I /:MSTITUT SENEGALAIS DE RECHERCHES AGRICOLES ...
I
/:MSTITUT SENEGALAIS DE RECHERCHES
AGRICOLES (1 .S.R.A.)
DEPARTEMENT DE RECHERCHES SUR LES
PRODUCTIONS ET IA SANTE ANIMALES
- - - - - - - - - -
CENTRE DE RECHERCHES
ZOOTECHNIQUES DE KOLDA
AN ALTERNATIVE SELECTION
SCHEME FOft NDAJUA CA-ITLE
IN SENEGAL
BY
Abda~ FALL
M.S. veterimry Doctor
REF. No 030/CRZ/KOLDA
MAI 1989.
.
1. INTRODUCTION
A breeding programme aimed at the genetic improvement of the Ndama
cattle in the Southern subhumid tse-tse infested zone of Senegal has been
underway since 1972 by the CRZ* at Kolda which is a research station of
the Senegalese Institute for Agricultural Research. Mass selection is perfor-
med on progenies of 200 heads of breeding females of the station herd.
Improved bulls are introduced into village herds.
Research efforts directed toward the genetic improvement of local breeds
have been sujected to critisism. The vulnerability of these programmes to
critisism stems from their complex and costly nature. More over, they are
long term undertakings and their impact on herd productivity is not only
slow to appear but also is difficult to evaluate. Issues raised are relative to :
1) What is the relevance the breeding programmes. i.e.
what is the position of the genetic factor as compared to other
constraints facing livestock productions ?
.-.f
Il) What should be the objectives and selection criteria for a multi-purpose
production system ?
_’
II 1) What is the most suitable breeding scheme capable of ensuring a wide
spread genetic progress in an extensive production system characterized
by a stressfull environment and seasonality in inputs and outputs.
This article attempts to give responses to the issues mainly for the
Ndama cattle in the context of the Southern part of Senegal. The objectives,
methodes, scheme and constraints of the ongoing selection programme are
examined and the herd productivity collected on station are analysed in order
to gain insights for the identification of avenues which cari lead to the design
of an efficient breeding programme.
l . .
I . . .
* CRZ = Centre de Recherches Zootechniques de Kolda.
- 2
II. THE ONGDING BREEDING PRDGRAMM
2.1 - (lbjabs and nK%hade.
At a national level, increased meat output has been the main objective
which dictated the orientation of research and developments efforts on lives-
teck. This objective has been translated into the design and ,implementation,
among other means such health tare,
of breeding programmes aiming at the
genetic improvement of the beef potential of local ruminant breeds through
mass selection. The animal base is composed of herd of limited number raised
on station. Improved bulls selected from this herd are introduced into village
herds.
2.2-Sekctioncrïteriaandsdreae
Crowth and reproductive performances as well as coat color and format
are used as discriminatory traits for the choice of future parents. The brown
coat color was presumed to be correlated to the Ndama attribute of trypanoto-
Iérance.
A first selection is performed on calves at 6 months of age on the basis
of coat color an weight grain. The latter is also used for a second selection
applieds on 18 months old calf bulls. An individual test is then operated for
6 months on calf bulls of highest?GFih constitute about 10 9, of the initial
stock.
A final selection occurs at the end of this phase
ndidate bulls are
ranked according to growth performances, weight gai:? efficiency, sperm
quality and format.
2.3 - EvaIuation of the breeding pmqraarape
As stated before, the evaluation of the breeding programme is a diffi-
cuit task. We Will howerer carry it out on the basis of two criteria (1) the
growth performances obtained on station and (Il) the number of improved
breeding mâles introduced into village herds. Even through not comprehen-
sive, this appraoch Will help identify main drawbacks of this programme.
l
. . . . . .
- 4
ResUlt!§
The analysis of variante laid out in table 1 shows the significance of
different factors included in the mode1 for weight at different ages. Least
squares means are shown in table 2.
EffkcS of sesson of bïrth
The season of birth have a significant effect on weights at all
ages exept for the weight at 24 months. Animais born from february to july
have heavier weight than those born the rest of the year. The variation of
weight due to season of birth depends of the period of the year (wet or dry
season) when the animal reaches a given age.
Effectofageofdam
The effect of age of dam is significant from birth to 12 months. At 12
months, progenies of cows of more than 12 years of age are 19 % lighter than
calves from 8 to 11 years old CO~S.
