Differences between Belclare and Suffolk ewes
in fertilization rate, embryo quality and
accessory sperm number after cervical
or laparoscopic artificial insemination
S. Faira, J.P. Hanrahanb, C.M. O’Mearaa, P. Duffya, D. Rizosa,
M. Wadea, A. Donovanb, M.P. Bolanda, P. Lonergana,
aDepartment of Animal Science, Centre for Integrative Biology, Conway Institute for Biomolecular
and Biomedical Research, University College Dublin, Belfield, Dublin 4, Ireland
bSheep Research Centre, Teagasc, Athenry, Co. Galway, Ireland
Received 15 July 2004
Ewe breed has been shown to have a major effect on pregnancy rates following cervical AI using
frozen–thawed semen. The main objective of this study was to examine the differences between
purebred Belclare and Suffolk ewes (multiparous) in fertilization rate, number of accessory sperm
and stage of embryo development on day 6 after cervical or laparoscopic AI with frozen–thawed
semen. In experiment 1, Belclare and Suffolk ewes were synchronized for 12 days and were either
cervicallyinseminated(year1:n = 28and31;year2:n = 16and15,respectively)orlaparoscopically
inseminated (year 2: n = 13 and 14). In experiment 2, superovulated Belclare (n = 4) and Suffolk
(n = 13)eweswerelaparoscopically inseminated.Alleweswereslaughtered6daysafterAI;oocytes/
embryos were recovered, morphologically graded and stained to assess the number of cells and
accessory spermatozoa. Data from both experiments were combined for statistical analysis. The
proportion of ewes with fertilized oocytes was significantly higher following laparoscopic AI
compared with cervical AI (54% versus 19%). More Belclare than Suffolk ewes yielded fertilized
oocyte(s)aftercervicalAI(34%versus10%,P < 0.02)buttherewasnodifferenceafterlaparoscopic
AI (62% versus 60%). From the ewes that yielded at least one fertilized oocyte the proportion of
Theriogenology 63 (2005) 1995–2005
* Corresponding author. Tel.: +353 1 716 7731; fax: +353 1 716 1103.
E-mail address: email@example.com (A.C.O. Evans).
0093-691X/$ – see front matter # 2004 Elsevier Inc. All rights reserved.
Belclare ewes with embryos at the morula/blastocyst stage was significantly greater than for Suffolk
ewes(94%versus59%,P < 0.02).AhigherproportionofBelclarethanSuffolkeweshadevidenceof
sperm reaching the site of fertilization following cervical AI (39% versus 15%, P < 0.02) but there
wasnodifference afterlaparoscopicAI(62%versus64%,P > 0.8).Amongsttheeweswithevidence
of sperm at the site of fertilization, laparoscopic AI resulted in a higher number of sperm per oocyte/
embryo or per ewe than cervical AI (P < 0.01). These results suggested that the difference in
pregnancy rate between Suffolk and Belclare ewes following cervical AI was due to: (i) sperm
traversing the cervix and uterus in a higher proportion of Belclare than Suffolk ewes, leading to a
higher incidence of fertilization and (ii) the lower developmental competence of fertilized oocytes
from Suffolk ewes.
# 2004 Elsevier Inc. All rights reserved.
Keywords: Sheep; Ewe breed; Cervical insemination; Fertilization rate; Embryo quality
The availability of an effective AI procedure utilizing frozen–thawed semen is the
essential key to the widespread use of AI in animal breeding programmes . The site of
deposition of frozen–thawed semen has a major effect on fertilization rate. Thus,
significantly higher fertility is generally obtained with laparoscopic AI using frozen–
thawed ram semen than following transcervical [2,3] or cervical insemination . Because
of the size and shape of the external os and the tortuous structure of the cervical canal ,
intrauterine insemination of sheep generally requires the use of laparoscopy. Transcervical
AI has been achieved using specially designed inseminating equipment [2,3] and
manipulation of the cervix using forceps . The number of reports in which transcervical
deposition of semen has been achieved is relatively low and there are concerns about the
potential trauma involved. Thus, while laparoscopic AI is a very effective technique, the
costs involved in terms of expertise, labour and equipment are high, limiting its use.
