Status of Artificial Insemination Service in Ethiopia

Desalegn G/Medhin1, b, Merga Bekana2, Azage Tegegne3, Kelay Belihu2

1
EMDTI; P.O.Box-1573, Debre Zeit, Ethiopia; [email protected];
2
AAU, Faculty of Vet. Med.; Debre Zeit, Ethiopia;
3
ILRI, Addis Ababa, Ethiopia

Abstract

A study on reproductive performance of 18 artificial insemination (AI) bulls, which were in semen pro-
duction during the study period and on the efficiency of AI operations were conducted at the National
Artificial Insemination Center (NAIC) and in ten purposively selected areas of five regional states. The
study was further conducted to assess problems and constraints associated with the AI service in Ethio-
pia. Evaluations of number of services per conception and conception rates to first insemination were
carried out. Pregnancy diagnosis was performed on 375 cows in the selected areas. Besides, 114 straws
of semen comprising of 61 from regions and 53 from NAIC were checked for semen motility to see if
there were any differences in motility due to handling between the center and the regions. Thirty AI
technicians, 17 technical staff of NAIC and 246 farmers were used for the collection of data using ques-
tionnaire surveys. Moreover, fifty-two high-level professionals who have stakes directly or indirectly
in the area of AI service have participated in the focus group discussions in the five regions. Mean (±
SE) scrotal circumference for Holstein, Jersey and crosses of Holstein with indigenous breeds at NAIC
were 39.71± 0.64, 41.00 ± 1.16 and 39.00 ±1.00 respectively. The results did not differ (p>0.05) among
the three breeds. Similarly, semen physio-morphological analysis revealed that there was no difference
(P>0.05) for the different semen characteristics except for the total morphological  defects  which  dif-
fered significantly (P<0.05) among the three breeds. The source, the selection procedures, and health sta-
tus  of  the  AI  bulls  at  NAIC  were  found  to  have  serious  problems.  Mean  (±  SE)  number  of  servic-
es  per  conception  and  conception  rate  to  first  inseminations  were  1.88 ± 0.07 and 27.06 ± 0.44,
respectively. These results differ significantly (P<0.001) among the five regions. Of the total cows used
during the study, 40.53% have been found to be repeat breeders. Mean (± SE) semen motility for the
NAIC and for the regions were 53.2% and 51.7%, respectively and there was no difference (P>0.05). The
overall results of the assessment via questionnaire surveys and focus group discussions at large have
indicated that artificial insemination is not doing well in all the regions and at national level at large.
It is inferred, therefore, that artificial insemination operation in Ethiopia is not a success and hence re-
quires urgent corrective measures.

Keywords: Semen; Semen motility; National Artificial Insemination Center; AI efficiency

1. Introduction
Agriculture (mainly crop and livestock production) is the mainstay of the Ethiopian economy em-
ploying approximately 85% of the total population. Livestock production accounts for approxi-
mately 30% of the total agricultural GDP and 16% of national foreign currency earnings (Lobago,
2007). The total cattle population for the rural sedentary areas of Ethiopia is estimated at 43.12
million, of which 55.41% are females (CSA, 2006). Out of the total female cattle population, only
151,344 (0.35%) and 19,263 (0.04%) heads are hybrid and exotic breeds, respectively. With an av-
erage lactation length of 6 months and an average daily milk production of 1.44 liters per cow, the
total milk produced during the year 2006/07 was recorded to be 2.634 billion liters (CSA, 2006).
This suggests that the total number of both exotic and hybrid female cattle produced through the
crossbreeding work for more than four decades in the country is quite insignificant indicating un-
successful crossbreeding through AI.

Artificial insemination (AI) has been defined as a process by which sperm is collected from the
Ethiopian Society of Animal Production

male, processed, stored, and artificially introduced into the female reproductive tract for the pur-
pose of conception (Webb, 2003). Semen is collected from the bull, deep-frozen and stored in a
container with Liquid Nitrogen at a temperature of minus 196 degrees Centigrade and made for
use. Artificial insemination has become one of the most important techniques ever devised for the
genetic improvement of farm animals. It has been widely used for breeding dairy cattle as the
most valuable management practice available to the cattle producer and has made bulls of high
genetic merit available to all (Webb, 2003; Bearden et al., 2004).

In Ethiopia, AI was introduced in 1938 in Asmara (the current capital city of Eritrea), the then
part of Ethiopia, which was interrupted due to the 2nd World War and restarted in 1952 (Yemane
et al., 1993). It was again discontinued due to unaffordable expenses of importing semen, liquid
nitrogen and other related inputs requirement. In 1967, an independent service was started in the
then Arsi Region, Chilalo Awraja under the Swedish International Development Agency (Sida).
The technology of AI for cattle has been introduced at the farm level in the country over 35 years
ago as a tool for genetic improvement (Zewdie et al., 2006). The present National Artificial Insemi-
nation Center (NAIC) was established in 1984 to coordinate the overall AI operation at national
level (GebreMedhin, 2005). The efficiency of the service in the country, however, has remained at
a very low level due to infrastructure, managerial, and financial constraints, as well as poor heat
detection, improper timing of insemination and embryonic death (Shiferaw et al., 2003).

Cattle breeding is mostly uncontrolled in Ethiopia making genetic improvement difficult and an
appropriate bull selection criteria have not yet been established, applied and controlled (Tegegn
et al., 1995). Although artificial insemination, the most commonly used and valuable biotechnol-
ogy (Webb, 2003) has been in operation in Ethiopia for over 30 years, the efficiency and impact
of the operation has not been well-documented (Himanen and Tegegn, 1998). Reproductive prob-
lems related to crossbreed dairy cows under farmers’ conditions are immense (Bekele, 2005). It is
widely believed that the AI service in the country has not been successful to improve reproductive
performance of dairy industry (Sinishaw, 2005). From the previous little studies, it has been found
that AI service is weak and even declining due to inconsistent service in the smallholder livestock
production systems of the Ethiopian highlands (Dekeba et al., 2006). The problem is more aggra-
vated by lack of recording scheme, wrong selection procedures, and poor management of AI bulls
associated with poor motivations and skills of inseminators (GebreMedhin, 2005).

The objective of the study was, therefore, to assess the status of AI service in Ethiopia and identify
problems and constraints associated with it; come up with pertinent and workable recommenda-
tions that could call upon decision makers and stakeholders to give the utmost attention to the AI
operation in the country.