.,. :
Effect of year of birth
The year of birth effect is significant for weights at all ages. Variation
of body weight accross years may be the result of the influence of many fac-
tors such as the changes in management practices with varying supplemental
levels, pasture availability dependant upon annual rainfall and the extent to
which pastures have been damaged by fire, the genetic progress and the
health situation.
Correlation and regression between means for weight (6 to 24 month
weight) and means for rainfall during the relevant period of the study were
negative and non-significant.
-5
TABLE. 1
- -._-
ANALYSIS OF VARIANCE OF BODY WEIGTH AT DIFFERENT AGES
----_.--..-
.-
_-. ..-. -. -^. ----._--__
..---_- _ _.
M E A N SQUARF S
d. f
d f
d . f
d . t
-
_.“._
-.-
d . t
.- - ..- .“. -_.__
--.---.
--.
__.
B i r t h
6 m
‘2 m
18 m
24 m
ISeason of i-iirtt
3
2 4 6 ” ’
3
9335+*
3
:972++
5
2932**
3
2 0 5 6
( A g e o f D a m
3
115**
3
1646**
3
i 701*
:
1638
3
1 t-559
l
jyear o f B i r t h
7
79s’”
7
2430**
6
211fj**
6
7397+*
5
17304**
1
227*”
1
3 4 9 4 +*
1
6298**
1
13051**
1
1s900**
1
S e a s o n X A g e o f D a m
9
13**
9
443s
9
993*
9
9 1 5
R,mainder
6 4 4
1 1
700
2 1 3
5 8 3
4 8 8
4 2 3
7 5 6
2 9 0
1 5 8 0
/
I .._
- - - - - --.---- --__
-. -- -... - .._ -_.-..
-.-- ------.-..-
/
** 1’ c .Ol
* P L.05
- 7
The regression of least square constants on year of birth (1 to 8) were
caculated to determine any linear trend in body weight as shown in table 3.
Table 3 : Correlation and regression of least-square constant for
weight on year of birth
Trait
Correlation
Regression 1
Birth
$81
.38*
6 month weight
- .62
- 1.32
12 month weight
-,27
-
.58
I
18 month weight
-.63
- 2 . 6 4
24 month weight
- ,.52
- 3.52
l
i
P < 0 5
Correlations and regressions are negative and non significant for the 8
years studied. These statisticç have proven to be positive and çignificant
from the previous analyçis of the data over the period lasting from 1974 to
1980. Table 4 shows the differences in growth performances during these
two periods.
Dramatic weight declines
have occured during the second period (1981
to 1988) as compared to weights in the first period of study (1974 to 1980).
Eventhough the correlations of least square constants on year of birth were
not significant their negative nature illustrate s a negative trend of cattle
body weight over years. With the non-estimation of genetic parameters (wich
Will be performed) it may be difficult to associate this phenemenon
to either
4
environmental or genetic factors. However some observations may lead to the
incrimination of environmental factors. The decline in growth performances
may be attributed to changes in management practices and paçture availabi-
Iity. The level of supplementation has substantially dropped during the second
period because of budget conçtraints. Moreover, tremendouç paçture areas
have been devaçtated by fire during 1985 and 1986,. The gradua1 degradation
of available grazing areas after 17 years of utilisation without any attempts to
improve them may contribute
to explain this situation. Hundreds of hectares
are being encroached
by Woody species with a consequent decline in grass
especies density. In addition, half of the station posture areas have been
c4;\\rr.rr II.-. L .^^_.,_^ ,c - --^-_-- --* --.--a
* .
- 7
The regression of least square constants on year of birth (1 to 8) were
caculated to determine any linear trend in body weight as shown in table 3.
Table 3 : Correlation and regression of least-square constant for
weight on year of birth
T r a i t
Correlation
l
R e g r e s s i o n (
l
Birth
i81
.38*
-.62
- 1.32
!
6 month weight
12 month weight
-,27
-
.58
18 month weight
-- .63
- 2.64
24 month weight
- ,.52
3.52
-
i
P < 0 5
Correlations and regressions are negative and non significant for the 8
years studied. These statistics have proven to be positive and significant
from the previous analysis of the data over the period lasting from 1974 to
1980. Table 4 shows the differences in growth performances during these
two periods.
Dramatic weight declines have occured during the second period (1981
to 1988) as compared to weights in the first period of study (1974 to 1980).
