many biologists proceeded to search for molecular genetic
homologies that could reveal previously unimagined
morphological homologies (Janies and DeSalle, 1999).
Geoffroy would have been thrilled. There are, however,
problems with the view that animals are all alike.
The most serious problem, in my view, is that
homologous genes may sometimes be involved in
the development of adult structures that are clearly
not homologous (Striedter and Northcutt, 1991).
For example, insect wings and vertebrate nervous
systems both depend on hedgehog function for normal
development, but this does not make neural
tubes and insect wings homologous (Bagun˜a and
Garcia-Fernandez, 2003). Instead, findings such as
this suggest that evolution tends to work with highly
conserved ‘master genes’ (Gehring, 1996) or, more
accurately, tightly knit assemblies of crucial genes
(Nilsson, 2004), which it occasionally reshuffles by
altering their upstream regulatory elements and/or
downstream targets.