Two sheep were fitted with permanent cannulus in the first cerebral ventricule and permanent electrodes were stitched to the wall of the reticulum. They were housed in individual pens and fed lucern pellets twice a day (total duration 1.30 hour) at fixed hours. Food intake was weighed. Chewing movements were recorded permanently by means of a submandibular balloon. Reticular motility was recorded by electromyography during feeding periods. Feeding behaviour and rumination were previously recorded during two control periods. No treatment was applied during the first control period of 6 days (Ta). During the second one (Tb), daily intracerebroventricular (ICV) injections of saline (0.2 ml)) were given for 4 consecutive days. During the following 4 days (Na period), the sheep received isovolumic ICV injections of nalorphine (1 mg per day). Saline was injected again ICV for the last 4 days of the experimental period (Tc). During the period of nalorphine administration there was an important fall in the circadian percentage of rumination of one sheep and a total suppression in the other, discarding the possibility of a volumetric effect. There was also a significant decrease of food intake in the two sheep but nalorphine did not modify the frequency of reticular contractions in any case. Regarding to the general behaviour, the drug induced a general excitation with bleating and increased responsiveness to stimuli. Oral activity included compulsive and continuous chewing movement, creaking the teeth, gnawing, nipping and an apparent activity of searching food for up to 15 hours. In conclusion, nalorphine and morphine effects upon the sheep are compared. The effects of the two drugs are exactly the same regarding to the decrease of rumination and feeding intake and the modification of general behaviour, but morphine decreases markedly the reticular rhythm whereas nalorphine does not. Besides, it seems that the animals get tolerant more readily to morphine than to nalorphine.