The following is a direct lift from the Wiki article on Modular Evolution with critical comments added in red.
Mosaic evolution (or modular evolution) is the concept that evolutionary change takes place in some body parts or systems without simultaneous changes in other parts. Another definition is the “evolution of characters at various rates both within and between species”.408 Its place in evolutionary theory comes under long-term trends or macroevolution.
By its very nature, the evidence for this idea comes mainly from palaeontology. It is not claimed that this pattern is universal, but there is now a wide range of examples from many different taxa. Some examples:
- Hominid evolution: the early evolution of bipedalism in Australopithecines, and its modification of the pelvic girdle took place well before there was any significant change in the skull, or brain size.
Unfortunately this narrow and hopeful view ignores the fact that paleontologists can identify an australopithecine from its skull and teeth, vertebrae and other body parts. So, no modular evolution here.
- Archaeopteryx. Nearly 150 years ago Thomas Henry Huxley compared Archaeopteryxwith a small theropod dinosaur, Compsognathus. These two fossils came from the Solnhofen limestone in Bavaria. He showed that the two were very similar, except for the front limbs and feathers of Archaeopteryx. Huxley’s interest was in the basic affinity of birds and reptiles, which he united as the Sauropsida. The interest here is that the rest of the skeleton had not changed.
And yet paleontologists can identify Archaeopteryx by the pelvis, skull, tail, etc, whether wings or feathers are present or not. The differences are present, even if subtle. So, no modular evolution here.
- Meadow voles during the last 500,000 years.
This example is reported at the population level. Nothing here strong enough to differentiate one species from another.
- The pterosaur Darwinopterus. The type species, D. modularis was the first known pterosaur to display features of both long-tailed (rhamphorhynchoid) and short-tailed (pterodactyloid) pterosaurs.
Unfortunately, Darwinopterus was not the transitional taxon between the long tails and short tails when you add more taxa to your analysis. Darwinopterus was an evolutionary dead end with no Cretaceous descendants. Sure it had a long skull and long neck, like pterodactyloids. But anurognathids had a short tail like pterodactyloids and only Andres thinks they were transitional. So, no modular evolution here. Convergence, yes!
- Evolution of the horse, in which the major changes took place at different times, not all simultaneously.
Evolution from tiny four-toed pre-horses to large one-toed modern horses indeed took place over many millions of years. Certain traits occurred simultaneously. Others waited until later. So, no modular evolution here.
Not sure what the reporter has in mind here, but certainly there were several clades that demonstrate convergence. Mammal experts are famous for being able to identify their specimens from a few teeth alone. So while other parts were evolving, so were the teeth. So, no modular evolution here.
If anyone can provide an example of modular evolution in vertebrates, please bring it to the attention of the Wikipedia author of this article. If modular evolution were indeed present, paleontologists would be confounded by one end of the body not matching the other. Phylogenetic analysis usually takes care of all such problems.
And convergence is out there ready to trip you up if you’re not careful.