"Breeding and Its Technology - Now And The Future"

A.O. McKinnon
Goulburn Valley Equine Hospital

Frozen Semen


Although it was known as early as 1776 that cooling spermatozoa held them in a dormant state and there followed a huge interest in the physics of cold, cryopreservation of spermatozoa remained an enigma until the serendipitous discovery of glycerol in 1948: "In the autumn of 1948 my colleagues Dr Audrey Smith and Dr C Polge were attempting to repeat the results which Shaffner, Henderson and Card had obtained in the use of laevulose solutions to protect fowl spermatozoa against the effects of freezing and thawing. Small success attended the efforts and pending inspiration a number of solutions were put away in the cold store. Some months later work resumed with the same material and negative results were again obtained with all of the solutions except one, which almost completely preserved motility in fowl spermatozoa frozen to -790C. This very curious result suggested that chemical changes in the laevulose possibly caused or assisted by the flourishing growth of mould which had taken place during storage had produced a substance with surprising powers of protecting living cells against the effect of freezing and thawing. Tests however, showed that the mysterious solution not only contained no unusual sugars but in fact contained no sugar at all. Meanwhile further biological results had shown that not only was motility preserved after freezing and thawing but, also, to some extent, fertilizing power. At this point, with some trepidation, the small amount (10-15 ml) of the miraculous solution remaining was handed over to our colleague Dr D Elliott for chemical analysis. He reported that the solution contained glycerol, water and a fair amount of protein. It was then realised that Mayer’s albumen- the glycerol and albumen of the histologist- had been used in the course of morphological work on the spermatozoa at the same time as the laevulose solutions were being tested and with them had been put away in the cold store. Obviously there had been some confusion with the various solutions although we never found out exactly what had happened. Tests with new material very soon showed that the albumen played no part in the protective effect, and our low temperature work became concentrated on the effects of glycerol in protecting living cells against the effects of low temperature (Parkes, 1956). Since the discovery of glycerol protecting cells during freezing and thawing, progress has been dramatic in the area of multiple species gamete preservation, however, there is still room for considerable improvement in the efficiency of the techniques. For example the number of bull spermatozoa killed or rendered immotile during the freeze/thaw process has changed very little since 1955.


Many advantages of freezing are similar to cooling. In addition some others exist:

1) A stallions breeding may continue while he is performing or even while recovering from an injury or illness.
2) Semen can be easily transported internationally.
3) Semen from valuable stallions can be stored indefinitely or even in the hope that breed regulations change to allow them to use frozen semen.


1) The primary disadvantage is that in many stallions, techniques used for processing, packaging, freezing, thawing and insemination of their frozen semen do not result in satisfactory pregnancy rates. Currently pregnancy rates from frozen semen average approximately 50% of the same stallions with fresh semen. However some stallions freeze quite well and some do not. In the future selection for those stallions that freeze well may be expected to improve the number of stallions with semen suitable for freezing and thus the per cycle pregnancy rate.

2) Costs. Veterinary charges for freezing of spermatozoa are quite high. At the GVEH our current charges are somewhere between $20-30 per insemination dose. This depends on a number of factors such as sperm numbers and quality.

3) Pregnancy rates from lay freezers (mobile freezing units) are poor and have helped create industry misconceptions about the difficulty of working with frozen equine semen. Based on the results of frozen semen, many countries were quick to learn that semen frozen by most mobile units was inferior to those samples frozen in major bull studs. A disadvantage of freezing at centres is loss of control of the stallion to the handling of other people. We have even had people send a handler in to manage the stallion (exercise etc) while the semen is being frozen.

4) Breeder misconceptions.

A) Success rates are high. This tends to increase the number of dissatisfied clients (usually those that have not been adequately informed prior to beginning the freezing of semen).

B) That anybody can freeze semen. Many shipments of frozen semen are arriving in Australia and occasionally the semen has been frozen by inexperienced personnel. These people often have had only a small amount of experience in sterile technique, laboratory procedures etc. Some of the samples are good, however most are inferior. Once again this leads to client dissatisfaction. We even have one person in Australia who categorically states that when dealing with their group, the semen is always frozen well and the problem must be with the lack of suitable veterinarians at the other end. Unfortunately most people in this category are very good communicators and very poor technicians.

C) Errors in identity. Many breeders are concerned that their mare may be bred with the wrong semen. Obviously well marked straws and DNA fingerprinting will eliminate these misconceptions.


A cryoprotectant such as glycerol allows a substantial percentage of spermatozoa in a sample to survive the freeze/thaw process and retain fertilising capability, whereas most cells are rendered incapable of fertilising an egg if frozen in the absence of a cryoprotectant. There are two general classes of cryoprotectants: 1) Penetrating cryoprotectants. These pass through the sperm membrane and act both intracellularly and extracellularly, and 2) Non-penetrating cryoprotectants. These act only extracellularly. Glycerol is the most common cryprotectant although DMSO and propylene glycol have been used with some cells. Non-penetrating cryprotectants include proteins such as in milk or egg yolk; sugars such as fructose, lactose, mannose, raffinose or trehalose; synthetic polymers such as polyvinylpyrrolidone or methyl cellulose and amides. Most penetrating cryoprotectants serve as both a solvent and a solute. All non-penetrating cryoprotectants, including proteins lipids and sugars are solutes or colloids and can not serve as a solvent. Compounds placed into a solvent, such as water, which dissolve and form a true solution or ionise are termed solutes.

