Primitive traits found in the anatomy of the Konik - is there truth behind the myth?

The very popular anecdote for the origin of the Konik is that it descends from the last Polish wild horses kept in a game park in the 19^th century. This story is very common and even repeated in some scientific papers, although the evidence for this scenario is very weak. I called it the “Konik myth”, and describe in this post why that story is actually very unlikely. Based on the historic and also genetic evidence, it seems that the Konik is nothing but a robust landrace with no close affiliation to alleged late-surviving Polish wild horses. 

There are, however, four papers that exanimate morphological details in several wild equines and domestic breeds that provide some clues that, it may seem, there is some truth behind the myth of the Konik because it displays several primitive morphological traits that many other domestic horse breeds do not. Is it maybe not a myth after all? Let’s dive into these morphological studies. 

 

Two of those papers concern the neck anatomy. The nuchal ligament lamellae attach to the cervical vertebrae C2 to C7 in equines such as the Przewalski’s horse, the donkey and the Grant’s zebra, and also in tapirs, suggesting that this is the basal condition for Equus [1,2]. In most domestic horse breeds observed the nuchal ligament lamellae attach from cervicals C2 to C5, with no attachments to C6 and C7. However, in two of four Konik individuals examined and one of one Bosnian Mountain horse, the lamellae attach also to C6 and C7 as in other equine species [1]. The authors consider both the Konik and the Bosnian Mountain horse “tarpan descendants”. As a reference for the “tarpan” ancestry of the Bosnian a book is listed, which I do not have access to. So I did google search on the alleged affiliations of the Bosnian to the European wild horse. It seems that there is the story that the Bosnian mountain horse descends from a mix of the “tarpan” (= the European wild horse, not those nebulous horses of the Russian steppe in the 18^th and 19^th century that were in fact called tarpan) and Mongolian horses. I do not know if there is any evidence for this story. Anyway, the authors conclude that the attachment of the nuchal ligament lamellae found only in those two breeds endorse the idea that these two breeds recently descended from European wild horses [1,2]. Later it was found that this condition is also present in the Australian Stock horse [2], therefore it is not limited to the two alleged “tarpan descendants”. 

The other anatomical detail that is supposed to link the Konik to wild horses is found in the distal limb. The interosseous muscles II and IV are thin ligamentous structures in most domestic horses, while in the Konik and Bosnian Mountain horse they are strong chord-like bands. It has been found that this is also the case in all donkeys, Przewalski’s horses and Grant’s zebras examined, what suggests it is a primitive trait and the anatomy makes it likely that it serves a function [3,4]. 

 

Does this mean that there is truth behind the alleged recent wild horse ancestry of the Konik and the Bosnian Mountain horse? At first glance it may indeed suggest that. But this is not necessarily the case. Regarding the neck anatomy, the primitive condition was found only in half of the Koniks observed, while the other half had the domestic condition. Furthermore, the presence in the Australian Stock horse shows that it is not limited to the Konik and Bosnian Mountain horse. It might also be found other horse breeds that were not examined. About 20 domestic horse breeds were examined (the Exmoor pony and Icelandic horse among them), but other breeds that might be interesting in this context were not. I am particularly thinking of landraces such as the Mongolian horse or the Finnhorse and Latvian horse (which were recently found to partially descend from the Przewalski’s horse [5] and therefore might have some primitive anatomic traits), or the Vyatka horse or the Hucule (one study suggests the Konik and Hucule might be related [6]). Consequently, these primitive traits might be found in more landraces than just the Konik and Bosnian Mountain horse. 

It also should not be forgotten that there is a chance that some or perhaps even all modern Konik lineages have introgression from the Przewalski’s horse. When Lutz and Heinz Heck carried out their “breeding-back” experiments, they crossed several pony breeds, also Koniks from Poland, with a Przewalski’s stallion [7]. After the Second World War, many Koniks from Poland were returned to Poland, including horses from the Heck brothers, which may have been part Przewalski’s horse. A rather strong hint for the Przewalski’s influence in Polish Koniks is the fact that some of them have short, erect manes (see here). I think the only explanation for that is Przewalski’s introgression. Koniks with upright manes are, among other locations, found at Popielno, the most important Polish Konik breeding site where most Koniks imported to other countries come from [8], including Dutch Koniks used in the studies cited above. A genetic test could bring clarification. In Germany, the situation is even more intransparent. On breeding sites such as the Wildpark Hardehausen, Heck horses were crossed with Przewalski’s horses once again to achieve upright manes, and since many breeders in Germany do not differentiate between Koniks and Heck horses, the situation is rather unclear. Now there are several herds of either Koniks or Heck horses in Germany with upright manes and a nebulous history, such as these here. 

