We compared the arousal and hyperactivity produced by intraperitoneal (i.p.) injections of thyrotropin-releasing hormone (TRH, pGlu-His-Pri-NH2; 10, 20, 30 and 60 mg/kg) and 0.3 and 2 mg/kg d-amphetamine (low and moderate amph., respectively) by measuring the occurrence of discrete behavioral items with a behavioral sampling and scoring method. To minimize extraneous variables affecting activity, rats were caged singly inside isolated observation chambers and tested in the daytime after a 2.5 h period of habituation. Under these conditions, vehicle (0.9% NaCl)-treated rats were inactive and either rested or slept through 80% of all time samples taken in the hour after injection. Both TRH and amph. produced significant arousal from sleeping, but TRH, at all doses tested, produced less arousal than moderate amph. and a pattern of behavioral responses which differed from both low and moderate amph. Moderate amph. produced marked increase in forward locomotion and rearing, but low amph. and TRH did not. Both TRH and low amph. increased grooming (perhaps simply by increasing wakefulness), but TRH failed to increase sniffing, a cardinal feature of ampha.-induced excitement. Unlike amph., TRH produced wet-dog shakes, piloerection, tail elevation and teeth chattering. Both mod. amph. and TRH significantly produced increased activity when compared to controls as assessed with photocell counts, though the amph. effect was more robust. The lack of arousal after i.p. injections of thyroid-stimulating hormone (10 I.U./kg) or melanocyte-stimulating hormone release-inhibiting factor (Pro-Leu-Gly-NH2; 60 mg/kg) is evidence that TRH-induced arousal is neither mediated by activation of the pituitary-thyroid axis nor by a non-specific effect of tripeptides generally.