolved under the stresses of the African environment,
Mauritania and Botswana, respectively (Winrock they are well adapted to the local conditions.
International, 1992). On the basis of dollar values used
by Winrock International (1992) for meat and milk,
monetary contributions of small ruminants to GDP were
estimated for 1988. Meat output contributed about Performance potential of indigenous
$1286 million while milk output contributed about breeds
$1162 million.
In general, livestock products do not contribute The broad genetic variability of African small ruminant
much to direct foreign currency earnings in African breeds enables them to survive under stressful
countries. The exceptions include countries like environmental conditions, including high disease

Small Ruminant Research Network Workshop 209

J.E.O. Rege

incidence, poor nutrition and high temperatures. flock owners deliberately delay first breeding (Wilson
Environmental pressure also maintains a wide range of et al, 1984). Poor nutrition increases the animals'
genotypes, each adapted to a specific set of susceptibility to diseases. Animals that have low
circumstances. This is clearly the species' genetic maintenance requirements and the ability to make
response to risk. Under on-station management, efficient use of poor quality forages are therefore at an
indigenous African sheep and goat breeds have shown advantage. There is a need for comparative evaluation
good growth and reproduction performances, indicating of indigenous and exotic breeds in African
their potential. This shows that they would respond to environments, taking input costs and output prices into
improvements in management, particularly nutrition account.
and disease control. This should be accompanied by
Most imported (temperate) breeds are raised on
selecting for improved performance. However, African
high potential agricultural land where crop farming is
livestock producers are not yet in a position to organise
breed improvement programmes. Governments should practiced. The critical role of indigenous breeds is in
therefore take responsibility for the development and providing the only means of using areas where other
implementation of selection schemes. forms of agriculture are not practical. It may not be
possible to improve some of these environments,
especially in the arid zone, to accommodate exotic
breeds. However, the strategy will most probably be
A case for evaluation and some kind of crossbreeding to take advantage of breed
complementarity and/or upgrading to allow gradual
conservation improvement in husbandry. Additionally, recent
developments in biotechnology indicate that direct
Studies indicate that the performance of small ruminants
transfer of genes may become a routine method for
could be improved through management. Additionally,
estimates of genetic parameters point to considerable germplasm improvement. Thus there is need to keep a
genetic variation within indigenous populations. This reservoir of these exceptional genes. Indigenous breeds
indicates the potential for genetic improvement through are crucially important today and will continue to be
selection. valuable in the future.
There has been a tendency to over-emphasise the
low productivity of indigenous breeds without due
consideration of some important characteristics of these
breeds. When the small size of these breeds and the harsh ILCA/FAO plan of action
environmental conditions under which they are raised
are taken into account, their productivity is impressive. Several expert panel meetings have addressed the issue
Comparative studies of indigenous and exotic breeds to of conservation of indigenous breeds, at both global and
determine their feed utilisation efficiency have not been regional levels. These meetings have involved, among
conclusive. However, it is known that breeds with high others, the Food and Agriculture Organization (FAO) of
maintenance requirements tend to lose the most weight the United Nations (UN), the Organization of African
and have the highest mortality rates under stress Unity (OAU) through its Inter-African Bureau for
conditions such as drought (Frisch, 1984). Most breed Animal Resources (IBAR), UN Environmental
comparison studies have concentrated on quantifying Programme (UNEP), National Agricultural Research
performance (e.g. live weights and milk production) but Systems (NARS) and other International Agricultural
not inputs. Indeed, the high-performing temperate Research Centres (IARCs), including ILCA. However,
breeds cannot survive under traditional management in efforts in support of Animal Genetic Resources (AGRs)
most African environments. Although the performance have mainly remained at the stage of studies to identify
of indigenous breeds under improved management has priorities. Recognising the need to go beyond studying
not been adequately assessed, there are indications that options and actually to implement them, in 1991 ILCA
they respond to improved husbandry. For example, launched a research project to characterise indigenous
Kolff and Wilson (1985) have reported that indigenous African AGRs. ILCA's research initiative is aimed at
documenting population sizes, biological performance
small ruminants were able to double their daily weight
and adaptive or other special characteristics of these
gain rate with only minimal improvements in nutrition
AGRs. This information can then be used to develop
and management. The ability to survive under adverse rational utilisation and conservation programmes. In
environmental conditions with low inputs makes February 1992, ILCA organised a research planning
indigenous breeds a low-risk choice. The low-risk factor workshop to work out methodologies and outline the
of resistant breeds is important where market prices are implementation process. The workshop was attended by
unstable or where the probability of death from NARS scientists and AGR experts from FAO, several
environmental stress is high (Frisch, 1984). In some IARCs and other institutions outside Africa. Details of
cases "low productivity" is an adaptive mechanism. For the AGR programme for Africa proposed by the FAO
example, delayed age at first parturition and extended were presented. It immediately became clear that, given
parturition intervals in semi-arid environments are the scale of the undertaking and the common objectives
mechanisms for coping with seasonal and often of ILCA and FAO's proposed programmes, a
unreliable feed availability. In such environments some collaborative programme would be more effective.

210 Small Ruminant Research Network Workshop

 Indigenous African small ruminants: a case for etc..

Subsequently, joint proposals have been prepared and Jahnke H E, Tacher G, Keil P and Rojat D. 1988. Livestock
funding is currently being sought. production in tropical Africa, with special reference to
the tsetse-affected zone. In: Livestock production in
tsetse-affected areas of Africa. Proceedings of a meeting
held in Nairobi, Kenya, 23-27 November 1987.
References 1LCA/1LRAD (International Livestock Centre for
Africa/International Laboratory for Research on Animal
Brown LH. 1971. The biology of pastoral man as a factor in
Diseases), Nairobi, Kenya. pp. 3-21.
conservation. Biological Conservation 3(2).
Kolff H E and Wilson R T. 1985. Livestock production in
Campbell D J. 1978. Coping with drought in Kenya
central Mali: The "Mouton de case" system of
Masailand: Pastoralists and farmers of the Loitokitok
smallholder fattening. Agricultural Systems
area. Working Paper No. 337. Institute for Development
Studies, University of Nairobi, Nairobi, Kenya. 16:217-230.

Coppock D L, Ellis J E, Wenpahl J, McCabe J D, Swift D M PrattDJandGwynneM D. 1977. Rangeland management and
and Calvin K A. 1982. A review of livestock studies of ecology in East Africa. Hodder and Stoughton, London,
the south Turkana ecosystem. Proceedings of the Small UK. 310 pp.
Ruminant Collaborative Research Support Programme, Rege J E O and Agyemang K. 1992. The role of indigenous
Kenya. Winrock International, Morrilton, Arkansas, cattle breeds in African livestock production systems.
USA. Paper presented at the 43rd Annual Meeting of the
Cossins N. 1983. Production strategies and pastoral man. In: European Association for Animal Production, Madrid,
Pastoral systems research in sub-Saharan Africa. Spain, 14-19 September 1992.
Proceedings of the IDRC/ILCA workshop held at ILCA,
Shafie M M. 1992. Morphological and anatomical
Addis Ababa, Ethiopia, 21-24 March 1983. ILCA characteristics ofsubtropical sheep and goats as means
(International Livestock Centre for Africa), Addis
of adaptation to hot climate. Paper presented at 43rd
Ababa, Ethiopia. pp. 2 1 3-23 1 . Annual Meeting of the European Association for Animal
Epstein H. 1971. The origin of domestic animals of Africa. Production, Madrid, Spain, 14-19 September 1992.
Africana Publishing Corporation, New York, USA. 719
Shwartz H J. 1983. Improved utilisation of arid rangelands
pp. through multiple species herds. Proceedings of the 5th
FAO (Food and Agriculture Organization of the United World Conference on Animal Production 2:625-626.
Nations). 1983. The State ofFood and Agriculture 1982.
FAO Agriculture Series 15. FAO, Rome, Italy. 203 pp. Wilson R T. 1991. Small ruminant production and the small
ruminant genetic resource in tropical Africa. FAO
FAO (Food and Agriculture Organization of the United Animal Production and Health Paper 88. FAO (Food and
Nations). 1987. World Crops and Livestock Statistics Agriculture Organization of the United Nations), Rome,
(1984-1985). Food and Agriculture Organization of the Italy. 231 pp.
United Nations (FAO), Rome, Italy
Wilson R T and Light D. 1 986. Livestock production in central
Frisch J E. 1 984. Genetic attributes required for efficient cattle
Mali: Economic character and productivity indices for
production in the tropics. In: Baker F H and Miller M E
traditionally managed goats and sheep. Journal of
(eds), Beef cattle science handbook. Volume 20.
Animal Science 62:567-575.
Westview Press, Boulder, Colorado, USA. pp. 244-270.
ILCA (International Livestock Centre for Africa). 1987. Wilson R T, Peacock D P and Sayers A R. 1984. Aspects of
ILCA 's strategy and long-term plan. ILCA, Addis reproduction in goats and sheep in south-central Kenya.
Ababa, Ethiopia. 99 pp. Animal Production 38:463-467.

Ingawa S A. 1986. Socio-economic aspects of Abet farming Winrock International. 1976. Proceedings of a Workshop on
households. In: von Kaufmann R, Chater S and Blench the Role of Sheep and Goats in Agricultural
R (eds), Livestock systems research in Nigeria's Development held at Winrock International Center,
subhumid zone. Proceedings of the Second Morrilton. Arkansas. USA, 15-17 November 1976.
ILCA/NAPRI Symposium held in Kaduna, Nigeria, 29 Winrock International Livestock Research and Training
October - 2 November 1984. ILCA (International Centre, Morrilton, Arkansas, USA. 43 pp.
Livestock Centre for Africa), Addis Ababa, Ethiopia. pp. Winrock International. 1992. Assessment of animal
227-242. agriculture in sub-Saharan Africa. Winrock
Jahnke H E. 1982. Livestock production systems and livestock International Institute for Agricultural Development,
development in tropical Africa. Kieler Morrilton, Arkansas, USA. 125 pp.
Wissenschaftsverlag Vauk, Kiel, Germany. 253 pp.

Small Ruminant Research Network Workshop 211

 Impact sur la productivité des ovins
Djallonké de la sélection des béliers sur leur taille
A. Oya1, C.V. Yapi2, M. Ouattara3
lSODEPRA-PNSO, 01 B.P 1, 366 Bouaké 01 (Côte d'Ivoire)
2Institut des savanes (IDESSA), 01 B.P. 633, Bouaké 01 (Côte d'Ivoire)
Laboratoire de pathologie animale, B.P. 1 366, Bouaké 01 (Côte d'Ivoire)

Résumé Ce programme recouvrait des activités d'encadrement

des propriétaires d'animaux et la gestion du Centre
Dans le but d'améliorer les performances de national ovin (CNO) ayant pour objectifs la production
croissance des ovins de race Djallonké, la Côte d'Ivoire de géniteurs performants, la mise à l'essai des
conduit depuis 1 976 un programme national de sélection techniques et du matériel d'élevage et la formation des
ovine (PNSO). Afin de mesurer l'impact de ce techniciens et des éleveurs.
programme sur la productivité des ovins Djallonké, une Après plusieurs années d'encadrement qui ont vu
étude est menée pour tester deux hypothèses: 1) la naître des éleveurs performants, il est apparu opportun
sélection sur la taille des béliers améliore la croissance de créer un programme national de sélection ovine
des ovins Djallonké; et 2) la sélection sur la taille des (PNSO). Cet organisme a pour mission de fournir aux
béliers n'a pas d'effets négatifs sur la santé des animaux éleveurs encadrés par la SODEPRA des reproducteurs
ni sur leur résistance ou tolérance aux maladies. L'étude mâles sélectionnés dans le but d'améliorer les
engagée vise à rechercher les indicateurs de maladies et performances de croissance du mouton Djallonké. Il
à comparer les performances de production et de apparaît aujourd'hui nécessaire de comparer les béliers
reproduction chez les descendants de béliers issus de sélectionnés aux béliers tout-venant, sur la base des deux
deux types génétiques: le type sélectionné et le type hypothèses suivantes:
non-sélectionné.
1. La sélection sur la taille des béliers améliore les
performances de croissance des ovins Djallonké;
Impact ofselection on size of 2. La sélection sur la taille des béliers n'a pas d'effets
rams on West African Dwarf négatifs sur la santé des animaux ni sur leur
résistance ou tolérance aux maladies.
sheep productivity Le présent rapport présente la démarche qui sera
suivie dans l'étude initiée par la SODEPRA et
Abstract l'IDESSA afin a) de mesurer l'impact que la sélection
A national sheep-breeding programme (PNSO) has sur la taille des béliers exerce sur les performances de
been conducted in Côte d'Ivoire since 1976 to improve croissance des ovins Djallonké et sur les caractères de
growth performance in West African Dwarf sheep. In reproduction des brebis; et b) d'établir les relations entre
order to measure the impact ofthis programme on sheep la sélection sur la taille des béliers et les indicateurs de
productivity, a study is being carried out to test the résistance aux maladies.
following two hypotheses: 1) that selection of ram size
improves growth of West African Dwarf sheep, and 2)
that selection of ram size has no adverse effect on the
health ofthe animals nor on their resistance to disease. Matériel et méthodes
The objectives ofthe study are to seek disease indicators
and compare productive and reproductive performance Les béliers sélectionnés du PNSO seront comparés
in the progeny oftwo types of ram, the selected and the à des béliers tout-venant achetés sur le marché local et
non-selected. dans les villages (béliers témoins). Des contrôles seront
effectués pour s'assurer de leur bon état sanitaire (sang,
sperme, selles).
Introduction Les deux types de béliers seront mis en lutte avec
des brebis tout-venant après une période de quarantaine.
Afin de rationaliser l'élevage ovin traditionnel en Cinq béliers de chaque type génétique seront introduits
Côte d'Ivoire, un programme national ovin (PNO) avait dans des lots de 20 brebis pendant 45 jours. Les béliers
été engagé et mis en oeuvre dès 1976 par la Société de rejoindront les brebis pendant la nuit seulement. Les
développement des productions animales (SODEPRA). cinq béliers sélectionnés seront de générations

Small Ruminant Research Network Workshop 213

A. Oya, C.V. Yapi, M. Ouattara

différentes pour pouvoir comparer les générations entre - sur 80 descendants (40 agneaux des deux sexes par
elles. lot): OPG et hématoscopie tous les mois;
Cette étude s'étendra sur trois cycles de sur l'ensemble des agneaux: sérologie à l'âge de 3
reproduction séparés par des intervalles de huit mois. mois, 9 mois et 1 an.
Les animaux seront conduits sur pâturage naturel Un fichier sera établi pour recenser tous les autres
pendant la journée et recevront 300 g d'aliments problèmes de pathologie. Tout cadavre sera acheminé
complémentaires (son de blé et tourteaux de coton) par au laboratore pour autopsie et recherches
complémentaires.
animal (bélier, brebis suitées) et par jour.
L'analyse statistique qui sera utilisée est celle de
Les paramètres de reproduction et de croissance
la méthode des moindres carrés. Les facteurs de
suivants seront relevés: variation suivants seront pris en compte:
- chez les brebis: la fertilité, la prolificité, la - pour les caractères concernant les performances de
fécondité, les poids avant la lutte et après croissance de l'agneau: le type de bélier, le sexe,
l'agnelage, la mortalité; le mode de naissance, la période d'agnelage, le
chez les agneaux: le poids à la naissance, le poids poids de la brebis;
au sevrage (90 jours), les poids à 30 j, 60 j, 80 j, - pour les données relatives aux béliers sélectionnés:
180 j, 240 j et 365 j, les âges à 20, 25, 30, 35 et les paramètres indiqués ci-dessus et le facteur
40 kg pour les mâles, la mortalité à la naissance génération du bélier;
(0-3 j), la mortalité entre la naissance et le sevrage pour les caractères concernant les performances de
(4-90 j) et la mortalité post-sevrage. reproduction des brebis: le type de bélier, la
Les indicateurs de maladies suivants ont été période de lutte, la période d'agnelage, l'année, le
retenus: poids de la brebis.
- la recherche des oeufs et larves dans les fèces par Les descendants femelles feront l'objet d'études
la technique de Mac Master (nombre d'oeufs par ultérieures visant à évaluer l'effet de la sélection sur les
gramme); aptitudes maternelles de la brebis Djallonké.
la recherche des parasites sanguins (Babesia ovis,
trypanosomes et anaplasmes) par frottis, goutte
épaisse et hématocrite;
- la recherche de témoins d'infection de la sphère
Résultats attendus
génitale telle que la chlamydiose, la brucellose Suivi de l'évolution du progrès génétique réalisé
(surtout pour l'épididymite contagieuse des ovins). entre les différentes générations de béliers sélectionnés:
Le protocole des prélèvements pour la prise en - détermination du gain génétique réalisé;
compte de ces indicateurs est le suivant: - évaluation du schéma de sélection mis en oeuvre
sur l'ensemble des béliers et brebis: sérologie, par le PNSO;
OPG, hématoscopie une fois sur un cycle (pendant - contribution à l'amélioration du plan de
la quarantaine); prophylaxie appliqué par les projets
- sur 40 brebis prises au hasard (20 brebis de chaque d'encadrement ovin de la SODEPRA dans les
lot): OPG et hématoscopie pendant la gestation; élevages de la base de sélection.

