With the advent of new information about alterations of cardiac gap junctions in disease conditions associated with arrhythmias, there have been major advances in the genetic and metabolic manipulation of gap junctions. In contrast, in naturally occurring cardiac preparations, little is known about cell-to-cell transmission and the subcellular events of propagation or about structural mechanisms that may affect conduction events at this small size scale. Therefore, the aim of this article is to review results that produce the following unifying picture: changes in cardiac conduction due to remodeling cardiac morphology ultimately are limited to changes in three morphologic parameters: (1) cell geometry (size and shape), (2) gap junctions (distribution and conductivity), and (3) interstitial space (size and distribution). In this article, we consider changes in conduction that result from the remodeling of cell size and gap junction distribution that occurs with developmental ventricular hypertrophy from birth to maturity. We then go on to changes in longitudinal and transverse propagation in aging human atrial bundles that are produced by remodeling interstitial space due to deposition of collagenous septa. At present, experimental limitations in naturally occurring preparations prevent measurement of the conductance of individual gap junctional plaques, as well as the delays in conduction associated with cell-to-cell transmission. Therefore, we consider the development of mathematical electrical models based on documented cardiac microstructure to gain insight into the role of specific morphologic parameters in generating the changes in anisotropic propagation that we measured in the tissue preparations. A major antiarrhythmic implication of the results is that an "indirect" therapeutic target is interstitial collagen, because regulation of its deposition and turnover to prevent or alter microfibrosis can enhance side-to-side electrical coupling between small groups of cells in aging atrial bundles.