Recent Posts

Read a Random Post



« | Main | »

Romanov Cocktail

By Daniel Hubbard | August 21, 2011

On July 19, 1918 at about 2 am, a firing squad left a puzzle for DNA to solve. Over the previous eighteen months, Russia had undergone two revolutions, capitulated to Germany and suffered an ongoing civil war between communist and anti-communist armies. The  first revolution led to the Tsar’s abdication. The second revolution and the civil war that followed led to the hurried murder of the Tsar and his family by the Bolsheviks, as a detachment of the anti-Bolshevik White Army approached. The original story is that after the family was shot and bayoneted by firing squad, the bodies were taken to an iron mine in the forest and burned.

There were other stories that circulated. One was that the bodies were not burned at the mine, just placed there and the next morning when rumors started to fly, it was decided that too many people knew too much. So the bodies were put back on the truck, taken farther into the forest and when the truck broke down, quickly buried in two shallow graves. A diary published in 1926 but then suppressed by Soviet authorities told a similar story. An account from the late 1920s by a Soviet poet, Vladimir Mayakovsky, a man who was trusted and allowed to travel freely, claimed that he had seen the burial site and that it was about nine miles outside Yekaterinburg down a specific abandoned road. An old police investigation of unexplained burials showed a small plot of forest covered with railroad ties.

When the amateur investigator who assembled this evidence got permission in the 1970s to visit the area under false pretenses, he secretly searched for and found the site and he found skulls with bullet holes. Having no way to proceed further, he eventually reburied the skulls. In the last days of the Soviet Union, the researcher, Alexander Avdonin, felt safe in using official channels to try to start an investigation into the site he’d found.

Eleven people had been executed that morning, the Tsar and his family and four servants. The grave proved to contain the remains of nine people not eleven. More work would be needed.

What’s in it for the Genealogist?

When I read about this research, I was thinking about genetic genealogy and the lessons just kept popping out at me. There really are some lessons to be learned here and in what is to come in the story. All sorts of evidence was considered. The man whose clandestine digging turned up the bones began with local stories he had heard as a child. A genealogist might think of them as oral history interviews, even if Avdovin wouldn’t have thought of it that way in his childhood. Moral: When you first start investigating your family history, interview relatives. Get the oral history.

There was evidence that all eleven people had been burned beyond recovery. There was even an early official visit to the iron mine during the civil war by a White Russian investigator that tuned up gems, scraps of clothing and bones in the mine. It looked open and shut. The Romanovs were there but were nothing more than ash and that was the end of the story. Moral: The obvious answer that comes early on isn’t necessarily the correct answer.

Nevertheless, persistent investigating turned up documents with different accounts of what had happened. Moral: Gather all the evidence you can, worry about contradictions later. Don’t disregard data that does not confirm what you think you know.

Once the bones where exhumed, many different angles were explored. The skeletons were compared to photographs to determine if the size of the bones and the facial characteristics matched the Romanovs. Moral: We still have not gotten to DNA analysis. A solid “paper trail” is needed to make DNA testing worthwhile and understandable once it is done.

Nuclear DNA was compared and the remains of an adult man and an adult woman were consistent with that man and that woman being the parents of the children whose remains were found. Moral: Until recently, testing nuclear DNA (other than the Y chromosome) was restricted to analyzing crime scenes and determining very close relationships like parent-child. Recently, it has come to be used for testing the probability of more distant relationships. It is interesting to note that part of testing the hypothesis that the remains belonged to the Tsar and part of his family was a check for internal consistency. If the remains were not even from a family, there would not be much point in trying to prove they came from the Tsar’s family.

Mitochondrial DNA from the woman matched the mitochondrial DNA of the children. Moral: Another internal consistency check but also notice that mitochondrial DNA testing has been used for a while now in genealogy and that it is the mother whose mitochondrial DNA should match her children’s mitochondrial DNA. The comparison will work for any child, son or daughter, as well as for any child of a daughter. The father’s mtDNA isn’t relevant.

To compare with DNA samples from other people, a blood sample was taken from Prince Philip who’s mother’s mother’s mother was also the Tsarina’s mother. Because they were related through purely mother-daughter relationships up to Prince Philip, his mitochondrial DNA should match the remains of the mother of the family in the grave. They matched. Moral: The mtDNA of the prince was, in a strict sense, used to determine the mtDNA of his well documented great-grandmother. Bones of a woman suspected to be a daughter of his great-grandmother should show the same mtDNA as the Tsarina’s mother if there is any chance that the bones come from the Tsarina. A DNA sample and a clear paper trail were used to check a case where the chain of evidence was much weaker.

What about finding the remains of nine people when there should have been eleven? The hunt for two more skeletons continued until they were found years later not far away from the larger grave. Moral: Try to explain discrepancies. Your hypothesis will hold up better if there is an explanation for the discrepancy or ideally, if further investigation shows that there is no discrepancy.

