At the present time the Kingston surname DNA project has fourteen sets of Y chromosome results from thirteen individuals, representing twelve distinct Kingston families. Two of the individuals tested (Group B) share a common known Kingston ancestor. Although none of the other eleven Kingston families can be connected by paper records (despite extensive genealogical research in most cases), seven of the families, all tracing their Kingston ancestors back to the western part of Co. Cork in southern Ireland, as a result of Y chromosome analysis are now known to be related. An eighth family, based in North America since the 1800s, has an oral tradition of Irish Kingston origins, and as a result of Y chromosome analysis this family is now also known to be related to the seven aforementioned families with Irish origins. We have grouped these eight families together as Group A. The families in Group A do not share a genetic match with the individuals assigned to Group B. The three other families tested have results which differ significantly from each other, and from the individuals in Groups A and B, and we can conclude that the project now contains members from five independent Kingston families.

Kit #68119 (67 markers tested with FTDNA) and kit #smgf2 (38 markers tested with the Sorenson project) are from the same individual. The same 30 markers have been tested by both laboratories, with identical results. The most distant known ancestors for kit #68119 (and therefore kit #smgf2) and kit #smgf1 (43 markers tested with Sorenson project) lived in Meenies townland (866 acres) in the parish of Drimoleague, Co. Cork, Ireland. The most distant known ancestors for kit #106714 (67 markers tested with FTDNA) and kit #107275 (67 markers tested with FTDNA) lived in Clodagh townland (1046 acres) which is also in the parish of Drimoleague. The most distant known ancestor for kit #99618 (25 markers tested with FTDNA) lived in the neighbouring parish of Caheragh. The most distant known ancestor for kit #93641 (25 markers tested with FTDNA) lived in the Dungannon/Drombeg area of Kilfaughnabeg parish, Co. Cork, Ireland, approximately 15 miles from Drimoleague. The most distant known ancestor for kit #109951 (67 markers tested with FTDNA) lived in Timoleague, approximately 30 miles from Drimoleague and 18 miles from Dungannon/Drombeg. None of these seven distinct families could be connected by researching paper records, but a relationship is established by Y chromosome analysis. This demonstrates how invaluable DNA testing can be, and it also excludes the possibility of a non-paternity event in the Kingston line of all seven participants in the last 200-300 years. The most distant known ancestor for kit #114954 (37 markers tested with FTDNA) emigrated to Canada in the first half of the nineteenth century. The oral tradition of this family is that he was born in Ireland, although his son?s 1880 US census return gave his parents? birth country as Scotland. The same census record gave his son?s wife?s parents? birth country as Ireland despite the fact that her maiden name, Crockett, is a Scottish surname, suggesting that the census return (completed by the enumerator and not by a member of the household) could very easily be incorrect and possibly switched the birthplaces of the parents? of the head of the household and his wife. Y chromosome analysis has proven that this Kingston family is closely related to the other seven, which supports the oral tradition and undermines the information in the census return.

While it may not be surprising that a relationship has been confirmed between the Kingston families from the Drimoleague area, it is altogether more interesting that the Timoleague family is also a match. Timoleague-based Kingstons have been documented as far back as the 1600s, when a Colonel Samuel Kingston received land there during the Cromwellian plantations. In a detailed study by A. Richard Kingston entitled ?The Origins of Co. Cork Kingstons?, published in the Journal of the Cork Historical and Archaeological Society, Volume LXXXVI, 1981, the author suggests that Colonel Samuel Kingston, like neighbouring beneficiaries of the Cromwellian plantations, may have come from Somerset in England. The author speculates that Timoleague and Drimoleague Kingstons may all be related, but that documentary evidence of this has not been found. This project has proven that (at least some) Timoleague and Drimoleague Kingstons are related. As for the Kingston family with origins in the Dungannon/Drombeg area of Kilfaughnabeg parish, which is approximately an equal distance from both Drimoleague and Timoleague, the answer to which of these communities they are more closely related to may come from analysis of mutations. The next step is to find matching Kingstons who know where in England their Kingston ancestors came from. On the basis of the results of non-Group A members of this project (see below) we can assume for now that the ancestors of Co. Cork Kingstons probably did not come from Devon, Hampshire, or Yorkshire.