All solutes or colloids in a solution, either within a spermatozoan or in the extracellular media, contribute to the osmotic properties of the solution. The higher the concentration of solutes the greater the osmotic pressure. Glycerol serves as a solute within the water and also penetrates into the sperm. Either in the extender or inside the spermatozoan, glycerol contributes to the osmotic pressure of the extender or cell. A non-penetrating cryoprotectant such as lactose is a solute but because of the nature of the molecule it does not penetrate through the plasma membrane of a living cell. Thus lactose can contribute to the osmotic pressure of the seminal extender but not that of the sperm. Non-penetrating cryoprotectants move water out of the sperm which results in dehydration and shrinkage. Solvent cryoprotectants such as glycerol are beneficial because they function as a solvent with a freezing point much lower than that of water. As extended semen is frozen crystals of pure water freeze into small blocks of ice between which are "channels" of unfrozen extender, remaining solvent plus solute. The sperm reside in these unfrozen channels. Despite this it is thought that the primary beneficial effect of glycerol is extracellular. The movement of water out of the sperm is probably good because it is a direct result of water loss thus the possibility of intracellular ice damage is reduced.

The process of thawing and insemination is potentially the most damaging to sperm due to rapid changes in volume. For example, from an environment rich in glycerol, lactose and salts of the extended semen with a high osmotic pressure (>1000 mOsmol/kg) to an environment in which the seminal extender is diluted in the uterus (» 300 mOsmol/kg), results in a rapid increase in volume for the sperm as water moves in to equilibrate with the high concentration of intracellular glycerol (left behind). Rapid swelling of the sperm may rupture the plasma membrane. The solution to the problem of rapid volume increase is to serially dilute the glycerol in the media surrounding the sperm in a controlled manner prior to insemination (similar to frozen/thawed embryos). Fortunately this is not necessary for most stallions spermatozoa as it would add another complex step to what needs to be a relatively simple procedure (to gain universal equine practitioner acceptance). However at least part of the problem of reduced fertility with cryopreserved sperm is a consequence of the apparent necessity to use a cryoprotectant such as glycerol. There is abundant evidence that glycerol is toxic to sperm (Pace et al. 1975), even when sperm are not subjected to the additional rigours of freezing and thawing. The outward movement of water is greater when sperm are cooled below 00C at a slow cooling rate than at a very rapid rate. Ideally the cooling rate results in sufficient but not excessive outward movement of water and perhaps the formation of a few non-lethal ice crystals within each sperm, but optimal survival after re-warming. In the event that a cooling rate is slower or faster than ideal the damage can be mitigated by appropriate adjustment of the warming rate. As a generalisation, the more rapid the cooling (freezing) rate, the more rapid should be the warming (thawing) rate. The ideal cooling and warming rates are also influenced by the composition of the extender, including the concentration of glycerol. Typically straws of stallion sperm are warmed at 7000C/min by immersion in 370C water for ³ 30 seconds or 40000C/min by immersion in water at 750C for 7 seconds. The thawing procedure recommended by personnel who froze the spermatozoa should be followed closely.

Techniques for freezing equine semen

Methods listed below are those presently used by the GVEH and at the Equine Reproduction Laboratory at CSU.

Straws. Semen is typically frozen in either 0.5ml or 5ml straws. We have also cut 5ml straws in half to freeze in 2.5ml straws. This has the advantage of reduced storage and the straws fit well on the standard 0.5ml semen freezing rack. Progressively motile spermatozoa % appear to be better for most stallions frozen in 0.5ml straws, however it is either inconvenient to freeze multiple straws per insemination dose (5-10) or potentially excessively damaging to spermatozoa to be centrifuged at a rate capable of providing 1 X 109 in each 0.5ml straw. Each clinician needs to decide what effect they can detect when freezing in each container available. There are other containers available for freezing. Initially sperm was frozen in pellets, then glass ampoules, then aluminium tooth-paste-like tubes and even thin sealed packs. Experience with at least two forms of the above is desirable rather than the use of multiple different containers.

Solutions and extenders. The solution and extenders used at the GVEH have been described previously (Amann et al. 1984; Pickett et al. 1993). All solutions are prepared accurately using high quality deionised water and reagent grade chemicals. We prepare freezing extenders and centrifuge media in bulk and freeze it. The freezing extender is frozen in 5 or 20 ml aliquot’s and is not stored for longer than 3 months. Centrifuge media and extenders are thawed on the morning required, prior to collection of semen. Unused media is discarded not refrozen.




