Ideal horses for rewilding: Exmoor, Konik or Przewalski?

In this post, I described a number of Tarpan-like horse breeds. None of these breeds is a complete stand-in for the Tarpan, but each of them is able to survive without human help and each of them resembles at least one colour morph of the Tarpan to a larger or lesser extent. That’s why most rewilding projects are convinced that no further breeding (more precisely, “breeding-back”) is necessary for "restoring" the wild horse in Europe. Speaking in terms of ecology and social behaviour, this might be true indeed. But in this post I want to share my thoughts on which kinds of breeds would give us an authentic picture of how the Tarpan looked like and behaved like, and what combination might be best for being rewilded.

As I explained here, the Tarpan did not have a uniform appearance. Although probably all Tarpans had a relatively small, strong and stocky body with a robust head and a short frizzy mane, there are five possible colour morphs known. Based on historic and genetic evidence, bay dun (like in the Przewalski’s horse) probably was the most common of these, followed by black dun. There is no evidence of a correlation of these morphs to a specific region (yet), so probably each of them were distributed all over Europe. Therefore I think that we have to consider more than one breed or type of breed if we want to create authentically Tarpan-like feral horse populations. One also has to consider that landraces usually show regional adaption to a specific habitat or climate and might do not as well in other regions. For example, the attempt to introduce Koniks into the Atapuerca mountains failed badly because the horses were not adapted to the mountainous habitat (and also the presence of wolves).

The Przewalski horse is, as we all know, the only living wild representative of Equus ferus. Its authentic wild horse features, be it phenotypic or behavioural, are certainly very desirable and their influence would compensate the domestic “maladaptions” of some otherwise authentic breeds. Surely some people would be against the influence of Przewalskis in the feral/wild European horse population because there are genetic differences (they have one more chromosome pair and separated from domestic horses during the middle Pleistocene), and they are adapted to a very cold, dry steppe environment. But pure Przewalskis are used in a number of grazing projects in Europe, and also semi-feral in Hortobagy (Hungary) and Atapuerca. Nevertheless, I am inclined not to use pure Przewalskis for releasing them into European wilderness among domestic horses because I think that a) it might be a waste as their native range is Asia and b) their different climatic adaptions might not be ideal in some regions. But F1 hybrids with suited, local primitive horses could be used instead. The use of crosses instead of pure individuals also has the advantage of not diminish the population of Przewalskis that are used for the conservation of this precious subspecies.

As I mentioned here, the ABU in Germany has Przewalski x Konik crosses which probably represent the most common phenotype of the Tarpan very well. They have the muscular stocky body that all these ponies have, a robust head, a short frizzy mane and a perfectly bay dun colour. They are most likely adapted fine to European climate. Since these crosses worked out well, imagine other combinations like Exmoor x Przewalski, or Hucule x Przewalski.

Konik x Przewalski on my flick album

The behaviour of Przewalski’s horses is, not surprisingly, very reminiscent of what is known about that of the Tarpan. They are difficult to handle or to tame and have a high potential aggression, and they are known to kill concurring domestic horses. But in the wilderness, they behave shy towards humans, and they also know how to defend themselves against predators. This is very desirable and needed, because tame released horses can cause problems in interaction with humans (according to some sources, Koniks remain tame even after living their whole life without human contact) and if the horses do not know how to defend themselves successfully against predators, you’ll get the same results as with the Koniks in Atapuerca. UPDATE: In Popielno and the Biesczcadzki Koniks defend themselves successfully against predators - even bears; they form a defensive circle around their foals and stallions try to chase away wolves. See f.e. here. On the other hand I might mention that there are domestic/feral horses which know how to defend themselves as well, such as Exmoors (f.e. they form a defensive circle around their foals, and they tend to be shier) and Garrano from Portugal (feral Garrano deal with wolves).

I agree that actual breeding-back is not really necessary in the case of the horse, because the suited breeds are very close to the desired archetype already. Instead, releasing a bunch of suited breeds together in an area so that they interbreed and letting mother nature do the rest might be sufficient already.

Therefore, my proposal of breeds and breed combinations is:

Iberia:

Garrano, Asturcon (used to predation, phenotypic closeness to the Tarpan but lack the dun factor), primitive Pryor Mountains mustangs and Przewalski crossbreeds (bring in the dun factor). This combination would result in a population that is experienced in living feral, dealing with predation, suited to the climate and contains all desired wild horse features.

Garrano during winter

Pryor Mountain Mustang

Northern Europe:

Exmoor (feral ancestry, phenotypic closeness to the Tarpan, but lacking the black gene and dun factor), Fjord horse (brings in the dun gene), Konik (brings in dun and black), Przewalski crosses. This population would be perfectly cold-adapted, have all the Tarpan features and contain animals that can deal with predators (not only the Przewalski is adapted to predation, but also Exmoor ponies were exposed to predators in previous centuries). The lack of dun is actually not a negative feature, because the presence of the dun factor within Holocene wild horse populations is not evaluated yet, and non-dun wild coloured horses like the Exmoor are perfectly camouflaged in wooded habitats.

Semi-feral Exmoor ponies

Fjord horses

Western, Central and Eastern Europe:

Exmoor, Konik, Hucule and Noriker (used to mountainous habitat), Przewalski crosses.

Noriker horse

Koniks

I am sure that everyone has a different opinion on this, and some might not agree with me at all and prefer regionalism, existing feral populations or the exclusive presence caballine horses. Until now it seems that each project uses its own breed, and I think this is good. It might lead to exactly the diverse horse population that contains all known Tarpan features as I am proposing it here. Only time will tell. But in my opinion, the fact that the Tarpan had a number of different colour morphs and that modern breeds have different local adaptions should be considered when compositing a new semi-feral horse population for a specific region.