2. Materials and methods

2.1. Study areas
The research was conducted at the NAIC and in ten selected areas of five regions, viz., Addis
Ababa, Southern Nations Nationalities and Peoples Region, Oromia, Tigrai, and Amhara (Figure
1). The regions were arbitrarily selected as regions 1, 2, 3, 4, and 5 starting from Addis Ababa (1)
up to Amhara (5). All the study areas were purposively selected because these areas are the ones
where AI service is widely practiced and they constitute wide range of agro- ecology.

The NAIC is the only place where semen is produced for use all over the country and is responsible
for the coordination of AI operations in the country. During the research work, between August
and October, 2007, 18 bulls were used for semen collection in the center.

88 17th ESAP-Proceedings
 Animal Breeding

Figure 1. Illustration of the study areas

2.2. Study population

All the bulls, which were in use for semen production (n=18) and all animal science professionals
(n=17) during the study period in the NAIC were involved in the study. Dairy farmers (n=246),
dairy cows and heifers (n=375), and AI technicians (n=30) were also used. Moreover, animal health
and animal production professionals in federal, regional, zonal and Woreda agriculture and rural
development offices, NGOs, research centers, and academic institutions (n=52) were represented
in the study population in the five regions.

2.3. Study design

A cross-sectional type of study supported by questionnaire survey, observation, focus group dis-
cussion, and laboratory analysis of semen was carried out from August 2007 to June 2008.

2.4. Sample size determination and sampling procedures

Sample sizes were determined by the following formula (Thrusfield, 2005)

Total cows (n) = Zα2 x p (1-p)

d2
Where:

P (expected prevalence) = 0.5

d (desired absolute precision) = 0.05

17th ESAP-Proceedings 89
Ethiopian Society of Animal Production

Zα = 1.96.

Stratified sampling was used to determine the total number of farmers and cows for each region
while systematic random sampling method was used to select cows and farmers within region.
Table 1 depicts the summary of the number of cows, farmers and number of participants for focus
group discussion in each study area during the research work.

Table -1: Distribution of farmers, cows, AITs, and FGD participants in the five regions

Region Woreda Sample size AITs Focus group discussion

Cows Farmers Number of Number of Number of Number of par-
groups participants groups ticipants
Addis Addis Ababa 87 83 6 6 1 9
Ababa
SNNPR Dale 69 34 1 2
Awassa 15 10 1 5 1 10
Oromia Tiyo 54 24 1 2
Ada’a 16 17 1 1 1 9
Tigrai Atsbi 15 16 1 4
Mekelle 55 22 1 5 1 10
Amhara Fogera 36 12 1 3
Bahir Dar _ 14 1 1
Zuria 1 14
Bahir Dar 28 14 1 1
Total 375 246 15 30 5 52
AITs= artificial insemination technicians; FGD= focus group discussion

2.5. Data collection

2.5.1. Semen quality analysis and evaluation of bulls

Physical and retrospective evaluations of semen producing bulls at NAIC, physio-morphological

assessment of semen in laboratory were conducted. All the 18 bulls were checked for locomatory
and other physical situations at standing status, during sexual stimulation and during mount-
ing. The bulls were graded as very good, good, or with hoof problem for locomatory organs while
they were graded as very good, good, adequate, or poor for libido according to (Hansen, 2006). In
grading locomatory status of bulls, the situation of hoof and moving ease was considered as crite-
ria. A bull graded as very good in locomatory means he has no any type of hoof problem and can
move and jump easily; good means that a bull has some untrimmed hooves but has no difficulty in
jumping or moving; while a bull graded under “hoof problem” means that the bull has untrimmed
hooves and lesions which seriously hinder his locomatory and jumping ease. The time elapsed
until sexual desire is fully aroused has been taken as the criteria for grading libido of each bull.
A bull showing libido in less than five minutes is graded as very good, in 5 minutes is graded as
good, in 5 to 10 minutes is graded as Adequate, while a bull showing sexual desire 10 to 30 min-
utes is graded as Poor. Scrotal measurement for all the 18 bulls was performed using a flexible
scrotal circumference tape (Hansen, 2006).

Three consecutive semen collections once per week were performed in the NAIC to determine the
semen quality of the bulls. The collection of semen was performed using artificial vagina accord-
ing to Bearden et al., (2004); Hafez, (1993). Semen characteristics were assessed in the semen
laboratory of the Centre. Stage- heated microscope with a magnification of 100x for mass activity,
200x for sperm motility, and 1000x for morphology were used. The values for semen parameters
were recorded every week while semen was collected. The parameters included were semen color,
volume, mass activity, individual motility, live-dead percentage, and morphological abnormalities

90 17th ESAP-Proceedings
 Animal Breeding

of the semen collected from AI bulls. Semen analysis was carried out in accordance with the pro-
cedures provided by Garner, (1991); Herman et al., (1994); Bearden et al., (2004).

Evaluation of spermatozoa motility was performed by taking semen samples from each region and
from the NAIC to determine if there are major changes in motility between NAIC and the regions
due to handling. The evaluation was conducted by transporting sample semen from each study
area to the semen laboratory at NAIC using a three-litter capacity container with periodically
controlling and filling liquid nitrogen and by making maximum care for the semen. Semen motil-
ity was determined in a total of 93 straws from 16 bulls, in which 49 belonged to regions and the
other 44 to NAIC. Moreover, 17 straws of new batch semen of which eight were from region 5 and
9 from NAIC were analyzed for motility to check why semen motility from region 5 was differently
so low.

2.5.2. Pregnancy diagnosis and assessment of services per conception

The number of services per conception and conception rates to first insemination in each study
area were determined by rectal palpations of all the cows and heifers to confirm pregnancies.

2.5.3. Questionnaire survey

Structured questionnaires were prepared to interview farmers, AI technicians, and animal health
and animal production professionals to collect data on the status of AI services and constraints
associated with the service. The questionnaire surveys to AI technicians comprised of two types.
First category consisted of questions that relate to the overall status of AI service in their respec-
tive weredas while the second one is directly related to their technical proficiency. Results related
to technical proficiencies of AI technicians which are above 85% were categorized as “Excellent”,
71 to 85 % “Very good”, 60 to 70 % “Good” while those below 60% were “Poor”.

2.5.4. Focus group discussion

A total of five groups were organized one in each region. Each group was made up of 9 to 14 profes-
sionals originating from development, academic and research institutions in the regions. Before
the start of the discussion, participants of each group were briefed about the objective of the study
and invited to participate in the discussion.

