Jefferson DNA Data

There was a long controversy regarding whether or not Thomas Jefferson could have fathered any sons by Sally Hemings. To attempt to answer this question, Y-chromosomal DNA was taken from living male claimed-descendants of Jefferson and Hemings.

Hemings Descendants for Testing
It was possible to locate male-line descendants of Sally Hemings' youngest son Eston Hemings. In addition, male-line descendants of Thomas Woodson were also located. Woodson's descendants maintain that he was the eldest son of Jefferson and Hemings, although there was no firm evidence to support this claim. Hemings' eldest son Beverly Hemings had no male descendents. Male-line descendants of Hemings' other son Madison Hemings have subsequently been located, but have not been tested. Shown in the figure below is the lineage of the one male-line descendant (H21) of Eston Hemings and the five male-line descendants (W55, W56, W69, W70, and W61) of Thomas Woodson.

Positive Controls
Thomas Jefferson did not have a surname-bearing, proven descendant survive as a positive control. Therefore, it was necessary to locate male-line descendants of Thomas Jefferson's paternal uncle, Field Jefferson. Five such descendants (J41, J42, J47, J49, and J50) were located and their DNA was analyzed.

Some considered Thomas Jefferson's nephews by his sister to be possible fathers of Hemming's children. It was also possible to locate three male-line descendants (C27, C29, and C31) of Samuel and Peter Carr -- the sons of Thomas Jefferson's sister suspected of being the fathers.

K2 haplogroup
In addition to the Hemings controversy, Jefferson's DNA is also of interest as it bears a rare haplogroup for Europeans. Jefferson's fame motivated further study of the K2 haplogroup, and its origins in the UK. The K2 haplogroup is most common in various locations in the middle east, so it is of some interest how small numbers of K2 individuals came to reside in the UK and elsewhere in Europe. A study by King et al in 2007 of 85 randomly selected men from the UK with the surname Jefferson found that they divided into a number of different haplogroups, suggesting the surname originated independently multiple times from different unrelated founders. However, two out of the 85, with paternal grandfathers from Yorkshire and the West Midlands respectively and no known familial links to the USA, were found to belong to haplogroup K2; and these two showed a perfect 17/17 match for STR values of the descendents of Field Jefferson. This confirms the plausibility of Jefferson's commonly accepted heritage.

The background level of the K2 haplogroup in the UK is typical of the random dispersal of small numbers of uncommon haplogroups throughout the world and could have had any ancient origin of arrival in the UK. See the BBC article on K2 haplogroups for more information. One theory regarding the origin of K2 concerns the Phoenicians, an ancient maritime trading culture that spread out across the Mediterranean from their home in what is now Lebanon. K2 makes up about 7% of the Y chromosome types found in Somalia, Oman, Egypt and Iraq. It is present in very low frequencies elsewhere and in Europe, but it is not entirely absent.

Results
The results of the 14 descendants are shown. Differences are highlighted with bold font. The five descendants of Field Jefferson (and proxies for Thomas Jefferson) have nearly identical Y-chromosome DNA alleles except for a single difference at J50. It is a reasonable assumption that this is a point mutation.

The descendant of Eston Hemings has the same set of Y-chromosome DNA alleles as the descendants of Field Jefferson. This supports the claim that Thomas Jefferson could have been the father of Eston Hemings. It is impossible to prove absolutely that no other Jefferson fathered the child. (1) That would be proving a negative, and (2) any male who had the same Y-chromosome as Thomas Jefferson (other descendants of a common male ancestor) could have been the father, provided that this person was in the immediate vicinity of Sally Hemings nine months before the birth of Eston Hemings. Some have argued that historical evidence implicates Thomas' brother, Randolph Jefferson, as the more likely father of Eston Hemings; but this suggestion has been sharply criticised by others who note that it is Thomas himself rather than Randolph that the evidence places as being under the same roof as Sally at the likely time of her conceptions. The Carr descendants have similar DNA among themselves but are clearly different from either the Jefferson or Hemings descendants. Four of the descendants of Thomas Woodson are quite similar among themselves but different from Jefferson and Hemings although they do have similarities to the descendants of the Carr line. One of the Woodson descendants is quite different from all of the other individuals which suggests that one of the genetic ancestors was not in the direct line from Thomas Woodson.

Family Pedigree Member Bi Allelic Markers Microsatellite STR Mini Satellite MSY1

Jefferson
 * J41 0000001 15,12,4,11,3,9,11,10,15,13,7 (3)5, (1)14, (3)32, (4)16
 * J42 0000001 15,12,4,11,3,9,11,10,15,13,7 (3)5, (1)14, (3)32, (4)16
 * J47 0000001 15,12,4,11,3,9,11,10,15,13,7 (3)5, (1)14, (3)32, (4)16
 * J49 0000001 15,12,4,11,3,9,11,10,15,13,7 (3)5, (1)14, (3)32, (4)16
 * J50 0000001 15,12,4,11,3,9,11,10,16,13,7 (3)5, (1)14, (3)32, (4)16

Hemings
 * H21 0000001 15,12,4,11,3,9,11,10,15,13,7 (3)5, (1)14, (3)32, (4)16

Carr
 * C27 0000011 14,12,5,12,3,10,11,10,13,13,7 (1)17, (3)36, (4)21
 * C29 0000011 14,12,5,11,3,10,11,10,13,13,7 (1)17, (3)37, (4)21
 * C31 0000011 14,12,5,12,3,10,11,10,13,13,7 (1)17, (3)36, (4)21

Woodson
 * W55 0000011 14,12,5,11,3,10,11,13,13,13,7 (1)16, (3)27, (4)21
 * W56 0000011 14,12,5,11,3,10,11,13,13,13,7 (1)16, (3)27, (4)21
 * W69 0000011 14,12,5,11,3,10,11,13,13,13,7 (1)16, (3)27, (4)21
 * W70 1110001 17,12,6,11,3,11,8,10,11,14,6 (0)1, (3a)3, (1a)11, (3a)30, (4a)14, (4)2
 * W61 0000011 14,12,5,11,3,10,11,13,13,13,7 (1)16, (3)28, (4)20

Allele assignments

 * {| class="wikitable" style="text-align:center; font-size: 90%"

!DYS393 !DYS 390 !DYS 19 !DYS 391 !DYS 388 !DYS 389I !DYS 392 !DYS 389II !DXYS 156Y
 * 13
 * 24
 * 15
 * 10
 * 12
 * 12
 * 15
 * 27
 * 12
 * }
 * }

Results of re-testing the original Jefferson descendent samples for additional STR markers were published by King et al in 2007. Together with DXYS 156Y (which was not included in the new panel), this gives the extended DNA signature:


 * {| class="wikitable" style="text-align:center; font-size: 90%"

! DYS 393 ! DYS 390 ! DYS 19 ! DYS 391 ! DYS 388 ! DYS 439 ! DYS 389i ! DYS 392 ! DYS 389ii ! DYS 437 ! DYS 460 ! DYS 438 ! DYS 461 ! DYS 462 ! DYS 436 ! DYS 434 ! DYS 435 ! DXYS 156Y
 * 13
 * 24
 * 15
 * 10
 * 12
 * 12
 * 12
 * 15
 * 27
 * 14
 * 10
 * 9
 * 11
 * 13
 * 12
 * 11
 * 11
 * 12
 * }
 * }

(Note: the value of DXYS 156Y was reported as 7 in the original paper. This is believed to translate to 12 in the convention now used by DNA testing labs and online databases).