Genetic study on the English TB ancestry of some Arabian sire lines
The findings of recent genetic research by Dr. Barbara Wallner on the sire lines in Arabian horses is likely to cause a lot of stir in the Arabian horse world, from racing “industry” circles to purist ones.
The research points to, among other findings, English Thoroughbred ancestry in the sireline of the Saudi/Qatari stallion Amer. Amer was the most successful sire of “Arabian” racehorses of his generation. The information is part of a larger study titled “The horse Y chromosome as an informative marker for tracing sire lines”.
It shows the y chromosome in Amer’s offspring displaying the same unique genetic mutation that characterizes the offspring of the English Thoroughbred stallion Whalebone. This mutation is not present in the y chromosome of other English TB male descendants of the Darley Arabian. The Darley Arabian is the sireline for Whalebone, and the main foundation sireline of the English TB breed overall. This means that the face-saving argument of “both Amer and Whalebone/Darley trace to an Arabian horse sire line” does not stand.
Many purist breeders must feel so vindicated. I do, for one. Thank god for genetic advances, and for the freedom of expression in some countries that allow such studies to continue to be published.
Just to clarify – the study itself only indirectly points at Amer. It very clearly maps out quite a few of the TB y chromosome haplogroups, including the Tb-dW1 groups, which is unique to the Whalebone-Eclipse-Darley sire line.
What stirs the pot re: Amer is that one of the researchers involved with this study, Dr. Barbara Welland (who works with the University of Veterinary Medicine in Vienna in the Institute for Animal Breeding and Genetics), gave a report on the 22nd on the Arabian y chromosome haplogroups, and revealed that her team had found that Amer’s sireline traced back to TB-dW1.
There is a follow-up paper forthcoming, which should prove to be quite an interesting read.
oh, i need to clarify my entry then..
Right ! Cause Amer is not in this one study…only Miecznik has been tested Tb in this one.
Amélie, just to confirm, are you referring to the 1931 Fetysz ‘son’? You must know something more than I’m gleaning from the article re: the identities of the horses tested.
Follow-up:
Had a chance to look in the Supplementary Table 10, which has all of the sirelines listed. Amer was not tested in this study, but:
Miecznik = Tb-d. From what I can discern, though, is that it means he’s in the Darley Arabian patriline, which could mean anything from 1) ‘he’s got Thoroughbred in him’ to 2) ‘he’s a direct descendant of a horse that was bred of the Darley Arabian before the DA made it out of the desert’ to 3) ‘Miecznik and the Darley Arabian have a common patrilineal ancestor’. The table showing the tail male lines didn’t specify what variation of Tb-d he was.
Follow up #2
The problem, however, is that Miecznik tested Tb-d, and later in the table, his supposed sire, Fetysz, tested Ao, which is a completely different haplogroup. According to this study, Miecsznik’s sire is… probably not Fetysz? Curious.
^ A similar switcheroo apparently happened with Mameluck, the Hadban Enzahi SE ‘son.’ See #9 below.
Some other interesting findings below–
1) https://www.allbreedpedigree.com/amurath — T_nonTb-d ; the entire Bairactar sireline is implicated, as well. Which means… Bask… Arax… Gwarny… Penthagonn & Pamir… Pasat…
2) https://www.allbreedpedigree.com/famulus2 — another son of Fetysz, this one tested Ao
3) Nurreddin II tested Ao in his sireline
4) Faruk, son of Kuhailan Abu Urkub, tested Ao
5) https://www.allbreedpedigree.com/gazal — T_nonTb-d ;; Hamdani el Semri sire line
6) https://www.allbreedpedigree.com/halid – TM Sad OA // T_nonTb-d
7) https://www.allbreedpedigree.com/kemir+i — TM Souakim // T_nonTb-d // an Al Khamsa ancestral element Europa horse
8) Skowronek is apparely Ao / Arabian sireline
9) https://www.allbreedpedigree.com/mameluck3 — apparently tests T_nonTb-d, but his sireline horses (Hadban Enzahi, Nazeer, Mansour, Gamil Manial) all test Ao. Another son of Nazeer, Morafic, tests Ao.
