In a message dated 7/8/2011 6:02:55 P.M. Central Daylight Time, weh8@verizon.net writes: These two groups are not independent. The 683 batch contains the 320 batch, but if they were significantly different, they would show it in these statistics. Almost by casual inspection a comparison of the two results proves my point that the two are statistically indistinguishable. Whether one uses the first batch or the second batch, the results should also be statistically the same. My results on the date of origin of M222, using both batches of data, also yield consistent results. Shouldn't this email be titled M222+ and M222 untested rather than M222-? There are no known M222- in the project except a few classified as M222- and pre M222 and none of these are included in any project spreadsheets. If the two are statistically indistinguishable my guess is that would mean there are few if any M222- in the project despite the lack of SNP tests. For those interested in such things the latest version of the M222 spreadsheet is online at the M222 site. It has a separate worksheet for SNP tests - the 37 marker worksheet is a subset of that group. John