Sir

The visual shock of last week's ‘intelligent design’ cover was matched by the perceptiveness of your News Feature (Nature 434, 1062–1065; 2005) on the seepage of this slyly religious ideology into science curricula. This stuff should certainly be kept out of high schools, but I am ambivalent about its presence in our universities, where free discussion is of greater value than correctness — political, scientific or otherwise.

On the other hand, I'd be properly rebuked and sanctioned for incompetence if I were to assert in my undergraduate physical-chemistry course that the intricate, precisely exponential distribution of velocities observed in a collection of gas molecules is simply too perfect and beautiful to have arisen from random collisions, but that we should instead consider a mechanism by which intelligent designers — let's call them “Maxwell's angels” — individually push on each molecule, while keeping in intelligent communication with each other to maintain this distribution.

One reason that scientists famously fail in rebutting ID is that we use the wrong analogies. Evolution is not a blind watchmaker or any other kind of engineer, but rather a short-order cook, and — looking at the phenomenally complicated structures — one who is less like Isaac Newton than Rube Goldberg or W. Heath Robinson.

A terrific argument against ID came to me recently after two consecutive talks, one on the Wnt signalling pathway, the next on G-protein crosstalk in control of cellular calcium. Just look at the details, and you'll immediately abandon all thoughts that biological systems were designed with any intelligence whatsoever.