A better understanding of the reasons for the generally low conception rate achieved
following cervical deposition of frozen–thawed semen is still an important objective
towards the goal of establishing a cost-effective and widely applicable AI procedure for
achieving acceptable pregnancy rates in sheep. That pregnancy rates in excess of 55% are
routinely obtained in Norway, under field-scale conditions over a number of years,
demonstrate that acceptable pregnancy rates are achievable [7,8]. Donovan et al. 
recently evaluated the procedures used in Norway under Irish conditions. They reported
higher pregnancy rates using fresh compared to frozen–thawed semen but found no
differences in pregnancy rate following natural or synchronized estrus. The same authors
found no difference between rams of Irish and Norwegian origin; however, they reported a
significant effect of ewe breed on pregnancy rate. This is consistent with studies
demonstrating that pregnancy rate following cervical insemination of frozen–thawed
semen is highly dependent on ewe breed, with values of 8, 28, 44 and 77% reported for
Suffolk, Texel, Belclare and Finnish Landrace, respectively . Further studies
examining breed differences in timing of ovulation concluded that there was no effect of
breed on the interval from the preovulatory LH surge to ovulation; however, there was a
substantial difference between Finnish Landrace and other breeds in the variability in the
S. Fair et al./Theriogenology 63 (2005) 1995–20051996
interval from pessary removal to ovulation . Reported anatomical differences of the
cervix amongst these breeds are consistent with breed differences in mature body size, but
are not correlated with differences in pregnancy rate .
The reason for the variation in fertility among ewe breeds following cervical AI with
frozen–thawed semen may be due to differences in sperm transport through the cervix and
uterus or due to early embryo mortality. The aim of the present study was: (i) to examine
in pregnancy rate after cervical AI with frozen–thawed semen , (ii) to evaluate ewe
breed effects on the number of sperm competing for fertilization by examining the number
of accessory spermatozoa bound to the zona pellucida and (iii) to assess the stage of
embryo development on day 6 after cervical or laparoscopic AI.
2. Materials and methods
2.1. Experimental design
Two experiments were conducted to compare fertilization rate, number of accessory
spermatozoa and to assess early embryo development in purebred Belclare  and
purebred Suffolk ewes on day 6 after cervical or laparoscopic AI with frozen–thawed
semen. These breeds were chosen because they were known to differ significantly in
pregnancy rate following cervical AI with frozen–thawed semen . Experiment 1 was
carried out over two breeding seasons. In year 1, multiparous Belclare (n = 28) and
Suffolk(n = 31)eweswerecervicallyinseminated.Inyear2,eweswereeithercervically
(Belclare n = 16, Suffolk n = 15) or laparoscopically (Belclare n = 13, Suffolk n = 14)
inseminated. Experiment 2 was carried out in year 2 (using superovulation and
laparoscopic AI) to confirm the differences in embryo quality observed in experiment 1.
TheestrouscyclesofBelclare(n = 5)andSuffolk(n = 13)ewesweresynchronizedfor12
days, following which, ewes were superovulated as described below. Crossbred ewes
(n = 29)weresynchronizedasdescribedbelowandnaturallymatedpriortoexperiment1
to provide a baseline measurement for recovery rate, fertilization rate and embryo
2.2. Collection, processing and freezing of semen
The semen used in this study was collected using an artificial vagina and was processed
and frozen as described by Donovan et al. . The processed semen was loaded into
0.25 mL Minitub straws (Minitub GMBH, Tiefenbach Germany), 200 ? 106spermatozoa
per straw and frozen in liquid nitrogen vapour in a programmable freezer before it was
plunged into liquid nitrogen for storage.
2.3. Synchronization of estrus
The study was carried out over two breeding seasons (September–February). All ewes
were maintained on pasture with free access to water at Lyons Research Farm, University
S. Fair et al./Theriogenology 63 (2005) 1995–20051997
College Dublin (538180N) for the duration of the experiments. To control the time of
ovulation, an intravaginal progestagen pessary (30 mg flugestone acetate; Chronogest,
Intervet, Boxmeer, The Netherlands) was inserted for 12 days. In experiment 1, ewes were
given 500 IU eCG (Intervet, Boxmeer, The Netherlands) at pessary removal, while in
experiment 2 ewes were given 1250 IU of eCG 48 h prior to pessary removal in a single
2.4. Animals and insemination
Semen from two rams of proven fertility was used and within any year for a given
experiment all the semen came from one ram. Straws containing 200 ? 106spermatozoa
were thawed in a water bath at 70 8C for 8 s immediately prior to insemination. Ewes were
cervically inseminated 56–57 h post-pessary removal in a standing position. The cervix
was located, via a speculum with a light source, and the semen was deposited as far as
possible into the cervix without using force (usually into the first fold) by means of an
insemination pipette with a bent tip (Minitub GMBH, Tiefenbach, Germany). For
laparoscopic AI, ewes were deprived of food and water for 16 h prior to insemination. In
experiment 1, ewes were laparoscopically inseminated 60 h post-pessary removal while in
experiment 2, due to superovulation, ewes were laparoscopically inseminated 48 h post-
pessary removal as described by Evans and Maxwell . The inseminating pipette was
introduced into the lumen of each uterine horn and 0.125 mL of frozen–thawed semen
containing approximately 100 ? 106sperm was deposited in each horn. Fertile rams were
introduced to the crossbred ewes 36–48 h post-pessary removal at a ratio of 1:8 and
remained with the ewes for 3 days.