Eventhough the correlations of least square constants on year of birth were
not significant their negative nature illustrate s a negative trend of cattle
body weight over years. With the non-estimation of genetic parameters (wich
Will be performed) it may be difficult to associate this phenemenon
to either
<
environmental or genetic factors. However some observations may lead to the
incrimination of environmental factors. The decline in growth performances
may be attributed to changes in management practices and pasture availabi-
lity. The level of supplementation has substantially dropped during the second
period because of budget constraints. Moreover, tremendous pasture areas
have been devastated by fire during 1985 and 1986. The graduai degradation
of available grazing areas after 17 years of utilisation without any attempts to
improve them may contribute
to explain this situation. Hundreds of hectares
are being encroached by Woody species with a consequent decline in grass
especies density. In addition, half of the station pasture areas h a v e b e e n
given up because of management constraints.
- 8
Table 4 : Overall least-square
means for weight at b i r t h , 6, 1 2 , 1 8
and 24 months from 1974 to 1988
Overall least-square mean (kg)
~---------------------
n------------------------
1974-1980
1980-1988
Birth
403
17.7
6 6 8
17.5
6 months
403
81.6
7 3 2
69.7
1 2 months
306
111.5
6 0 6
87.7
1 8 months
306
144.0
4 4 6
103.9
2 4 months
189
175.0
303
124.7
Effktofsex
The effect of sex was significant for weight at all ages with females
lighter than males.
Effect of the interaction season of bitht qe of dam
Weights at 6 and 12 months are significantly influenced by this interac-
tion. Heaviest weights are obtained by calves born between february and july
and wich are progenies of 8 to 11 old CO~S. Lightest calves are born between
august and january and are progenies of Young (2 - 3 years) and old (more
than 12 years) CO~S.
2.3.2 - Canstraints of ihe breedinq programme
2.3.2.1 - The first resultats of the analysis of cattle body weight on station
give the evidence that a breeding programme under circumtances
of a low nutritional plane may not be efficient in terms of increased
productivity which is the main stated objective. The depression of
growth performances attributed to the effects of environmental fac-
tors indicate that if a sustained and significant genetic progress is
to be made in a stressfull
environment such as that prevailing at
Kolda, a great deal of efforts should conconmitantly to the selection
programme, be directed to trie improvement of the nutritional an
disease situation.
- 9
2.3.2.2 -’ Si-nce the beginning of the breeding programme at Kolda an annuat
mean of 5 breedings males have been introduced into village herds.
This is a major factor limiting the impact of the selection programme
based on natural mating with regard to the size of the target regio-
nal herd totaling an estimated number of 350 000 Ndama breeding
females.
2.3.2.3 - The limited size of the animal base raised on station causes a reduc-
tion of the genetic variability and the selection intensity which are
important elements of the selection differential.
111. AN ALTERNATIVE BREEDING PROCWE
The stop of the genetic programmes and the focus on other more constrai-
ning factors facing the livestock sector has been advocated in Senegal. The
genetic improvement of livestock is a complex and time-consuming process
which can’t be postponed. Advantages such as the introduction of innova-
tions into village herds brought about through the genetic programme justify
the allocation of ressources and efforts to its implementation. It is however
fundamental to design an efficient breeding programme with a clear definition
of objectives, scheme and organisation which cari fit the production system and
apt to satisfy the needs of ail persons involved in the process.
3.1 - Objectives
The planning of a breeding programme requires a clear formulation of
objectives to ensure its success. The objectives of Iivestock owners must be
taken into account. In the traditional production system cattle is raised for
many purposes. Milk production is a major function of Ndama cattle. Mill-i
yield
is therefore used as selection criterion by cattle keepers. Growth
performances are used as well. The consideration of meat and milk out puts
may meet the livestock producers and national needs.
J
. . . . . .
- 10
Adaptability of Ndama cattle in the tsetse infested environment deserves
a great deat of attention.
A dual purpose (beef and milk) animal with a high degree of tolerance
to trypanosomiasis seems to be the most relevant genotype in the Kolda
context.
The main objectives of the breeding programme cari then be stated as
follow :
1) the improvement of the beef potential of Ndama cattle assuming that this
is not antagonic to the development of its traction power ;
2) the improvement of its milk potential ;
3) the increase of the tolerance of Ndama cattle to trypanosomiasis.
3.2 -Methodandcriterïa
Constraints associated to the use of artificial insemination dictates the
adoption of mass selection.
Future parents Will be selected on the basis of the following criteria :
1) dam milk yield ;
2) growth performances ;
3) an adaptability criteria which is yet to be identified.
