Margaret McKelvey married Charles McIlwain

I don’t want to sound too pessimistic but it’s very unusual to be able to trace Irish ancestry back to the 1600s. It does sometimes happen but normally only for nobility or others with lots of land or wealth likely to make the pedigree researchable. The other 99.5% of our population can’t do that. The records simply don’t exist.

You are enquiring about the ancestry of Margaret McKelvey who died in 1680. We don’t have any birth, death or marriage records in Ireland that are likely to include her. (You haven’t said what denomination she was but at that distance it probably doesn’t matter. No denomination has records for someone born c 1640 or earlier.)

You mention Duncan taking his mother’s surname. Perhaps. I can’t say. But you probably need to know that McIlwain is a variant of McKelvey (according to MacLyaght’s The Surnames of Ireland). So it’s the same name. We have loads of interchangeable names in Ireland. Robinson/Robertson; Connor/Nogher. I could provide many more examples. I think someone may have got confused with this family when doing the research. They may have assumed the 2 different names on different documents indicated the father’s surname and the mother’s maiden name. Perhaps not.

There are an impressive 989 trees on Ancestry with this family. Many show Margaret’s maiden name as Adams, not McKelvey. That said, I don’t think the trees I looked at are very accurate. I have not done any detailed research but an initial impression is they may be faulty. In the case of the Furman tree, it shows Duncan born in Ballymena, Co Antrim in 1680. He then apparently had a daughter Elizabeth 150 miles away in Fermanagh in 1695 (when he would have been 15). His father Charles reportedly died in Donegal in 1696 (120 miles from Ballymena). Duncan then had a daughter Margaret in Tyrone c 1700. He married Ann Kyle in 1701 in Bracky parish in Fermanagh (no such parish) but there is a townland of Bracky in Derryvullan parish.

Father Charles was reportedly born in Ballymena, Co Antrim in 1660. 4 siblings were reportedly born in Ayr (Scotland) between 1663 and 1670. Then sibling Joseph was born in Aughnacloy in Co Tyrone c 1675 followed by Duncan in Ballymena in 1680. Then Charles reportedly died in Donegal in 1696. There’s an awful lot of travelling going on here. Why? There’s a gravestone on Findmygrave which reportedly confirms Charles’ death (I can’t actually read the inscription). But I am worried by the fact that it looks to me to be a 20th century gravestone not 17th century and it also looks to be Catholic whereas I suspect your family were not, if they came from Ballymena or Scotland. But the family movements make no sense at all. Leaping about between Scotland and various parts of Ireland. I think folk may have been guessing events off the internet. Making the few early Irish & Scottish records that do exist fit their expectations.

But the fact remains there are no records in Ireland for the mid 1600s that are likely to trace your line back further. (Most Irish research comes to a stop around 1800 for that reason, never mind in the 1600s). But before worrying about that, I’d be inclined to check your tree back step by step looking for perhaps more reliable information about their ancestry.

Possibly DNA testing may be a way of matching with others who have additional information about where the family originate. Family Tree DNA reportedly has more people with Ulster roots than any other company. That obviously increases the chances of finding a match. You might want to try them or, if you have already tested, you can transfer your results to them for no fee.

The North of Ireland Family History Society is running an Ulster DNA project and can offer FTDNA testing kits at a reduced price. http://www.nifhs.org (Go to DNA project on the website).