We can generate a modal haplotype for Group A, which constructs a haplotype composed of the most common results for each marker among the eight members in the group. The modal haplotype is the closest approximation we can make to what the haplotype of the common ancestor of these eight families might have been. It is based on statistics rather than fact, because we cannot directly test an individual who would have lived several hundred years ago. All eight families in Group A are a perfect 12/12 match. All eight families have been tested for the same 25. None of the families tested need necessarily possess this modal haplotype exactly, but as it happens five kits (kit #93641, kit #99618, kit #109951, kit #114954, and kit #smgf1) do match the 25 marker modal haplotype. Each of the other three families deviates by just one marker from the 25 marker modal haplotype, but in each case this is by a different marker: for kit #68119 it is DYS447, for kit #106714 it is DYS455, and for kit #107275 it is DYS464. The likelihood is that these ?mutations? occurred in recent generations, and it means that these three members have a genetic distance of one from the 25 marker modal haplotype (and from each of the five members of Group A who match the 25 marker modal haplotype exactly) and a genetic distance of two from each other. Five of the eight families have been tested for the same 37 markers. These include the three families that deviated from the 25 marker modal haplotype by one marker each. Two of these three families deviate by a second marker in the 26-37 marker range: for kit #68119 it is DYS607 and for kit #107275 it is CDYb. Kit #106714 does not deviate any further from the modal haplotype. The other two families tested for these 37 markers (kit #109951 and kit #114954) match the 37 marker modal haplotype exactly. Four of the eight families have been tested for 67 markers (kit #68119, kit #106714, kit #107275, and kit #109951), and these include the three families that deviated from the 25 marker modal haplotype (two of which deviated by a further marker for markers 26-37). The three families that deviated from the 25 marker modal haplotype have identical results for markers 38-67, but the fourth family (kit #109951), which matched the 37 marker modal haplotype exactly, deviates from the 67 marker modal haplotype by two markers (DYS444 and DYS531). Data relating to mutation rates is available for the first five of the seven mutations mentioned above (no mutation rate data is available yet for DYS444 or DYS531). Four of the five are fast-mutating markers and just one (DYS455) is described as a slow-mutating marker.

When the families are compared at the maximum number of markers which they have had tested they have no closer matches in the entire FTDNA database (148,500 Y chromosome results as of February 2009) than that to each other, with one exception (Branley, see below). Using FTDNATiP analysis to calculate the probability that the families share a common ancestor within 12 generations, the range of probabilities for all eight families in Group A is from the least probable at 77.85% (kit #107275 and both kit #93641 and kit #99618 at 25 markers) to the most probable at 99.55% (kit #114954 and kit #109951 at 37 markers). The one closer non-Kingston match is between kit #107275 and a Mr Branley (37 markers tested with FTDNA) at 37 markers, with a probability of sharing a common ancestor within 12 generations of 78.4%. Other Group A members show only slightly greater genetic distances from Mr Branley. Possible explanations include a non-paternity event (ie a child taking the surname of somebody other than their natural father, eg adoption, illegitimacy, extramarital relationships, a family altering their surname) or a common ancestor who lived before surnames were first adopted. The closeness of the genetic match would perhaps argue against the latter possibility.

All nine kits in Group A have been estimated to belong to haplogroup R1b1b2 (known as R1b1c until 2008), and one of these kits (kit #68119) has subsequently undergone deep clade testing which confirms this haplogroup ? with a more specific designation of R1b1b2a1b (M153- M222- M269+ M37- M65- P312+ P66- SRY2627- U106- U152-). R1b1b2 is by far the commonest haplogroup in Western Europe. R1b accounts for 70% of the tested male population of Southern England and over 90% of the tested male population of Ireland. The vast majority of these are R1b1b2.

