29 December 2011

MacGregor DNA Project annual update 2012

MacGregor DNA Project annual update 2012

Welcome to the 2012 update of the MacGregor DNA project. For this update, as a result of the nature of the test results which have appeared over the past year, I am concentrating on three groups: Greig/Gregg/Grigor; MacGregor distant and Septs. There have been only a few Ian Cam MacGregor new results this year and a few upgrades from 37 to 67 markers. At the end of this article I comment on what relationships if any can be inferred from those results. I will also comment on relationships to be found in the various subgroups that are listed on the results page at www.familytreedna.com/public/macgregor. This year’s update does therefore not include discussion of the Ian Cam group, the Irish related group and the Viking groups which have all been covered in detail in recent updates. At the moment I am not including discussion of McAdam(s), Grier and Gregory results as there are separate projects for these which have many more results than those listed in the MacGregor project. However as always I am happy to provide commentary on these by e-mail.

The three groups which I am dealing with this year should not be thought of as containing participants who were all related in the recent past. Most of the branches of the trees presented here are several thousand years old before they converge on a single ancestor who was the original founder. This means that there are participants who share the same name but not the same genetic signature within the time of the adoption of surnames (from approximately 1300-1350). However what the charts do suggest is some definite and some possible connections which participants may wish to explore further. Often these connections are through the same surname even if spelt differently (when only church clerks could write how they chose to spell a name was sometimes idiosyncratic). Occasionally the trees suggest a connection between individuals with different names. This can be simply an accident of genetic mutation but could suggest a connection relating to the late adoption of a surname, or of adoption.

For the two groups after the Gregg/Greig/Grigor tree I have drawn two charts which include the same information. The first chart has no labelling apart from the kit number and the first couple of letters of the name of the earliest known ancestor (I had to adapt a couple of entries but these should be obvious). The second chart contains the information supplied concerning the earliest known ancestor of the line. In some of the charts these labels can be quite close together but hopefully comparative use of the two charts should make it clear what label belongs with which branch.

In these charts I include the MacGregor of Glencarnoch entry (kit 2124) and the typical Gregory signature for the largest group of that family (kit 179683) again for comparison.

Chart 1 Gregg/Grieg/Grigor (excludes the Gregors near Perth who changed to MacGregor in the late 18th/early 19th centuries)

This chart suggests:

- Gregg kits 137236, 76121 and 110496 are closely related – unfortunately no ancestral information has been supplied

- Gregg kits 130191 and 6979 are closely related and seem to have an connection with Tipperary in Ireland

If these two lines (and Gregor 44975) are connected it will probably be at the earliest time of surnames

- 214992 Gragg, 7489 Gregg and 141020 Gregg are all related through a common ancestor. Interestingly this is one instance where two quite different spellings are from the same family. There appears to be a connection therefore between Kilkenny County in Ireland and South Carolina

- Gragg 158127 and 81282 are related and appear to come from Antrim in Ireland

- Greig 9690 and Grigor 131056 are probably not related by surname – there is a possibility that both families originate in the north east of Scotland. The fact that the MacGregor of Glencarnoch line is attached to this branch has been discussed in an earlier update – the suggestion is that this connection dates from what has been known in the past as the Dark Ages (c500-1000).

- Gregg 64662 and Greig 45360 are related and appear to have a Scottish connection

All the other lines are almost certainly separate genetic lines and not related directly to those discussed above. Chart 3 is a chart which takes the entries above and places them in a grid which estimates the time to the most recent common ancestor. The numbers should only be used as a guide since DNA mutates randomly. The grid is generated by Dean McGee’s Y-Utility found at http://www.mymcgee.com/tools/yutility.html?mode=ftdna_mode. What is obvious is that the time distances between individuals extend well beyond the time of the adoption of surnames.

Chart 2 Gregg/Grieg/Grigor Time to Most Recent Common Ancesto

Part 2

MacGregor (Distant) charts (includes the Gregors near Perth)

In the following charts what is immediately clear is that these are a good number of families called M(a)cGregor who are separated from each other genetically and that their common ancestor existed before surnames.

