The potent in vivo vasodilator actions of acetylcholine were first described by Henry Dale exactly 100 years ago. However, isolated rings from blood vessels usually constrict to acetylcholine, a conundrum that exercised Robert Furchgott as early as the 1950s, when he wrote an extensive review on the pharmacology of vascular smooth muscle. It was not until 1980 that Furchgott solved the conundrum, with the publication of his classic paper demonstrating the obligatory role of endothelium in the relaxation of vascular smooth muscle caused by acetylcholine. It then took a further 7 years for Salvador Moncada’s group to show that NO, enzymatically formed from arginine, was the endothelium-derived relaxing factor (EDRF) described by Furchgott. At the same time Masashi Yanigasawa and colleagues discovered the potent endothelium-derived constriction factor (EDCF) endothelin-1 (ET 1). However, unlike NO, it is unlikely that ET-1 plays a highly dynamic role in the control of vascular tone. Several other candidate EDRFs (not least prostacyclin and H2O2) and EDCFs (e.g. other prostanoids) exist. Amongst EDRFs, some cause vasorelaxation by inducing hyperpolarisation of smooth muscle, and were thus named endothelium-derived hyperpolarising factors (EDHFs). The best characterised of these are epoxy-eicosatrienoic acids (EETs). Particularly in resistance arteries, there is evidence of myo-endothelial gap junctions directly linking electrical activity between endothelial and smooth muscle cells. Tudor Griffith in the 1990s pioneered studies to demonstrate that these gap junctions were an important route by which endothelium could directly hyperpolarise smooth muscle and cause vasorelaxation without the need for secreted factors. Subsequently others (particularly Steven Segal, Kim Dora and Chris Garland) have shown that this route of electrical coupling between endothelium and smooth muscle is also required to enable efficient conducted vasodilatation – the physiological process whereby local vasodilation induced at one site is conducted rapidly and co-ordinately upstream. While there is now a wealth of data indicating that lack of NO bioavailability (often due to uncoupling of NO synthase due to loss of its required cofactor tetrahydrobopterin) underlies impaired endothelium-dependent vasodilatation in disease states such as diabetes, more recent data increasingly suggest that the relative importance of other individually characterised EDRFs and EDCFs in the control of vascular tone depends critically on the vascular bed being studied and its state of health or disease.