2.6. Data management and statistical analysis

Data was analyzed using SPSS (release 11.05, 2002) and was summarized using descriptive sta-
tistics. Mean comparison was done by one way ANOVA while rank comparison was made by
Mann-Whitney U test.

3. Results
3.1. Morphological characteristics of semen and bull evaluation
Most of the bulls were under the category of very good libido (77.8%) and good locomatory organs.
Four bulls (22.2%) had hoof problem affecting their locomotion.

The scrotal circumference of bulls in the NAIC didn’t vary (p>0.05) significantly between breeds
(Table 3). However, the highest value for scrotal circumference was observed in Jersey cattle fol-
lowed by Holstein-Friesian and crosses of Holstein-Friesian and indigenous cattle.

17th ESAP-Proceedings 91
Ethiopian Society of Animal Production

Only total morphological defects showed significant difference between the breeds of bulls (p<0.05).
The highest proportion of defects was observed in Jersey bulls (18.8%) followed by crossbred
(13.33%) and Holstein bulls (8.5%).

Table 2: Summary of libido and locomatory organs evaluations of bulls at NAIC

Parameter Bull breeds Attributes Count Proportion (%)

Locomatory Holstein Friesian (HF) Very good 0 0

(n=10) Good 9 90
Hoof problem 1 10
Jersey(n=5) Very good 0 0
Good 4 80
Hoof problem 1 20
Cross(HF*Local) Very good 0 0
(n=3)
Good 1 33.3
Hoof problem 2 66.7
Libido Holstein Friesian (HF) Very good 10 100
(n=10)
Good 0 0
Adequate 0 0
Poor 0 0

Jersey(n=5) Very good 4 80

Good 1 20
Adequate 0 0
Poor 0 0
Cross(HF*Local) Very good 1 33.3
(n=3)
Good 2 66.7
Adequate 0 0
Poor 0 0

n= number of bulls

Table -3: Scrotal circumferences of semen producing bulls at NAIC 

Parameter Bull breeds Mean(SE) 95%CI P-value

Holstein Friesian(n=7) 39.71(0.64) 38.14-41.29

Scrotal circumference 0.44
Jersey (n=3) 41.00(1.16) 36.03-45.97

HF*L crossbred 39.00(1.00) 26.29-51.71

Bulls (n=2)
n= number of bulls

Table 4: Semen characteristics of bulls at NAIC

Parameters Holstein-Friesian Jersey Crosses of HF with indig-

enous
Mean (SE) 95% CI Mean (SE) 95% CI Mean (SE) 95% CI
Volume(Ml) 8.87(0.90) 6.84-10.90 5.68(0.37) 4.67-6.70 8.00(1.85) 0.03-15.97
Concentration (109) 1.40(0.97) 1.18-1.62 1.24(0.09) 0.98-1.50 1.23(0.38) -0.42-2.89
Mass activity (1-5)a 3.45(0.21) 2.97-3.93 3.38(0.12) 3.05-3.70 2.80(0.49) 0.67-4.92
Live cell percentage 75.90(1.20) 73.19-78.61 74.40(1.81) 69.39-79.41 76.33(2.85) 64.08-88.59
Fresh individual motility 73.97(3.06 67.04-80.90 75.60(0.87) 73.18-78.02 74.33(1.86) 66.35-82.32
Individual motility at dispatch 54.50(3.83) 45.83-63.17 59.00(5.10 44.84-73.16 43.33(3.33) 28.99-57.68

92 17th ESAP-Proceedings
 Animal Breeding

Major morphological defects 2.27(0.93) 0.16-4.38 6.58(3.08) -1.98-15.14 3.33(1.33) -2.40-9.07

Total morphological defects 8.50(1.85) 4.32-12.69 18.80(2.85) 10.89-26.72 13.33(4.67) -6.74-33.41
a1=very poor, 2=poor, 3=good, 4=very good, 5= excellent

The color of semen varied from milky to watery. The milky color accounted for 41.2% and the
creamy color for 37.3%. There were also semen from some bulls with bloody (7.8%), watery (3.9%),
and yellow (9.8%) colors. The results indicated that 14.29 % of the semen colors for the Holstein
bulls were bloody while 22.22 % for that of the crosses were watery.

Table 5: Proportion of semen colors for semen produced during the study period at NAIC

Parameter Bull breeds Attributes Count Proportion (%)

Semen colors Holstein Friesian (HF) Milky 12 42.86

(n=30 )
Creamy 9 27.78
Bloody 4 14.29
Watery 0 0
Yellow 5 17.86
Jersey(n=12) Milky 8 66.67
Creamy 4 33.33
Bloody 0 0
Watery 0 0
Yellow 0 0
Cross(HF*Local) Milky 1 11.11
(n=9 ) Creamy 6 66.67
Bloody 0 0
Watery 2 22.22
Yellow 0 0
Overall averages(n=49 ) Milky 21 41.17
Creamy 19 37.25
Bloody 4 7.84
Watery 2 3.92
Yellow 5 9.80
n =number of straws

3.2. AI organization and efficiency of AI service

The questionnaire survey revealed that 82% of the NAIC technical staff confirmed that there are
no appropriate collaborations and communications between the NAIC, regional agriculture bu-
reaus and other stakeholders. Similarly, participants of the focus group discussions in all study
regions unanimously supported the above idea. Again, 88.2% of the NAIC staff said that the AI
service is not doing appropriately at national level and that there was no structured and proce-
dural way of recruiting semen-producing bulls by the center and selection was being done from
only one farm. Seventy one percent of the respondents revealed that bulls selection and manage-
ment at NAIC is not being done according to recommended procedures and all agreed that the
Center is not regularly and appropriately checking bulls for reproductive diseases. Eighty eight
% of the respondents confirmed that there was no policy that guides the distribution of semen in
the country.

Results of the focus group discussions have revealed that there were no functionally effective re-
sponsible bodies at the respective regional and wereda levels to coordinate the AI services and no
proper mechanisms of controlling indiscriminate inseminating/breeding. Moreover, it was found
with full consensuses that there were problems associated with the AI service with regard to prop-
erly carrying out responsibilities by the NAIC, the regional, and the wereda agriculture bureaus

17th ESAP-Proceedings 93
Ethiopian Society of Animal Production

in that these institutions did not perform their responsibilities appropriately to correct the con-
tinuing problems associated with the AI service in the country. .