2.5. Assessments of reproductive performance
All ewes were slaughtered in a commercial abattoir 6 days after AI. Their reproductive
temperature of the tracts as close to body temperature as possible. The number of corpora
lutea (CL) on each ovary was recorded to determine ovulation rate.
Each uterine horn was then flushed from the base of the horn to the utero-tubal junction
using 20 mL phosphate-buffered saline (PBS) containing 3% fetal calf serum (v/v). The
flushings from each uterine horn were pooled in a 50 mL polypropylene conical tube.
Oocytes/embryos were located using a stereomicroscope. Oocyte recovery rate was
calculated on a per ewe basis as the number of oocytes divided by the number of CL.
Fertilization rate was also defined on a per ewe basis: 1 if the ewe yielded at least one
fertilized oocyte and 0 if only unfertilized oocytes were recovered (all 1-cell embryos were
deemed to be unfertilized).
Embryos were morphologically classified under the light microscope as being good or
those that had developed to the early morula stage or further, and poor quality embryos
were those that had not developed to the early morula stage. This classification was based
on the stage-for-age of the embryo  and the assumption that embryos that had not
developed to the early morula stage or further by day 6 would be unlikely to establish a
S. Fair et al./Theriogenology 63 (2005) 1995–20051998
pregnancy. The incidence of morula/blastocyst on a per ewe basis was defined as a
binomial variable for those ewes with at least one fertilized oocyte recovered (=1 if at
least one morula/blastocyst was represented among the recovered oocytes and zero
otherwise). For the estimation of the number of cells per embryo and of the number of
accessory sperm per oocyte, all oocytes/embryos were placed on a slide, air-dried, and
fixed in 100% ethanol overnight. They were stained using Hoechst 33342 (10 mg/mL in
2.3% sodium citrate) and a coverslip was applied. They were then visualized under an
epifluorescence microscope and the number of cells was counted, as was the number of
accessory sperm attached to the zona pellucida. After staining, all embryos classified as
good using the above criteria had ?32 cells while the majority of the poor embryos had
2.6. Data analysis
Data from experiments 1 and 2 were combined for statistical analysis. Because
crossbred ewes were only used to provide baseline measurements, they were not included
in the statistical analyses. The statistical models used for all analyses had an effect for each
of the eight combinations of experiment, year, breed and method of AI (see Table 1).
Appropriate linear contrasts between the resulting mean values were used to evaluate the
effects due to method of AI and ewe breed within method of AI as follows:
- AI method was tested using the means for year 2 of experiment 1;
Ovulation rate and oocyte recovery rate were analysed using the least-squares proc-
edures (Proc GLM of SAS, 1996) and was based only on ewes that ovulated and with both
ovaries recovered. Fertilization rate and embryo quality were analysed as binomial events
using a generalized linear model (Proc GENMOD of SAS; SAS, 1996)  with a logit
Because the sperm count data were clearly not normally distributed and did not
approximate to any distribution for which there was a defined transformation to yield
normality, the analysis of these data was done in two stages with the ewe being the
experimental unit in both. For the first stage, ewes were classified as 1 if at least one
fertilized oocyte was recovered or if a sperm was found on any of the oocytes recovered
from that ewe – ewes that failed both these criteria were classified as 0. Ewes that
failed to yield an oocyte at recovery were excluded as having no information. The
second stage involved only those ewes that had evidence of sperm at the site of
fertilization (classified = 1 at first stage). Each animal was classified within experiment
by year as to whether the value for sperm count was above the median for that experi-
ment by year (=1) or not (=0) and the results analysed as a binomial event. Percentage
values quoted in the results were all derived from back-transformation of the appropriate
estimates on the logit scale and thus will not be exactly equal to the numerical results
in Tables 1–3.
