Hepatic-portal infusion of .77 M NaCI (.5% aWl resulted in a shorter latency to drink than did jugular infusion for both sham-operated rats and rats subjected to abdominal vagotomy. However, .5-h water intake was not differentially. affected by the route of hypertonic saline infusion in either group. Hepatic-portal infusion of an isotonic saline load resulted in greater water intake than did jugular infusion for both vagotomized and control rats, and the latency was shorter in vagotomized rats, but not in control animals, after hepatic-portal infusion than after a jugular infusion. Hypertonic saline was a more potent dipsogenic stimulus than isotonic saline for both routes of infusion, and vagotomy reduced water consumption produced by osmotic challenge. The results provide some evidence for a hepatic drinking system by demonstrating a differential effect of hepatic-portal and systemic saline infusions. However, this effect persists after abdominal vagotomy and possibly 'is mediated wholly, or in part, by other visceral innervation. In addition, abdominal vagotomy reduced water intake, but not latency to drink, induced with osmotic challenge given via either vascular route, an effect that may possibly be due to alterations in gastrointestinal function or to loss of afferent input from hypothetical visceral osmo-or sodium-sensitive cells that modulate drinking.Recent experimental results have provided evidence that receptors located within the circulation of the liver detect perturbations in water and electrolyte balance and activate appropriate homeostatic mechanisms (Sawchenko & Friedman, 1979). Electrophysiological investigations have provided evidence for the existence of osmo-or sodium-sensitive cells in several species (Adachi, Niijima, & Jacobs, 1976;Andrews & Orbach, 1974;Niijima, 1969). In addition, a number of studies have demonstrated the greater efficacy of hepatic-portal infusions of saline loads than of systemic infusions in eliciting physiologicalcompensatoryresponses (Daly, Roe, & Horrocks, 1967;Haberich, Aziz, & Nowacki, 1965;Lydtin, 1969;Passo, Thornborough, & Rothballer, 1973). However, relatively few investigations have considered the behavioral consequences of the activation of hepatic sodium-or osmo-sensitive mechanisms. Lin and Blake (1971) hepatic-portal infusion of hypertonic saline decreased saline drinking, whereas infusion of hypertonic saline through the vena cava, as well as infusion of an equiosmotic sodium-free solution through either vein, failed to alter intake in a test of rats deprived of fluid for 24 h and then presented simultaneously with both water and saline. More recently, Blake and Lin (1978) demonstrated that, following right cervical vagotomy, hepatic-portal infusion of hypertonic saline no longer decreased saline drinking preference' although this denervation did not affect water or saline intake during control infusions or during a 24-h period of ad-lib food and water access.Thus, there has accrued evidence for a hepatic mechanism that is sensitive to fluctuations i...