Glucose (g)



Na citrate (g)



Disodium EDTA (g)



Na bicarbonate (g)



Table 3. Composition of the Citrate-EDTA, and Glucose-EDTA media used for centifugation and extender for freezing equine semen (Adapted from (Cochran et al. 1984)).

Lactose solution (11% w:v)

50 (ml)

Glucose EDTA solution (Table 3)

25 (ml)

Egg Yolk

20 (ml)


5 (ml)

Equex STM

0.5 (ml)

Table 4. Composition of the Lactose-EDTA media used as an extender for freezing equine semen (Adapted from (Cochran et al. 1984)).

Centrifugation. The gel free portion of the ejaculate and is diluted to (approximately) 50 X 106 spermatozoa per ml with centrifuge medium prewarmed to 370C. After thorough mixing the suspension is transferred into 50 ml centrifuge tubes and centrifuged for 15 min at approximately 1500 RPM (400g). The exact conditions of centrifugation may have to be varied to produce a proper pellet and depends on the concentration of spermatozoa and even its motility. Low motility and concentration samples will form a firmer pellet in less time. The supernatant is discarded taking care not remove any sperm from the pellet and the semen is reconstituted in the appropriate volume of freezing extender (Table 4), which has been placed on the bench from the incubator (at 370C) at the beginning of centrifugation. At this time both the centrifuged sample and the extender have cooled to between 24-280C (providing centrifuge holders are prewarmed). Note: For the first few times a particular stallion is frozen the concentration of spermatozoa remaining in the centrifuge medium should be used to determine the number of spermatozoa lost through processing. The results can be used to determine if a longer or faster centrifugation is necessary (without further depression in post centrifugation motility). A 10% reduction in spermatozoal motility associated with centrifugation is acceptable. At the GVEH our aim is to have one billion (109) spermatozoa per insemination dose, regardless of the progressive motility. This is based on trial freezes with acceptance only of stallions with a post freeze progressive motility of >30%. At least one study (Volkmann et al. 1987) has demonstrated significantly better pregnancy rates by increasing the number of PMS from 175 x 106 to 249 X 106.

Packaging. Straws should be labelled with the stallions name, the processing center and the date of collection. 0.5 ml polyvinylchloride straws should be sealed with PVC powder or heat sealed at the opposite end to the cotton plug. 5 ml or 2.5 ml straws can be sealed with colored plastic beads or ball bearings. When loading semen, failure to have an air bubble within the fluid column, to allow for rapid expansion of ice crystals during thawing, may result in the straw exploding.

Cooling. Although not universally accepted, we find that for most stallions, slow cooling of the straws from room temperature to 50C at a rate of » -0.1/min improves the progressive motility of the frozen/thawed spermatozoa.

Freezing. The use of programmable freezers produces the most consistent freezing of straws, however straws have also been successfully frozen over a liquid nitrogen vapour. To freeze over a vapour a specific quantity of liquid nitrogen is placed in a styrofoam container, usually up to a designated line, relative to the height of the freezing rack. For 0.5ml straws, positioning the straws between 1-8 cm above the surface has been advocated. Straws remain at this position for 5-10 min and are then plunged into liquid nitrogen. Considerable experimentation with each individual stallion is necessary to provide the most optimum cooling rate. Five ml straws are placed 1-3 cm above the level of liquid nitrogen for 15-20 min and then plunged. When the straws are in the liquid nitrogen vapour they can cool to -1600C if left long enough. They are stored at -1960C in the liquid nitrogen. Mechanical freezers usually cool at a rate of 100C/min to -150C and then from 250C/min from -15 to -1200C. Straws are held at -1200C for 2-3 min and then plunged into liquid nitrogen.

Thawing. A variety of different regimes have been proposed.

1) For 0.5 ml straws (PVC) they can either be thawed by placing them in a 370C water bath for 30 seconds or >, or into 750C for exactly 7 secs then into 370C for at least 5 secs (Pickett et al. 1993).

2) 2.5 ml straws are recommended to be thawed for 5 min in a 370C waterbath.

3) 5 ml straws are placed in a 500C waterbath for 45 sec and then into 370C for at least 10 sec.

Post thaw evaluation. Post thaw quality of spermatozoa representing each ejaculate processed should be evaluated approximately 0-7 days after the freeze. One straw representing each freeze should be examined. The entire contents of the straw should be discharged into prewarmed extender at 370C. The extender semen ratio should be adjusted to give a final concentration for motility evaluation of 20 X 106. Motility estimates should be made at 0 and 30 min after thawing (semen maintained at 370C). Criteria for acceptability is commonly 35% PMS at thaw and ³ 40% PMS at 30 min. A minimum accepted dose is 200 X 106 motile spermatozoa. The decision as to whether to accept semen of lesser quality will depend on factors such as stallion age and value of the offspring etc. Utilising the above techniques above our aim is always to provide at least 300 X 106 PMS per insemination dose.

Facilities available for the use of artificial insemination of horses.