The evaluation of the technical know-how of AI technicians showed that most of them were in the
category of good (56.7%) and very good (26.7%). The remaining were evaluated as poor (10%) and
excellent (10%). Regarding their motivations, 86.67% of the respondents were totally unmotivated
while the remaining had some motivation (13.33). Nearly all (96.7%) of the AI technicians respond-
ed that they never got on job trainings and other incentives. About three fourth (73.3%) of the AI
technicians were not giving service during weekends while others did so on personal agreements
(26.3%) in which they receive additional payments from the farmers. Most of the AI technicians
complained that liquid nitrogen is not readily available (86.7%) and the remaining 13.3% had no
problem in getting liquid nitrogen. More than three fourth of the AI technicians (76.7%) believed
that there is a risk of indiscriminate insemination while the others either did not have any idea
about the problem (13.3%) or believed it can be controlled. Seventy percent of the AITs confirmed
that they started their careers as AITs within the last ten years while the more experienced ones
had abandoned and/or changed their jobs. Ninety-three percent said that they do not believe that
NAIC was doing its responsibilities properly. Eighty seven percent of the AITs revealed that farm-
ers were willing to pay more fees for the services provided they get reliable and quality services.
On the other hand, 80 percent of them revealed that they do no get necessary supports by the
respective wereda and regional bureaus of agriculture to perform their duties appropriately. Simi-
larly, 83 percent said that AI service delivery was not consistent in their respective areas. Ninety
seven percent of the AITs revealed that they were neither satisfied nor happy with their jobs as
AITs because of the very little attention given to it by all responsible bodies.

Ninety three percent of the farmers who participated in the questionnaire surveys bitterly dem-
onstrated that they were not getting reliable and consistent AI service and 81% of them explained
the reasons for that as absence of service on weekends & holidays, shortage of AITs and shortage
of inputs. In relation to this, 64% of them said that they pass estruses of in heat cows without
breeding them while 34% said that they use bull/ natural mating at times where they do not get
the AI service. The study has clearly confirmed that 93.1% of all the farmers participated in the
study areas showed dissatisfaction with the overall AI service. In many places, it was observed
that farmers trek their cows for more than 28km round trip to get the AI service though they
couldn’t be sure to get the service after traveling for such a long distance. This was happening in
many areas. The majority of farmers (91.5%) confirmed their willingness to pay more fees for the
service provided they get reliable, efficient, and effective services. Moreover, 90% of the AIT’s who
took part in the study confirmed that the AI service is not doing well in their respective areas due
to many impeding situations (6.7% responded “AI is doing well” and 3.3% of them abstained from
giving any judgment).

In Bahir Dar, where artificial insemination service was being delivered by a private AIT only,
farmers were supposed to pay not less than 30 Birr per single insemination as compared to 3-5
Birr in other areas of the country. Nevertheless, in the same place with the same AIT, the NSC for
Jerusalem Children’s and Community Development Organization was found to be as low as 1.3.
This was found to be because of the special arrangements made between the AIT and the organiza-
tion in which the AIT was paid 50 Birr per pregnancy and not per insemination which is believed
to motivate the technician to improve his conception rates to the best of his capacity.

Ninety percent of the farmers said that they usually have herd health problems, which directly
and indirectly have impacts on the efficiency of the AI service. The diseases of major importance
in order of prevalence were mastitis (39%), tuberculosis (24%), problems associated with calving
(17%), and others. In relation to this, 76 percent of the farmers said that they do not have easy
access to animal health services while 21 percent said they more or less get the service.

94 17th ESAP-Proceedings
 Animal Breeding

The necessity of an efficient national body responsible to coordinate AI service was emphasized by
the NAIC personnel (94.1%), and unanimously by all AI technicians and all the focus group discus-
sion groups. Most of the respondents in the questionnaire survey (82.4%), all the AI technicians,
and all the groups in focus group discussion confirmed that there was no appropriate collaboration
and regular communication between NAIC and stakeholders. Lack of breeding policy, absence of
herd recording system, inefficient management at national level and lack of clearly defined share
of responsibilities among stakeholders were the constraints mentioned by 70.6% of the AI person-
nel at NAIC and all the groups in the focus group discussion. Lack of ownership and follow-up to
the AI service, inadequate manpower and lack of commitment, lack of attention and incentives to
AIT’s, limitation of inputs and facilities and structural problems were also mentioned by 29.4% of
the NAIC personnel, all of the AI technicians and all the groups in the focus group discussion.

Regarding the possible solutions, constituting a national breeding policy and strategy, restructur-
ing the service at all levels, creation of awareness, diversifying the service, and establishment of a
system of monitoring and evaluation, and encouraging the private involvement in the sector were
emphasized by 52.94% of the NAIC personnel, all of the AI technicians and all the groups involved
in the focus group discussion. Strong stakeholder participations, establishment of ownership to
the sector, integration of AI service with health and feeds packages, re-establishment of incentive
mechanisms to AITs and coordination of efforts to alleviate the technical management problems
of the NAIC were also mentioned by 11.76% of the NAIC personnel, all the AI technicians and all
the groups in the focus group discussion.

3.3. Assessment of efficiency of AI operations

Semen motility at NAIC (53.2%, n=44) and in the regions after dispatch (51.7%, n=49) was not sig-
nificantly different (p>0.05). The result indicated that there was no significant difference (p>0.05)
in semen motility among the regions as well (Table 6).

However, a special case was observed whereby the average value of motility for semen obtained
from region 2 (SNNP) was as low as 20%. Similarly, the average motility for semen obtained from
region 5 was as low as 32.5%. Nine straws of semen for the same bulls as that of region 5 were
taken and checked for motility from NAIC for assessing the reasons why the motility in region 5
was so low and it revealed that the average motility was low, too ( 38%).

Table -6: Effect of handling of semen between  NAIC & regions on semen motility

Parameter Region n Mean (SE) 95%CI P-value

1 22 51.59(1.49) 48.50-54.68
Semen motility
2 6 50.00(3.65) 40.61-59.39 0.77
3 15 51.67(1.74) 47.94-55.39
4 6 54.17(2.01) 49.01-59.33
Overall re-
gions 49 51.73(0.97) 49.78-53.69

NAIC 44 53.18(1.01) 51.05-55.31 0.32

n =number straws used for the analysis of semen motility

The number of services per conception (NSC) and conception rates to first insemination (CRF) in
the different study areas are presented in Tables 7 and 8 respectively. There was significant dif-
ference among the study regions in NSC (p<0.001). The least NSC was observed in Amhara (1.55)
followed by Addis Ababa (1.70) and Oromia (1.78). Highest value for NSC was recorded for Tigrai
(2.47). The values of NSC did not take in to account the 152 cows (40.53% of the total sampled
cows) which were found to be repeat breeders and not considered during the analysis for NSC.

