General: Earth Clones (The Good, The Bad, and The Undone)Y’all probably knew this one was coming. This is not a discussion anyone likes to have on Sagan 4. But I’m also about to make a pretty bold claim.

The wolf shrew is the most realistic case of convergent evolution that occurred during its time.
I know what you’re thinking. This is literally a wolf, right? Well, what if I told you that it isn’t? While it looks vaguely doggish and has somewhat wolfy behavior, this is not what makes a wolf. The wolf shrew has a proportionally large, muscular head, a short neck, and a long thick tail. The nose is large and not particularly dog- or wolf-like at all. It’s a facultative biped and has an opposed hallux for stability. Whatever this creature is, though it has “wolf” in the name, it is certainly no wolf.
The wolf shrew is recognizable, but not exact. It has decidedly non-wolf characteristics, both intentional and not. It can be likened to a variety of large predators on Earth that are
not wolves, such as mesonychids and thylacoleonids, but is not a clone of either of those either. This makes it a far more realistic Earth clone than some of the others which evolved around the same time (and sorry Cheatsy, I’m gonna rip into some of your childhood drawings a bit)...

The hornhog is
literally just a babirusa with a sail and plent features. It even has “shock absorbers” that look like nostrils to complete the look. While convergent evolution can and does produce familiar forms sometimes, this is far too extreme--it is incredibly rare for it to produce something like animal X but group Y, especially when novel ornamentation gets involved. Usually, the results of convergent evolution are a mixture of characteristics from multiple familiar groups working together to make something both familiar and unique, something that the wolf shrew accomplished but that the hornhog did not.
So, what about fixing or undoing bad earth clones like that, making them more alien? Well, there’s certainly some completely
wrong ways to do it.

The climbing cantro was an attempted “fix” of an alleged earth clone, the cantro. The fact that the cantro was most certainly not any more an earth clone than this aside, the climbing cantro takes the completely wrong approach to alienification. It randomly moved all its eyes into a fragile, skull-snapping position. It lost bipedalism and got small arm-like feet for no reason. Its direct descendant, the chastro, became a foreleg biped for no reason other than to “alienify” it even more. The changes that were made to “fix” it make no logical sense, the organism is even more broken than before and shouldn’t even be able to function, and the changes don’t even do a good job of making it more alien because it somehow looks even more like a wolf than any other dromaeocanid, even the wolf shrew itself.

Another great example of how not to undo an earth clone is the kangatwail. The approach taken here seems to have been to duplicate body parts, giving it three pouches and a forked tail. This is about the same level of “alien” as your average star trek humanoid, and the creature is still fundamentally just a kangaroo with extra stuff slapped on.
So, how
does one fix a bad earth clone?
First, let me state that I do not support wiping out earth clones or alienifying them just for the sake of getting rid of the terran characteristics. This is just an example of how it is possible to make something that is not an earth clone from an earth clone with only plausible, natural changes.

In the thaw following the snowball event, excluding an odd offshoot of pipents that was recently retconned to have survived, the Trogagon and its descendants became the last of the non-gundi nodents still alive. The Trogagon has just a single feature that makes it unlikely to produce anything resembling an actual rodent ever again--its incisors basically became a beak for cracking crystal flora and don’t resemble the “buck tooth” ancestral to nodents at all. The majority of its descendants have followed suit, and now have beaked faces rather than rodent-like. This one little change alone makes any plausible elaboration of the species unrecognizable as once having been a rodent clone, as they converged on faces more typical of birds, ornithischians, and turtles than of any mammal.
And speaking of plausible changes resulting in a more alien appearance from non-alien ancestors, we must also talk about tuskents!

Tuskents are also derived from nodents. They weren’t alienified on purpose--they are the result of a natural progression into becoming a marine creature which eventually “unwhaled” back to land. The entire hand and foot had been modified into polydactyl, wooden, claw-based paddles convergent with swimming beetles or those nail-swimming things from Snaiad, and the original elbows and knees became immobile. As a consequence, in living secondarily terrestrial species the entire forearm and lower leg are hand and foot, respectively, and they walk on the two innermost digits of each limb and have a very large number of dewclaws.
All of these things came about as adaptations for the time without intent to make something super alien, just to adapt to new environments and new niches, and suddenly this aberrant branch also became the only survivors of the once mighty pipents. In a sense, they can be thought of as the monotremes of nodents: They’re the most derived, yet most early stem-tuskents are no more alien or aberrant than the lineages leading up to modern trogagons and gundis. It’s unlikely that tuskents will ever produce forms resembling their earth clone ancestors no matter how much convergent evolution occurs, standing as an example of the second way to make non-earth clones from earth clones: by creating a lineage that gets so derived that it’s unrecognizable, yet is also completely plausible.
That’s how seed worlds like Serina do it, after all.
(these are canaries!)