Characterisation of the Connemara Pony Population in Ireland

Report presented to
The Department of Agriculture, Food and Rural Development,
Agriculture House, Kildare St., Dublin 2.

Report funded by
The Department of Agriculture, Food and Rural Development,
Agriculture House, Kildare St., Dublin 2.

Deirdre Feely B.Agr.Sc.1,
Patrick Brophy MVB MRCVS1,
Katherine Quinn M.Agr.Sc1.

1Department of Animal Science and Production,
Faculty of Agriculture,
University College Dublin,
Belfield, Dublin 4.

Introduction

The Connemara Pony is numerically a small breed, with approximately 2,000 breeding females and 250 breeding males in Ireland. Traditionally, the Connemara was a working pony and enjoyed a prominent role in agricultural life in the West of Ireland. However, in the middle of the last century farming practices changed, and as machinery was introduced, the role of the working pony became redundant. The Connemara Pony has maintained its popularity by establishing a position in the showing and riding industry.

The Connemara Pony Breeders’ Society was founded in 1923. The main objectives of the Breed Society are the encouragement, development and maintenance of the Connemara Pony as a pure breed. Since its formation the Society has also been responsible for the publication of the Connemara Pony Stud Book. The breed is now recognised throughout the world as a top class performance pony and 17 different countries have established their own Breeders’ Societies.

As yet, there is little concern regarding the number of pure bred foals produced annually. However, the Connemara Pony Stud Book has been closed since 1964 and the practice of overusing popular sires is prevalent throughout the history of the breed. This could potentially led to a narrowing of the gene pool, high levels of inbreeding and a loss of genetic diversity within the population.

The main objectives of this project were to demographically and genetically characterise the Connemara Pony population, with specific emphasis placed on how past breeding practised have affected the present genetic composition of the breed. Height trends were also analysed in attempt to establish evidence of genetic erosion.

Methodology

A total of 20,032 records were used in the characterisation of the Connemara Pony population. These records were obtained from the Breed Societies’ database and from Dan-Axel Danielsson of the Swedish Connemara Pony Society.

The study focused on two reference populations. The first reference population consisted of 2,316 registered ponies born between 1993 and 1996 inclusive. This reference population represents the current breeding population of Connemara Ponies. The second reference population contained 2,844 foals born between 1998 and 2001 inclusive. This reference population contained records of both registered and non-registered ponies and represents the future breeding stock of the Connemara Pony. It should be noted that approximately one third of the non-registered ponies in this reference population will not be subsequently registered as Connemara Ponies.

The reference populations were characterised both demographically and genetically. The demographic characterisation is a description of a population in numerical terms and enables us to see how the size and structure of the population has altered over time. The main parameters estimated as part of the demographic characterisation included the number of registered ponies born year, the sex ratio, the average generation interval and family size. The genetic characterisation of a population is conducted to determine the level of genetic diversity within the population. As part of the genetic characterisation the average inbreeding coefficient and average relationship coefficient for animals in the reference populations were calculated. The contributions made by the ancestors of the reference populations, the number of founders, the effective number of founders and the effective number of ancestors were also estimated in order to measure the level of genetic diversity within the population. The influence that the Thoroughbred, Arab, Irish Draught and Welsh Cob breeds had on the reference populations was calculated. Height trends in the Connemara Pony were also analysed.

Summary of results

  • Demographic characterisation of the population
  • Up until 1959 the number of registered ponies born annually was consistently below 100. There was a steep increase in the number ponies born per year between 1959 and 1970, peaking at 568 in 1970. The population size was reduced considerably between 1971 and 1980, but recovered again between 1980 and 1996. At present there is approximately 800 pure bred foals born annually, however only 60 to 70% of these are subsequently registered as Connemara Ponies.
  • In recent years the ratio of registered mares to stallions deteriorated considerably. In 1980 for every 8.53 registered mares born there was one registered stallion born. In 1994 this ratio increased to 30.5 registered mares per stallion.

