Genetics

Modern research into the genetics of horses and ponies largely supports what the rest of this site has explored through historical documentation, but the study of genetics is still evolving.

The past two decades have resulted in an explosion of research in the field of equine genomics. With the creation of the original marker maps in the horse, subsequent sequencing and annotation of the complete equine genome and the availability of genomic tools to investigate specific traits and diseases, the study of equine genomics has rapidly accelerated.... Undoubtedly, the next decade will continue to see an increase in the amount of available DNA tests for horses. (Finno & Banasch, 2013)

With newer methods constantly being developed, older studies are sometimes, not surprisingly, shown to be in error as the new techniques are refined.

A study of the origins of the domestic horse (Jansen et al., 2002) concluded 'that several distinct horse populations were involved in the domestication of the horse' and that breeds can be grouped in distinct phylogenetic clusters. The study was based on mitochondrial DNA (mtDNA), genetic material which is passed down the female line and exhibits strong geographical characteristics. One cluster contained a high proportion of animals derived from Barb (North Africa) and Iberian stock, including American mustangs. Another contained only pony breeds from the British Isles, Scandinavia and central Europe, including Exmoor, Highland, Fjord, Icelandic and Connemara. (The Fell is identified in the lower left quadrant in Figure 2, mtDNA phylogenetic tree chart; labelled LFEFE1 as a branch off from Exmoor, JEX60.)

An American study, using Randomly amplified polymorphic DNA (RAPD) "marker" analysis, also concluded that horse and pony breeds fall into natural geographic clusters. These can be grouped as:

breeds developed in North America;
breeds related to the Arab;
breeds related to the Andalusian (Iberian);
and northern European breeds including heavy horses and ponies.

The phylogenetic tree produced by this study can be seen here (Cothran, 1995) which should be read from left to right - the left side is the ancient trunk or source, and branches split off towards the right into the most recent "branches" of the tree with the breeds at their tips.

diagram showing phylogenetic tree of relationships between Fell and other breeds

This analysis showed the Fell and Dales, predictably, as sister breeds, and first cousins to the Shire and Clydesdale, second cousins to the Highland, and third cousins to the Hackney.

Subsets of British and Irish pony breeds have formed part of a wider dataset for microsatellite or mitochondrial analysis, although the sample sizes were small and sample origins were often not stated or were from daughter export populations (Jansen et al., 2002; Luis, Juras, Oom, & Cothran, 2007; Leroy et al., 2009; Georgescu, Manea, Dudu, & Costache, 2011; van de Goor, Haeringen, & Lenstra, 2011). (Winton et al., 2020)

Papers using different methods, such as genetic distance, have shown different and more distant relationships with the other breeds in the UK, apart from the Dales:

The Dales and Fell are neighbouring pony breeds, which have their area of origin in the Pennine hills, on the eastern slopes and western slopes respectively. There are records of exchange of breeding animals between the breeds in the first half of the twentieth century, and this is evident in their close genetic relationship (genetic distance 0.289). Both breeds owe part of their ancestry to the extinct Galloway (...) but the Fell is a more distinctive breed which has suffered less introgression. (Alderson, 2005)

What is interesting in this particular analysis is that the Fell is less closely related to the other breeds of horses or ponies in the British Isles. Its distance from the Highland is particularly surprising (as is the lack of support for any Clydesdale influence on the Dales; this is a myth that dogs the Dales breed in the same way that the "Roman Friesian" dogs the Fells):

(The Fell) is significantly distant genetically from breeds in all groups, especially the Thoroughbred (1.008), Suffolk (1.096) and Highland (1.034). ... Clydesdale influence on the Dales Pony has been postulated, but their genetic distance (0.606) does not support this contention. (Alderson, 2005)

Most recently the following paper (Winton et al, 2020), reporting Microsatellite analysis of native British and Irish pony breeds, has found that "Ancestral maternal diversity was maintained by most populations, particularly the Fells and Welsh ponies, which exhibited rare and ancient lineages."

[Scientific terms in this section: An allele is one of two, or more, forms of a given gene. A haplotype is a group of alleles that are inherited together from a single parent.]

The Section C and Section D clearly showed a close relationship for all networks, as did the Fell and Highland ponies.

...the [Welsh] Section D and Fell groups had the highest number of haplotypes (25 and 20, respectively) and also the highest haplotypic diversity values, while the Fell had the highest overall nucleotide diversity.

