Although first described in the final decade of the 19th century, the axis responsible for atrioventricular conduction has long been the source of multiple controversies. Some of these continue to reverberate. When first described by His, for example, many doubted the existence of the bundle we now name in his honour, while Kent suggested that multiple pathways crossed the atrioventricular junctions in the normal heart. It was Tawara who clarified the situation, although many of his key definitions have not universally been accepted. In key studies in the third decade of the 20th century, Mahaim then suggested the presence of ubiquitous connections that provided "paraspecific" pathways for atrioventricular conduction. In this review, we show the validity of these original investigations, based on our own experience with a large number of datasets from human hearts prepared by serial histological sectioning. Using our own reconstructions, we show how the atrioventricular conduction axis can be placed back within the heart. We emphasise that newly emerging techniques will be key in providing the resolution to map cellular detail to the gross evidence provided by the serial sections

Those using the mouse for the purposes of electrophysiological research presume that the morphology of the conduction axis is comparable with the human arrangement. As yet, however, to the best of our knowledge, no direct comparison has been made between the species. By comparing our extensive histological findings in the human heart with comparable serially-sectioned datasets prepared from adult murine hearts, we aimed to provide this information. When comparing the gross anatomy, we used three-dimensional datasets of neonatal mice hearts prepared using episcopic microscopy. The overall cardiac architecture is comparable, although the mouse has a persistent left superior caval vein draining via the coronary sinus. An inferior pyramidal space, and an infero-septal recess, are both present in the murine heart, although not as well developed as in the human heart. The overall arrangement of the conduction axis is similarly comparable, albeit with subtle differences reflecting the incomplete wedging of the subaortic outflow tract in the murine heart. Most significantly, the findings in both species reveal the presence of extensive superior septal pathways, which perhaps explain the finding of base-to-apex activation of the ventricular mass known to occur in the murine heart.