Second-order conditioning (SOC) in pigeons, but not rats, appears to involve an association between the second-order stimulus (S2) and the first-order stimulus (S1). Nairne and Rescorla (1981) suggested it was the use of stimuli from the same modality that promoted an association between S2 and SI in pigeon SOC studies. In support of their hypothesis, they demonstrated that pigeons, like rats, did not form an association between S2 and SI when these stimuli were from different modalities. In this study, we sought to determine whether rats, like pigeons, would associate S2 with SI when these stimuli shared the same modality. Female Lister rats injected with LiCI after consuming .12M saline solution (S1) showed an aversion to a 15% sucrose solution (82) that was subsequently paired with the saline. This was so regardless of whether S2 and 81 had been presented sequentially (Experiment 1) or simultaneously (Experiment 2). Only in Experiment 2, however, did extinction of the aversion to saline diminish the aversion to sucrose; that is, employing stimuli from the same modality was not a sufficient condition, of itself, to allow rats to associate 82 with S1.Second-order conditioning (SOC), in which one stimulus (S2) elicits a conditioned response (CR) as a result of being paired with another stimulus (S 1) whose ability to act as a reinforcer stems from prior association with an unconditioned stimulus (US), has been the subject of several research papers in recent years. Of particular relevance to the experiments reported here are those studies that attempted to establish whether SOC involves an association between S2 and S1 such that S2 elicits whatever response is appropriate to S1.The strategy employed by these studies was to manipulate S1 (once SOC had been established to 52) in such a way as to alter the CR elicited by S1 and then test the response to S2. A change in the CR to S2 corresponding to that produced in the CR to S 1 would have provided evidence of an association between S2 and S1. This procedure produced conflicting results, however, depending upon whether the subjects were rats or pigeons. Experiments using an autoshaping procedure with pigeons