As frozen/thawed semen has a short half-life, insemination must take place close to ovulation. This necessitates frequent examination of the mare per rectum as often as may be required to accurately predict ovulation. It is considered that manual palpation alone will not provide sufficiently reliable evidence on when to predict ovulation and thus facilities should include the capabilities of ultrasonographic examination of follicular development.

Facilities should be available to breed the mare under cover with adequate restraint to avoid damage to the mare or personnel involved in with the procedure. Restraining devices such as stocks should be immediately adjacent to the semen handling area or laboratory and should be constructed to avoid damage to the semen by temperature fluctuation or contamination before insemination.

Semen Handling: Stallion spermatozoa are fragile, easily damaged and short lived. They are susceptible to cold-shock, direct sunlight, many antibiotics and low levels of residues that may be left on the liners of artificial vaginas, glassware and other receptacles with which the semen may come in contact. Care should be taken to assure that all contact surfaces are made of non-spermicidal material, are thoroughly cleaned and have been rinsed with de-ionised water.

Before insemination, all handling and preparation of semen should take place in a clean room close to the insemination area. This room should be heated to maintain a minimum temperature of at least 240C and the following equipment must be available:

a) a microscope (preferably with a heated stage) for evaluating viability of spermatozoa (progressively motile sperm) at 200 x magnification and preferably with phase contrast,

b) an incubator maintaining temperature at 370C , and

c) a water bath for thawing spermatozoa which is capable of accurately maintaining the temperature specified for thawing different batches of semen.

Mare Insemination: The mare should be cleaned and prepared following normal procedures for artificial insemination. Prior to thawing semen, all equipment should be prepared. Care should be taken to avoid spermicidal preservatives in lubricants by using KY gel. If the thawed semen needs to be removed from its receptacle (ie. glass ampoule or 5 ml straws) prior to insemination, then care should be taken to avoid using syringes with rubber plungers (the lubricant in these syringes has been demonstrated to be spermicidal).

Management of mares inseminated artificially with frozen semen.

Try to ensure that only mares of normal fertility are bred with frozen semen.

Pre-breeding Evaluation: The requirements for pre-breeding gynaecological examination will vary according to the veterinarian and status of the mare. Most mares should be cultured and demonstrated negative for venereal pathogens (Klebsiella pneumoniae, Pseudomonas aeruginosa and Taylorella equigenitalis). Ultrasonographic examination should be used to determine that the mare has no signs of uterine inflammation (fluid). Other examination procedures such as uterine cytology, biopsy, endoscopic exam, digital manipulation of the cervix, etc. may be necessary depending upon the history of the mare to be bred, her age and recent foaling complications etc.

Timing of insemination with frozen semen: Although the frozen thawed spermatozoa from some stallions may survive 24 hours or more after insemination, best results are achieved if the semen is inseminated in a period from 12 hours before, until a maximum of 6 hours after ovulation. Insemination greater than 6 hours after ovulation results in a decreased chance of conception and an increased incidence of early embryonic death. Because each breeding unit of frozen semen is expensive to produce, it is recommended mares are bred a maximum of 2 times per cycle, and preferably once. There are various ways that a mare can be managed during the insemination period to optimise the time of insemination and vary according to economic considerations, facilities and veterinary experience:

A) Ultrasonographic examination every 6 hours. Mares are inseminated immediately ovulation has been detected.

B) 12-24 hourly examination. Ultrasonography is used to determine when the follicle(s) is immediately pre-ovulatory. The mare is inseminated immediately and re-checked in 12-24 hours. If ovulation has not occurred 24 hours later, the mare is re-inseminated at each inspection until ovulation has occurred. This will result in the use of more semen.

C) Ovulation induction. When ultrasonographic evaluation of a mare in early oestrus reveals the presence of 30-35 mm follicle and she is treated with 3000 IU of hCG (i.v.) then most mares will ovulated in 36 + 4 hours. Under these circumstances, the mare is examined at hCG + 24 hours. The mare may be inseminated at this time and re-inseminated 24 hours later if she has not ovulated (hCG + 48 hours). An alternative method is to inseminate all mares once at hCG + 36 hours. The advantages of using hCG are that it limits the number of examinations, while still allowing flexibility in relation to time of insemination. If pregnancy is not established, then subsequent cycles may be best managed with the use of another ovulation-inducing agent (GnRH -McKinnon et al. 1992) in those mares where hCG antibodies may result in a varied or less predictable response.

Artificial insemination: The mare should be restrained and then prepared for breeding as hygienically as possible and clean water should remove all traces of spermicidal antiseptic soaps, etc. After the frozen semen has been thawed according to the suppliers recommendations, it should be inseminated immediately. When the breeding dose is contained in one 0.5 ml straw, it will be difficult to evaluate the motility of the spermatozoa from samples bred. In these cases, recommendations are to sacrifice 1-2 straws in every 10 processed to assess the motility from each ejaculate. On other occasions, when semen needs to be placed inside a syringe prior to insemination, it is possible to check the motility after breeding from residues left within the A.I. pipette or syringe. Accurate evaluation will only be possible after the semen has been warmed to 37oC for 5 minutes (use a warmed slide and cover slip). If possible, insemination should be performed as far up the uterine horn ipsilateral to the ovulatory follicle as possible.