17th ESAP-Proceedings 95
Ethiopian Society of Animal Production

There was also significant variation between the regions in CRF. The highest CRF was observed
in Addis Ababa (40.23 %) and the least was found in Tigrai (7.14%).

Table 7: NSC by region

Parameter Region N N1 Mean(SE) 95% CI P-value

Addis Ababa 64 109 1.70 (0.14) 1.42-1.98

NSC SNNPR 44 84 1.91(0.19) 1.53-2.29 0.000

Oromia 46 82 1.78 (0.06) 1.03-1.28
Tigrai 47 116 2.47 (0.12) 2.04-2.51
Amhara 22 34 1.55 (0.17) 1.19-1.90

Total 223 425 1.88 (0.07) 1.74-2.02

n =total number of pregnancies

N1 total number of inseminations

Table 8: CRF by region

Parameter Region N N1 Mean (SE) 95% CI P-value

Addis Ababa 87 35 40.23 (0.50) 39.25-41.21

SNNPR 84 28 33.33 (0.13) 33.07-33.59

Oromia 70 24 34.29 (0.37) 33.56-35.02 0.000

CRF (%) Tigrai 70 5 7.14 (0.85) 5.47-8.81

Amhara 64 13 20.31(1.25) 18.31-22.31

Total 375 105 27.06(0.44) 26.19-27.92

n =total number of first inseminations
N1 total number of pregnancies to first services

4. Discussions
In the present study, it is found that the bulls at NAIC originated from one farm and no standard
guidelines are used during selection. This situation is supported by the reports given by Agegnehu
et al. (2008) and this was not because of lack of guidelines and standards rather failure to follow
the already set guidelines. Agegnehu (2007) and Zewde (2007) indicated that AI bulls need to be
carefully recruited and the frequency of tests for performance and for diseases at NAIC were not
sufficient. According to the international animal health code (2001) of the Office International
des Epizooties (OIE)), as cited by Agegnehu (2007), donor and teaser animals should be tested for
bovine brucellosis, bovine tuberculosis, bovine viral diarrhea, infectious bovine rhinotracheitis,
Campylobacter fetus, and trichomoniasis. However, it was found that NAIC did not follow this
procedure. It is obvious, therefore, that there is a high risk of disseminating sexually transmitted
diseases.

The problems associated with hoofs of the semen producing bulls at NAIC is believed to be due to
miss handling of the bulls. The results of the evaluation of libido of the bulls were encouraging but
endeavors should have been made to upgrade the libido status of those found within the “good”
category to higher ones. Different reasons for low libido have been given by Arthur et al., (1982)
which include age, heredity, environment and disease or defects of the penis in which case the
penis fails to make contact with the vulva either because of insufficient protrusion or because it is
deviated ventrally or laterally.

96 17th ESAP-Proceedings
 Animal Breeding

The mean value of the scrotal circumferences for the three breeds of bulls at NAIC ranged from
39.71 to 41cm with no sign of difference (P>0.05) between the three breeds. This is believed to be
due to the ages of the bulls of the three breeds which was above 30 months and above such ages
the scrotal circumference becomes very similar for both Bos taurus and Bos indicus breeds (Han-
sen, 2006). As regards to the overall evaluation of semen characteristics, results indicated that
there were no significant differences between the three breeds for all the semen traits except for
the total morphological defects. The highest total morphological defect rate is observed in Jersey
bulls followed by crosses and Friesian bulls . However, the rate of total morphological defects in
this study (8.50% for Friesian and 13.33% for crosses) is lower than the reports of Sinshaw (2005)
(16.10% for Friesian, and 18.01% for crosses).

The volume of semen are not significantly different (p>0.05) among the three breeds which is
not in line to the results described by Sinishaw (2005) and Sori (2004) who reported 6.77% for
Friesian breeds and 4.4 % for indigenous breeds, respectively. The absence of significant differ-
ence among the three breeds could be due to individual differences in performance or differences
in sample size among the three breeds. According to Zewdie et al. (2005), the volume of bull’s
semen varies between ejaculates, individual bulls, breed, and age. Other reasons, which could
contribute to differences in semen volume, include nutritional status, geographic locations, season
of years of study, method of semen collection and handling of bulls during collection and procedure
and frequency of collection (Salisbury et al., 1978; Hafez, 1993; Blezinger, 1999; Andrabie et al.,
2002). Sorensen (1979), Sinishaw (2005), and Sori (2004) reported similar findings as the current
research for the other parameters (semen motility, semen concentration, live cell percent, and
morphological defects). 

The results of the assessment of semen colors for all the three breeds of bulls in percentages was
found to be 41.2, 37.3, 7.8, 3.9, and 9.8 for milky, creamy, bloody, watery and yellowish, respec-
tively. The proportions of milky and creamy colors do not differ much from the findings of Sin-
ishaw (2005), while they differ from that of Sori (2004) (77.61% and 1.49% respectively). The other
color proportions differ much from those of Sori (2004) (17.92% watery, 1.49% bloody, and none
for watery) and Sinishaw (2005) (none for watery, and bloody, 20%yellowish). The presence of se-
men with abnormal colors indicates problems associated with the process of spermatogenesis and
pathologies on the accessories and reproductive organs and the reproductive passage ways. The
normal color of bull semen ranges from thick whitish to slightly yellowish fluid whose consistency
is mainly determined by the number of spermatozoa it contains (Herman et al., 1994).

The outcome of the assessment of AI technicians regarding their technical knowledge was 10%,
26.7 %, 53.3%, and 10%, respectively for excellent, very good, good, and poor. This result and the
finding that most of the technicians do not get on job trainings shows that there was some deficit
indicating a need for upgrading the capacity of technicians through giving proper trainings par-
ticularly for those who were under the category of the poor technical expertise.

The study has found an alarming result with motivations of the AI technicians in which 86.67% of
them have indicated that they are not motivated to work as AI technicians due to associated prob-
lems and constraints. This is fully supported by the reports of the FAISETD (2007) that indicated
a very high turnover of AI technicians all over the country. The situation is closely associated to
the discontinuation of in-service trainings and incentive mechanisms which had been practiced
during the past.

Again, it was revealed that all responsible bodies from federal to wereda levels, particularly the
NAIC, are not giving proper attention to the AI service which indicates that decision makers need
to work hard to improve the situation of the AI operation at national level.