S. Fair et al./Theriogenology 63 (2005) 1995–2005 1999
S. Fair et al./Theriogenology 63 (2005) 1995–2005
Summary of ovulation rate, fertilization and embryo development in purebred Belclare and Suffolk ewes on day 6 after cervical or laparoscopic AI with frozen–thawed
Cervical AI (year 1) Cervical AI (year 2)Laparoscopic AI (year 2) Laparoscopic AI
superovulation (year 2)
BelclareSuffolk Belclare Suffolk BelclareSuffolk BelclareSuffolk
No. of ewes
Ovulation rate ? S.E.
Ewes yielding oocytes/embryos
Ewes yielding fertilized oocytes
Ewes with morula/blastocysts
No. of oocytes/embryos recovered
No. of oocytes fertilized
No. of morula/blastocysts
3.54 ? 0.3
2.34 ? 0.3
4.0 ? 0.4
3.77 ? 0.5
2.29 ? 0.4
7.0 ? 0.8
5.08 ? 0.5
2.47 ? 0.4
In year 1, one ovary from each of two Belclare and two Suffolk ewes was lost at
slaughter and therefore their data were excluded from the ovulation rate and recovery rate
analyses. Following laparoscopic AI in experiment 2, one Belclare ewe failed to ovulate
and was thus excluded from the analyses. Following natural mating only oocytes/embryos
from 20 of the 29 crossbred ewes were stained to provide data on fertilization rate, embryo
quality and sperm numbers.
ewes (3.8 versus 2.4, P < 0.001). Embryo recovery rate varied from 68 to 90% (Table 1)
among the treatment combinations and was not significantly affected by breed or method
With respect to fertilization rate, the interaction between breed and method was not
statistically significant. The proportion of ewes with fertilized oocytes was higher
following laparoscopic AI than after cervical AI (54% versus 19%, P < 0.02). Following
S. Fair et al./Theriogenology 63 (2005) 1995–2005 2001
Fertilization rate and proportion of ewes with fertilized oocytes that reached the morula/blastocyst stage
Experiment and year Artificial
(no. of ewes)
Ewes with fertilized
oocytes reaching the
Experiment 1, year 1Cervical
Experiment 1, year 2 Cervical
Experiment 2, year 2Laparoscopic
aObserved % of ewes with at least one fertilized oocyte.
bObserved % of ewes.
Classification of ewes based on sperm numbers per oocyte for ewes with evidence of sperm at the site of
Experiment and year Artificial insemination
(no. of ewes)
Total ewes with
evidence for sperm
Ewes with sperm
count > median (%)
Experiment 1, year 1Cervical
Experiment 1, year 2Cervical
Experiment 2, year 2Laparoscopic
natural mating, 67% of ewes yielded fertilized oocytes. A greater proportion of Belclare
than Suffolk ewes yielded fertilized oocyte(s) after cervical AI (34% versus 10%,
P < 0.02). There was no evidence for a difference between the breeds following
laparoscopic AI (Belclare 62%, Suffolk 60%; P > 0.9).
The model used for the analysis of the proportion of ewes with fertilized oocytes that
progressed to the morula/blastocyst stage only had effects for ewe breed and experiment-
by-year-by-method becausethe fullmodel specified inSection2could notbe usedbecause
for some combinations of breed with experiment-by-year-by-method all ewes had morula/
blastocysts (Table 2). However, the reduced model provided a good fit to the data as
evidenced by a scaled deviance of 0.9 (38 d.f.). The reduced model provided a good fit for
the data. Method of insemination had no effect on the proportion of ewes with fertilized
oocytes that progressed to the morula/blastocyst stage (91% versus 70%, for laparoscopic
versus cervical AI, respectively; P > 0.2). A higher proportion of Belclare than Suffolk
ewes yielded fertilized oocytes which reached the morula/blastocyst stage (94% versus
59%, P < 0.02). The corresponding value for naturally mated crossbred ewes was 78%
(14 of 18 ewes).
The interaction between breed and AI method for the proportion of ewes with evidence
of sperm at the site of fertilization was not significant. A greater proportion of ewes had
evidence of sperm at the site of fertilization following laparoscopic AI (54%) than after
cervical AI (28%) butthis differencewas not significant (P = 0.06). A higher proportion of
Belclare ewes had evidence of sperm reaching the site offertilization following cervical AI
than did Suffolk ewes (39% versus 15%, P < 0.02) but not after laparoscopic AI (62%
versus 64%). The corresponding value for crossbred ewes following natural mating was
60%. Of the ewes with evidence of sperm at the site of fertilization, the median values for
average number of sperm per oocyte were 0.65 and 2.00 for cervical AI in years 1 and 2 of
experiment 1, respectively, and 9.50 and 3.20 for laparoscopic AI in experiments 1 and 2,
respectively. The median number for naturally mated ewes was 36 (n = 12 ewes).
