A curious thing: I reported yesterday on this lovely work on the genetic regulation of pelvic limbs in fish, and got a comment that the "modularity and surprisingly robust flexibility" of the system was evidence of design. Quite the contrary, I see evidence of mechanisms that permit integrated evolution of organisms, with no designer required. There's another example briefly described in a news article in this week's Science (Pennisi, 2004) that describes work presented at a Cold Spring Harbor conference on the evolution of developmental diversity. It expands further on this matter.
Cichlid fishes are examples of a recent, rapid radiation into new forms. One of the reasons behind their success seems to be the adaptability of their jaws, which have allowed them to diversify into many ecological niches by changes in their feeding apparatus (Albertson et al., 1999, 2003). What's becoming clear is that the modular and interlinked network of regulatory genes does not hinder change, it facilitates concordant change in the patterns of expression of multiple genes to produce an integrated morphology of the jaw.