Chart 3 – unlabelled MacGregor distant

There are several reasons why different families adopted the same surname and previous updates have explored some of these in depth. Briefly, families adopted surnames to replace patronymics (so Donald McGregor would replace Donald son of John son of Ewan, or sometimes Donald son of Duncan son of Gregor), or, a surname was adopted to be part of the clan group – especially if living on land owned by a powerful or influential Laird or Chief. There are, for example, quite a few people called Grant today because their ancestor lived on the land owned by the Laird of Grant. Sometimes these changes of name are recorded in the parish registers when a child is baptised but often not.

Chart 4 MacGregor distant - labelled

I will discuss this chart from left to right clockwise starting in the lower left corner.

- kit 5356 is very distant from all the others and goes off the page for the same distance again since this result belongs to a completely different haplogroup (group which defines a more common genetic origin). All the others in this group belong to haplogroup R1b which dates back approximately 12,000 or more years so 5356 is connected many thousands of years even before that. There is no significance to the fact that 213368 and 15199 connect to this line. These are single result families at this point and not connected in time of surnames

- 155381, 99676, 186233, 164124, and 158870 – all these, from the evidence, are related. It seems likely that this family originated in Rosshire

- 170627 (no genealogical details) and 75529 Drummond alias McGrigor in Kincardine, Perthshire may be related in the period of surnames but all the others in that group (117991, 98515, 60181, 79517 and 179182) are either not related within the surname period or are connected in the centuries before proscription of the clan (from 1603)

- 13403 (John born 1753) and 130631 (no details) are probably related to each other within the last 400 years – it is less likely that 20630 James Ross MacGregor is other than distantly related to them

- 153633 James born in Forgue and the Gregory line are quite separate from all others and not connected in surname time – the connection is certainly 2000 or more years ago.

- 27811 (James Gregor Moneydie) and 161907 (no details) are definitely related, as are 7183 John Gregor Bankfoot, 2726 Alexander MacGregor 1770 and 138485 (no details). Charts in earlier updates have suggested that these five should be considered as a group and may have their origins using the surname Gregor in parishes near to the town of Perth.

- 132396, 167829, and 90803 (no details supplied for these) – if there is a connection here it must be very early or DNA has been subject to increased mutation

- 178662 Daniel 1814 and 100843 Alexander in Creich – likewise this is either an early connection or DNA has been subject to increased mutation

- 95193 James in Rathven could be related to 94589 Charles in Clackmannanshire but again this would be either distant or DNA has mutated quickly, but 94589 is closely related to 158917 and 129009 no details for either)

- 65444 and 35981 Patrick McGregory can only be related distantly

- 153297, 4151 and 198351 are all related and probably descend from three brothers born 1769, 1772 and 1766 respectively. Since this is my own family I can be quite sure about most of the detail in it. What is significant and therefore may hold true for other families too is the significant amount of mutation that has occurred in each branch over the last 200 years. The computer program still links the three lines together but the short time distance and the relatively long twigs that are produced over such a short time suggests that similar rates of mutation could be found in other families. This is why it is hard to say that some families are definitely NOT related within the period of surname adoption.

- The next branch containing MacGregor of Glencarnoch 2124 and 17711 John, Milnathort, 122079 James, Greenville, 121543 (no detail), 28296 Donald McComas and 17621 James McKinlay all appear to be closely related but the problem here is that we believe there has been no mutation in 2124 in 600 years which would put connections earlier than the adoption of surnames. This may explain why James McKinlay and MacGregor of Glencarnoch are related as there is a tradition of connection between these clans pre surnames. Alternatively McKinlay could be an alias adopted by a MacGregor but again it would have to be very early in the 2nd Millennium that this happened.

