Barton, As I would understand it, if we have a group who all share an IBD segment, then there would be, somewhere way back in history, a Common Ancestor for the group who was responsible for this segment. If you don't look at the whole group, parts of the group might have inherited the segment through a much more recent common ancestor. In your group of 4, one of your matches might be a 5th cousin, descended from the first child of the CA, while the other three might be your first or second cousins, all descended by the same route for many generations. You and your near cousin have a common ancestor who is much more recent than the common ancestor for the whole group. I don't think there's any reason to think that the descent had to split into 4 strands right at the start, as you say here. Regards, Paul Rakow > Let?s take that example (a 5th cousin as the most distant in the TG). > For me to believe that very well may be the CA, I, and my 3 matches would > need descend from 4 different children of the 4th great grandparent ? or > something pretty close (one of them from the same child but different > grandchildren, say). Otherwise, who is to say the CA is not some other > CA I share with one of my nearer cousins in the TG? On Tue, October 27, 2015 03:53, Barton Lewis bartonlewis@optonline.net [DNA-NEWBIE] wrote: > > > > When you say, ?I believe all the IBD segments in a TG come from the same > ancestor,? that?s where things get messy (for me). Previously, there?s > been discussion of different levels of cousins matching on a TG ? a 2nd > cousin; 3rd cousin once removed; and 5th cousin, say. On its face, that > sounds kind of specious; they?re all still descended from a CA, if it?s > the same family, right? But that?s just it: we don?t know for sure they > are ? if they?re from an endogamous population especially ? and unless > they have a very, very robust tree all the way back to the level of the > root CA. > > > > Let?s take that example (a 5th cousin as the most distant in the TG). > For me to believe that very well may be the CA, I, and my 3 matches would > need descend from 4 different children of the 4th great grandparent ? or > something pretty close (one of them from the same child but different > grandchildren, say). Otherwise, who is to say the CA is not some other > CA I share with one of my nearer cousins in the TG? We early colonials > have lots of endogamy; and ancestors and lines are going to be > duplicated, even (and maybe) especially where we don?t see them. >