So introgression from the Przewalski’s horse might explain the primitive traits seen in the Konik. The fact that the primitive condition of the neck ligaments was found in only half of the Koniks examined might endorse this explanation. 

Another possibility is, since the strong distal limb ligaments apparently serve a function according to the authors, that this primitive trait was never lost in the ancestors of Konik, Bosnian Mountain horse and whatever other landraces that might have them. So that this trait was retained rather than the result of recent wild horse introgression. 

 

All in all, those studies do not necessarily suggest a recent wild horse ancestry for the Konik and the Bosnian Mountain horse. These primitive traits might also be found in other landraces that were not examined, and partly it might be explained by Przewalski’s horse introgression in the case of the Konik. 

The Konik myth can only be true if 1. the horses kept at the Zamoyski game park were indeed wild horses (for which there is no direct evidence), 2. if those horses were indeed donated to the local farmers (for which there is only one dubious historic source and which is unlikely for historic reasons), 3. if there was still wild horse influence in the local farm horses after more than 100 years of breeding (which is unlikely as well). So the Konik myth is actually a less likely scenario than Przewalski’s introgression or other landraces/horse breeds which were not examined having the same primitive traits. More research including other landraces would have to be done in order to clarify that. 

 

References

 

[1] May-Davis et al.: The disappearing lamellae: implications of new findings in the family Equidae suggest the loss of nuchal ligament lamellae on C6 and C7 occurred after domestication. 2018. 

[2] May-Davis et al.: Rare finding of a full nuchal ligament lamellae with attachment points from C2-C7 in one Australian stock horse. 2019. 

[3] May-Davis et al.: A primitive trait in two breeds of Equus caballus revealed by comparative anatomy of the distal limb. 2019. 

[4] May-Davis et al.: Comparative distal limb anatomy reveals a primitive trait in 2 breeds of Equus caballus. 2019. 

[5] Kvist & Niskanen: Modern Northern domestic horses carry mitochondrial DNA similar to Przewalski’s horse. 2020. 

[6] Cieslak et al.: Origin and history of mitochondrial DNA linages in domestic horses. 2011. 

[7] Heck: The Breeding-back of the Tarpan. 

[8] Tadeusz Jezierski, Zbigniew Jaworski: Das Polnische Konik. 2008.

 

 

The post-domestic wildtype: Is the Przewalski's horse wild or feral?

EDIT: A 2021 study found that the Botai horses were not domestic, therefore the Przewalski's horse is still a predomestic, never domesticated wild horse. That explains why it has no domestic traits. 

The Przewalski's horse - wild or feral?

The Przewalski’s horse was long considered the last strain of wild horses that remained undomesticated, while all other populations have been wiped out by man. But as most of my readers will know, this view has been challenged in recent years. A study by Gaunitz et al. 2018 found that the modern Przewalski’s horse descends from the domesticated horses of the Botai culture, the earliest domesticated horses known so far [5]. As it seems, these domesticated horses were not of the caballus/ferus subspecies, but of the przewalskii subspecies. Assuming this is correct, the modern Przewalski’s horse has to be considered a feral form of horse, and not a genuine wild horse, according to the authors of the study. I, however, think that there is still a way to consider the Przewalski’s horse a wild animal. 

 

Domestic animals all show similar traits as a result of domestication. This is called the “domestication syndrome”, which I covered here several times on my blog (go here or here). Classic symptoms of the domestication syndrome are a spotted/piebald coat colour, paedomorphy (the retention of juvenile traits into adulthood), less intense fight-or-flight reaction (i.e. the animals become more tame and agreeable, even trainable), reduced sexual dimorphism, earlier maturity, loss of seasonal adaptions and a reduced brain volume [1,2]. You find these traits in all currently domesticated mammals, be it dogs, cats, sheep, goats, domesticated silver foxes or domestic horses of the ferus/caballus subspecies. Even feral domestic horses, such as mustangs in North America or brumbies in Australia, show vestiges of domestication (f.e. piebald coat colour). Does the Przewalski’s horse show any of these typically domestic traits, being a once-domesticated horse? 

