Just hazarding a guess based on pterosaurs and birds: 1) bipedal configuration, fingers get smaller – or maybe not in the case of bats which probably remained inverted and arboreal throughout (Sharovipteryx, Coelophysis); 2) fingers get longer, folding mechanism added (unnamed fenestrasaur, Velociraptor), 3) fingers (or fingers + feathers) get much longer (eudimorphodontids, Archaeopteryx) and become volant, no longer used for grasping, climbing, walking.
Perhaps.
<If you look at Ptilocercus and Nandinia, both are quadrupedal climbers, but both hold their prey with their hands. Both enjoy inverted locomotion. Nandinia likes to jump out of trees. Ptilocercus has a fairly naked tail. Not much, but it’s a start.>
It is my humble opinion that you're reading too much into this argument. There are 1) convergences in anatomy, and 2) convergences in behavior.
Of course, Drs. Martin and Feduccia, but added to the suite of morphological characters, these add evidence to the case.
Jaime, I'm just giving you the cherries on top. There's so much more in the dataset.