04 September 2005

Where are we now? Update #6

Background to the Update

This update has been done as the number of participants in the Clan Gregor DNA Project approaches 150. We have benefited form a change in Family Tree DNA’s group structuring arrangements whereby some members of the separate Gregory, Grier, and Stirling Projects have decided to have their results also shown on the Clan Gregor Project pages. We welcome these individuals and some of their earliest ancestry statements will appear in this report and in due course.

As a result of the larger numbers involved in the project now, I have decided to try to make sense of the various groupings which seem to be developing, and to, at least confront some of the anomalies which seem to be emerging.

This update uses the arrangement of results presented in Table 1 below. It is suggested that the easiest way to follow this is to actually print out Table 1 (in colour if possible) and refer to it directly when reading the following discussion.

The Table has been colour coded. The colours in themselves have no significance but they are used to separate different family groups – sometimes these are quite small, single surname groups, or for larger grouping of different surnames which have some elements in common. Participants’ results are uniformly referred to by their kit number.

Table 1 - Marker Comparison (click to enlarge)

Haplogroup Q

At the beginning of the chart, we have one Bennett (15716). This DNA signature is normally associated with Native Americans (originating in East Asia) but a comment from Dr David Faux of Ethnoancestry is significant here. He says: 'I was given access to the skeleton findings of a Shetland database a while back and was astounded to learn that 8% of the sample were typed at haplogroup Q - the same percentage as haplogroup I. My first "Q" emerged recently in a participant with a patronymic surname...Clearly the distribution pattern points to a Norwegian source dating back to Viking times.'

Eastern European Connection

McGregor (5356) and the two brothers (7422 and 9338) belong to a DNA haplotype J2 that is found in Britain, but rarely. It is most common in Eastern European countries, leading to speculation that it is either from gypsy background or, possibly, from Eastern European soldiers stationed in Britain during the Roman occupation in the first three centuries AD. These two (or rather three) MacGregor results are more than likely too distant from each other to be from the same individual within the historic period. MacGregor (Horn (19304) is probably also part of this group but the DNA profile could also be I1b (which is possibly Scandinavian in origin). Only a haplogroup test would confirm this. Bennett N6998 is also almost certainly in the J haplogroup since the DNA signature is like kits 7422 and 9338 who have confirmed their haplogroup as J2.

The Irish/Scottish Viking Group

The group Magee (208290) to Nevins (2314) are quite closely related and are all from haplogroup I, the Viking haplogroup, specifically Norwegian (from the occurrence of the specific numbers). It is possible that these are descendants of the Viking kin group which settled in the Northwest Scotland/Isle of Man/East Ireland after the Viking invasions of the 9th century AD. Although the common ancestor of these individuals may be some time back, possibly two millennia – there is no question that the kin group was a very tight one – being all descended from a single individual and probably aware of their connections.

The Grier/Greer subgroup (shaded in green) almost certainly represent descent from one individual of that name, with the possible exception of Greer 17630 who probably has too many DNA mutations (21 out of 25) to make him related directly to this group within the time of the adoption of surnames. To be definite about this, it would be necessary to see 37 marker results from this whole group. However, it is striking that an individual can have so many mutations more than others yet still be part of the surname group, derived from the common Viking ancestor. As we will see later with the Stirlings, this cannot easily be explained, except possibly through geographic proximity at the time of adoption of surnames.

All those who have tested hereafter up to (but not including) Orr in Table 1 are from the haplogroup R1b – perhaps THE most frequently found male haplogroup in Europe. It is common across Spain, France, Germany, the Low Countries, Britain, and Ireland and, as I have written before, represents spread out after the last Ice Age but all having a common ancestor who lived 20 to 30 thousand years ago.

Other MacGregor Lines: Back to Those Named MacGregor

Kits 24470 and 11773 could well be related. Kit 17790 might belong to this group.
These 3 are in yellow.

Kits 20123 and 1774 are related: this is almost certainly the Glengyle family. Denoted by blue.

Kits 13429 to 22187 might well be related to each other more recently (by which I mean from 1603 to 1850). This is hard to prove without a better paper trail. This group is in green.

Kit 4714 seems to be currently an isolated result and therefore connected more distantly in time (in the 15th century for example). This entry is white.

Kits 3346 to 10454 seem to be very closely related, sharing the same DNA sequences for the most part with the Chief’s line. If these individual are not just examples of very ‘stable’ DNA then they must be related more closely to the Chief’s line and to each other. This group is given in light yellow.