Although the coat colour variation of the Przewalski’s horse was greater before the genetic bottleneck event (f.e. more different nuances of brown), piebald individuals have never been documented. All known individuals seem and seemed to be of the bay dun base colour, which is also found in other equine species such as the kiang or onager. 

Skull of a Pleistocene wild horse from Denmark (left), a Przewalski's horse (middle) and an Exmoor pony (right)

Also, the Przewalski’s horse cannot be considered paedomorphic. A comparison between the skull of a Przewalski’s horse, a wild Pleistocene horse that surely was undomesticated and a robust domestic breed, the Exmoor pony, shows that the Pleistocene skull and the Przewalski’s horse’s skull look very similar while the Exmoor pony skull clearly has a shorter snout, larger eyes and a less massive lower jaw. The Exmoor pony’s skull has to be considered paedomorphic, while the Przewalski’s horse’s skull shows no signs of paedomorphy (it has smaller eyes, a longer snout, a massive lower jaw). 

Do Przewalski’s horses have a reduced brain volume? The brain volume of Przewalski’s horses has never been compared to that of an older wild horse, but domestic horses have 14% less brain case capacity and 16% less brain weight than wild-living Przewalski’s horses [3]. The brains of Przewalski’s horses therefore are larger than those of domesticated horses (it has to be noted that Przewalski’s horses living in zoos have smaller brains though [3]). The brain volume of wild-living Przewalski’s horses are therefore probably not smaller than those of predomestic wild horses, or at least not significantly otherwise it would be comparable to that of the domestic horse. 

Przewalski’s horses reach maturity later than domestic horses. While domestic stallions reach maturity between the 12. and 20. month, mares reach it after 12 to 18 month. Przewalski’s horses reach maturity after two years (Wikipedia). This is not that much of a difference, but it is later. It is not known how long it took predomestic horses to reach maturity. According to Wikipedia, other wild equines such as the onager also reach maturity at the age of two years. 

Sexual dimorphism is not as considerable in equines as in, for example, wild bovines. In all species and subspecies of extant equines, including both the Przewalski’s horse and the domestic horse, the males are slightly larger than the females. 

Both the Przewalski’s horse and the domestic horse have retained some degree of seasonality in their reproductive circle, most foals are born in spring. Therefore it cannot be claimed that the Przewalski’s horse lost seasonal reproductive adaptions as many domestic animals did. 

Regarding the behaviour of the Przewalski’s horse, it can be said that there are noticeable differences to the domestic horse in temperament and trainability (while the social behaviour is more or less identical). Przewalski’s horses are considerably more aggressive, especially the stallions. These often fight very harshly in a herd, sometimes to death [4]. They react very aggressively to other equines in their range and may even attack their mares, and humans (while free-ranging domestic horses do not attack humans) [4]. While free-ranging domestic horses can be transported without sedating, Przewalski’s horses have to be sedated [4]. Because of their aggressive and mostly untameable behaviour, Przewalski’s horses are much more difficult to handle than feral domestic horses. Nothing in their behaviour would suggest that the Przewalski’s horse is not a wild but a feral horse. Bunzel-Drüke et al. write in 2011: “Keeping Przewalski’s horses should be left to specialists. These animals are not just another horse breed, but a genuine wild animal, which is to be handled totally different than feral domestic horses due to the aggressiveness of the stallions”[4].  In rare cases, Przewalski’s mares have been trained, such as the individual Vaska, which was trained to be ridden. This can also be done with zebras (go here for a video), which are undomesticated, so it is possible to do this with a wild equine and hence the (limited) trainability of single Przewalski’s horses is not an argument for a domestic nature. As outlined above, the behaviour of Przewalski’s horses differs from that of feral domestic horses by the high level of aggression and the difficulty to handle it.

