By B. W. Pickett, Ph.D.
Professor Emeritus, Colorado State University


To quote Dr. R.H. Foote, one of the most noted reproductive physiologists of our era,  "Artificial insemination (AI) was the first great biotechnology applied to improve reproduction and genetics of farm animals."  During the past 200 plus years tremendous strides have been made in AI of numerous species. But AI has been particularly slow in horses due to the fact that until very recently breed registries have been opposed to its use. Therefore, techniques, procedures, and particularly extenders for stallion semen have been copied from those used in other species. Workers at Cornell University introduced the word EXTENDER because the medium they were using to breed cows "enhanced and extended the usefulness of semen." Therefore, the primary reasons for using an extender with stallion semen are as follows:

Contributions and Attributes of an Extender:

1.)To increase the volume of the ejaculate. The progress that has been made in AI would not have been possible without the use of extenders to increase the volume of the ejaculate so that numerous females could be bred with one sample of semen. Extenders also reduce the loss of spermatozoa in the equipment and the build-up of toxic by-products of metabolism.

2.)To permit effective antibiotic treatment of semen containing pathogenic or potentially pathogenic organisms, which reduces the spread of venereal diseases and uterine infections. For maximum effect, antibiotics must be added to the extender before it is added to the semen.

3.)To prolong the survival of spermatozoa. Semen can be shipped nationally and internationally.  Thus, a significant percentage of mares are bred with shipped (transported) semen in comparison to only a few years ago.  This eliminates the cost and stress of shipping a mare and/or foal to a breeding farm.

4.)To protect spermatozoa from unfavorable environmental conditions. Obviously, a sample of semen will warm or cool more slowly in a larger volume than a smaller volume.

5.)To aid in proper evaluation of sperm motility. An extender will disperse the sperm to permit the observation of the movement of individual cells. Almost universally, sperm quality is determined by the percentage of progressively moving spermatozoa.  The accuracy of this determination is dependent upon: a) clarity of the extender and, b) quality of the microscope used to view the spermatozoa.

6.)To increase pregnancy rate. Higher pregnancy rates per cycle are obtained when semen is extended immediately after collection and before insemination. Spermatozoa will also remain fertile longer in an extender than seminal plasma.

Considering the six reasons previously listed, it is obvious that selection of the appropriate extender is one of the most important decisions for a breeding program.

In general, the extender you have selected must contain the following materials or attributes:

1.)Materials to protect spermatozoa against changes in temperature, especially cold. Stallion spermatozoa possess little or no inherent resistance to sudden changes in temperature much below room temperature.  The addition of egg yolk, milk and/or the proteins, lipids, micro- and macromolecules contained in egg yolk and/or milk, permit stallion spermatozoa to survive adverse conditions, particularly when stallion semen is to be cooled for shipment.

2.)An osmotic pressure compatible with spermatozoa. The optimal osmotic pressure depends upon the combination of ingredients in the extender and the ratio of electrolytes to nonelectrolytes.

3.)Proper balance of mineral elements. Unfortunately, all of these ingredients have not been identified.

4.)Proper combination of nutrients. Since the motility of spermatozoa requires energy, it has been assumed that the ideal extender must contain ingredient(s) to provide this energy.  However, it is possible that the necessary material is contained in the spermatozoa, or seminal plasma. Although under certain circumstances, ingredients in an extender may be essential.

5.)Chemicals for neutralizing the toxic products produced by spermatozoa. It would appear that buffering agents are essential for extenders to prolong sperm motility. If so, milk and/or milk products contain these chemicals.

6.)Materials to stabilize enzyme systems and provide integrity of membranes. A properly buffered extender, containing appropriate macromolecules, will stabilize membranes and prevent their leakage from the sperm.

7.)A carrier free of infectious organisms and any other materials detrimental to motility. This means that the ingredients for formulation of the extenders must come from a reliable source or sources, and they must be stored properly. When compounding the extender, it must be done in a "clean" environment. Following the mixing of the ingredients, they must be packaged and stored to prevent contamination and deterioration. In selecting an extender get all the information that you can about source and quality of materials, handling and storage procedures as well as quality control of each individual batch.

As previously indicated, motility of spermatozoa is the most common method for estimating fertility, but it is only an estimate.  Fortunately, after hundreds of failures, some new procedures for determining fertility of spermatozoa in the laboratory have begun to show some promise.  Such a development would be most beneficial, because the cost of truly evaluating the effect of an extender on fertility by breeding mares is time-consuming and astronomically expensive due to the number of females (mares) that must be bred under controlled conditions.

  Generally, if motility of the sperm is satisfactory many mares are bred before the true fertility of the extender is accurately known.  For example, presented in Table 1 are the results of a limited study done with an extender widely used for breeding cattle, that was adapted for stallion semen AI, based only on motility of the spermatozoa. The extender containing the 0.349% Tris supported motility of stallion spermatozoa as well as or better than any extender we ever evaluated. However, only 15.4% of the mares became pregnant when the extender was tested under field conditions. (Presented in Table 2, Stallion A represents appropriate reproductive efficiency per cycle).

Table 1. Effect of Tris Seminal Extenders on Pregnancy Rate in Mares

a -Numbers in parentheses are the number of mares that became
b -Percentages within rows are different (P<0.10).

When AI first became prevalent in cattle, the most important problem to solve was a method to store semen long enough for shipment to inseminators in the field. This required an extender. First, an egg yolk-phosphate extender was developed, followed by one with egg yoke-citrate, which was used for many years, and is still used to some extent with frozen semen.

