Reassessing evolutionary relationships of scleractinian corals

Abstract

The widely accepted family tree of Scleractinia published by Wells, based on a combination of morphological coral taxonomy and the fossil record, has recently been revised by Veron. It is now possible to test the validity of some of the conclusions reached by these and other authors by the use of molecular techniques. This paper reviews the results to date. Studies of ribosomal DNA have shown that the Scleractinia are monophyletic, i.e. derived from the same ancestral taxon. Extensions of this same data set now indicate that the Poritidae and Dendrophylliidae, with their fossil antecedents, may each warrant separate suborder status. They further suggest (a) that the Suborder Faviina (faviids, mussids and their allies) should probably be retained as a monophyletic group and (b) that Wells' original account of the isolated position of the Pocilloporidae and Astrocoeniidae is correct. These conclusions all accord with Veron's family tree. However, the Fungiina, even after removal of the Poritidae, are unlikely to be a monophyletic group at suborder level. The molecular data further show that externally observable morphological characters used in the taxonomy of extant corals distinguish families more reliably than do internal micro-skeletal characters frequently used in coral palaeontology.