10) https://www.allbreedpedigree.com/mersuch+i — T_nonTb-d
11) https://www.allbreedpedigree.com/netron — Netron ( < Patron, Aswan, Nazeer) tests Ao. Related to #9
12) https://www.allbreedpedigree.com/o+bajan8 – T_nonTb-d
13) Ofir / Kuhailan Haifi = Ao
14) https://www.allbreedpedigree.com/ruchan — Kurus OA // Ao // hujjah apparently in the info tab for Kurus OA
15) https://www.allbreedpedigree.com/ruzgar30 — Hilaluzzaman // Ao
—————————
The thing about T_nonTb-d is that, without further specification, it looks like this could be: [Tb-o] for either the Byerley Turk or the Goldolphin Arabian sire lines; [Tu] which clearly has Oriental ancestors but primarily denotes the Franches-Montagnes; and [Ta] which does also point to Arabians. [Tb] is also found in Akhal Tekes, and supposedly hearkens to the Turkoman horses.
But, I'm only barely making sense of the other Supplementary tables, so someone more versed will have to break down the Tb-d variants and the T_nonTb-d variants if they're shown in the other tables.
Follow up to #6 of my pending post:
It seems that all of the other horses in the Albatur sireline, and several other of his sons, test Ao, meaning that Halid is not of the Sad OA sireline, and a different horse is his sire.
Gee whiz o_o
After being giddy with excitement at first, I’m now a bit hesitant to draw strong conclusions from Supplementary Table 10, as some of the early Thoroughbreds (e.g. Florizel, Cade) are incorrectly assigned to Tb-d, when they should both read Tb_nonTb-d; comparing the data in the table to Figure 3 in the article, Florizel and Cade are both correctly assigned there, to Tb-oB1 and Tb-oB3b respectively. I would be very keen on a follow-up paper to this study for sure.
Moira,
yes you are correct about Miecznik being a Darley descendant and guess number 1 is correct (2 and 3 are to be forgotten since there is no Tb-d outside the Darley descent). Whether he is from the Mercury (W2) or Pot8os (W2) branch does not matter much IMO (although to be fair Pot8os is more likely since they are so very much more numerous – Flying Childer’s branch is very unlikely since it is to be found mostly in the US standarbreds). Indeed he is not a son of Fetysz.
This is correct also about Mameluck being not a son of his father.
And correct about Halid not being a son of his father either.
Those are the 3 pedigree errors we have here (for Arabians…there is a good amount in the Thoroughbreds too).
Right so beside this the T lines whether they are T non Tb-d (Tb-o, Tu or Ta) or Tb-d are all Turks. Sadly Ta is the perfect anagram for “Turkomans registered as Arabians”. As for Tu, Franches-Montagnes have either Thoroughbred sirelines or Anglo-Normand (and within Anglo-normand the “oriental” stallions used at Le Pin to build the breed were Turks : although they are very often badly referred to as Arabians).
Kate,
You should consider things the other way around ^^ Always stick to the table! Well actually yes it is the Figure 3 that is a bit misleading because it does mix good pedigrees and errors without pointing at all of them. This type of Figures are very often misleading in genetic studies…They are supposed to help summarize results but it does mix them up a little too much to be clear most of the time -_-‘
As for Cade (Godolphin Sireline) it is not very common so it is hard to say if it was actualy Tb or Tb non-Tb-d at first. I agree the Tb non Tb-d is more likely. And the Tb-d is probably a swap to the Darley line within the Hanoverian/Morgan bloodlines tested (with a pedigree mistakes within the studbooks between Protector and Zernbog or even later). This one really needs further testing.
As for Florizel, we know for fact that the Herod line is Tb non Tb-d because of extensive testing over Highflyer’s and Woodpecker’s branches. It is extremely likely that the Florizel line is not from Herod but another swap of pedigrees for a Darley line. Note that Diomed blood is to be found mostly within American breeds such as Quarter Horses, Saddlebreds, Standardbreds, Morgans and American Trotters…all of which breeds did not maintain studbooks early enough to hope for a perfect pedigree record.
