You can’t pick your relatives


In yesterday’s post I mentioned that breeders have a new way of finding closely related breeding groups. Genetic markers allow scientists to map out the relationships between the different breeds. This type of analysis was how the Abaco Barbs were identified as belonging the the Colonial Spanish breeding group. This can be really useful for feral herds like the Abacos where there are conflicting theories regarding the origin of the horses.

It is also being used to identify unique populations, with genetically distinct profiles, to target for preservation. This was recently done with a relatively large group of French breeds. The study is open access, so it can be read here:

Genetic diversity of a large set of horse breeds raised in France assessed by microsatellite polymorphism

Studies like this can turn up surprising results. In this particular one, the distance between two breeds that many would have assumed to be closely related – the Percheron and the Boulonnais – were actually quite distant. Both breeds are large, grey drafters from the same part of the world. Some historical accounts even suggest that the latter was used to create the former. And yet the Percheron is more closely related to the Norman Cob (technically a light breed) and the stout, silver dapple Comtois than to the Boulonnais. Here is a chart from that paper showing one method used to group the breeds studied.


Breed legend: PS – Thoroughbred, AA- Anglo-Arabians, SF – Selle Francais, TF – French Trotter, APPAL – Appaloosa, QH – Quarter Horse, AR – Arabian, LUS – Lusitano, PRE – Pura Raza Espanol, AB – Arab-Barb, BA – Barb, LAND – Landais, CAM – Camargue, POT-Pottock, PFS – French Pony, CO – Connemara, WAB – Welsh Pony, NF – New Forest, MER – Merens, BR – Breton, COBND – Norman Cob, COMT – Comtois, PER – Percheron, HAF – Haflinger, POIT – Mulassier, ARD – Ardennais, AUX – Auxois, TDN – Trait du Nord, BOUL – Boulonnais, FRI – Friesian, FJ – Fjord, SHE – Shetland, IS – Icelandic, PRW – Przewalski

Results like this challenge some of our assumptions about breeds. The authors of the study note that a few of the breeds clustered in groups that are different from French registry classifications. Those appear in italics on the chart. The Camargue, considered a warmblood breed in France, falls into the pony breeds, while the Merens, Halflinger and Friesian all cluster with the draft breeds. (To be fair the Friesian is a bit of an outlier there, as it is in almost any equine relationship chart.)

Sometimes the results of these kinds of studies vary a bit in the details, depending on the specific samples used (this one used a pretty large set) or the specific markers being studied. Others are really consistent from study to study, like the grouping of Nordic breeds, highlighted here in pink.


Of course, it is heresy in many Fjord and Icelandic circles to suggest that these horses are ponies. For that matter, the other group that falls into this same cluster (although not used in this study) is the Miniature Horse, which many admirers adamantly insist is not a pony either. Of course, it is hard to maintain that position when your closest relative is the quintessential pony, the Shetland.

Swapping the sections of the chart around, though, shows that the graphs for these Nordic breeds look more like the section of the graph for the other ponies than for the light breeds. Those are the breeds most Americans imagine when the word horse, and not pony, is used.

This is still really new science, but these kinds of papers have been appearing with increasing regularity. Hopefully they will one day provide an even clearer picture of how the different breeds developed and are related. But even with what we know now, it is possible to make more educated guesses about what needs to be preserved, and what might be the best path to take for those breeds with limited numbers. The new information will probably require that we lose some of the mythology that has surrounded many of our breeds, but the benefit should be healthier horses in the long term.

(Fjord image from Mirk-Stock and used with permission, original chart from the open access paper linked in this post.)


9 Responses to You can’t pick your relatives

  1. Jacqueline Ferrigno September 2, 2011 at 11:38 am #

    Oh gods, this reminds me of one of my High School Science Fair Projects. I was comparing phenotype and genetic data of the Florida Cracker Horses to and assortment of breeds which at that time people thought were closely related. I wasn’t able to obtain genetic data myself or any samples without filing a binder full of paper work, but Dr. Cothran already had genetic data, so he was able to mail me what he had concluded. Turned out that the closest relative to the Florida Cracker Horse, was the Welsh Pony. Talk about a freaking mind blower.

    Speaking of which, Cr. Cothran was supposed to publish an article on the Florida Cracker Horse some time ago, I wonder if he finished getting through all that red tape.

    • Jacqueline Ferrigno September 2, 2011 at 3:34 pm #

      I just got an email back from Dr. Cothran, The paper is published in the journal Animal Genetics, but it is not in print yet. He did send me a PDF file however, looks like there has been new information since I last looked. If you are interested in it, I can ask if I can forward you a copy. Or if it comes out in print soon, this is the information.

      A microsatellite analysis of five Colonial Spanish horse populations
      of the southeastern United States
      E. K. Conant, R. Juras and E. G. Cothran
      Department of Veterinary Integrative Biosciences, Texas A&M University, College Station, TX 77843-4458, USA

      • The Equine Tapestry September 3, 2011 at 5:54 am #

        I would love to see a copy. That is really odd about the Welsh Pony, since that breed has traditionally been found in specific regions in the United States, but Florida wasn’t one that comes to mind. It is one of those breeds that had an import wave and then a period of dormancy, though… long enough for that first wave of imports to be “lost” (in stud book terms, but generally horses always go *somewhere*).

      • Jacqueline Ferrigno September 3, 2011 at 5:49 pm #

        Yeah, I was floored. My hypothesis blew up in my face LOL. He had given me an explanation for that years ago, but it had been so long. I will have to go through my burned CD’s and hope that I have a copy of that report. We had a house fire in 2005 and it fried the computer, but my science fair board survived with some smoke damage lol. This dendogram in the new charts I don’t believe looks familiar, as this one has the Florida Cracker being closest to a Walker. But I haven’t had a chance to read the entire file yet. I have sent off an email, and hopefully hear back after the holiday, tho he’s pretty good about checking emails.

      • Mindy Barrick September 3, 2011 at 6:57 pm #

        I found a summary for the article and that alone is interesting. Here is the summary:

  2. The Equine Tapestry September 3, 2011 at 8:08 pm #

    Hi Mindy! Thank you for posting that link. That’s really interesting about the connections between the Tackies and the Bankers. I will have to dig up the pictures of I have taken of the two groups for comparison.

  3. Jacqueline Ferrigno September 5, 2011 at 9:46 pm #

    Just got Dr. Cothrans permission, said it’s published and feel free to share it, so I sent you an email titled Florida Cracker Horse.

  4. hmh December 20, 2011 at 11:28 am #

    Hi, I know this article is a few months old, but I just wanted to comment on a few things.. I don’t agree with your conclusion that the Fjords+Icelandics look more like the ponies than the horses- they look pretty distinct at everything but k=2, which is just separating all breeds based on two assumed ancestral populations (which appears to be light horses vs. heavy horses, not pony vs. horse). The ponies really look like a mess, but there’s no substantial shared colors (and thus ancestral populations) between them at the FJ/IS. I think the Shetlands are clumping with them for so long because there’s too many ancestral populations for the numbers of individuals, and they’re more related to the FJ/IS than other distinct breeds are related to eachother. That being said, I really don’t like the figure from the paper.. they made the graph with values averages over the whole population, but it’s also possible to show values for each individual, which shows much more about the population.. but oh well. I really enjoy your blog posts, but I figured I’d just try to clear that up.


  1. What is tölting? | - September 10, 2011

    […] Another reason given is that they have “big” personalities like horses. But as Lesli Kathman discovered in a genetic diversity report published in the Genetics Selection Evolution Journal, the Icelandic is closely related to the […]