Total-evidence approach. Throw it all together and see which signal is strongest.
The biogeography doesn't seem right. Dispersal of semi-volant microtheropods around the J-K boundary would be predominantly E -> W IIRC and they'd have to do a generous bit of ocean crossing for which they were not well-equipped.
Then maybe the dispersal just happened earlier.
What seems certain is that highly interesting marerial is liable to occur in the E slope of the Urals. And perhaps also in some former islands in C Europe, but I think many have been rather well-explored.
There are extremely few sites from the J-K boundary.
What do you mean? Fusion in adults is the plesiomorphic condition here, and a joint the derived one. (Also found in *Rahonavis* and some or all of the Chinese long-tailed "birds", I forgot which ones. Not found in *Sapeornis* or confuciusornithids.)
What I mean is how to score the difference between enantis and euornis. Not as a binary character (fusion/joint).
How well is tarsometatarsus evolution documented?