Post-artificial insemination examination: Mares should routinely be examined 24 hours after breeding to ensure ovulation has occurred and that there is no evidence of uterine fluid which may necessitate intra-uterine or systemic anti-bacterial treatment. Post-breeding surgery such as Caslick may be performed at this time. Ultrasonography should be used 14-15 days after breeding to determine if the mare has conceived, has twins, etc.

Pregnancy rates.

The only accurate indicator of fertility is pregnancy rate per cycle. Pregnancy rate at the end of the season is influenced by the number of cycles that mares are bred, the fertility per cycle and in many cases good or poor management practices. Table 5 shows the theoretical pregnancy rate for matings with a theoretical probability of between 10% and 70% pregnancy rate per cycle. Examination of table 5 shows that a stallion at 20% pregnancy rate/cycle can have 80% of his mares pregnant after 7 cycles which would not be different from a stallion with an expected outcome of 40% per cycle after three cycles or two cycles at 50% pregnant per cycle. Similarly 88% of mares would be pregnant after three cycles at 50% per cycle, 4 cycles at 40% per cycle, 6 cycles at 30% per cycle and 9 cycles at 20% per cycle.

Pregnancy rate per cycle








Cycle 1
























































Table 5. The effect of multiple cycles on cumulative pregnancy rate for matings of various theoretical outcomes (pregnancy rate per cycle).

Many studies and many semen centres report the overall pregnancy rate only and from the above discussion it should be clear that the pregnancy rate per cycle is the important figure to ascertain to recognise the true fertility of the sample. This becomes even more important when we are advising clients on the purchase of a set number of straws. If the fertility is only 30% per cycle and the number of straws retailed is 3 then the probabilityof having a mare pregnant in this scenario is only 66% after three attempts (cycles).

Comparisons between pregnancy rates of reported different breeding trials are generally invalid because of limited number of mares per stallion. Refer to table 2 to determine that if the true fertility is 35% pregnancy per cycle and only twenty mare cycles were evaluated, then the actual pregnancy rate may be anywhere from 14%-56% and still be within the 95% confidence limits. Even breeding 100 mare cycles only reduces the 95% confidence interval to between 25 and 45 mares pregnant. In addition, many studies have failed to provide data from control inseminations with fresh semen from the same stallion at the same time of year; frequently there are differences in the number of spermatozoa per insemination and the timing of insemination relative to ovulation (Pickett et al. 1993). Some examples of these studies are listed below.

Nishikawa (1975) presented data on pregnancy (conception rates) averaging 56.3% from studies between 1966-73. These were overall pregnancy rates from frozen semen. As an interesting aside, in the process of developing a suitable extender to freeze with, he looked at the fertility of sperm that had been cooled to 40C and stored for 8 hours and demonstrated an average pregnancy rate of 67.3%. They used egg yolk as an extender and this may be the first report of commercial cooling and transporting of equine semen.

Pace and Sullivan (1975) examined the effect of time of insemination, number of spermatozoa and extender components on the pregnancy rate of mares bred with frozen semen and found that fertilization rate was highest when mares were inseminated within 12 hours of ovulation, foaling rate was improved by increasing the number of motile spermatozoa per insemination (80 X106) and that pregnancy rate was improved by reducing glycerol from 7% to 2% and delaying freezing for two hours after extension. They also demonstrated the toxicity of glycerol by reporting a pregnancy rate per cycle for mares inseminated with unextended semen of 75%, extended semen without glycerol of 50% and extended semen with 7% glycerol of 35%.

The most notable early work with frozen equine semen performed in Europe was by Merkt and Klug (Merkt et al. 1975; Klug et al. 1975). Merkt et al reported on the results of the long term storage of semen frozen by pellets and later a pregnancy rate per cycle of 22% for a total number of mares pregnant of 50% at the end of the season. Martin et al. (1979) compared dilution and pellet freeze to centrifugation with and without extender. They were first to freeze in large volume (4ml) straws and obtained a pregnancy rate per cycle of 63%. Tischner (1979) using aluminium packages and freezing a non centrifuged sperm rich fraction reported a pregnancy rate per cycle of 32% for frozen compared to 52% for fresh, however the numbers were low (4/44 and 11/17, respectively).

Researchers in Colorado (Cochran et al. 1983) reported pregnancy rates from three stallions of 17%, 33% and 35% per cycle compared to 60%, 62% and 64% per cycle for fresh semen from the same stallions respectively. In addition they (Cristanelli et al. 1985) reported that the best extender based on % PMS after thawing was non clarified, lactose-EDTA-egg yolk extender containing 20% egg yolk and 4% glycerol. However we should always be wary of studies that use as an end point PMS rather than pregnancy rate, as progressive motility only measures a sperms ability to swim not whether it has survived the damages of freeze/thaw well enough to still be capable of fertilisation (Amann, 1989; Amann et al. 1987). Also in Colorado in 1983 procedures had improved such that from 36 mare cycles frozen semen resulted in 50%, 56% and 51% pregnancy rate per cycle, to give frozen as % of fresh of 75%, 84% and 100% respectively (mean 86%) (Pickett et al. 1993; Amann et al. 1987). However, as a good demonstration of how individual stallions can effect results, the next year that using the same procedures and different stallions the results were 10%, 30% and 61% for frozen as a % of fresh (mean 40%) and the individual pregnancy rates of frozen semen were 8%, 21% and 48% per cycle (mean 30% per cycle). Further experiments were performed to ascertain if minor variations had occurred to explain the differences. It was finally concluded that it was individual stallion differences.