You can also check both the original paper on : https://www.ncbi.nlm.nih.gov/pubmed/30988347
Always better to check from original material
All in all the study provides an enormous amount of info including much enlightenment on various levels : horse domestication, human selection over horse, breeds relationships, pre-DNA testing pedigree mistakes…I am still working on the outcomes 😮 May try for a blog entry if Edouard is ok ^^
Amélie, thank you for your response! I have been reading the original paper, and it was when I looked into the supplementary material in addition that I paused and went “Hmm.”
If I understand, the Florizel bloodline is being traced in this study to Diomed, through his on-paper descendants in non-Thoroughbred horses? I am trying to work out what the father/son pairings in the supplementary table mean (my assumption is that they aren’t testing the remains of eighteenth and nineteenth century horses, but rather testing horses that trace to various stallions in each individual lineage).
As for Cade, the fact that the supplementary data has him as Tb-d, but Precipitation etc as Tb_nonTb-d is a problem, as Cade is the tail male ancestor of all surviving Thoroughbred descendants of the Godolphin. I was a little surprised that the Godolphin himself was not tested, as we know exactly where his remains are, at Wandlebury.
Re the Turkoman thing, I was looking into that yesterday, and found a variety of European sources from the nineteenth century, which consistently claimed that the Turkoman horses had received at least two, possibly three, large scale infusions of Arabian blood. Armin Vámbéry (1865) Travels in Central Asia says “By origin the Turkoman horse is Arabian, for even at the present day those of the purest blood are known by the name Bedevi (Bedoueen)”, and another one of the sources, I can’t remember which but will look it up, claims that the Turkmen tribes repeatedly brought in Arabian horses, as they felt their own horses lost their quality after three or four generations without the Arabian influence. How reliable these sources are, I do not know, but I do know that following the Russian conquest of Turkmenistan, many of the tribes’ horses were confiscated or killed, to render the nomads weaker, so the breed has gone through a bottleneck of sorts in recent times.
The other thing that dawned on me yesterday, while comparing sources on Turkomans and Turks and Arabians, was that in the books I was reading, which went back to the 1500s, none of the authors knew about Turkoman horses until the 1800s, and that when Turkomans were included in encyclopaedias, they received a separate entry from the Turks, which were characterised as horses from Turkey. It took me a while to understand what was going on there, until I noticed that the sources on the Turkomans all placed them east of Persia, which was at odds with the Ottoman empire during this time period.
Taking into account the location of Turkmenistan, and the cities mentioned by travellers amongst the Teke, Yomud and other Turkmen tribes, it also crossed my mind that the chances of Turkoman horses being in Syria during the eighteenth and nineteenth century craze for Oriental horses in Europe were very slim, as to reach Syria they would have to cross the mountain ranges of Iran, come over the Zagros into modern Iraq, and then cross the Tigris and Euphrates. Which is vastly more travelling through vastly more difficult country than the migrations of the Bedouins. Not impossible, but it seems unlikely.
Which led me back to what haplogroup T means, why it is present in Arabians, Tekes, and Thoroughbreds. My current hypothesis, pending further reading of journal articles, histories, and travelogues, is that T is an Arabian haplotype. It just happens to be the haplotype of the horses found in Syria prior to the northwards migrations of the Anazah and Shammar from the Nejd, in response to the rise of the al-Sauds. Ao, my hypothesis goes, is the haplotype found more commonly in the Anazah horses (the fabled Nejdis). This would explain why T is found in breeds that, according to paper sources, have Arabian ancestry that predates the 1800s, such as the Thoroughbred and the Akhal-Teke, but why Ao is found in Arabians that trace to horses taken from the Anazah or from the Wahhabis (such as the founding of Muhammad Ali’s stud in Egypt).
As hypotheses go, it’s imperfect – the Darley Arabian is connected to the Fedaan, who are an Anazah tribe, if I remember correctly; he is also a Ma’naqi, according to a letter written by Thomas Darley in December 1703: “He is about fifteen hands high, of the most esteemed race among the Arabs, both by sire and dam, and the name is called Mannicka”. Which is interesting in light of the assertion that the Ma’naqi was cross-bred with Turkish stallions in the 1700s! And there are certainly other options to explain haplogroup T in Arabian horses. But I think genetic studies are like statistics: it’s all very well to have a quantitative analysis, but without the context, the qualitative analysis, the numbers aren’t very useful. Conversely, the genetics help to keep the history grounded, and allow us to sift through all the narratives surrounding events to find the kernels of truth/reality within them.
