OK, first, I apologize for the "may" crack, if it offended. I'm looking back over the thread and agree with Tom's 10 March 09 post (and Bill's 12 March 09 post) I find it hard to believe there are only two testing scenarios: the Plan's "3 months, next entry date" census, or the Code's "12 months, 1/1 or 7/1 entry date" census. I think anything in between if fair game. Not naming names or choosing employees, but picking a cutoff (alternative entry) date. (I think the Code's provision is to prescribe a maximum, not a one-and-only). I think if I could legally design a Plan's eligibility to match the testing method I want to use to test, I should be able use that method. I would exclude "all" hired after the cut-off date I pick. I would include all otherwise excludable participants hired before the cut-off date.