Muller reported an overall pregnancy rate of 64% in 1982 and in 1987, he reported a pregnancy rate of 56%. More importantly he reported on differences between the fertility of individual stallions (29% -62%) and pregnancy rates from 0-79% associated with different technicians. In France (Rousset et al. 1987) it was demonstrated that for semen evaluation the highest repeatability was % of motile spermatozoa in both raw and extended semen prior to or after storage and that in relation to season and frequency of ejaculation on suitability of stallion semen for freezing it was demonstrated (Magistrini et al. 1987) that winter was probably the best time to freeze semen due to lower volumes of gel free semen and a high sperm concentration. However, it was noted that sexual behaviour was not as good and that optimisation of frequency of ejaculation and season did not change a "bad freezer" into a good one.

In South Africa (Volkmann et al. 1987) it was demonstrated using pony stallions that an overall pregnancy rate of 55% per cycle was possible. Pregnancy rates were significantly improved from 44% to 73% per cycle, when the number of progressively motile spermatozoa per insemination were increased from 175 X 106 to 249 X 106. They recommended that serving with a post thaw progressive motility of > 30% was appropriate and that insemination be performed every second day with a minimum of 220 X 106, once preovulatory follicle was identified.

In Texas (Kloppe et al. 1988) a study was designed to evaluate the effect of insemination timing on the fertilisation capacity of frozen/thawed semen. Mares were inseminated with either fresh semen every other day, frozen/thawed semen every other day or frozen/thawed semen within 6 hr after ovulation for a pregnancy rate of 70%, 60% and 55% per cycle respectively (N=20/group, P> 0.05). This coupled with the work from Wisconsin (Woods et al. 1990a; Pace et al. 1975), suggest that insemination should be performed as close as possible to ovulation and that not after 6 hr. In France, Arnaud Evain of Equi-technic quoted by (Pickett et al. 1993), showed a pregnancy rate per cycle for those stallions breeding more than 15 mares with frozen semen of between 26% and 56% and an end of season pregnancy rate of between 22% and 89%.

In Canada (Samper et al. 1986), 177 mares were bred from 32 frozen ejaculates from 9 stallions (mean number of mares per stallions was 19) for a first cycle pregnancy rate of between 32-70% (mean 50.5%) and a final number of pregnant mares of 60-90% (mean 72.4%). All mares were inseminated with 1 X 109 spermatozoa regardless of the quality. Recently at Colorado (Jasko et al. 1992c) embryo recovery rates were reported for thermoelectrically cooled semen of 74% (31/42) and 56% (23/41) for frozen thawed semen. Individual stallion pregnancy rates for frozen semen (3 stallions) were 47%, 57% and 64% per cycle for a percentage of frozen compared to cooled of 75%, 67% and 82% with an overall result of 74%.

Although more mares have been bred with frozen semen in China and Russia than the rest of the world combined, there is little solid data apart from total pregnancy rates or personal communications (Pickett et al. 1993). Results for both countries are around 65% of the mares becoming pregnant. For example Piao and Wang (1988) reported on the breeding of 59,111 mares and jennies between 1982-1987 for an average conception rate of 66.3% (total number pregnant) and a per cycle pregnancy rate of between 45-60% (average 53.1%).

Recently in Australia, Dowsett et al. (1995) reported on the use of non centrifuged semen frozen in glycerol/UHT skim milk/egg yolk/fructose extender after being cooled, transported and held in extender around three hours. They used a minimum of 300 X 106 live, normal spermatozoa per insemination from each of four stallions. Seventeen mares were bred for a total of 44 cycles for a pregnancy rate per cycle of 31.8% and an overall pregnancy rate of 82% (14/17). Although the pregnancy rate per cycle was quite low it did demonstrate that semen from some stallions, in some circumstances, may be frozen without centrifugation.

From the above it should be obvious that frozen semen can routinely achieve around 40% pregnancy rate per cycle and in some stallions it will be much better. Occasional stallions will not have their semen remain fertile after the process, no matter what the selection criteria are and what techniques are used. Percent progressive motility post thaw is not an accurate estimator of fertility and the only acceptable criteria is the pregnancy rate per cycle. Probably at least 50 cycles should be evaluated before trying to compare treatments or stallions.