214 Small Ruminant Research Network Workshop

 Caractères phénotypiques de différenciation et de
croissance des agneaux de race pure Djallonké et croisés
Sahélien x Djallonké
C.V. Yapi

Institut des savanes (IDESSA), département Elevage

B.P. 633, BouakéOl (Côte d'Ivoire)

Résumé showed birth weight (BW), ear length (EL) and tail
length (TL) to be traits for differentiating the two breed
Pour déterminer les caractères phénotypiques groups. These traits were positively correlated (tel-TL
permettant de différencier un agneau Djallonké de race = 0.707; rEL-BW = 0.699; m-BW = 0.627). The best
pure d'un agneau croisé Sahélien x Djallonké, une discriminant variable was the ear length.
analyse factorielle discriminante et une analyse de Notwithstanding their younger age, the cross-
variance ont été réalisées sur 64 agneaux contemporains breds had better performances in all traits measured.
des deux types génétiques. L'analyse discriminante a The analysis ofvariancefor preweaning growth showed
révélé que les caractères qui permettent de différencier that breed group significantly affected weight
les deux types génétiques étaient le poids à la naissance (P<0.001 ) only at an early age (from birth to 60 days).
(PN), la longueur des oreilles (LO) et la longueur de la The effect ofbreed group disappeared at weaning. Live
queue (LQ), caractères entre lesquels il existe une weight between birth and weaning rosefrom an average
corrélation positive étroite (rLOJLQ = 0,707; rLO_PN 2.17 kg to 7.62 kg in crossbreds, andfrom 1.76 to 7.28
= 0,699; rLQ_PN = 0,627). La longueur de l'oreille kg inpurebreds. Live weight, average daily weight gain,
constituait la meilleure variable discriminante. ear length and tail length at 60-70 days averaged 6.78
Pour tous les caractères mesurés, les agneaux kg, 65.5 g/d, 12.4 cm and 21.5 cm, respectively, for
croisés présentaient les meilleures performances bien crossbreds, and 6.55 kg, 62.9 g/d, 9.9 cm and 17.6 cm,
qu'étant les moins âgés. L'analyse de variance sur la respectively, for purebreds.
croissance présevrage indiquait que l'effet du type
génétique sur le poids n'était très important (P<0,001)
qu'au jeune âge (de la naissance à 60 jours). Au sevrage,
l'effet dû au type génétique avait disparu. De la
1. Introduction
naissance au sevrage, le poids vif des agneaux croisés La race Djallonké, une ressource génétique sûre
passait de 2,17 kg à 7,62 kg en moyenne et celui des pour la Côte d'Ivoire, se caractérise par une précocité de
agneaux de race pure de 1,76 kg à 7,28 kg. A l'âge de reproduction des jeunes béliers et antenaises, une
60-70 jours, le poids vif, le gain moyen quotidien, la aptitude à se reproduire tout au long de l'année, une
longueur de l'oreille et la longueur de la queue avaient adaptation aux conditions de l'environnement tropical
des valeurs moyennes de 6,78 kg, 65,5 g/j, 12,4 cm et humide et surtout une tolérance à la trypanosomiase
21,5 cm pour les croisés et de 6,55 kg, 62,9 g/j, 9,9 cm animale, conséquence d'une longue période de sélection
et 17,6 cm pour les agneaux de race pure. naturelle.
Malgré ces qualités, elle présente un poids vif
adulte dépassant rarement 40 kg et une croissance très
Phenotypic differentiation and lente des animaux, ce qui limite la rentabilité de la
growth traits in purebred West production de viande.
Face au souhait des éleveurs de posséder ou de
African Dwarf lambs and their mettre sur le marché des animaux de grand format, on
crosses with Sahel lambs assiste dans les zones urbaines à des pratiques de
croisement entre des moutons de race Djallonké et des
Abstract animaux de races originaires des pays du Sahel
communément appelées "sahéliennes". A l'âge adulte,
Discriminantfactorial analysis and analysis ofvariance le poids vif de ces animaux de grande taille peut
were made on 64 contemporary West African Dwarf atteindre 90 kg (Charray et ai, 1980; Fitzhugh et
(WAD) and WAD x Sahel lambs to determine phenotypic Bradford, 1983).
traits for differentiating purebred WAD from their Le département Elevage de l'Institut des savanes
crosses with Sahel sheep. The discriminant analysis (IDESSA) à Bouaké a lancé une étude ayant pour

Small Ruminant Research Network Workshop 215

C.V. Yapi

objectifs la définition des caractéristiques phéno- Le poids à la naissance et l' âge au sevrage ont été utilisés
typiques des croisés Sahélien x Djallonké et l'évaluation comme covariables dans l'analyse du poids au sevrage.
de leurs performances par rapport à celles des Djallonké
de race pure.
Le présent article présente les résultats de la
première année de l'étude sur les caractères phéno- 3. Résultats et discussion
typiques de différenciation et la croissance présevrage
des agneaux Djallonké de race pure (D X D) et croisés
Sahélien x Djallonké (S x D). 3.1 Caractères phénoty piques de
différenciation

2. Matériel et méthodes L'analyse factorielle discriminante révèle un

pourcentage de 80% d'animaux bien classés parmi les
Un bélier sahélien dont les caractéristiques agneaux Sahélien x Djallonké et de 86% parmi les
correspondaient à celles du mouton Peul a été acheté sur agneaux Djallonké. Le fait que les agneaux croisés
le marché à bétail de Bouaké et mis en lutte sur un lot soient classés parmi les agneaux de race pure indique
de 26 brebis Djallonké. Par ailleurs, 4 lots de 17 à 20 que la plupart d'entre eux sont phénotypiquement plus
brebis Djallonké ont été affectés d'un bélier Djallonké proches des moutons sahéliens. Une faible proportion
chacun; les luttes ont eu lieu simultanément du 22 mai des croisés avait les caractéristiques typiques de la race
au 6 juillet 1991, soit pendant 45 jours. Il s'agissait de Djallonké.
luttes naturelles, sans traitement hormonal préalable des Le poids à la naissance (F = 1 8,30), la longueur
brebis. Les béliers rejoignaient les lots de brebis pendant de l'oreille (F = 25,12) et la longueur de la queue (F
la nuit seulement. = 20,94) confirment cette classification (tableau 1). La
Les agnelages ont eu lieu en plein air d'octobre à meilleure variable discriminante était la longueur des
novembre 1991. Un numéro était attribué aux agneaux oreilles, pour laquelle la valeur de F était la plus élevée.
à la naissance et le poids, le sexe et le mode de naissance La corrélation était de 0,707 entre la longueur des
étaient immédiatement enregistrés. Le sevrage des oreilles et celle de la queue; de 0,699 entre la longueur
agneaux a eu lieu le 29 janvier 1992 à un âge moyen de des oreilles et le poids à la naissance; et de 0,627 entre
90 jours. la longueur de la queue et le poids à la naissance. Le
Les animaux ont été alimentés sur pâturage de périmètre thoracique de l'agneau s'étant révélé très
savane naturelle pendant la journée. Le soir, les brebis fortement corrélé avec le poids (r - 0,959), le GMQ (r
recevaient 50 à 100 g de tourteau de coton par animal. = 0,852) et la longueur scapulo-ischiale (r = 0,800), il
De la paille de riz était distribuée pendant la saison pourrait être utilisé comme indice d'expression du poids
sèche. ou du GMQ lorsque la pesée est difficile à réaliser.
Une série de mesures ont été prises sur les agneaux Bien que constituant la meilleure variable
entre l'âge de 40 et de 90 jours. Ces mesures ont ensuite discriminante, la longueur des oreilles ne peut à elle
été analysées par la méthode d'analyse factorielle seule permettre de classer les agneaux. Son association
discriminante du programme STAT-ITCF (ITCF, 1 987) avec les autres variables s'avère très importante pour la
afin de déterminer les caractères phénotypiques classification des agneaux, étant donné que le T de
permettant de différencier les deux types génétiques S Hoteling (tableau 1) est supérieur à la valeur de F pour
x D et D x D. Les variables mesurées étaient le poids à la longueur des oreilles.
la naissance, l'âge à la pesée, le poids au jour de la pesée, A la naissance, les agneaux croisés Sahélien x
le gain moyen quotidien (GMQ) de la naissance au jour Djallonké pesaient 0,5 kg de plus que les agneaux
de la pesée, le nombre total de taches, le pourcentage de Djallonké. A un âge plus avancé (63-71 jours), la
taches sur le corps à l'exception de la tête et des différence de poids entre les deux types génétiques
extrémités des pattes, le rapport entre le nombre de diminuait et n'était plus que de 0,23 kg, avec une
taches sur le corps et le nombre total de taches, la différence de la croissance quotidienne de 2,63 g/j
longueur de l'oreille, la longueur du cou, la longueur (tableau 1). La longueur des oreilles des croisés avait
scapulo-ischiale, la longueur de la queue, et le périmètre 3 cm de plus que celle des Djallonké et la queue, 4 cm
thoracique. de plus.
Les agneaux ont été pesés à 1 5, 33 et 7 1 jours et au La queue des Djallonké se situe au-dessus ou au
sevrage. Les poids aux âges types de 30, 60 et 90 jours niveau du jarret et celle des croisés au niveau ou en
et les différents gains moyens quotidiens ont été calculés dessous du jarret, dans une proportion de 50% chaque.
à partir des différentes pesées. Aucun agneau Djallonké n'avait la queue en dessous du
Une analyse par la méthode des moindres carrés jarret et aucun croisé ne l'avait au-dessus du jarret, sauf
(Harvey, 1 988) a été effectuée pour déterminer les effets chez des animaux mal classés. Par ailleurs, avec un
dus au sexe (mâle ou femelle), au mode de naissance pourcentage de taches sur le corps de 9,83% contre
(simple ou double), au type génétique (race pure ou 23,93% chez les croisés (tableau 1), les agneaux
croisée) et à l'interaction entre le sexe et le type Djallonké présentaient en général une robe plus
génétique sur les différents poids et GMQ des agneaux. uniforme.

216 Small Ruminant Research Network Workshop

 Caractères phénotypiques de différenciation et de .

Tableau 1. Valeurs de F et moyennes des caractères mesurés chez des agneaux Djallonké et Sahélien x Djallonké

F (probabilité associée) Moyennes

Caractères (T2 = 58,72)
DxD SxD
Poids à la naissance (kg) юко-,01) 2,12 2,64
Age(j) 4,46 (3,87) 71 63
Poids (kg) 0,17(68,29) 6,55 6,78
GMQ(g/j) 0,13(72,11) 62,87 65,5
Nombre total de taches 1,49(22,54) 6 5
Pourcentage de taches (%) 5,32 (2,32) 9,83 23,93
Longueur des oreilles (cm) 2L12 (0,00) 9,89 12,39
Longueur du cou (cm) 0,05(81,98) 17,12 17,31
Longueur scapulo-ischiale (cm) 2,33(13,10) 35,07 37,21
Longueur de la queue (cm) 2&24(0,01) 17,58 21,51
Périmètre thoracique (cm) 0,29 (60,24) 43,74 44,49

3.2 Performances de croissance Tous types génétiques et modes de naissance

confondus, il n'y avait pas de différence entre le poids
Il ressort de l'analyse de variance que le sexe, le des agneaux mâles et femelles de 30 jours au sevrage
mode de naissance et le type génétique avaient un effet (tableau 3). A l'exception du GMQ entre 30 et 60 jours,
très significatif sur le poids à la naissance. L'influence la croissance pondérale des femelles a été plus rapide
de ces facteurs diminuait avec l'âge, puis disparaissait que celle des mâles, mais cette différence n'était pas
au sevrage (tableau 2). Les poids de la naissance au statistiquement significative. En particulier, les croisés
sevrage des agneaux croisés étaient nettement femelles avaient une croissance de la naissance à 30
supérieurs à ceux des agneaux de race pure (tableau 3). jours (98,6 g/j) significativement supérieure à celle des
S 'agissant des gains moyens quotidiens, seuls les effets
croisés mâles. De la naissance au sevrage, le poids des
dus au sexe et à l'interaction type génétique x sexe
agneaux croisés augmentait plus rapidement que celui
étaient importants au jeune âge (P<0,05), les autres
des agneaux de race pure, mais la différence n'était pas
facteurs n'ayant pas eu d'effet sur le gain moyen
quotidien (tableau 4). statistiquement significative (tableau 5).
Les différences de poids entre les agneaux issus de Les faibles performances des croisés enregistrées
naissances simples et ceux issus de naissances à l'IDESSA par rapport à celle obtenue par d'autres
gémellaires n'étaient significatives (P<0,01) qu'entre 0 institutions pour le mouton du Vogan par exemple
et 30 jours, encore que la quasi-totalité des agneaux ait (Amégée 1983, 1984) pourraient s'expliquer par
été de naissance simple. Sur 59 agneaux, 4 seulement l'inadéquation de l'environnement sanitaire et
étaient issus de naissances gémellaires. D n'y a eu alimentaire qui empêche les croisés d'extérioriser tout
aucune naissance double parmi les croisés. leur potentiel génétique.

Tableau 2. Analyse de variance du poids des agneaux de la naissance au sevrage

Source de Degré de
variation liberté Naissance 30 j 60j 90j Sevrage
Sexe 1 изг 0,33 1,3 0,01 0,39
Mode de naissance 1 иб*" 5,29* 4,67 0,87 0,12
Type génétique I 1,78*" 9,63" 12,17** 3,27 0,56
Sexe x Type génétique 1 0,29 2,53 0,91 0,08 0,08
Régression
Poids naissance 1 2,67
Age au sevrage 1 8,01
Résiduelle 0,11 0,95 1,81 2,53 2,73
(d.d.l.) (54) (50) (43) (34) (32)
*** P<0,001 ** P<0,01 * P<0,05

Small Ruminant Research Network Workshop 217

C.V. Yapi

Tableau 3. Moyennes estimées du poids des agneaux de la naissance au sevrage (kg)

Moyenne générale 1,97 4,06 5,64 7,28 7,50
±0,09 ±0,29 ±0,46 ±0,64 ±0,71
Sexe
Mâle 2,15a 3,97 5,85 7,31 7,32
Femelle l,79b 4,15 5,42 7,26 /,59
Mode de naissance
Simple 2,33a 4,70a 6,32 7,64 7,60
Double l,60b 3,42b 4,96 6,93 7,31
Type génétique
Djallonké l,76b 3,57b 4,99b 6,93 7,28
Croisés 2,17a 4,55a 6,29a- 7,63 7,62
Au sein d'une même colonne, les moyennes affectées des mêmes lettres ne sont pas significativement différentes.

Tableau 4. Analyse de variance des GMQ des agneaux de la naissance au sevrage

Source de Degré de
variation liberté Naiss.-30j 30-60 j 60-90 j Naiss.-90j Naiss.-Sevrage
Sexe 1 3306 6383** 977 69 40
Mode de naissance 1 1209 670 59 2 12
Type génétique 1 3068 1290 0 67 73
Sexe x Type 1 5092* 4199* 256 96 30
Régression
Poids naissance 1 7
Age au sevrage 1 80
Résiduelle 1099 897 497 299 330
(d.d.l.) (50) (43) (34) (34) (32)
*P<0,01 *P<0,05

Tableau 5. Moyennes estimées du GMQ des agneaux de la naissance au sevrage (g/j)

Moyenne générale 69,5 55,4 41,7 59,1 58,8
±9,79 ± 10,30 ±8,99 ±6,97 ±7,78
Sexe
Mâle 70,3a 35,7
60,5 57,5 57,4
Femelle 40,5b
78,6 47,8 60,7 60,1
Type génétique x sexe
Mâle 63,0b 51,5b 38,8 57,8 56,6
Djallonké
Femelle 58,6b 45,9 44,7 57,2 57,0
Mâle 58,0b 89,2a 32,5 57,2 58,2
Croisé
Femelle 98,6a 35, lb 50,8 64,2 63,2
Au sein d'une même colonne, les valeurs affectées d'une même lettre ne sont pas significativement différentes.

4. Conclusion pondérale diminue avec l'âge et disparaît au sevrage. Il

est probable que la faible production laitière de la brebis
Phénotypiquement, le croisé Sahélien x Djallonké se Djallonké ne permette pas à l' agneau croisé de maintenir
rapproche plus du mouton sahélien que du Djallonké. sa supériorité de poids jusqu'à un âge avancé, malgré les
Cette ressemblance se manifeste dans la longueur des potentialités génétiques qu'il recèle. Il s'ensuit que
oreilles, la longueur de la queue et une présence l'amélioration de la productivité ne devrait pas
importante de taches sur le corps. Le poids à la naissance s'appuyer uniquement sur les performances pondérales
est plus élevé chez les croisés Djallonké x Sahélien que mais aussi sur la valeur laitière de la brebis.
chez les Djallonké de race pure, mais cette supériorité

218 Small Ruminant Research Network Workshop

 Caractères phénotypiques de différenciation et de ...

Fitzhugh, et G.E. Bradford. Westview Press, Boulder,

Références Colorado (E.-U.). p. 23 à 52.
Amégée Y. 1983. Le mouton de Vogan (croisé Djallonké x Harvey W.R. 1988. User's Guide for LSMLMW PC-Version.
Sahélien) au Togo. Revue d'élevage et de médecine Ohio State University, Columbus, Ohio (E.-U.). 79 p.
vétérinaire des pays tropicaux 36(l):79-84.
Charray J., Coulomb J., Haumesser J.B., Planchenault D. et
Amégée Y. 1984. Le mouton de Vogan (croisé Djallonké x Pugliese P.L. 1980. Les petits ruminants d'Afrique
Sahélien) au Togo. I. La production lactée et ses relations centrale et d'Afrique de l'Ouest: Synthèse des
avec la croissance des agneaux. Revue d'élevage et de connaissances actuelles. IEMVT (Institut d'élevage et
médecine vétérinaire des pays tropicaux 37(l):82-90. de médecine vétérinaire des pays tropicaux),
Fitzhugh H.A. et Bradford G.E. 1983. Productivity of hair Maisons-Alfort (France), Ministère de la coopération,
sheep and opportunities for improvement. In: Hair Paris (France). 295 p.
Sheep of Western Africa and the Americas. A Genetic ITCF (Institut technique des céréales et des fourrages). 1987.
Resourcefor the tropics. Publié sous la direction de H.A. Manuel d'utilisation de STATITCF. ITCF, Versailles
(France).

Small Ruminant Research Network Workshop 219

 Sheep production in Zambia: A review of sheep breeding
W.N.M. Mwenya

Department ofAnimal Science, University ofZambia

Box 32379, Lusaka, Zambia

Abstract population found in Zambia. The sheep found in the

commercial sector are the exotic type, namely,
Sheep breeding in Zambia is reviewed, particularly the Blackhead Persian, Dorset Horn, Dorper and Suffolk.
use of evaluation, selection and mating systems to The commercial sheep flocks are concentrated in
improve the economic traits of sheep. Lusaka and Central Provinces, near the main
Programmes to improve sheep production by consumption centres for marketed sheep.
evaluating and selecting the best animals should The sheep are kept for meat. Off-take in the
concentrate on the traditional sector. Open national commercial sector is 25% while in the traditional sector
nucleus flocks should be established for the Fat- tailed it is unknown. Most of the mutton consumed in urban
indigenous sheep which are widespread throughout the areas is supplied by the commercial sector. The annual
country. The improved genotype would be disseminated per capita consumption of mutton is estimated at 0.7 kg
by the extensive use of rams from the national nucleus of which the rural areas account for 80%. The marketed
flock. supply of meat from sheep is relatively small as it
accounts for less than one per cent of the total meat
supply. There is therefore a need to improve sheep
Production ovine en Zambie: production in order to increase annual per capita
consumption of mutton. Better breeding, nutrition,
étude de la reproduction et de la management, disease control and marketing can all
sélection des moutons contribute to this. This study reviews improving sheep
production through breeding.
Résumé
Cet article passe en revue les programmes de
reproduction et de sélection des ovins en Zambie, et
Breeding in production sectors
particulièrement les systèmes d'évaluation, de sélection
et de lutte utilisés pour améliorer les paramètres de Traditional sector
production des ovins.
Flock sizes in the traditional sector vary from a few to
Les programmes d'amélioration de la production about 50 animals. This sector keeps the two indigenous
ovine par l'évaluation et la sélection des meilleurs breeds, Fat-tailed and Thin-tailed sheep. The more
animaux devraient être axés sur le secteur traditionnel widespread of these is the Fat-tailed sheep, which
et des troupeaux reproducteurs nationaux ouverts de accounts for 63% of the indigenous sheep. Formal
moutons indigènes à queue grasse devraient être animal breeding practices such as evaluation and
constitués. Le génotype amélioré de ce type de mouton, selection of individual animals for performance are not
qui se rencontre sur l'ensemble du territoire, serait practised. All ages and sexes of sheep are left to run
disséminé par le recours à grande échelle à des béliers together at all times and mating is at random within the
issus de ce troupeau reproducteur national. flock.