- 173181, 185487, 126138, 29834, 200914, 137093 and 84081are all related in some way to Rev William McGregor through collateral lines or forebears who lived a few generations before him

- 174561 is supposed to descend from Rob Roy MacGregor but the DNA suggests otherwise

Part 3

Sept names

In a way the Sept names are easier to deal with in that one does not expect these to be close together. A Sept was created by linking a surname with a clan, usually by tradition, and it is quite possible that individuals with the same surname are not related to each other (as for example with Black and King in these charts).

Chart 5 – Sept names unlabelled

Once again I begin at the bottom left and this time with the participants called Turk. There is no question that these individuals are all related and connected with Antrim in Northern Ireland. However the family has a tradition of actually coming from Brig o’ Turk in the Trossachs, Scotland. Their DNA signature suggests a connection with MacGregor of Glencarnoch that may have been before surnames were common. If this was a name adopted as a result of proscription then it was probably based on geographical proximity to the MacGregors and a shared genetic heritage going back into the ‘Dark Ages’.

Chart 6 – Sept names labelled

In the next group, the Skirlings, 16710, 13678 and 13635 are closely related, and the rest: 2124 MacGregor of Glencarnock, 205188 Enoch Murray, 23501 James Stirling Inverarity, 81803 Brown and 129382 Black are probably all related within the period up to the adoption of surnames rather than after it, so again this may demonstrate geographical proximity and a common ancestor during the ‘Dark Ages’.

203008 White and 179484 would have a common ancestor probably 2,000 or more years ago. There are no other matches for these in the project at this point. A similar comment applies to the group 205721 Shanklin, 61249 Walker, 3963 Black and 179683 Gregory. This is the same for 208986 Whyte and 168162 Magee (this last has a different genetic origin from the larger group which will be discussed later). The same applies to 126778 Black and 95320 West. 104782 Gregson 50101 White, and 10279 McPeters are singles with no matches in the project at 37 markers.

120116 and 115596 are not related to others in the project apart from, surprisingly 27575 King, but are closely related to each other. 17586 McGeorge and 158878 Dougall are not closely related – it is just that their respective DNA has not undergone much mutation. 41034 and 188467 both share the surname Black and look as though they could be distantly related but their individual DNA has mutated rather quickly if that is the case. 74820 and 179850 have the same known ancestor and are indeed closely related genetically.

All the other DNA profiles are unrelated to each other and the rest of the group with the possible exception of 32205 White and 114313 Lackie who may share a common ancestor in the last 600 years. 108644 Black, White 151731 and 166858 Grigsby are from completely different haplogroups than the R1b shared by all the other participants in this group. They all belong to haplogroup G but even in that group they are quite distant from each other. According to Wikipedia:

Various estimated dates and locations have been proposed for the origin of Haplogroup G. The National Geographic Society places haplogroup G origins in the Middle East 30,000 years ago and presumes that people carrying the haplogroup took part in the spread of the Neolithic. Two scholarly papers have also suggested an origin in the Middle East, while differing on the date. Semino et al. (2000) suggested 17,000 years ago. Cinnioglu et al. (2004) suggested the mutation took place only 9,500 years ago.

Finally in this group, all the participants, apart from the one already mentioned, who bear the name Magee, McGehee, Mackgehee and McGee are all closely related having no doubt descended from the ancestor who emigrated to the United States in the 17th century.

General Comments on other results:

In the Ian Cam line kit 191228 was received this year. This line has sufficient mutations to suggest that it has developed independently for some hundreds of years, possibly since the time of proscription. However the DNA may imply an ancestral connection with kits 119330 (Malcolm b 1780 Buchanan), and James McPherson (bc 1752 Abernethy) and 94903 (no details) going back to the early 1600s.

Kit 217464 (no details) is also new this year. It is hard to place this exactly as a result of a very distinctive mutation in locus 29 (19 where most have 24 or 23). This means there could be a connection with 94903 (no details). Alternatively there could be a connection through the locus 32 mutation (to 19) with 152478 (no details), or 90446 (Duncan b 1795 Paisley), or less likely, 119330 (Malcolm b 1780 Buchanan) and 4714 (Duncan MacGregor Fortingall 1773?).