Early Embryonic Death

Initialy there was concern that an increase in early embryonic death had been reported. More recently a combined report (Pickett et al. 1987) demonstrated that although EED was higher in mares bred from frozen semen, no significant differences existed (21% versus 14%, P>0.1).

Experiences with frozen semen

Semen has been frozen from hundreds of stallions now at the GVEH. In all cases tested, pregnancies have been achieved. The results have varied from good (85% per cycle) to poor (10% per cycle), depending on the stallion. There has been increasing demand for freezing and breeding with Australian stallions (locally and internationally). This has resulted in the establishment of numerous facilities for the purposes of attracting clients. There are huge differences in the abilities of the personnel and of course equipment between centres and clients should be aware that a good sales person does not always mean a good result. The GVEH is currently an approved centre for freezing semen for international export. We have successfully preserved semen for export (based on test breeding with our research mares) semen from ponies Standardbreds and Warmbloods.. The costs associated with the procedure and the careful counseling of clients about expected success rates has kept many people away. Particularly those that just want to store some semen before they castrate their stallion or those who worry about their stallion dying.

Experiences with semen imported into Australia have also been variable. Fertility from frozen semen will vary with freezing technique, mare fertility, inherent stallion fertility and management conditions. For example at the GVEH in 1993 semen was imported from a variety of countries. Results below are reported exactly as they occurred (Table 6). They were selected from a number of breeding stallions that we were involved with (frozen semen only) to demonstrate the point of variable stallion fertility. Clearly people who bred to the Paint or Pony stallions were very happy, whilst the Percheron breeder was not. The Percheron semen was frozen by an unidentified technician and the other semen was frozen by experienced personnel. The excellent fertility of the Paint stallion most likely reflects his inherent fertility. All mares were treated identically at the GVEH in relation to ovulation induction and post insemination treatments etc and all the mares were of a similar fertility cross-section. So, the results most likely reflect differences in fertility of the samples of spermatozoa rather than mare or management related differences. Those differences may be freezing technique or stallion related.



Number of mares bred

1st cycle fertility

Overall fertility

Cycles per pregnancy




7/9 (78%)

9/9 (100%)





4/8 (50%)

8/8 (100%)





1/10 (10%)

2/10 (20%)


Table 6. Results from frozen semen at the GVEH in 1993.


Amann, R.P. and B.W. Pickett 1984 An overview of frozen equine semen: Procedures for thawing and insemination of frozen equine spermatozoa. Special Series 33:1-30.(Abstract)

Amann, R.P. and B.W. Pickett 1987 Principals of cryopreservation and a review of cryopreservation of stallion spermatozoa. J. Equine. Vet. Sci. 7:145-173.

Amann, R.P. 1989 Can the fertility potential of a seminal sample be predicted accurately ? J. Androl. 10:89-98.

Bowen, J.M. 1969 Artificial insemination in the horse. Equine. Vet. J. 1:98-110.

Cochran, J.D., R.P. Amann, E.L. Squires, and B.W. Pickett 1983 Fertility of frozen-thawed stallion semen extended in lactose-EDTA-egg yolk extender and packaged in 1.0-ml straws. Theriogenology, 20:735-741.

Cochran, J.D., R.P. Amann, D.P. Froman, and B.W. Pickett 1984 Effects of centrifugation, glycerol level, cooling to 5 C, freezing rate and thawing rate on the post-thaw motility of equine sperm. Theriogenology, 22:25-38.

Cristanelli, M.J., R.P. Amann, E.L. Squires, and B.W. Pickett 1985 Effects of egg yolk and glycerol levels in lactose-EDTA-egg yolk extender and of freezing rate on the motility of frozen-thawed stallion spermatozoa. Theriogenology, 24:681-686.

Douglas-Hamilton, D.H., R. Osol, G. Osol, D. Driscoll, and H. Noble 1984 A field study of the fertility of transported equine semen. Theriogenology, 22:291-304.

Dowsett, K.F., B.L. Dunn, L.M. Knott, and J.W. Cooper 1995 Observations on the use of deep frozen stallion semen. Aust. Eq . Vet. 13:53-60.

Francl, A.T., R.P. Amann, E.L. Squires, and B.W. Pickett 1987 Motility and fertility of equine spermatozoa in a milk extender after 12 or 24 hours at 20C. Theriogenology, 27:517-526.

Jasko, D.J., J.A. Hathaway, V.L. Schaltenbrand, W.D. Simper, and E.L. Squires 1992a Effect of seminal plasma and egg yolk on motion characteristics of cooled stallion spermatozoa. Theriogenology, 37:1241-1252.

Jasko, D.J., J.M. Martin, and E.L. Squires 1992b Effect of insemination volume and concentration of spermatozoa on embryo recovery in mares. Theriogenology, 37:1233-1239.

Jasko, D.J., D.M. Moran, M.E. Farlin, E.L. Squires, R.P. Amann, and B.W. Pickett 1992c Pregnancy rates utilizing fresh, cooled, and frozen-thawed stallion semen. Proc. AAEP, 649-660.