Anyway, I will no doubt pop back in here at some point with a newly changed opinion, after doing more reading, but that is my current train of thought, for whatever it is worth.
Okay: Time for some background which will explain things.. Here’s the deal- Arabians, Turkomans, and Caspian ponies( horses) all three descend from the ancestral Afro-Turkish horse. The Afro Turkish horse was one of the 4 types of horses that survived the last ice age and it appears also survived early domestication. The other three were the Tarpan basically extinct nowadays, the proto warmblood or mossback( really called that ) and the Draft subspecies. The three non Afro Turkics have been bred together in differing combinations since domestication to give us our current modern breeds. The only horses that are descended completely or even mostly from one of the four founder types,Tarpan, proto warmblood, Draft, and Afro Turkick, are Asil Arabians. So it makes sense to reason that evidence of common ancestry genetic markers will surface. This also explains why Many Iberian horses show Male Arabian Y chromosome evidence in spite of Andalusian breeders fatuous claims that Barbs and Andalusians are not related to Asil Arabs. For that assertion to be true the 700 year long plus arab occupation of Spain would have to have never happened..The Adalusian beeeders mental gymnastics are more likely the result of misplaced Spanish nationalism.
It also helps to remember that Thoroughbreds were built by combining 16th century Hobby mares, with Barbs, Turkomans, and a few Arabians. Both Barbs and Hobbys contain draft blood. As far as we know Turkomans had no draft blood nor do Asil Arabians.
best
Bruce Peek
I haven’t yet fully examined the study or supplemental material. So, at the risk of sounding like an idiot, I’ll put this thought out there anyway.
Did the researchers sample Akhal-Teke horses rather than Turkomans from Iran? If so, there may be the possibility of a Thoroughbred sireline in the A-T breed.
Convoluted!
Jenny, the researchers cite a 2017 paper by Wallner et al. “Y Chromosome Uncovers the Recent Oriental Origin of Modern Stallions”, Current Biology 27, regarding the Akhal-Teke origin of the Thoroughbred. In the 2019 paper, they say “Previously, we identified HG Tb as a signature of the Turkoman horse, an ancient horse population from the steppes of central Asia”, in reference to the 2017 paper.
The relevant section in the 2017 paper is this: “To identify the origin of Tb, we extended our samples by including the Akhal-Teke, the remnant of the Turkoman horse, and found that Tb is the most frequent HT among 78 Akhal-Teke males (81%, Figure 4B). Thus, Tb is likely of Turkoman origin and spread widely by English Thoroughbred stallions. Additionally, the presence of Tb in many European breeds with no documented influence of English Thoroughbred stallions shows the influence of Turkoman stallions, independent of the English Thoroughbred.”
As a historian, I convince I find that unconvincing, particularly given the history of the Akhal-Teke, who had an open stud book until 1932. And yes, I agree, it is entirely possible that Tb could be the result of Thoroughbred introgression into the Akhal-Teke breed; the stallion Absent, who won Olympic gold in dressage in 1960, has a line to the Thoroughbred stallion Fortingbrass, son of Fontenoy (Faugh-a-Ballagh x Fete), and there were other documented Thoroughbred stallions used in the breed in the early part of the twentieth century.
Exactly, I second Jenny’s point. English TB is one of the components of the modern Akhal Teke breed. So it’s indeed possible that Akhal Tekes tested trace to English TB in the male line, and that these in turn trace to Turkoman horses in the male line.
“If I understand, the Florizel bloodline is being traced in this study to Diomed, through his on-paper descendants in non-Thoroughbred horses? I am trying to work out what the father/son pairings in the supplementary table mean (my assumption is that they aren’t testing the remains of eighteenth and nineteenth century horses, but rather testing horses that trace to various stallions in each individual lineage).”
This is absolutely correct.
“As for Cade, the fact that the supplementary data has him as Tb-d, but Precipitation etc as Tb_nonTb-d is a problem, as Cade is the tail male ancestor of all surviving Thoroughbred descendants of the Godolphin. I was a little surprised that the Godolphin himself was not tested, as we know exactly where his remains are, at Wandlebury.”
