Introduction
Molecular clocks and the fossil record— towards consilience?
Molecular clocks or the fossil record: which approach tells evolutionary time? The papers
presented in this volume stem from a joint Palaeontological Association/ Systematics
Association symposium held at the Third Biennial Meeting of the Systematics Association,
at Imperial College, London on 5 September 2001, which brought together
palaeontologists and molecular biologists with the aim of addressing the disparity between
molecular and palaeontological perspectives on evolutionary time.
Resolving the timing and tempo of major evolutionary radiations represents one of the
most prominent and controversial debates currently underway in evolutionary biology.
Although this temporal dimension has traditionally been the preserve of palaeontology,
with a burgeoning molecular database that is progressively more representative of the tree
of life, attempts to establish a molecular timescale for evolutionary history have become
ever more sophisticated. The so-called ‘molecular clock’ works, at least at a simplistic
level, by calibrating molecular distance between a pair of organisms to time using a
reliable fossil estimate. By extrapolating this scaling it is possible to estimate the time of
divergence between other taxa, for which the record is perhaps unreliable, or simply as a
test of fossil-based estimates. Because of the unlikelihood of the fossilization of the earliest
representatives of taxa, even the apparently reliable fossil records of taxa used in
calibrating the clock can provide only a conservative estimate of the true divergence date,
and so it is to be expected that molecular clock estimates should also be conservative.
However, this is not the case; molecular clocks almost always provide divergence
estimates that are not just older, but considerably so (e.g. Hedges, Chapter 2), and
sometimes as much as double the age of fossil-based estimates. By implication the fossil
record is not just incomplete, but represents only half of evolutionary history, and the
latter half at that.
No one would defend total completeness of the fossil record, but it is the degree to
which evolutionary history is unrepresented by the fossil record that is in question (Fortey
et al., Chapter 3; Benton, Chapter 4). Indeed, the gross disparity between molecular and
fossil-based estimates has led to a period of introspection amongst the palaeontological
community, and the development of a variety of methods for the evaluation of the quality
of the fossil record. These include attempts to assess confidence in the fossil record as a
reflection of the true time of origin of a particular clade (e.g. Benton, Chapter 4;
Donoghue et al., Chapter 10). These methods are becoming progressively more realistic
and, inevitably, more complex in their application (e.g. Tavaré et al. 2002), but they do
provide a means of testing the timing of evolutionary events within the bounds of