http://www.spacelab.net/~catalj/box/behe.htm addresses Behe's argument:
"Michael Behe thinks he can detect design in biochemical systems and structures that are "irreducibly complex" (IC), but that conclusion is wrong.
"By irreducible complexity I mean a single system which is composed of several interacting parts that contribute to the basic function, and where the removal of any one of the parts causes the system to effectively cease functioning. An irreducibly complex system cannot be produced gradually by slight, successive modifications of a precursor system, since any precursor to an irreducibly complex system is by definition nonfunctional." [*]"
"By his definition, many systems we see around us are IC, and yet have developed gradually and without any overarching design. Think of large cities, market economies and especially ecosystems. So given an IC system, it will either be a product of design, or of an undirected, stochastic process. The truth is, we should expect Darwinian evolution to produce such systems in biology, and not be surprised to find them. The underlying processes are called co-adaptation and co-evolution, and they have been understood for many years. Biochemical pathways are not build-up one step at a time in linear assembly-line fashion to meet some static function. They evolve layer upon layer, contingency upon contingency, always in
flux, and retooling to serve current functions. The ability of life to evolve in this fashion has itself evolved over time. Detecting IC does not indicate design, and therefore Behe's hypothesis collapses."
And now the fallacy:
The Fallacy of Conclusion by Analogy
When it comes to explaining science to the public, analogies and metaphors are essential tools of the trade. We all can better understand something new and unusual, when it is compared to something we already know: a cell is like a factory, the eye is like a camera, an atom is like a billiard ball, a biochemical system is like a mouse trap. An A is like a B, means A shares some conceptual properties with B. It does not mean A has all the properties of B. It does not follow that what is true for B is therefore true for A. Analogies can be used to explain science. But analogies cannot be used to draw conclusions or falsify scientific theories. Yet Behe commits this fallacy throughout his book. For example:
1.A mousetrap is "irreducibly complex" - it requires all of its parts to work properly.
2.A mousetrap is a product of design.
3.The bacterial flagellum is "irreducibly complex" - it requires all of its parts to work properly.
4.Therefore the flagellum is like a mouse trap.
5.Therefore the flagellum is a product of design.
Another great website can be found at
http://www.cbs.dtu.dk/dave/Behe.html
And http://www-polisci.mit.edu/bostonreview/BR21.6/orr.html discusses some of the issues as well
Read the part
"Reducible Complexity
The first thing you need to understand about Behe's argument is that it's just plain wrong. It's not that he botched some stray fact about evolution, or that he doesn't know his biochemistry, but that his argument-as an argument-is fatally flawed. To see this we need to first get clear about what kinds of solutions to irreducible complexity are not open to Darwinism. "
For some interesting information relevant to the discussion here.
Lungs:
"The point is there's no guarantee that improvements will remain mere improvements. Indeed because later changes build on previous ones, there's every reason to think that earlier refinements might become necessary. The transformation of air bladders into lungs that allowed animals to breathe atmospheric oxygen was initially just advantageous: such beasts could explore open niches-like dry land-that were unavailable to their lung-less peers. But as evolution built on this adaptation (modifying limbs for walking, for instance), we grew thoroughly terrestrial and lungs, consequently, are no longer luxuries-they are essential. The punch-line is, I think, obvious: although this process is thoroughly Darwinian, we are often left with a system that is irreducibly complex. I'm afraid there's no room for compromise here: Behe's key claim that all the components of an irreducibly complex system "have to be there from the beginning" is dead wrong. "
More at http://www-polisci.mit.edu/bostonreview/evolution.html