Yes, the Godolphin line is pretty much split into two hypothesis which are hard to sort. The Tb non Tb-d is more likely since only the Zernrborg lines appears to be Tb-d. I guess they had to make a choice and picked Tb-d. My guess is that there’s been a swap for the Darley line after Protector with good candidates being Adeptus, Devil’s Own or Goldschaum sireline. The TB side of the test always shows Tb non Tb-d anyway. As for testing from bone remains it was not included in this study. It does require a very different process from testing from live animals. Most likely there will be more focused studies in the future regarding the Godolphin sireline which will sort out which side of the family tree is correct.
Making a separate post for the Turk thing ^^
Merci, Amélie! That clears up a good deal; I do hope that at some point the Godolphin’s remains can be tested, as I am really interested in his sire line. (Sentimental reasons, blame it on Marguerite Henry’s King of the Wind, even though it perpetuates Eugène Sue’s fictional romance.) I am looking forward to your post on the Turk/Turkoman thing!
Alright…so let’s start first with the European literature from the 19th century.
I have been also reading a lot of it and I can say that you will find many many examples in the Studbooks and in general texts of confusion between Arabians and Turks.
You can also find examples that there was exchanges and breeding of both breeds in both areas (meaning inside the Arabian Penninsula and outside). In Damoiseau’s writing for example you can read that the Fid’ans were also breeding Turkomans horses alongside Arabians. The thing is…as much as it was very clear for the local people…it was not for the Europeans most of the time. These guys were buying remount stallions…they picked what looked good to them without much consideration for “Asil” status at least until the 1870’s/1880’s.
If you ask an Arab man if the Arabian is the father of all horses including Turk horses he will tell you yes…and if you ask the same to a Turkish person about the Turk horses being the father of all horses including Arabians he will tell you yes as well -_-‘ So literature is sadly not much of an help.
This is when modern genetics comes into play…
Thanks to our ability to build nice tools from algorithms…the main purpose of the above mentioned study was to determine scientifically the family tree of our modern sire lines via the mutation rate of the Y chromosome. The same has been done several years ago for tail females via MtDNA which enables us to know that all tail females in horses are pre-domestic occurrences and thus cannot give us a lot of information regarding horse breeding or selection by humans. The tail males however are providing us with a very different story. As you can see on figure 4 and for the sake of comparison will add the findings regarding tail females.
We do now know that :
– The last common maternal ancestor to all horses is about 157 000 years old before the beginning of the latest ace age after which maternal lines starts to split.
– The last common paternal ancestor to all horses is about 77 000 years old after the beginning of the lastest ace age after which paternal lines starts to split.
Mutations rates for sirelines in horses is estimated to be around 10 generations on a time lapse of about 400 years (with current technology in this study)
– There is no common maternal ancestor to all the domestic horses (horses minus Przewalskii) since the F Haplogroup is about 11 000 years old (about at the end of the latest ice age) and some domestic breeds have been found with older Haplogroups. This is a very good indicator to support the theory of restocking from wild mares during domestication and semi-feral breeding. It does also supports the latest discoveries that Przewalskii horses are also descent of domesticated horses bred by the Botaï (making them not actual “feral” horses).
– The latest common paternal ancestor to all the domestic horses (horses minus Przewalskii) is about 17 000 years old. This does support the latest researches that from all the sire lines domesticated by humans only one survived to this present day. The undergoing researches regarding genetic “disposition factors” in correlation to tail males on older remains is to be expected quite interesting as a side note.
– About 3.500 years ago the common paternal ancestor to all domestic horses sire line had been divided into two different tail males. The North East Asian group O including Yakut and Mongolian horses and the common paternal ancestor to all the West domestic horses. This does support the fact that at least 3.500 years ago humans were actively breeding horses and were able to support selective decisions. As a side note these tail males have been found to carry specific genes enabling better fat stocking among other things. It does also gives a first indication on a geographic location for male domestication.
– About 2000 years ago the common paternal ancestor to all West domestic horses sire line had been divided into two different tail males : The Northern West domestic horse which included ancestors of northern draft types of all sizes (ancestors of Icelandic, Shetlands, Swedish Drafts etc.) and the Southern “Crown” domestic horse.
