*1 Cf. PN *tutu.4, *tugi, *tuugaki "signal fire"
*6 PAN *(tt)u9(tt)u9 "light, kindle, burn" (Bst. 1979).
*8 Note. Given Blust's reconstruction, the original basic form must have been *TUTU, with transitive/passive *TUTU-GI-A. A common reanalysis makes this *TUTUGI-A. *TUGI could then result from interpreting *TUTUGI as CV-reduplicated. And *TUGI-A would be the transitive/passive of this. This does not account for the long *U in the last form, which is well supported throughout PN. Do we have a general rule for this? or could it be a reduction from *TUTUGIA, dropping the medial T? In any case, *TUTU and *TUUGIA are by far the most widespread forms, and in several languages are an active/passive pair. I guess I would favour retaining *TUTU.4 and *TUUGIA as the two entries, suitably cross-referenced. The question then becomes what to do with the *TUTUGI and *TUGI forms -- to include them with *TUUGIA on the grounds that they include the historical thematic consonant, or are post-PPN backformations; or with *TUTU on the grounds that they were probably grammatically variants of the base rather than the suffixed form? (Clk).