LaparoscopicAI resulted inahighernumberofspermperoocyte/embryo(ortotalperewe)
than cervical AI (P < 0.01). However, there was no effect of breed, irrespective of method
of insemination, on the number of sperm per oocyte/embryo or total number of sperm per
ewe (P > 0.3).
The higher proportion of Belclare than Suffolk ewes with evidence of sperm reaching
the site of fertilization following cervical AI with frozen–thawed semen is probably the
finding was that failure of fertilized oocytes to progress to the morula/blastocyst stage was
affected by ewe breed and, thus, was also a contributory factor to the lower pregnancy rate
in Suffolk ewes following cervical AI with frozen–thawed semen.
The accessory sperm associated with oocytes/embryos recovered on day 6 post-
insemination are considered to represent the number of sperm competing for fertilization
during the time the oocyte was receptive . Accessory sperm have been shown to
provide qualitative and quantitative insights into the characteristics of potential fertilizing
S. Fair et al./Theriogenology 63 (2005) 1995–20052002
sperm in cattle [18,19]. In agreement with other reports, laparoscopic AI resulted in better
fertilization rates than cervical AI [20,21] which may be a reflection of the significantly
cervical AI. The significantly lower fertilization rate following cervical AI must therefore
be at least partially due to impeded sperm transport through the genital tract .
The significantly higher proportion of Belclare than Suffolk ewes with evidence of
sperm at the site of fertilization following cervical AI probably explains the higher
fertilization rate in the Belclare ewes. Because there was no difference between the breeds
in the number of sperm per oocyte/embryo or total per ewe (for ewes with evidence of
sperm at the site of fertilization) the difference between Suffolk and Belclare ewes
following cervical AI was presumably due to sperm got past the cervical barrier in a
significantly higher proportion of Belclare ewes. This interpretation is consistent with the
finding that following laparoscopic AI there was no difference in fertilization rate between
with evidence of sperm at the site of fertilization or in the number of sperm per oocyte/
embryo or total per ewe. This would suggest that when sperm are deposited near the site of
fertilization, they are equally capable of fertilizing oocytes in Belclare and Suffolk ewes.
Another major finding of this study is that during the first 6 days after AI, the
developmental rate of Suffolk embryos was significantly lower than that of Belclare
embryos, as evidenced by the significantly lower proportion of fertilized oocytes from
partially explain the significant effects of ewe genotype on pregnancy rate following
cervical AI with frozen–thawed semen [10,12] and also the parallel differences reported
following natural mating . Most embryonic wastage occurs within the duration of an
estrous cycle and there is a negative linear relationship between the number of eggs shed
and rate of embryo loss . Embryo transfer experiments involving recipients from
different breeds or lines have generally failed to reveal significant differences in the
maternal contribution to embryo survival . However, there are exceptions to this; other
studies have provided evidence for such effects with a notably high embryo survival in the
Romanov breed .
In conclusion, the inability of sperm to get to the site of fertilization in sufficient
numbers to fertilize the oocyte(s) is a major constraint of cervical AI using frozen–thawed
semen. We have shown that the efficiency of sperm transport is poorer in a higher
proportion of Suffolk than Belclare ewes and fertility is further compromised by poorer
embryo quality. We speculate that the difference in sperm transport may be due to a
difference in the finely tuned endocrine control of sperm transport mechanisms , while
the lower developmental competence in Suffolk than Belclare embryos may be due to a
difference in the quality of the oocyte being ovulated or due to differences in the uterine
This work was funded by the Department of Agriculture and Food under the Research
Stimulus Fund (RSF 061).
S. Fair et al./Theriogenology 63 (2005) 1995–2005 2003
 Donovan A, Hanrahan JP, Lally T, Boland MP, Byrne GP, Duffy P, et al. AI for sheep using frozen–thawed
semen. Teagasc, Sheep Research Centre, Athenry, Sheep Series 11; 2001. ISBN 1 84170 152 1.
 Wulster-Radcliffe MC, Wang S, Lewis GS. Transcervical artificial insemination in sheep: effects of a new
transcervical artificial insemination instrument and traversing the cervix on pregnancy and lambing rates.
