FAMILY GROUPS TRADITIONALLY ATTRIBUTED TO CAPT. THOMAS GRAVES (GEN. 169)

(DNA Test Results from FTDNA)

As the result of many years of research, Capt. Thomas Graves, 1608 settler in Jamestown, VA, was believed to have 3 sons, John, Thomas, and Francis. Adventurers of Purse and Person, Virginia, 1607-1624/5, third edition, 1987, documents the first 4 generations of this family, and the fourth edition, published in 2007 as a 3-volume set, documents the first 6 generations. However, tracing the lines down to the present day has been a daunting task. Many records are lost or not yet readily available. Y-chromosome DNA testing has been performed on as many descendants as possible to try to learn more about this family, to verify lineages, and to help find earlier ancestry. The results of this DNA testing have been surprising. The chart below shows the four major family groups that have been identified. The DNA test results show that it is almost certain that each of these four groups is descended from a different immigrant ancestor. That observation leads to the conclusion that probably one of these groups is descended from Capt. Thomas Graves and the other three are descended from another ancestor. Because the line of John², Ralph³, Ralph⁴ is well documented and carries the name Crowshaw down through a number of generations, and that was supposedly the surname of the wife of Capt. Thomas Graves, that seems most likely to be the line of his descendants.

One question this situation raises is, assuming these conclusions are correct, what happened to the direct male lineages from the other documented sons and grandsons of Capt. Thomas Graves.

The DNA results of the tested descendants of John², Ralph³, Ralph⁴ exactly match the results for the descendants of Thomas Graves of Hartford, CT (gen. 168) and his brother Deacon George Graves (gen. 65). The documentation for the early generations of this line seems solid, supporting the premise that this is the true line of Capt. Thomas Graves.

The results from the descendants of John², Thomas³, John⁴ almost exactly match (24 of 25) the results for the descendants of William Greaves of Whitfield, Northamptonshire, England (gen. 47). Since there is not good documentation from John⁴ back to John², it is not difficult to believe that this lineage is in error. The most uncertain link in the early generations of this line is John⁴ as a son of Thomas³, and that is where any error probably is.

For Francis², it is now believed that the documented youngest child of Capt. Thomas Graves was a daughter, Frances Graves, as discussed in an appendix to genealogy 169, and the male Francis Graves was a son of some other immigrant, presently unknown. We unfortunately have only two tested descendants of Francis Graves (with a third one in process as of May 2007), but the name Francis was carried on in other families (including genealogy 150) whose tested descendants exactly match the results for the Francis descendant. The results from Francis² also exactly match the results of descendants of Rear Adm. Thomas Graves (gen. 28), John Graves of Concord, MA (gen. 166), John Greaves of St. Mary’s Co., MD (gen. 247), and others.

The most surprising result is for the line of Thomas², Thomas³, John⁴, John⁵. The documentation for this line is fairly good, and it was expected that it would match the John², Ralph³ line. We have not yet found an English family whose DNA result matches this line. However, we may be able to find and test descendants of other branches of this family group, so it is still possible that the ancestral haplotype for this part of the family may change. On the other hand, another possibility is that Thomas² was an adopted or illegitimate son of Capt. Thomas Graves, since his descendants are the only ones not yet matching a Graves or Greaves family in England. In any case, we should eventually find a match either with a Graves or Greaves family in England or with another surname.

More lines from Francis² and Thomas² need to be tested. Although that will provide more confidence that the test results represent the ancestor (Francis or Thomas), that will not affect the conclusions for the lines already tested. There are too many differences between the DNA results of the family groups in too few generations to hold out much hope that any of these groups will be shown to descend from a common immigrant ancestor. A summary of the test results and further analysis and discussion is below.

	Locus	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	16	17	18	19	20	21	22	23	24	25	26	27	28	29	30	31	32	33	34	35	36	37
ID#	Surname																											G	Y	Y									Gene-	Gener-
																												A	C	C									alogy	ations
	& DYS #					3	3				3		3		4	4							4	4	4	4		T	A	A					C	C				from
		3	3		3	8	8	4	3	4	8	3	8	4	5	5	4	4	4	4	4	4	6	6	6	6	4	A	I	I	4	6	5	5	D	D	4	4		common
		9	9	1	9	5	5	2	8	3	9	9	9	5	9	9	5	5	4	3	4	4	4	4	4	4	6	H	I	I	5	0	7	7	Y	Y	4	3		ancestor
		3	0	9	1	a	b	6	8	9	i	2	ii	8	a	b	4	5	7	7	8	9	a	b	c	d	0	4	a	b	6	7	6	0	a	b	2	8
1391	Graves	13	24	14	11	11	14	12	12	13	13	13	29	18	9	10	11	11	25	15	19	29	15	15	17	17													169	4
1620	Graves	13	23	14	11	11	15	12	12	12	13	13	29	17	9	10	11	11	25	14	19	29	15	15	16	16	12	11	19	23	16	15	17	18	36	40	12	12	169	4
31933	Graves	13	23	14	11	11	14	12	12	12	13	13	29	17	9	10	11	11	25	15	19	29	15	15	17	18													169	5
1872	Graves	13	24	16	11	11	14	12	12	12	13	13	29	18	9	10	11	11	25	14	19	29	15	15	16	17													169	4
2231	Greaves	13	23	14	11	12	15	12	12	12	13	13	29	17	9	10	11	11	25	14	19	29	15	15	16	16													47	8

The number of generations in the preceding table is from the earliest ancestor of each group of descendants for which an inferred haplotype could be determined. (For more detail, see www.gravesfa.org/DNAchart169.pdf. That chart no longer includes the group that contains ID 1872.) As can be seen on DNAchart169.pdf, each group of descendants includes DNA test results from multiple descendants, and the ID number in the above table is representative only. The last row in the above table (with ID 2231) is from genealogy 47 (William Greaves of Whitfield, Northamptonshire), and was included because it appears to share a common ancestor with the group represented by ID 1620. The number of generations for ID 2231 is the minimum, and could be more. Depending on the assumptions, the number of generations for ID 1872 could be 9 or 1, so 4 was put in the table as a compromise.

Probabilities for a specific number of mutations have been calculated using a computer program developed by Ann Turner. Results of that calculation are shown below for 8 and 9 transmission events (also called mutation opportunities). The number of transmission events between two descendants of a common ancestor is the sum of the generations on the two lineages. For example, the number of events between the line represented by ID 1391 and that represented by 1872 is 4 + 4, for a total of 8, and the actual difference between these two haplotypes is 4. The table below shows that the probability of 4 mutations in this number of opportunities is only 0.00072, extremely low.

No. of Transmission Events	Expected No. of Mutations	Probability of a specific no. of mutations
		0	1	2	3	4	5
8	0.4	.67032	.26813	.05363	.00715	.00072	.00006
9	0.5	.63763	.28693	.06456	.00968	.00109	.00010

The phylogenetic diagram below was generated using Fluxus Network 4.1.1.1 from www.fluxus-engineering.com/sharenet.htm. The median joining technique was used.

This page revised 22 May 2007.