R. Reisza and Johannes Mullera, 2004, Trends Genet 20 : 237-241.

aDepartment of Biology, University of Toronto at Mississauga, 3359 Mississauga Road North, Mississauga, ON L5L 1C6, Canada


The fossil record serves a crucial function as an external calibration for genomic clocks and molecular evolutionary timescales. Although certain portions of the vertebrate fossil record are accurate, there is always uncertainty in establishing a divergence time because the fossils can only provide evidence of the first appearance of the descendants of a split, and by definition they underestimate the date of the true evolutionary event. Nevertheless, recent workers have used fossil data for establishing external calibration dates without taking into account this limitation or the quality of the relevant fossil record. In this article, we present evidence that the fossil record is inadequate for proposing a reliable external calibration date for the mammal–bird split, the most widely used evolutionary event for the study of molecular evolutionary timescales. Instead, we propose the bird–lizard split as an alternative major evolutionary event, one of several potential external calibrations. Finally, we argue for closer, direct collaboration between paleontologists and molecular biologists.