[The Shetland] is notably more closely related maternally to the Fell than to any other British or Irish breed analyzed here... This relationship has since diverged for the more rapidly evolving SSR markers.

Fell ponies displayed very high maternal diversity, with a broad distribution of haplogroups, and moderately high nuclear diversity. ... The Fell has shared maternal lineages with Sections A, C, and D, and to a lesser extent a relationship with the New Forest and Shetland. This is in contrast to the SSR data, where the Fells show the closest relationship to the Highland. ... It is likely given the results found here, that sharing of stock occurred between these regions, with introductions of ponies of Fell-Galloway origin into the Scottish Highlands via drovers, thus influencing the development of the modern Highland pony breed. It is apparent that this happened largely according to male-biased crossing, as the maternal data do not show significant haplotype sharing between the Highlands and Fells. The Fell ponies showed the closest autosomal relationship with the Welsh Ponies and Cobs (particularly Section C), the Connemara, and the Highland. The Fell and the Welsh Sections share a strong phenotypic trait of a notably fast trotting ability. ... The Fell and the Welsh populations have maintained the greatest maternal diversity of the ancestral British ponies, with the presence of rare ancestral haplotypes and unique derived haplotypes. The retention of ancient mitochondrial diversity is likely a function of the isolated nature of these upland ponies and the fact that foreign introductions have largely been limited to male-biased influences.

The Fell ponies face the challenge of reducing the proportion of FIS carriers in the population, but great care must be taken to ensure unique ancestral maternal haplotypes are not lost in the process. (Winton et al, 2020)

Practical Application of Genetic Research

Genetic research, of course, is not confined to the evolution of horse breeds. Its practical application targets both useful mutations (such as the "gaitkeeper" gene DMRT3 which controls whether a horse typically performs lateral or diagonal gaits) and damaging ones such as inherited diseases.

Extensive research has been done into the causes of equine diseases with challenging names such as hyperkalemic periodic paralysis, severe combined immunodeficiency, overo lethal white syndrome, junctional epidermolysis bullosa, glycogen branching enzyme deficiency, malignant hyperthermia, hereditary equine regional dermal asthenia, and polysaccharide storage myopathy.

However, Fell, Dales and coloured horses have, as far as I know, only one inheritable disease factor, Foal Immunodeficiency Syndrome or FIS.

Foal Immunodeficiency Syndrome

FIS is caused by a single mutation in the sodium/myo-inositol cotransporter gene (SLC5A3). This gene plays a vital role in the regulatory response in many tissues including lymphoid tissues.

FIS is an autosomal recessive trait, meaning a foal can only be affected if the foal inherits the disease from both parents. Animals that are carriers do not have any symptoms associated with FIS and will live normal healthy lives. However, if they breed, they are likely to pass on a copy of the defective gene to their offspring. Affected foals, which have inherited a copy of the mutated allele from both parents, suffer a variety of difficulties: they appear to be normal at birth, but develop infections within a few weeks. The infections are resistant to treatment, and foals die or are put down by three months of age.

table of outcomes of various breeding combinations
The Fell Pony Society's Health page on its web site links to various documents which explain the inheritance (such as the above diagram) including the following, written in 2006 by Bob Charlton:

Every foal inherits half of its genetics from each parent. If, in BOTH of these parents, one gene is defective (made so by the presence of the syndrome configuration of alleles that make up the gene), then there is a one in four chance that the resulting foal will suffer from the syndrome and will die.

There has been much confusion about what a one in four chance actually is. Some people have misinterpreted it as being that for every four foals a carrier mare gives birth to, if mated with a carrier stallion, one will be a syndrome foal. NOT SO - it is EACH foal that has a one in four chance of being a syndrome foal.

In the mid-1990s Paul May MRCVS realised that the various presentations he was seeing of sick foals were in fact a syndrome with one root cause. Research was conducted between 1996 and 2010 at the University of Liverpool's Leahurst campus, principally by Prof. Derek Knottenbelt, Prof. Gareth Thomas, Prof. Stuart Carter, Dr June Swinburne and Dr. Laura Fox-Clipsham. They announced a marker test in early 2010, which now enables breeders to identify carrier animals and make informed decisions about breeding.

The Society currently (2020) uses Weatherbys Scientific to handle this testing.