Packed cell volume (PCV) might serve as an indicator of trypanosomiasis
“the repeatability of the monthly PCV of cattle in contransting situations
ranged from 0.26 to 0.39 with a mean value of 0.32. This indicates the possi-
bility of a heritability estimate sufficient to justify the cost of a selection
programm. ILCA, 1986, p. 78”.
l
. . . . . .
- 11
Figure 1 gives the proposed selection scheme. It is character,ied b y
- the test of cal
bulls from village herds on station
*
- the shift of the present station role to a number of multiplication herds.
The multiplication herds are those traditional herds that have been single
out beacause of their better management practices in terms of health tare and
feeding regime they apply.
They Will be converted into commercial herds
receiving higher ranking bulls or semen from the station and saling breeding
males to others herders.
Males progenies of high milk yieding cows from village herds are tested
on station from 10 to 18 months of age. A final selection is perfoemed at the
end of the test on the basis of their growth performances. Best bulls are
directed into multiplication herds.
3 . 1 - Requi-ts
The level and variability of cattle performances under village circumtance!
as well as the significance of the effects of factors causing variation are not
precisely known. These information are needed as inputs to an efficient
selection.
They cari however be obtained through the design and the implementation of
a recording system which Will have the role to collect and analyse data pertai-
ning to village herds productivity.
A monitoring scheme invulving 2 000 heads
of cattle is presently conducted by the CRZ/Kolda. A. substantial number of
herds are also monitored by the SODEFITEX which is a developement project
intervening in that zone ? For the selection programme to have a regional
impact it is required that this monitoring scheme be strenghten
with person-
nel, logistic means and equipement with the involment if all institutions concer-
ned with cattle production in that region.
A second requirement deals with the organisation of lives teck owners
involved in the breeding process. The “Groupement d’lnteret Economique :
r IMPROVED
7
*
PACKAGES
- HEALTH
*
STATION
4
VILLAGE HERS
1
I
Selection on
1
Dam milk
1
1
YOUNG MALES
FROM VILLAGE
-._-_
<
-4 -
i
/
C-----Adaptation (2 months)
1
/
i
-Test on growth (6 months)
î
/
/
ASelection at 18 months on growth
I
r
u
-.
w
IMPROVED
L3 BULLS
4
L
1
I
Figure 1 : The selection Scheme
- 12
CIE” in the new formulation of the cooperative system in rural Senegal. Its
structure and functionningof fer opprotunities to farmers for them to have
access to bank loans for the purchase of basic inputs such as veterinary
products, supplemental feed, bulls, etc.. o It is also a valuable tool at their
disposa1 to facilitate livestock owners to market their products. An efficient
functionning of the selection scheme cari be acheived through the organisa-
tion of livestock owners into CIE framework.
3 . 5 - Cawtraints
The lack .of data on village herds productivity makes difficult the selec-
tion of calves only on the basis of their genetic merit. Correction factors
needed to adjust performances in order to minimize the effects on environmen-
tal factors are not available.
Difficulties associated to the application artificial insemination Will retard
the selection process. Artificial insemination should be attempted in the multi-
plication herds with high management standards.
The available personnel, logistic means and equipement does not permit
the wide spread impact of the selection programme.
3 . 5 - hlplerentation
Two major phases are identified for the implementation of this selection
scheme :
Phase I (2 to 3 years)
- calves are selected on the basis of their dam milk yield estimated by lives-
teck owners ;
- a recording and analysis system is designed and implemented ;
- livestock owners are organized into CIE ;
- multiplication herds are identified ;
- series of 100 calves are tested each year.
. ..!...
- 13
Phase II
- The system of recording and analysis is working fairly well ;
- the firt results of the programme are evaluated ;
help
- findings from the firsts phase/adjust and improve the selection programme.
:i. .i:
- 15
R E F E R E N C E S
1 - F A L L ( A . ) , D I O P ( M . ) , S A N D F O R D ( J . ) , W I S S O C A ( Y . J . ) , D U R R I N ( J . ) ,
TRAIL (J. C.M.) (1982) - Evaluations of the productivitiec of Djallonke sheep
and Ndama cattle at the Centre de Recherches Zootechniques Kolda, Senegal.
ILCA Research Report n03.
2 - HARVEY (R.W.) (1987) - Parmcard, Paramter tard generation Program for
LSMLMW Muxed Mode1 Least-Squares and Maximum Likehhood Computer Prograr
3 - ILCA (1986) - The African Trypanosolerant Livestock fdetwork. Indications
from results. 1983-1985.