Kenney, R.M., R.V. Bergman, W.L. Cooper, and G.W. Morse 1975 Minimal contamination techniques for breeding mares:Technique and preliminary findings. Proc. AAEP, 327-335.

Kloppe, L.H., D.D. Varner, R.G. Elmore, K.N. Bretzlaff, and J.W. Shull 1988 Effect of insemination timing on the fertilizing capacity of frozen/thawed equine spermatozoa. Theriogenology, 29:429-440.

Klug, E., H. Treu, H. Hillman, and H. Heinze 1975 Results of insemination of mares with fresh and frozen stallion semen. J. Reprod. Fertil. Suppl. 23:107-110.

Magistrini, M., P. Chanteloube, and E. Palmer 1987 Influence of season and frequency of ejaculation on production of stallion semen for freezing. J. Reprod. Fertil. Suppl. 35:127-133.

Martin, J.C., E. Klug, and A.R. Gunzel 1979 Centrifugation of stallion semen and its storage in large volume straws. J. Reprod. Fertil. Suppl. 27:47-51.

McKinnon, A.O., R.W. Brown, R.L. Pashen, P.E. Greenwood, and J.R. Vasey 1992 Increased ovulation rates in mares after immunisation against recombinantbovine inhibin alpha-subunit. Equine. Vet. J. 24:144-146.

McKinnon, A.O., A.M. Nobelius, S.T.D. Figueroa, J. Skidmore, J.R. Vasey, and T.E. Trigg 1993 Predictable Ovulation in Mares Treated with an Implant of the GnRH Analog Deslorelin. Equine. Vet. J. 25:321-323.

Merkt, H., E. Klug, D. Krause, and H. Bader 1975 Results of long-term storage of stallion semen frozen by the pellet method. J. Reprod. Fertil. Suppl. 23:105-106.

Muller, Z. 1982 Fertility of frozen equine semen. J. Reprod. Fertil. Suppl. 32:47-51.

Muller, Z. 1987 Practicalities of insemination of mares with deep-frozen semen. J. Reprod. Fertil. Suppl. 35:121-125.

Nishikawa, Y. 1975 Studies on the preservation of raw and frozen horse semen. J. Reprod. Fertil. Suppl. 23:99-104.

Pace, M.M. and J.J. Sullivan 1975 Effect of timing of insemination, numbers of spermatozoa and extender components on the pregnancy rate in mares inseminated with frozen stallion semen. J. Reprod. Fertil. Suppl. 23:115-121.

Parkes, A.S. 1956 Preservation of living cells and tissues at low temperatures. Proc. Int. Cong. Anim. Reprd. AI. 69-75.

Piao, S. and Y. Wang 1988 A study on the technique of freezing concentrated semen of horses (donkeys) and the effect of insemination. Proc. Int. Cong. Anim. Reprd. AI. 3:286a-286c.

Pickett, B.W., E.L. Squires, and A.O. McKinnon 1987 Procedures for collection, evaluation and utilization of stallion semen for artifical insemination. 2:1-125.(Abstract)

Pickett, B.W. and R.P. Amann 1993 Cryopreservation of semen. In: Equine Reproduction. A.O. McKinnon and J.L. Voss, eds. Lea & Febiger, Philadelphia, pp. 769-789.

Rousset, H., P. Chanteloube, M. Magistrini, and E. Palmer 1987 Assessment of fertility and semen evaluations of stallions. J. Reprod. Fertil. Suppl. 35:25-31.

Samper, J. C. Hellander, J.C. and B.G. Crabo 1991 Relationship between the fertility of fresh and frozen stallion semen and semen quality measured as sperm motility and with glass wool/shephadex filters. J. Reprod. Fertil. Suppl. 44:107-114.

Squires, E.L., C.K. Barnes, H.S. Rowley, A.O. McKinnon, B.W. Pickett, and R.K. Shideler 1989 Effect of uterine fluid and volume of extender on fertility. Proc. AAEP, 25-30.

Tischner, M. 1979 Evaluation of deep-frozen semen in stallions. J. Reprod. Fertil. Suppl. 27:53-59.

Varner, D.D., T.L. Blanchard, C.L. Love, M.C. Garcia, and R.M. Kenney 1988 Effects of cooling rate and storage temperature on equine spermatozoal motility parameters. Theriogenology, 29:1043-1054.

Volkmann, D.H. and D. van Zyl 1987 Fertility of stallion semen frozen in 0.5-ml straws. J. Reprod. Fertil. Suppl. 35:143-148.

Voss, J.L. and B.W. Pickett 1976 Reproductive management of the broodmare. 1-29.(Abstract)

Voss, J.L. 1993 Human chorionic gonadotropin. In: Equine Reproduction. A.O. McKinnon and J.L. Voss, eds. Lea & Febiger, Phildelphia, London, pp. 325-328.

Woods, J., D.R. Bergfelt, and O.J. Ginther 1990a Effects of time of insemination relative to ovulation on pregnancy rate and embryonic-loss rate in mares. Equine Veterinary Journal, 22:410-415.

Return to Angus McKinnon Title Page


back to top