Pharyngula: How creationism should be taught in the classroom:
"The way I actually teach genetics is essentially a temporal series of criticisms. I start with Darwin's pangenesis for a little historical background, and tell them this is wrong, and here's why, criticizing it on the basis of it's ad hoc nature and its failure to fit experimental observations. Then I introduce Mendel, and we see his view of particulate, quantifiable inheritance, and how it superseded Darwin, and then I show how parts of it are wrong, with experiments that show how it fails, which leads into linkage. And then I show how some of our initial concepts of chromosomal inheritance are wrong, with work done on extrachromosomal factors. Step by step, we build a case for a complex and detailed understanding of the rules of heredity by experiment…where even the experiments that go 'wrong' (that is, don't show us the results we expected from existing theory) help us acquire a deeper understanding of the process.
In a way, it's a pretty ruthless business. Weak handwaving, of the sort that Darwin was doing in his theory of inheritance, doesn't cut it and gets chopped apart savagely with the bloody cleaver of experiment. Creationism is far, far weaker than Darwin's 19th century proposal, so you can guess how it fares.
When the proponents of creationism ask that their nonsense be taught in school, there is an implicit expectation that the scientists will put away their implements of destruction and suspend the savagery while their delicate little flower of unsupportable fluff is discussed reverentially. That is not going to happen. If it did, it wouldn't be a science class."