Kit 13165 may go with the above group or the ‘original’ Stirling group.

Kits 12511 and 24029 may be related more directly to each other or they may be separate family groups. These two are in darker yellow.

Kits 12596 to 26174 (in orange) appear to share common characteristics which may indicate family connections.

Kits 12436 and 17711 (given in puce) are not directly related to each other but seem to be a wider and more ancient connection to the MacGregor group as discussed in the update before this one.

This brings me on to those individuals in group R1b who are not immediately connected to the ‘original’ MacGregor line.

Part-takers/Those with Similar DNA

The Turks (kits 11655 and 21757) are related to each other and have a resemblance to the MacGregor line. They are traditionally called after Brig o’ Turk in Perthshire where there has been a strong MacGregor presence for many centuries and where the Glengyle family had land at one time. The DNA suggests that the family origin story may indeed be accurate.

The Greers 10589 and 35624 are related to each other almost certainly but not to the earlier Greer group. These could be descendants of Grierson of Lag in Dumfriesshire as suggested by the family genealogies.

The MacGregor/McGregor group from 2726 to 27811 have many general DNA matches with other surnames, but, are almost certainly from the same family. From paper evidence this would be a family using the surname Grigor and living in the vicinity of Perth in the 17th and 18th centuries. This group are coloured dark blue. Two of this group are known cousins.

The next group of MacGregors/Greig/Gregg/McGhee are not related to each other and represent single families who have either adopted the name MacGregor or use the version now found. They share a common ancestor with the MacGregor line but before the development of surnames though there is still the possibility that the connection goes back only as far as the time of Kenneth MacAlpine.

Of the remaining results in this section McAdams 3133 and 9567 appear to be related; Gregorys 25690 and 36102 likewise; McGregor 29834 and 36394 are potentially related, and 7783 may be part of this group.

The next group – from Grierson 7874 to Gregg 20673 have very close genetic similarities, which, from the paper trails, suggest a possible recent Irish origin and descent from a common ancestor pre surnames. The two Griersons 7874 and 33323 are almost certainly directly related post surname adoption. The same is true for Mc/MacGregors 4715, 2944, 17929 – a suggestion here of a second, smaller and more recent MacGregor surname origin which could bring in the Griersons too may give at least one reason for the persistent story of connection between Grierson and MacGregor.

The fours McAdams 16568 to 11256 are likewise probably similarly related to each other in recent time

MacGregor 36727 and Gregg 20673 seem to be separate families though tied into the overall high level of connectedness pre surnames.

Finally in the R1b sections there is the Foxtons who are not related to anyone else but ARE related to each other. These are given in yellow.

A new grouping, coloured in green, is the Orr family. I have mentioned these before in previous updates. Their haplogroup is not R1b but R1a , they are all related and have identified an Irish origin for the extended family. By the way, Orr 16713 is not of this family group unless there was an adoption or illegitimacy in that family.

Next we come to the two McAdams 3714 and 5223. These have identified a common ancestor but their actual DNA origin is from the comparatively rare haplogroup E3b suggesting an Eastern European or middle eastern origin. These families could descend from Roman auxiliaries or Neolithic farmers who had come into Britain.

Older Viking
The final group from Mitchell 36808 to MacGregor 31436 are not related to the main MacGregor lines or indeed any of the others so far mentioned, expect that they have an ‘older’ form of the DNA sequence associated with Haplotype I – the Viking group – but suggestive of a different settlement pattern than that of the Irish/NW Scotland group. Possibly the ancestors of these came into Eastern England/Scotland from different Norwegian origins than the earlier group.

Chart A - DNA linkages (click to enlarge)

Chart A above shows DNA linkages based on distance from each other. It is interesting to see which names group together. The bunching of Stirling is particularly significant showing a common time depth for origins. For the main line MacGregor, only the chiefly line (2124) is given.

[McGre(g) is short for MacGregor, Stir for Stirling and Dougal for MacDougall of the Isles, McPet for McPeters, Grego for Gregory]

Earliest known ancestor detail for the new Stirling Group

13635 - Feltus Barrow Stirling, Jr., earliest known ancestor Andrew Skirling, b. Kinnettles, ca 1714, married Jean Nevay ca 1735, d. Nov. 1780, interred Kirkton of Kinnettles. Descended through Andrew's son Alexander Stirling (bapt. 07 Jul 1751, Kinnettles), who emigrated prior to the American Revolution and eventually settled at West Feliciana Parish, Louisiana.

