the trap of assuming that similar features must be convergent, a tendency recognized long
ago in the memorable words of Derek Ager: ‘The fashionable fixation for homeomorphy
in many groups brainwashed many of us into thinking: “if they look alike they cannot be
related”’ (Ager 1993, p. 90). At the very least, the last common ancestor of Bilateria must
have possessed muscles and a head, suggesting it was not a tiny planktonic organism. If the
Precambrian was really characterized by the slow diversification of the stem-groups of the
bilaterian phyla, then ad hoc arguments must be employed to explain why animals that can
be reconstructed to be of large size and biomechanically complex left neither a trace nor
body fossil record. The common argument (e.g. Thomas 1999) that the presence of phyla
characterized by small body size that have no fossil record (e.g. kinorhynchs) suggests that
similar animals could have ‘silently’ existed in the Precambrian misses the point: the body
plan features of bilaterians such as muscles and the coelom could not have evolved in such
an organism (see discussion in Budd and Jensen 2000). Phylogenetic reconstruction of the
history of such features, combined with consideration of the biomechanical regime in
which they would have been adaptively useful, concisely rules out the possibility of the
most ancestral bilaterians being tiny and/or planktonic. The only possibility is that the
body plan features of groups of phyla were acquired independently of each other, a feature
of some earlier attempts to reconcile molecular and fossil evidence (e.g. Bergström 1989;
Davidson et al. 1995, and, most recently, Erwin and Davidson 2002). Even so, this can
hardly apply to important features of the deuterostome taxa mentioned above.
Further doubt is cast on this model by the increasingly recognized presence of
members of the stem-groups of bilaterian phyla in the Cambrian (e.g. Budd 1993; Mooi et
al. 1994; Conway Morris and Peel 1995; Holmer et al. 2002; Williams and Holmer
2002). If they are preserved in the Cambrian, why not any in the Precambrian? One is
forced to return to the view that, if there truly is a long Precambrian tail of bilaterian
diversification, it was of animals that should have been capable of leaving a trace or body
fossil record. The presence of Ediacaran fossils and many other sediments of suitable age
and type in the terminal Proterozoic means that one cannot suggest lack of outcrop as a
reason for the lack of evidence (Darwin 1985). Similarly, lack of suitable facies (as in Gale
et al. 2001) cannot be invoked over the long period of time required in the Precambrian
to obliterate any benthic fossil record. In particular, the excellent and microscopic
preservation in the Duoshantuo Formation and similar sediments, none of which yield
convincing bilaterians, is compelling evidence that bilaterians, even tiny ones, were not
then widespread. One must therefore fall back on the idea that rarity of the taxa meant
that the fossil record simply did not pick them up. The question to be addressed, then, is
whether the early stages of the bilaterian (abundance) radiation were truly of this nature—
or, indeed, are radiations in general?
The Cambrian explosion as an ecological radiationDespite some important peculiarities, the events at the beginning of the Cambrian, the
‘Cambrian explosion’ as defined by Budd et al. (2001), should be approached as any other
major evolutionary event, that is, from the initial perspective of the neodarwinian
synthesis. Undoubtedly, the ecological theory of adaptive radiation (Simpson 1953;
Schluter 2000) is the most dominant theory in this field, despite the existence of more
GRAHAM E.BUDD AND SÖREN JENSEN 177