Commercial sector
Introduction
Commercial sheep production started in 1979
Sheep in Zambia are produced by two farming sectors. (Productive Farming, 1981). Commercial farmers keep
The commercial sector has income generation as the Blackhead Persian, Dorset Horn, Dorper and Suffolk.
main priority of sheep farming. The traditional sector The Dorper is the most common breed because it has
places less emphasis on income and more on nutrition good characteristics such as a long breeding season, high
and subsistence. The traditional sector owns 64% of milk yield in the ewe and a good quality carcass
Zambia's 59,720 sheep. In this sector the sheep are (Productive Farming, 1983). Suffolk rams are used in
mainly of the Fat-tailed and Thin-tailed types, which are two- and three- breed crosses as the terminal sire to
indigenous. Traditional sheep production is improve lamb growth rate and carcass weight.
concentrated in the Southern, Eastern and Luapula Commercial farmers have claimed that slaughter
Provinces, accounting for 63% of the traditional sheep weights for Suffolk x Dorper crosses are 30% higher

Small Ruminant Research Network Workshop 221

W.N. M. Mwenya

than those for pure Dorper (Productive Farming, 1981). production of sheep; to upgrade the standards and the
There is no evaluation and selection taking place on efficiency of operation of both commercial and
these farms to accompany crossbreeding. traditional producers; and significantly expand the
average size of off-take from traditional flocks
(MAWD, 1986). The production targets are to increase
Breeding in research sheep population by an annual rate of 3.3% and to
increase slaughtering by an annual rate of 7%.
In 1 970 breeding work started at the Mazabuka Regional The strategy for achieving these objectives
Research Station, situated in the southern part of includes the improvement of marketing facilities and
Zambia. Its purpose was to evaluate pure Dorper and price incentives, improving extension and animal health
Blackhead Persian as meat animals. Traits considered services, strengthening applied research activities and
were birth weight, pre-weaning mortality rate, weaning establishing strategically located sheep multiplication
weight (at four months) and post-weaning weight gain units for distributing improved genotypes.
of male castrates up to the age of three years (MAWD, In Zambia, material, money and personnel
1 970). Dorper lambs were heavier at birth (4.2 kg) than resources are very limited. Improvement through
Blackhead Persian lambs (3.1 kg). However, the
breeding should concentrate on the traditional sector
pre-weaning mortality rate was better for the Blackhead
which has the largest population of unimproved sheep.
Persian ( 1 0%) than for the Dorper (24.5%). The average
weaning weight of single and twin Blackhead Persian Improvement could therefore make the largest impact.
lambs was 18.9 kg for females and 20.9 kg for males. Commercial farmers already have unlimited access to
At 3.5 years of age the Dorper castrates were about 58% improved genotypes through the import of proven rams,
heavier than the Blackhead Persian and the carcass although they do not evaluate and select individual ewes
weight was about 49% heavier (MAWD, 1972-73). for performance.
In May 1972 another sheep project was initiated at Breeding work in the traditional sector should
the Mansa Research Station, situated in the northern part concentrate on establishing an open nucleus flock for
of Zambia. The aim of this project was to evaluate the Fat-tailed sheep in each of Southern, Eastern and
meat production potential of indigenous sheep as Luapula Provinces. In these flocks evaluation and
purebred and as crossbreds to Dorper and Blackhead selection should concentrate on fertility, growth and
Persian sheep. The project was abandoned the same year viability traits. The improved genotypes would be
because of high mortality rates, the reasons for which disseminated by the extensive use of rams from the open
were unknown. nucleus flocks.
To improve carcass weights of marketed sheep
from the traditional sector, on-farm assessments of the
Breeding in extension best breed of sire in the terminal cross of indigenous
ewes with exotic breed should be carried out
In 1969 the Government established a National Stud concurrently with the establishment of open nucleus
Farm of Dorper and Blackhead Persian near Lusaka in flocks. Possible sire breeds to be assessed would be
the central part of the country. Breeding stock from the Dorper and Blackhead Persian.
Stud Farm were sold to farmers to establish small sheep
units in different parts of the country. Due to high
mortality rates and lack of expert management, the
Farm was abandoned in 1977. References
In 1985, the Government drafted a proposal to set MAWD (Ministry of Agriculture and Water Development),
up a ranch to keep indigenous breeds. The animals were Fourth National Development Plan. 1986. Agricultural
to be evaluated and selected for growth and hardiness to Sector: Sheep and Goats. pp. 50-52.
survive village conditions. Sires from the ranch were to MAWD (Ministry of Agriculture and Water Development).
be used on village ewes to improve the quality of village 1970. Animal Husbandry Section. Annual Report. pp.
flocks. This proposal has not been implemented. 100-114.
MAWD (Ministry of Agriculture and Water Development).
Annual Report. 1972/73. Animal Husbandry Section.
Future breeding work pp. 110-111.
The Government's main development objectives for Productive Farming. 1981. Sheep Field Day. 10 pp.
sheep are: to increase substantially the commercial Productive Farming. 1983. Dorper Field Day. 17 pp.

222 Small Ruminant Research Network Workshop

 Prospects for breeding for resistance to endoparasites in
small ruminants in Africa — a new ILCA research
programme
R. L. Baker', L. Reynolds2, A. Lahlou Kassi3, J.E.O. Rege3
Tekelye Bekelye , E. Mukassa-Mugerwa and B. Rey
lILCA, P. O. Box 46847, Nairobi, Kenya
2ILCA, P. O. Box 80147, Mombasa, Kenya
3ILCA, P. O. Box 5689, Addis Ababa, Ethiopia
IEMVT, France, seconded to ILCA, P. O. Box 5689, Addis Ababa, Ethiopia

Abstract chez les ovins et les caprins. La variabilité génétique de

cette résistance entre races de moutons non africains est
This paper briefly reviews the evidence for genetic bien établie. Certains documents font état d'une
resistance to endoparasites both between and within variabilité interrace en Afrique. Toutefois, la
breeds of sheep and goats. There is extensive evidence quasi-totalité des rapports publiés à ce jour pèche par
for genetic variation in resistance among breeds of sheep l'inadéquation du schéma expérimental, notamment
outside Africa. Some evidence exists for breed variation pour ce qui est du petit nombre d'animaux
in Africa. Almost all published reports to date suffer échantillonnés dans chaque race et du manque
from poor experimental design, particularly in terms of d'information sur la manière dont les animaux ont été
the small numbers of each breed sampled and lack of échantillonnés. Les preuves de l'existence de variations
information on how the animals were sampled. The génétiques de la résistance aux endoparasites à
evidence for genetic variations in resistance to l'intérieur d'une même race sont plus convaincantes.
endoparasites within breeds is more convincing. L'héritabilité est en moyenne de 0,35. A la lumière de
Heritabilities average about 0.35. In the light of this ces observations, le CIPEA a engagé un programme
evidence, ILCA has developed a Pan-African panafricain de recherche pluridisciplinaire visant à
multidisciplinary research programme to investigate étudier et à caractériser la résistance génétique aux
and characterise genetic resistance to endoparasites in endoparasites chez certaines races indigènes de
some indigenous African sheep and goat breeds. This moutons et de chèvres en Afrique. Parmi celles-ci
includes the evaluation of: Menz and Horro sheep breeds figurent les moutons Menz et Horro d'Ethiopie; les
in Ethiopia; Dorper and Red Maasai sheep and Galla moutons Dorper et Massai et la chèvre Galla et la petite
and Small East African goats in coastal Kenya; Peul Peul chèvre de l'Afrique de l'Est dans la région côtière du
(Fulani) and Djallonké sheep and West African Dwarf Kenya; les moutons peul peul et Djallonké et la chèvre
and Sahel goats in Senegal . The experimental design and naine d 'Afrique occidentale ainsi que la chèvre du Sahel
research protocol for this research programme is au Sénégal. Le dispositifexpérimental et le protocole de
outlined and some preliminary results of the research in recherche de ce programme sont exposés dans leurs
the Kenyan coastal site are presented. grandes lignes et certains des premiers résultats des
recherches effectuées dans la zone côtière du Kenya sont
présentés.
Perspectives de sélection sur la
résistance aux endoparasites
Introduction
chez les petits ruminants
Gastrointestinal nematodes impose severe economic
africains — un nouveau constraints on sheep and goat production in pastoral
programme de recherche du systems worldwide. Losses occur through mortalities,
reduced production due to subclinical parasitism and
CIPEA direct costs associated with control (Barger, 1982;
ILCA, 1992). Existing procedures to control
Résumé endoparasites rely almost exclusively on curative or
preventive treatment with anthelmintics. However,
Le présent article brosse un tableau rapide des widespread and indiscriminate use of these potent drugs
facteurs qui démontrent l'existence d'une résistance has resulted in the emergence of anthelmintic-resistant
génétique aux endoparasites inter- et intraspécifique strains of nematodes (Waller, 1991). There are also

Small Ruminant Research Network Workshop 223

R.L. Baker et al

increasing environmental concerns about chemical related to the Djallonké sheep (Bradford and Fitzhugh,
residues in meat and on pasture resulting from 1983).
anthelmintic use.
In most countries in sub-Saharan Africa,
endoparasite control options are limited. Access to Within-breed genetic variation
anthelmintics by smallholder livestock producers is
often restricted due to expense or unavailability. In Baker et al ( 1 992) tabulated and reviewed the substantial
addition, communal pastoral systems do not allow for body of evidence reporting heritability estimates for
regulated grazing as a means of lowering exposure to resistance to internal parasites in sheep and goats.
infective larvae on pasture. These considerations Resistance has commonly been assessed by either faecal
egg counts (FEC) or haematocrit (packed cell volume
provide an impetus to develop alternative or
— PCV). FEC has been shown to be a good indicator of
complementary control options which eliminate or at actual worm burdens (especially in young animals) but
least reduce the reliance on anthelmintics. Exploiting can depend on the parasite species. PCV is a good
genetic variations in host resistance to endoparasites is indicator of worm burdens for blood-sucking parasites
an attractive option both to control endoparasites and such as Haemonchus contortus.
improve the production efficiency of small ruminants. From the 14 Australian and New Zealand studies
that were reviewed by Baker et al (1992) the average
heritability for a single FEC measurement following
Genetic resistance natural or artificial infection was 0.32. Heritabilities
were similar for natural and artificial modes of infection
and infection with a number of different parasite genera
Breed differences (both single genus infections and mixed infections).
The average heritability of PCV (0.35) was similar to
There have been many reports since the mid- 1930s of that for FEC. The advantage in taking more than one
substantial variations among sheep breeds in resistance measurement of FEC depends on the repeatability
to internal parasites, particularly to Haemonchus estimate. The repeatability within a single artificial
contortus, Ostertagia circumcincta and Trichostongylus infection with Haemonchus contortus is high (0.60) and
colubriformis. Gray (1991) tabulated and summarised relatively little is to be gained in selection accuracy from
23 publications on this subject and this was expanded to more than one measurement (Woolaston et al, 1991).
34 studies in a review by Baker et al (1992). For
However, with natural challenge in New Zealand, the
example, the East African Red Maasai (Preston and
Allonby, 1978, 1979) and the Florida Native, St Croix, repeatability between two separate infections separated
Barbados Blackbelly and Navajo (Courtney et al, 1984; by an anthelmintic treatment is lower at about 0.30 to
Knight et al, 1973; Gamble and Zajac, 1992) are all 0.40. This results in a heritability of 0.53 for the average
relatively resistant to H. contortus in comparison with of two FECs taken in lambs at about five and seven
European breeds such as the Rambouillets or Merinos. months old (Baker et al, 1991).
There are also indications that some of the West African Estimates of genetic parameters (i.e. heritabilities
sheep and goat breeds, such as the Djallonké sheep and and repeatabilities) of resistance to endoparasites in
the West African Dwarf goat, may be relatively resistant sheep and goats in Africa are limited. Those available
to both internal parasites and trypanosomiasis (Osinowo were reviewed by Baker et al, (1992). Only three
and Abubakar, 1989; Smith, 1989). heritability estimates are available (Rohrer et al, 1991;
Nearly all the studies reviewed by Gray ( 1 99 1 ) and Baker, 1992 (unpublished)). These range from 0.22 to
Baker et al, (1992) are characterised by very poor 0.40, similar to those found in Australia and New
experimental design, due to both the numbers of animals Zealand. There are more repeatability estimates
of each breed tested, and lack of information on how the available. These are quite variable, ranging from 0.05 to
breeds were sampled. In addition, very few of the studies 0.42. In some studies the low repeatability estimates for
took account of variations among sires within breeds. FEC have been shown to be affected by the egg counting
The magnitude of the differences between sires can be procedure used. The improved modified McMasteregg
the same as the largest differences between breeds (Gray counting technique (MAFF, 1977) is recommended and
et al, 1987). Many of the breed differences reported usually gives higher repeatability estimates. It is
could reflect a single sire effect and hence should be important that the egg counting technique is
interpreted cautiously. standardised in any experimental study on resistance to
While many of the publications on breed variation endoparasites and the procedure finally used is clearly
for resistance to disease can be criticised due to their documented in subsequent publications.
poor experimental design, it is reassuring that some A number of divergent selection experiments in
breeds have been identified as resistant in a number of sheep for resistance or susceptibility to a number of
independent studies. This applies particularly to the different internal parasite genera are currently being
Florida Native, St Croix and Red Maasai sheep breeds. carried out. In Australia there are five studies, in New
It is very likely these breeds have a real resistance to Zealand three studies and in France one study. They
internal parasites. It is worthy of note that the St Croix were reviewed by Baker et al (1992). Successful
sheep breed originated from West Africa and is probably response to selection has been reported in some of these

224 Small Ruminant Research Network Workshop

 Prospects for breeding for resistance to etc...

studies (Baker et al, 1991; Cummins et al, 1991; added to the study to provide a breed comparison with
Woolaston et al, 1991; Windon, 1991). This confirms the Galla goats. Beginning in 1993 the Galla and Small
the moderate to high heritabilities for resistance to East African does (about 120 of each) will be mated to
endoparasites. five bucks of each of the same breeds in a diallel
crossbreeding design.
In Senegal both on-farm and on-station research
ILCA's research project components are planned. The on-farm study will
monitor Peul Peul (Fulani) sheep and Sahel goats in the
northern Sahelian arid zone and Djallonké sheep and
Objectives West African Dwarf goats in the southern humid zone.
This study was initiated in July 1 992 and utilises the PPR
The primary objective of the first phase of this research (Pathology and Productivity of Small Ruminants)
project is to identify and quantify genetic variations in
programme developed by ISRA (Institut Sénégalais de
resistance to internal parasites between and within some
indigenous breeds of sheep and goats in Africa. Other Recherche Agricole) and IEMVT (Institut d'Elevage et
important objectives include: determining the de Médécine Vétérinaire des Pays Tropicaux). The
specificity of genetic variations in resistance to different on-station component is being developed in
parasite genera; undertaking parasite epidemiology collaboration with the ILCA Trypanotolerance Thrust.
studies; establishing the relations between FEC, PCV It is currently proposed that this will be initiated in 1993
and actual worm burdens; and carrying out a at the ISRA Kolda research station in the southern humid
comparative economic evaluation of the use of breeding zone of Senegal. Initially the Peul Peul and Djallonké
and chemotherapy for parasite control (ILCA, 1991, sheep breeds will be evaluated, possibly in a diallel
1992). crossbreeding design.
The second phase of this project will investigate
immunological and genetic mechanisms of resistance to Experimental protocol
endoparasites. Breeds or groups of animals that are most
resistant and most susceptible to endoparasites will be The main experimental protocol that will be used to
used. This may identify immunological or genetic assess resistance to internal parasites at all research sites
markers of resistance. A desirable marker would be one will involve measurements of faecal egg counts (FEC)
that does not need infection and that would not be and packed cell volume (PCV) in young lambs and kids
influenced by other environmental or physiological from weaning (three months) to one year old. All
factors. Direct selection for resistance using animals will be given an anthelmintic treatment
parasitological measurements (e.g. FEC or PCV) (drenched) at weaning. Then they will be left
requires artificial challenge or unchecked field infection undrenched on pasture until a monitor group of about 50
with endoparasites. It is time-consuming, subject to lambs or kids (which are sampled weekly) reach a
sampling errors and may interfere with other traits under critical FEC which averages 1 ,500 epg (eggs per gram)
selection. in sheep and 1 ,000 epg in goats. At this stage all animals
in the group will have faecal and blood samples taken
and will then be drenched. The procedure will be
Experimental design repeated until the animals reach one year of age, about
three or four samplings, depending on the environmental
Phase 1 of the research was initiated in 1990 in coastal conditions. Ewes will be monitored at critical times
Kenya and in 1992 in Ethiopia and Senegal. throughout the reproductive cycle (e.g. mating, three
In Ethiopia, Menz and Horro sheep breeds are months post-mating, just before parturition and 1, 2 and
being evaluated at ILCA's Debre Berhan research site. 3 months post-parturition).
Each year 340 ewes and 1 0 rams of each breed will be At all sites both faecal cultures and worm counts
evaluated. Each breed will be split into two groups (150 from slaughtered animals will be undertaken to estimate
ewes mated and 20 unmated) and mated to lamb in two the numbers and relative frequency of different parasite
distinct seasons of the year (i.e. wet and dry). In addition genera. At each faecal sampling time a buffy coat test
to the primary objective of assessing resistance to for trypanosomiasis will be undertaken and blood stored
internal parasites, other studies using the same sheep for subsequent antigen identification of infection (in the
will assess parasite epidemiology, undertake Kenyan and Senegal sites).
reproductive physiology studies, measure food intake
and investigate fat metabolism and gut capacity.
In coastal Kenya (near Mombasa), during 1 990 and
1991 , approximately 200 Dorper and 200 Red Maasai x
Some preliminary Kenyan results
Dorper ewes were mated to 12 Dorper and 12 Red Results from the first crop of lambs born in this study at
Maasai rams in a diallel crossbreeding design. In 1992 Diani Estate (near Mombasa) in mid- 1990 showed that
a group of about 100 Red Maasai ewes were added to Red Maasai crosses required fewer drenches than
the study, also being mated to both sire breeds. At the Dorper lambs and had a higher survival rate to one year
same site Galla goats have been monitored in 1990 and of age (Reynolds et al, 1992). In this crop of lambs,
1 99 1 . In 1 992 a group of Small East African goats was animals were treated with an anthelmintic when their

Small Ruminant Research Network Workshop 225

R.L. Baker et al

PCV level dropped below 21%. This results in severely lamb crops. The small differences in live weight in
censored data which makes it difficult to assess favour of heavier Dorper lambs were not significant.
resistance in terms of FEC and PCV. For all subsequent There is also preliminary evidence from this data
lamb crops at Diani the new protocol described earlier for within-breed genetic variation. Restricted
was used which ensures FEC and PCV is measured in maximum likelihood (REML) animal model analyses
all animals at each sampling time. for lambs of all genotypes born in 1991 and 1992 were
Differences among breed groups for lambs born in carried out. Heritability estimates (± SE) were 0.21 ±
1991 (yearlings) and in 1992 (6-month old lambs) of 0.13 for FEC at weaning, 0.18 ± 0.1 1 for FEC in five-
resistance and production traits are shown in Tables 1 month old lambs and 0.24 ±0.16 for live weight in five-
and 2, respectively. month old lambs.

