TOP: The most favoured hypothesis of basal gnathostome relationships prior to the discovery of Entelognathus. Osteichthyans "reinvent" boniness.
BOTTOM: The aftermath of discovery based on our favoured hypothesis. Osteicthyans "inherit" boniness. Note that the exact phyletic position of Entelo is still slightly ambiguous with regards to other arthrodire-grade placoderms.
(I am not serious)
A friend of mine linked me here, and if you don't mind a question from a total layperson -- what led to the idea that the bone content of the skull and fin girdles disappeared, then later reappeared in the first place? Hindsight is 20/20, of course, but wouldn't it have been simpler to posit an intermediate form from the start at which the two branches that lost and retained these diverged, as shown at the bottom?
I'm guessing that this was considered and rejected in favor of alternatives considered more likely, but I'm curious as to why.
Sorry if this is a stupid question. I'm not a evolutionary biologist (or in fact any kind of biologist)!
Okay - The acanthodian Acanthodes has long been recognised to have a very osteichthyan-like braincase, thus suggesting a close kinship between ray-finned fishes and acanthodians.
Acanthodes is one of the very last (Permian) and most specialised of acanthodians. Recent analyses have shown that some earlier acanthodians (eg. Ptomacanthus) do not have bony-fish style neurocrania. So most recent studies have resolved a paraphyletic Acanthodii, with some acanthodians closer to sharks and others closer to bony fishes (as seen in the upper tree). This reinforces the scenario of a "shark-like" (acanthodian-like) common ancestor of modern jawed vertebrates.
With Entelognathus added, all the acanthodians bunch together on the chondrichthyan stem, away from the bony fishes. This may imply that the last acanthodians in the Permian (the ones that hadn't evolved into sharks) convergently developed a osteichthyan-like braincase.
Also worth noting
* Acanthodians we're for a long time the only known gnathostomes from the Silurian, thus were considered by many to be close to the ancestral gnathostome stock. Now we know a much wider range of jawed fishes were around in the Silurian.
* Some prominent palaeontologists favoured the hypothesis of a close kinship between sharks and placoderms exclusive to acanthodians and bony fishes.
* For a long time actinopterygians were considered more "primitive" than sarcopterygians. The most "primitive" actinopterygians from the Devonian (eg. Cheirolepis) have superficially acanthodian-like tiny scales. Recent discovery of Silurian osteichthyans like Guiyu show that big-scales is the default setting for the bony fishes and the scales of Cheirolepis are derived.
There are some other characters as well, but hopefully you get the idea - with an increasing amount of complete fossils from the Silurian (where previously we had none), many of the assumptions of what is anatomically "advanced" and what is not (based on Devonian fossils) are being questioned.
Also don't forget, that sharks have long been dogmatically viewed as "archaic" or "living fossils" while bony-vertebrates (ie. the group containing humans) must be "advanced" - sort of a holdover from old 'great chain of being’-style philosophies. Even with the advent of modern comparative and phylogenetic techniques, this thinking has clouded our vision.