With one exception all the McFarlanes in the Ian Cam group, however spelt, are related to each other genetically as well as having a distinctive signature that suggests a MacGregor connection.

At the outset I promised a comment on each of the smaller groups listed in the DNA project page at www.familytreedna.com/public/macgregor.

All these comments relate to a comparison of the Y chromosome as the results appear on the FtDNA chart. It is possible that a male adopted the name of his mother rather than that of his father. Any relationship between such males cannot be verified genetically at the moment.

The two Andersons are not closely related genetically. The Gruar/Grewar pair ARE closely related genetically. 34571 M(a)cDonald is not closely related genetically to 90496 and N12949 but these last two are related genetically within the period of surnames. All the Magruders in the project are related to each other. 147767 McLachlan and 149586 McLaughlin are not related to each other or to 153144 McLaughin and 84701 McLaughlin. However I would expect these last two to be related through a quite recent common ancestor since it looks like 84071 has faster mutating DNA up to this point. The McNee/McNie pair ARE closely related and should look to find the common ancestor. 187136 Mitchell is not related to the other two in the group. A2881 could increase to 37 markers which would probably confirm a connection with 126082 Mitchell.

In the Moore group the four individuals are probably not genetically related in the time of surname adoption although it is hard to be absolute about 18134 who has only tested 12 markers. The two individuals in the Mustard group ARE closely related. In the Nichols group it is possible that 112863 and 202529 are related within the time of surname adoption but in that case would have more rapidly mutating DNA. The other two are not related either to each other in the Y chromosome and indeed 65000 belongs to the G haplogroup whereas the others are R1b.

I have spoken about the Orrs in previous updates. The two Reids are not related genetically, nor are the three Smiths (and 24436 has Viking DNA). In the Stirling (other) group 52255 Starling and 41885 Starling are closely related. The other two in the group are not related to each other or to these two.

At the moment I have not attempted to draw a chart of the ungrouped entries but will do this during the year for report in the next update. In the meantime if anyone in the ‘Ungrouped’ section would like to indicate to me if they have matches within the MacGregor project subgroups at 35/37 and 63/67 and above I will run their results within the group to which their matching project member belongs.

Finally a note about 67 matches. Keith MacGregor (Clan Gregor Society North America representative) has raised a question with me on the issue of public availability of contact information on order that individuals can contact directly (from the charts I generate) others who appear closely related through the various branches. Unfortunately as a result of the data protection laws in Britain I cannot release that information but I do give the kit number so that members of the project can easily identify each other using when using the Matches section of their personal DNA webpages on Family Tree DNA.

I would point out, however, that not all individuals with 67 marker matches are members of the MacGregor project, and in fact with a few exceptions those on the matches grid who show 5 to 7 mutations away from an individual often have surnames which are not related to Clan Gregor at all. These are not generally people whose ancestors adopted aliases but indicate how little genetic mutation has occurred in the last 2000 years in many families’ DNA. More distant 67 matches will almost certainly NOT have the characteristic marker mutations which distinguish members of the MacGregor bloodline. In particular we would expect to see the fourth and fifth markers as 10,10, the 12th marker usually at 30, and the 29th marker usually at 24. It would be unusual in the MacGregor bloodline profile for ALL of these markers to show mutation.

These reports are always longer than I imagine they are going to be and I try to be as clear as possible taking account of the fact that DNA mutates randomly and therefore the prediction of connections cannot be an exact science. As always I am prepared to answer questions from participants with regard to individual results. Please contact me on richardmcgregor1ATyahoo.co.uk.

The trees are generated using the program Splitstree4 for which the following reference applies:

D.H. Huson and D. Bryant ‘Application of Phylogenetic Networks in Evolutionary Studies’, Molecular Biology and Evolution, 23(2):254-267, 2006.

No comments:

Post a Comment