Regarding the first group you will find interesting correlations to gaited breeds but moreover animals historically selected for draft use in accordance to archaeological findings in Northern Europe supporting it.
Interesting enough these breeds are bearing a strong connection with Celtic and German populations considering both the geographic localisation and time frame.
The second group called the “Crown” is what is of most interest to us since it does include our famous “orientals”.
It is to be noted however that every one of these major groups are including at least one Asian bred (by Asian I mean east of the Caspian sea) which does reinforce the above mentioned theory of a Asian domestication for male lineages.
– Right…so what about the Crown now…The Crown has been fully divided now into 3 clades : T (Turks), H (Iberian/North African) and A (Arabians). What you have to understand here is that (again the figure might be misleading here) is that one is not the ancestor of the others. They were all created from that same roots and arose pretty much at the same time over a VERY short amount of generations (3/4 – symbolized by the grey rounds on the figure). What it does mean is that the latest common ancestors to all three was at best born 1700 years ago.
Now…if you consider human history where were we 1700 years ago? And who on earth had the power to create such specific clades in such a short amount of time. There is only 2 answers : Persians (under Sassanid rule) for the Turkomans (T) and Romans (for the North African (H) and South European/Levant clades (A)).
Note also that we know from prior studies on tail males that there was a small amount of exchange between both mostly from T into A (but not the other way around). Which does also support the fact that T horses were to be found and bought in the Levant.
So after all most modern horses have a Roman ancestor (pretty much like humans which does makes sense)…
For the sake of it the Ta group (which is I am sorry again is absolutely not closely related to A) is about 900 years old which is consistent with the rise of the Turkish Empire…if you need even more clues of their provenance ^^
The most interesting part of the A group story is how first “waves” were introduced very early on in West European countries (both in Ad and Ao) which strongly indicates some sort of breeding program on a very large scale by the Romans. This is how we can find Arabian sirelines now in breeds considered as “old European” types.
It is also to be noted that the most recent “wave” of Ao is from a sireline who is about 1200 to 1300 years old being consistent with the rise of the Arab culture.
Whereas on the other hand recent T lines “waves” are at best 600 years old but more often around 200 years old, thus matching absolutely perfectly to creation of the Thoroughbred bred.
Regarding the H sireline, modern waves are about 500 years old so they are consistent with the ruling of Northern Africa by the Ottoman Empire probably enabling trading of North African horses with West European countries better.
i think you should turn this comment into a separate blog entry Amelie
That was the first idea yes ^^ Will put this with general “mistakes” findings so far into one
Once again, thank you, Amélie, for a very clear, informative and thorough response. I am looking forward to your blog entry.
I agree that 19th century European literature is not the most reliable of sources, and in the older works there is sooo much confusion about Turks and Arabians, which Arabians often assuming near mythical status. Re the Fedaan breeding Turkoman horses as well, that is very interesting, and I will read Damoiseau this weekend. As for the lack of asil considerations in the horses selected, I waiver there, as there is a transcribed hujjah in Thomas Pennant’s 1776 “British Zoology”; the hujjah is dated 1722, and was explicitly acquired because the horse’s purchaser wanted to be certain of the horse’s authenticity. Of course, this does not mean that every European went to the desert in search of genuine and pure Arabians, nor does it mean that the Darley and Godolphin Arabians were themselves necessarily asil. This is mostly me quibbling on a point! And you are quite right that the army remount expeditions just wanted Arab horses, without regard for purity of blood.
I also am in complete concord that T and Ao are not closely related; I suppose my phrasing was misleading when I said that T was (in my hypothesis) an Arabian haplotype. What I meant was that T could possibly be a haplotype of the horses found in the northern parts of Arabia Deserta, regardless of its unrelatedness to Ao, because science and genetics do not make an Arabian, but a cultural definition does. (I think genetic tests are great to have as parentage verification now, and to pick out accidental identity-swaps and deliberate fraud where we have records to compare them to, but I think applying them retroactively to horses from the early 1700s and drawing firm conclusions is … I don’t know, I’m still searching for the right way to express this, but it makes me uncomfortable, by applying anachronistic criteria and terms and standards, not to mention ones that were not relevant to the cultural definition of an Arab horse, at least as I understand it.)