 

All in all, when taking a close look at the Przewalski’s horses’ biology, it becomes evident that it does not possess any of the typically domestic traits found in domesticated animals. On the basis of behaviour, the Przewalski’s horse is clearly distinguishable from feral domestic horses. The Przewalski’s horse was, ever since its discovery, always viewed as the wildtype of horse and nobody had the suspicion it might be feral instead of genuinely wild before the Gaunitz et al. study came up. There is only one morphologic difference to the domestic Botai horses the modern Przewalski’s horses are supposed to descend from: the limb elements seemingly became less robust than in their ancestors. This is not necessarily a typically domestic trait and could also be the result of genetic drift. The modern Przewalski’s horses may descend from a domesticated population, but this domestication process obviously could not have been very intense, otherwise the modern Przewalski’s horse would have clear domestic traits. And not to forget, the Botai horses were from 5500 years ago. These ancestors must have run wild again at some point since the Przewalski’s horse was exclusively known in its free-ranging state. Assuming this happened not long ago after the domestication event, the modern Przewalski’s horse was exposed to natural selection for – at maximum – 5500 years. It seems that the Przewalski’s horse became fully dedomesticated during these millennia, after being in a domesticated stage that was not intensely domestic in the first place. 

How to categorize the Przewalski’s horse, then? It seemingly never was very intensely domesticated, it must have become fully dedomesticated after millennia in the wild and consequently does not possess any domestic traits today. Yet, it is not predomestic as it has a domestic history. 

In my Dedomestication series, I coined the term “post-domestic” (as opposed to predomestic) for a fully dedomesticated animal population. It would be a wildtype, but a new, secondary wildtype – shaped by nature, just as the predomestic wildtype. My concept of the post-domestic wildtype was theoretical, as I did not have any unambiguous examples for this state – feral cattle, feral pigs or feral domestic horses living today are not there yet, and it is questionable if the dingo can be truly considered a post-domestic wildtype or just a feral dog. But now that we have the Przewalski’s horse as probably the only once-domesticated animal that became fully dedomesticated during millennia of natural selection, there is a very good candidate for a post-domestic wildtype. 

Regardless of whether the concept of a dedomesticated, post-domestic wildtype will ever become established in the zoological community, I consider the Przewalski’s horse a wild animal. Not a predomestic, but a post-domestic one. The establishing of these terms is only paperwork, dedomestication is undoubtedly a real evolutionary process that is sadly under-studied. 

 

References

 

[1] Trut, 1999: Early Canid Domestication: The Farm fox experiment. 

[2] Dobney & Larson, 2005: Genetics and animal domestication: new windows on an elusive process.

[3] Röhrs & Ebinger, 1998: Are zoo Przewalski horses domesticated horses? 

[4] Bunzel-Drüke, Finck, Kämmer, Luick, Reisinger, Riecken, Riedl, Scharf & Zimball: „Wilde Weiden: Praxisleitfaden für Ganzjahresbeweidung in Naturschutz und Landschaftsentwicklung“. 2011 

[5] Gaunitz et al.: Ancient genomes revisit the ancestry of domestic and Przewalski's horses. 2018. 

 

Bulls from Oostvaardersplassen and Slikken van Flakkee

The Oostvaardersplassen Heck cattle population has been covered multiple times on this blog. It is an interesting population because the cattle have been living there free of human influence since the 1980s and therefore would serve as an important case study for dedomestication if they would only be studied. 

The bull at 0:57 is very interesting. Not only do its horns face more forward than usual for Heck cattle (especially for the founding individuals of this population), but also its legs are longer and the trunk shorter and less massive. Also the skull is not as short as in usual Heck cattle. All in all it is more aurochs-like than usual Heck bulls. It is well possible that this is an actual result of natural selection, as the proportions and horn orientation are probably not the result of phenotypic plasticity. 

The bulls in this video are less long-legged, but their horn orientation is interesting as well. It approaches the aurochs-like 70-60° relative to the snout. When you look at older photos of Heck bulls from the OVP, you see bulls with much more upright and straightly curved horns. To me, this is another hint that evolution is at work at the OVP, and resulting in a regression towards wildtype-like phenotypes. 

The Heck cattle population in Slikken van Flakkee is interesting as well. As the cattle at OVP, they are free to breed for themselves without any artificial selection. This means there are some selective pressures working on the cattle. And just like the Heck cattle at OVP, they are more slender and more long-legged than usual Heck cattle. 