Table 1. Least squares means for live weight, PCV, FEC and mortality for lamb breed genotypes born in 1991.

Lamb breed Yearling lambs Mortality

(Sire breed x dam breed) (birth-1 yr) (%)
No. lambs Weight (kg) PCV(%) FEC (epg)
Dorper x Dorper (D) 56 21.7 27.3 1846 45
Dorper x (RM x D) 54 21.7 27.3 1419 37
Red Maasai (RM) x Dorper 68 21.7 29.7 1112 23
Red Maasai x (RMxD) 77 21.0 29.2 1197 23
Significance
Sire breed (SB) ns *** * ***
Dam breed (DB) ns ns ns ns
SBxDB ns ns ns ns
'.ns = P>0.05; * = P<0.05; »** = P<0.001.

Table 2. Least squares means for live weight, PCV, FEC and mortality for lamb breed genotypes born in 1992.
6-month old Mortality
lambs
Lamb breed No. Weight PCV FEC (birth-6 mo)
(Sire breed x dam breed) lambs (kg) (%) (epg) (%)
Dorper x Dorper (D) 44 12.2 21.9 3817 40
Dorper x (RMxD) 57 11.5 23.5 2467 20
Dorper x Red Maasai (RM) 4 11.0 23.2 85 43
RM x Dorper 47 11.9 24.3 1688 25
RM x (RMxD) 62 11.5 24.5 1759 22
RMxRM 7 11.7 26.3 1333 13
Significance
Sire breed (SB) ns ** ns *
Dam breed (DB) ns ns ns ns
SBxDB ns ns * ns
'ns = P>0.05; * = P<0.05; " = P<0.01.

Only 18 Red Maasai ewes were mated in 1992 References

which resulted in a small number of Dorper x Red
Maasai and straightbred Red Maasai lambs. Subsequent Baker R L, Watson T G, Bisset S A, Vlassoff A and Douch P
matings at Diani will ensure that about 100 Red Maasai G С. 1 99 1 . Breeding sheep in New Zealand for resistance
ewes are mated. Despite the small number of to internal parasites: Research results and commercial
straightbred Red Maasai lambs evaluated there is clear application. In: Gray G D and Woolaston R R (eds),
evidence that an increasing proportion of Red Maasai Breeding for disease resistance in sheep. Australian
blood results in more resistance to endoparasites than Wool Corporation, Melbourne, Australia. pp. 19-32.
Dorper lambs. This is shown by higher PCV, lower FEC Baker R L, Lahlou Kassi A, Rege J E О, Reynolds L, Bekele
and lower mortality. This trend was consistent in both T, Mukassa-Mugerwa E and Rey В. 1992. A review of
genetic resistance to endoparasites in small ruminants

226 Small Ruminant Research Network Workshop

 Prospects for breeding for resistance to etc..

and an outline of ILCA's research programme in this Osinowo O A and Abubakar В Y. 1 989. Appropriate breeding
area. Proceedings of the 10th Scientific Workshop of the strategies for small ruminant production in West and
Small Ruminant Collaborative Research Support Central Africa. In: Adeniji K O (ed). Improvement of
Programme, Nairobi, Kenya. (In press) small ruminants. Proceedings of the Workshop on the
Improvement of Small Ruminants in West and Central
Barger I A. 1982. Helminth parasites and animal production.
Africa. OAU (Organization of African Unity), Nairobi,
In: Symons LEA, Donald A D and Dineen J K (eds),
Kenya. pp. 71-84.
Biology and control of endoparasites. Academic Press,
Sydney, Australia, pp. 133-155. Preston J M and Allonby E W. 1978. The influence of breed
on the susceptibility of sheep and goats to a single
Bradford G E and Fitzhugh H A. 1983. Hair sheep: A general
experimental infection with Haemonchus contortus.
description. In: Fitzhugh H A and Bradford G E (eds),
Veterinary Record 103:509-512.
Hair sheep of Western Africa and the Americas. A
Genetic Resource for the Tropics. Westview Press, Preston J M and Allonby E W. 1979. The influence of breed
Boulder, Colorado, USA. pp. 3-22. on the susceptibility of sheep to Haemonchus contortus.
American Journal of Veterinary Research 33:817-823.
Courtney C H, Parker C F, McLure K E and Herd R P. 1984.
A comparison of the periparturient rise in faecal egg Reynolds L, Baker R L, Sherington J and Njubi D. 1992.
counts of exotic and domestic ewes. International Resistance to gastro-intestinal parasites in Dorper and
Journalfor Parasitology 14:377-38 1 . Red Maasai-Dorper crossbred sheep. Proceedings ofthe
Cummins L J, Thompson R L, Yong W K, Riffkin G G, 10th Scientific Workshop of the Small Ruminant
Collaborative Research Support Programme, Nairobi,
Goddard M E, Callinan A P L and Saunders M J. 1991.
Genetics of Ostertagia selection lines. In: Gray G D and Kenya. (In press)
Woolaston R R (eds), Breedingfor disease resistance in Rohrer G A, Taylor J F, Davis S K, Waruiru R M, Ruvuna F,
sheep. Australian Wool Corporation, Melbourne, Mwandotto В A J, McGuire T and Rurangwirwa F. 1 99 1 .
Australia. pp. 11-18. The use of randomly amplified polymorphic DNA
Gamble H R and Zajac A M. 1992. Resistance of St Croix markers in analysis of susceptibility to Haemonchus and
lambs to Haemonchus contortus in experimentally and Coccidia. In: Proceedings of the Ninth Small
naturally acquired infections. Veterinary Parasitology Ruminant-CRSP Scientific Workshop held at
41:211-225. Ketri-Muguga, Kenya, 27-28 February 1991. KARI
(Kenya Agricultural Research Institute)/SR-CRSP
Gray G D. 1991. Breeding for resistance to Trichostrongyle (Small Ruminant Collaborative Research Programme),
nematodes in sheep. In: Owen J В and Axford R F E Kenya. pp. 71-85.
(eds), Breeding for disease resistance in farm animals.
CAB (Commonwealth Agricultural Bureaux) Smith О В. 1989. Health package for the smallholder farmer
International, Wallingford, UK. pp. 139-161. in West and Central Africa. In: Adeniji K O (ed),
Improvement of small ruminants. Proceedings of the
Gray G D, Presson В L, Albers G A A, Le Jambre L F, Piper Workshop on the Improvement of Small Ruminants in
L R and Barker J S F. 1987. Comparison of within - and West and Central Africa. OAU (Organization of African
between-breed variation in resistance to haemonchosis Unity), Nairobi, Kenya. pp. 21 1-221.
in sheep. In: McGuirk В J (ed), Merino improvement
programs in Australia. Australian Wool Corporation, Waller P J. 1991. The status of anthelmintic resistance in the
Melbourne, Australia. pp. 365-369. world. Its impact on parasite control and animal
production. Paper prepared for the FAO Expert
ILCA (International Livestock Centre for Africa). 1991. Consultation on Helminth Infections of Livestock in
Proceedings of the Research Planning Workshop on Developing Countries (AGA/HIL/91/12), FAO (Food
Resistance to Endoparasites in Small Ruminants, 5-7 and Agriculture Organization of the United Nations),
February 1991, Addis Ababa, Ethiopia. 78 pp. Rome, Italy. 19 pp.
ILCA (International Livestock Centre for Africa). 1992. Windon R G. 1991. Resistance mechanisms in the
Genetic resistance to gastrointestinal parasitism in small
Trichostrongylus selection flock. In: Gray G D and
ruminants. ILCA Project Proposal (2.2/01-02/01 ). Woolaston R R (eds), Breedingfor disease resistance in
Knight R A, Vegors H H and Glimp H A. 1973. Effects of sheep. Australian Wool Corporation, Melbourne,
breed and date of birth of lambs on gastrointestinal Australia. pp. 77-86.
nematode infections. American Journal of Veterinary
Woolaston R R, Windon R G and Gray G D. 1991. Genetic
Research 34:323-327.
variation in resistance to internal parasites in Armidale
MAFF (Ministry of Agriculture, Fisheries and Food). 1977. experimental flocks. In: Gray G D and Woolaston R R
Manual of veterinary parasitological laboratory (eds), Breeding for disease resistance in sheep.
techniques. Technical Bulletin 18. Her Majesty's Australian Wool Corporation, Melbourne, Australia.
Stationery Office, London, UK. 129 pp. pp. 1-9.

Small Ruminant Research Network Workshop 227

 Estímate of genetic and phenotypic parameters for
the Dorper, Red Maasai and their crosses
P.M. Kiriro
Egerton College, Njoro, P.O. Box 536, Njoro, Kenya

Abstract Le génotype des animaux, l'année, la saison et le

mode de naissance, le sexe des agneaux et le mode
Data accumulated over 10 years (1971-1981) by the d'élevage influaient (P<0,01) sur tous les paramètres
Ministry of Livestock Development and FAO/UNDP de croissance. L 'hétérosis direct et I 'hétérosis maternel
(Food and Agriculture Organization of the United intervenaient pour une part importante dans le poids à
Nations/United Nations Development Programme) la naissance (4,50 et 4,89%), le poids au sevrage (10,95
were analysed to estimate the growth performance of et 6,50%) et le gain moyen quotidien (9,36 et 8,93%).
Dorper and Red Maasai sheep and their crosses. The La valeur estimée de l'héritabilité était de 0,14 ± 0,8,
heritability, heterosis, phenotypic and genotypic 0,08 ±0,05 et 0,10 ± 0,06 respectivement pour le poids
correlations were determined for birth weight, average à la naissance, le poids au sevrage et le gain moyen
daily gain and weaning weight. quotidien. Cesfaibles valeurs montrent que la sélection
The animals' genotype, year, season and type of n 'est pas le moyen à retenir pour améliorer les poids au
birth, sex of lamb and its rearing affected (P<0.01) all sevrage et les gains moyens quotidiens. D'après les
the growth parameters. Direct and maternal heterosis valeurs calculées des corrélations génétiques et
made sizeable contributions to birth weight (4.50 and phénotypiques des paramètres de croissance des
4.89%), weaning weight ( 1 0.95 and 6.50%) and average agneaux Dorper, on pourrait obtenir un accroissement
daily gain (9.36 and 8.93%). The heritability estimates des poids à la naissance et des gains moyens quotidiens
were 0.14 ± 0.8, 0.08 ± 0.05 and 0.10 ± 0.06 for birth en sélectionnant sur le poids des agneaux au sevrage.
weight, weaning weight and average daily gain,
respectively. These low estimates indicate that selection
will not be an appropriate option for improving weaning
weights and average daily gains. The genetic and
Introduction
phenotypic correlation estimates of the growth Maasai sheep are part of a wide variety of indigenous
characteristics of Dorper lambs suggest that birth weight fat-tailed hair sheep scattered over many parts of Kenya,
and average daily gain could be increased by selecting Uganda and Tanzania. They have a medium red-brown
for heavier lambs at weaning. body. Maasai sheep and particularly Red Maasai sheep
are found in large numbers in the Rift Valley Province
of Kenya. Red Maasai sheep attracted the attention of
Estimation des paramètres livestock scientists because of their large population,
wide distribution, uniformity of body colour and
génétiques et phénotypiques hardiness. It was also observed that with slightly better
chez les moutons Dorper, Red nutrition, disease control and selection, it was possible
for the Red Maasai sheep to increase in body size, grow
Maasaï et les produits de leur faster and produce more milk.
croisement The Dorper sheep were first introduced into Kenya
in 1952 at the then Katumani research station of the
Ministry of Agriculture. As a breed they were developed
Résumé in South Africa between 1945 and 1950, mainly at
Les données accumulées pendant dix ans Grootfontein , from Dorset Horn and Blackhead Persian
(1971-1981) par le Ministère du développement de sheep. They have a medium-white body with a black
l'élevage et la FAO/PNUD (Organisation des Nations head, neck and legs. The Breed Society was formed in
Unies pour l'alimentation et l'agriculture/Programme South Africa in 1950. By 1970, Dorper sheep had spread
des Nations Unies pour le développement) au Kenya ont to Eastern, Rift Valley and Coast Provinces of Kenya.
été analysées afin d'estimer les performances de In the 1970s more Dorpers were imported from Southern
croissance des moutons Dorper et Red Maasaï et des Africa for research and multiplication.
produits de leur croisement. L'héritabilité, I'hétérosis et When the Kenyan Government decided to initiate
les corrélations phénotypiques et génotypiques pour le a sheep and goat development project in collaboration
poids à la naissance, le gain moyen quotidien et le poids with FAO/UNDP in 1970, the Red Maasai was
au sevrage ont été déterminés. identified as the most important indigenous breed of

Small Ruminant Research Network Workshop 229

 P.M. Kiriro

sheep. Since then, short-term investigative research

aimed at characterising the Red Maasai has been
Materials and method
conducted. Studies have been carried out to determine
the relationships between body weight and body History
measurements (FAO, 1975). Studies to compare Red
Maasai and the Blackhead Persian have also been The studies that generated this data were carried out at
Naivasha National Animal Husbandry research station
conducted in Tanzania. There were considerable
by officials of the Ministry of Livestock Development
differences in body weights and average daily gains at and FАО/UNDP between [Link] study was
different stages of growth and between sexes. conducted at the Ol'Magogo division of the station. The
Red Maasai were bred pure to ascertain their foundation flock was established from Dorper and Red
production potential. They were also crossbred with Maasai sheep already in the country. Dorper sheep were
Dorper and Hampshire rams to incorporate the superior from Mariakani, Katumani and Kiboko research
production traits of the two exotic breeds into the Red stations. The Red Maasais were purchased from
Maasai sheep. The Dorper sheep was also used to Nanyuki, Gilgil and Baringo (Figure 1).
improve other indigenous fat-tailed sheep in many parts
of arid Kenya. The Dorper was considered suitable to Activities on foundation flock
Kenyan conditions because it was developed from a
desert fat-rumped sheep (Blackhead Persian) very The following activities were carried out:
similar to the Somali or the Blackhead sheep of Northern pure breeding of Dorper and Red Maasai to expand
Kenya. According to Erasmus ( 1 968), the Dorper has an the flocks;
ideal body conformation that adapts it to dry conditions.
short-term breed characterisation studies;
It is also hardy, has a high quality carcass, high lambing
percentages, good mothering qualities and is extremely studies on the effects of frequent mating (joining
suitable for upgrading haired breeds of sheep. every eight months) on productivity;
The Dorper is probably the breed in Kenya that has disease tolerance; and
been studied the longest in research stations. In 1975,
crossbreeding Dorper and Red Maasai to measure
the Sheep and Goat Development Project published 15
the improvement in productivity and develop a
years of production data on Dorper sheep. The data
new breed.
showed that the Dorper has a birth rate of about 84%, a
24% twinning rate, a birth weight of about 3.5 kg,
average daily gain of about 1 17 g and lamb mortality of Data collected and records
15%.
The breeding policy and the entire management
So far, however, no work has been done to
strategies were designed to collect the maximum
establish the level of some economically important amount of data on the dam, sire and lambs. The data
genetic parameters, e.g. heritability estimates of birth collected in this study covered the pre-mating and
weight, average daily gain and weaning weight. The mating periods, lambing, lamb rearing, weaning,
amount of hybrid vigour resulting from crossbreeding postweaning and treatment records. This study reports
Dorper with Red Maasai sheep has also not been the results of data analysed up to weaning.
established. A breeding project was initiated at the
OTMagogo division of the National Animal Husbandry Statistical analysis
Research Station, Naivasha and managed by the Sheep
and Goat Development Project/FAO in 1970. It was Data on growth characters were analysed by ANOVA
expected to generate data to help investigate the using SAS (1982). The statistical model was: Yijklmn =
importance of Dorper and the Red Maasai and their U + B¡ + (YW); + Rk + Si + Em + eijklm where U is overall
crosses. The sheep data from OTMagogo were analysed mean, В is breed of lamb, (YW) year of birth-season
to fulfill these objectives: effect, R is type of birth, S is sex of lamb effect, E is age
of lamb effect and e is the random error. The effects of
• to estimate breed means for birth weight, average crossbreeding on growth characters were analysed using
daily gain and weaning weight of Dorper, Red a model similar to that used by Robinson et al (1981)
Maasai and their crosses; which was executed using SAS (1 982). Analyses for the
• to estimate heritability of birth weight, growth rate genetic parameters, i.e. heritabilities and correlations for
growth characters were conducted using the Least
and weaning weight; Square Maximum Likelihood program (LSML76) of
• to estimate genetic and phenotypic correlations Harvey (1982). Heritabilities and correlations analyses
between birth weight, average daily gain and were restricted to the Dorper breed because of higher
weaning weight; and numbers of observations of sires, and to lambs whose
sires were known. The genetic and phenotypic
• to estimate heterosis of birth weight, average daily correlations were calculated by paternal half-sib
gain and weaning weight. procedure (Falconer, 1981).

230 Small Ruminant Research Network Workshop

 Estimate of genetic and phenotypic parameters for etc..

Figure 1. Geographical location ofNaivasha.

