Reinforcement strengths of different intensities of brain stimulation were assessed by means of a two-member behavioral chain. A variable interval schedule of 30 sec was the first-member, and five lever presses, each rewarded with stimulation, was the second. It was found that response rate on the VI schedule continued to increase beyond the intensity value which produced peak rate on the second-member, self-stimulation lever. It was concluded (1) that brain-stimulation reinforcement strength cannot be assessed adequately by means of selfstimulation rate, and (2) the chaining technique employed in the present experiment appears promising as an analytical tool in brain-stimulation research. Finally, some aspects of the data suggested a fatigue or stimulation-adaptation phenomenon.It has been assumed that changes in the intensity of rewarding electrical stimulation of the brain alters behavioral reinforcement strength (Sidman, Brady, Boren, Conrad, and Schulman 1955). One of the earliest tests was to observe changes in self-stimulation rate on a continuously available response lever as a function of stimulation intensity. (Throughout this report, "self-stimulation rate" refers to the procedure of determining response rate when each lever press is followed by electrical stimulation with the animal pacing its own responses.) In extensive studies of the problem (e.g., Reynolds, 1958), it has been found that self-stimulation rate increases with increases in intensity up to a point. Beyond a point for each animal, self-stimulation rate declines with further increases in intensity. The question remains as to whether the decline in self-stimulation rate at higher intensities reflects a decrease in reinforcement strength and/or the development of aversive properties. Possibly the decrease in self-stimulation rate at high intensities reflects neither of those events, but rather a slowing of the rate due to some other effect of stimulation, such as seizures or respondent interference with ongoing operants. Valenstein (1964) has discussed the problem in some detail. ' Skinner, 1957). A twomember behavioral chain, with self-stimulation in the second member, permits a comparison between self-stimulation rate and the characteristics of first-member performance. The first member of the behavioral chain employed in the present paper was a variable interval of 30 sec, and changes in VI response rate were compared with changes in self-stimulation rate as the intensity of brain stimulation was varied& METHOD Three male albino rats with chronically implanted, stainless steel electrodes aimed at the vicinity of the mammillary bodies of the posterior hypothalamus were adapted to the experimental situation and given preliminary training in a two-lever rat box. The final baseline chain performance was as follows. The left-hand lever was continuously available while the right-hand lever was retracted out of the box. Responding on the left-hand lever resulted in the insertion of the retracted lever on a variable interval schedule of 30 s...