/:MSTITUT SENEGALAIS DE RECHERCHES
AGRICOLES (1 .S.R.A.)
DEPARTEMENT DE RECHERCHES SUR LES
PRODUCTIONS ET IA SANTE ANIMALES
- - - - - - - - - -
CENTRE DE RECHERCHES
ZOOTECHNIQUES DE KOLDA
AN ALTERNATIVE SELECTION
SCHEME FOft NDAJUA CA-ITLE
IN SENEGAL
BY
Abda~ FALL
M.S. veterimry Doctor
REF. No 030/CRZ/KOLDA
MAI 1989.
.
1. INTRODUCTION
A breeding programme aimed at the genetic improvement of the Ndama
cattle in the Southern subhumid tse-tse infested zone of Senegal has been
underway since 1972 by the CRZ* at Kolda which is a research station of
the Senegalese Institute for Agricultural Research. Mass selection is perfor-
med on progenies of 200 heads of breeding females of the station herd.
Improved bulls are introduced into village herds.
Research efforts directed toward the genetic improvement of local breeds
have been sujected to critisism. The vulnerability of these programmes to
critisism stems from their complex and costly nature. More over, they are
long term undertakings and their impact on herd productivity is not only
slow to appear but also is difficult to evaluate. Issues raised are relative to :
1) What is the relevance the breeding programmes. i.e.
what is the position of the genetic factor as compared to other
constraints facing livestock productions ?
.-.f
Il) What should be the objectives and selection criteria for a multi-purpose
production system ?
_’
II 1) What is the most suitable breeding scheme capable of ensuring a wide
spread genetic progress in an extensive production system characterized
by a stressfull environment and seasonality in inputs and outputs.
This article attempts to give responses to the issues mainly for the
Ndama cattle in the context of the Southern part of Senegal. The objectives,
methodes, scheme and constraints of the ongoing selection programme are
examined and the herd productivity collected on station are analysed in order
to gain insights for the identification of avenues which cari lead to the design
of an efficient breeding programme.
l . .
I . . .
* CRZ = Centre de Recherches Zootechniques de Kolda.
- 2
II. THE ONGDING BREEDING PRDGRAMM
2.1 - (lbjabs and nK%hade.
At a national level, increased meat output has been the main objective
which dictated the orientation of research and developments efforts on lives-
teck. This objective has been translated into the design and ,implementation,
among other means such health tare,
of breeding programmes aiming at the
genetic improvement of the beef potential of local ruminant breeds through
mass selection. The animal base is composed of herd of limited number raised
on station. Improved bulls selected from this herd are introduced into village
herds.
2.2-Sekctioncrïteriaandsdreae
Crowth and reproductive performances as well as coat color and format
are used as discriminatory traits for the choice of future parents. The brown
coat color was presumed to be correlated to the Ndama attribute of trypanoto-
Iérance.
A first selection is performed on calves at 6 months of age on the basis
of coat color an weight grain. The latter is also used for a second selection
applieds on 18 months old calf bulls. An individual test is then operated for
6 months on calf bulls of highest?GFih constitute about 10 9, of the initial
stock.
A final selection occurs at the end of this phase
ndidate bulls are
ranked according to growth performances, weight gai:? efficiency, sperm
quality and format.
2.3 - EvaIuation of the breeding pmqraarape
As stated before, the evaluation of the breeding programme is a diffi-
cuit task. We Will howerer carry it out on the basis of two criteria (1) the
growth performances obtained on station and (Il) the number of improved
breeding mâles introduced into village herds. Even through not comprehen-
sive, this appraoch Will help identify main drawbacks of this programme.
l
. . . . . .
- 4
ResUlt!§
The analysis of variante laid out in table 1 shows the significance of
different factors included in the mode1 for weight at different ages. Least
squares means are shown in table 2.
EffkcS of sesson of bïrth
The season of birth have a significant effect on weights at all
ages exept for the weight at 24 months. Animais born from february to july
have heavier weight than those born the rest of the year. The variation of
weight due to season of birth depends of the period of the year (wet or dry
season) when the animal reaches a given age.
Effectofageofdam
The effect of age of dam is significant from birth to 12 months. At 12
months, progenies of cows of more than 12 years of age are 19 % lighter than
calves from 8 to 11 years old CO~S.
