developmental constraint - senecio7の日記

Arthur W 2004 The effect of development on the direction of evolution: toward a twenty-first century consensus. Evol Dev 6:282-288.

Gould & Lewontin (1979) ... was a deliberate attempt to provoke a response from the neo-Darwinian school
one of the many responses they got came from Cheverud (1984), who ... argued that "the genetic variance/covariance matrix of quantitative genetic theory measures developmental constraints"
though it does not explain them

the phrase "developmental constraint" has, in my view, been unhelpful because it suggests a negative role for development in the evolutionary process
Gould (1989) attempted to solve this problem by urging biologists to adopt a broad usage of "constraint" that included positive as well as negative effects
he supported his argument with a seventeenth century quotation in which "constrained" clearly meant "compelled"
the fact that he had to resort to such an old usage to make his point says it all

recently, I proposed a different solution
first, we need a "neutral" umbrella term to refer to all changes in ontogeny caused by mutations of developmental genes − developmental reprogramming
second, given that some kinds of reprogramming are likely to be easier to produce than others because of the nature of the genotype–phenotype map (Wagner and Altenberg 1996), we need a term for this "ease variation" that will be interpreted as including both positive and negative effects − developmental bias
third, we can now regard constraint as one of two subcategories of bias, referring to developmental trajectories that are "hard" to produce
the other subcategory − "easy" changes − can be called developmental drive

if, summed over a very large number of different mutations of the gene concerned, the probability of being rerouted in some directions is higher than for other directions, then there is developmental bias
developmental bias is likely to be the rule rather than the exception in all species of multicellular organisms
ubiquity of occurrence does not ensure a directional evolutionary effect
whether or not developmental bias has such an effect is exactly what is at issue
the common theme underlying all manifestations of bias is that the variation upon which natural selection acts is highly structured rather than random

one much more important caveat ... renders my "developmental bias affects the direction of evolution" assertion a hypothesis rather than a statement of fact
this concerns the structure of the adaptive landscape
we have little information on the "shapes" of adaptive landscapes in nature
an important avenue for future research lies in attempting to quantify the structure of actual, as opposed to imaginary or artificial, adaptive landscapes
it has been argued by Beldade et al. (2002) that selection can break bias
... argues for a dominant role of natural selection, rather than internal constraints, in shaping existing variation
I agree with this conclusion in relation to the particular experiment conducted
it would be unwise to use it as a basis for generalizing beyond these specific experiments
conclusions reached on the basis of artificial selection cannot be applied to natural selection
natural selection is not trying to do anything
it just happens
it does not deliberately put fitness maxima in places that will test the breakability of bias
we end up with the crucial question of what shapes of adaptive landscapes and what forms of bias prevail in nature, and how these map to each other

one reason why I suspect that developmental bias is even more important in the origin of novelties than in ordinary evolution is that it seems likely that the frequency distribution of "magnitudes of mutational effect on development" is shifted upward in such origins
this is not to say that novelties involve "macromutations"
a shift in this distribution − a little less "micro" a little more "meso" − probably characterizes the origin of many novelties in many lineages

bias will be the norm rather than the exception in all populations of all species
that is not to say that there is no need to gather further evidence of bias
there is such a need, especially in the case of biases in embryogenesis
most of the "correlated characters" work done by quantitative geneticists concerns postembryonic development
population-level studies of embryos have been almost nonexistent to date
the most pressing need is to obtain evidence not just for the existence and commonness of bias but rather for its claimed directional effect on the evolutionary process