34° 35° 36° 37° 38° 39° 40° 4IC 42° E
—i— —i—

5°N
SUDAN
ETHIOPIA

LakeTurkona
-3°

UGANDA

KENYA

ELDAMA RAVINE
О NANYUKI
О
KISUMU ° OL JORO OROK
0 NAKURU
° GILGIL
O NAIVASHA
Lake Victoria

° NAIROBI

O MACHAKOS

NZANIA ^\ О Kl BOKO

O/MALINDI

Scale N5 000 000

miles 50 0 50 Ю0 MARIAKANI,
O/ MOMBASA
kms 50 0 50 ЮО 150

Indian Ocean
5° S

Results and discussion of all the breed groups (Table 1). The magnitude of
difference in growth characters was closely associated
with divergence in genotype (Figures 2 and 3). Lambs
Genetic and environmental effects on which were heavy at birth performed better than light
growth characters lambs.
Breed of lamb affected (P<0.001) birth weight (BW), Year of birth-season affected (P<0.001) birth
weaning weight (WW) and average daily gain (ADG) weight, weaning weights and average daily gain. The

Small Ruminant Research Network Workshop 231

P.M. Kiriro

Table 1. Least squares means and standard errors for birth weight (BW), weaning weight (WW) and average daily gain (ADG)
for each breed and sex of lamb.
Body characters (kg, SE)
Category BW WW ADG
Breed of lamb
Dorper (D) 3.64 (0.04)a 19.50(0.23)" 159.35(2.27)"
Red Maasai (R) 2.90 (0.05)c 15.56 (0.28)c 127.06 (2.78)d
DxR 3.33 (0.06)b 17.51 (0.32)b 138.61 (3.23)c
DxDR 3.64 (0.05)a 19.63 (0.26)" 161.59 (2.60)b
DxDDR 3.49 (0.1 3)b 20.82 (0.70)" 169.91 (7.06)"
D x DDDR 3.60 (0.1 3)a 19.44(0.71)" 161.69 (7. 19)b
DDR x DDR 3.58 (0.05)a 19.80 (0.30)a 162.88 (2.99)b
Sex
Male 3.53 (0.05) 19.44(0.27) 158.80(2.70)
Female 3.38 (0.05) 18.34(0.27) 150.10(2.71)
Letters on left of x represent sire breed.
SE = standard error.
abed = Means within a column carrying different superscripts are significantly different (P<0.001 ).

Figure 2. Growth trends ofDorper (D), Red Maasai (R) and Figure3. Effects of percentage of Dorper in the cross on
their crosses. BW, WW and ADG. Percentages ofDorper in the
cross are represented bv: 0 = Red Maasai (R); 100
Body weight (kg) = Dorper (D); 50 = DxR; 75 = DxDR; 87.50 =
DxDDR; and 93. 75 = DxDDDR.
25
Body weight (kg) Growth rate per day (g)
25 170
20

20 160
15

15 - 150
10 -

10 140
BW

ADG
WW
130
BW WW
Growth characters

effect was closely related to rainfall. Type of birth, 120

rearing and sex of lamb affected (P<0.001) all growth 0.00 50.00 75.00 87.50 93.75 100.00
characters. Percentage

Heterosis Maternal heterosis estimates of 0.16 kg (4.89%),

1.14 kg (6.50%) and 12.8 g (8.93%) were obtained for
birth weight, weaning weight and average daily gain,
Average direct heterosis shows that crossbred lambs
respectively. The results show that maternal heterosis
were heavier by 0. 1 5 kg (4.5%) for birth weight, 1 .92 kg
contributed significantly to birth weights, weaning
(10.95%) for weaning weight and 13.8 g (9.36%) for
weights and average daily gains of crossbred lambs.
average daily gain. The study shows that heterosis for
average daily gain was almost similar to that obtained Similar observations were reported by Farid et al ( 1 977),
for weaning weights. Similar observations were made Holtman and Bernard (1969) and Willham (1970).
by Sidwell and Miller (1971a, 1971b), Holtman and All lambs from known sires were used in the
Bernard (1969) and Singh et al (1981). heritability estimates.

232 Small Ruminant Research Network Workshop

 Estimate of genetic and phenotypic parameters for etc...

Heritability estimates Middleburg, C.P. The Non-wooled Mutton Breeds.

Government Printer, South Africa, Pretoria. pp. 55-63.
Heritability estimates (h ) were obtained from 1202 Falconer D S . 1981. Introduction to quantitative genetics. 2nd
observations (Table 2). The heritability estimates were edition. Longman, London, UK. 340 pp.
low, indicating that it would be difficult to improve FAO (Food and Agriculture Organization of the United
weaning weights and average daily gains through Nations/Sheep and Goat Development Project). 1975.
selection. Similar observations were reported by Johor 15 Years' production data on Dorper sheep at Katumani
and Norton (1977), Suvindra and Tomar (1982) and Research Station, Machakos. Technical note 12.
Shelton and Menzies (1968). The uniformity of the Katumani Research Station, Machakos, Kenya.
growth traits of the flock might have led to low
Farid A, Makarechian M and Sefidbakht N. 1977.
heritability estimates for weaning weights and average Crossbreeding of Iranian fat-tailed sheep: Lamb
daily gains. performance of Karakul, Mehraban and Naeni breeds.
Journal ofAnimal Science 44:542.
Correlation estimates Holtman W В and Bernard С 1969. Effects of general
combination ability and maternal ability of Oxford,
Genetic and phenotypic correlations of growth Suffolk and North Country Cheviot breeds of sheep.
characters for Dorper lambs are shown in Table 3. Journal ofAnimal Science 28: 155.
Positive genetic correlations were obtained Johor K S and Norton N W. 1977. Genetic study on birth and
between birth weight and weaning weight and between weaning weight of Suffolk and Torghee sheep. Indian
weaning weight and average daily gain. Veterinary Journal 54:902.
Phenotypic correlation between birth weight and Robinson O W, McDaniel В T and Rincon В T. 1981.
weaning weight and weaning weight and average daily Estimation of direct and maternal additive and heterotic
gain were positive. effects from crossbreeding experiments in animals.
Journal ofAnimal Science 52:45.
Conclusions SAS (Statistical Analysis System). 1982. SAS user's guide:
Statistics. SAS Inc., Cary, North Carolina. USA, 404 pp.
The results obtained imply that little change in either
Shelton M and Menzies J W. 1 968. Genetic parameters of some
birth weight or average daily gain would be expected by
performance characteristics of range fine wool sheep.
selecting for these traits. They are genetically and Journal ofAnimal Science 27: 1219.
phenotypically only very slightly negatively correlated.
However, significant improvements to birth weights and Sidwell G M and Miller LR. 1971a. Production in some pure
average daily gains could be obtained by selecting for breeds of sheep and their crosses. I. Reproductive
heavier weaning weights. efficiency in ewes. Journal of Animal Science
32:1084-1089.

References Sidwell G M and Miller LR. 1971b. Production in some pure

breeds of sheep and their crosses. II. Birth weights and
Erasmus L S. 1968. The small stock industry in South Africa. weaning weights of lambs. Journal of Animal Science
Agricultural Research Institute of the Karoo Region, 32:1090-1094.

Table 2. Estimate of heritabilities (h ) and standard errors of growth characters of Dorper lambs.

Number Number of
of sires observations (lambs) h2(SE)
Character
Birth weight 44 1202 0.138(0.060)
Weaning weight 44 1202 0.08 (0.051)
Average daily gain 44 1202 0.107(0.055)
All Iambs from known sires were used in the heritability estimates.

Table 3. Genetic and phenotypic correlations of growth characters for Dorper lambs.

Growth character Birth weight Weaning weight Average daily gain

Birth weight 0.546 -0.038

Weaning weight 0.199 0.838

Average daily gain -0.022 0.942

Genetic correlations above diagonal, phenotypic below.
Correlations were estimated from 1202 observations.

Small Ruminant Research Network Workshop 233

P.M. Kiriro

Singh В P, Arora C L, Bohra S D J, Bapna D L and Nivsarkas Suvindra К and Tomar N S. 1982. Genetic studies on
A E. 1981. Notes on the variation in the pre- and post- Corriedale sheep in India. Indian Veterinary Journal
weaning body weights of Malpura and Sonad, and their 59:948.
crosses with Suffolk and Dorset. Indian Journal of Wfflhain R L 1970 ^^ co ences of crossbreeding.
Ammal Sconces 26:261. Journal ofAnimal Science 30:690-693.

234 Small Ruminant Research Network Workshop

 Productivity of Blackhead Persian sheep in Tanzania
S.M. Das1, J.E.O. Rege2 and Mesfin Shibre2

Zonal Research and Training Centre, Mpwapwa, Tanzania

ILCA, Addis Ababa, Ethiopia

Abstract modifié par la saison (P = 0,59). Tous les indices ont

montré une tendance à la baisse durant la plus grande
Weight of lambs weaned (at 150 days) per ewe per year partie de la période d'étude (de 1966 à 1981). Un léger
(Index I), per kg liveweight of ewe per year (Index II), redressement des indices II et III a été observé, sans
per kg metabolic body weight ( w ' ) of ewe per year qu 'il se soit maintenu par la suite. L 'agnelage de saison
(Index III), and per lifetime of ewe regardless of ewe sèche était associé aux niveaux les plus élevés de
weight (Index IV), were computed for 3022 ewes at productivité que reflètent les indices I (P<0,01) et II
Malya and West Kilimanjaro research centres in (P = 0,0). Le niveau de production observé à Malya
Tanzania. The data analysed covered a period of 24 dépassait (P<0,01) toujours celui enregistré à West
years (1965-1988). Least squares means for the four Kilimanjaro, avec des poids supérieurs de 3 kg, 152 g,
respective indices were 24.6 kg (SE 3.19), 675 g (SE 0,45 kg et 7,2 kg respectivement pour les quatre indices.
1 5.7) 1 .8 kg (SE 0.04) and 63 kg (SE 1 .6). Year, seasons Bien que les poids au sevrage relevés à Malya aient été
and location significantly (at least P<0.05) affected all plus élevés que ceux enregistrés à West Kilimanjaro,
the indices except index IV which was not affected une grande partie des différences interstations entre les
(P=0.59) by season. All indices showed a tendency to indices de productivité a été attribuée aux différences de
decline during most of the study period (1966 through mortalité des agneaux avant le sevrage.
1981). Indices II and HI exhibited a small recovery
thereafter but this was not subsequently maintained. Dry
season lambing was associated with the higher
productivity indicated by index I (P<0.01) and index II Introduction
(P=0.0). Productivity at Malya was consistently superior
(P<0.01) to that at West Kilimanjaro, being 3 kg, 152 g, In Tanzania, the majority of traditional sheep are kept
0.45 kg and 7.2 kg better, respectively, for the four extensively, mostly in mixed livestock crop farming
indices. Although Malya was superior to West systems. The Blackhead Persian (ВНР) sheep comprise
Kilimanjaro in lamb weaning weight, a large proportion a large proportion of the adapted sheep breeds. The ВНР
of the between-station differences in productivity sheep were acquired in the early 1960s (Das and
Sendalo, 1991) with the purpose of improving the
indices was attributed to differential preweaning lamb
indigenous sheep population through crossbreeding.
mortality. The breed is mainly located at government research
centres and parastatal farms and is used for meat
production.
Productivité du mouton Overall meat production from sheep is affected by
Blackhead Persian en Tanzanie several variables, the most important of which are the
fertility, prolificacy and mothering ability of the ewes
Résumé and lamb survivability and weight. Productivity has
been measured from these traits by using indices which
Les poids d'agneaux sevrés (à 150 jours) par account for possible pre-weaning dam effects (Wilson,
brebis et par an (indice I), par kilo de poids vifde brebis 1987). Productivity indices can be used for selection
par an (indice II), par kilo de poids métabolique (F"' ) purposes as well as for comparing flocks when measured
de brebis par an (indice III) et par carrière de femelle either for a fixed period of time or over the entire life
des brebis, quel que soit le poids de la brebis (indice IV), time of the ewe. To incorporate feed requirements,
ont été calculés pour 3 022 brebis aux stations de productivity indices may be calculated as weight (kg) of
recherche de Malya et de West Kilimanjaro en lambs weaned per unit weight (kg) of ewe live weight
Tanzanie. Les données analysées avaient été recueillies or unit of ewe metabolic weight (kg) at parturition.
sur une période de 24 ans (1965-1988). Les moyennes In this study, the productivity of ВНР sheep at both
des moindres carrés des quatre indices étaient research centres (Malya and West Kilimanjaro) was
respectivement de 24,6kg (±3,19), 675 g(± 15,7), 1,8 kg calculated using indices (I to III) proposed by Wilson
(± 0,04) et 63 kg (± 1,6). L'année, la saison et la station (1987). A fourth index was developed to calculate total
avaient un effet significatif (au moins P<0,05) sur tous lamb mass production per lifetime of ewe. This index
les indices, exception faite de l 'indice IV qui n 'était pas could also be expressed as the lamb mass produced for

Small Ruminant Research Network Workshop 235

S.A. Das, J.E.O Rege and Mesfin Shibre

each lambing when mean flock size is used in the Index I x 1000
denominator. female postpartum weight
Index П1 = weight (kg) of young produced per kg
Materials and methods metabolic weight of female per year
The records of ВНР sheep used in this study were Index I
collected from two Livestock Research Centres —
female postpartum weight0 73
Malya and West Kilimanjaro in Tanzania. The Malya
Research Centre lies at approximately 4 S latitude and Index rV = total litter weight (kg) at 150 days per
37°E longitude, at an altitude of about 1050 metres female during lifetime
above sea level. The Centre covers an area of 1800 ha
of which 1400 have natural vegetation of savannah Lamb mass (kg)
grassland. The natural vegetation at Malya comprises Index ГУ
Combretum wood species and grass species of average number of parturitions per flock
Bothrichloa, Hyparrhenia, Themeda and Cynodon. A
few paddocks are sown with Chloris gayana. The analyses were carried out by fixed effects
The West Kilimanjaro Research Centre lies models using the General Linear Model (GLM)
approximately 3°S latitude and 39°E longitude, at an procedure of SAS (1987) software package. The fixed
altitude of about 1270 metres above sea level. It lies on effects included in the model were year and season of
the rain shadow region of Kilimanjaro mountain and parturition, age of ewe and location. The residual mean
covers an area of 5300 ha. Pennisetum schimperi is the square was used as the error term to test the significance
dominant grass species in the natural pastures. of all differences evaluated among classes. Where
Cultivated pastures include Digitaria scalprum and analysis of variance depicted significant differences for
Bothrichloa spp. variables, linear contrasts of least squares means were
At both Centres, sheep were grazed from 0800 to estimated to test pairwise differences within factors.
1700 hours and were housed indoors at night. Sheep
were bred in both the wet and dry seasons. The wet
breeding season extended from March to May, while the Results and discussions
dry breeding season extended from September to
November. Each breeding season consisted of 50-70
Productivity indices I, II and III
days and the ratio of rams to ewes in breeding groups
was about 1 :40. Ewe lambs joined the mating groups at Overall least squares means of indices, I, П and III are
the age of 72 weeks. Lambs were ear-tagged soon after presented in Table 1 . Year, season of birth and location
birth and left to suckle their dams during grazing until significantly (P<0.05) affected the three indices, but age
weaning at 16 weeks of age. Lambs were separated at of ewe at parturition was not significant (P>0.05) for any
weaning by sex into different weaner flocks. Records of the indices. All indices tended to decline over the
taken on lambs included birth and weaning weights and years from 1969 onwards. However, while index I
weights at 24, 48 and 72 weeks. Ewes were weighed at (Figure 1) exhibited a small recovery in the 1986/87
mating, lambing and weaning. period, indices II (Figure 1) and III (Figure 2) showed
substantial recoveries in two periods — 1982 and
Statistical analyses 1986/87. Indeed productivity in these periods increased
beyond the base year (1966) levels. On the other hand,
The productivity indices were constructed for individual index IV (Figure 2) never recovered to levels attained in
ewes using data on corrected litter weight at 1 50 days, the late 1960s.
lambing interval, survival of young and post-partum Year effects on indices were related to climatic
weight of ewe. Lamb live weights were adjusted to 150 factors, especially the amount of annual rainfall. The
days and additionally preadjusted for effect of sex. The decrease in availability of fodder in years of low rainfall
corrected weights at 150 days were used to calculate tended to lower weaning weights of lambs and affected
productivity indices I, II and III. The fourth index, index reproductive performances of ewes, thereby increasing
IV, was defined as the lifetime productivity per ewe in parturition intervals. This explains variations in the
terms of lamb mass produced. To obtain a measure of productivity of ewes between years. Periods
lamb mass (kg) per female per lambing, Index IV was characterised by increased lamb mortality, increased
further divided by the average number of parturitions per lambing intervals and reduced lamb growth are expected
flock. The four productivity indices were calculated as
to be associated with low productivity. A similar
follows:-
relationship between year of parturition and pro
Index I = weight (kg) of young per female per year ductivity of station-managed sheep has been reported by
litter weight (kg) at 1 50 days x 365 Murayi et al (1987) in Rwanda. Murayi et al (1987)
attributed differences between years to amount of
subsequent parturition interval
rainfall and availability of forages.
Index II = weight (g) of young produced per kg Ewes that lambed in the dry season had signi
live weight of female per year ficantly higher productivity indices than those lambing

236 Small Ruminant Research Network Workshop

 Productivity of Blackhead Persian sheep in Tanzania

Table 1. Least squares means of productivity indices 1, 11 and 111 by year, season, age and location sub-classes.

