In a Pavlovian conditioning procedure, rats were exposed to an odor conditioned stimulus (CS) and then were given morphine with its effect serving as the unconditioned stimulus (US). After four CS-US pairings, the CS was tested alone to assess the presence of an analgesic conditioned response (CR) using a hot-plate test. In Experiment 1a, two groups were conditioned by pairing either 10 mg/kg morphine or saline with an odor CS. In Experiment 1b, two groups were given an odor CS paired or unpaired with 10 mg/kg morphine. These results established that an odor cue can support a morphine-induced analgesic CR. Experiment 2 characterized the dose-effect curve (0, 3, 10, and 30 mg/kg morphine) using an odor conditioning procedure. The dose-effect curve showed an inverted U-shaped function, with the 10 mg/kg morphine group having significantly longer paw-lick latencies compared to all other groups. This finding contrasts with the monotonically ascending dose-effect curve for the analgesic unconditioned response (UR) to morphine.