Anyway, to get back to the point I was trying to clarify about T and Ao, since the Arabians of the Middle East have an impressive diversity of matrilines, why shouldn’t they also have had two patrilines that may or may not have fallen, roughly, into northern horses and southern horses? This is a genuine question, as it is the hypothesis I am testing out by reading as much as I can at the moment, so if there is a good refutation, then I don’t mind if it is refuted.
Regarding the origins of T, H and A, that’s a very good point about the empires of the ancient world. My knee-jerk reflex was “But the Romans were terrible cavalry-men and farmed that responsibility out to their allies”, because I have focused on late Republican Rome for the most part, but once I looked at the third and fourth century Imperial army composition, I discovered that Roman cavalry had improved quite a bit. The journal articles keep saying as well – and this is of interest! – that under the Roman Empire at this time, there were Imperial stud farms in Spain, Thrace and Cappadocia. The horses in Spain would likely have been a mixture of the local Iberian landraces, which were prized even then, some of the famous Libyan horses (Mauretania had one of the largest cavalry forces in the Imperial period, probably because of the neighbouring Garamantes), and presumably some of the Gallic horses, but I am speculating. Thracian studs may well have drawn on the horses of the steppe tribes, as well as the local horses, which had been famed for speed since Homeric times (Thracian mares were said to be impregnated by the north wind).
As for the Cappadocian studs, the region was known in antiquity as the land of beautiful horses, and Cappadocia was another Roman province which fielded a very large cavalry force, because of its proximity to the steppes, and to the Parthians and Sasanians. Of course, Anatolia, the Levant, Mesopotamia and Persia have a very long tradition of horse warfare, chariots, cavalry, etc, going back to the Bronze Age, and I spent two hours last night going through article after article talking about battles and cavalry tactics, etc. It was good reading.
I also had a moment when I learned that Imru’ al-Qays may have met with Justinian, who was very interested in Arabia, particularly southern Arabia, mostly for its value as a thorn in the flesh of Khosrau. He even sent a gift of Cappadocian horses to Himyar, so, yes, Roman horses definitely found their way into Arabia, as did, I imagine, Persian horses.
Re A sire lines finding their way into West Europe, there is a rather old (1911) description of equine remains at a Roman fort in Britain, which included one specimen with a head said to be more attractive than the finest Arabian. This was taken as evidence at the time for the mobility of North African and Levantine horses around the Roman empire; it would be interesting to find out what has happened to these remains since and test them too. (I am aaaalll for more genetic tests to reconstruct movements of horses.)
The waves of T lines 600 years ago are interesting, as those nineteenth century European sources claimed that Timur/Tamurlane was responsible for distributing some four thousand Arabians (mares in some sources, no sex specified in others) amongst the Turkmen tribes. This tallies fairly well with the DNA evidence, I think, not so much regarding the specifics of this story (which I have my doubts about), but more along the lines of the first wave of T coinciding, roughly, with the incursions of Timur, who (allegedly) had an interest in moving horses around as breeding stock.
So there you have a sort of stream of consciousness from me, and I would love to write more, but I do have to step out now for the morning. I am really enjoying this discussion, and am keenly anticipating your blog entry, so that the learning and debate can continue.
Kate, you wrote: “Since the Arabians of the Middle East have an impressive diversity of matrilines, why shouldn’t they also have had two patrilines that may or may not have fallen, roughly, into northern horses and southern horses? This is a genuine question.”
Could that be the hitherto unsubstantiated Raswan story of some of the ‘Anazah having migrated northwards in the 1700s and bred their asil mares to Turkmen stallions? Could it be that Raswan was credible on this?
That is a really interesting idea, Edouard. That could match – Amélie mentioned Damoiseau saying that the Fid’ans bred Turkomans as well as Arabians, which is the tribe the Darley Arabian is associated with (I’m assuming Fid’an = Fedaan); the Darley Arabian’s sire and dam are of the strain “called Mannicka”; there is a 1722 hujjah signed at Acre for a horse who is a “Monaki Shaduki”; and the Ma’naqi is the suspect strain according to Raswan, isn’t it?
