This grows more interesting as the length of this thread grows!
Who would have thought that a form as "common"! as a FutPerf would generate a thread now into its 3g page?! Thanks Ken, for digging out the TLG data. I can't access TLG from home and haven't been back on campus, but you've saved me a lot of work. I still need to think through all of it.
As for Smyth, I'm a bit puzzled, so Dr. "Carl-Smyth"
may need to take me in hand here. The discussion in 447 pertains to what I've usually called "iotat redup." which is the normal form for the present in the μι verbs, in contrast to "normal"/ε redup. used in the perf. And 448 lists only 2 verbs, ἀγω and φερω, that have a duplicated syllable at the beginning of the stem in 2Aor--but I've never thought of them as "reduplicated" forms that have the same morphology as perfects. Perhaps I've just used "reduplicated" in a narrower sense (maybe narrower than I should?).
I already noted Smyth's listing in 1955:
But then look at Smyth, §1955, who lists 10 future-perfect monolectic forms, including 3 κεκ- forms: ἀναγεγράψομαι, δεδήσεται, κεκλῄσεται, πεπράξεται, εἰρήσεται, κεκλήσομαι, μεμνήσομαι, κεκτήσομαι, τεθνήξω, and ἑστήξω. He does refer to some of these (κεκλήσομαι, μεμνήσομαι, κεκτήσομαι in particular) as being "used like a simple future," but that is a statement of pragmatic function, not morphological formation.
Likewise on p. 703 in Smyth's Verb List he shows κραζω with a futurePerfect form κεκραξομαι, noting that it is "as fut."--which I assume means "is used with the same meaning as a future" (similar to his preceding comment, "2 perf κεκραγα as pres.").
Maybe I'm missing some other data in Smyth somewhere or not understanding what I've cited correctly, but regardless of what may be true of other forms listed in other posts above, it sure sounds to me like Smyth is treating the reduplicated κεκ- forms of κραζω as legitimate, monolectic, futurePerfect forms of κραζω.
So what have I gotten twisted up here?