It sounds like the principal identifying feature of a dinosaur and bird is the shape of the hips and this separates them from the lizards.
So the dinosaurs split off from the lizards, and then later their scales turn into feathers and they shrink to bird size. The shrinkage across the whole group is easily explained, the big ones are destroyed by a comet, climate change, etc. But how does the entire class of animals develop feathers? We seem to be pretty confident that they were not feathered when they split off from the lizards. But by the time they would have developed feathers, there would have been thousands or tens of thousands of different species of dinosaur. If one develops feathers, he’s not mating with any other species than his own. Feathers would have to independently rise across the entire dinosaur line (at least among the small ones), which seems highly improbable.
I could suggest that, because of the shape of the hips, dinosaurs are inherently bipedal and maybe scales are more heavy than they are worth. Having lighter feathers allows you to run on two legs faster and being faster is better than being armored. So, there, we have an evolutionary force that would act similarly across the entire set of animals who are bipedal.
But it seems like we should still see at least a few true dinosaurs, scaled bipedal animals. With all of nature’s diversity, I would think that scales would have won out in at least a few cases? Maybe they exist?
If not, is there any other explanation for the full transfer? Is there sufficient overlap between species that even though A and C can’t make babies, A and B can and B and C can, so eventually the whole set of dinosaurs are able to transfer one mutation (e.g. feathers) across the entire group? A and C are, effectively, mating with each other through a slow process of indirect cross-species mating?
And what about the beak?