Y-STR clustering alone was not able to answer our questions regarding the various 21,21 families. Yes, this method is able to predict future SNP results many times, but not always. "AS-generics" were previously grouped in one big blob, and now are turning out positive for all sorts of different SNP's. Besides, there are individuals whom the I1 project had predicted to be positive for a specific SNP and who turned out to be negative for it (Z58 comes to mind). Y-STR clustering does not answers all questions. That is why SNP tests are always necessary. Both things are complimentary. One shouldn't exclude the other. Aaron Torres [DNA] Fw:  New I1 Haplogroup Project: Z58+ and Z63+/T2 To: <y-dna-haplogroup-i@rootsweb.com>, <genealogy-dna@rootsweb.com> Message-ID: <ABAE0B04A3FB486C9A23DE9A56BBD499@kenPC> Content-Type: text/plain; format=flowed; charset="iso-8859-1";     reply-type=original [[Not exactly; the Z snps have so far mainly given formal credentials to the I1 clades earlier found by examining the off-equilibrium clustering of extended STR haplotypes (sort of like giving marriage licenses to common-law couples on their 50th anniversaries), and helping sort out the time relationships of those clade foundings. The I1 generics will be sorted out in a more timely fashion with the Z snps, but this will still be a very slow process, and it could be speeded up by the (more costly) process of extensions to 111 STRs. KN ]]