 Wulster-Radcliffe MC, Lewis GS. Development of a new transcervical artificial insemination method for
sheep: effects of a new transcervical artificial insemination catheter and traversing the cervix on semen
quality and fertility. Theriogenology 2002;58:1361–71.
 King M, McKelvey W, Dingwall W, Matthews K, Gebbie FE, Mylne M, et al. Lambing rates and litter sizes
following intrauterine or cervical insemination of frozen/thawed semen with or without oxytocin admin-
istration. Theriogenology 2004;62:1236–44.
 Halbert GW, Dobson H, Walton JS, Buckrell BC. The structure of the cervical canal of the ewe.
 Halbert GW, Dobson HJSW, Buckrell BC. A technique for transcervical intrauterine insemination of ewes.
 Olesen I. Effects of cervical insemination with frozen semen on fertility and litter size of Norwegian sheep.
Livestock Prod Sci 1993;37:169–84.
 Paulenz H, Soderquist L, Adnoy T, Nordstoga A, Gulbrandsen B, AndersenBerg K. Fertility results after
different thawing procedures for ram semen frozen in minitubes and mini straws. Theriogenology
 Donovan A, Hanrahan JP, Kummen E, Duffy P, Boland MP. Fertility in the ewe following cervical
insemination with fresh or frozen–thawed semen at a natural or synchronised oestrus. Anim Reprod Sci
 Donovan A, Hanrahan JP, Duffy P, Byrne GP, Boland MP. Cervical AI with fresh and frozen–thawed semen:
effect of ewe breed. Irish J Agric Food Res 1999;38:267.
 Donovan A, Hanrahan JP, Duffy P, Boland MP. AI in sheep: breed differences in timing of ovulation. Irish J
Agric Food Res 2000;39:3.
 Hanrahan JP. Aspects of reproduction performance in small ruminants – opportunities and challenges.
Reproduction Suppl 2003;61:15–26.
 Hanrahan JP. Reproductive performance of Belclare sheep: a new composite breed. In: Proceedings of the
40th European Association for Animal Production. 1989. 4 pp. (Paper S6).
 Evans G, Maxwell W. Salamon’s artificial insemination of sheep and goats. Sydney: Butterworths; 1987.
 Wintenberger-Torres S, Sevellec C. Atlas of the early development of the sheep embryo, vol. 51. Versailles:
INRA Publ.; 1987.
 SAS. SAS/STAT software: changesand enhancements through release 6. 11. Cary, USA:SASInstitute, Inc.;
 Saacke RG, DaltonJC,Nadir S, Nedel RL,Bame JH.Relationshipof seminaltraits andinseminationtime to
fertilisation rate and embryo quality. Anim Reprod Sci 2000;60–61:663–77.
 Saacke RG, DeJarnette JM, Bame J, Karabinus DS, Whitman SS. Can spermatozoa with abnormal heads
gain access to the ovum in artificially inseminated super- and single-ovulating cattle? Theriogenology
 DeJarnette JM, Saacke RG, Bame J, Vogler CJ. Accessory sperm: their importance to fertility and embryo
quality, and attempts to alter their numbers in artificially inseminated cattle. J Anim Sci 1992;70:484–91.
 Hill JR, Thompson JA, Perkins NR. Factors affecting pregnancy rates following laparoscopic insemination
of 28447 Merino ewes under commercial conditions: a survey. Theriogenology 1998;49:697–709.
 Salamon S, Maxwell WMC. Frozen storage of ram semen II Causes of low fertility after cervical
insemination and methods of improvement. Anim Reprod Sci 1995;38:1–36.
 Hanrahan JP. Pregnancy rate in Suffolk and Texel sheep to natural mating at either synchronized or natural
oestrus. Reprod Domestic Anim 2003;38:358.
 Hanrahan JP. Selection for increased ovulation rate, litter size and embryo survival. In: Proceedings of the
Second World Conference on Genetics Applied to Livestock Production; 1982. p. 294–309.
S. Fair et al./Theriogenology 63 (2005) 1995–2005 2004
 Ricordeau G, Razungles J, Lajous D. Heritability of ovulation rate and level of embryonic losses in the
Romanov breed. In: Proceedings of the Second World Conference on Genetics Applied to Livestock
Production; 1982. p. 591–5.
 Hawk HW. Sperm survival and transport in the female reproductive tract. J Dairy Sci 1982;66:2645–60.
S. Fair et al./Theriogenology 63 (2005) 1995–20052005