Productivity index
I (kg) IKg) HI (kg)
Variable Level N Mean se N Mean se N Mean se
Overall 3022 24.62 3.19 1304 675.0 15.68 1304 1.76 0.04
Year (0.0001) (0.0001) (0.0001)
1965 12 29.45 3.19 - - - - - -
1966 52 26.52 1.59 19 631.1 73.9 19 1.63 0.19
1967 134 34.29 1.01 72 797.4 39.3 72 2.12 0.10
1968 110 28.82 1.09 63 707.5 40.6 63 1.86 0.10
1969 137 30.56 0.95 106 829.3 31.9 106 2.21 0.08
1970 145 24.79 0.94 104 699.7 31.7 104 1.84 0.08
1971 129 27.05 1.00 97 768.6 33.4 97 2.01 0.09
1972 132 18.68 0.98 97 582.8 32.9 97 1.53 0.07
1973 150 23.21 0.91 123 635.3 28.5 123 1.67 0.08
1974 127 21.86 0.98 100 626.0 31.4 100 1.64 0.08
1975 130 24.57 0.97 87 738.8 33.5 87 1.93 0.09
1976 153 23.00 0.89 70 518.7 37.5 70 1.37 0.09
1977 235 24.93 0.75 73 610.5 36.3 73 1.61 0.09
1978 171 21.19 0.86 48 533.8 45.0 48 1.43 0.11
1979 184 22.20 0.84 36 542.6 52.4 36 1.44 0.14
1980 189 26.82 0.81 18 580.3 73.4 18 1.52 0.19
1981 202 24.42 0.79 13 487.1 86.7 13 1.31 0.22
1982 139 20.69 0.94 8 905.3 110.1 8 2.29 0.28
1983 148 20.45 0.91 21 682.5 69.9 21 1.78 0.18
1984 113 21.62 1.05 43 725.3 48.8 43 1.85 0.13
1985 89 18.68 1.18 48 599.1 45.8 48 1.48 0.12
1986 77 22.46 1.28 25 729.7 63.3 25 1.79 0.16
1987 60 28.58 1.44 31 962.3 56.9 31 2.35 0.15
1988 4 25.99 5.44 2 631.3 217.0 2 1.71 0.55
Season
Dry 1365 25.65 0.46 637 699.4 19.0 637 1.82 0.05
Wet 1657 23.58 0.45 667 650.1 18.6 667 1.69 0.05
Age of ewe (yr)
<2 471 23.65 0.57 165 639.2 27.8 165 1.66 0.07
2+-3 904 24.31 0.46 375 661.9 20.5 375 1.73 0.05
3+-4 664 25.07 0.50 289 673.1 21.8 289 1.75 0.06
4+-5 483 23.90 0.57 206 634.5 25.3 206 1.65 0.07
5+-6 273 24.97 0.73 132 700.1 30.1 132 1.82 0.08
6-<4 150 25.91 0.94 85 724.2 36.7 85 1.88 0.09
>7 77 24.49 1.29 52 692.0 46.4 52 1.81 0.12
Location
Mal ya 1723 26.15 0.43 719 751.0 18.8 719 1.98 0.05
WK 1299 23.08 0.47 585 599.0 18.5 585 1.53 0.05
Figures in parenthesis are tail probabilities for the F-tests.
N = number of observations.
se = standard error.
WK = West Kilimanjaro.

Small Ruminant Research Network Workshop 237

S.A. Das, J.E.O Rege and Mesfin Shibre

in the wet season. This is because lambs born in the dry longer parturition intervals than older ewes. Similar
season were weaned in the middle of the subsequent wet results have been reported in sub-Saharan Africa
season so were able to obtain adequate milk from their (Wilson, 1987; Lebbie, 1991). Differences between
dams and could graze the lush abundant forages locations were significant for these productivity indices.
available. However, lambs born in the wet season were Ewes had lower indices at West Kilimanjaro compared
weaned mid-dry season. They gained less weight by to those at Malya. The lower productivity indices at
weaning time due to scanty and low quality fodder. In a West Kilimanjaro could be attributed to: lower weaning
weight of lambs; higher preweaning mortality of lambs;
study of the effect of season of parturition on
and the presence of older ewes at the station.
productivity in Sudan Desert sheep, Suleiman et al
(1990) observed that productivity indices were higher
for ewes lambing in the wet summer than in other Index IV and mean lambs mass
seasons.
Overall least squares means of index IV and lamb mass
In the present study, the age of ewe did not per lambing are presented in Table 2. Index IV tended
significantly affect any of the productivity indices, to decline from 1969 onwards, with a small recovery in
though younger ewes tended to have numerically lower the 1978-1980 period (Figure 2). Ewes at Malya had
indices. This was mainly because younger ewes had significantly higher Index IV values and lamb mass per

Table 2. Least squares means ofproductivity index IV and lamb mass per lambing by year, season, age and location sub-classes.

Index IV, kg Lamb mass per lambing, kg

Effect Level N Mean se N Mean se
Overall 987 63.03 1.62 987 20.81 0.22
Year (0.0001) (0.0001)
1966 11 18.67 12.93 11 21.63 1.74
1967 16 28.99 10.74 16 26.52 1.44
1968 31 35.67 7.62 31 21.95 1.03
1969 21 62.46 9.29 21 18.84 1.25
1970 19 84.67 9.73 19 22.26 1.31
1971 29 101.05 8.11 29 21.73 1.09
1972 42 67.66 6.55 42 21.76 0.88
1973 38 52.17 6.97 38 19.19 0.94
1974 35 85.63 7.09 35 23.94 0.96
1975 55 85.01 5.71 55 21.31 0.77
1976 29 85.62 7.99 29 20.47 1.08
1977 28 86.38 8.15 28 19.41 1.09
1978 68 41.95 5.20 68 14.32 0.70
1979 51 54.96 6.02 51 19.99 0.81
1980 72 77.04 5.07 72 23.75 0.68
1981 59 81.67 5.56 59 23.97 0.75
1982 60 77.06 5.52 60 22.05 0.74
1983 51 62.39 6.06 51 18.97 0.81
1984 98 64.33 4.31 98 19.01 0.58
1985 39 40.85 6.90 39 16.49 0.93
1986 38 47.18 6.81 38 20.39 0.91
1987 36 54.67 7.19 36 18.78 0.97
1988 61 52.80 6.15 61 21.53 0.83
Season
Dry 375 64.05 2.49 375 20.13 0.34
Wet 612 62.01 2.45 612 20.49 0.33
Location
Malya 600 66.64 2.08 600 22.56 0.28
WK 387 59.42 2.79 387 19.05 0.37
N = number of observations.
se = standard error.
WK = West Kilimanjaro.

238 Small Ruminant Research Network Workshop

 Productivity of Blackhead Persian sheep in Tanzania

Figure 1. Trends in productivity indices I and II during the Figure 2. Trends in productivity indices III and IV during
study period. the study period.

Index I, kg Index III, kg

2.6 -i
40
a) Index I c) Index III
2.4 -

35
2.2

30 2

1.8 -
25 -
1.6

20 -
1.4 -

1.2 -^i—l—i—i—l—l—l—l—l—I—I—l—l—l—l—l—i—i—l—i—i—i—i—i—
15 i i i i i i i i i i i i i i i i i i i i i i i r~
1966 68 70 72 74 76 78 SO 82 84 86 88
1965 67 69 71 73 75 77 79 81 83 85 87
Year of lambing
Year of lambing

Index II, g Index IV, kg

1000 -i 120
d) Index IV
b) Index II
900 100 -

800 80

700 - 60

600 40

500 20 -

400 -i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i i i i i 0 ~H—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i—i i i i i i i r~

1966 68 70 72 74 76 78 80 82 84 86 88 1966 68 70 72 74 76 78 80 82 84 86 88
Year of lambing Year of lambing

lambing than those at West Kilimanjaro. The superiority especially during the preweaning period, could also be
of Malya was mainly due to lower weaning weights of improved in order to increase production.
lambs born at West Kilimanjaro resulting from poor
nutrition.

References
Conclusion Das S M and Sendalo D S. 1991. Small ruminant research
High variations in productivity indices show that the highlights in Tanzania (1960-1989). Ministry of
productivity of ВНР sheep in Tanzania can be improved. Agriculture, Livestock Development and Co-operatives,
Strategies should be developed to improve the Dar es Salaam, Tanzania. 40 pp.
management systems, such as improving pastures and Lebbie S H В . 1 99 1 . A na¡ysis ofthe reproductive performance
forages and general nutrition. Ewes should be fed of indigenous small ruminants in selected countries in
concentrates during the suckling period. Lamb care, sub-Saharan Africa. Addis Ababa, Ethiopia. 65 pp.

Small Ruminant Research Network Workshop 239

S.A. Das, JE. O Rege and Mesfin Shibre

Murayi T, Sayers A R and Wilson R T. 1987. La productivité Suleiman A H, Sayers A R and Wilson R T. 1990. Evaluation
des petits ruminants dans les stations de recherche de ofShugor, Dubasi and Watish subtypes ofSudan Desert
l 'Institut des sciences agronomiques du Rwanda. ILCA sheep at the El-Huda National Sheep Research Station,
Research Report 15. ILCA (International Livestock Gezira Province, Sudan. ILCA Research Report 18.
Centre for Africa), Addis Ababa, Ethiopia. 58 pp. ILCA (International Livestock Centre for Africa), Addis
SAS (Statistical Analysis System). 1987. Statistical analysis Ababa, Ethiopia. 30 pp.
system user's guide: Statistics. SAS Institute Inc., Cary, Wilson R T. 1987. Productivity of traditionally managed small
North Carolina, USA. ruminants in an agro-pastoral system in northern
Burkina Faso. Tropical Agriculture (Trinidad) 64: 163—
165.

240 Small Ruminant Research Network Workshop

 Some production parameters of Blended goats in
semi-arid regions of Tanzania
E.T.R. Lyatuu, S.M. Das and J.I. Mkonyi

Livestock Production Research Institute

P. O. Box 202, Mpwapwa, Tanzania

Abstract taux butyreux moyen pour les deux stations était de

2,9%.
Data on blended goats (a three-way cross of Kamorai
(55%), Boer (30%) and indigenous Tanzanian (15%)
goats) were collected over a period of four years Introduction
(1987-1990) at Malya and Kongwa Research Stations.
These records were analysed to assess the production Raising cattle in arid and semi-arid regions has many
potential of Blended goats. Birth weights, weaning constraints which tend to limit the supply of animal
weights, weight at 72 weeks, milk yield, lactation length protein available to rural farmers in these areas. As goats
and butter fat content of milk were used as parameters. are more efficient than cattle in feeding and other
managerial aspects, they offer more potential for meat
At Malya the average birth weight was 2.34 kg, the and milk production in these areas.
average weaning weight 12.5 kg and the average weight Increasing the meat and milk production of
at 72 weeks was 28.0 kg. At Kongwa the averages were indigenous goats through improved management and
2.31, 1 1 .6 and 27.4 kg for birth weight, weaning weight increased feed resources is unlikely to be effective under
and weight at 72 weeks, respectively. Milk yields ranged semi-arid conditions. Therefore, genotypes should be
from 90.0 kg to 175.5 kg during an average lactation of developed which are capable of producing more meat
236 days. Season of kidding, age at kidding and birth and milk within existing environmental constraints.
type all significantly (P<0.05) influenced lactation Although production traits are moderately heritable, it
yield. Average milk butter fat content for both stations would take time to achieve substantial increases by
was 2.9%. selection within existing indigenous genotypes because
the initial level of production is low. More rapid
increases can be expected from crossbreeding with
Paramètres de production chez exotic genotypes.
Blended goats are a three-way cross between
les chèvres croisées des régions Kamorai (55%), Boer (30%) and indigenous Tanzanian
goats (15%). They have phenotypically large pendulous
semi-arides de la Tanzanie ears and bigger bodies than indigenous goats (Das,
1989). Earlier reports on Blended goats have been
Résumé presented by Das (1989, 1992).
Data collected over four years (1987-1990) at
Des données sur des chèvres croisées (croisement
Malya and Longwa Research Stations were analysed to
trois-voies entre des caprins de race Kamorai [55%],
evaluate the growth and dairy potential of Blended
Boer [30%] et indigène de Tanzanie [15%]) recueillies
goats.
sur une période de quatre ans (1987-1990) aux stations
de recherche de Malya et de Kongwa ont été analysées
en vue de l 'évaluation du potentiel de production de ces
chèvres. Les paramètres utilisés étaient: les poids à la Materials and methods
naissance, au sevrage et à 72 jours, la production
laitière, la durée de la lactation et le taux butyreux. Study sites
A Malya, le poids à la naissance était en moyenne The two sites were the Livestock Research Centre,
de 2,34 kg, le poids au sevrage de 12,5 kg et le poids à Malya in north-western Tanzania and the Pasture
72 jours de 28,0 kg. A Kongwa, ces poids moyens étaient Research Centre, Kongwa, central Tanzania. They both
respectivement de 2,31, 11,6 et 27,4 kg. La production have a semi-arid climate with savannah grassland
laitière variait entre 90,0 kg et 175,5 kg au cours d'une consisting mainly of Commiphora and Combretum
lactation moyenne de 236jours. La saison de mise-bas, woody species and Eragrostis, Aristida, Bothriochloa,
l'âge à la mise-bas et le type de naissance avaient un Hyparrhenia, Themeda and Cynodon grasses. A few
effet significatif (P<0,05) sur la production laitière. Le paddocks at Malya are sown with Chloris gayana, while

Small Ruminant Research Network Workshop 241

E.T.R. Lyatuu, SM. Das and J.I. Mkonyi

at Kongwa, paddocks are sown with Cenchrus ciliaris performance (daily milk yield, lactation milk yield,
and some are sown with Siratro and Leucaena lactation length and % butter fat).
leucocephala. A linear model was used for analysing data on
weight for age and lactation milk yield. The model
Management of animals included fixed effects of seasons (wet, dry), birth type
(single, twin), sex (male, female), dam age (2-3, 4-5,
Management at both locations was similar. Kids were 5-6, 6-7 and 7 years) and interaction between main
eartagged and weighed at birth and allowed to suckle in effects.
the morning. They were reared indoors and provided
with fodder and water ad libitum until weaning at 16
weeks. Does were milked once in the evening. After
weaning, does were milked twice a day. Kids grazed on Results and discussion
natural pastures from 0800 to 1500 hours in addition to
receiving a daily concentrate ration (maize/rice bran and Growth parameters
cotton seed/sunflower seed cake) of 30 to 50 grams per
kid. Female kids were bred for the first time at 72 weeks Weights at birth, weaning and 72 weeks for Malya were
of age. Thereafter they were bred twice a year with 2.34 + 0.05 kg, 12.5±0.12 kg and 28.0±0.73 kg,
mating seasons of 70 days each corresponding to the dry respectively. Corresponding values at Kongwa were
(September - November) and wet (March - May) 2.31±0.08 kg, 11.6±0.34 kg and 27.4±0.16 kg,
seasons. The buck:doe ratio was 1 :30. The animals were respectively. Birth weights of Blended kids were higher
vaccinated, dipped and dewormed regularly to control than the 2.1 kg reported for indigenous goats and
disease, while clinical diseases were treated on Kamorai x indigenous crosses (Das and Sendalo, 1991 ).
occurrence. This was also true for weights at 72 weeks. The
indigenous and Kamorai x indigenous crosses have been
Data collection and analysis reported to weigh 11.7 kg and 17.5 kg, respectively (Das
and Sendalo, 1991). Birth and weaning weights were
A total of 893 records on live weights at different ages also higher than those reported for Adal and for Saanen
and 263 milk records of Blended does from both stations x Adal goat kids in Ethiopia (Kassahun et al, 1989) and
were analysed in this study. Kids were weighed at birth, those reported for Sudanese goats (Wilson, 1976).
weaning and at 72 weeks of age. Does were weighed at Tables 1 and 2 show the effect of seasons, type of
mating and kidding. Pre-weaning daily milk yield was birth, sex and age of dam on the weight (kg) of Blended
estimated by doubling the evening milk yield. kids at Malya and Kongwa.
Parameters analysed included weight for age (at birth, Kids born in the wet season were heavier from birth
weaning and at 72 weeks of age), and lactation to 72 weeks old than were those born in the dry season,

Table 1. Effect of season, type of birth, sex and age of dam on weight (kg) of Blended kids at Malya.
Birth weight Wean ing weight 72-weeks weight
Variable n mean n mean n mean
Season:
wet 148 2.52 122 12.8 97 28.2
dry KU 2.48 93 12.3 59 27.5
Average SE 0.08 0.26 1.02
Birth type:
96 2.73a- 13.5a- 30.4a
single 81 56
2.04b 11.2b 27.1b
twin 156 134 90
Average SE 0.09 0.31 1.08
Sex:
2.11a- 27.9a
female 115 106 12.3 84
137 2.57b 109 29.5b
male 12.7 62
Average SE 0.06 0.39 1.12
Age of dam:
2-3yrs 67 2.15a- 51 27.4
11.8 33
4-5yrs 110 2.22a-
95 12.4 61 28.2
6-7yrs 41 2.47b 37 29
11.9 29.0
2.41b
> 7 yrs 34 32 12.9 23 28.5
Average SE 0.09 0.32 1.03
ab =Means within a column (for each variable) with different superscripts differ significantly (P<0.05).

242 Small Ruminant Research Network Workshop

 Some production parameters of Blended goats in.

Table 2. Effect of season, type of birth. sex and age of dam on weight (kg) of Blended kids at Kongwa.
Birth weight Weaning weight 72-weeks weight
Variable n mean n mean n mean
Season:
wet 72 2.36 56 11.8 33 26.6
dry 55 2.28 39 11.5 25 27.9
Average SE 0.11 0.49 1.21
Birth type:
single 51 2.64a 42 12.3a 28 29.3a
twin 76 2.01b 53 10.6b 30 26.5b
Average SE 0.17 0.56 1.45
Sex:
female 69 2.18a 51 9.3a 35 26.8a
male 58 2.43b 44 12.7b 23 28.2b
Average SE 0.15 0.52 1.33
Age of dam:
42 2.06a 31
2-3yrs 10.7 18 26.4
4-5yrs 46 2.19a 38 10.4 26 28.1
6-7yrs 25 2.34b 16 12.2 10 28.3
<7yrs 14 2.42b 10 11.5 4 27.7
Average SE 0.14 0.55 1.41
ab = Means within a column (for each variable) with different superscripts differ significantly (P<0.05).

but mean differences were not significant (P<0.05). Das the effect of location were obtained by Das and Mkonyi
and Sendalo (1992) reported similar findings on (1990). These differences may be attributed to
improved goat kids at Malya. They also found that differences in management, particularly in feeding
season had a significant influence (P<0.05) on both birth subsystems. In Kongwa the availability of maize bran
and weaning weights. This could be attributed to the fact was erratic, but at Malya rice polish and cottonseed cake
that during the wet season, gravid does tend to gain were abundant.
weight from the lush green forages available at the farm.
Kids born single at both sites were significantly Dairy parameters
(P>0.05) heavier than twins at birth, weaning and at 72
weeks of age. Males were significantly (P<0.05) heavier A total of 263 does were studied for their dairy potential,
than females at birth and at 72 weeks old at Malya (Table 174 at Malya and 89 at Kongwa. Average daily milk
1). At Kongwa, males were superior in all liveweight yield per doe was 0.69 kg and 0.66 kg for Malya and
categories (Table 2). The effects of birth type and sex on Kongwa respectively. Mean lactation milk yield at
birth weight, weaning weight and mature weight agree Malya was 170.6 (SE 4.57) kg for a mean lactation
with those reported by Chawla and Bhatnagar (1983) for length of 245 (SE 3.11) days. At Kongwa, mean
Barbari and Beetal goats in India. Twins had lower birth lactation milk yield was 153.4 (SE 5.42) kg for a mean
lactation length of 23 1 (SE 4.38) days.
weights than singles, probably due to their smaller size
Daily milk yields at both sites were much lower
and weight in the uterus. After birth, single kids had an
than those reported for Beetal goats (0.9 kg) (Verma and
advantage over twins as twins had to compete for the
Chawla, 1987) and Norwegian goats (1.1 kg) (Das and
milk from their dam. Single kids had sufficient milk for
Sendalo, 1991).
growth to weaning. Higher weights at maturity for single
The values for lactation milk yield were higher
kids could be explained by this advantage and the
than those reported for Toggenburg and Anglo-Nubian
positive correlation between weaning and mature
crosses (85 kg and 104 kg in 143 days respectively), but
weights. lower than other dairy goats such as Saanen and
Age of dam significantly (P<0.05) influenced birth Norwegian. Mean lactation milk yields of Blended goats
weight of kids, with older does producing heavier kids were also lower than those reported for Jamnapari goats
than younger does. (202 kg in 191 days) by Sands and McDowell (1978).
The interaction between age of dam and type of This could be due to the low genetic potential for milk
birth for birth weights was significant (P<0.05) at production of the Boer and indigenous goats which
Malya. Other interactions between main effects were not crossed to form Blended goats.
significant. The lack of effect of season x birth type Table 3 shows the effect of type of birth, season
interaction on liveweight implies that singles showed and age of dam on the dairy potential of Blended goats
superiority irrespective of season of birth. at Malya and Kongwa.
Differences between sites were significant Does with single kids had higher lactation milk
(P <0.001) for weaning weight only. Similar results of yields and longer lactation lengths at Malya. At Kongwa

