"No reason to doubt the Romanov DNA Testing"



T.J. Parsons, PhD

Former  Chief Scientist Armed Forces DNA Identification Laboratory


I am thoroughly unconvinced by the Knight study.  I note a few of the reasons below, and point out some fallacies relating to the paper that may in part have to do with the Stanford group’s lack of familiarity with how the Romanov DNA testing was conducted.

1.  The suggestion that the Stanford study served as any type of replication of the Romanov testing is wholly inaccurate.  They tested none of the Romanov material in question: the sample they tested was simply a finger from a relative of Alexandra's, not from the same environmental context as the Romanov remains.

2.  The suggestion that the "chain of custody" of this finger (residing in a museum after having been recovered from a pit, and identified by a family member) is superior to that of the Romanovs is not even slightly credible.  This old finger is put forward as a better reference sample than a modern day blood sample collected directly from Prince Phillip, Duke of Edinburgh, that was collected with thorough, formal chain of custody by the Forensic Science Service, a world renowned forensic laboratory.  As is mandatory in proper forensic DNA investigations, this blood sample was collected only after all testing on the Romanov remains had been completed.  The theory that the bones were contaminated by DNA that matches Prince Phillip, then, is not reasonable, especially given that this sequence is very unusual—-as Knight et al note, it has not been seen before.  However, I don't doubt the authenticity of the Stanford finger as coming from Alexandra's sister: a sequence matching Prince Phillip was obtained as a minor component of the DNA recovered by the Stanford team;  by far the best explanation for this is that it is the authentic mtDNA sequence of Alexandra's maternal line, and that most of the DNA recovered by the Stanford team was from a more recent contamination.  Such a result is hardly surprising given that the Stanford lab did not use facilities dedicated to contamination-free ancient DNA testing, but says nothing about the Romanov testing that was performed in qualified forensic laboratories.

3.  The Knight study fails to refer to the nuclear DNA results obtained by Gill et al.; these results were consistent with the putative Romanov family group (mother, father, offspring), and certainly could not be obtained by random contamination.  Recovery of nuclear DNA from old remains is difficult and uncertain, but has been demonstrated now in many laboratories, including our own.  For example:  our laboratory often has good success with old material recovered from temperate climates, but very poor success from warmer climates.  I understand the ground temperature in Siberia where the Romanov remains were buried is rather low... Our recent experiments published in Science (Hofretier et al, 2004) show that it is, in fact, quite reasonable to obtain 1200bp mtDNA fragments from old forensic bones recovered from cooler climates such as Alaska.  We were able to do this quite easily, even though just such a result is what Knight et al. claim as “impossible.”

4.  The Knight paper claims that the Gill et al. did no cloning: results of cloning experiments are clearly described in the Gill et al. paper.

5.  The Knight paper overlooks the fact that an exceptionally sensitive technique for recovering ancient DNA was used by Gill et al:  nested PCR.  None of the ancient DNA references that Knight et al hang so much weight on relate to the use of this technique.  The Forensic Science Service (Gill et al) has had more than a decade of experience using this exact method on highly degraded, old, and trace material, and I'm unaware that anyone in the forensic DNA community doubts their capabilities in this regard, or the authenticity of the Romanov testing.  Our recent experiments (Hofreiter et al., 2004) showed that even for fragmented DNA where standard techniques were unable to amplify 1200 bp fragments, mtDNA recovery was successful with the same nested PCR approach as was used by Gill et al, 1994.  The central argument of Knight et al—that 1200 bp fragments can’t be amplified from old bones—is therefore wrong on two counts:  1)  it isn’t true (as shown by our results on bones from Alaska), and 2)  the Gill et al nested PCR technique is not dependent on successful standard amplification of 1200bp fragments in the first place.

6.  The Knight paper does not reference our work (Ivanov et al) where we actually replicated testing on the Tsar, using standard ancient DNA techniques (rather than the nested PCR of Gill et al).  The bones were actually in rather good shape, although clearly old, and yielded degraded but amplifiable DNA very much as we would expect from our vast experience with such material.  The case was not extremely difficult, but the results clearly did not come from modern high molecular weight contaminants.  

A second critical portion of our testing that was also ignored by the Knight paper was that we also tested bones of the Tsar's brother Georgi.  These bones were officially exhumed from St. Peter and Paul's Cathedral specifically for our study long after the Gill et al study had been completed.  Our results from Georgi Romanov reproducibly (different extractions, different bone samples, multiple amplifications) indicated a different heteroplasmic ratio at position 16169, while matching at all other positions.  The DNA results from Georgi and Tsar Nicholas were not consistent with coming from the same individual, or the same contaminant, although they were highly consistent with coming from brothers from a maternal lineage in which heteroplasmy was segregating at position 16169.  

7.  Large portions of the Knight et al paper contained allegations of "inconsistencies" regarding the Romanov case that have nothing to do with the DNA science that was actually performed at Stanford, and that are outside of the expertise and experience of both the Stanford researchers and myself.  Sadly, I found that references in the paper put forward to substantiate the allegations, were simply references to more allegations.  There is virtually no theory of conspiracy that has not been alleged in relation to the politically-charged and sensationalized Romanov case, so such references would not be hard to assemble.  However, the DNA results speak for themselves:  when taken together, it is a fantastically unrealistic suggestion that these results could be the result of contamination, either random due to error, or intentional due to a vast conspiracy of all-knowing evil scientists.


Sincerely, Tom Parsons

Thomas J. Parsons, Ph.D.

Chief Scientist

Armed Forces DNA Identification Laboratory


  • P. Gill et al., Identification of the remains of the Romanov family by DNA analysis. Nature Genet., 6, 130-135 (1994).
  • Hofreiter, M, Loreille, O., Ferriola, D., and Parsons, T.J.  2004.  “Ongoing Controversy over Romanov Remains.”  Science 306:407-408.
  • Ivanov et al., Mitochondrial DNA heteroplasmy in the Grand Duke of Russia Georgij Romanov establishes the authenticity of the remains of Tsar Nicholas II.  Nature Genet., 12, 417-420 (1996).
  • A. Knight et al., Molecular, forensic and haplotypic inconsistencies regarding the identity of the Ekaterinaberg remains.  Ann. Hum. Biol.  Online prEview  (2004).