Demographic analysis of the reference populations

  • In the 1993 to 1996 reference population the number of registered ponies born per year increased from 548 in 1993, to 590 in 1996. In the 1998 to 2001 reference population the number of foals born per year appeared to decrease from 866 in 1998, to 441 in 2001. However, it is assumed that this is due to a lack of records for foals born in more recent years rather than an actual drop in annual foal production.
  • 181 different sires produced the 2,316 animals in the 1993 to 1996 reference population. The number of progeny per sire ranged from 1 to 140. 214 different stallions produced the 2,844 animals in the 1998 to 2001 reference population. The number of foals per stallion ranged from 1 to 151.

Age profile of the sires and dams of the reference populations

  • The average age of sires as higher for the more recent reference population. For example, 7% of the sires of the 1993 to 1996 reference population were under 5 years of age, but in the 1998 to 2001 reference population only 2% of the sires were under 5 years of age.
  • A larger proportion of younger dams produced the more recent reference population. 15% of the dams of the 1993 to 1996 reference population were 20 years or older, while only 5% of the dams of the 1998 to 2001 reference population were of that age group.


Generation interval

The generation interval is defined as the average age of the parents when their offspring are born.

  • The average generation interval between parents and offspring in the 1993 to 1996 reference population was 10.51 years. The average generation interval for the 1998 to 2001 reference population increased slightly to 10.59 years.
  • For both reference populations the generation interval between sires and their offspring was approximately 2 years longer than the generation interval between dams and their offspring.
  • Between 1980 and 2000 the average generation interval increased by 16%, from 8.98 years to 10.44 years.
  • The generation interval calculated is similar to the generation intervals found in other horse populations. The longer generation interval between sires and their progeny may indicate that breeders have a preference for older and proven sires, or it may merely reflect that stallions tend to commence breeding later in life.

Family size

For the purpose of the analysis family size was defined as the number of ‘breeding’ offspring per sire and dam. Offspring were deemed ‘breeding’ if they had produced at least one registered offspring themselves. In an ideal situation, family sizes would be balanced, giving each breeding animal an equal chance of producing their own replacements in the next generation.

  • On average each sire produced 2.88 ‘breeding’ male and 9.94 ‘breeding’ female offspring. Paternal family sizes were found to be extremely unbalanced, with a large proportion of the breeding population being produced by a small pool of stallions. For example, 10% of sires produced 55% of the ‘breeding’ female offspring and 30% of the ‘breeding’ male offspring.
  • On average each dam produced 1.24 ‘breeding’ male and 1.77 ‘breeding’ female offspring. The maternal family sizes showed less variation compared to paternal family sizes, as dams are biologically limited to producing one foal per year.
  • Paternal family sizes were very unbalanced. This is likely to cause a loss in the genetic variation of the breed, an increase in the relationship among animals in future generations, and a rise in the level of inbreeding.
  • Genetic characterisation of the population


Pedigree completeness

Pedigree completeness is an important parameter as the accuracy of the genetic characterisation is largely dependent on the quality of the records used in the analysis. Pedigree completeness was measured by determining the proportion of ancestors known per generation. The complete generation equivalent was also used to measure pedigree completeness and is defined as the average number of complete generations recorded.

  • Both reference populations’ pedigree data was practically 100% complete for the first 3 generations, i.e., almost 100% of all parents, grandparents and great grandparents were known. After the 5th and 6th generations, the proportion of known ancestors steadily decreased.
  • There complete generation equivalents for the 1993 to 1996, and the 1998 to 2001 reference populations, were 6.15 and 6.59 respectively.
  • The level of pedigree completeness for the animals in the reference populations was deemed to be relatively high when compared to corresponding studies.


Inbreeding

The coefficient of inbreeding measures the probability that an animal receives identical genes by descent from its sire and dam.