.,. :
Effect of year of birth
The year of birth effect is significant for weights at all ages. Variation
of body weight accross years may be the result of the influence of many fac-
tors such as the changes in management practices with varying supplemental
levels, pasture availability dependant upon annual rainfall and the extent to
which pastures have been damaged by fire, the genetic progress and the
health situation.
Correlation and regression between means for weight (6 to 24 month
weight) and means for rainfall during the relevant period of the study were
negative and non-significant.
-5
TABLE. 1
- -._-
ANALYSIS OF VARIANCE OF BODY WEIGTH AT DIFFERENT AGES
----_.--..-
.-
_-. ..-. -. -^. ----._--__
..---_- _ _.
M E A N SQUARF S
d. f
d f
d . f
d . t
-
_.“._
-.-
d . t
.- - ..- .“. -_.__
--.---.
--.
__.
B i r t h
6 m
‘2 m
18 m
24 m
ISeason of i-iirtt
3
2 4 6 ” ’
3
9335+*
3
:972++
5
2932**
3
2 0 5 6
( A g e o f D a m
3
115**
3
1646**
3
i 701*
:
1638
3
1 t-559
l
jyear o f B i r t h
7
79s’”
7
2430**
6
211fj**
6
7397+*
5
17304**
1
227*”
1
3 4 9 4 +*
1
6298**
1
13051**
1
1s900**
1
S e a s o n X A g e o f D a m
9
13**
9
443s
9
993*
9
9 1 5
R,mainder
6 4 4
1 1
700
2 1 3
5 8 3
4 8 8
4 2 3
7 5 6
2 9 0
1 5 8 0
/
I .._
- - - - - --.---- --__
-. -- -... - .._ -_.-..
-.-- ------.-..-
/
** 1’ c .Ol
* P L.05
- 7
The regression of least square constants on year of birth (1 to 8) were
caculated to determine any linear trend in body weight as shown in table 3.
Table 3 : Correlation and regression of least-square constant for
weight on year of birth
Trait
Correlation
Regression 1
Birth
$81
.38*
6 month weight
- .62
- 1.32
12 month weight
-,27
-
.58
I
18 month weight
-.63
- 2 . 6 4
24 month weight
- ,.52
- 3.52
l
i
P < 0 5
Correlations and regressions are negative and non significant for the 8
years studied. These statisticç have proven to be positive and çignificant
from the previous analyçis of the data over the period lasting from 1974 to
1980. Table 4 shows the differences in growth performances during these
two periods.
Dramatic weight declines
have occured during the second period (1981
to 1988) as compared to weights in the first period of study (1974 to 1980).
Eventhough the correlations of least square constants on year of birth were
not significant their negative nature illustrate s a negative trend of cattle
body weight over years. With the non-estimation of genetic parameters (wich
Will be performed) it may be difficult to associate this phenemenon
to either
4
environmental or genetic factors. However some observations may lead to the
incrimination of environmental factors. The decline in growth performances
may be attributed to changes in management practices and paçture availabi-
Iity. The level of supplementation has substantially dropped during the second
period because of budget conçtraints. Moreover, tremendouç paçture areas
have been devaçtated by fire during 1985 and 1986,. The gradua1 degradation
of available grazing areas after 17 years of utilisation without any attempts to
improve them may contribute
to explain this situation. Hundreds of hectares
are being encroached
by Woody species with a consequent decline in grass
especies density. In addition, half of the station posture areas have been
c4;\\rr.rr II.-. L .^^_.,_^ ,c - --^-_-- --* --.--a
* .
- 7
The regression of least square constants on year of birth (1 to 8) were
caculated to determine any linear trend in body weight as shown in table 3.
Table 3 : Correlation and regression of least-square constant for
weight on year of birth
T r a i t
Correlation
l
R e g r e s s i o n (
l
Birth
i81
.38*
-.62
- 1.32
!
6 month weight
12 month weight
-,27
-
.58
18 month weight
-- .63
- 2.64
24 month weight
- ,.52
3.52
-
i
P < 0 5
Correlations and regressions are negative and non significant for the 8
years studied. These statistics have proven to be positive and significant
from the previous analysis of the data over the period lasting from 1974 to
1980. Table 4 shows the differences in growth performances during these
two periods.
Dramatic weight declines have occured during the second period (1981
to 1988) as compared to weights in the first period of study (1974 to 1980).
