statistical confidence. Methods have also been devised to assess congruence between the
order of stratigraphic appearance of taxa in the fossil record and the sequence of evolutionary
branching as inferred from phylogenetic hypotheses constructed independently of
temporal data. These represent powerful independent tests of the fossil record, although
they are not without their attendant problems (Paul, Chapter 5). All of these techniques
can be used to identify and qualify the veracity of potential calibration points for
molecular clock analyses.
The fossil record has also become better understood through significant advances in
systematics and classification with, for instance, the gradual elimination of paraphyletic
groups. But perhaps the greatest advances have been made through the better
understanding, application and realization of the implications of total- and crown-group
concepts (Jefferies 1979; Budd & Jensen, Chapter 9; Donoghue et al., Chapter 10; Dyke,
Chapter 12). Hitherto enigmatic fossil taxa that have resisted classification because they
exhibit only a subset of the characters necessary for inclusion in groups defined on the
basis of their living members (crown-groups) have subsequently been recognized as
reflecting interim stages in the assembly of crown-group body plans. The augmentation of
such fossil taxa to their nearest living crown-group constitutes the more universal total-
group, and the difference between total- and crown-groups discriminates the paraphyletic
stem-group. The recognition of stem-groups is important not only because it provides an
understanding of the steps through which clade divergence proceeded, but also because
stem taxa are frequently older than their fossil crown-group relatives and therefore lead
frequently to significant extensions to the inferred time of divergence of one crown-group
from another. The discrimination of stem-from crown-group taxa has also proved to be
profitable because it provides more appropriate time estimates for clade divergence in
comparison with those derived from molecular clock analyses which are, rather ironically,
usually framed within the context of total-groups.
Reanalysis and revision of the fossil database has raised the possibility for
rapprochement between fossil- and molecular-based approaches. Generally, however,
optimism has not been rewarded. In short, two main camps have arisen amongst
practising palaeontologists, delimited by (a) those who accept the molecular clock
estimates, concluding that the fossil record is corrupted by its reliance upon negative
evidence (Fortey et al., Chapter 3) and, (b) those who find that the fossil record exhibits
no evidence for a significant deterioration in quality of the fossil record through time, and
conclude that molecular clock estimates are spurious, for one reason or another (Benton,
Chapter 4; Wellman, Chapter 7; Budd & Jensen, Chapter 9). Indeed, if there is any
agreement, it is that pictures of evolutionary history widely accepted less than ten years
ago are incorrect.
Molecular clocks, in turn, are not without their own attendant problems (Rodríguez-
Trelles et al., Chapter 1; Fortey et al., Chapter 3). Although the molecular clock originally
found support in Kimura’s neutral theory of molecular evolution, observed rates of
substitution are generally much higher than expected. In attempting to allow for this,
various supplementary hypotheses to the neutral theory have been invoked, the validity of
which are explored by Rodríguez-Trelles et al. (Chapter 1). However, the theoretical
basis underlying most recent analyses is more empirical, relying instead upon a so-called
‘law of large numbers’ (Rodríguez-Trelles et al., Chapter 1; e.g. Hedges, Chapter 2).
2 INTRODUCTION