  • The average inbreeding coefficient for animals in the 1993 to 1996 reference population was 4.49%. This had increased to 4.65% for the 1998 to 2001 reference population.
  • 5.31% of the animals in the 1993 to 1996 reference population had inbreeding coefficients under 2%. The proportion of ponies in the 1998 to 2001 reference population with inbreeding coefficients under 2% was 2.67%.
  • The increase in inbreeding from 1980 to 1990 was 0.93% and was very similar to the expected rate at which inbreeding would increase under random mating conditions.
  • The average inbreeding coefficient for animals born in 1980 was 3.19%, this has increased steadily, reaching 4.65% in 2000.
  • The level of inbreeding detected in the Connemara Pony population was high in relation to most comparable studies. The actual increase in inbreeding corresponded to the theoretical increase in inbreeding expected if mating were at random, indicating that breeders did not take sufficient steps to avoid the mating of related animals.

Average relationship coefficient

The average relationship coefficient measures the proportion of genes that animals have in common.

  • The average relationship coefficient among animals in the 1993 to 1996, and 1998 to 2001 reference populations were 10.26% and 10.66% respectively. The relationship among the sires of both of the reference populations was almost 1% higher than the relationships among the dams of the reference populations.
  • Considering the average relationship between two first cousins is 12.5%, the average relationship among animals in the reference populations was extremely high. The average relationship among the sires of the reference populations was higher than the relationship among the dams of the reference populations, implying that stallions selected for breeding are of similar ancestry or breeding lines.

Contributions made by ancestors 

Important ancestors were identified by calculating the marginal contributions made by ascendants to the reference populations. The marginal contribution is the contribution made by an ancestor that is not already explained by another animal.

  • Carna Bobby was the most important ancestor to both of the reference populations, with a marginal contribution of 13.81% to the 1993 to 1996 reference population and 13.93% to the 1998 to 2001 reference population. Dun Lorenzo and Carna Dun were the next most important ancestors, contributing approximately 10.5% and 8.5% respectively to the genes of the reference populations.
  • Overall the contributions made by the ancestors of the reference populations were found to be very unequal with 6 ancestors contributing 50% of the genes to both of the reference populations
  • The imbalance of the contributions made by ancestors implies that future generations are at risk of further losses in genetic variation.

Number of founders, effective number of founders A founder is defined as an ancestor with unknown parents or the unknown parent where only one parent is unknown. It is assumed that all founders are unrelated and all of the genes in the populations emanate from these founders. The effective number of founders is a theoretical number defined as the number of equally contributing founders that would be expected, given the level of genetic diversity that exists in the reference population. The more balanced the founder contributions are to the reference population, the more the effective number of founders will approach the actual number of founders.

  • There were 351 founders for the 1993 to 1996 reference population and 342 founders for the 1998 to 2001 reference population.
  • The effective number of founders for both of the reference populations was 35.8.

The discrepancy between the actual number of founders and the effective number of founders is expected to decrease the amount of genetic diversity in the present population relative to what would have transpired had all founders contributed equally.

The effective number of ancestors

The effective number of ancestors is a theoretical number that supplies us with the minimum number of ancestors needed to explain the complete genetic diversity of the reference population. Unlike the effective number of founders, the effective number of ancestors accounts for bottlenecks in the pedigree. The closer the effective number of ancestors is to the effective number of founders, the smaller the impact past bottlenecks have had on the genetic diversity of the population.

  • The effective number of founders for both of the reference populations was approximately 18.
  • This indicates that past bottlenecks have adversely affected the genetic diversity of the animals in the reference populations.

The breed composition of the reference populationsA small number of Thoroughbred, Arab and Irish Draught stallions sired registered Connemara Ponies in the 1940s and 1950s. There were also two stallions in the pedigree file that were known to have Welsh Cob genes. The proportion of genes that the animals in the reference populations possessed, originating from these stallions, was estimated to obtain the influence that the Thoroughbred, Arab and Irish Draught had on the reference populations..