The video shows some bulls fighting. They have more aurochs-like proportions than usual Heck cattle, and also have a hump, which is rare in Heck cattle. I particularly like the bull at 3:53, the curly hair gives it a kind of wild appearance. 

Both the populations at OVP and Slikken van Flakkee are very interesting and valuable populations. A study could be done that documents the phenotypic changes in the herd over the decades in order to have an empirical support for the regression towards the wildtype. A pity that nobody that has the possibilities seems to be interested in such a study (dedomestication is definitely under-studied). 

A Tauros cow at Milovice, Czech Republic

As the Tauros Programme does not have a gallery or any other way of presenting their animals thoroughly on their website, finding photos of Tauros cattle always requires a google search. This time, I found a photo of a Tauros cow at Milovice, Czech Republic, from 2020 on Wikimedia Commons by Michal Köpping: 

Photo by Michal Köpping

The breed combination is a mystery as usual in Tauros cattle. But I suspect it has Maremmana ancestry because of the horns, perhaps mixed with Sayaguesa. Maybe it is Sayaguesa x (Highland x Maremmana), or something completely different. 

Regarding its qualification as a "breeding-back" cow, the trunk looks too long (consequently it is not as high-legged as it should be). The horn curvature is wrong as it has those outwards-curving horns of Maremmana or Highland. The colour would be OK if it would not have those tiny white spots on the belly. When not selected against, these tiny white spots can multiply in later generations and result in completely piebald offspring (this happened with some Heck cattle at Oostvaardersplassen). 

The deficiencies described here unfortunately go for quite a lot of Tauros cattle individuals, it would require strict selection to fix that. The Tauros Programme describes on their website that they will have a quantity building-phase first and then a quality building-phase, I do not know if this is still the plan of the project. 

Cattle, aurochs and brain volume

Domestication goes hand in hand with certain changes in morphology, physiology, genetics and other aspects. These changes are very similar in all domesticated mammals, and therefore called the “domestication syndrome”. Typical symptoms of the domestication syndrome are paedomorphy, changes in colour, endocrinology, physiology, a loss of genetic fitness, reduction in sexual dimorphism and also brain volume. Cattle are a prime example for the domestication syndrome. They are paedomorphic compared to the aurochs, changed in morphology, colouration, body size and show a reduced sexual dimorphism. Whether or not the brain volume also declined during domestication was not tested previously, but is very likely as this has been found in all other domesticated mammal species and cattle show all the other symptoms of the domestication syndrome as well. 

Recently, the brain volume of aurochs has been compared to cattle of multiple breeds using well-preserved skulls. The results have been presented in a paper by Balcarcel et al. 2021. They looked at the brain volume of aurochs and compared them to very derived (dairy and beef) cattle breeds, Chillingham and White park cattle, Lidia (= bullfighting cattle in the paper) and also Heck cattle. 

What they found is that domestic cattle do have smaller brains than the aurochs. Very derived breeds show the greatest reduction in brain size (from 30,6% to 24,9%), Chillingham and White park cattle 18,2% and Lidia only a 15,3% reduction. [1] The authors hypothesize that the less human contact the cattle have the larger is the brain volume. The reduction of brain size of Lidia is considerably lower not only than that of the other cattle breeds but also than that of other domesticated species (f.e. cats 24%, pigs 34%, dogs 29%) [1]. 

This supports my assumption that Lidia, Spanish fighting cattle, are the least derived taurine cattle breed on this world. Heck cattle showed no outstanding pattern in the study [1]. To me this is not surprising considering the breeds it descends from – some of them were derived breeds like Angeln or Murnau-Werdenfelser. 

 

What does this mean for “breeding-back”? For once, it confirms that it would be a waste if Spanish fighting cattle would not play a role in “breeding-back”, considering the obvious value of the breed (go here for my idea in which form the breed could contribute). It would be interesting to know the reduction of brain size of breeds heavily used in “breeding-back” such as Sayaguesa and other Iberian primitive breeds. Perhaps those are similar to Chillingham and White Park cattle. Podolian cattle (including Maremmana and Hungarian grey which are used in “breeding-back”) show a reduction of 21-31%. All in all I do not expect the other “breeding-back” cattle (that means Taurus, Tauros and Auerrind) to have a reduction of brain size as small as Lidia. Perhaps they would be closer to Chillingham cattle or somewhere between Chillingham cattle and more derived breeds, or maybe in the same range as Podolian cattle. It would have to be tested. 

