which molecular phylogenies and ‘clocks’ may be used to test them. The latter come with
another battery of qualifications. We are particularly concerned with what has been
termed the Cambrian evolutionary ‘explosion’. Even here, we find confusion in what data
might or might not support the idea of an ‘explosion’. For example, Ayala et al. (1998) in
a highly critical essay estimated the protostome-deuterostome split at 670 Ma and the
echinoderm-chordate split at 600 Ma, claiming that these supported palaeontological
estimates. In fact, these estimates allow for a>130 myr phylogenetic fuse before the base
of the Cambrian.
Evolutionary radiationsMajor evolutionary radiations are apparently rapid bursts of evolution—often prompted
by ‘ecological release’—recorded by the short-term appearance of many morphologies
(and hence high-level taxa) in the fossil record. Familiar though they are, in some
examples a question that remains is whether the radiation coincided with the actual
appearance of all of the major clades, or whether some or all required an antecedent
period wherein their relevant synapomorphies were acquired one by one. This is what
Cooper and Fortey (1998) termed the ‘phylogenetic fuse’ and Jablonski and Bottjer (1990)
the ‘macroevolutionary lag’. During this formative period of time it has been suggested
that the groups in question may have been rare, and/or small, or unlikely to fossilize for
some other reason. On the contrary, some analyses of the fossil record (Sole et al. 1999;
Benton et al. 2000) have suggested that it is adequate to record true ranges, and that rarity
is not a problem. None the less it remains difficult to understand the significance of an
‘absence’ below a radiation in a way that is not the same as an ‘absence’ above an alleged
extinction, since only in the latter case do we know the previous existence of a clade, and
can hence statistically assess the increasing improbability of its survival above its last
record. The taphonomic rules may have been different during ‘fuse’ phases.
It is not unreasonable to invoke an early innovative phase involving organisms of small
population size, and/or small absolute size (Martin and Palumbi 1993). It has been
claimed that evolutionary rates are higher in small taxa, not least because reproductive
rates correlate negatively with size. Morphological change may also be accelerated among
highly vicariant and isolated populations. The problem is that direct fossil evidence of such
a phase is, in principle, likely to be rare, and hence impossible to distinguish from other
reasons for absence. This is why independent testing of divergence times from molecular
evidence becomes so important.
However, it is possible to test the ‘fuse’ idea by looking at a converse example: where
the early representatives of a major radiation are known to be large animals. The best case
is probably terrestrialization and the origin of tetrapods. The Devonian complex of
animals in this transition (for a recent summary, see Ahlberg and Johansen 1998; Ahlberg
2001) including the well-known Ichthyostega are about one metre long and robust.
Although their occurrences may be localized, they are also easily fossilizable by virtue of
the habitat in which they lived. They also record a selection of what might loosely be
described as ‘hopeful monsters’. So far as we are aware, there is no suggestion that there
was a prior history of major clades in the ‘fish’-tetrapod transition, and the sister taxa of
early tetrapods are selectable from near contemporaries in the osteolepiforms. In short, if
PHYLOGENETIC FUSES AND EVOLUTIONARY ‘EXPLOSIONS’ 43