The couplons of the cardiomyocyte form nanospaces within the cell that place the L-type calcium channel (Ca(v)1.2), situated on the plasmalemma, in opposition to the type 2 ryanodine receptor (RyR2), situated on the sarcoplasmic reticulum. These two molecules, which form the basis of excitation-contraction coupling, are separated by a very limited space, which allows a few Ca(2+) ions passing through Ca(v)1.2 to activate the RyR2 at concentration levels that would be deleterious to the whole cell. The limited space also allows Ca(2+) inactivation of Ca(v)1.2. We have found that not all couplons are the same and that their properties are likely determined by their molecular partners which, in turn, determine their excitability. In particular, there are a class of couplons that lie outside the RyR2-Ca(v)1.2 dyad; in this case, the RyR2 is close to caveolin-3 rather than Ca(v)1.2. These extra-dyadic couplons are probably controlled by the multitude of molecules associated with caveolin-3 and may modulate contractile force under situations such as stress. It has long been assumed that like the skeletal muscle, the RyR2 in the couplon are arranged in a structured array with the RyR2 interacting with each other via domain 6 of the RyR2 molecule. This arrangement was thought to provide local control of RyR2 excitability. Using 3D electron tomography of the couplon, we show that the RyR2 in the couplon do not form an ordered pattern, but are scattered throughout it. Relatively few are in a checkerboard pattern--many RyR2 sit edge-to-edge, a configuration which might preclude their controlling each other's excitability. The discovery of this structure makes many models of cardiac couplon function moot and is a current avenue of further research.