I would caution against over interpretation of the results to date as the sample size is still quite small. For example, if someone could and was willing to go on a campaign to solicit and pay for all of their STR 67 near matches, e.g. I have 145 within a GD of 7, then I suspect I could single handedly skew the R-M222 tree so that the *FGC4087 subbranch under FGC4077/FGC4078* would appear to have become dominant. So perhaps a better estimate would be to browse through the STR-67 matches for all of the current members of the R-M222 project, collect the list of names, then eliminate overlaps so as not to overcount and see what the totals then look like using just the SNP headers we now have. (One would have to have project admin status to do this in order to access the individual Y-STR match lists from the R-M222 GAP page, to download, and to collate the data.) Doing this would help to reduce any sampling or testing bias with the current R-M222 subgroup test cohorts. That is it is unknown whether by chance or design if there are one or more groups who have done more SNP testing than others. This exercise would also give an estimate of the total number of R-M222 testers in the FTDNA database whether they have or have not joined the R-M222 project. The results may or may not show the same distributions as the current SNP testers headcounts. Walter Freeman On 5/4/2014 1:17 PM, David Wilson wrote: > I don't think it is news to anyone that S660/DF109 is the monster sub-branch > within M222, but I was just idly pushing some numbers around on a warm > weekend morning and realized that it may be an even larger group than I had > calculated the last time I went through this exercise. Perhaps others would > be interested in these stats. > > > > Iain Kennedy's tree now has 117 individuals in it, if I counted them > correctly. (I did this hastily, so I may be off by one or two.) > > > > Of these, the S568 group (now including PF1169) contains 7 individuals. > > > > The FGC4077 group (which has yet to produce a major subhaplogroup to divide > its members) contains 15 individuals. > > > > The 97 others are all located in S660. > > > > In short, about five-sixths of M222 individuals are in S660; the other > one-sixth are split between the two smaller groups, with FGC4077 twice the > size of S568. > > > > Within S660, if you exclude the major S588 and DF85 branches, just under > one-fourth are S660* and members of the recently identified small branches. > The S588 population makes up a little over one-fourth and the DF85 branch > about one half of the entire S660 group. Of that large latter branch, DF97 > is the largest subordinate classification. With 32 members it makes up > two-thirds of the DF85 branch by itself and amounts to more than one-fourth > of the entire M222 population. DF97 may be that large because of statistical > oversampling of some surnames and kindred groups, so these percentages may > change a bit as time goes by. > > > > David Wilson > >