In the USA one of the first extenders used for stallion semen was a mixture of half and half cream and gelatin. This was replaced by milk, both skim and whole. Although milk was an excellent extender for cattle semen it had to be heated to detoxify one of the ingredients before it was usable.  It was ultimately determined that heating was also necessary when it was used with stallion semen. This was almost immediately followed by the discovery that non-fat dried skim milk could replace heated milk extenders.   Consequently, most of the mares bred utilizing AI is with semen extended with a dried milk extender, commonly known as E-Z MixinŽ.  When the horse industry adopted transported semen, procedures for cooling and shipping had to be developed.  These developments required a number of years before reasonable fertility was obtained.  Today the primary extender used to breed mares both immediately and after semen is collected and after cooling, is a dried skim milk extender (E-Z MixinŽ). It is the only extender that has been thoroughly evaluated with respect to both motility and fertility, utilizing a sufficient number of mares for meaningful conclusions.

Considerable research is currently in progress to isolate the specific component or components in milk that are responsible for fertilization, i.e., pregnancy. We can expect some new extenders to be available in the near future.

However, to avoid many costly mistakes, resulting in poor fertility, it is imperative that any new extender, regardless of the hype, be thoroughly tested in the field.  This testing must include a sufficient number of mares, to assure the breeder that if poor fertility occurs it will not be the fault of the extender.  Further, it is necessary to test every batch that is made to assure that quality control is maintained. Each batch should be identified so that if the extender is suspected, that specific batch can be identified, but please remember that no extender will provide maximum reproductive efficiency unless the instructions for handling are followed.

It is not uncommon for an extender for stallion semen AI to be selected because it was recommended by a "friend" that you happened to see at a meeting, horse show, races, etc. The conversation may start by "how many of your mares did you get pregnant last year?" Your friend says 97%. That has got be good, right? Well let's have a look at the results in Table 2.

Presented in Table 2 are the records of two hypothetical stallions (A and B), each bred to 100 mares.  This emphasizes how erroneous some assumptions can be made from overall pregnancy rates only. No one would dispute that overall pregnancy rates of 99 and 97% for Stallions A and B, respectively, are excellent. However, upon closer observation, a first-cycle pregnancy rate of 65% for Stallion A vs. 35% for Stallion B represents a very large economic difference.  After only two cycles, Stallion A had settled 88% of his mares compared to only 58% for Stallion B. After four cycles, Stallion A had settled 99 % of his mares, whereas Stallion B had settled only 82%. Therefore, the cost of each pregnancy obtained by Stallion B would be much greater than for those impregnated by Stallion A. This cost is borne primarily by the mare owners, and is associated with the routine management cost such as, shipping semen, teasing, palpating and/or ultrasounding. Further, the average date of foaling would be later, resulting in less marketable foals, and creating a group of mares that will always have a late foal unless held open for a year. In this example, overall pregnancy rates were virtually identical, but there was a tremendous difference in fertility.

Table 2. Pregnancy Rates of Two Hypothetical Stallions (A and B). Stallion A had a 65% pregnancy rate per cycle, while Stallion B's pregnancy rate per cycle was 35%. These calculations represent pregnancies determined by ultrasonography at 15 days post-ovulation.

It required 1.53 cycles for Stallion A to obtain a pregnancy and 1.83 breedings compared to 2.86 and 3.16, respectively for Stallion B.  The difference in cycles and breedings per pregnancy represents 50% to 60% more work in the breeding shed for Stallion B and his breeding crew.  In addition to the real costs for the owners with mares to Stallion B, there was a loss of morale within the breeding shed crew, and a poor reputation for the farm and stallion for coming years.

When only overall pregnancy rates are known, or even foaling rates, fertility can be grossly misleading. Further, and most important these poor pregnancy rates may go unrecognized unless excellent records are kept. This scenario could easily happen if one selects an extender without thorough investigation, particularly including pregnancy rates per cycle compared to a known extender. Although there are many laboratory tests for fertility, and many more being developed, the only true test for fertility is pregnancies per cycle or some similar measurement.

Those individuals who utilize extenders that have not been thoroughly tested remind me of the old song, "I'm Always Chasing Rainbows."

About the Author...

B. W. Pickett, Ph.D. is Professor Emeritus, Colorado State University.

   Tris (%)
2.4 0.349
Cycle No. Mares Inseminated Pregnancy Rate (%) No. Mares Inseminated Pregnancy Rate (%)
1. 17(9) a 52.9 b 18 (3) 16.7
2. 15 (4) 26.7 8 (1) 12.5
Total 32 (13) 40.6 b 26.4 15.4
  Stallion A   Stallion B
Pregnancy Rates No. of Mares Pregnant Pregnant Total  (%) No. of Mares Pregnant Pregnant Total  (%)
Overall 99 100 99 97 100 97
Cycle 1 65 100 65 35 100 35
Cycle 2 23 35 66 23 65 35
Cycle 3 8 12 66 15 42 36
Cycle 4 3 4 75 9 27 33
Cycle 5 - - - 6 18 33
Cycle 6 - - - 4 12 33
Cycle 7 - - - 3 8 37
Cycle 8 - - - 2 5 40
Cycles/Pregnancy 1.53 - - 2.86 - -
Cycles/Pregnancy 1.83 - - 3.16 - -
Animal Reproduction Systems

Contact Us
Order Information
ARS Newsletters
View Cart
Product Documentation
Contacting Tech Support
How-to Videos
Copyright ©2000-2014 Animal Reproduction Systems, Inc. All Rights Reserved