Should brain size become another criterion for “breeding-back”? Well, theoretically, yes, it would be beneficial. But without having Lidia in the mix, which has a drastically smaller reduction of brain size than the other breeds, selection for larger brains would probably not be that fruitful since the difference between the other breeds is not that large. Apart from that, what makes it basically impossible, is the fact that the brain volume cannot be measured in the living animal. Thus, it is not possible to pick a bull with a particularly large brain and use it for breeding, as the large brain can only be diagnosed post mortem. At least I cannot think of a method to measure brain volume in a living cattle individual. 

 

An important question is whether the original brain size of the aurochs is a requirement for living in the wild. I do not think so. While there would certainly be a selective pressure towards larger brains in a cattle population in the wild, it is certainly not a vital requirement for the cattle to have the same encephalization as the aurochs in order to survive. This is proven by the many feral cattle populations on this world, which often descend from usual farm cattle. Furthermore, it is very likely that selection would lead to a secondary enlargement of the brain in a feral cattle population, as the dingo has a larger encephalization than (other) domestic dogs [2]. Thus, it is not really a problem that “breeding-back” cannot actively breed for larger brain size. I am still fond of the idea of practising selection on a neurologic/behavioural level though, by selecting those cattle that exhibit a wildtype-like behaviour (go here for the post). This might perhaps also influence brain size. 

By the way, I am credited for the silhouette of my aurochs cow drawing they use in the paper, which is the first time that my name is mentioned in a peer-reviewed scientific paper. 

Literature 

[1] Balcarcel et al.: Intensive human contact correlates with smaller brains: differential brain size reduction in cattle types. 2021.

[2] Smith et al.: Brain size/body weight in the dingo (Canis dingo): comparisons with domestic and wild canids. 2018. 

Another aurochs drawing

Recently I finished another aurochs drawing. I was inspired by the photo of a Pinzgauer bull for the pose. I did it with a ballpoint pen and coloured it in GIMP. 

The head and horns are based on the Berlin skull. The colour of the dorsal stripe is grey, as the historic text by Sigismund von Herberstein suggests. 

New photos of Auerrind bulls

Recently, some new photos of Auerrind bulls have been published on Facebook. 

Apollo (Maremmana x Watussi) © Felix Hohmeyer

This is Apollo, the Maremmana x Watussi cross bull. He is going to cover the herd of Sayaguesa cows and the Pajuna cow. I am looking forward to those crosses. 

© Felix Hohmeyer

© Felix Hohmeyer

This is Alvarez, the Sayaguesa x Watussi bull. His horns have a prime volume, they would need to curve more inwards however. This trait might be added by Chianina (some of the Chianina cows of the Auerrind project have inwards-curving horns), therefore I think (Sayaguesa x Watussi) x (Sayaguesa x Chianina) is the most promising combination of the whole project so far. The first bull of this combination has been born recently, as it was announced on Facebook. 

Benito (Sayaguesa x Maremmana) next to a Sayaguesa x Chianina cow © Claus Kropp

Benito, the Sayaguesa x Maremmana bull is going to cover the Sayaguesa x Chianina, Chianina and Sayaguesa x Grey cattle cows now. These combinations might be good as well, but there is the risk that the recessive dilution factors removing the red pigment in the coat that is found in Maremmana and Chianina might dilute the coat colour. 

New second-generation Auerrind calf born

Claus Kropp recently announced on Facebook that a new second-generation calf of the Auerrind project was born. It is of the combination (Sayaguesa x Watussi) x (Sayaguesa x Chianina), the - in my opinion - most promising combination of the project. It is a bull calf. 

Now the herd of the Sayaguesa x Chianina, pure Chianina and Sayaguesa x Grey cattle is being covered by the Maremmana x Sayaguesa bull Benito. 

Personally, I would have given Alvarez (the Sayaguesa x Watussi bull) several seasons to cover the Sayaguesa x Chianina cows, to produce more than just three of this combination. The larger the numbers, the larger the probability that there are calves that inherited the desired traits in the right combination (f.e. the large body size of Chianina in combination with the horn volume from Watussi and so on). But let's see how the three calves of this season will develop.