It was found that 94.1% of NAIC personnel, 100% of AI technicians, and full consensus of the focus
group discussions supported the necessity of an efficient national responsible body to coordinate

17th ESAP-Proceedings 97
Ethiopian Society of Animal Production

the AI service. This result agrees with the reasons identified as problems of AI by Chupin and
Schuh (1992) and Zewdie et al. (2006).

Again, the outcome of the study revealed that there is no appropriate collaboration and communi-
cations among stakeholders in which the NAIC failed to do which consequently contributed to the
unsuccessfulness of the service.

The findings regarding the constraints associated with the AI service at national level/regional lev-
els are in agreement with the suggestions of Tegegn et al. (1989), Mukassa-Mugerwa et al. (1989),
Garner (1991), Jokinen and Bertilson (2003), Bearden et al. (2004), Bekele (2005), Gebremedhin
(2005), Gibson et al. (2006), ESAP (2007) and Zewde (2007). This indicated that the NAIC was not
functioning well and was consistently losing the confidence of stakeholders in the country.

In the present study, there is no difference in semen motility among the regions, and between the
regions and the NAIC due to differences in handling. This justifies that there is no significant effect
on semen motility due to handling differences. However, the motility observed in four straws of se-
men obtained from region 2 (SNNP) is as low as 20% on the average. The reason for this condition
is believed to be that the semen was produced from 50% bulls and distributed without demand to
the region and was kept for many years without being utilized. Similarly, the mean semen motili-
ties for sample semen obtained from region 5 is as low as 32.5% and that of sample semen from
the same bulls taken from NAIC is as low as 38%. Even though statistically significant differences
(p<0.05) between the two was found the value for both was lower than the recommended lower
values, 40% (Zewdie et al., 2005). The reason was hypothesized to be either of or a combination
of two facts. The first assumption was that the center shifted from using Laiciphos to Andromed
as semen extenders during the production of this particular batch and the changes might have
occurred during the transition periods due to lack of experiences in using new chemicals and/or
fresh laboratory technicians. The other possible reason assumed to cause the difference was also
the shifting of the long-experienced laboratory technicians by new ones during that time.

The overall mean number of services per conception obtained in this study ( 1.88) as a national
average is lower than the reports of Abate (2006) (2.48) but within the range reported by Bacha
(2007) (1.7-2.7) and higher than the finding of Kaziboni et al. (2004) (1.65-1.74). The values of
number of services per conception in the current study did not take in to account 152 cows (40.53%
of the total sampled cows) which were found to be repeat breeders. The very high repeat breeding
condition in the country is believed to be a serious problem and needed to be seriously addressed.
This finding and the associated causes do coincide with the reasons for the high numbers of ser-
vices per conception as the results of problems associated with poor semen quality, poor semen
handling practices and poor insemination practices (Bekana et al., 2005; Macpherson, 1968; Mc-
Dowell, 1972; Negussie, 1992). Other reasons were such as discontinuation of incentives to AI
technicians, season of breeding, management factors in relation to estrus detection, timing of
insemination and skill of pregnancy diagnosis were also indicated by other reports (Haile-Mariam,
1987; Swensson et al., 1981; Denbarga, 2005; Haile-Mariam et al., 1993; Negussie et al., 1998;
Damron, 2000 and Shiferaw et al., 2003).

The difference in the number of services per conception among the five regions was significant. It
was understood during the study that the NSC could have dramatically increased than the actual/
obtained result if it was possible to include in the study all the cows/heifers, which were continu-
ously being sold due to repeat breeding in many farms.

The reason why the semen motility for region 5 was very low while the NSC was in good status is
that the assessment of NSC was done for semen produced earlier than the production of the last
batch which was used for the motility analysis.

98 17th ESAP-Proceedings
 Animal Breeding

The conception rates to first inseminations in general were poor ranging from 7.14 to 40.23%. The
differences for these parameters among regions were also highly significant. This is also an indi-
cator of very low level of fertility or reproductive efficiency, which could be due to AI inefficiency
among other reasons. The number of services per conception is an indicator of reproductive effi-
ciency/ conception rates (Albero, 1993; Bekele et al., 1991; Haile-Mariam et al., 1993; Negussie et
al., 1998; Shiferaw et al., 2003).

5. Conclusion and recommendations

AI service in Ethiopia has been given little or no emphasis at the federal, regional or wereda levels
during the last years even though it is a widely practiced animal biotechnology all over the world.
The most important constraints associated with AI service in Ethiopia include less structural
linkage between AI Center and service giving units, absence of collaborations and regular com-
munications between NAIC and stakeholders, lack of breeding policy and herd recording system,
inadequate resource in terms of inputs and facilities, and absence of incentives and rewards to
motivate AI technicians. The values for semen volume, concentration, live-dead percentage, fresh
individual motility, and thaw motility, were within the recommended values whereas the values
for morphological defects among the jersey and crossbred bulls and mass activity were not encour-
aging. The locomatory organs and libido of AI bulls were in general good but needed to be checked
regularly and get improved for some of them particularly the status of hoof problems within the
bulls. The repeat breeding situation was a very alarming finding. Similarly, the conception rates
were found to be very low and discouraging at large. The source of AI bulls particularly the HF
bulls was only one farm that was highly tuberculosis-infected farm amongst others, which is an
obvious hazard to the AI operation in the country.

Serious attention should be given by all concerned bodies (policy makers, researchers, practitio-
ner, development workers, etc)

Based on the above conclusions the following are recommended:

•• Efficient national body responsible to coordinate and monitor AI service, herd recording and
also livestock breeding programs should either be reorganized or formed as new;
•• Mechanisms should be devised to increase the involvement of stakeholders in the activities of
AI in the country. Professional associations should critically work in close collaborations with
the Ministry of Agriculture and Rural Development in formulating policies and implementa-
tion strategies;
•• Selection of bulls for AI should strictly follow the standard guidelines and procedures set for
the purpose and also the national livestock development policies of the country;
•• The private sector should be encouraged to be involved in the AI service sector but with strict
control by an active breeding policy;
•• Establishment of a functional breeding policy and strategy should be given at most priority
and each stakeholder and professional should work hard towards its implementation;
•• Import semen of the desired quality for the immediate use in accordance with the rules and
regulations for the import of genetic materials to be followed by creating reliable source of se-
men producing bulls through reestablishing the Milk recording Scheme of the center in a more
strengthened status;
•• The AI service provision should be restructured in such a way that it responds well to the
breed improvement programs of the country. It should be well organized with clearly defined
duties and responsibilities of stakeholders;
•• Endeavors should be made to improve the current status of conception rates at large by im-
proving the efficiency of AI and heard health;

17th ESAP-Proceedings 99
Ethiopian Society of Animal Production

•• Establishment of cattle collection centers in selected areas in the different regional states to
practice natural mating using proven bulls could be considered as an alternative to the AI;
•• Further study should be conducted on assessment of the fertility levels of herds at national
level.
Acknowledgement
This research was conducted in collaboration with the Faculty of Veterinary Medicine of Addis
Ababa University, the National Artificial Insemination Center of the Ministry of Agriculture and
Rural development, Improving Productivity and Market Access of Ethiopian Farmers (IPMS)/
ILRI, and Sida/ SAREC.