Eventhough the correlations of least square constants on year of birth were
not significant their negative nature illustrate s a negative trend of cattle
body weight over years. With the non-estimation of genetic parameters (wich
Will be performed) it may be difficult to associate this phenemenon
to either
<
environmental or genetic factors. However some observations may lead to the
incrimination of environmental factors. The decline in growth performances
may be attributed to changes in management practices and pasture availabi-
lity. The level of supplementation has substantially dropped during the second
period because of budget constraints. Moreover, tremendous pasture areas
have been devastated by fire during 1985 and 1986. The graduai degradation
of available grazing areas after 17 years of utilisation without any attempts to
improve them may contribute
to explain this situation. Hundreds of hectares
are being encroached by Woody species with a consequent decline in grass
especies density. In addition, half of the station pasture areas h a v e b e e n
given up because of management constraints.
- 8
Table 4 : Overall least-square
means for weight at b i r t h , 6, 1 2 , 1 8
and 24 months from 1974 to 1988
Overall least-square mean (kg)
~---------------------
n------------------------
1974-1980
1980-1988
Birth
403
17.7
6 6 8
17.5
6 months
403
81.6
7 3 2
69.7
1 2 months
306
111.5
6 0 6
87.7
1 8 months
306
144.0
4 4 6
103.9
2 4 months
189
175.0
303
124.7
Effktofsex
The effect of sex was significant for weight at all ages with females
lighter than males.
Effect of the interaction season of bitht qe of dam
Weights at 6 and 12 months are significantly influenced by this interac-
tion. Heaviest weights are obtained by calves born between february and july
and wich are progenies of 8 to 11 old CO~S. Lightest calves are born between
august and january and are progenies of Young (2 - 3 years) and old (more
than 12 years) CO~S.
2.3.2 - Canstraints of ihe breedinq programme
2.3.2.1 - The first resultats of the analysis of cattle body weight on station
give the evidence that a breeding programme under circumtances
of a low nutritional plane may not be efficient in terms of increased
productivity which is the main stated objective. The depression of
growth performances attributed to the effects of environmental fac-
tors indicate that if a sustained and significant genetic progress is
to be made in a stressfull
environment such as that prevailing at
Kolda, a great deal of efforts should conconmitantly to the selection
programme, be directed to trie improvement of the nutritional an
disease situation.
- 9
2.3.2.2 -’ Si-nce the beginning of the breeding programme at Kolda an annuat
mean of 5 breedings males have been introduced into village herds.
This is a major factor limiting the impact of the selection programme
based on natural mating with regard to the size of the target regio-
nal herd totaling an estimated number of 350 000 Ndama breeding
females.
2.3.2.3 - The limited size of the animal base raised on station causes a reduc-
tion of the genetic variability and the selection intensity which are
important elements of the selection differential.
111. AN ALTERNATIVE BREEDING PROCWE
The stop of the genetic programmes and the focus on other more constrai-
ning factors facing the livestock sector has been advocated in Senegal. The
genetic improvement of livestock is a complex and time-consuming process
which can’t be postponed. Advantages such as the introduction of innova-
tions into village herds brought about through the genetic programme justify
the allocation of ressources and efforts to its implementation. It is however
fundamental to design an efficient breeding programme with a clear definition
of objectives, scheme and organisation which cari fit the production system and
apt to satisfy the needs of ail persons involved in the process.
3.1 - Objectives
The planning of a breeding programme requires a clear formulation of
objectives to ensure its success. The objectives of Iivestock owners must be
taken into account. In the traditional production system cattle is raised for
many purposes. Milk production is a major function of Ndama cattle. Mill-i
yield
is therefore used as selection criterion by cattle keepers. Growth
performances are used as well. The consideration of meat and milk out puts
may meet the livestock producers and national needs.
J
. . . . . .
- 10
Adaptability of Ndama cattle in the tsetse infested environment deserves
a great deat of attention.
A dual purpose (beef and milk) animal with a high degree of tolerance
to trypanosomiasis seems to be the most relevant genotype in the Kolda
context.
The main objectives of the breeding programme cari then be stated as
follow :
1) the improvement of the beef potential of Ndama cattle assuming that this
is not antagonic to the development of its traction power ;
2) the improvement of its milk potential ;
3) the increase of the tolerance of Ndama cattle to trypanosomiasis.
3.2 -Methodandcriterïa
Constraints associated to the use of artificial insemination dictates the
adoption of mass selection.
Future parents Will be selected on the basis of the following criteria :
1) dam milk yield ;
2) growth performances ;
3) an adaptability criteria which is yet to be identified.
