Development: Three Grades of Ontogenetic Involvement 187mark of selection. So development explains the distribution of form, only insofar
as it isnon-adaptive. Those who wish to accord a significant role to development
in the explanation of biological form often find themselves in the uncomfortable
position of having to deny what to many seems self-evident, that so much of bio-
logical form is adaptive.^5 The debate as it is played about between adaptationism
and its adversaries often takes this form ([Gould and Lewontin, 1979]; [Sober,
1996]).
This conception of the role of development in evolution is what I call the ‘First
Grade of Ontogenetic Involvement’. I believe it is the orthodox position on the
relevance of ontogeny for the process of evolution. First grade ontogenetic involve-
ment is predicated upon the tripartite distinction among evolutionary processes
handed down from the sub-organismal version of the modern synthesis. Develop-
ment plays no role in the inheritance, recombination or initiation of variants in
genotype space. Nor can it alter the adaptedness of populations, by driving a pop-
ulation up an adaptive peak in phenotype space; that is selection’s job. Given the
presumed fragmentation of the component processes of evolution, the only con-
tribution developmental processescouldmake to evolution is that of a constraint
against the adaptation-promoting power of selection.
There is ample reason to resist the fragmentation of evolution.^6 For one thing,
there is no particularly compelling reason to suppose that replicators or genes are
exclusively the units of inheritance. The principal motivation for this conviction
resides in what Matteo Mameli [2005] calls the ‘conception-donation’ model of
inheritance. On this view whatever is inherited is ‘donated’ by one parent or
another at conception. Clearly, everything donated to an offspring at conception
is a replicant derived from a parent.^7 But there is little reason to suppose that
only these things can be inherited.
As we have seen, all selection requires of inheritance is the reliable recurrence
of phenotypes within familial lineages. Replication may well guarantee this, but
there is little reason to think thatonlyreplication does. There are myriad mecha-
nisms by which such reliable recurrence may be secured. Variant phenotypes can be
passed on by imprinting [Mameli, 2004], learning, cultural transmission ([Jablonka
and Lamb, 2004]; [Boyd and Richerson, 2005]), the constancy of ecological con-
ditions [Gilbert, 2001] and crucially any number of epigenetic and developmental
mechanisms [Jablonka, 2001]. Replication may be inheritance, but inheritance is
not replication.
This point is elegantly illustrated by a parable offered by Mameli [2004]. A
population of butterflies lays its eggs on a certain species of host plant. The
(^5) The adaptedness of form seems so self-evident (at least to some) that it inspired John May-
nard Smith to declare “The main task of any evolutionary theory is to explain adaptive com-
plexity, i.e. to explain the same set of facts which Paley used as evidence of a Creator [Maynard
Smith, 1969].
(^6) Notwithstanding the gifts of the three ‘Weissmen’ discussed above.
(^7) I take it that anything that is copied and passed on is a replicant; this category includes
more than just genes. Methylation patterns, maternal cytoplasm, cell membranes [Jablonka and
Lamb, 2005] all qualify on this definition.