The two new skeletons were tested. Their nuclear DNA was consistent with their being a son and a daughter in the family from the first grave. The boy’s yDNA was compared to the yDNA of the father and they matched. Moral: The test that is most useful in genealogy can only be applied to males and can only compare males who are related to each other only through men. Here that is easy, a son and a father should match.* Be careful though. As useful as it is, a yDNA test only supports or refutes descent from a common male ancestor along an all-male line. The farther back the common male ancestor, the greater the chance of mutations that give some idea of the time scale involved but those mutations don’t come so often that the test can determine exactly how many generations separate a pair of test subjects from their common ancestor. A yDNA test alone could not tell the difference between the father-son relationship of the Tsar to his heir and the uncle-nephew relationship between the Tsar and a brother’s son. It takes the oral history, the paper trail and in this case the other DNA tests to produce a preponderance of evidence that the boy was the Tsar’s son Alexei.

For good measure, the Tsar’s brother was exhumed and tested. He died of natural causes and was given a well-documented royal burial long before the revolutions. As brothers, their mtDNA should match. It did. In fact, both samples shared a condition called heteroplasmy. The quick and dirty explanation of that term is that though one human cell will contain only one copy of its nuclear DNA, it will contain hundreds and hundreds of copies of the mtDNA. That makes it possible for some copies of mtDNA from the same cell to show a mutation and the others to lack the mutation. It is a rare condition and the mtDNA sample from the Tsar’s brother and the bones not only both showed heteroplasmy but showed the same mutation. Moral: Some similarities and differences between DNA samples are much more significant than others, and you have to read up on whatever is relevant to your problem to understand what it means.

Some other DNA tests were also performed. Perhaps the most interesting were tests of a shirt and a handkerchief stained with the Tsar’s blood after an assassination attempt in Japan in 1891. The handkerchief, kept in a Japanese museum did not match the bones. The shirt, which had been hidden away in a Russian archive, did match. Moral: We aren’t likely to find blood samples of ancestors stored away like that but we do need to think about provenance, who controlled the evidence and how it was handled. In this case, the samples were obviously not cared for with DNA testing in mind. One contradicted the hypothesis that the bones came from the Tsar but the chance of contamination with other DNA is so high that the mismatch is not strong evidence. The other sample, which had been left alone for decades, showed  a match. In more typical cases for a genealogist we have to question things such as if the hand written records we have found are original, or copies made by hand from a decaying original that no longer exists. The farther we are from the original, the greater the risk of errors being introduced. We have to ask if crazy Aunt Mildred ever owned the family Bible and if she had, might she have “improved” the entries?

Any sufficiently complex hypothesis will need quite a mix of evidence and analysis to back it up. As much data as possible of as many different types needs to be found. Different types of data have different strengths and different ways of supporting each other. Here oral history and paper trails both clear and murky come into play. Different types of DNA testing triangulate from father to son, mother to children, parents to offspring, identifiable people to mysterious bones.  Almost all point to the same conclusion or have reasonable explanations when they do not. This research really did involve a remarkable cocktail of evidence.


* a boy and his paternal grandfather should also match but there is no reason a boy and his maternal grandfather should.

Twitter It!

Topics: Genealogy, Methods | 4 Comments »

Twitter It!

4 Responses to “Romanov Cocktail”

  1. MN Family Historian Says:
    August 21st, 2011 at 12:33 pm

    Great case study how traditional genealogy and DNA testing can be used in a complementary manner. However, I do have one quibble. You wrote “This kind of testing isn’t useful for genealogy”. That’s no longer true. Autosomal DNA testing is probably more useful than Y-DNA or mtDNA now. It also can’t ‘prove’ a specific relationship, but it can estimate the degree of relationship. Plus, it can match relatives from any branch of a family tree. We’re no longer limited to the straight paternal and maternal lines.

  2. Greta Koehl Says:
    August 21st, 2011 at 9:00 pm

    This investigation and the fascinating path it took is a great choice as a parallel to the use of genealogical evidence – and I will always remember the phrase “cocktail of evidence.”

  3. Daniel Hubbard Says:
    August 21st, 2011 at 9:26 pm

    MN… , Thanks for your comment. Yes, it is true that I shouldn’t have written that autosomal DNA (the nuclear DNA that gets shuffled between generations) isn’t useful for genealogy. That has changed. On the other hand, I’m not sure that it is more useful. It certainly has different use cases but it also has different limitations. Since people don’t always read the comments, I’ve also updated the post.

  4. Daniel Hubbard Says:
    August 21st, 2011 at 10:18 pm

    Thanks Greta. Or should I say “bottoms up?”