Relative importance of molecular, neontological, and paleontological data in understanding the biology of the vertebrate invasion of land

Abstract

Meyer and Wilson's (1990) 12S rRNA phylogeny unites lungfish and tetrapods to the exclusion of the coelacanth. These workers also provide a list of morphological features shared in common between modern lungfish and tetrapods, and they conclude that these traits were probably present in their last common ancestor. However, the exquisite fossil records of the abundant extinct lungfishes and rhipidistians show that at least 13 out of Meyer and Wilson's 14 supposed ancestral traits were not present in the last common ancestor of lungfishes and tetrapods. Using extant taxa to infer ancestral morphologies is fraught with difficulties; just like molecular sequences, ancestral character states of morphological traits may be severely overprinted by subsequent modifications. Modern lungfish are air-breathing nonmarine forms, yet their Devonian forebears were marine fish that did not breathe air. Fossils dating from the time of origin of tetrapods in the Devonian offer the only hope of understanding the morphological innovations that led to tetrapods; morphological analysis of the "living fossils," the coelacanth and lungfish, only lends confusion.