16710 - Wayne L. Stirling, earliest known ancestor John Skirling, b. ca 1730, probably in Barry, Scotland, died aft. 1784, Barry or Panbride, Scotland. Descended through John Skirling's grandson, James (bapt. 17 Jul 1792, Panbride) who emigrated in to Upper Canada 1840 with his wife, Elizabeth Lawrence, and their children, Three of his brothers (George, Alexander and William), their wives and children emigrated at the same time.

13678 - John Anderson Stirling, earliest known ancestor David Skirling, b. ca 1745, married Janet Wallace 01 Aug 1773, Dunnichen, d. 15 Mar 1798, Dunnichen. Family still living in Scotland, though branches did emigrate to New Zealand and the United States.

23501 - William Stewart Stirling, earliest known ancestor James Stirling (or Skirling), b. ca 1740, married 26 Dec. 1773, Kinnettles, Margaret Whyte of Kinnettles, he was from Inverarity, d. before 1858. Family still living in Scotland, with some branches in the United States.

A Brief Discussion of the New Charts

Note: The 25 marker charts (1 through 3) have all the results to date for writing (early August 2005). However some recently reported 37 upgrades have not yet been included in the 37 marker charts. These will appear in the 37 diagrams for the next update. It is necessary to fix a cut off point for updates or else they would never get done with the constant influx of results coming in.

Table 2 lists all the 25 marker results I have in my hands to date except that where there are individuals sharing results (either exactly, or with very few mutations) I have grouped these all together into one ‘super result’ (for example 29MacG2124 – is 29 MacGregor results together within kit 2124 with mutations discounted).

Table 2 -25 Marker Results (click to enlarge)

This is to allow participants who do not fit into these ‘super groups’ to see where they lie. Thus Charts 1 and 2 are not specifically intended for members of the ‘supergroups’ but for the others. Most of these ‘supergroups’ have 37 marker results and therefore appear in the later charts (3, 4, and 5).

Chart 2 fits onto the bottom of Chart 1. Chart 1 is the diagrammatic representation of the haplogroup R1b for members of the project. Most project members belong to this haplogroup. Note also the haplogroup E3b. This is a quite separate haplogroup distant from R1b – on the chart the connecting line from R1b has been foreshortened for space purposes making it look as if there is a shorter time distance between them.

Note: In Charts 1 and 2, kits are identified by 3 letters of the surname and last three letters of the kit. The program only allows 6 letter identifiers. Usually, I will use the whole kit number.

Chart 1 - Haplogroup R1b (click to enlarge)

Chart 2 shows members of the haplogroup R1a (always Orr) and haplogroup I – there are two branches of this haplogroup. The branch leading to Nevin, etc., is the ‘younger’ branch in Britain – supposed to be associated with the Viking invaders of the 9th century (to Ireland and the Western Isles).

Chart 2 - Haplogroups R1a and I (click to enlarge)

Chart 3 shows all the 37 marker results from the MacGregor project excluding those who match the bloodline and are called MacGregor. All these are ‘within’ the red star labelled 2124. What is remarkable is how dense the next layer out is called Stirling. Although Stirling was a known alias, I cannot currently explain the DNA similarity with the MacGregor line in so many cases.

Chart 3 - 37 Marker Results: Excluding MacGregors (click to enlarge)

Chart 4, also 37 markers, is the full set of 37 results (but note that the computer programme only shows kit 2124 where the DNA match is exact - thus 2909, for example, matches 2124 exactly, so 2909 does not appear independently). The clustering effect of the MacGregor bloodline results (in a star-like pattern) is normal for strongly related results. The lower kits are named because they did not fit, for the most part, into Chart 3. The Gregory (?) is my indication that I am not sure why the program chose to place these results with effectively the Viking Group when the results themselves suggest membership of the R1b haplogroup.

Chart 4 - 37 Markers: Full Set (click to enlarge)

Chart 5 gives a labelled version of, primarily, the MacGregor main group. Again, exact matches are taken into the red circle for 2124. The circles SCO can be ignored . They refer to the process of contracting the connections between different results to avoid large numbers of lines on the page.

Chart 5 - MacGregor Main Group (click to enlarge)