Due thanks also goes to all who directly and indirectly contributed to the successful accomplish-
ment of the research work.

Reference
Abate H. (2006): Evaluation of artificial insemination service efficiency and reproductive performance in
North Gondar Zone, Ethiopia. MSc Thesis. Alemaya University, Alemaya.

Agegnehu B. (2007): Performance evaluation of semen production and distribution team for 2006/07 fiscal
year.NAIC, Addis Ababa.

Agegnehu B., Desalkegn H., and Alehegn W. (2008): Study report on the Tuberulosis status of Holeta Dairy
Farm, NAIC. Addis Ababa.

Albero M. (1993): Comparative performance of F1 FriesianXZebu heifers in Ethiopia. Anim Prod 37,
247-252.

Andrabi S.M.H., Naheed S., Khan L.A., Ullah N. (2002): Semen characteristics of crossbred (Friesian x Sahi-
wal) and Sahiwal young bulls maintained under subtropical conditions of Punjab. Pakistan Veterinary
Journal 23, 100-102.

Arthur G.H., Noakes D.E., Pearson H. (1982): Veterinary Reproduction and Obstetrics. Theriogenology. 5th
ed. ELBS/Bailliere Tindall. Pp. 417-479.

Bacha B. (2007): Sub clinical endometritis and its effect on reproductive performance in crossbred dairy cows
in Debre Zeit. MSc Thesis, Addis Ababa University, Faculty of Veterinary Medicine.

Bearden H.J., Fuquary, J.W., Willard, S.T. (2004): Applied Animal Reproduction. 6th ed. Mississippi
State University. Pearson, Prentice Hall. Upper Saddle  River, New Jersey 07458. Pp. 155-233.

Bekana M., Gizachew A., and Regassa F. (2005): Reproductive Performance of Fogera Heifers Treated with
Prostaglandin F2a for Synchronization of Estrus. Tropical Animal Health and Production. 37, 373-
379.

Bekele T. (2005): Calf Sex Ratios in Artificially Inseminated and Natural Mated Female Crossbred Dairy
Herd. In: proceedings of the 13th annual conference of the Ethiopian   Society of Animal Production.
Addis Ababa, Ethiopia, Pp. 225-230.

Bekele T., Kasali O.B., Alemu T.(1991): Reproductive problems in crossbred cattle in Central Ethiopia.Anim
Prod Sci. 26, 41-49.

Blezinger S. (1999): Cattle Today: Many factors affect bull performance. http://www.CattleToday.com.

100 17th ESAP-Proceedings

 Animal Breeding

Chupin D. and Schuh, H. (1992): Survey of present status of the use of artificial insemination in developing
countries. Animal Production and Health Division, FAO, Rome, Italy.

CSA, Central Statistics Agency, Federal Democratic Republic of Ethiopia (2006): Agricultural Sample Sur-
vey 2006/07, volume II, Report on livestock and livestock characteristics. Statistical Bulletin 388. Addis
Ababa, Ethiopia. Pp. 9-10, 25- 27.

Damron W.S. (2000): Introduction to Animal Science: Global, Biological, Social and Industry Perspectives.
Oklahoma State University. Prentice Hall, Upper Saddle River, New Jersey 07458. Pp. 221-224.

Dekeba A., Ayalew W., Hedge P.B., and Taddese Z. (2006): Performance of the Abernossa Ranch in
the production of Ethiopian Boran x Holstein crossbred dairy heifers in Ethiopia. In: Ethiopian Journal
of Animal production 6, 33-53.

Denbarga Y. (2005): A Study on reproductive performance of crossbred dairy cows in smallholder dairy farms
in South East Shoa Zone of Oromia Region, Ethiopia. MSc Thesis. Addis Ababa University, Faculty of
Veterinary Medicine.

ESAP, Ethiopian Society of Animal Production (2007): Policy and Research Recommendations and final dec-
laration of the 15thannual Conference of the Ethiopian Society of Animal Production. ESAP recommen-
dations series No.2. Addis Ababa, Ethiopia.

FAISETD, Field AI service Extension and Training Department, NAIC (2007): Millennium report. National
AI Center, Addis Ababa.

Garner D.L. (1991): Artificial Insemination. In: Cupps, P.T. (ed.), Reproduction in Domestic Animals. 4th
ed. San Diego, California: Academic Press Inc., Pp.251-274.

GebreMedhin D. (2005): All in one: A Practical Guide To Dairy Farming. Agri-Service Ethiopia Printing
Unit, Addis Ababa. Pp. 15-21.

Gibson J., Gamage S., Hannote O., Iniguez L., Maillard JC. Rischkowsky B., Semambo, D., Toll J. (2006):
Options and Strategies for the Conservation of Farm Animal Genetic Resources: Report of an inter-
national Workshop (7-10November 2005, Montpellier, France). CGIAR System-wide Genetic resources
Programme (SGRP)/Biodiversity International, Rome, Italy. Pp 15-16, 35-36.

Hafez E.S.E. (1993): Reproduction in Farm Animals. 6th ed. Philadelphia: Lea and Febiger. Pp. 405-439.

Haile Mariam M. (1987): Evaluation of growth and reproductive performances of Boran cattle and their
crosses with Friesian at Abernossa, Ethiopia. MSc Thesis, Alemaya University of Agriculture, Dire
Dawa, Ethiopia.

Haile Mariam M., Banjaw K., GebreMeskel J. and Ketema T. (1993): Productivity of Boran Cattle and their
Friesian crosses at Abernossa Ranch, Ethiopia. I. Reproductive performance and prevailing mortality.
Trop Anim Health & Prod. 25, 239-248.

