In the mammalian myocardium, potassium (K(+)) channels control resting potentials, action potential waveforms, automaticity, and refractory periods and, in most cardiac cells, multiple types of K(+) channels that subserve these functions are expressed. Molecular cloning has revealed the presence of a large number of K(+) channel pore forming (alpha) and accessory (beta) subunits in the heart, and considerable progress has been made recently in defining the relationships between expressed K(+) channel subunits and functional cardiac K(+) channels. To date, more than 20 mouse models with altered K(+) channel expression/functioning have been generated using dominant-negative transgenic and targeted gene deletion approaches. In several instances, the genetic manipulation of K(+) channel subunit expression has revealed the role of specific K(+) channel subunit subfamilies or individual K(+) channel subunit genes in the generation of myocardial K(+) channels. In other cases, however, the phenotypic consequences have been unexpected. This review summarizes what has been learned from the in situ genetic manipulation of cardiac K(+) channel functioning in the mouse, discusses the limitations of the models developed to date, and explores the likely directions of future research.