Small Ruminant Research Network Workshop 243

E.T.R. Lyatuu, SM. Das and J.I. Mkonyi

Table 3. Effect of type of birth. season and age of dam on dairy potential of Blended goats at Malya and Kongwa.
Lactation milk yield (kg) and length
Malya Kongwa
mean mean mean mean
Variable n (kg) (days) n (kg) (days)
Type of birth:
176.9a 251a
single 96 51 155.7 233
165.2b
twin 78 238b 38 152.2 228
Average SE 7.34 6.25 8.44 7.61
Seasons:
99 178.6a 247a 158.7a 235a
wet 56
163.1b 240b 146. lb 221b
dry 75 33
Average SE 8.23 7.82 8.63 8.75
Dam age (yrs):
2-3 45 161.8a 236a 33 144.5a 228a
166.2a 241a 151.8b 230b
4-5 XI 29
27 172.3b 247b 153.9b 232b
6-7 15
176.7b 246b 155.3b 232b
7 yrs 21 12
Average SE 7.68 6.27 8.77 7.21
ab = Means within a column (for each variable) with different superscripts differ significantly (P).

there was no difference in milk yield due to birth type. Blended goats have advantages over indigenous
Does kidding in the wet season had significantly goats for live weight and milk production. They could
(P<0.05) higher lactation milk yields and longer be used to improve the dairy and meat production by
lactation lengths than those kidding in the dry season. selection and crossbreeding. The introduction of
This is probably due to the availability of lush, green Blended goats and their crossbreeds selected for high
pastures during the wet season compared to low quality milk yields could solve the problem of insufficient milk
forages in the dry season. in semi-arid areas of Tanzania. This could also improve
The effect of age of dam was significant (P<0.05) the economic status of goat keepers throughout the
for lactation milk yields and lactation lengths at both country.
sites, with older dams having more milk and a longer Improving feeding and management of practices
lactation period than young does. Variations in milk (housing, disease control) is essential for optimum goat
yield due to age of dam were similar to those reported productivity. More studies on the quality of milk and
by Steine (1975); lowest yield occured in the first meat produced, reproductive efficiency, disease
lactation and highest in the fifth. resistance, nutrition and other managerial aspects are
Average butter fat content was 2.9% and 2.91% necessary for the establishment and characterisation of
for milk from goats in Malya and Kongwa respectively. Blended goats.
This is less than that reported for Norwegian goats but
higher than Norwegian x indigenous crosses (Das and
Sendalo, 1991). The interactions between main effects Acknowledgment
were not significant.
The authors are grateful to the Commissioner for
Research and Training for permission to present this
paper. Also we would like to thank the International
Foundation for Science (IFS) based in Sweden for
Conclusion and recommendations funding this study. Last but not least we would like to
give our special thanks to scientists and technicians at
The results of the evaluation of some production traits Malya, Kongwa and Mpwapwa for their assistance
of Blended goats show that Blended goats have the during the course of this study.
potential to produce both milk and meat for human
consumption.
Values for lactation milk yield, daily yield and
milk butter fat content of milk are lower in Blended References
goats than in other dairy goats in the subtropics. Chawla D S and Bhatnagar D S. 1983. Milk production
However, the high variation in lactation milk yield potential of various crossbred goats under stall - fed
indicates that improvements could be made through condition in tropics. Asian Journal of Dairy Research
breeding and selection. The higher live weights and 2(3):173-178.
wide variation among Blended goats indicate their Das SM. 1989. Preliminary results on evaluation and breeding
potential for meat production. of blended dairy goats in Tanzania. In: Wilson R T and

244 Small Ruminant Research Network Workshop

 Some production parameters of Blended goats in.

Azeb Melaku (eds), African small ruminant research Kassahun Awgichew, Yibrah Yacob and Fletcher I. 1989.
and development. Proceedings of a conference held at Productivity of purebred Adal and quarterbred Saanen x
Bamenda, Cameroon, 18-25 January 1989. ILCA Adal goats in Ethiopia. In: Wilson R T and Azeb Melaku
(International Livestock Centre for Africa), Addis (eds), African small ruminant research and
Ababa, Ethiopia. pp. 536-545. development. Proceedings of a conference held at
Bamenda, Cameroon, 18-25 January 1989. ILCA
Das S M and Mkonyi J I. 1990. Evaluation of Blended goats
(International Livestock Centre for Africa), Addis
for dairy production in semi-arid regions of Tanzania.
In: Proceedings of the 81 Tanzania Veterinary Ababa, Ethiopia. pp. 510-523.
Association Scientific Conference held at Arusha, Sands M and McDowell R E. 1978. The potential of the goat
Tanzania, 4-7 December 1990. (In press). for milk production in the tropics. Cornell Inter, tional
Agriculture Mimeograph No. 60. Cornell University,
Das S M and Sendalo D S. 1991. Small ruminant research
Ithaca, New York, USA. 53 pp.
highlights in Tanzania (1960-1989). Ministry of
Agriculture, Livestock Development and Co-operatives, Steine T. 1975. Factors affecting traits of economic importance
Dar es Salaam, Tanzania. 40 pp. in goats. Meldinger fra Norges Landbruk shogekole
Das S M and Sendalo D S. 1992. Comparative performance 54(2):30.
of improved meat goats in Malya, Tanzania. In: Rey В, Verma N K and Chawla D S. 1987. Factors affecting test day
Lebbie S H В and Reynolds L (eds). Small ruminant yield and monthly milk yield of Beetal, Alpine and
research and development in Africa. Proceedings of the crossbred goats: A note. Indian Journal of Animal
First Biennial Conference of the African Small Production and Management 3: 146-147.
Ruminant Researach Network, ILRAD, Nairobi, Kenya,
Wilson R T. 1976. Studies on livestock of Southern Darfur,
10-14 December 1990. ILCA (International Livestock
Sudan. IV. Production traits in goats. Tropical Animal
Centre for Africa), Nairobi, Kenya. pp. 445-452.
Health and Production 8:221-232.

Small Ruminant Research Network Workshop 245

 Performances de reproduction des caprins dans
les systèmes d'élevage traditionnels
des zones communales au Zimbabwe
F. Monicat, J. Keravec, P.M. Borne

Veterinary Goat Project, Veterinary Research Laboratory

[Link] 8101, Harare (Zimbabwe)

Résumé des caprins sont minimales: la fertilité annuelle était de

86% pour les multipares et de 3 1 % pour les femelles de
Un suivi individuel et par troupeau des moins de 2 ans; une seule mise-bas était enregistrée par
performances des caprins a été effectué dans neuf zones an, en milieu de saison sèche; la première mise-bas
communales représentatives de la variabilité des intervenait après 22 mois; la prolificité moyenne était de
systèmes de production rencontrés en milieu 120% pour les multipares et le taux de fécondité moyen
traditionnel au Zimbabwe. Du 1er septembre 1990 au 1er
de 110%.
novembre 1991 , un total de 6 950 caprins appartenant à
254 exploitations ont été suivis individuellement sur une 3) Les zones communales où le milieu naturel est
base bimensuelle. fortement dégradé, avec une forte densité de
population et des contraintes alimentaires imposées
Les performances de reproduction se sont révélées par les systèmes de culture.
très différentes suivant les environnements naturels et
socio-économiques dans lesquels évoluaient les Cette situation s'observe dans les zones dominées
animaux. Trois situations prédominantes ont pu être par les cultures et dans les zones péri-urbaines à forte
observées. densité de population. Aux restrictions alimentaires de
saison sèche imposées par la dégradation du milieu
1) Les zones communales d'implantation récente,
naturel s'ajoutent les mesures de lutte contre la
caractérisées par un environnement naturel non
divagation des animaux pendant la saison des pluies afin
dégradé, une faible pression démographique et une
de protéger les cultures. Au milieu de la saison des
faible densité animale.
pluies, les animaux sont enfermés pendant une partie de
Dans ces zones où une brousse arbustive dense la journée. La reconstitution des réserves est ainsi
occupe la majeure partie du territoire, la présence d'une interrompue et ne reprend qu'après les récoltes, quand
strate arbustive buissonnante permet aux femelles de les animaux sont relâchés. Pour la plupart des femelles,
reconstituer leur état corporel en fin de saison sèche, les fécondations - quand elles ont lieu - surviennent à
avant les premières pluies. Les fécondations ont lieu au la fin de la saison des pluies. Les performances de
tout début de la saison des pluies et les potentiels de reproduction s'en ressentent fortement: le taux de
reproduction des caprins s'extériorisent pleinement: le fertilité annuelle était de 73% pour les multipares et de
taux de fertilité annuelle était de 97% chez les multipares 17% pour les femelles de moins de 2 ans; une seule
et de 62% chez les femelles de moins de 2 ans; le mise-bas était enregistrée par an, en milieu de saison
dessaisonnement était important, avec une moyenne de sèche; les premières mises-bas intervenaient après 22
trois mises-bas en deux ans réparties sur la saison des mois; la prolificité moyenne était de 120% pour les
pluies et la saison sèche; la première mise-bas multipares et le taux de fécondité moyen de 100%.
intervenait entre 12 et 18 mois; la prolificité était en Dans ces systèmes d'élevage soumis à de lourdes
moyenne de 150% pour les multipares et le taux de contraintes, 35% des chevreaux meurent avant l'âge de
fécondité de 190%. 1 an.
2) Les zones communales d'implantation ancienne Il ressort des résultats enregistrés que les
dotées d'une végétation fortement dégradée et performances des troupeaux sont d'autant plus
soumises à une forte pression animale. homogènes et plus médiocres que les contraintes de
Dans ces zones, où le surpâturage et l'exploitation l'environnement naturel et socio-économique sont plus
intensive du bois par la population ont entraîné la quasi- importantes. A l'inverse, une diminution des contraintes
disparition de la strate arbustive buissonnante, les du milieu se traduit par une plus grande variabilité des
femelles doivent attendre la pousse de l'herbe en début performances par troupeau. Ce sont alors les techniques
de saison des pluies pour reconstituer leurs réserves. et pratiques individuelles mises en oeuvre par les
Elles ne sont fécondées qu'en milieu de saison des éleveurs qui permettent l'expression du potentiel des
pluies. Dans ce cas, les performances de reproduction

Small Ruminant Research Network Workshop 247

F. Monicat, J. Keravec, P.M. Borne

growing at the beginning of the rainy season to

Reproductive performance of regenerate their reserves. They do not conceive until the
goats in traditional livestock middle of the rainy season. Thus reproductive
performance is poor. There is: an annualfertility rate of
systems in communal areas in 86% for multiparous does and 31% for those below two
Zimbabwe years old; a single kidding in the middle of the dry
season; first kidding after 22 months; mean fertility of
Abstract 1 .2for multiparous does; and afecundity rate averaging
110%.
Individual and herd performances of goats have been
monitored in nine communal areas in Zimbabwe which C) Heavily degraded natural environment with high
are representative ofthe various traditionally managed population density and food constraints due to
production systems. Between 1 September 1990 and 1 cropping systems.
November 1991, 6950 goats belonging to 254 farms This type ofsituation isfound in areas where crops
were monitored twice monthly on an individual basis. dominate and in peri-urban areas with high human
population density. In addition tofood restriction during
Results show that reproductive performance
the dry season, due to the degradation of the natural
varied markedly with the natural and socio-economic
environment, measures are taken to protect crops
environment of the animals. Three situations
against wandering animals during the rainy season. In
predominate.
the middle of the rainy season, animals are locked up
A) Recently established communal areas where the for part of the day. Rebuilding body reserves is held up
natural environment has not become degraded, with and is resumed after the harvesting period, when the
low population pressure and low animal density. animals are released. When conception does take place,
In these areas most ofthe land is covered by shrub it occurs at the end of the rainy season, thereby heavily
bushland. A zone of brushwood enables females to affecting reproductive performance. An annual fertility
recover their body condition by the end of the dry rate of 73% for multiparous and of 17% does females
season. Conceptions occur at the very beginning of the below 22 months; a single kidding in the middle of the
rainy season and the goats ' reproductive potential is dry season; first kidding after 22 months; mean fertility
fully achieved with an annual fertility rate of 97% for of 1.2 for multiparous does; and a fecundity rate
multiparous and 62% for does below two years of age; averaging 100% are typical. In these heavily
there is little seasonality of breeding, with an average constrained livestock systems, 35% ofthe kids die before
of three kiddings in two years spread over the rainy and they are one year old.
the dry seasons; first kidding occurs between 12 and 18 Results show that the more the environment
months; there is mean fertility of 1.5 for multiparous
imposes natural and socio-economic constraints the
does; and the fecundity rate averages 190%.
more herd results become uniform and poor. On the
B) Old communal areas with heavily degraded other hand, when animals are not submitted to heavy
vegetation and high animal pressure. environmental constraints, considerable variability in
In these zones, overgrazing and overexploitation herd performances is recorded. The animals ' potential
of wood lead to the near disappearance of the zone of is achieved through the individual techniques and
brushwood and females have to wait for the grass practices of their owners.

248 Small Ruminant Research Network Workshop

Posters - Abstracts
 Posters Abstracts

Données sur l'association coccidies-helminthes au niveau

du polyparasitisme gastro-intestinal des moutons Djallonké en Côte d'Ivoire
Mme M.S. Hounzangbé-Adoté

B.P. 1170, Cotonou (Bénin)

Les examens coprologiques effectués sur trois liée à l'état sanitaire, l'âge et le sexe du troupeau. Le
troupeaux de moutons Djallonké, élevés dans la région polyparasitisme coccidies-helminthes induit un
de Bingerville révèlent la présence des parasites équilibre biologique avec une prédominance aux
gastro-intestinaux lorsque les animaux ne sont pas strongles. La rupture de cet équilibre augmente la
régulièrement déparasités. L'intensité parasitaire est densité de la population et l'activité des coccidies.

Effect of concentrate to forage ratio on intake and digestibility in sheep

E.H Goromela , S.J.K. Nyakalo and E.E. Ndemanisho

Livestock Production Research Institute, P. O. Box 202, Mpwapwa, Tanzania

Department ofAnimal Science and Production
Sokoine University ofAgriculture, P.O. BOX 3004, Chuo Kikuu, Morogoro, Tanzania

Abstract There were significant differences (P<0.05) in feed

intake between treatments I and III but not ( P>0.05)
Eight male Blackhead Persian sheep weighing between between treatments II and IV. Animals on treatment II
26 and 30 kg, and of between 1 and 2 years of age, were demonstrated significantly higher (895.9+ 14.9 g/day)
used to determine the effect of concentrate to forage average daily feed intake than those on treatment I
ratio on intake and digestibility. Four dietary treatments (724.4 ± 12.1 g/day). There was a significant difference
(I, II, III and IV) on the basis of concentrate to forage (P<0.05) between treatment means in terms of growth
ratio of 20:80, 30:70, 40:60 and 50:50, respectively, rate. Animals on treatment II had a higher growth rate
were formulated. In a completely randomised design (110.0+ 17.8 g/day) than animals on treatments I, III
experiment, two animals were randomly allotted to each and IV. Nitrogen balance and apparent digestibility
treatment and fed the diets ad libitum in individual crates coefficients for DM, OM, CP, ЕЕ, NFE, Ash, NDF and
throughout the 56 days of the experiment. The ADF were not affected by treatments.
performance parameters measured were feed intake,
digestibility and nitrogen balance.

Performance of dairy goats in Mgeta Morogoro, Tanzania

D.R. Nkungu2, G.C. Kifaro3 and LA. Mtenga3
2
Zonal Research and Training Centre, Central Zone, [Link] 202, Mpwapwa
The Department ofAnimal Science and Production, Sokoine University ofAgriculture
[Link] 3004, Chuo Kikuu, Morogoro

Abstract 3 months, 3 to 6 months and 6 to 12 months of age,

respectively. In both traits, male and single kids were
A study was conducted to evaluate the performance of superior to females and kids born as twins, respectively.
Norwegian dairy goats and their crosses under
Kids with 87.5% and 50% Norwegian blood levels
smallholder farmer conditions in Mgeta village, performed better than the 100% Norwegian kids. The
Morogoro, Tanzania.
overall mean for lactation milk yield (kg) was 142.0 ±
Means and standard deviation were estimated for 120.0, the lactation length was 158.6 ±89.9 days and
both kid and dam traits studied. The overall kid mean milk yield per day was 0.87 ±0.4 litres. Kidding
weights (kg) were 2.7 ± 0.4, 13.3 ± 2.5, 20.9 ± 4.5, and interval (days), twinning rate (%) and litter size were
37.1 ± 9.3 at birth, 3 months, 6 months and 12 months 374.0 ± 98.9, 388.0 ±71.0, 41 .5% and 1.5 respectively.
of age, respectively. The average daily gains (kg) were Dam and kid mortalities were 7.8% and 7.5%,
1 16.0 ±31.9, 86.5 ± 30.1 and 87.9 ±31.3 from birth to respectively.

