In the years leading up to the turn of the nineteenth century, most debates relating to conduction of the cardiac impulse revolved around whether the process was myogenic or neurogenic. Stannius then showed unequivocally that conduction was myogenic, but his studies failed to clarify the anatomic substrates for initiation and dissemination of the cardiac impulse. Indeed, anatomical studies in the final decade of the 19th century had served only to confuse the issue. In 1893, Wilhelm His the younger, working in Leipzig, proposed that a solitary muscular fascicle crossed the plane of atrioventricular insulation. Stanley Kent, a physiologist working in London, suggested that multiple pathways permitted atrioventricular conduction in the normal heart. The confusion was sufficiently great for Arthur Keith, one of the foremost anatomists of the time, to express scepticism at the existence of the purported atrioventricular bundle. He was then sent, by James Mackenzie, the account of the atrioventricular conduction axis produced by Sunao Tawara, a Japanese pathologist working in the laboratory of Ludwig Aschoff, at Marburg in Germany. As Keith explained, with Tawara’s account to guide him, he was able to demonstrate unequivocally the atrioventricular bundle, and to trace it from the atrioventricular node to the ventricular conducting fascicles. It was no exaggeration for Keith to state that the discovery of the atrioventricular conduction axis by Tawara ushered in a new epoch for cardiac research. Keith had not only been asked by Mackenzie to verify the findings of Tawara, but also by Wenckebach to explore the structure of the junction between the superior caval vein and the right atrium. Stimulated by the findings of Tawara, and helped by Martin Flack, a medical student at the time, he re-examined the cavoatrial junctions of several mammalian hearts. It was Flack who made the crucial breakthrough, finding a wonderful structure within the terminal groove, a structure that was found in all animals studied, including man. Thus was found the sinus node. Controversy then shifted to the internodal atrial myocardium, but Aschoff and Monckeberg, at an important meeting held in Erlangen, Germany, in 1910, established the criteria for anatomic recognition of histologically specialised tissues that retain their relevance even today. They pointed out that the nodes were histologically discrete, and could be traced through serial sections, whilst tracts shared both these features, also being insulated from the adjacent areas of working myocardium. Using these criteria, we can now recognise the structures identified by Kent as remnants of an extensive array of specialised tissue in the developing heart which become sequestrated within the atrial vestibules in the postnatal heart. All the works of these giants of the past, therefore, retain their relevance for our current understanding of the anatomic disposition of the conduction tissues of the heart.