Hansen, G.R. (2006): Managing Bull Fertility in Beef Cattle Herds. Animal Science Department. Florida Co-
operative, Extension, Service, Institute of Food and Agricultural Sciences, University of Florida. http://
edis.ifas.ufl.edu.

Herman H.A., Mitchell J.R., Doak G.A. (1994): The Artificial Insemination and Embryo Transfer of Dairy
and Beef Cattle: A Handbook and Laboratory Manual. IPP, Interstate  Publishers, INC. Danville, Il-
linois. Pp. 3-7; 45-244.

Himanen A., Tegegn A. (1998): A Proposal for Establishment of a National Milk Recording and Herd Regis-
tration Scheme in Ethiopia. Ministry of Agriculture, Addis Ababa, Ethiopia.

17th ESAP-Proceedings 101

Ethiopian Society of Animal Production

Jokinen S. and Bertilsson, L. (2003): Observations on the National Dairy Herd  Performance Monitoring
Program of the National Artificial Insemination Centre (NAIC). Swedish University of Agricultural Sci-
ences (SLU), ISSN 1402-3237. Pp. 21-27.

Kaziboni S., Kusina N T, Sibanda S., Makuza S., Nyoni O and Bhebhe(2004): Performance of artificial in-
semination in smallholder dairies of Nharira-Lankashire in Zimbabwe. Animal Science Department,
University of Zimbabwe. Mount Pleasant, Harare, Zimbabwe.

Lobago F. (2007): Reproductive and Lactation Performance of Dairy Cattle in the Oromia Central Highlands
of Ethiopia with Special Emphasis on Pregnancy Period. Doctoral thesis, Swedish University of Agricul-
tural Sciences, Uppsala.

Macpherson J.W. (1968): Semen Placement effects on fertility in bovines. J. Dairy Sci., 51:807.

Mc Dowell, R.E. (1972): Improvement of livestock production in warm climate. W.H. Freeman and company.
Sanfransisco, U.S.A.

Morrow D.A., Roberts S.J. and Mc Entee K. (1985): Postpartum ovarian activity and involution of the uterus
and cervix in dairy cattle. I. Ovarian activity. The Cornell Veterinarian. 59, 173-190.

Mukassa-Mugerwa E., Tegegn A. Mattoin M., and Cecchini G. (1989): Effect of Estrus Synchronization with
Prostaglandin F2a in Ethiopian highland Zebu (Bos indicus) cows. Animal Production. 48, 367-373.

Negussie E. (1992): Reproductive performance of local and crossbred dairy cattle at Assela Livestock farm.
Arsi, Ethiopia. MSc Thesis, Alemaya University of Agriculture, Ethiopia.

Negussie E., Brannag E., Banjaw K., and Rottmann O.U. (1998): Reproductive performance of dairy cattle
at Assela Livestock farm, Arsi, Ethiopia. I: Indigenous cows versus their F1 crosses. J. Anim Breed
Genet.115, 267-280.

Roberts S.J. (1985), Veterinary Obstetrics and Genital Diseases. Theriogenology. 2nd ed. CBS Publishers &
Distributors, India. Pp. 604-755.

Salisbury G.W., Van Demark N.L., Lodge J.R.(1978): Physiology of reproduction and artificial insemination
of cattle. 2nd ed. San Francisco, W.H. Freeman. Pp 385-479.

Shiferaw Y., Tenhagen B.A., Bekana M. and Kasa T. (2003): Reproductive performance of crossbred dairy
cows in different production systems in the central highlands of Ethiopia. Trop Anim Health Prod. 25,
551-561.

Sinishaw W. (2005): Study on semen quality and field efficiency of AI bulls kept at the National Artificial
Insemination Center. MSc thesis,  Addis Ababa University, Faculty of Veterinary Medicine, Debre
Zeit.

Sorensen A.M. Jr. (1979): Animal Reproduction: Principles and Practices. Texas A & M university. Mc
Graw-Hill Publishing Company. Pp. 85-107; 292-330.

Sori H. (2004): Evaluation of Semen Parameters in Ethiopian Indigenous Bulls Kept at Kality, Artificial
Insemination Centre. Master’s Thesis. Addis Ababa University, Faculty of Veterinary Medicine, Debre
Zeit.

Swenson C., Schanr J., Branag E., and Meskel L.B. (1981): Reproductive performance of Zebu and crossbred
cattle.World Anim.Rev.38, 31-36.

Tegegn A., Entwist K.W., and Mugerewa Mukassa (1992a): Nutritional influence on growth and on set of
puberty in Boran- Friesian cross bulls. Theriogenology. 35, 1005-1016.

102 17th ESAP-Proceedings

 Animal Breeding

Tegegn A., Kassa T., and Mukassa-Mugerwa E. (1995): Aspects of bull production with emphasis on cattle
in Ethiopia. П. Sperm production capacity and semen characteristics. In: Proceeding of the Third Na-
tional Conference of Ethiopian Society of Animal production. Pp. 83-99.

Tegegn A., Warnick A.C., Mukassa-Mugerwa E., and Ketema H. (1989): Fertility of Bos indicus and Bos
indicus x Bos taurus crossbred cattle after estrus synchronization. Theriogenology. 31, 361-369.

Thrusfield M.(2005)Veterinary epidemiology. 3rd ed. Blackwell Science Ltd., Oxford, Great Britain. Pp. 182-
198.

Webb D.W. (2003): Artificial Insemination in Cattle. University of Florida, Gainesville. IFAS Extension,
DS 58. Pp. 1-4.

Yemane B., Chernet T., Shiferaw T. (1993): Improved Cattle Breeding. National  Artificial Insemination
Centre. Addis Ababa, Ethiopia. Pp. 15.

Zewde E. (2007): Artificial insemination and its implementation. Ethiopian Society of Animal Production
(ESAP). Addis Ababa, Ethiopia. Pp. 7-14, 29, 45.

Zewdie E., Deneke N., FikreMariam D., Chaka E., HaileMariam D., Mussa A.(2005): Guidelines and proce-
dures on bovine semen production. NAIC, Addis Ababa.

Zewdie E., Mussa A., Melese G.M., HaileMariam D., Perera B.M.A.O. (2006): Improving artificial insemina-
tion services for dairy cattle in Ethiopia. In: Improving the reproductive management of smallholder
dairy cattle and the effectiveness of artificial   insemination services in Africa using an integrated ap-
proach. International Atomic Energy Agency (IAEA). Pp. 17-19.

17th ESAP-Proceedings 103