Abnormal vascular tone is responsible for many of the complications seen in cirrhosis making the identification of the pathophysiology of abnormal dilatation a major focus in hepatology research. The study of abnormal vascular tone is complicated by the multiple vascular beds involved (hepatic, splanchnic, peripheral, renal and pulmonary), the differences in the underlying cause of portal hypertension (hepatic versus pre-hepatic) and the slow evolution of the hyperdynamic state. The autonomic nervous system, circulating vasodilators and abnormalities in vascular smooth muscle cells (receptors, ion channels, signalling systems and contraction) have all been implicated. There is overwhelming evidence for an overproduction of NO (nitric oxide) contributing to the peripheral dilatation in both animal models of, and in humans with, cirrhosis and portal hypertension. This review focuses on the proposal that endotoxaemia, possibly from gut-derived bacterial translocation, causes induction of NOS (NO synthase) leading to increased vascular NO production, which is the primary stimulus for the development of vasodilatation in cirrhosis and its accompanying clinical manifestations. The current controversy lies not in whether NO production is elevated, but in which isoform of NOS is responsible. We review the evidence for endotoxaemia in cirrhosis and the factors contributing to gut-derived bacterial translocation, including intestinal motility and permeability, and finally discuss the possible role of selective intestinal decontamination in the management of circulatory abnormalities in cirrhosis.