Molecular clocks, molecular phylogenies and the origin of phyla

Protein, RNA and DNA sequences have been widely used to construct phylogenies and to calculate divergence times using a molecular clock. Reliance on molecular information is particularly attractive when fossil evidence is missing or equivocal, as in the Cambrian metazoan radiation. I consider the applicability of molecular clocks and phylogenetic analysis of molecular data to the origin of metazoan phyla, and conclude that molecular information is often ambiguous or misleading. Amino acid sequences are of limited use because the redundancy of the genetic code masks patterns of descent, while in a nucleotide sequence only four potential states exist at each site (the four nucleotide bases). In each case, homoplasy may often go undetected. The application of a molecular clock to resolve the timing of the metazoan radiation is unwarranted, while molecular phylogenetic reconstruction should be approached with care. A potentially more useful technique for phylogenetic reconstruction would be the use of patterns of genome structure and organization as characters.