Small Ruminant Research Network Workshop 251

Posters Abstracts

Pastures for small ruminants

A.J. Mwilawa and M. M.S. Lugenja

Ministry ofAgriculture
LPRI, P.O. Box 202, Mpwapwa, Tanzania

About 99% of all sheep and goats in Tanzania are raised those in the central plateau, species composition is
traditionally on natural pastures. The quality of these limited and therefore the animal performance is usually
pastures varies with season, age and species of plants in poor for large parts of the year. Under natural and
association. Pastures in the periphery of the country and cultivated pasture, sheep and goats are complementary
highlands are usually rich in plant species composition, to one another in their grazing behaviour. Feeding
making them excellent for the production of preference does not exist under stall-feeding.
well-finished animals all year round. However, for

Growth and reproduction performance of Blackhead Persian sheep

in Tanzania
D.S.C. Sendalo, S.M. Das and L.A. Mtenga

Ministry ofAgriculture
Livestock Development and Cooperatives
P.O. Box 2066, Dar es salaam, Tanzania

Data collected at the Livestock Research Centres, Malya and 33.24 kg, respectively. Year had significant effect
and West Kilimanjaro, Tanzania, to determine the on birth weight, weaning weight and weights at 24, 48
effects of genotype and environment on birth weight, and 72 weeks of age. Also season significantly
weaning weight and weights at 24, 48 and 72 weeks of influenced all weights except for birth weight and
age were analysed. Records on birth weight (n=6207), weight at 24 weeks of age. Birth type had significant
weaning weight (n=4444), weights at 24 weeks effect on birth weight (P<0.001) and weaning weight
(n=3494), 48 weeks (n=2633) and 72 weeks of age (P<0.05) but did not significantly affect the post-
(n=1874) were collected and analysed for fixed effects: weaning weights. Age of the dam significantly
year of birth, season of birth, sex, birth type, age of dam influenced birth weight and weaning weight of lambs.
and location of differences. Differences due to location were highly significant for
The overall unadjusted mean weight at birth and weights at birth, weaning, 24 and 72 weeks of age.
weaning were 2.68 and 1 5 . 1 5 kg, respectively. The mean Lambs born at Malya were heavier at all ages than those
weights at 24, 48 and 72 weeks of age were 18.19, 22.93 at West Kilimanjaro.

Role of goats in the utilisation of rangelands of Tanzania

M.L. Kusekwa, A.J. Mwilawa and S.W. Chenyambuga

LPRO, Box 202, Mpwapwa, Tanzania

Abstract Experience with goat raising among the agro-

The rangelands resource of Tanzania is vast, covering pastoralists has shown that these animals are herded
an area of 60 million hectares. This area is capable of alongside cattle and sheep. The sheep and cattle tend to
supporting over 20 million Tropical Livestock Units graze while the goats mostly browse, thus bringing
(TLUs) and millions of game animals. There are about about a balanced utilisation of the rangelands with a
6.4 million goats in the country and they, together with concomitant woody weed control effect.
sheep (estimated 3.0 million), produce 30,000 tonnes of
red meat for the human population.

252 Small Ruminant Research Network Workshop

 Posters Abstracts

Increased productivity of indigenous goats through an organised extension

approach in Newala
[Link]

Ministry ofAgriculture
Regional Agriculture/Livestock Office
P.O. Box 512, Mtwara, Tanzania

The environment and vegetation in Newala district are improved goat husbandry practices to farmers
favourable for goat production but unfavourable for particulary in the areas of internal parasites, housing,
cattle production; the major constraints being lack of nutrition, disease prevention and breeding stock
water and suitable grazing; also cattle ownership is management. The project operates, in conjunction with
beyond the means of most village farmers. Goats are the district livestock and extension (NALERP) offices,
therefore appropriate to the environment and to the by the provision of training and field materials and
farmer. equipment inputs to field staff, village meetings, use of
Baseline studies indicated a high kid mortality rate, extension aids (booklets, posters, subject handouts and
being 45% for kids up to 12 months of age. Major causes model goat houses), assistance to initiate primary school
of death appeared to be helminthiasis (Haemonchus and womens' group goat projects and provision of
spp.), poor nutrition and bad housing. tree/forage planting materials.
The long-term objectives of the project are to The number of goats in the district has increased
improve family nutrition, farmer income and family from 64,000 (1984) to 144,000 (1991). The number of
welfare by reducing goat mortality, increasing pro households owning goats has increased from 15%
duction, thereby increasing the meat supply, and (1984) to 30% (1991); the average flock size of
encouraging milk production. The short-term objectives seven has not changed. Ownership of goats provides
of the project are to communicate simple and low cost security.

Browse plants as feed resource for ruminants in Botswana

A.A. Aganga, J.S. Kiazolu and CM. Tsopito

Botswana College ofAgriculture

P. Bag 0027, Gaborone, Botswana

Abstract and end of each season. Small ruminants in particular

depend heavily on browse, especially during the dry
Eleven browse species in all were sampled from two season. Nutrient composition of the browse plants in the
vegetational locations. Six browse species were loamy soil areas varied in CP content from 72.2g/kg DM
obtained from sandveld, (Mmantshwabisi) while nine for Peltophorum africana to 165.7g/kg DM for Acacia
were sampled from loamy soils (Makgobokogobo). mellifera pods during the dry season. CP content ranged
Four browse species were common to the two from 78.9g/kg DM for Peltophorum africana to
vegetational zones. The browse plant components 177.0g/kg DM for Acacia fleckil pods in the sandveld
(leaves, twigs and pods) were analysed for proximate during the dry season. Phosphorus content ranged from
composition, and also evaluated for in vitro dry matter 0.58g/kg DM for Dichrostachys cineria leaves (L) to
digestibility and nylon bag dry matter degradability at 1 .5g/kg DM for Acacia fleckil (L) in loamy soil area
24h, 48h, 72h and 96h incubation periods using during the dry season. Phosphorus content of
cannulated Simmental steers. The browse samples were Peltophorum africana was least with 0.53g/kg DM and
analysed for dry matter (DM), ash, crude protein CP, Acacia fleckil tops the list with 1.33g/kg Dm of
crude fibre and fat. Minerals analysed were P, K, Ca, Na, phosphorus during the dry season.
Mg, Mn and Fe. Sampling was done three times each in
both wet and dry seasons i.e. at the beginning, middle

Small Ruminant Research Network Workshop 253

Posters Abstracts

Production de viande des ovins en milieu paysan

Tahirou Tangara

DRSPR/IER
B.P. 9030, Bamako (Mali)

Résumé La durée de l'engraissement a été de 66 ± 4 jours

pour l'ensemble des deux lots d'animaux. Le gain de
Dans le cadre de la présente étude, des analyses ont
poids journalier était de 85,22 ± 24,87g pour le lot
été effectuées pour déterminer les performances de 121
ovins mâles engraissés en stabulation dans des petites expérimental contre 56,55 ± 40,03 g pour le lot témoin.
exploitations agricoles de la zone semi-aride du Mali à Le niveau de bénéfice économique par tête
l'approche de la fête de Tabaski à Banamba. Ces travaux embouchée/paysan était en moyenne de 5,430 FCFA et
ont été complétés par des études économiques et 4,483 FCFA respectivement pour le lot témoin et le lot
financières destinées à évaluer la rentabilité de expérimental.
l'opération. En zone semi-aride la rentabilité de l'embouche
n'est pas liée seulement au gain de poids.

Etablissement d'une formule barymétrique chez le mouton à queue

grasse de l'Afrique de l'Est utilisable en milieu rural
G. Sibomana & C. Kayotse

ISAR - Programme petits ruminants

B.P. 138, Butare (Rwanda)

Résumé Après l'élimination de la hauteur au garrot et du

tour spiral, les données sur le périmètre thoracique et le
La présente étude a pour but de tester en milieu poids ont été soumises à quatre modèles de régression:
rural, la formule barymétrique établie au Centre La régression linéaire simple Y = a + bx
National de l'Elevage de Songa à partir des données - La régression quadratique Y = ax + b + с
recueillis sur le mouton local élevé en station. La régression exponentielle Y - exa+ x
Pour cela un échantillon expérimental de 400 - La régression multiplicative T = ax
moutons a été prélevé en milieu rural. Chaque mouton Les résultats d'analyse de ces derniers font
faisant partie de l'échantillon a fait l'objet de relevé de ressortir que le modèle de régression multiplicative
poids et de trois mensurations corporelles en l'occurence permet le meilleur ajustement du périmètre thoracique
le périmètre thoracique, la haute au garrot et le tour au poids et que par conséquent il peut être avanta
spiral. geusement utilisé dans l'établissement de la formule
La comparaison de trois mesures corporelles au barymétrique.
niveau de leurs intensités de liaison avec le poids, Une table de conversion périmètre thoracique -
confirment les résultats trouvés en Station en révelant poids utilisable en milieu rural est proposée.
que le périmètre thoracique reste la mensuration la plus
fortement corrélée au poids.

254 Small Ruminant Research Network Workshop

 LIST OF PARTICIPANTS

BENIN EGYPT
Mme M.S. Hounzangbé-Adoté Ferial Hassan
B.P. 1170 ARC-Technology Transfer Building
COTONOU 9 Gamaa Street
GIZA

BOTSWANA
ETHIOPIA
A.A. Aganga
Botswana College of Agriculture O.O. Akinbamijo
P. Bag 0027 ILCA
GABORONE P.O. Box 5689
ADDIS ABABA

BURKINA FASO
Aimé Nianogo J.E.O. Rege
Institut du Développement rural ILCA
Université de Ouagadougou P.O. Box 5689
B.P. 7021 ADDIS ABABA
OUAGADOUGOU
S. Tembely
BURUNDI ILCA
M. Ba P.O. Box 5689
Projet caprin Ngozi ADDIS ABABA
B.P. 1041
BUJUMBURA
C. Beyene
P.O. Box 3487
CAMEROON ADDIS ABABA
R. Njwe
Department of Animal Science
J. Walsh
University of Dschang
ILCA
B.P. 222
P.O. Box 5689
DSCHANG
ADDIS ABABA

CONGO
A. Lahlou-Kassi
A. M. Batalou ILCA
B.P. 2499 P.O. Box 5689
BRAZAVILLE ADDIS ABABA

B.I. Shapiro
COTE DTVOIRE
ILCA
A. Oya P.O. Box 5689
SODEPRA-PNSO ADDIS ABABA
01 B.P. 1366
BOUAKE01
GHANA
C.V. Yapi A.K. Tuah
Institut des savanes (IDESSA) University of Science and Technology
01 B.P. 633 Department of Animal Science
BOUAKE01 KUMASI

Small Ruminant Research Network Workshop 255

List of Participants

GUINEE N.A. Urio

M. Pathé Diallo ILCA
Direction de l'élevage P.O. Box 46847
B.P. 559 NAIROBI
CONAKRY
R.L. Baker
KENYA ILCA
P.O. Box 46847
A. Abate
NAIROBI
University of Nairobi
Department of Animal Production
P. O. Box 29053 J.T. Musiime
NAIROBI OAUABAR
P.O. Box 30786
W.O. Odenya NAIROBI
Egerton University College
Animal Science Department
P.O. Box 536 L. Reynolds
NJORO ILCA
P.O. Box 80147
M. Mucuthi MOMBASA
Laikipia Research Programme
P.O. Box 144
NANYUKI MALAWI
S. Karua
P.M. Kiriro Department of Animal Science
Egerton University College University of Malawi
P.O. Box 536 Bunda College of Agriculture
NJORO P.O. Box 219
LILONGWE
R. Shavulimo
Egerton University College
Department of Animal Health J.W. Banda
P.O. Box 536 Department of Animal Science
NJORO University of Malawi
Bunda College of Agriculture
P.P. Semenye P.O. Box 219
Nutrition and Management Project LILONGWE
P.O. Box 252
MASENO
MALI
K. Miller Mamadou Doumbia
Small Livestock Project Station de recherche zootechnique, Sotuba
P.O. Box 2160 B.P. 262
NYERI BAMAKO

J. Ndikumana Tahirou Tangara

ILCA B.P. 9030
P.O. Box 46847 DRSPR/IER
NAIROBI BAMAKO

CM. Mwendia E. Olalokou

KETRI ILCA-Bamako
P.O. Box 362 B.P. 60
KIKUYU BAMAKO

256 Small Ruminant Research Network Workshop

 List of Participants

MAROC RWANDA
Derquaoui L. G. Sibomana
Dept. de reprod. animale et d'I.A. ISAR-Programme petits ruminants
Institut agronomique et vétérinaire Hassan II B.P. 138
B.P. 6202, RABAT-INSTITUTS BUTARE

NIGER SENEGAL
A. Yenikoye Mme Maïmouna Cissé
Faculté d'agronomie Laboratoire national de l'élevage
Université de Niamey et des recherches vétérinaires (ISRA)
B.P. 237/10896 B.P. 2057
NIAMEY DAKAR

NIGERIA Magatte Ndiaye

O.A. Osinowo B.P. 2057
National Animal Production Research Institute LNERV-ISRA-HANN
Ahmadu Bello University, DAKAR
ZARIA Fax: 21 18 79

A.A. Ademosun SIERRA LEONE

P.O. Box 182 I.C. Hassan
ILE-OLUJI Departmental Chief Veterinary Officer
Ondo State 9C Thompson Bay
Carlton Carew Road
FREETOWN
[Link]
Department of Veterinary Medicine
University of Nigeria
SWAZILAND
NSUKKA
B.H. Ogwang
APH Department
K.A. Majiyagbe University of Swaziland
National Veterinary Research Institute P.O. LUYENGO
[Link] 155
VOM near JOS
Plateau State TANZANIA
Ben Moshi
Principal Secretary
I.F. Adu Ministry of Agriculture, Livestock
Department of Animal Production and Health and Cooperatives
P.M.B 2240 P.O. Box 2066
ABEOUKUTA DAR ES SALAAM

A. Ikwuegbu L.A. Mtenga

ILCA Sokoine University of Agriculture
P.M.B. 2248 [Link] 3004
KADUNA Chuo Kikuu
MOROGORO

NORWAY
N.R. Standal L.E.G. Mboera
Agricultural University of Norway Livestock Training Institute
P.O. Box 5025 P.O. Box 51
1430 AS MPWAPWA

Small Ruminant Research Network Workshop 257

List of Participants

S.M. Das A.S. Mejooli

Zonal Research and Training Centre RALDO
Livestock Production Research Institute P.O. Box 3084
[Link] 202 ARUSHA
MPWAPWA

E.T.R Lyatuu A.A. Aboud

Livestock Production Research Institute DASP
P.O. Box 202 Sokoine University of Agriculture
MPWAPWA P.O. Box 3004
Chuo Kikuu
E.H. Goromela MOROGORO
DASP
Sokoine University of Agriculture
[Link] 3004 A. J. Mwilawa
Chuo Kikuu Ministry of Agriculture
MOROGORO LPRI
P.O. Box 202
MPWAPWA
D.R. Nkungu
DASP
Sokoine University of Agriculture Kagenguka
P.O. Box 3004 FSR
Chuo Kikuu Box Ilonga
MOROGORO KILOSA

D.S.C. Sendalo
Commissioner of Research and Training D.B. Mpiri
Ministry of Agriculture Ministry of Agriculture
Livestock Development and Cooperatives P.O. Box 2066
P.O. Box 2066 DAR ES SALAAM
DAR ES SALAAM
A. Mahuyemba
M.L. Kusekwa LMU
LPRO P.O. Box 195
P.O. Box 202 KAHAMA
MPWAPWA

N. Ole-Lengiongi
W.C.H. Mleche
KILIMO/MIFUGO
Ministry of Agriculture
P.O. Box 3084
Regional Agriculture/Livestock Office
ARUSHA
P.O. Box 512
MTWARA
K.J. Boki
KILIMO
V.G. Rushalaza P.O. Box 9152
Research and Training Department of DAR ES SALAAM
Ministry of Agriculture
Pasture Research Centre N.J.S. Mollel
P.O. Box 5 Sefian Agricultural Research Institute
KONGWA P.O. Box 6024
ARUSHA
Jim Crees
Ministry of Agriculture F. Akarro
ODA, Nairobi Uyole Agricultural Centre
P.O. Box 9192 Box 400
DAR ES SALAAM MBEYA

258 Small Ruminant Research Network Workshop

 List of Participants

A.B. Mtambuki S. Nyakalo

Uyole Agricultural Centre Sokoine University of Agriculture
P.O. Box 400 P.O. Box 3004
MBEYA MOROGORO

J.K.A. Bee O.S.J. Swai

LRC P.O. Box 1661
P.O. Box 147 BUKOBA
SANYA JUU

R. Saidi
D.N. Msafiri LRC
LITI P.O. Box 147
P.O. Box 51 SANYA JUU
MPWAPWA

WEST KILIMANJARO
J.K.K. Msechu G.R.M. Nsengwa
LPRI VIC
P.O. Box 202 P.O. Box 290
MPWAPWA IRINGA

R.P. Mbwile
LPRI H. Kyomo
P.O. Box 202 LITI Tengeru
MPWAPWA P.O. Box 3101
ARUSHA
J.K. Shiyumba
UAC
P.O. Box 400 D.M. Lubambe
MBEYA LITI Tengeru
P.O. Box 3101
J.K. Murro ARUSHA
UAC
P.O. Box 400 D.M. Komwihangilo
MBEYA LPRI
P.O. Box 202
G.L. Komba MPWAPWA
ACLD
[Link] 512
MTWARA J.I. Mkonyi
LPRI
Saluum P.O. Box 202
MOA MPWAPWA
P.O. Box 512
MTWARA A.R.M. Kaliba
LPRI
P.O. Box 202
P.A. Mkonyi MPWAPWA
MOA
P.O. Box 2066
DAR ES SALAAM TCHAD
R. Lancelot
J. Ngowi CIRAD-EMVT
FMC International Agriculture Laboratoire de Farcha
P.O. Box 81 18 B.P. 433
ARUSHA N'DJAMENA

Small Ruminant Research Network Workshop 259

List of Participants

TOGO ZAMBIA
Pessinaba W.N.M. Mwenya
Centre d'appui technique de Kolokope Department of Animal Science
B.P. 65 University of Zambia
ATAKPAME P.O. Box 32379
LUSAKA

UGANDA
ZIMBABWE
L. Sieffert
J.L.N. Sikosana
Makerere University - Kampala
Faculty of Veterinary Medicine Matopos Research Station
P.O. Box 7062 Private Bag K5137
KAMPALA BULAWAYO

P.R. Hatendi
U.K. Grasslands Research Station
С Hendy P. Bag 3701
NRI MARONDERA
CHATHAM
Kent J. Kategile
ME4 4TB ILCA-Harare
P.O. Box 3211
HARARE
USA
R.C. Noble F. Monicat
Tuskegee University Veterinary Goat Project
Department of Agricultural Science P.O. Box 1378
Milbank Hall #312 HARARE
TUSKEGEE
AL 36088 J. Keravec
Veterinary Goat Project
P.O. Box 1378
ZAIRE HARARE
Kafumbo Monga wa Ilunga
Université de Lubumbashi C. Shumba
B.P. 3473 AGRITEX
LUBUMBASHI 1 P.O. Box 354
MASVINGO

260 Small Ruminant Research Network Workshop

ISBN 92-9053-284-Х Printed at ILCA, Addis Ababa, Ethiopia