  • The Thoroughbred was the most influential of the foreign breeds, accounting for approximately 6% of the genes of the reference populations. The Arab, Irish Draught and Welsh Cob accounted for approximately 3.7%, 1.2% and 0.9% of the genes of the reference populations respectively. Approximately 88% of the genes in the animals in the reference population are assumed to be Connemara Pony.
  • The majority of the animals in the reference populations possessed at least some Welsh Cob, Thoroughbred and Arab genes. Approximately 50% of the animals in the reference populations had Irish Draught in their ancestry.

Height trends in the Connemara PonyThe traditional role of the Connemara Pony was as a versatile working animal. The main emphasis was on producing ponies with strength, hardiness, good bone and intelligence. However, as farming became increasingly mechanised during the middle of the last century, the role that the Connemara Pony had secured as a working animal began to disappear. The breed has survived by gaining a reputation as a performance animal and establishing a place in the showing and riding industry. In order to adapt to present market demands the breed is moving away from a traditional type to a ‘modern’ type of animal that is taller and lighter boned. As the breed moves away from the traditional type valuable genes may be lost, along with the characteristics that distinguish the Connemara Pony from other equine breeds. This process is known as genetic erosion and is a problem confronting many present-day equine breeds. Height trends in the Connemara Pony were analysed in an attempt to identify the occurrence of genetic erosion.

  • The average height at time of inspection of registered Connemara Ponies born in 1970 was 135.33cm. This increased steadily over the years, reaching 142.88cm in 1997.
  • Between 1972 and 1997 the average height of registered mares and stallions at the time of inspection increased by 8cm and 5.5cm respectively. Between 1975 and 1995, ponies bred outside Co. Galway were significantly taller than ponies bred inside Co. Galway (141.29cm +/- 4.56 versus 139.23cm +/- 5.16; P=0.0001). However, between 1991 and 2000 there was no difference in height between the two groups (142.49cm +/- 3.98 versus 142.38cm +/- 4.18; P=0.3834).
  • The analysis of height trends confirmed that Connemara Ponies are growing taller. It is assumed that this is a consequence of both improved environmental conditions and selection for taller ponies.

Conclusion

Following the analysis, it appears that the Connemara Pony breed is being confronted with two problems. Firstly, the survival of the traditional type of breed is under threat, and secondly, the genetic diversity of the breed is diminishing.

Today, the riding industry is an important outlet for Connemara Ponies. However, there is concern that this industry is instigating a shift from the traditional type of pony, to a taller, ‘modern’ type. The traditional type of Connemara Pony is perfectly adapted to the environment in which it developed and is completely distinct from other equine breeds. It may be necessary to safeguard against market forces inciting the disappearance of the traditional type of pony, which is a valuable national resource, and once lost can never be recovered.

The results generated from the characterisation of the Connemara Pony population indicate that past breeding practices have caused a significant loss in the breeds’ genetic diversity. To ensure that the genetic variation in the breed does not recede to a detrimental level, breeding policies need to be altered.

In future, it is vital that sire family sizes become more balanced, giving all stallions a better opportunity to breed their own replacements in the next generation. This would help to control the level of inbreeding and genetic diversity within the population.

The stallions used for breeding are closely related to each other and tend to be of similar ancestry or breeding lines. From a genetic diversity perspective it may be advantageous to have a pool of breeding stallions that are less related to each other to bestow a variety of genes to the proceeding generations.

As relationships among animals in the present population is high the mating of related animals is inevitable. Breeders must be very vigilant in respect to the stallions that they use for breeding to ensure that a minimal amount of inbreeding is practised.

There are 16 different countries, outside Ireland, that have formed their own Breeders’ Societies and maintain their own stud books. A study is presently being undertaken to characterise the Connemara Pony populations in a number of these countries. It is hoped that these animals may be a source of genetic variability that could be used to widen the gene pool of the Irish Connemara Pony population.

Adequate genetic diversity is vital for the long term health and viability of any population. Thus, it is vital that breeding practices are altered in order to secure the future prosperity of the Connemara Pony breed.