I love all of this, but Charles and I came to a conclusion some time ago: we were breeding from horses that were representative of what the bedouin of the greater Arabian deserts considered asil at the turn of the 19th-20th century, and if we could hold that type, that was good enough for us.
Of course, Jeanne, I don’t think there is another viable alternative to that conclusion.
Kate, yes, the Raswan story is that some ‘Anazah Bedouins (the Salqa of the Ananzah is one tribe he mentions) bred their asil Ma’naqi mares to Turkmen stallions. I never gave credence to that story, but there may be something there in the light of the genetic findings.
Does anyone have the original version of what Raswan said about the Turkoman blood? Would love to see it.
Getting back to Miecznik — obviously Miecznik himself has been dead for years and was not tested. His sire line, as I understand it, exists today through two of his sons: Aquinor and Doktryner. Were multipile horses from the Aquinor and Doktryner sire lines tested, and do these two lines both test Tb-d?
Nobody has commented on the *Nureddin II sire line testing Ao? How many branches of that sire line exist?
I’d be willing to bet its probably in one of the Index books. Can’t guess which one though.
best
Bruce Peek
One source of Raswan on the Ma’naqi and the Turkoman is his article in the March/April 1942 edition of Western Horseman “Arabian horses in relation to Turks and palominos”. In it, he says:
“SHAYKH Muhammad’ Abd Ar-Rahman At-Tay, whom I visited last in 1928 in northern Mesopotamia told me that his MUNIQI HADRUJ, racing type Arabians, came from a cross with a stallion that one of his forefathers brought from Asterabad by way of Bushire, Persian Gulf, Basra. A Turcoman on his pilgrimage to the Holy City had visited his tribe and became an inseparable friend of the old Sheykh. Returning to the Caspian Sea the Turcoman persuaded his Bedouin friend to join him and they left with a number of slaves, gazelle hounds, falcons and Arabian horses. Two years later the Tayi chieftain returned with his entourage and brought a number of Turcoman horses, among them a splendid sire that was bred to a MUNIQIYAH mare.”
Kate, what a coincidence that you were writing down the quote from Raswan about Muhammad al-Abd al-Rahman of Tayy at the same time I was writing about the Saqlawis of his brother Talal al-Abd Rahman..
Thank you, Kate!
Regarding Raswan, the Muniqis and Turcoman issue, I think there was a Khamsat article many years ago by the Dirks on this topic. They debunked this connection of Raswan’s by talking to the Tai directly about it. Raswan misunderstood the story on this topic if I remember correctly.
Kate : Regarding the origins of the T, H and A. If I understand correctly there seems to be much puzzlement as to how some of the European Pony based breeds came up with A sire lines.. You are quite correct about the Romans being terrible cavalrymen. The A sire line got into the European stock much earlier than the late Roman Imperial period. The Phoenecians started selling horses along their coastal trading posts all the way out to and around the straight of Gibralter, and north along the coast of France, across into Britain and even southern Ireland. They were exchanged for Tin- incredibly necessary to make Bronze.. The first crosses of horses from todays Lebanon onto the pony breeds and light farm horses gave piebald and spotted coat patterns and some gaited horses too. Later the crossbreds were blended into the British and Irish Hobbys. The Hobbys after the late middle ages were used to form the Dam lines of the British Royal mares- the early distaff side of the Thoroughbred.
Also the historical record does indeed confirm that Romans- Eastern Empire variety by this time,used Arab Nomads as allies in their never ending wars against the Sasanids or Persians who also hired Nomadic horsemen from greater Arabia as allies too. So it makes sense that their horses would have to at least a small extent found their way into a Roman remount breeding system. Why have a large breeding system you ask in light of the Romans being terrible Cavalrymen- because the Roman road system needed horses to move goods. And a well maintained road system meant the economy, on the scale of an empire could respond to supply and demand.
best
Bruce Peek
@Lyman – I suspect that would be worth a reprint, if it’s from as long ago as you’re saying!
Can someone please explain the genetic description of T_nonTb-d and how it pertains to the Bairactair sire line? I cannot find that information anywhere in the Figure tables!