Packed cell volume (PCV) might serve as an indicator of trypanosomiasis
“the repeatability of the monthly PCV of cattle in contransting situations
ranged from 0.26 to 0.39 with a mean value of 0.32. This indicates the possi-
bility of a heritability estimate sufficient to justify the cost of a selection
programm. ILCA, 1986, p. 78”.
l
. . . . . .
- 11
Figure 1 gives the proposed selection scheme. It is character,ied b y
- the test of cal
bulls from village herds on station
*
- the shift of the present station role to a number of multiplication herds.
The multiplication herds are those traditional herds that have been single
out beacause of their better management practices in terms of health tare and
feeding regime they apply.
They Will be converted into commercial herds
receiving higher ranking bulls or semen from the station and saling breeding
males to others herders.
Males progenies of high milk yieding cows from village herds are tested
on station from 10 to 18 months of age. A final selection is perfoemed at the
end of the test on the basis of their growth performances. Best bulls are
directed into multiplication herds.
3 . 1 - Requi-ts
The level and variability of cattle performances under village circumtance!
as well as the significance of the effects of factors causing variation are not
precisely known. These information are needed as inputs to an efficient
selection.
They cari however be obtained through the design and the implementation of
a recording system which Will have the role to collect and analyse data pertai-
ning to village herds productivity.
A monitoring scheme invulving 2 000 heads
of cattle is presently conducted by the CRZ/Kolda. A. substantial number of
herds are also monitored by the SODEFITEX which is a developement project
intervening in that zone ? For the selection programme to have a regional
impact it is required that this monitoring scheme be strenghten
with person-
nel, logistic means and equipement with the involment if all institutions concer-
ned with cattle production in that region.
A second requirement deals with the organisation of lives teck owners
involved in the breeding process. The “Groupement d’lnteret Economique :
r IMPROVED
7
*
PACKAGES
- HEALTH
*
STATION
4
VILLAGE HERS
1
I
Selection on
1
Dam milk
1
1
YOUNG MALES
FROM VILLAGE
-._-_
<
-4 -
i
/
C-----Adaptation (2 months)
1
/
i
-Test on growth (6 months)
î
/
/
ASelection at 18 months on growth
I
r
u
-.
w
IMPROVED
L3 BULLS
4
L
1
I
Figure 1 : The selection Scheme
- 12
CIE” in the new formulation of the cooperative system in rural Senegal. Its
structure and functionningof fer opprotunities to farmers for them to have
access to bank loans for the purchase of basic inputs such as veterinary
products, supplemental feed, bulls, etc.. o It is also a valuable tool at their
disposa1 to facilitate livestock owners to market their products. An efficient
functionning of the selection scheme cari be acheived through the organisa-
tion of livestock owners into CIE framework.
3 . 5 - Cawtraints
The lack .of data on village herds productivity makes difficult the selec-
tion of calves only on the basis of their genetic merit. Correction factors
needed to adjust performances in order to minimize the effects on environmen-
tal factors are not available.
Difficulties associated to the application artificial insemination Will retard
the selection process. Artificial insemination should be attempted in the multi-
plication herds with high management standards.
The available personnel, logistic means and equipement does not permit
the wide spread impact of the selection programme.
3 . 5 - hlplerentation
Two major phases are identified for the implementation of this selection
scheme :
Phase I (2 to 3 years)
- calves are selected on the basis of their dam milk yield estimated by lives-
teck owners ;
- a recording and analysis system is designed and implemented ;
- livestock owners are organized into CIE ;
- multiplication herds are identified ;
- series of 100 calves are tested each year.
. ..!...
- 13
Phase II
- The system of recording and analysis is working fairly well ;
- the firt results of the programme are evaluated ;
help
- findings from the firsts phase/adjust and improve the selection programme.
:i. .i:
- 15
R E F E R E N C E S
1 - F A L L ( A . ) , D I O P ( M . ) , S A N D F O R D ( J . ) , W I S S O C A ( Y . J . ) , D U R R I N ( J . ) ,
TRAIL (J. C.M.) (1982) - Evaluations of the productivitiec of Djallonke sheep
and Ndama cattle at the Centre de Recherches Zootechniques Kolda, Senegal.
ILCA Research Report n03.
2 - HARVEY (R.W.) (1987) - Parmcard, Paramter tard generation Program for
LSMLMW Muxed Mode1 Least-Squares and Maximum Likehhood Computer Prograr
3 - ILCA (1986) - The African Trypanosolerant Livestock fdetwork. Indications
from results. 1983-1985.