Of particular interest in the results of all nine kits is DYS393=12. Only 2% of the R1b haplogroup in Western Europe display this result, most of the others showing DYS393=13. DYS393=13 is known as the Atlantic Modal Haplotype (AMH) or Haplotype 15. DYS393=12 is known as Haplotype 35. Haplotype 35 is found with greatest frequency in Eastern Europe and Western Asia, decreasing towards Western Europe. Haplotype 15 shows the opposite trend, and is present with greatest frequency in Western Europe. There is a theory that members of Haplotype 35 are descended from those who found shelter in Anatolia during the Last Glacial Maximum while members of Haplotype 15 are descended from those who spent that period in Iberia. There are a number of theories as to how Haplotype 35 found its way to Britain and Ireland. These include mass migration of Roman soldiers of Alan and Sarmatian origin to Northern England (a theory favoured by the Border Reivers DNA Study), migration of Sephardic Jews, and Pictish origin (some authorities believe that the Picts originally came from Scythia). Although DYS393 is a very slowly mutating marker, it is also possible that DYS393=12 simply represents a recent mutation from the more common DYS393=13, but its presence in all nine kits means that this mutation could not have occurred in at least the past 200-300 years.

Kit #138797 (12 markers tested with FTDNA) and kit #138799 (12 markers tested with FTDNA) are first cousins, whose most distant known ancestor lived in St Peter, Leeds, Yorkshire, England. Interestingly, although Y chromosome analysis essentially supports their relationship, they do differ from each other at one of the 12 markers tested (DYS439). DYS439 is described as a fast-mutating marker, and clearly the mutation must have occurred either in one of the two individuals tested or in their fathers. This highlights one of the weaknesses of FTDNATiP analysis, according to which these individuals have an estimated probability of just 32.64% of sharing a common ancestor within 12 generations, despite the fact that we know they shared a paternal grandfather. These individuals are both estimated to belong to haplogroup I1. Haplogroup I1 is most commonly found in Scandanavian populations, and a relatively small percentage of the British population belong to this haplogroup. There are a number of opposing theories regarding its presence in Britain, with many supporting the important role of Viking and Norman migration. More recently the Genographic Project has reported that the origins of this haplogroup can be placed within the Iberian peninsula during the Last Glacial Maximum.

The most distant known ancestor for kit #107117 (37 markers tested with FTDNA) lived in Barnstable, Devon, England in the 1500s. The most distant known ancestor for kit #107546 (25 markers tested with FTDNA) lived in Eversley, Hampshire, England in the 1700s, although we cannot be sure that he was born there. Both are, like the eight Group A families, estimated to belong to haplogroup R1b1b2. However, both have DYS393=13 and are therefore not of Haplotype 35. The most distant known ancestor for kit #142045 (37 markers tested with FTDNA) is believed to have come from Stoke Damerel, Devon, England, although confirmation of this is awaited. Kit #142045 is estimated to belong to haplogroup I2b1. Haplogroup I2b1 is relatively uncommon in Britain and it is said to have similarities in distribution within the Balkan states and Western Europe to haplogroup I1 (see Group B above) but with the important difference that haplogroup I2b1 is uncommon in Scandanavia.

Although the individuals tested with kit #107117 and kit #142045 both trace back to Devon, they are not related to each other. Barnstable, in the north of the county of Devon, and Stoke Damerel, in the south of the county of Devon, are approximately 60 miles apart.

The results of these three families neither match each other nor the families in Group A and Group B, and therefore represent three further independent Kingston families, bringing our total in this project to five unrelated Kingston families. This supports our earlier hypothesis that the Kingston surname was probably adopted independently by families from different places called Kingston. The addition of these three families widens the spectrum of Kingstons represented in this project and increases the chances of future participant families finding a useful match.