Running tally
What you just did has a name
The rate at which a phenotype evolves is not just about selection. It is also about how many mutations can produce it. Webbed feet and eye loss are easy — any of several genes can be tweaked and you get the phenotype. They arise again and again. The placenta is hard — it required a specific transcription factor binding a specific transposon that happened to land in exactly the right place. It evolved roughly twice in the history of vertebrates. Hinged jaws and the tetrapod limb are harder still; those evolved once each.
The pattern in the tally reflects two separate multipliers on the per-generation rate of evolution:
- Mutation target size. How many bases, how many genes, can mutate and still produce the phenotype?
- Selectability. Conditional on producing the phenotype, how strongly favored is the mutant?
A trait with a huge mutation target that is strongly favored in some environments — eye loss in caves, webbed feet in aquatic lineages — appears in every clade the environment recurs in. A trait with a tiny mutation target — hinged jaws, true flight, bone — appears once or twice in the entire history of life and never again. The parallel-evolution count is a clean empirical readout of this distinction.