Abstract
Recently the crystal structure of a potassium channel has been determined (1). The emerging structural features turned out to be a beautiful illustration of both the power and limitation of the approach of structure—function analysis by mutagenesis. The crystallographic data essentially confirmed models developed on the basis of mutagenesis data; in particular, the lining of a critical part of the channel pore, the selectivity filter, by the P segment was verified. It also exemplifies how much more detail the crystal structure can give as compared to the crude information even the most sophisticated mutagenesis approach can yield.